article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
the uniform electron gas , or jellium model , is an archetypal example in solid - state physics and many - body theory .
its treatment , in the hartree fock ( hf ) approximation , can be found in classic textbooks @xcite , where , we learn that the hf equations applied to the ground state of the jellium , admit plane wave solutions with energy - wavevector dispersion relation given by , @xmath0 @xmath1 is the fermi wavevector , @xmath2 .
the single - particle energy @xmath3 is the sum of the free - electron energy , @xmath4 , and the single - particle exchange energy .
the fermi wavevector @xmath1 is often expressed in terms of the mean radius per particle @xmath5{9\pi/4 k_f^3}$ ] @xcite ; for typical values of @xmath6 in metals , the two terms in ( [ disphf ] ) are comparable in size .
it is well known in the literature that the dispersion relation ( [ disphf ] ) has a logarithmically divergent derivative at the fermi energy , shown in fig .
[ fig : hf ] . another marked difference between the free electron result and the hf solution for jellium , evident in fig .
[ fig : hf ] , is the considerably increased bandwidth of the hf dispersion . finally , it is well known that in the hf approximation the dos for jellium vanishes at the fermi level ( fig .
[ fig : hf ] ) , since the dos is inversely proportional to the derivative of the dispersion .
the zero in the dos at the fermi level suggests that jellium is a semimetal , in obvious disagreement with experimental evidence for simple metals , such at sodium or aluminium , which are described accurately by the jellium model . in the literature ,
the qualitatively wrong description of jellium in the hf approximation is attributed to the long range of the coulomb repulsion @xcite .
it is well known that the flawed description can be corrected by the introduction of electronic many body correlation effects @xcite , which screen the bare coulomb potential and thus eliminate the unphysical divergent derivative of the dispersion relation , the zero in the dos at the fermi level , and also reduce the bandwidth of the hf dispersion relation of jellium . in an effort to understand whether hf s lack of screening actually plays a role
, we revisit the hf study of jellium , attempting to correct the qualitative errors of the hf description , but without including any form of electronic correlation . for this purpose
, we employ slater s hyper - hf ( hhf ) theory for the ground and the excited states of an @xmath7-electron system @xcite .
specifically , we use the single - particle hhf equations by gidopoulos and theophilou @xcite , who considered an @xmath7-electron system described by a hamiltonian @xmath8 and then variationally optimised the average energy @xmath9 of all configurations ( @xmath7-electron slater determinants @xmath10 ) constructed from a basis set of @xmath11 spin - orbitals , @xmath12 .
the aim in hhf theory is to obtain approximations , at the hf level of description , for the ground and the excited states of an @xmath7-electron system .
these states are represented by @xmath7-electron slater determinants , constructed from a common set of spin - orbitals .
obviously , to have the flexibility to describe excited states , the number of spin - orbitals ( @xmath11 ) must exceed the number of electrons .
for example , say we are interested to approximate the ground and excited states of the helium atom . in the hf ground state of he ,
the @xmath13 orbital ( @xmath14 ) is doubly occupied . to study a couple of excited states ,
we need at least one more spin - orbital and the next one is : @xmath15 . with the three available spin - orbitals , @xmath16
, we can form three configurations for the he atom ( two - electron slater determinants ) : @xmath17 $ ] , @xmath18 $ ] , @xmath19 $ ] .
in hhf theory , we variationally optimise the three common spin - orbitals simultaneously , by minimising the sum of the expectation values @xmath20 .
the minimisation leads to the hhf single - particle equations for the three spin - orbitals .
it turns out that these equations resemble the ground - state hf equations for the lithium atom ( three electrons ) but with a weakened coulomb repulsion between the three electrons , to keep balance with the nuclear charge which has remained that of the he nucleus . in general , in hhf
theory @xcite for an @xmath7-electron system , one considers a set of @xmath11 orthonormal spin - orbitals , with @xmath12 . on this spin - orbital basis set
one may define , @xmath21 , @xmath7-electron slater determinants . the derivation of the single - particle hhf equations in ref .
is based on theophilou s variational principle @xcite , @xmath22 where @xmath23 are the @xmath24 lowest eigenvalues of the @xmath7-electron hamiltonian @xmath8 .
an extension of the variational principle , with unequal weights in the sums in ( [ theo ] ) was proposed by theophilou @xcite , and independently by gross , oliveira , kohn @xcite .
these variational principles can be derived as special cases from the helmholtz variational principle in statistical mechanics @xcite . in particular
, the inequality in ( [ theo ] ) arises as the high temperature limit of the helmholtz variational principle .
optimisation of the @xmath11 spin - orbitals @xmath25 to minimise the sum of the energies on the l.h.s . of ( [ theo ] ) leads to the following single - particle equations for the spatial part of the spin - orbitals @xcite ( in atomic units ) : @xmath26 \varphi_i({\ensuremath{\mathbf{r } } } ) \nonumber \\ & + { 1 \over \lambda } \ , \sum^r_{j=1}\left[j_j({\ensuremath{\mathbf{r } } } ) - \delta_{s_j , s_i } k_j({\ensuremath{\mathbf{r } } } ) \right]\varphi_i({\ensuremath{\mathbf{r } } } ) = \lambda_i \varphi_i({\ensuremath{\mathbf{r } } } ) \ , , \end{aligned}\ ] ] where , @xmath27 and @xmath28 are the coulomb and exchange operators respectively .
@xmath29 signifies the attractive potential of the nuclear charge . for @xmath30 , eqs .
[ subhf ] reduce to the familiar ground - state hf equations . in eqs .
[ subhf ] , the orbitals @xmath31 , with @xmath32 , are correctly repelled electrostatically by a charge of @xmath33 electrons .
in contrast to the ground - state hf case @xcite , this holds true even for the orbitals that are not occupied in the hhf ground - state slater determinant as long as these orbitals are variationally optimised , i.e. , for @xmath31 , with @xmath34 .
furthermore , the orbitals that are left variationally unoptimised , i.e. , @xmath31 , with @xmath35 , are repelled by a charge of @xmath36 electrons . in the hhf equations , the well - known asymmetry in the treatment of the variationally optimised and unoptimised orbitals by the nonlocal exchange operator
@xcite is still present , but softened ( for large @xmath37 ) , compared with ground - state hf .
we note that for @xmath38 , koopmans theorem @xcite ceases to hold for the hhf equations .
the hhf equations ( [ subhf ] ) have the form of ground - state hf equations for a virtual system of @xmath11-electrons , where the electronic coulomb repulsion is multiplied by @xmath39 : @xmath40 .
therefore , the calculation of the optimal spin - orbitals to represent the ground and excited - states of an @xmath7-electron system in the hhf approximation , is reduced to the calculation of the ground - state hf orbitals of a fictitious system with a greater number of electrons @xmath12 , and scaled down electronic coulomb repulsion .
a related approach is the `` super - hamiltonian method '' by katriel @xcite .
finally , before applying the hhf equations to jellium , we remark that correlated , approximate eigenstates of the hamiltonian @xmath8 can be obtained by diagonalising the matrix @xmath41 @xcite , where @xmath10 are the @xmath7-electron hhf slater determinants .
this configuration - interaction method employs the hhf spin - orbitals , which are optimised to represent on equal footing the ground and the excited states of @xmath8 .
similarly to the hf ground state , the hhf equations for jellium admit plane wave solutions .
it follows that the ground - state @xmath7-electron slater determinant and its total energy are the same in the hf and hhf approximations .
although the hf and hhf equations for jellium admit the same solution for the orbitals , the dispersion relations for the single - particle energies @xmath42 and @xmath43 differ .
in particular , the hhf dispersion , @xmath44 , results from an optimisation that involves a broader range of wavevectors than the hf dispersion @xmath42 . following the standard treatment in textbooks @xcite , it is straightforward to work out directly the solution of the hhf eqs .
[ subhf ] . here
, we exploit the similarity of eqs .
[ subhf ] with ground - state hf equations of an @xmath11-electron system , to obtain that the hhf dispersion relation , @xmath43 , will be given by ( [ disphf ] ) with the single - particle exchange energy scaled down by the factor @xmath45 : @xmath46 @xmath47 is the fermi wave vector of the virtual @xmath11-electron system , @xmath48 dividing @xmath49 , and taking the thermodynamic limit , @xmath50 , with the ratio @xmath37 fixed , we obtain : @xmath51 substitution of the above into eq .
( [ krdisprel0 ] ) yields the desired expression for the single particle energy levels of jellium , in terms of @xmath37 and the fermi wavevector @xmath1 of the actual @xmath7-electron system : @xmath52 the dispersion relation in eq .
( [ disprelation0 ] ) reduces to the ground - state hartree - fock result for @xmath53 , and to the free electron dispersion relation , @xmath54 , in the limit @xmath55 ( see fig . [
fig : ss ] ) . for increasing @xmath37 ,
the bandwidth of the hhf dispersion , @xmath56 , decreases compared to the ground - state hf dispersion , @xmath57 . for @xmath55 the exchange term in hhf dispersion vanishes and @xmath43
reduces to the free - electron result .
importantly , the wave vector at which the logarithmically divergent derivative occurs is shifted from @xmath1 to @xmath47 , such that the divergence no longer occurs at the fermi energy of the physical @xmath7-electron system , when the number of optimised orbitals is @xmath58 .
the dos , @xmath59 , can be obtained directly from eq .
[ disprelation0 ] @xcite and is given by the parametric equation : @xmath60^{-1}\ , .
\label{dos}\end{aligned}\ ] ] @xmath61 is expressed in terms of @xmath47 ( rather than @xmath62 ) to keep the notation simple . for any finite @xmath37 ,
the dos still vanishes .
however , as shown in fig .
[ fig : ss ] , the zero in the dos occurs at the fermi energy of the fictitious @xmath11-electron system , @xmath63 , rather than the fermi energy of the physical system @xmath64 .
in metals , screening is an important effect that reduces the range of the effective repulsion between electrons , shielding any charge at distances greater than a characteristic screening length . in the literature of many - body theory @xcite and solid - state physics @xcite , where jellium is a paradigm , the qualitatively flawed description of metals by the hf approximation is attributed to the long - range nature of the coulomb interaction , which , combined with the neglect of correlation , deprives from the hf equations the flexibility to model the phenomenon of screening .
this understanding of hf s failure is further supported by the softening of the divergence in the slope of @xmath3 , after replacing the bare coulomb potential in the hf nonlocal exchange term by a screened coulomb potential @xcite . on the other hand , in the theoretical chemistry literature
, it is well known that the hf nonlocal exchange term , in finite systems , gives rise to several counterintuitive results , reminiscent of the hf anomalies in jellium , which are not associated with hf s lack of correlation .
for example , handy et .
al @xcite disproved the widely held view @xcite , that the hf nonlocal exchange potential decays as @xmath65 at large distances . in particular , in ref .
it was demonstrated that the asymptotic decay of an occupied hf orbital @xmath31 with eigenvalue @xmath66 is not @xmath67 , as would be expected from the @xmath65 tail of the exchange potential . on the contrary , in hf all the occupied orbitals @xmath31 decay uniformly at large distances away from the system , regardless of their energy eigenvalue .
as already mentioned , it is also widely known that the hf exchange operator deals with the occupied and unoccupied orbitals in the ground - state hf slater determinant in an asymmetric way @xcite : for an @xmath7-electron system , the ( variationally optimised ) occupied orbitals are repelled electrostatically by a charge of @xmath33 electrons , while the ( variationally unoptimised ) unoccupied orbitals feel the stronger repulsion of @xmath7 electrons , making the unoccupied orbitals too diffuse , and raising their energy eigenvalue to unphysically high values @xcite .
it follows that the energy to excite an electron from an occupied hf orbital @xmath31 to a virtual hf orbital @xmath68 , keeping the other occupied orbitals frozen , must be smaller than the eigenvalue difference @xmath69 .
this is because the energy @xmath70 incorporates the coulomb interaction of the orbital @xmath68 , hosting the electron after excitation , with all the occupied orbitals , including the orbital @xmath31 accommodating the same electron before excitation . in this sense
, the coulomb interaction of the orbitals @xmath31 and @xmath68 can be interpreted as a form of self - interaction raising the energy of the virtual level @xmath70 .
it is similar to the `` ghost '' self - interaction discussed in ref . .
this interpretation is consistent with viewing the virtual hf energies as single - particle levels of the @xmath7-electron system .
alternatively , by extending the proof of koopmans theorem @xcite , it can be shown that @xmath71 is equal to the negative of the electron affinity of the system to bind an electron at the virtual orbital @xmath68 , see e.g. ref . .
the interpretation of the virtual energies as negative affinities amounts to regarding the unoccupied levels as virtual levels of an @xmath72-electron system .
we note that for jellium , where the occupied and virtual orbitals are plane waves , the self - interaction error of the virtual energy levels ( discussed above ) vanishes in the thermodynamic limit .
therefore , it makes sense to consider that the hf unoccupied levels represent virtual single - particle levels of the @xmath7-electron system , and to study their dispersion relation and density of states . a consequence of the asymmetry in the treatment of occupied versus unoccupied orbitals in the hf equations , is presented by bach et al .
@xcite , who prove that , for a finite system , the highest occupied spin - orbital in the ground - state unrestricted - hf slater determinant , is nondegenerate : a nonzero gap separates it from the lowest unoccupied spin - orbital , even in systems with an odd number of electrons , in contrast to physical expectation .
recently , hollins et al . , using two different methods , the optimised effective potential , or exact exchange method @xcite , and the local fock exchange potential ( lfx ) method @xcite , showed that it is possible to omit correlation and still obtain an accurate dispersion relation @xmath3 for simple metals , provided the exchange potential term is local @xmath73 @xcite .
the lfx potential , defined as the local exchange potential with the hf ground - state density @xcite , is particularly interesting in our context : even though the determination of the lfx potential is based on the same ground - state hf calculation that gives a very poor prediction for the dispersion of simple metals ( e.g. na , al ) , the band - structure of the lfx potential @xcite almost coincides with the band - structure of the local density approximation in density functional theory @xcite , which , by construction , is very accurate for these systems . with regard to the vanishing of the hf dos of metals at the fermi energy , hollins et al .
@xcite argue that it appears to be an artifact of the hf nonlocal exchange operator .
our work supports this point of view , by studying the nonlocal hf exchange term directly .
we find that the well - known anomalies of the hf description of jellium are still present in the solution of the hhf equations .
however , the location of the divergent derivative of @xmath44 and the zero in the dos are no longer at the fermi level of the actual @xmath7-electron system . instead
, they are positioned at the border separating the variationally optimised and unoptimised orbitals . by choosing to variationally optimise a very large number of orbitals , the unphysical zero in the dos can be pushed to very high energies , avoiding completely the window of single - partice energies that can be of any relevance to the ground state of jellium .
therefore , it is no longer justified to relate the mobile divergence in the derivative of the hhf dispersion and the travelling zero in the hhf dos of jellium , with the lack of electronic correlation in the hhf approximation , even though these anomalies can still be removed by screening the coulomb repulsion in the exchange term .
this strengthens the view that the failure of the ground - state hf solution for jellium is also an artifact of the nonlocal exchange operator .
at least part of the explanation of the divergent derivative of the dispersion seems to be the asymmetry in the treatment of the variationally optimised and unoptimised orbitals : as the wavevector @xmath74 crosses @xmath47 from below , the plane wave solutions of the excited - state hhf equations are subjected to a discontinuous drop of the nonlocal exchange term and a correspondingly discontinuous increase in the coulomb repulsion from the hartree term .
this raises the single - particle eigenvalue @xmath44 to higher energies , diminishing the dos in the neighborhood of @xmath75 .
the same mechanism operates in the divergent slope of the ground - state hf dispersion @xmath42 . in conclusion
, we suggest that the qualitative failure of the hf approximation for jellium is unrelated to the lack of correlation in the hf approximation . instead , this failure is another example in the list of counterintuitive results caused by the nonlocality of the hf exchange operator .
our work complements the work of hollins et al . in ref .
and our conclusion is contrary to what is currently written in almost any textbook in the fields of many body theory and solid state physics . to the best of our knowledge , slater gave the only hint so far in the literature that the hf failure may be unrelated to electronic correlation . in his textbook on `` insulators , semiconductors and metals ''
@xcite he writes that `` it can not be in accordance with experiment to write the total energy of the electronically excited state of the crystal as a sum of one - electron energies of the type of ... '' ( cf . the hf solutions ) .
`` a great deal of effort has gone into explaining this apparent paradox connected with the free - electron theory of electrons in metals .
... it is the impression of the present author , however , that we do not need to look for any deep and profound explanation '' .
j. kohanoff , n.i .
gidopoulos , `` density functional theory : basics , new trends and applications '' , in the _ handbook of molecular physics and quantum chemistry _ , edited by stephen wilson , vol . * 2 * , part 5 , chapter 26 , pp 532568 , john wiley & sons , ltd , chichester ( 2003 ) | we revisit the hartree - fock ( hf ) calculation for the uniform electron gas , or jellium model , whose predictions divergent derivative of the energy dispersion relation and vanishing density of states ( dos ) at the fermi level are in qualitative disagreement with experimental evidence for simple metals .
currently , this qualitative failure is attributed to the lack of screening in the hf equations . employing slater s hyper - hartree - fock ( hhf ) equations , derived variationally , to study the ground state and the excited states of jellium , we find that the divergent derivative of the energy dispersion relation and the zero in the dos are still present , but shifted from the fermi wavevector and energy of jellium to the boundary between the set of variationally optimised and unoptimised hhf orbitals
. the location of this boundary is not fixed , but it can be chosen to lie at arbitrarily high values of wavevector and energy , well clear from the fermi level of jellium . we conclude that , rather than the lack of screening in the hf equations , the well - known qualitative failure of the ground - state hf approximation is an artifact of its nonlocal exchange operator .
other similar artifacts of the hf nonlocal exchange operator , not associated with the lack of electronic correlation , are known in the literature . |
quantum phase transitions ( qpt s ) @xcite , i.e. transitions taking place at zero temperature , have recently become a subject of intense research in theoretical and experimental condensed matter physics .
these transitions are driven by zero temperature quantum fluctuations and are accessed by varying a non - thermal control parameter @xmath7 , like chemical composition , magnetic field , pressure , etc .
@xcite , at a critical value @xmath8 , known as quantum critical point ( qcp ) .
physical properties associated with quantum fluctuations are quite distinct from those associated with thermal fluctuations responsible for conventional temperature - driven phase transitions . for example , in metallic systems , the material at qcp shows an unconventional non - fermi liquid ( nfl ) behaviour @xcite .
the existence of such critical points has been proved to hold the key to many technologically exploitable phenomena , e.g. the occurrence of high - temperature superconductivity @xcite , behaviour of two - dimensional electron gases @xcite , etc .
perhaps the most easily understandable class of a qpt is the ferromagnet - to - paramagnet ( fm - pm ) transition at 0 k in itinerant electron systems as a function of the strength of exchange interaction between electron spins @xcite .
the fm - pm qpt occurring in metallic ni@xmath0pd@xmath1 alloy system across its qcp ( @xmath6 = 0.026 ) is one such example , wherein signatures of quantum criticality have been observed in bulk macroscopically @xcite and microscopically @xcite , and in nanodimensions macroscopically @xcite . several theories , including quantum ising model , quantum rotor model , hubbard model and heisenberg - spin model , with renormalization group approaches and space - time scalings , exist for understanding qpt s and associated critical behaviours of different kinds of systems undergoing qpt @xcite .
computational studies of qpt s , however , are posed with various formidable challenges including ( i ) the problem of quantum many - particle , ( ii ) occurrence of the transition only in the thermodynamic limit of infinite system size , and ( iii ) anisotropic space - time scaling at qcp @xcite .
the first of these , i.e. the quantum many - particle problem , has the possibility to be addressed with sophisticated approximation methods like hartree - fock or density functional theory ( dft ) @xcite .
these theories , however , neglect fluctuations , which are crucial for understanding phase transitions including qpt , and hence would fail in describing the quantum fluctuations associated with qpt s @xcite .
nonetheless , it is still intriguing to explore what dft computations of the non - fluctuating ground state say vis - a - vis the qpt occurring in a material system .
dft results have a relevance also in the case of the usage of hubbard model , where these are to be used as an initial input .
although some computational methods , like quantum monte carlo , density matrix renormalization group and dynamical mean field theory have been proposed for finding qcp and predicting quantum critical behaviour @xcite , none have hitherto been utilized for computations to the best of authors knowledge .
thus , using dft for computations of ground state properties of systems undergoing a qpt and comparing the results with experiments would perhaps be the first , although still very small , step in this direction . in this scenario and with the above motivation
, we take in this work the ni@xmath0pd@xmath1 alloy system to represent the class of materials exhibiting qpt , compute the ground state electronic and local magnetic properties of the system across the qcp by dft , and seek any correlation with the experimental quantum - fluctuation driven properties reported in the literature @xcite .
the full - potential linearized augmented plane wave ( flapw ) method of dft as implemented in the code wien2k @xcite was employed to perform the present calculations .
the generalized gradient approximation as introduced by perdew , burke and ernzerhof @xcite was taken as the exchange - correlation functional .
pure pd and ni crystal structures were constructed in the cubic fm@xmath9 m space group .
the lattice parameters were then optimized by calculating the total energy at different volumes and fitting the result to the birch murnaghan equation of state @xcite for both pd and ni . in order to construct the structures of ni@xmath0pd@xmath1 alloys with @xmath10 and in the vicinity of @xmath6 ,
the following strategy has been adopted : out of very many possible combinations @xmath11 of generating pd supercells of sizes @xmath12 , only the smallest supercell volume combinations have been taken in such a manner that replacing just the ( 0 , 0 , 0 ) pd by ni suffices to create the particular ni@xmath0pd@xmath1 alloy .
the purpose of taking the smallest supercells has been to minimize the computational cost .
since the study is a comparative one , computational accuracy for any individual @xmath4 is not very crucial , and hence this strategy of computation should be good enough .
the @xmath4 values and the corresponding supercells are listed in tab .
the equilibrium lattice constants for each of the supercells were once again obtained by the same method as for pd and ni .
further , the atomic coordinates were relaxed in all the impurity cases to reduce the atomic forces to less than 1 mry / au . in the flapw method
, the basis function is set differently in two regions - non overlapping muffin - tin sphere of radius r@xmath13 , and the remaining interstitial region - around an atom . in the muffin - tin region ,
the potential is taken to be atom - like and atomic spherical wave functions are used . in the interstitial region , on the other hand , the electrons behave like plane waves . in present calculations , the r@xmath13 value for both ni and pd atoms are taken to be 2.5 au .
the maximum multipolarity of the spherical wavefunctions has been taken as 10 , while the length @xmath14 of the maximum wavevector for the plane - wave basis set has been limited by @xmath15 = 7.0 .
the charge density was fourier expanded up to @xmath16 = 12 .
for sampling of the brillouin zone , a k - mesh with 110 irreducible k - points in the irreducible wedge of the brillouin zone was used . for each concentration
both unpolarized and spin - polarized self - consistent field calculations were performed .
.[tab1]values of ni concentration @xmath4 and the corresponding supercell size ( @xmath12 ) . [ cols="<,<,<",options="header " , ] +
the first step of analysis is to see which of the non - magnetic and ferromagnetic states is more stable for a particular @xmath4 according to the calculations , and whether the inferences are consistent with the existing experimental reports @xcite . for this
, the difference @xmath17 between the total energies from the spin - polarized ( e@xmath18 ) and unpolarized ( e@xmath19 ) calculations ( @xmath20 ) for each composition has been calculated and plotted against @xmath4 in fig . 1 .
two features can be identified in the figure : ( i ) all the points corresponding to @xmath21 supercells vary more or less smoothly with @xmath4 , while if @xmath22 cases are categorized separately , the point corresponding to the @xmath23 supercell of this category is scattered far away .
for this reason , it would not be inappropriate to consider just the @xmath21 supercells for further analyses .
therefore , only the @xmath21 supercells are being analysed hereafter .
( ii ) for small @xmath4 , the points corresponding to the @xmath21 supercells can be shown to follow a @xmath24 dependence , as displayed in the inset of fig .
this is almost the same as the variation @xmath25 of the fm - pm transition temperature @xmath26 with @xmath4 as reported by nicklas _
_ @xcite . with these kinds of highly simplified structures , approximating @xmath27 to @xmath24 must be acceptable .
this gives further strength to the strategy of taking only @xmath21 supercells for further analyses .
with @xmath4 for various supercells .
inset : small-@xmath4 portion of the curve along with a @xmath24 fit.,scaledwidth=60.0% ] figure 2a shows the total densities of states ( tdos s ) from unpolarized calculations for all the studied compositions with @xmath21 supercells ( tab .
an examination of the figure suggests that on incremental introduction of ni , the shape of the pd ( @xmath28 ) tdos gets modified systematically in such a way that each of the three major peaks , as well as the whole tdos , broadens .
this is an indication of some hybridization between certain pd and ni orbitals @xcite .
this aspect will be investigated further later . at this point
, it would be interesting to see the behaviour of the ` tdos at the fermi energy ' @xmath29 ) with @xmath4 , as displayed in fig .
2b . from the figure , a power - law dependence of n(e@xmath30 ) on @xmath4 similar to that of @xmath26 on @xmath4 as reported by nicklas _ et .
@xcite can be observed beyond @xmath31 . assuming the stoner exchange integral @xmath32 for the onset of ferromagnetism @xcite to be slightly less than the 1/@xmath33 ) value for pure pd , i.e. , @xmath34
, we can say that ferromagnetism appears in the system at @xmath31 and gets stronger according to a power law with @xmath4 , in agreement once again with the existing report of dependence of @xmath26 on @xmath4 @xcite .
the assumed value @xmath35 of @xmath32 is not different from the value ( 0.39 ) reported by hong and lee @xcite . this way , taking @xmath21 supercells for the analyses
is further justified .
the tdos s from spin - polarized calculations are shown in fig .
qualitatively the same inferences can once again be drawn for each spin component as for the corresponding unpolarized cases .
it would further be interesting to see the @xmath4 dependences of the magnetic moments on ni and its nearest neighbour pd , as derived from the spin - polarized calculations and shown in fig .
an overall increasing trend of the moments on both the atoms with incremental addition of ni to pd can be observed for small ni concentrations from the figure .
this observation can be understood as follows : ni as an impurity in pd host is known to induce ferromagnetic spin - polarization of pd conduction electrons and give rise to giant moments in its vicinity @xcite . this way , moments on both pd and ni are likely to increase with ni doping on an average for low ni concentrations . however , a smooth variation may not be obtainable from the calculations ( as is observable from fig .
3b ) because the model structures taken in the calculations are far different from and far simpler than real experimental samples @xcite . we would now investigate the afore mentioned hybridization between certain pd and ni orbitals . according to the calculations , the local dos s of ni and pd are dominated by 3@xmath3 and 4@xmath3 partial dos s ( pdos s ) , respectively ( not shown ) .
it would thus be sufficient to compare the 3@xmath3 pdos of impurity ni and 4@xmath3 pdos of its nearest neighbour ( nn ) pd . the pd 4@xmath3 pdos s for @xmath36 and ni 3@xmath3 pdos s for @xmath37 are shown in fig .
4a . that there are variations in both the types of pdos s with @xmath4 , which in turn are indicative of ni 3@xmath3-pd 4@xmath3 hybridization , is apparent at the first sight of the figure . for further analysis
, we adopt the method of semi - quantifying hybridizations from dos s as used in our previous work @xcite . for this
, we determine the width ( @xmath38 ) , centroid ( @xmath39 ) and height ( @xmath40 ) of the ni 3@xmath3 pdos s and the width ( @xmath41 ) and centroid ( @xmath42 ) of the nn pd 4@xmath3 pdos s , and then look at their variations with @xmath4 .
an increase in hybridization then would be signalled as a concurrent widening and lower energy shifts of the centroids of the two pdos s , and simultaneous increase and decrease of the width and height of a particular pdos @xcite .
plots of @xmath38 and @xmath41 versus @xmath4 , @xmath38 and @xmath40 versus @xmath4 , and @xmath39 and @xmath42 versus @xmath4 are shown in figs .
5a , 5b and 5c , respectively .
the above mentioned variations , characteristics of an increasing ni 3@xmath3-pd 4@xmath3 hybridization with @xmath4 , can be readily observed from the figures .
this leads to the first outcome of the study that the ni 3@xmath3-pd 4@xmath3 hybridization increases monotonically on introducing ni impurities incrementally into pd matrix .
existence of ni 3@xmath3-pd 4@xmath3 hybridization in diluted @xmath43ni alloys has been reported earlier in a photoemission study @xcite , while such @xmath4-dependence has not been studied . and
nn pd 4@xmath3 partial densities of states.,scaledwidth=40.0% ] however , there is another interesting observation associated with all the curves in figs .
5a - c : apart from the systematic variations with @xmath4 as mentioned above , each curve possesses an anomaly in the vicinity of @xmath2 .
a close observation suggests that @xmath38 has a somewhat steep rise around @xmath6 and then increases with a power - law like trend , if a linear background is subtracted .
this can be likened with the power - law @xmath4 dependences of @xmath26 and @xmath44 and hence may have a correlation with the occurrence of magnetic order in pd on ni doping .
a concurrent but opposite trend is shown by @xmath40 and @xmath42 .
this reverse nature of the two quantities is quite likely from the hybridization point of view as discussed above .
however , @xmath41 and @xmath39 exhibit a rather sharp peak and a similar dip , respectively , around the qcp , over and above the quite apparent monotonic power - law - like increase or decrease with @xmath4 .
these peak - like anomalies certainly can not arise due to the presence of quantum fluctuations at the qcp , since such fluctuations are not implementable in the calculations .
but , then there is no explanation for the occurrence of these exceptions . keeping in mind the smooth variation of the total energy of the supercells and the concurrent behaviours of other quantities with the onset of magnetic order ,
the anomalies can perhaps also not be discarded as mere artefacts .
this way , there appears to be a definite relation between the orbital hybridization of the non - fluctuating ground state and the emergence of quantum fluctuations in the ni@xmath0pd@xmath1 alloy system .
exactly what this relation could be can however not be determined or explained at this stage . and @xmath38 , ( b ) width @xmath38 and height @xmath40 , and ( c ) @xmath42 and @xmath39 , with @xmath4 .
power - law behaviours and the @xmath41 peak in ( a ) are schematically indicated ; for others , these features are qualitatively apparent and hence are not shown.,scaledwidth=40.0% ]
we have computed electronic and local magnetic properties of the non - fluctuating ground state of ni@xmath0pd@xmath1 alloy system across the quantum critical point by _ first - principles _ density functional theory .
the simplest model structures constructed with the smallest possible @xmath21 pd supercells , the ( 0 , 0 , 0 ) atom of which is replaced by ni , have been chosen for the study .
the primary reason to choose the @xmath21 structural constructions has been that the total energy for these varies smoothly with @xmath4 .
the density of states at the fermi energy follows a power - law with @xmath4 similar to that obeyed by the experimental ferromagnetic to paramagnetic transition temperature .
the local magnetic moments on the impurity ni and nearest neighbour pd atoms have a monotonic increase with @xmath4 .
the study gives rise to two main outcomes ( i ) the ni 3@xmath3-pd 4@xmath3 hybridization increases monotonically , and according to the occurrence of magnetic order , on introducing ni impurities incrementally into pd matrix , and ( ii ) the hybridization is anomalous at the qcp , which is indicative of a possible scenario of the existence of a definite relation between the orbital hybridization and the emergence of quantum fluctuations in the ni@xmath0pd@xmath1 alloy system .
p. blaha , k. schwarz , g. madsen , d. kvasnicka and j. luitz 1999 wien2k : an augmented plane wave + local orbitals programme for calculating crystal properties , techn .
universitt wien , austria , isbn 3 - 9501031 - 1 - 2 . | _ first - principles _ density functional theory computations of electronic structure and local magnetic properties of the non - fluctuating ground state of ni@xmath0pd@xmath1 alloy system around its quantum critical point @xmath2 have been performed .
the density of states at the fermi energy and certain other parameters characterizing the ni 3@xmath3 - pd 4@xmath3 hybridization apparently follow power - laws with @xmath4 similar to that obeyed by the reported ferromagnetic to paramagnetic transition temperature .
the width of pd @xmath5 density of states ( dos ) and centroid of ni 3@xmath3 dos show peak - like anomalies in the neighbourhood of @xmath6 , and so indicate a possible scenario of the existence of a definite relation between the orbital hybridization and the emergence of quantum fluctuations in the system . ni@xmath0pd@xmath1 alloy , quantum phase transition , hybridization anomaly |
the effective field theory of qcd , called chiral perturbation theory ( chpt ) @xcite , has become a very successful methodology for low - energy hadron physics .
this effective field theory is based upon the goldstone character of the low - lying pseudoscalar mesons and incorporates the appropriate symmetries of the strong interaction . in this approach ,
one obtains a systematic expansion in powers of external momentum and light quark masses , or equivalently in the number of loops .
however , in order to be renormalized order by order , additional low - energy constants have to be introduced at each order in the chiral expansion .
these low - energy constants are not fixed by chiral symmetry alone but have to be determined phenomenologically .
at one - loop order in the chiral expansion only a small number of low - energy constants have to be introduced @xcite .
these are known rather accurately @xcite and have been used in order to relate different observables to each other .
therefore , one - loop chpt provides finite predictions for many different low - energy observables in terms of only a very limited amount of empirical input .
even though this has been remarkably successful , the two - loop corrections are important in order to improve the accuracy of the predictions . however , two - loop computations are very laborious and have until now only been accomplished in a limited number of cases .
furthermore , the total number of low - energy constants increases dramatically at this order in the chiral expansion @xcite . since an extensive phenomenological analysis of these new two - loop low - energy constants is not available at present , they have so far been evaluated either from the given process or from underlying models .
since chpt provides a systematic perturbative expansion in powers of external momentum and light quark masses , it will satisfy unitarity perturbatively .
this implies that the deviation from exact unitarity at a given order provides an estimate of the energy range where the truncation of the chiral expansion is applicable , i.e. the energy range where higher order corrections can be neglected .
although these higher order corrections could in principle be calculated and thereby increase the energy range of the chiral expansion , chpt is in practice limited to the first few powers in the loop expansion .
this is due to the rapid increase in both computational effort and new low - energy constants at each additional order .
thus , the unitarity restriction will give a major limitation on the applicability of chpt .
an important special case of this fact is in the presence of resonances where higher order unitarity corrections become essential .
therefore , in order to extend the range of applicability of chpt , unitarity will be of the utmost importance . a rather general way to impose unitarity on the chiral expansion
is by the inverse amplitude method ( iam ) @xcite , which can be justified by the use of unitarity in a dispersive approach .
since this method combines unitarity and dispersion relations with the chiral expansion , it is likely that it will be more powerful than chpt alone .
the iam applied to one - loop order in the chiral expansion has been extensively analysed @xcite and has indeed proven to be a successful method to extend the range of applicability of one - loop chpt .
however , the iam is not restricted to the one - loop approximation but can in general be applied to any given order in the chiral expansion .
therefore , it might well be that the iam applied to a given order can be improved by applying it to the next order .
if this turns out to be the case , the iam could be considered as a systematic way of improving chpt order by order in the chiral expansion .
the iam was originally applied to @xmath0 scattering and the pion form factors to one loop in chpt @xcite . by now
, the full two - loop @xmath0 scattering amplitude has been evaluated by a dispersive analysis @xcite and a field theory calculation @xcite , respectively .
the pion form factors have also been evaluated to two loops some time ago @xcite .
therefore , it is now possible to apply the iam to two loops in the chiral expansion for both @xmath0 scattering and the pion form factors . in the present paper ,
this extension of the iam is analysed and compared to both the one - loop case and one- and two - loop chpt in order to study whether the iam is a systematic way of improving chpt .
section [ sec : pipi ] is devoted to @xmath0 scattering , sec .
[ sec : form ] to the pion form factors , and the conclusions are given in sec .
[ sec : con ] .
@xmath0 scattering is of fundamental theoretical importance , both because it only involves the self - interaction of the lightest strongly interacting particle , and because it plays a significant role for low - energy hadron phenomenology .
the @xmath0 scattering amplitude can be expressed in terms of a single function @xmath1 as @xmath2 where @xmath3 , @xmath4 , and @xmath5 are the mandelstam variables .
the definite isospin amplitudes @xmath6 are given in terms of the function @xmath1 as @xmath7 and the projection of these isospin amplitudes onto partial waves is given by @xmath8 in the elastic region @xmath9 , unitarity implies that the partial waves satisfy the following relation @xmath10 where @xmath11 is the phase - space factor . as a consequence of eq .
( [ eq : pfuni ] ) , the partial waves can be parametrized in terms of real phase shifts @xmath12 as @xmath13 - 1 \right\ } .\ ] ] the elastic unitarity relation ( [ eq : pfuni ] ) is in practice useful up to the @xmath14 threshold @xcite . above this energy
the extended unitarity relation becomes @xmath15 and @xmath16 is the corresponding phase - space factor . at still higher energies , other important inelastic channels
must be included and the unitarity relation has to be generalized further .
the inelastic channels can be included in the parametrization ( [ eq : parpha ] ) by adding an imaginary part to the phase shifts . in the framework of chpt
, the @xmath0 scattering amplitude is obtained as an expansion in powers of external momentum and light quark masses @xmath17 where @xmath18 is of @xmath19 .
the leading order term @xmath20 corresponds to the current algebra result derived by weinberg @xcite , whereas @xmath21 is the one - loop correction determined by gasser and leutwyler @xcite .
recently , the two - loop correction @xmath22 has also been evaluated by a dispersive analysis @xcite and a field theory calculation @xcite , respectively .
these two approaches agree insofar as the general structure of the amplitude is concerned
. however , the field theory calculation contains additional information regarding the dependence of the chiral low - energy constants which is not obtained in the dispersive analysis .
the full @xmath0 scattering amplitude to two loops in chpt is given in ref .
@xcite where the contributions of @xmath23 have been omitted here . at one - loop order , there are four low - energy constants @xmath24 , @xmath25 , @xmath26 , and @xmath27 , whereas six additional low - energy constants @xmath28 , @xmath29 , @xmath30 , @xmath31 , @xmath32 , and @xmath33 have to be introduced at two - loop order .
the superscript @xmath34 indicates that these low - energy constants depend on the renormalization scale @xmath35 , whereas the full amplitude is scale independent .
the projection onto partial waves of the chiral @xmath0 scattering amplitude ( [ eq : chptexam ] ) gives the following expansion @xmath36 these partial waves will satisfy the elastic unitarity relation ( [ eq : pfuni ] ) perturbatively , i.e. one has to two - loop order @xmath37 the first inelastic channel is the four pion intermediate state , which is a three - loop effect of @xmath23 . other inelastic channels with more pions are of even higher order in the chiral expansion .
this implies that the perturbative elastic unitarity relations to two loops will hold at least up to the @xmath14 threshold .
in fact , in the framework of su(2 ) chpt the relations ( [ eq : ppuni ] ) are satisfied for all energies above threshold , whereas in su(3 ) chpt they have to be generalized above the @xmath14 and @xmath38 thresholds .
the generalization above the @xmath14 threshold is obtained from the perturbative interpretation of eq .
( [ eq : epfuni ] ) @xmath39 the further extension of these perturbative relations above the @xmath38 threshold is obtained in a straightforward manner .
the two - loop @xmath0 scattering amplitude has hitherto only been evaluated within the framework of su(2 ) chpt @xcite .
this implies that in the derivation of the iam applied to two loops in the chiral expansion , the elastic approximation is used .
however , the limitations of this approximation and the possible extension to su(3 ) chpt will also be discussed in some detail . from the fundamental principles of s matrix theory
, it can be shown that the @xmath0 partial waves are analytic in the complex @xmath40 plane with a right or unitarity cut for @xmath41 and a left cut for @xmath42 .
therefore , the @xmath0 partial waves can be expressed in terms of dispersion relations with a number of subtractions to ensure the convergence of the integrals .
the same is also true for the inverse of the partial waves or more exactly for the function @xmath43 . neglecting possible pole contributions arising from zeros in the partial waves @xcite
, the four times subtracted dispersion relation for the iam is @xmath44 four subtractions are used in order to ensure that two - loop chpt satisfies a similar dispersion relation . on the right cut , the elastic unitarity relation ( [ eq : pfuni ] )
gives @xmath45 , i.e. this part can be computed exactly .
the same is not true for the left cut , which , however , can be approximated by means of two - loop chpt . expanding the function @xmath46 to two loops in the chiral expansion @xmath47
, one finds that the left cut can be approximated by @xmath48 .
finally , the subtraction constants can also be evaluated to two - loop order in the chiral expansion . denoting these subtraction constants
@xmath49 , @xmath50 , @xmath51 , and @xmath52 , the following dispersion relation is obtained for the iam @xmath53 ds'}{s'^4(s'-s - i\epsilon ) } .\end{aligned}\ ] ] this dispersion relation implies that the partial wave @xmath54 can be obtained from the two - loop chiral expansion .
however , the relation between the iam and two - loop chpt can be simplified by writing down a four - subtracted dispersion relation for the function @xmath55 .
since this is the function @xmath46 expanded to two - loop order , the left cut and subtraction constants in the dispersion relation for @xmath56 will be the same as in eq .
( [ eq : dispinv1 ] ) . for the right cut ,
the perturbative elastic unitarity relations ( [ eq : ppuni ] ) imply that @xmath57 .
thus , the following dispersion relation is obtained @xmath58 ds'}{s'^4(s'-s - i\epsilon ) } , \end{aligned}\ ] ] which is exactly the same as in eq .
( [ eq : dispinv1 ] ) .
therefore , the iam to two loops in the chiral expansion gives the partial wave @xmath54 in the simple form @xmath59 since this is formally equivalent to the [ 0,2 ] pad approximant applied on chpt , the expression ( [ eq : pade2 ] ) will coincide with the chiral expansion up to two - loop order .
however , in deriving ( [ eq : pade2 ] ) two - loop chpt has only been used for the left cut and the subtraction constants , whereas the important unitarity cut was computed exactly .
this is in contrast to chpt where the unitarity cut was only computed perturbatively . as a consequence of this ,
the [ 0,2 ] pad approximant ( [ eq : pade2 ] ) satisfies the elastic unitarity relation ( [ eq : pfuni ] ) exactly and not only perturbatively as in chpt .
the iam can in general be applied to any given order in the chiral expansion , but has up to now almost entirely been restricted to the one - loop approximation @xcite . in this case , the iam gives the partial waves in the form @xmath60 which is formally equivalent to the [ 0,1 ] pad approximant .
this form also satisfies unitarity exactly and coincides with the chiral expansion up to one - loop order .
it is derived in the same way as described above , the only difference being that the left cut and subtraction constants in eq .
( [ eq : dispinv1 ] ) are now approximated only to one - loop order in the chiral expansion .
this shows how the iam to two loops improves the iam to one loop , even though the dominantly unitarity cut is computed exactly in both cases .
the above discussion rests upon the assumption that the leading order term @xmath61 does not vanish .
in fact , this is only true for the lowest partial waves with @xmath62 , whereas the higher partial waves start at one - loop order in the chiral expansion . in these cases ,
the iam must be slightly modified by writing down a dispersion relation for the function @xmath63 . since perturbative
unitarity implies that @xmath64 is real on the right cut for these higher partial waves , the elastic unitarity relation ( [ eq : pfuni ] ) gives @xmath65 .
hence , the right cut can also in these cases be computed exactly , whereas the left cut and subtraction constants can be approximated by means of chpt as above .
however , in order to relate the dispersion relation for the iam to a corresponding one for chpt , one has to go to three loops in the chiral expansion .
this is due to the fact that for the higher partial waves the imaginary part on the right cut starts at @xmath23 in chpt .
should such a three - loop calculation be undertaken in the future , the iam could also in these cases be directly related to chpt in the same way as discussed above .
it would be desirable if the iam could be applied beyond the elastic approximation .
the first important inelastic channel in @xmath0 scattering opens up at the @xmath14 threshold and is due to the @xmath14 intermediate state . at a slightly higher energy
the @xmath38 inelastic channel will also appear , but this is a rather unimportant effect . only at still higher energies will the four - pion intermediate state begin to be significant , together with other inelastic channels .
the @xmath14 and @xmath38 intermediate states are already present at one - loop order in su(3 ) chpt @xcite , and will therefore also appear in an extension of the present two - loop calculation to the su(3 ) case .
thus , it might be that the iam applied to two loops in the su(3 ) chiral expansion could include these inelasticities . in order to try to include the four - pion intermediate state , one would need to apply the iam to three loops since this effect is of @xmath23 in the chiral expansion .
therefore , the following discussion is restricted to the @xmath14 and @xmath38 inelastic channels .
in fact , it will be limited to the important @xmath14 case since the extension to the @xmath66 case is rather obvious .
to include the @xmath14 inelastic channel , the extended unitarity relation ( [ eq : epfuni ] ) has to be used above the @xmath14 threshold in computing the right cut in eq .
( [ eq : dispinv1 ] ) , whereas the other parts are computed in exactly the same way as before . thus , since the relation ( [ eq : epfuni ] ) gives @xmath67 , this expression has to be used above the @xmath14 threshold in eq .
( [ eq : dispinv1 ] ) .
however , in this case it is not possible to compute the inelastic part @xmath68 exactly , so this part must be evaluated to two loops in the su(3 ) chiral expansion @xmath69 .\end{aligned}\ ] ] in order to simplify the relation between the iam and chpt for the su(3 ) case , the right cut in eq .
( [ eq : dispchi ] ) has to be evaluated with the inclusion of the @xmath14 intermediate state . from the extended perturbative unitarity relations ( [ eq : eppuni ] )
, one finds that @xmath70 is given by an expression exactly similar to eq .
( [ eq : inelaex ] ) .
thus , the inelastic case also gives a result of the form ( [ eq : pade2 ] ) , the only difference being that in this case the chiral su(3 ) expansion is used . in a similar way , by expanding the inelastic part @xmath68 to one - loop order , the iam gives a result of the form ( [ eq : pade1 ] ) with chpt evaluated within su(3 ) .
in the presence of the @xmath14 inelasticity , the unitarity cut can not be computed exactly , but one has to expand the inelastic part to a given chiral order . since the inelastic part starts at the @xmath14 threshold , the one- and two - loop chiral expansions would seem to lead to rather unreliable approximations .
however , it is important to realize that for the inelastic iam the exact extended unitarity relation has been applied before the inelastic part is approximated by the chiral expansion . furthermore , if one compare the expansion of the inelastic part for the iam ( [ eq : inelaex ] ) with the corresponding expansion for chpt ( [ eq : eppuni ] ) , the two - loop correction in eq .
( [ eq : inelaex ] ) is likely to be more suppressed compared to the two - loop correction in eq .
( [ eq : eppuni ] ) .
therefore , one might expect that the expansion of the inelastic quantity @xmath68 converges rather rapidly , i.e. the one- and two - loop expansion could be used over a large energy range .
indeed , as will be shown in the case of the pion form factors , the expansion of a similar quantity works very well up to rather high energies . in order to evaluate
how well the inelastic iam agrees with the extended unitarity relation ( [ eq : epfuni ] ) , one needs to determine the @xmath71 partial waves . to do this ,
the iam could be applied to one- and two - loop order for these partial waves . in this way
one finds that the inelastic iam satisfies eq .
( [ eq : epfuni ] ) exactly , provided that the quantity @xmath68 is determined by the leading order term .
thus , the iam is a rather consistent approach , but there is of course no way to know a priori to what extent this method can reproduce the inelasticities .
the chiral expansion depends on a number of low - energy constants not fixed by chiral symmetry alone .
these have to be determined before chpt and the iam can be compared with experimental data . in the present paper , this comparison is limited to the lowest @xmath0 partial waves with @xmath62 , since they are the best known experimentally .
since the chiral low - energy constants depend on the renormalization scale @xmath35 , this scale has to be chosen before the low - energy constants can be fixed .
this scale will be set at the mass of the @xmath72(770 ) resonance , i.e. @xmath73 mev .
in addition , throughout this paper the values @xmath74 mev and @xmath75 mev are used for the pion decay constant and pion mass , respectively .
the @xmath0 partial waves depend on four low - energy constants @xmath24 , @xmath25 , @xmath26 , and @xmath27 to one - loop order in the chiral expansion .
these have been determined phenomenologically rather accurately in one - loop chpt @xcite with results displayed in the first column in table [ table1 ] . for chpt to two loops , these low - energy constants should in principle be re - evaluated taking into account the two - loop corrections . however , since a thorough reevaluation seems rather out of reach at the moment , the one - loop values of @xmath24 , @xmath25 , @xmath26 , and @xmath27 are also used in two - loop chpt .
the additional low - energy constants to two loops @xmath28 , @xmath29 , @xmath30 , @xmath31 , @xmath32 , and @xmath33 may be estimated with the assumption that they are dominated by resonance contributions at the scale of the resonance , an assumption that is well satisfied in one - loop chpt @xcite .
in particular , including contributions from vector and scalar exchange together with terms of @xmath76 coming from su(3 ) one - loop chpt @xcite , one obtains the values for these two - loop low - energy constants given in the second column in table [ table1 ] @xcite .
the iam to a given order depends on the same low - energy constants as chpt to the same order . however , the phenomenological determination of these low - energy constants by the iam do not necessarily coincide precisely with the values obtained in chpt .
this is due to the fact that the iam contains higher order unitarity corrections , which will effect the phenomenological determination of the low - energy constants . some of the low - energy constants in the iam to one and two loops may be determined phenomenologically by fitting the experimental @xmath0 phase shifts . in this
fitting , the data obtained from @xmath77 decays @xcite are used together with the data from high statistics medium energy experiments @xcite . for a recent discussion of the experimental @xmath0 scattering data
@xcite .
the low - energy constants @xmath24 and @xmath25 in the iam to one loop may be determined in this way , whereas the fit will be rather insensitive to the precise values of the other low - energy constants @xmath26 and @xmath27 . for these latter low - energy constants ,
the values obtained in one - loop chpt are also used in the iam to one loop . in the case of @xmath24 and @xmath25 ,
one is used to fix the @xmath72(770 ) resonance defined to be at @xmath78 , whereas the other is determined by fitting the @xmath79 and @xmath80 experimental phase shifts for @xmath81 gev .
this will give an acceptable fit with values of @xmath24 and @xmath25 given in the third column in table [ table1 ] , whereas if one increase the energy range to @xmath82 gev , the fit will not be acceptable . however , the iam to two loops might improve the agreement with the experimental phase shifts up to @xmath83 gev . in this case
the fit is mainly sensitive to @xmath24 , @xmath25 , @xmath32 , and @xmath33 , so the other low - energy constants will be given by the same values as in two - loop chpt . with the use of the @xmath72(770 )
resonance to fix @xmath33 and varying @xmath24 , @xmath25 , and @xmath32 around the values used in two - loop chpt , these low - energy constants are determined by fitting the @xmath79 and @xmath80 experimental phase shifts up to @xmath83 gev . in this case
one obtains an acceptable fit with values of the low - energy constants given in the fourth column in table [ table1 ] .
there is no point in extending this fit further up in energy since inelastic effects will become essential above approximately 0.9 gev .
the low - energy constants in table [ table1 ] have not been assigned any error bars since it is difficult to estimate these error bars taking into account the higher order corrections . in addition , it is not possible to assign reliable error bars on the two - loop low - energy constants estimated from resonance exchange .
nevertheless , from table [ table1 ] it is observed that the values of the low - energy constants in chpt and the iam to both one and two loops are rather consistent with each other .
however , some variations are indeed to be expected due to the different treatments of the higher order corrections . close to threshold
, the @xmath0 partial waves can be parametrized in terms of scattering lengths @xmath84 and slope parameters @xmath85 @xcite . in table
[ table2 ] , the values obtained from chpt and the iam are compared to the experimental information for these threshold parameters . it is observed that one - loop chpt gives relatively large corrections to the leading order current algebra results , whereas the additional two - loop chpt corrections are somewhat smaller .
however , the difference between one and two - loop chpt is in general larger than the corresponding difference between the iam to one and two loops , which supports the expectation that the iam to one loop already includes the most important unitarity corrections .
in fact , the iam to both one and two loops compares remarkably well with two - loop chpt for the @xmath86 s wave , where the unitarity corrections are important even at low energies .
for the other channels , where the higher order unitarity corrections are less important at low energies , the iam compares in general rather well with both one and two - loop chpt . regarding the comparison with the experimental information , both the iam and chpt to one and two loops are quite consistent with the data .
unfortunately , the precision of these data does not allow one to distinguish the higher order corrections from the one - loop contribution , but this might be improved with the forthcoming accurately determination of @xmath87 from the measurement of the lifetime of the @xmath88 atom @xcite .
even though the chiral expansion only satisfies unitarity perturbatively , chpt is not in general restricted to the threshold region .
however , the deviation from exact unitarity will give an estimate of the actual energy range where the given truncation of the chiral expansion is applicable .
this deviation is shown in the argand diagrams in fig .
[ fig1 ] for the three lowest partial waves .
this shows that unitarity is indeed improved at low energies at each additional order in the chiral expansion . on the other hand ,
two - loop chpt begins to deviate significantly from exact unitarity above approximately 0.5 gev for all three partial waves .
therefore , two - loop chpt should only be trusted up to about this energy , whereas the energy range for one - loop chpt is somewhat smaller .
despite the fact that chpt only satisfies unitarity perturbatively , it is possible to define a chiral expansion for the phase shifts @xmath12 @xcite . in fig .
[ fig2 ] , the phase shift difference @xmath89 given by the iam and chpt is compared to the experimental data obtained from @xmath77 decays @xcite .
the iam to one and two loops gives very similar results in the displayed energy region .
furthermore , since these results compare very well to both two - loop chpt and the central experimental data , the higher order unitarity corrections are indeed very well approximated by the iam . as for one - loop chpt
, this appears to be systematically below the central experimental data , although it is still consistent with these data within the error bars . at da@xmath90ne
, it is expected that these error bars can be significantly reduced from new high statistics measurements of the @xmath77 decays @xcite . from this
, it will also be possible to determine accurately some of the low - energy constants in the iam from independent observables @xcite and thereby test the consistency of the iam in great detail . in fig .
[ fig3 ] , the phase shift @xmath79 is shown below @xmath91 gev .
it is observed that the iam to two loops agrees rather well with the data up to approximately 0.9 gev , whereas the iam to one loop only describes the data well up to about 0.7 gev . for one and two - loop chpt ,
both begin to disagree with the data somewhat above 0.5 gev , which is consistent with the unitarity requirement .
in fact , the iam to two loops describes the data rather well up to energies where the elastic approximation will no longer be applicable .
therefore , before one tries to apply this method to even higher energies , the @xmath14 inelasticity must be included in the iam as discussed previously . however , whether this inelastic iam will actually be able to reproduce the @xmath92(980 ) resonance , which is closely related to the @xmath14 inelastic channel , has to await further investigations . in fig .
[ fig4 ] , the phase shift @xmath80 is shown below @xmath93 gev . in this case
, the iam to two loops also seems to improve the iam to one loop at higher energies , even though the data are not very conclusive .
in addition , the iam also in this case successfully extends the range of applicability of chpt . in this channel ,
the first important inelastic effect is due to the four pion intermediate state , which should begin to be significant somewhat above 1.2 gev .
thus , the iam to two loops describes the data well up to energies where this inelasticity should be included .
it also seems that two - loop chpt is rather consistent with the data all the way up to about 1 gev .
however , this is in conflict with the unitarity requirement , which shows that two - loop chpt can only be trusted up to approximately 0.5 gev . finally , in fig .
[ fig5 ] the phase shift @xmath94 is shown below @xmath93 gev .
this channel is dominated by the @xmath72(770 ) resonance , which was used in the determination of the low - energy constants for the iam .
however , it is observed that the iam is not only consistent with the experimental data close to the resonance mass , but also for energies rather far away from this resonance .
furthermore , for energies somewhat above 1 gev the iam to two loops improves the iam to one loop , although inelastic effects should begin to be significant at these energies . as for chpt
, this only describes the low - energy tail of the @xmath72(770 ) resonance .
it has recently been shown that the iam gives the correct analytical structure in the complex @xmath40 plane @xcite .
of course , this method produces the appropriate cuts on the first riemann sheet , since these cuts are already present in chpt .
however , contrary to chpt , the iam can also produce poles on the second riemann sheet , where resonances are related to such poles in the vicinity of the real axis .
indeed , the iam gives the correct pole structure associated with the @xmath72(770 ) resonance .
furthermore , consistent with recent phenomenological analysis @xcite , this method also gives a pole in the @xmath86 s wave responsible for the strong final - state interaction in this channel .
in fact , these poles have only been obtained for the iam to one loop @xcite but the same should also be true in general . since this analytical structure is not trivial at all , this strongly supports the applicability of the iam .
the pion vector form factor ( @xmath95 ) provides important information about the internal structure of the pion .
another quantity of interest is the pion scalar form factor ( @xmath96 ) defined by the matrix element of the quark density .
these form factors depend on one kinematical variable @xmath40 , giving by the square of the four - momentum transfer .
they are analytical functions in the complex @xmath40 plane with a unitarity cut starting at the @xmath0 threshold .
the form factors will only be considered in the isospin limit , and the scalar form factor will be normalized according to @xmath97 .
thus , they may be expanded around @xmath98 as @xmath99 where @xmath100 is a generic symbol for the vector and scalar form factors . in the elastic region ,
one has the following unitarity relation @xmath101 which implies that the phase of @xmath100 will coincide with the @xmath0 phase shift @xmath12 in accordance with watson s final - state theorem @xcite .
more precisely , in the elastic region the phase of @xmath95 will coincide with the @xmath0 @xmath102 p phase shift @xmath94 , whereas the phase of @xmath96 will be given by the @xmath86 s phase shift @xmath79 . above the @xmath14 threshold , the elastic unitarity relation ( [ eq : ffuni ] )
has to be modified due to the important @xmath14 inelasticity .
this extended unitarity relation can be written as @xmath103 where @xmath104 is the properly normalized kaon vector or scalar form factor . at still higher energies , other important inelastic channels
must be included and the unitarity relation has to be generalized further . the chiral expansion of the pion vector and scalar form factors can be written as @xmath105 where @xmath106 is of @xmath107 . since the scalar form factor is normalized according to @xmath97 , one has for the leading order term @xmath108 for both form factors . regarding the one - loop correction @xmath109 ,
this has been determined by gasser and leutwyler both in su(2 ) chpt @xcite and in the su(3 ) case @xcite .
for the two - loop correction @xmath110 , this part has been evaluated by a dispersive analysis @xcite in the elastic su(2 ) approximation .
this dispersive approach is based upon the two - loop perturbative interpretation of the elastic unitarity relation eq .
( [ eq : ffuni ] ) @xmath111 .\end{aligned}\ ] ] these relations are satisfied for all energies above the @xmath0 threshold in the su(2 ) approximation .
therefore , in this case the full two - loop pion form factors may be expressed in terms of a dispersion relation as @xcite @xmath112 + { \rm re}t^{i(1)}_l(s')\right\ } ds'}{s'^3(s'-s - i\epsilon ) } .\ ] ] for the vector form factor , the subtraction constants @xmath113 and @xmath114 can be written as @xmath115 , \nonumber \\ c_v & = & \frac{1}{16\pi^2f^2_{\pi}}\left [ \frac{1}{60m^2_{\pi } } + \frac{\bar{f}_2}{16\pi^2f^2_{\pi}}\right ] , \end{aligned}\ ] ] whereas in the case of the scalar form factor , they may be expressed in the following way : @xmath116 , \nonumber \\ c_s & = & \frac{1}{16\pi^2f^2_{\pi}}\left [ \frac{19}{120m^2_{\pi } } + \frac{\bar{d}_2}{16\pi^2f^2_{\pi}}\right ] .\end{aligned}\ ] ] the subtraction constants are given in terms of the scale - independent one - loop low - energy constants @xmath117 and @xmath118 and the additional two - loop coupling constants @xmath119 , @xmath120 , @xmath121 , and @xmath122 . these coupling constants could in the future be expressed in terms of chiral logs and two - loop renormalized chiral low - energy constants . the dispersive analysis is not restricted to the elastic su(2 ) approximation , but may be generalized to the inelastic case as well .
the important @xmath14 inelasticity can be included in this approach by considering the two - loop perturbative extended unitarity relations @xmath123 \nonumber \\ & & + \sigma_{(k)}(s ) \left [ { \rm re}f^{(1)}_{(k)}(s)t^{i(0)}_{l(k)}(s ) + f^{(0)}_{(k)}(s){\rm re}t^{i(1)}_{l(k)}(s ) \right ] .\end{aligned}\ ] ] in the inelastic case , these expressions have to be used above the @xmath14 threshold in the dispersion relation eq .
( [ eq : cform ] ) .
the other inelastic channels may also be included in the dispersive approach , but these will only be important somewhat above the @xmath14 threshold . also for the form factors , the elastic approximation will be used in the derivation of the iam applied to two loops in the chiral expansion .
however , some short remarks on the possible inclusion of inelasticities will also be given in this case .
the starting point for the iam applied to the pion form factors is to write down a dispersion relation for the inverse of the form factor @xmath124 . in this dispersion relation , the elastic unitarity relation ( [ eq : ffuni ] ) gives @xmath125 .
unfortunately , this part can not be computed exactly , as was the case for @xmath0 scattering , but it may be expanded to two - loop order in the chiral expansion @xmath126 $ ] .
the subtraction constants can also be evaluated by expanding the function @xmath46 to two - loop order as @xmath127 .
thus , neglecting any pole contribution arising from possible zeros in the form factors @xcite , the following dispersion relation is obtained for the iam @xmath128 + { \rm re}t^{i(1)}_l(s')\right\ } ds'}{s'^3(s'-s - i\epsilon ) } .\ ] ] in order to simplify the connection between the iam and two - loop chpt , one can write down a dispersion relation for the function @xmath129 . since this is the function @xmath46 expanded to two - loop order in the chiral expansion , the subtraction constants in the dispersion relation for @xmath56 will be the same as in eq .
( [ eq : fdisp1 ] ) .
furthermore , the perturbative elastic unitarity relations ( [ eq : fpuni ] ) give @xmath130 $ ] , so the following dispersion relation is obtained for chpt @xmath131 + { \rm re}t^{i(1)}_l(s')\right\ } ds'}{s'^3(s'-s - i\epsilon ) } .\end{aligned}\ ] ] since this is identical to eq .
( [ eq : fdisp1 ] ) , the iam to two - loop order in the chiral expansion can be written as @xmath132 even though the unitarity cut in the iam for the pion form factors can not be computed exactly , it is still expected that this method is superior to the truncation of the chiral expansion .
this is based upon the fact that the expansion of @xmath133 used in the iam converges significantly faster than the corresponding expansion of @xmath134 used in chpt .
actually , even the one - loop expansion @xmath135 is a rather good approximation over a relatively large energy region @xcite . with this approximation and the @xmath0 partial waves given by eq .
( [ eq : pade2 ] ) , the iam to two loops will satisfy the elastic unitarity relation ( [ eq : ffuni ] ) exactly .
hence , the iam is a rather consistent way of imposing the unitarity restriction on the chiral expansion .
the iam was originally applied to the pion form factors in the one - loop approximation @xcite , which gives the result @xmath136 with the @xmath0 partial waves given by eq .
( [ eq : pade1 ] ) , this form will also satisfy the elastic unitarity relation ( [ eq : ffuni ] ) exactly , provided that the one - loop approximation @xmath135 is satisfied .
however , in this case @xmath133 and the subtraction constants are only expanded to one - loop order in the chiral expansion .
thus , even though the one - loop expansion of @xmath133 is a rather good approximation , it is clear that the iam to two loops will improve this approximation .
the first important inelastic channel is due to the @xmath14 intermediate state . in order to include
this inelastic channel in the iam , the extended unitarity relation ( [ eq : fefuni ] ) has to be used above the @xmath14 threshold .
this gives @xmath137 , which can be evaluated to two - loop order in the chiral expansion @xmath138 + { \rm re}t^{i(1)}_l(s)\right\ } \nonumber \\ & & -\sigma_{(k)}(s)\left\ { t^{i(0)}_{l(k)}(s)\left [ f^{(0)}_{(k)}(s ) -2f^{(0)}_{(k)}(s){\rm re}f^{(1)}(s)\right.\right .
\nonumber \\ & & \left.\left .
+ { \rm re}f^{(1)}_{(k)}(s)\right ] + f^{(0)}_{(k)}(s){\rm re}t^{i(1)}_{l(k)}(s)\right\ } .\end{aligned}\ ] ] this expression has to be used above the @xmath14 threshold in the dispersion relation ( [ eq : fdisp1 ] ) for the iam . in order to simplify the relation between the iam and chpt in this inelastic case , the perturbative extended unitarity relations ( [ eq : fepuni ] ) may be used to evaluate @xmath70 to two - loop order in the chiral expansion . since this will give an expression exactly similar to eq .
( [ eq : finelaex ] ) , the inelastic iam to two loops can also be written in the form ( [ eq : fpade2 ] ) with the @xmath14 inelasticity included in chpt .
in the same way , by expanding to one - loop order in the chiral expansion , the inelastic iam to one loop will be given by a form similar to eq .
( [ eq : fpade1 ] ) .
as was the case in the elastic approximation , it is also expected that the inelastic higher order corrections are substantially more suppressed for the iam than for chpt .
however , it still has to be investigated whether the iam is actually able to correctly describe the @xmath14 and other inelasticities .
the pion vector form factor is well known experimentally both in the time - like region @xmath139 @xcite and in the space - like region
@xmath140 @xmath141 @xcite . as for the scalar form factor
, this is not directly accessible to experiment .
however , it has been determined in terms of the experimental phase shifts for the @xmath142 system by a dispersive analysis @xcite . in the following
, the theoretical form factors will be limited to the elastic approximation when compared to the experimental information .
the vector form factor to one loop depends on the single scale - independent low - energy constant @xmath118 .
this can be fixed for both chpt and the iam from the experimental value of the pion charge radius @xmath143 @xmath144 @xcite . with the central experimental value ,
one obtains the results for @xmath118 displayed in the first and third column in table [ table3 ] , which is consistent with independent information on this low - energy constant @xcite . for two - loop chpt
, the value of @xmath118 should in principle be re - evaluated taking into account the two - loop corrections .
however , since a thorough re - evaluation seems rather out of reach at the moment , the one - loop value of @xmath118 is also used in two - loop chpt .
therefore , from the central experimental value of the pion charge radius @xcite , one obtains the result that the two - loop contribution to @xmath145 contained in the coupling constant @xmath119 vanishes . for the other coupling constant @xmath120 ,
this can be fixed from the value @xmath146
@xmath147 @xcite with a result displayed in the second column in table [ table3 ] . finally , for the renormalized low - energy constants @xmath24 , @xmath25 , @xmath26 , and @xmath27 occuring in the @xmath0 partial waves to one loop
, they will be given by the values shown in the second column in table [ table1 ] .
for the iam to two loops , the low - energy constant @xmath118 and the coupling constants @xmath119 and @xmath120 may be determined simultaneously by fitting the time - like experimental data below @xmath83 gev @xcite together with the space - like experimental data @xcite . for the other renormalized low - energy constants @xmath24 , @xmath25 , @xmath26 , and @xmath27 , the values shown in the fourth column in table [ table1 ] are used .
this gives a rather good fit with values of @xmath118 , @xmath119 , and @xmath120 displayed in the fourth column in table [ table3 ] .
for the the pion charge radius @xmath145 and the low - energy parameter @xmath148 also shown in table [ table3 ] , it is observed the the predictions for the iam to two loops agree remarkably well with the experimental information @xmath143 @xmath144 @xcite and @xmath146 @xmath147 @xcite .
in addition , with this method one finds that the two - loop correction to @xmath145 contained in the coupling constant @xmath119 is approximately 20% of the one - loop contribution , which is quite reasonable . from table
[ table3 ] , it is furthermore observed that the prediction for @xmath148 obtained from the iam to one loop also agrees remarkably well with the experimental information , whereas the corresponding prediction obtained from one - loop chpt is far too small .
the vector form factor @xmath149 is shown for @xmath150 @xmath151 @xmath141 in fig .
the iam to two loops describes the data very well both in the time - like and in the space - like regions .
in fact , also the space - like data for @xmath152 @xmath141 @xcite , which were not included in the fit , agree well with the iam to two loops .
of course , the small isospin violating @xmath153 mixing is not reproduced by this method , since the isospin limit has been applied throughout this paper .
in addition , this method should only be reliable below approximately @xmath154 @xmath141 , since the elastic approximation has be used for the vector form factor .
as for the iam to one loop , this agrees well with the space - like experimental data .
however , in the time - like region this method only approximates the @xmath72(770 ) resonance , as has been discussed in greater detail in ref .
nevertheless , it is obvious that the iam to both one and two loops significantly improves the behavior of chpt , which only accounts for the low - energy tail of the @xmath72(770 ) resonance .
the agreement between one and two - loop chpt and the space - like experimental data @xcite can be slightly improved by fitting these experimental data @xcite , but this will not change the range of applicability of chpt significantly . in fig .
[ fig7 ] , the predictions for the phase @xmath94 are shown below @xmath91 gev .
only the iam are shown since one and two - loop chpt will give the chiral @xmath0 i=1 p phase shifts to leading and one - loop order respectively .
for the iam to two loops , the phase agrees very well with the experimental data and gives a resonance at @xmath155 gev defined to be at @xmath78 .
furthermore , since this method also agrees well with the results of the iam for the @xmath0 i=1 p phase shifts shown in fig .
[ fig5 ] , the iam to two loops satisfies the elastic unitarity relation ( [ eq : ffuni ] ) very well indeed .
as for the iam to one loop , since this only approximates the experimental data , the elastic unitarity relation ( [ eq : ffuni ] ) will be somewhat violated in this case .
however , the iam to one loop is also quite consistent with a resonance structure @xcite and will give a resonance at @xmath156 gev .
thus , the iam to both one and two loops are rather consistent approaches to extend the range of applicability of chpt in the case of the vector form factor . the normalized scalar form factor to one loop is given in terms of the single scale - independent low - energy constant @xmath117 .
the same low - energy constant also occurs in @xmath0 scattering to one loop , where the central value @xmath157 obtained from the ratio @xmath158 @xcite was used . changing to scale - independent low - energy constants @xcite gives @xmath159 , which is used both in chpt and the iam to one loop .
the same value of @xmath117 is also used in two - loop chpt , whereas the coupling constants @xmath121 and @xmath122 are determined from the values @xmath160 @xmath144 and @xmath161 @xmath147 @xcite , respectively . for the additional low - energy constants to two loops
@xmath24 , @xmath25 , and @xmath26 , they will be given by the values shown in the second column in table [ table1 ] .
for the iam to two loops , the low - energy constant @xmath117 and the coupling constants @xmath121 and @xmath122 can not be determined in the same way as was applied for the vector form factor .
this is because the scalar form factor is not directly accessible to experiment .
however , in the elastic region the phase of the scalar form factor is given by the @xmath0 i=0 s phase shift , which is known experimentally .
therefore , the coupling constants @xmath121 and @xmath122 may be determined by fitting the experimental phase shift @xmath79 below @xmath83 gev @xcite , whereas @xmath117 can be fixed from the requirement @xmath160 @xmath144 . with the remaining low - energy constants @xmath24 , @xmath25 , and @xmath26 given by the values in the fourth column in table [ table1 ] , this gives an acceptable fit for the @xmath0 i=0 s phase shift . in table
[ table4 ] , the results for the low - energy constant @xmath117 and the coupling constants @xmath121 and @xmath122 are shown both for chpt and the iam . as for the iam to two loops , the value of @xmath117 is slightly different from the one - loop value @xmath159 .
this might well be due to the inclusion of the higher order corrections in the determination of this low - energy constant in the two - loop case .
however , the value of @xmath117 for the iam to two loops is still consistent with the value @xmath162 obtained from the ratio @xmath158 to one - loop order in chpt @xcite . in fact , using the value @xmath163 in the iam to two loops , the previous numerical analysis of @xmath0 scattering and the pion vector form factor will only change insignificantly .
the largest numerical difference will be for the @xmath0 threshold parameters , but the values given for the iam to two loops in table [ table2 ] will at most change by 5% . in the future it might be possible to determine the two - loop value of @xmath117 from independent observables .
the corresponding values of the pion scalar radius @xmath164 and the low - energy parameter @xmath165 are also given in table [ table4 ] .
for the pion scalar radius , the predictions obtained from chpt and the iam to one loop agree rather well with the result from the dispersive analysis @xmath160 @xmath144 ( solution b ) @xcite .
for the low - energy parameter @xmath165 , the predictions from the iam to one and two loops agree within 10 - 15% with the result @xmath161 @xmath147 ( solution b ) @xcite , whereas the corresponding prediction from one - loop chpt is too small .
the scalar form factor is known up to @xmath166 gev from the dispersive analysis in refs .
@xcite . in fig .
[ fig8 ] , solution b from this dispersive analysis is compared to the scalar form factor obtained from the iam and chpt .
it is observed that the iam to two loops agrees very well with the dispersive analysis over the whole energy region , whereas the same is only true for the iam to one loop below approximately 0.5 gev . as for two - loop chpt , this works rather well up to about 0.4 gev , while one - loop chpt only agrees with the dispersive analysis below the @xmath0 threshold .
this is due to the strong final - state interaction , which makes the higher order corrections important even at low energies . in fig .
[ fig9 ] , the phase @xmath79 obtained from the iam is shown below @xmath91 gev .
for the iam to two loops , it is observed that below approximately 0.9 gev , the phase agrees rather well with both the experimental data and the two - loop iam result for the @xmath0 i=0 s phase shift shown in fig .
thus , the iam to two loops satisfies the elastic unitarity relation ( [ eq : ffuni ] ) rather well . slightly above 0.9 gev
, the iam to two loops shows a very rapid phase variation of approximately @xmath167 , which might be a signal indicating that this method could reproduce the @xmath92(980 ) resonance .
however , in order to investigate this the @xmath14 inelasticity must be included in the iam . as for the iam to one loop
, this method only describes the data well at rather low energies .
the agreement could be improved at somewhat higher energies by reducing the value of the low - energy constant @xmath117 within the error bars @xmath162 , but this will at the same time give a value for the pion scalar radius less consistent with the dispersive result @xmath160 @xmath144 . finally , the phase difference @xmath89 between the scalar and vector form factors is shown in fig .
[ fig10 ] below @xmath168 gev .
the iam to two loops agrees very well with both the central experimental data and the results of the iam for the @xmath0 phase shift difference shown in fig .
the iam to one loop also agrees with the experimental data within the error bars , but is slightly different from the corresponding phase shift differences shown in fig .
as already pointed out , it is expected that forthcoming kaon facilities like da@xmath90ne can reduce the error bars on @xmath89 significantly @xcite , which will allow one to test the consistency of the iam at low energies in greater detail .
the iam is based upon the use of unitarity and dispersion relations together with the chiral expansion .
it has previously been shown that this method applied to one - loop order successfully extends the range of applicability of chpt .
however , the iam is not restricted to the one - loop approximation but can in general be applied to any given order in the chiral expansion . in the present paper , the extension of the iam to two loops has been analysed in the case of @xmath0 scattering and the pion form factors . the analysis has mainly been restricted to the elastic approximation but the possible extension to the inelastic case has also been discussed in some detail . from the derivation of the iam
applied to two loops in the chiral expansion , it is found that this approach indeed improves the iam to one loop both for @xmath0 scattering and in the case of the pion form factors . for @xmath0 scattering ,
a comparison with the experimental data shows that the iam applied to two loops is in fact rather consistent with the experimental data up to energies where inelasticities become essential . at somewhat lower energies ,
the two - loop approach compares well with the one - loop approximation , and in the threshold region they both agree with chpt . in the case of the pion form factors
, it is also found that the iam to two loops is consistent with the experimental information up to the first important inelastic threshold .
for the iam to one loop , although this also successfully extends the range of applicability of chpt , it only approximates the form factors at higher energies .
thus , the present analysis suggests that the iam is indeed a rather systematic way of extending the range of applicability of chpt .
however , in order to investigate this in greater detail , the iam to two loops should be applied to other processes . in this way
, some of the two - loop low - energy constants in the iam might be determined from independent observables .
there are also other ways of extending the range of applicability of chpt and work is still in progress in this subject .
however , at present the iam provides one of the most intriguing methods to improve chpt in a systematic manner .
the author thanks a. miranda and g. c. oades for discussions and comments , and j. bijnens for a private communication regarding the two - loop low - energy constants estimated from resonance exchange . the support from the faculty of science , aarhus university is also acknowledged .
s. weinberg , physica * 96a * , 327 ( 1979 ) .
j. gasser and h. leutwyler , ann .
* 158 * , 142 ( 1984 ) ; nucl .
b250 * , 465 ( 1985 ) .
h. leutwyler , ann .
( n.y . ) * 235 * , 165 ( 1994 ) .
j. bijnens , g. colangelo , and j. gasser , nucl .
* b427 * , 427 ( 1994 ) . h. w. fearing and s. scherer , phys .
d * 53 * , 315 ( 1996 ) .
t. n. truong , phys .
lett . * 61 * , 2526 ( 1988 ) .
a. dobado , m. j. herrero , and t. n. truong , phys .
b * 235 * , 134 ( 1990 ) ;
t. n. truong , phys . rev .
lett . * 67 * , 2260 ( 1991 ) ; a. dobado and j. r. pelez , phys .
d * 47 * , 4883 ( 1993 ) ; z. phys .
c * 57 * , 501 ( 1993 ) .
t. hannah , phys .
d * 51 * , 103 ( 1995 ) ; * 52 * , 4971 ( 1995 ) .
t. hannah , phys .
d * 54 * , 4648 ( 1996 ) .
a. dobado and j. r. pelez , `` the inverse amplitude method in chiral perturbation theory , '' report no .
hep - ph/9604416 .
m. knecht _
et al . _ ,
b457 * , 513 ( 1995 ) .
j. bijnens _ et al .
b * 374 * , 210 ( 1996 ) .
j. gasser and u. g. meissner , nucl .
b357 * , 90 ( 1991 ) .
m. alston - garnjost _ et al .
* 36b * , 152 ( 1971 ) . s. weinberg , phys . rev
* 17 * , 616 ( 1966 ) .
v. bernard , n. kaiser , and u. g. meissner , nucl . phys .
* b357 * , 129 ( 1991 ) . g. ecker _
et al . _ ,
nucl . phys . *
b321 * , 311 ( 1989 ) ; g. ecker _
et al . _ ,
b * 223 * , 425 ( 1989 ) .
j. bijnens , ( private communication ) .
l. rosselet _
et al . _ ,
d * 15 * , 574 ( 1977 ) .
s. d. protopopescu _ et al .
d * 7 * , 1279 ( 1973 ) .
b. hyams _ et al .
b64 * , 134 ( 1973 ) ; w. ochs , thesis , ludwig - maximilians - universitt , 1973 ( unpublished ) .
p. estabrooks and a. d. martin , nucl .
b79 * , 301 ( 1974 ) .
m. j. losty _
et al . _ ,
b69 * , 185 ( 1974 ) .
w. hoogland _
et al . _ ,
b126 * , 109 ( 1977 ) .
d. morgan and m. r. pennington , in _ the second da@xmath90ne physics handbook _ , edited by l. maiani , g. pancheri , and n. paver ( infn , frascati , 1995 ) , vol . 1
, p. 193 . m. m. nagels _ et al .
* b147 * , 189 ( 1979 ) . b. ananthanarayan , d. toublan , and g. wanders , phys .
d * 53 * , 2362 ( 1996 ) .
et al . _ , `` lifetime measurement of @xmath88 atoms to test low - energy qcd predictions , '' report no .
cern / spslc 95 - 1 .
j. gasser and u. g. meissner , phys .
b * 258 * , 219 ( 1991 ) . m. baillargeon and p. j. franzini , in _ the second da@xmath90ne physics handbook _ , edited by l. maiani , g. pancheri , and n. paver ( infn , frascati , 1995 ) , vol . 1 ,
particle data group , r. m. barnett _
et al . _ ,
d * 54 * , 1 ( 1996 ) .
k. m. watson , phys . rev . *
95 * , 228 ( 1954 ) .
j. gasser and h. leutwyler , nucl .
b250 * , 517 ( 1985 )
. l. m. barkov _
et al . _ ,
nucl . phys .
* b256 * , 365 ( 1985 ) ; l. m. kurdadze _
et al . _ , yad . fiz . *
40 * , 451 ( 1984 ) [ sov .
. phys . * 40 * , 286 ( 1984 ) ] ; g. v. anikin _
_ , report no .
inp 83 - 85 ( unpublished ) ; s. r. amendolia _ et al .
_ , phys . lett . *
138b * , 454 ( 1984 ) ; i. b. vasserman _
_ , yad . fiz .
* 33 * , 709 ( 1981 ) [ sov
. phys . * 33 * , 368 ( 1981 ) ] ; a. quenzer _
et al . _ ,
phys . lett . *
76b * , 512 ( 1978 ) .
na7 collaboration , s. r. amendolia _ et al .
_ , nucl . phys . *
b277 * , 168 ( 1986 ) . c. j. bebek
_ et al . _ ,
d * 17 * , 1693 ( 1978 ) .
j. f. donoghue , j. gasser , and h. leutwyler , nucl .
b343 * , 341 ( 1990 ) .
j. bijnens and f. cornet , nucl .
b296 * , 557 ( 1988 ) ; g. colangelo , m. finkemeier , and r. urech , phys .
d * 54 * , 4403 ( 1996 ) . l. v. dung and t. n. truong , `` equivalence between vector meson dominance and unitarised chiral perturbation theory , ''
report no .
hep - ph/9607378 .
.values of the low - energy constants at the @xmath72(770 ) scale used for one - loop chpt ( chpt1 ) , two - loop chpt ( chpt2 ) , the iam to one loop ( iam1 ) , and the iam to two loops ( iam2 ) . for a discussion on how these low - energy constants have been determined
see the text . [ cols="^,^,^,^,^ " , ] | the inverse amplitude method is analysed to two - loop order in the chiral expansion in the case of @xmath0 scattering and the pion form factors .
the analysis is mainly restricted to the elastic approximation but the possible extension to the inelastic case is also discussed in some detail .
it is shown how the two - loop approach improves the inverse amplitude method applied to one - loop order in the chiral expansion . for both @xmath0 scattering and the pion form factors , it is in fact found that the inverse amplitude method to two - loop order agrees remarkable well with the experimental data up to energies where inelasticities become essential . at somewhat lower energies ,
the two - loop approach compares well with the one - loop approximation , and in the threshold region they both agree with chiral perturbation theory .
this suggests that the inverse amplitude method is indeed a rather systematic way of improving chiral perturbation theory order by order in the chiral expansion . |
percolation theory is one of the simplest models exhibiting a phase transition @xcite . due to its simplicity
, it provides fundamental insights that facilitate the understanding of many phenomena across biology , physics , and geophysics @xcite , and is recently experiencing a revival of interest @xcite . in this paper
we use the theory of percolation to study the properties of a random spatial network model inspired by axonal guidance in neuronal networks .
the central nervous system is an intricate , disordered network composed of tiny processing units , the neurons , connected to each other in a complex manner @xcite .
neurons themselves have a rather complicated dynamics @xcite and are composed mainly of three parts : _ dendrites _ , which provide the input signals , a central body called _ soma _ , and the _ axon _ , which transmits the output and can reach up to one meter in length @xcite .
most models of neuronal dynamics @xcite agree that the topology of the connections between neurons plays an important role . as a consequence ,
several possible network structures have been investigated , ranging from fully connected graphs @xcite with homogeneous or random couplings @xcite to graphs in which the connectivity depends on the distance between neurons @xcite .
the development of the mammal brain is clearly very complex , starting from the neural plate and following several steps .
it is characterized by an initial exponential growth driven by neuronal migration , differentiation , and axonal guidance @xcite .
after its formation , in the years of maturity , the number of neural connections in all areas of the brain dramatically decreases , a process known as pruning @xcite . in this paper we abstract from such complexity and study a simple , stylized model to describe how the formation of a giant cluster of neurons depends on a few parameters that in principle could be measured experimentally .
the mechanism we propose for the formation of a bond between two neurons takes two effects into account .
first , in the neocortex the majority of connections between neurons are short - range ( with only few connections that make the network small - world @xcite ) , so it is reasonable , at least as a first approximation , to connect only neurons that are close to each other .
second , since the axon is much longer than the size of the body of the neuron , it is clear that the direction of the axon must play an important role @xcite .
therefore , we allow a neuron to be connected only with other neurons that fall within a certain angle around the direction of its axon . our simplified characterization of a neuron is shown in figure [ fig : example ] .
neurons are represented as disks equipped with a spin : the radius of the disks accounts for the finite size of neurons , while the spin for the directionality of the axon .
on one hand , the proposed model falls in the category of a continuum ( two - dimensional ) disk percolation model , which has been studied in a variety of contexts and also for different shapes as an alternative to disks @xcite .
however , an important difference is that we account for the presence of directionality in the formation of bonds ( due to the direction of the axon ) , meaning that the underlying network is directed . on the other hand , models of directed percolation
have been previously investigated both on a lattice to explain the flow in porous materials @xcite and on complex networks @xcite . in our case disks
are placed in a euclidean space , which is in stark contrast both with the former , characterized by the regular geometry of a lattice , and with the latter , in which there is no embedding in a metric space .
therefore , our model can be put at the intersection between continuum percolation , directed percolation , and random networks , and studying its behavior is interesting in itself .
the presence of a giant percolating cluster is particularly relevant in view of the aforementioned pruning process , which can be depicted as a removal of bonds between neurons .
since the underlying network is directed , one could in principle choose different definitions for the giant , percolating cluster of disks .
we focus on the emergence of a _ basin of influence _ in which every neuron belonging to the giant cluster can propagate a signal , i.e. information within the basin .
therefore , it represents the set of neurons which are mutually causally connected and , incidentally , it mirrors the definition of _ fully recurrent _ neural networks .
c + such property is also important in the context of artificial neural networks @xcite .
the recent success of deep learning algorithms @xcite suggests that a layered feedforward ( i.e. acyclic ) structure inspired by the topological structure of realistic neuronal cell networks in the visual cortex @xcite can significantly improve the performance of artificial neural networks .
thus , one might argue that the presence of a hierarchical structure is key for performing fast learning artificial tasks . from this perspective , the existence of giant strongly connected cluster is detrimental , as it rules out the presence of a layered feedforward structure .
the paper is organized as follows .
we first introduce the model and the relevant definitions in section [ sec : model ] . in section [ sec : analysis ] we analyze the model numerically and explore their universality classes by means of finite - size scaling .
conclusions follow in section [ sec : conclu ] .
we consider a simple , stylized model to describe the mechanism governing the formation of a giant cluster of neurons .
we represent a neuron as a disk with a given radius .
the region inside the disk encloses the soma , the dendrites , and the axon . to each disk
we also associate a direction ( a unit vector which we call _ spin _ ) that models the spatial orientation of the axon .
we then choose a simple set of rules to decide if two disks belong to the same cluster .
we start with implementing a proximity constraint , meaning that two disk can be part of the same cluster only if they overlap .
however , even if two neurons are close , they might not connect , because of the orientations of their axons .
therefore we add the following directional constraint : we form the link @xmath0 between two overlapping disks @xmath1 and @xmath2 if the center of disk @xmath2 is in the half - plane spanned by an angle @xmath3 around the direction of disk @xmath1 .
intuitively , it means that looking in the direction of disk @xmath1 we can detect the center of disk @xmath2 within an angle @xmath3 , i.e. in a limited range of angles in front of us .
we can see that such prescription is not symmetric : the existence of the link @xmath0 does imply the existence of the reverse link @xmath4 .
the directionality in the process that leads to the creation of the network is intended to mirror the physiological directionality of signals in neuronal networks , which are transmitted from the soma through the axon .
such problem naturally falls within the framework of random geometric graphs .
more formally , we consider a two - dimensional square box whose sides have length @xmath5 in which we drop @xmath6 disks at random locations .
we denote with @xmath7 the spatial coordinate of disk @xmath1 and with the unit vector @xmath8 its spin .
all disks have radius @xmath9 .
we now build a directed graph associating to each disk a node in the graph and using the aforementioned rules to draw edges between nodes . denoting with @xmath10 the element @xmath11 of the adjacency matrix of the graph
, we can write both the proximity and the directional constraint in the following compact form : @xmath12 where @xmath13 is the heaviside function , i.e. @xmath14 for @xmath15 , and @xmath16 otherwise , and @xmath17 is the vector square root of 2-norm
. we can see that the first @xmath18 implements the proximity constraint : it is different from zero only if the vector @xmath19 ( which connects the center of disk @xmath1 with the center of disk @xmath2 ) is shorter than @xmath20 , the diameter of a disk , i.e. only if the two disks overlap .
the second @xmath18 implements the directional constraint . to see it , it is sufficient to note that the scalar product @xmath21 is simply equal to the cosine of the angle between the spin @xmath8 and the unit vector connecting the center of disk @xmath1 with the center of disk @xmath2 . therefore , this term is larger than zero precisely when such angle is in the interval @xmath22 $ ] .
let us now mention a crucial result in the context of random graphs : let us scatter @xmath6 balls of radius @xmath23 in an hypercube of @xmath24 dimensions and linear size 1 , and let us link two balls if they overlap .
it has been proven @xcite that , for @xmath25 and in the limit @xmath26 , there exists @xmath27 such that for @xmath28 almost surely an infinite cluster of balls appears , while for @xmath29 almost surely it does not , a behavior that closely resembles a percolation transition . since our model generalizes the aforementioned mechanism by taking into account the spin direction , it is reasonable to expect a similar transition to occur .
therefore , we adopt the same scaling @xmath30 for the radius of disks and we define @xmath31 as the normalized radius of disks at which an infinite cluster of disks appears in the limit @xmath26 .
we also introduce @xmath32 , the average cluster size for a given value of @xmath33 ( excluding the largest `` infinite '' cluster @xcite ) , @xmath34 , the probability that an arbitrary disk belongs to the infinite cluster for a given value of @xmath33 , and @xmath35 the correlation length is defined from @xmath36 , where @xmath37 is the correlation function , i.e. the probability that a point identified by the vector @xmath38 applied to a point in a finite cluster belongs to the same finite cluster . ] .
close to the critical point @xmath31 , such quantities are expected to behave as power laws in the variable @xmath39 : @xmath40 finally , let us point out that rescaling the radius ( @xmath30 ) and keeping the linear size of the box fixed ( @xmath41 ) is equivalent to keeping the radius constant ( @xmath42 ) and rescaling the box ( @xmath43 ) , which is more intuitive when dealing with quantities such as the correlation length .
it is worth to remark that in our case edges have a direction . on a directed graph
there are several possible definitions of clusters , corresponding to different kinds of connectedness .
following @xcite , we identify clusters with _ strongly connected subgraphs_. more precisely , given a _ directed _ graph @xmath44 , being @xmath45 the set of nodes of cardinality @xmath6 and being @xmath46 the set of directed edges with maximal cardinality @xmath47 , a strongly connected subgraph @xmath48 of @xmath49 is a collection of nodes and directed edges of @xmath49 , such that there exists a directed path between each pair of vertices in @xmath48 ( i.e. in both directions ) .
a _ maximally _ strongly connected subgraph is the maximal subgraph that is strongly connected .
maximally strongly connected subgraphs will serve as our definition of percolating clusters .
so far we have not specified any distribution for the direction of spins , which is defined by the angle @xmath50 that @xmath8 forms with the horizontal side of the box .
we choose a general approach in which a single parameter allows us to interpolate between the case in which all spins are aligned along a privileged direction and the case in which the angles @xmath50 are uniformly distributed in in the interval @xmath51 $ ] .
the rationale of such choice is to provide a unified framework encompassing both feed - forward networks ( resembling the visual cortex @xcite ) , in which connections have a strong directionality , and recurrent networks @xcite , in which cycles appear frequently . without loss of generality , we choose a privileged direction defined by the unit vector @xmath52 parallel to the horizontal side of the box and we fix such direction _ on average _ , without assuming any further knowledge on the probability distribution of @xmath53
. hence @xmath53 will be the maximum entropy distribution compatible with @xmath54 being fixed on average : @xmath55 @xmath56 being the lagrange multiplier controlling the fluctuations around the chosen direction and @xmath57 being a normalization factor , which is readily determined : @xmath58 where @xmath59 is the modified bessel function of the first kind . in figure
[ fig : spin_dis ] we show @xmath53 for different values of @xmath56 . for small values of @xmath56
the distribution is concentrated around @xmath60 , the privileged direction . as @xmath56 increases , fluctuations around such direction increase ( @xmath61 is an increasing function of @xmath56 ) , and for @xmath62 the uniform distribution is recovered .
a realization of random positions and orientations of 100 disks ( disks boundaries are not shown to enhance readability ) is shown in figure [ fig : disex ] . , the angle that spins form with the x axes for different values of @xmath56 , the temperature controlling the fluctuations around such direction . ] . ]
we perform extensive numerical simulations to characterize the percolation transition . we compute numerical estimators of the strength of the infinite cluster @xmath34 and of the average cluster size @xmath32 .
@xmath34 is estimated simply as the fraction of disks in the largest cluster , while @xmath32 is estimated as the average of all cluster sizes , after the removal of the largest one .
we generate multiple realizations of the random locations of disks within the box , and by averaging both quantities over such realizations we define @xmath63 and @xmath64 .
we use open boundary conditions .
periodic boundary conditions would not simplify our numerical analysis , but could introduce a bias towards larger sccs , because of the edges formed across the boundary .
such bias would be especially strong when the process of the formation of edges is highly directional .
in fact , either for @xmath65 or for @xmath66 the cycles needed to form a scc can be only formed through the boundary .
let us focus initially on single values of @xmath56 and @xmath3 , which we set respectively equal to one and @xmath67 . in figure
[ fig : largest_and_average ] we plot @xmath68 and @xmath69 as functions of @xmath33 , the normalized radius of disks .
we see that for small values of @xmath33 , @xmath68 is very close to zero , while for large values of @xmath33 , @xmath68 approaches one , signaling the presence of a giant strongly connected component ( gscc ) .
the transition between the regimes becomes sharper and shaper as the size of the system increases . as regards @xmath69 , for small values of @xmath33
the average cluster size is slightly above one , showing that all disks form isolated clusters . as @xmath33 increases
, @xmath70 reaches the peak value @xmath71 and then decreases .
the reason is that the average cluster size is computed by excluding the largest cluster , whose size diverges in the limit @xmath26 . in a system of finite size , the appearance of the gscc has the effect of reducing the size of the other clusters , effectively leading to a decrease of @xmath70 . as a consequence ,
the position @xmath72 of the peak @xmath71 is a good estimator of the critical value of @xmath33 for a system of finite size . ,
fraction of disks in the largest cluster ( _ left _ ) and @xmath70 , average cluster size ( _ right _ ) as a function of @xmath33 , the normalized radius of disks .
every point is averaged over 100 realizations , @xmath73 , and @xmath74 .
@xmath75 goes from zero to one as @xmath33 increases and the transition becomes sharper and sharper as the number of disks @xmath6 increases .
the peak @xmath71 of the average cluster size corresponds to the appearance of a gscc ( see main text).,title="fig : " ] , fraction of disks in the largest cluster ( _ left _ ) and @xmath70 , average cluster size ( _ right _ ) as a function of @xmath33 , the normalized radius of disks . every point is averaged over 100 realizations , @xmath73 , and @xmath74 .
@xmath75 goes from zero to one as @xmath33 increases and the transition becomes sharper and sharper as the number of disks @xmath6 increases .
the peak @xmath71 of the average cluster size corresponds to the appearance of a gscc ( see main text).,title="fig : " ] in order to estimate the critical value of @xmath33 we use the binder cumulant method @xcite .
the idea is to build a quantity that _ at the critical point _
does not depend on the size of the system .
the important observation is that in system of finite size at the critical point the correlation length @xmath35 is approximately the linear size of the system @xmath76 , from which it follows that @xmath77 , @xmath78 , and @xmath79 .
if we now build the quantity @xmath80 we have that @xmath81 . using the hyperscaling relation @xmath82
, we can see that @xmath83 does not depend on @xmath84 .
therefore , if we plot @xmath85 vs @xmath33 for different sizes of the system
all curves will cross at @xmath31 . for illustrative purposes we focus again on @xmath73 and @xmath74 and , from the left panel of figure [ fig : binder_and_gammanu ]
we see that the curves corresponding to different sizes of system cross each other approximately at @xmath86 . in order to estimate the ratio @xmath87 we use finite - size scaling .
let us recall that @xmath88 , hence diverging as @xmath89 .
for a system of finite size , the value @xmath71 corresponding to the peak in a plot of @xmath69 vs @xmath33 can be used as an estimator of @xmath90 .
hence , a log - log plot of @xmath71 vs @xmath6 should yield points disposed on a straight line with slope equal to @xmath91 .
the error on the ratio @xmath87 can be computed simply by propagating the error on the linear fit . from the right panel of figure
[ fig : binder_and_gammanu ] we can see that for @xmath92 and @xmath93 the linear fit is satisfactory , resulting in the following estimate of the ratio : @xmath94 , with @xmath95 .
, binder cumulant with respect to @xmath33 , the normalized radius of disks , for different values of @xmath6 , the number of disks .
every point is averaged over 100 realizations , @xmath73 , and @xmath74 .
all curves intersect approximately at @xmath96 , which is therefore the estimate for the critical point . _
right : _ finite - size scaling to estimate the ratio @xmath87 for @xmath73 and @xmath74 .
log - log plot of @xmath71 , the peak value of the average cluster size with respect to @xmath6 .
the linear fit is good , yielding @xmath94 , with @xmath95.,title="fig : " ] , binder cumulant with respect to @xmath33 , the normalized radius of disks , for different values of @xmath6 , the number of disks .
every point is averaged over 100 realizations , @xmath73 , and @xmath74 .
all curves intersect approximately at @xmath96 , which is therefore the estimate for the critical point . _
right : _ finite - size scaling to estimate the ratio @xmath87 for @xmath73 and @xmath74
. log - log plot of @xmath71 , the peak value of the average cluster size with respect to @xmath6 .
the linear fit is good , yielding @xmath94 , with @xmath95.,title="fig : " ] the aforementioned analysis can be repeated for a range of values of @xmath97 , allowing us to draw a phase diagram using @xmath33 and @xmath97 as parameters . from the left panel of figure [ fig : phase_and_flow ]
we see that @xmath31 increases as @xmath97 increases . in the region above the curve there is a gscc component , while in the region below the curve there is no cluster
whose size is of order @xmath98 .
the quantity @xmath97 is the inverse temperature measuring the fluctuations around the average direction of spins ( which is the x axis ) .
as @xmath97 increases , fluctuations are suppressed and most spin will tend to be aligned .
hence , disks with a larger radius will be needed to form a gscc .
in general , we expect this to be true for the following reason . at @xmath99 ,
it is easy to see that the spins are all oriented in the same direction with probability one .
this implies that a disk can be connected only with disks that are `` in front of '' it , within the range of angles @xmath22 $ ] , along the direction of the external field , meaning that whenever a path from disk @xmath1 to disk @xmath2 exists , there will be no path from disk @xmath2 to disk @xmath1 .
hence , the maximal strongly connected component will have size equal to one .
for @xmath56 small but strictly positive , the probability of having directed paths going back is small but non - zero , implying that one needs a large number of disks , or a large radius , to create paths going back , so that the maximal strongly connected component has a size comparable with the total number of disks . .
@xmath31 are estimated with the method of binder cumulant , averaging every @xmath33 point over 100 realizations and using @xmath100 .
the region above the curve corresponds to the phase in which a gscc appears .
the error on @xmath31 is the interval separating @xmath31 from the values of @xmath33 on its left and right . _
right : _ flow of the ratio of critical exponents @xmath87 .
errors bars span @xmath101 confidence intervals and are computed as errors on linear fits as the one in the right panel of figure [ fig : binder_and_gammanu].,title="fig : " ] .
@xmath31 are estimated with the method of binder cumulant , averaging every @xmath33 point over 100 realizations and using @xmath100 .
the region above the curve corresponds to the phase in which a gscc appears .
the error on @xmath31 is the interval separating @xmath31 from the values of @xmath33 on its left and right . _
right : _ flow of the ratio of critical exponents @xmath87 .
errors bars span @xmath101 confidence intervals and are computed as errors on linear fits as the one in the right panel of figure [ fig : binder_and_gammanu].,title="fig : " ] in the right panel of figure [ fig : phase_and_flow ] we plot the flow of the ratio @xmath87 with respect to @xmath97 , for two values of @xmath3 .
interestingly , we see that the ratio @xmath87 is not constant with respect to @xmath97 , meaning that at least one of the critical exponents is changing with the temperature .
an intuitive motivation of such behavior is the following : @xmath56 changes the degree of anisotropy of the model by interpolating between a model in which links are formed only along a preferred direction ( @xmath102 ) and a model in which no such direction exists ( @xmath103 ) . for @xmath103
the probability of drawing a given network realization is invariant under spatial rotations .
on the contrary , for @xmath102 this is not the case , as the probability of drawing any configuration with spins sensibly deviating from the preferred direction quickly goes to zero .
hence , by changing @xmath56 and therefore by gradually breaking a symmetry of the system , we expect the critical behavior of the model to be radically different .
indeed this is what we observe in the right panel of figure [ fig : phase_and_flow ] , in which the ratio @xmath104 has different behaviors for small and large values of @xmath56 .
finally , we can see from figure [ fig : phase_and_flow ] that a smaller value of @xmath3 implies that @xmath31 becomes larger , as it is more more difficult to establish edges in the network .
interestingly , both for small and for values of @xmath56 , the behavior of @xmath104 seems not to depend on @xmath3 , while in the intermediate region a statistically significant discrepancy can be observed . estimating the values of the critical exponents is possible , at least in principle .
in fact , once the ratios are known , the knowledge of one exponent , i.e. @xmath105 is sufficient to compute all of them .
@xmath105 is usually also estimated by means of finite - size scaling . using the scaling relation @xcite @xmath106 , where @xmath107 is a scaling function
, we see that at the peak @xmath71 , @xmath108 must be equal to some constant @xmath109 , meaning that : @xmath110 , where we remind that @xmath72 is the value of @xmath33 corresponding to @xmath71 .
however , in our simulations the error on @xmath72 turns out to be quite large , being equal to the resolution with which values of @xmath33 are scanned in the numerical simulations . with the present resolution , the relative error on @xmath105 is easily around @xmath111 . increasing the resolution leads to noisy estimates of @xmath72 , which are likely to improve only using a larger number of realizations over which each data point is averaged . in summary
, more extensive simulations would be needed for a precise estimation of the critical exponents . strictly speaking
, this difficulty prevents a direct verification of the hyperscaling relation .
however , the binder cumulant evaluated at the critical point does not depend on the system size if and only if the hyperscaling relation holds , which is consistent with the crossing we observe in the left panel of figure [ fig : binder_and_gammanu ] .
in this paper we studied the emergence of a giant strongly connected component in a model of overlapping disks endowed with spin representing the direction along which connections between the centers of disks are formed .
the model is inspired by the connectivity properties of neuronal networks , in which axons prolongate to connect to other neurons dendrites .
the motivation for studying the strongly connected component , rather than the connected one , comes from the interest in the subset of neurons that mutually influence each others . in the terminology of machine learning ,
the strongly connected giant component represents a fully recurrent neural network . our results ,
albeit restricted to the two - dimensional case , point towards a few interesting facts .
first , we find that for a large range of temperatures there exist a critical percolation threshold which we were able to obtain numerically , providing a phase diagram of the model .
we note that , while in the case of the undirected networks the density of connections is sufficient to determine whether one is in the percolating phase or not , in the case of directed network one needs to measure not only the density of connections , but also the average orientation of the axons .
in addition , we provide strong evidence that the critical exponents change with the temperature , i.e. with the parameter controlling the fluctuations around a privileged direction for spins . as a consequence , rather than a single universality class , as we vary the temperature we span a family of universality classes .
this feature has been already observed in other models @xcite .
we believe our simplified model could be of interest in the context of the pruning process taking place in the mature brain , which can be interpreted as a reverse percolation process in which connections are indeed removed .
in fact , by sufficiently decreasing the density it is possible to cross the critical line and transition to the phase in which there is not a giant strongly connected component .
further investigations are required to understand whether the observed behavior occurs also in more realistic settings , as a three - dimensional spatial embedding and the presence of long - range connections between neurons .
the first case , also due to the scaling of the number of disks implied by the penrose theorem @xcite , is much more computationally intensive . in the second case , one could follow the approach of small - world networks in which connections are randomly re - wired with a certain ( small ) probability . however , one should carefully account for the role played by directionality , which could be disrupted by such re - wirings .
intuitively , one would expect that the presence of long - range connections could facilitate the emergence of a gscc . finally , as already explained in section [ sec : analysis ] , a precise estimations of all critical exponents also requires considerable computational resources .
we plan to characterize the behavior of all such extensions in future studies .
f. caravelli would like to thank leo kadanoff for various useful comments during this work at the alumni conference at perimeter institute , and the london institute of mathematical sciences for hospitality .
we also would like to thank rory finnegan and luca dallasta for useful comments on the presentation of the results .
the work of f. caravelli was supported by invenia labs .
m. bardoscia acknowledges support from : fet ip project multiplex ( 317532 ) .
f. caccioli acknowledges support of the economic and social research council ( esrc ) in funding the systemic risk centre ( es / k002309/1 ) .
99 shante v k s , kirkpatrick s , 1971 _ adv .
phys . _ * 20 * 85 stauffer d , aharony a , 1994 _ introduction to percolation theory _ ( taylor & francis , london ) berkowitz b , ewing r p , 1998 _ surv
. geophys . _ * 19*(1 ) 23 balberg i , 1987 _ philos . mag .
b _ * 56*(6 ) 991 nagler j , levina a , timme t , 2010 _ nat .
phys . _ * 7 * 265 da costa r a , dorogovtsev s n , goltsev a v , mendes j f f , 2010 _ phys
lett . _ * 105 * 255701 noh j d , park h , 2004 _ phys .
e _ * 69 * 016122 schrenk k j , araujo n a m , andrade j s , herrmann h j , 2012 _ sci .
* 2 * 348 cho y s , hwang s , herrmann h j , kahng b , 2013 _ science _ * 339 * 1185 andrade r f , herrmann h j , 2015 arxiv:1507.07614 saberi a a , 2015 _ phys . rep . _ * 578 * 1 markram h et al , 2015 _ cell _ * 163*(2 ) 456 gerstner w , kistler w , naud r , paniski l , 2014 _ neuronal dynamics _ ( cambridge university press , cambridge ) abbott
l f , 1999 _ brain res .
bull . _ * 50*(5 - 6 ) 303 hodgkin a l , huxley a f , 1952 _ j. physiol .
_ * 117*(4 ) 500 fitzhugh r , 1955 _ bull .
biophys . _
* 17 * 257 nagumo j , arimoto s , yoshizawa s , 1962 _ proc
. ire _ * 50 * 2061 jaeger h , 2002 ( _ advances in neural information processing systems _ vol 15 ) , ed s becker , s thrun , k obermayer ( mit press , cambridge , usa ) maass w , natschlaeger t , markram h , 2002 _ neural comput . _ * 14*(11 ) 2531 kandel e r , schwartz j h , jessell t , 2012 _ principles of neural science _
( mcgraw - hill , new york ) angelucci a , bressloff p c , 2006 _ prog .
brain res .
_ * 154 * 93 stiles j , jernigan t l , 2010 _ neuropsychol .
rev . _ * 20*(4 ) 327 navlakha s , barth al , bar - joseph z , 2015 _ plos comput .
biol . _ * 11*(7 ) e1004347 lefort s , tomm c , sarria j c f , petersen c c , 2009 _ neuron _ * 61*(2 ) 301 franze k , guck j , 2010 _ rep .
phys . _ * 73 * 094601 gawlinski e t , stanley h e , 1981 _ j. phys . a : math .
gen . _ * 14*(8 ) l291 garboczi e j , snyder k a , douglas f , thorpe m f , 1995 _ phys .
e. _ * 52 * 1 quintanilla j , torquato s , ziff r m , 2000 _ j. phys . a : math .
_ * 33*(42 ) l399 sasidevan v , 2013 _ phys .
e _ * 88 * 022140 hinrichsen h , 2000 _ braz . j. phys . _ * 30*(1 ) 69 cardy j l ,
sugar r l , 1980 _ j. phys . a : math .
_ * 13 * l423 takeuchi k a , kuroda m , chate h , sano m , 2007 _ phys .
* 234503 hinrichsen h , 2000 _ adv . phys . _ * 49*(7 ) 815 schwartz n , cohen r , ben - avraham d , barabasi a - l , havlin s , 2002 _ phys .
e. _ * 66 * 015104 bogu m , ngeles serrano m , 2005 _ phys . rev .
e _ * 72 * 016106 agliari e , cioli c , guadagnini e , 2011 _ phys . rev
. e _ * 84 * 031120 hbert - dufresne l , allard a , young j - g , dub l j , 2013 _ phys . rev . e _ * 88 * 062820 restrepo j g , ott e , hunt b r , 2008 _ phys
lett . _ * 100 * 058701 hopfield j j , 1982 _ proc .
sci . _ * 79 * 2554 hinton g e , osindero s , teh y - w , 2006 _ neural comput .
_ * 18 * -7 1527 hartley t , lever c , burgess n , okeefe j , 2014 _ phil .
b _ * 369*(1635 ) 20120510 watts d j , strogatz s h , 1997 _ nature _ * 393 * 440 .
penrose m , 2003 _ random geometric graphs _ ( oxford university press , oxford ) rintoul d , torquato s , 1997 _ j. phys .
gen . _ * 30*(16 ) l585 newman m e j , strogatz s h , watts d j , 2001 _ phys .
e _ * 64 * 026118 lamme v a , super h , spekreijse h , 1998 _ curr .
neurobiol . _ * 8*(4 ) 529 sporns o , 2010 _ networks of the brain _
( mit press , cambridge , usa ) janoos f , li w , subrahmanya n , morocz i , wells w , 2012 ( _ advances in neural information processing systems _ vol 25 ) , ed f pereira , c j c burges , l bottou , k q weinberger ( curran associates , red hook , usa ) . galassi m et al , 2015 gnu scientific library reference manual , version 2.1 , https://www.gnu.org/s/gsl/manual/gsl-ref.pdf siek j g , lee l - q , lumsdaine a , 2001 _ the boost graph library : user guide and reference manual _ ( addison - wesley , boston ) tarjan r e , 1972 _ siam j. comput .
_ * 1*(2 ) 146 binder k , 1981 zeitschrift fr physik b * 43*(2 ) 119 | we propose a continuum model of percolation in two dimensions for overlapping disks with spin . in this model the existence of bonds is determined by the distance between the centers of the disks , and by the scalar product of the ( randomly ) directed spin with the direction of the vector connecting the centers of neighboring disks .
the direction of a single spin is controlled by a `` temperature '' , representing the amount of polarization of the spins in the direction of an external field .
our model is inspired by biological neuronal networks and aims to characterize their topological properties when axonal guidance plays a major role .
we numerically study the phase diagram of the model observing the emergence of a giant strongly connected component , representing the portion of neurons that are causally connected .
we provide strong evidence that the critical exponents depend on the temperature . |
since the galactic cosmic rays are hampered by the moon , a deficit of cosmic rays in its direction is expected ( the so - called `` moon shadow '' ) .
the moon shadow is an important tool to calibrate the performance of an air shower array .
in fact , the size of the deficit allows a measurement of the angular resolution and the position of the deficit allows the evaluation of the absolute pointing accuracy of the detector .
in addition , charged particles are deflected by the geomagnetic field by an angle inversely proportional to their energy .
the observation of such a displacement provides a direct check of the relation between shower size and primary energy , thus calibrating the detector .
a detailed monte carlo simulation of cosmic ray propagation in the earth - moon system is mandatory to understand the moon shadow phenomenology and to compare the observed westward displacement with the expectations in order to disentangle the geomagnetic effect from some possible experimental biases .
when the emission of photons by a given gamma - ray source has to be simulated , a classical approach can be followed : once its energy spectrum is known , it is enough to decide for how long the source emission has to be reproduced to calculate the number of expected events which have to be sampled .
of course , if the response of an earth - based array has to be simulated , for each one of these primary particles also the induced shower and the detector response have to be calculated .
the easiest way to do that , is to shoot up the particles from the detector to the source .
this trick allows to concentrate the efforts of the simulation only on the showers which effectively construct the signal .
nonetheless , this method can not be used for the moon - shadow as it is , because of two main reasons . 1 .
the moon shadow is not a signal , but a `` lack '' in the background .
the effect is provoked by charged particles and not by photons .
this implies that we must take into account the effect of the electro - magnetic fields on their trajectories .
the first argument is the more relevant one .
in fact , if we wanted to remain faithful to the approach of simulating only the particles which reach the detector , we should compute the showers of the background sorrounding the moon and then take off the ones coming exactly from within moon . in such terms , the simulation is likely unfeasible , at least for low energy threshold experiments . to be precise ,
if the detector energy threshold is a few hundreds of gev , even by considering a square sky window sorrounding the moon not so large ( e.g. @xmath0 ) , too many showers should have to be simulated to reproduce even just one year of data taking .
it is evident that the moon shadow requires a different strategy of simulation .
it is better to treat the moon like a standard source and then reverse the amplitude of the signal . in the end
, both the gamma - source case and the moon shadow case reduce to a perturbation of the cosmic - rays background .
the only difference is the sign of the perturbation . because of their electric charge the cosmic rays
do not proceed straight from the moon to the earth , but their trajectories are bent by the geomagnetic field .
since this effect plays a crucial role in the final result of the physics analysis , it must not be underestimated during the simulation . especially for those particle which have low energy ,
the bending effect is very strong and a realistic prediction is possible only if the simulation accounts correctly for the intensity and the direction of the geomagnetic field .
what is more , notice that when the particles are sent back to the moon from the detector , the sign of their charge must be reversed to properly reproduce the direction of the deviation .
the air showers development in the atmosphere has been generated with the corsika v. 6.500 code including the qgsjet - ii.03 hadronic interaction model for primary energy above 80 gev and the fluka code for lower energies @xcite .
cosmic ray spectra have been simulated in the energy range from 30 gev to 1 pev following the relative normalization given in @xcite , resulting from a global fit of main experimental data .
about @xmath1 showers have been sampled in the zenith angle interval 0 - 60 degrees .
the secondary particles have been propagated down to a cut - off energy of 1 mev . at present , the only experiment able to observe the moon shadow with high statistical significance and energy threshold well below 1 tev is argo - ybj @xcite .
therefore we reproduced an ideal detector placed in yangbajing ( 4300 m a.s.l . ) having geometrical features similar to argo - ybj and a duty - cycle of @xmath2 .
we used a geant4-based code @xcite .
the trigger threshold has been set to @xmath3 charged particles over the whole detector .
the core positions have been randomly sampled in an energy - dependent area large up to 10@xmath4 @xmath5 10@xmath4 m@xmath6 , centered on the detector .
it has been already noticed that if a primary cosmic ray ( energy @xmath7 , charge @xmath8 ) traversing the geomagnetic field is observed by a detector placed in yangbajing , its trajectory shows a deviation along the east - west direction@xcite which in first approximation can be written as : @xmath9}\ ] ] the sign is set according to the usual way to represent the east - west projection of the moon maps ( see fig .
[ fig : simulatedmaps ] ) . the eq .
[ eq : dipoledisplacement ] can be easily derived by assuming that the geomagnetic field is provoked by a pure dipole laying in the centre of the earth ( see appendix ) .
nonetheless this approximation , which is derived for nearly vertical primaries , is not enough when the primaries energy is below few tev . to perform a numerical estimate of the bending effect ,
it is necessary to adopt a model of the magnetic field in the earth - moon system .
the roughest one is the so - called virtual dipole model ( vdm ) , whose name is self - explaining .
a better choice is the tsyganenko - igrf model ( t - igrf hereafter ) @xcite , which accounts for both internal and external magnetospheric sources . we compared the effect on the particle trajectories of vdm and t - igrf , in both cases finding non negligible differences with respect to the @xmath10 $ ] formula . among the two models themselves , we observed discrepancies up to the @xmath11 level ( @xmath12 ) for sub - tev primary energies , mainly due to the field intensity near the earth surface .
since the t - igrf model accounts for more factors , we refer to it hereafter
. are also shown.,width=288 ] in fig .
[ fig : ewdisplacements ] you can appreciate the analytical trend ( eq . [ eq : dipoledisplacement ] ) together with the actual east - west displacement calculated applying the t - igrf model for protons and he nuclei .
the analytical approach clearly underestimates the east - west deviation .
[ fig : geomagnetictrend ] shows the differences of the t - igrf - induced deviation with respect to the leading term @xmath13}$ ] .
the upper ( lower ) panel contains such a residual deviation along the east - west ( north - south ) direction as a function of the primary energy .
altough there are no effects for energies @xmath14 , below few tev the residual displacement can reach @xmath15 . notice that unlike the analytical approach would suggest , the north - south deviation of the single primary is non - null , being zero only on average . by following the procedure described before , we can obtain the moon shadow maps represented in fig .
[ fig : simulatedmaps ] . ) .
the color scale represents the intensity of the signal.,scaledwidth=40.0% ] there can be appreciated the displacement induced by the geomagnetic field .
the long tail of the left part of the up - right map is made by the lowest energy particles ( below 1 tev ) which are more deviated .
concerning the bottom - left map , the detector by itself provides the smearing of the signal , leaving intact the circular symmetry , as expected .
it is possible to study the effects of the finite angular resolution of the detector and of the geomagnetic field separately .
as already noticed , if we consider the magnetic deviation but not the detector , the circular symmetry of the signal is broken only along the east - west direction ( see fig . [ fig : simulatedmaps ] , top - right map ) . that make us confident that only the smearing due to
the angular resolution affects the signal along the north - south direction , thus allowing its determination .
actually , we stress that what we determine by considering the spread of the signal along the north - south direction , can not be properly named angular resolution , because the moon is not a point - like source and its own finite angular width ( half a degree ) contributes to the spread .
the superposition of the two effects can be easily visualized in case of gaussian point spread function ( psf ) : @xmath16 where the root mean square of the signal @xmath17 is related to the variance @xmath18 of the psf .
the contribution of the moon size to the rms is ( not ) dominant when @xmath19 is low ( high ) , i.e. at high ( low ) particle multiplicities .
just to be explicit , the difference between @xmath17 and @xmath19 is @xmath20 if @xmath21 , less than @xmath22 if @xmath23 , and only @xmath24 if @xmath25 .
finally , it is worth to invert the switching order of the detector effect and of the geomagnetic field . in fig .
[ fig : effectoftheangularresolution ] the effect of the angular resolution on the east - west projection of the moon is shown . because of the signal asymmetry induced by the magnetic field , such an effect provides not only the smearing , but also a further displacement of the signal peak .
the west tail of the shifted signal , in fact , has a larger weight than the sharp east one and tends to pull the signal in its direction . in fig .
[ fig : ewdisplacement ] the total eastward displacement is plot versus the number of detected particles .
we reproduced the moon shadow effect as observed by an ideal detector located at high altitude .
much attention has been devoted to the simulation of the geomagnetic field bending the particle trajectories .
we have been able to disentangle the contribution to the final signal of the geomagnetic field and of the detector psf respectively .
we quantified the contribution of the finite moon disc to the angular resolution estimation . for a comparison with the data collected by the argo - ybj experiment ,
see @xcite .
here is shown how to obtain the formula [ eq : dipoledisplacement ] . since only the magnetic field
is supposed to act upon the particles trajectories , the lorentz equation can be written as : @xmath26 where : * @xmath27 is the particle position at the time @xmath28 ; * @xmath29 and @xmath30 are the initial position and velocity of the particle ; * @xmath31 and @xmath7 are its charge and its ( constant ) energy ; * @xmath32 is the magnetic field , which in principle can vary both with respect to the position and to the time . if it is possible to write down an explicit functional form for @xmath33 , an attempt to solve the equation [ eq : integrallorentzequation ] can be made . on the contrary ,
especially when the variation with the time can not be easily summarized with an analytical formula , a numerical solution is unavoidable .
the equation [ eq : integrallorentzequation ] explicitly shows the perturbation induced by the magnetic field on the straight trajectory ( @xmath34 ) .
this suggests an iterative method to determine the solution , which can be expressed as the series : @xmath35 where @xmath36 is the unperturbed ( straight ) trajectory and for the higher orders holds : @xmath37 where @xmath38 is the displacement from the unperturbed trajectory at the time @xmath28 .
being content with the first - order approximation , it can be obtained : @xmath39 or : @xmath40 where @xmath41 is the integral of the magnetic field along the straight trajectory , whose value depends only on the time of the motion ( @xmath28 ) and on its initial conditions ( @xmath29 and @xmath30 ) .
since the phenomenon studied concerns ultra - relativistic particles and fixing for a moment the initial position and the final time , the equation [ eq : firstorderdisplacement ] becomes : @xmath42 in short , on a first approximation the displacement depends only on the ratio charge - to - energy of the primary and on the initial direction of its ultrarelativistic motion ( versor @xmath43 ) .
let us consider only the lowest order of the geomagnetic field multipoles - expansion , i.e. the _ dipole term _ : @xmath44 where @xmath45 has intensity @xmath46 t m@xmath4 and the south magnetic pole is supposed to have coordinates @xmath47 south , @xmath48 east . by setting @xmath49 ( _ vertical direction approximation _ ) and integrating from yangbajing to a distance @xmath50 earth s radii , the equation [ eq : dipoledisplacement ] is immediately obtained .
99 d. heck et al . ,
report * fzka 6019 * 1998 . b. wiebel - sooth et al . ,
_ a&a _ * 330 * 389 , 1998 .
r.iuppa et al . , these proceedings s. agostinelli et al . , nim a 506 ( 2003 ) 250 - 303 tibet as@xmath51 coll .
, astroparticle physics 28 ( 2007 ) 137 - 142 n. a. tsyganenko and m. i. sitnov , journal of geophysical research , vol .
112 , a06225 | an accurate montecarlo simulation of the deficit of primary cosmic rays in the direction of the moon has been developed to interpret the observations reported in the tev energy region until now .
primary particles are propagated trough the geomagnetic field in the earth - moon system . the algorithm is described and the contributions of the detector resolution and of the geomagnetic field are disentangled . cosmic rays , moon shadow analysis , monte carlo simulation |
since the mid 1990 s , several compounds of pr have been found to exhibit heavy fermion behavior . the crystalline electric field ( cef ) ground state of the pr@xmath5 ions in these compounds appears to be a @xmath6 nonmagnetic doublet that carries an electric quadrupole moment .
it is conceivable that the heavy fermion state in these pr compounds could originate from the interaction between the pr@xmath5 electric quadrupole moments and the charges of the conduction electrons
. this would be the electric analogue of the exchange interaction between the magnetic dipole moments of ce or u ions and the conduction electron spins that is widely believed to be responsible for the heavy fermion state in most ce and u heavy fermion compounds .
in fact , such a mechanism was proposed by cox in 1987 @xcite to account for the non - fermi liquid temperature dependences of certain normal state physical properties of the heavy electron superconductor ube@xmath11 .
the pr compounds that display heavy fermion behavior include prinag@xmath12 @xcite , prfe@xmath13p@xmath1 @xcite , and , possibly , prfe@xmath13sb@xmath1 @xcite . about a year ago , we reported that the compound pros@xmath0sb@xmath1 exhibits superconductivity with a superconducting critical temperature @xmath14 k that apparently develops out of a heavy fermi liquid with a quasiparticle effective mass @xmath15 , where @xmath16 is the mass of the free electron @xcite . as far as we know , pros@xmath0sb@xmath1 is the first example of a heavy fermion superconductor based on pr ; all of the other known heavy fermion superconductors are compounds of ce or u. the superconducting state appears to be unconventional in nature and may consist of two distinct superconducting phases @xcite .
an ordered phase , presumably of magnetic or quadrupolar origin , occurs at high fields @xmath17 tesla and low temperatures @xmath18 k @xcite , suggesting that the superconductivity may occur in the vicinity of a magnetic or quadrupolar quantum critical point ( qcp ) . in an effort to obtain information about the interactions that are responsible for the heavy fermion state and superconductivity in pros@xmath0sb@xmath1
, we have performed measurements of various normal and superconducting state properties of this compound as a function of temperature , pressure , and magnetic field @xcite .
analysis of magnetic susceptibility @xmath19 , specific heat @xmath20 , electrical resistivity @xmath21 , and inelastic neutron scattering measurements within the context of a cubic crystalline electric field ( cef ) yields a pr@xmath5 energy level scheme that consists of a @xmath6 nonmagnetic doublet ground state that carries an electric quadrupole moment , a low lying @xmath7 triplet excited state at @xmath8 k , and @xmath9 triplet and @xmath10 singlet excited states at much higher temperatures ( @xmath22 k and @xmath23 k , respectively ) @xcite .
this scenario suggests that the underlying mechanism for the heavy fermion behavior in pros@xmath0sb@xmath1 may involve the interaction of pr@xmath5 electric quadrupole moments with the charges of the conduction electrons , rather than pr@xmath5 magnetic dipole moments with the spins of the conduction electrons .
it also raises the possibility that electric quadrupole fluctuations play a role in the superconductivity of pros@xmath0sb@xmath1 . in this paper
, we briefly review the current experimental situation regarding the heavy fermion state , the superconducting state , and a high field , low temperature phase that is apparently associated with magnetic or quadrupolar order in pros@xmath0sb@xmath1 .
the first evidence for a heavy fermion state in the filled skutterudite compound pros@xmath0sb@xmath1 emerged from specific heat @xmath20 measurements on a pros@xmath0sb@xmath1 pressed pellet ( formed by pressing a collection of small single crystals in a cylindrical die ) at low temperatures .
specific heat data in the form of a plot of @xmath24 vs @xmath25 between @xmath26 k and @xmath27 k for the pros@xmath0sb@xmath1 pressed pellet from refs . @xcite and @xcite are shown in fig .
the @xmath20 data have been corrected for excess sb derived from the molten sb flux in which the crystals were grown .
the line in the figure represents the expression @xmath28 , where @xmath29 and @xmath30 are electronic and phonon contributions , respectively , and @xmath31 is a schottky anomaly for a two level system consisting of a doublet ground state and a triplet excited state at an energy @xmath32 above the ground state .
the best fit of this expression to the data yields the values @xmath33 mj / mol k@xmath34 , @xmath35 mj / mol k@xmath36 ( corresponding to a debye temperature @xmath37 k ) , and @xmath38 k. superimposed on the schottky anomaly is a feature in the specific heat due to the onset of superconductivity at @xmath14 k which is also observed as an abrupt drop in @xmath21 to zero and as a sharp onset of diamagnetism in @xmath19 . the feature in @xmath39 due to the superconductivity is also shown in the top inset of fig . [ heat ] along with an entropy conserving construction from which the ratio of the jump in specific heat @xmath40 at @xmath41 , @xmath42 mj / mol k@xmath34 , has been estimated . using the bcs relation @xmath43
, we obtain another estimate for @xmath44 of @xmath45 mj / mol k@xmath34 .
the value of @xmath46 is larger than that reported in ref .
@xcite due to the correction of the @xmath20 data for the excess sb ( about @xmath47 percent of the total mass ) .
this value is comparable to that inferred from the fit to the @xmath24 vs @xmath25 data in the normal state above @xmath41 , and indicative of heavy fermion behavior .
a similar analysis of the @xmath20 data taken at the university of karlsruhe on several single crystals of pros@xmath0sb@xmath1 prepared in our laboratory yielded @xmath48 mj / mol k@xmath34 , @xmath49 k , @xmath50 k , and @xmath51 , much higher than the bcs value of @xmath52 and indicative of strong coupling effects @xcite .
it is interesting that we also find a large value of @xmath53 in recent @xmath20 measurements on one single crystal of pros@xmath0sb@xmath1 at ucsd .
although the values of @xmath44 determined from these experiments vary somewhat , they are all indicative of a heavy electron ground state and an effective mass @xmath3 .
divided by temperature @xmath25 , @xmath24 , vs @xmath25 for a pros@xmath0sb@xmath1 pressed pellet .
the line represents a fit of the sum of electronic , lattice , and schottky contributions to the data .
upper inset : @xmath54 vs @xmath25 near @xmath41 for a pros@xmath0sb@xmath1 pressed pellet ( @xmath55 is the electronic contribution to @xmath56 ) .
lower inset : @xmath24 vs @xmath25 near @xmath41 for pros@xmath0sb@xmath1 single crystals , showing the structure in @xmath40 near @xmath41 .
data from refs .
@xcite.,width=302 ] further evidence of heavy fermion superconductivity is provided by the upper critical field @xmath57 vs @xmath25 curve shown in fig .
[ phase ] @xcite .
the orbital critical field @xmath58 can be derived from the slope ( -19 koe / k ) of the @xmath57 curve near @xmath41 and used to estimate the superconducting coherence length @xmath59 via the relation @xmath60 , where @xmath61 is the flux quantum .
the fermi velocity @xmath62 can be obtained from the bcs relation @xmath63 and used to determine the effective mass @xmath64 by means of the expression @xmath65 . using a simple free electron model to estimate the fermi wave vector @xmath66 , an effective mass @xmath3
is obtained @xcite .
calculating @xmath44 from @xmath64 yields @xmath67 mj / mol k@xmath34 , providing further evidence for a heavy fermion state in pros@xmath0sb@xmath1 . )
phase diagram of a pros@xmath0sb@xmath1 single crystal showing the regions exhibiting superconductivity ( sc ) and the high field ordered phase ( hfop ) .
the boundary delineating the sc phase is based on measurements of @xmath68 , while the hfop phase boundary is based on @xmath68 , @xmath69 , @xmath70 , and @xmath71 measurements ( see text ) .
the dashed line , derived from a peak in @xmath72 vs @xmath25 , is a measure of the splitting between the highest and lowest zeeman levels of the pr@xmath73 ground and @xmath7 excited states , respectively ( see text for further details ) .
data from refs .
@xcite.,width=302 ] recently , sugawara et al .
@xcite performed de haas - van alphen effect measurements on pros@xmath0sb@xmath1 .
they found that the topology of the fermi surface is close to that of the reference compound laos@xmath0sb@xmath1 and is explained well by band structure calculations .
in contrast to the similarity in the fermi surface topologies of pros@xmath0sb@xmath1 and laos@xmath0sb@xmath1 , the cyclotron effective masses @xmath74 of pros@xmath0sb@xmath1 are enhanced up to @xmath75 times relative to those of laos@xmath0sb@xmath1 .
the sommerfeld coefficient @xmath44 estimated from the fermi surface volume and the value of @xmath74 , assuming a spherical fermi surface , is @xmath76 mj / mol k@xmath34 , which is two to three times smaller than the value of @xmath44 inferred from the normal and superconducting properties of pros@xmath0sb@xmath1 .
our studies of laos@xmath0sb@xmath1 single crystals reveal superconductivity with a @xmath41 of @xmath77 k.
the @xmath19 data for pros@xmath0sb@xmath1 exhibit a peak at @xmath78 k and saturate to a value of @xmath79 @xmath80/mol as @xmath81 , indicative of a nonmagnetic ground state . at temperatures above @xmath82
k , @xmath19 is strongly t - dependent , as expected for well defined pr@xmath5 magnetic moments . in the analysis of the @xmath19 data ,
interactions between pr@xmath5 ions and hybridization of the pr @xmath83f and conduction electron states were neglected , while the degeneracy of the hund s rule multiplet of the pr@xmath5 ions was assumed to be lifted by a cubic crystalline electric field ( cef ) and to have a nonmagnetic ground state . according to lea , leask , and wolf ( llw ) @xcite , in a cubic cef , the pr@xmath84 hund s rule multiplet splits into a @xmath10 singlet , a @xmath6 nonmagnetic doublet that carries an electric quadrupole moment , and @xmath9 and @xmath7 triplets .
it was assumed that the nonmagnetic ground state of the pr@xmath5 ions corresponds to either a @xmath10 singlet or a @xmath6 nonmagnetic doublet @xcite .
although reasonable fits to the @xmath85 data could be obtained for both @xmath10 and @xmath6 ground states , the most satisfactory fit was obtained for a @xmath6 nonmagnetic doublet ground state with a @xmath7 first excited triplet state at @xmath86 k and @xmath9 and @xmath10 excited states at @xmath87 k and @xmath88 k , respectively .
inelastic neutron scattering measurements on pros@xmath0sb@xmath1 @xcite reveal peaks in the ins spectrum at @xmath89 mev ( @xmath90 k ) and @xmath91 mev ( @xmath92 k ) that appear to be associated with transitions between the @xmath6 ground state and the @xmath7 first and @xmath9 second excited states , respectively , that are in good agreement with the pr@xmath5 cef energy level scheme determined from the analysis of the @xmath19 data . as noted above , the schottky anomaly in the @xmath20 data on pros@xmath0sb@xmath1 taken at ucsd and at the university of karlsruhe @xcite can be described well by a two level system consisting of a doublet ground state and a low lying triplet excited state with a splitting of @xmath93 k , a value that is comparable to the values deduced from the @xmath19 and ins data .
while a magnetic @xmath7 pr@xmath5 ground state ( @xmath9 is not a possible ground state for a cubic system in the llw formulation ) could also produce a nonmagnetic heavy fermion ground state via an antiferromagnetic exchange interaction ( kondo effect ) , the behavior of @xmath21 of pros@xmath0sb@xmath1 in the normal state does not resemble the behavior of @xmath21 expected for this scenario . for a typical magnetically - induced heavy fermion compound
, @xmath21 often increases with decreasing temperature due to kondo scattering , reaches a maximum , and then decreases rapidly with decreasing temperature as the highly correlated heavy fermion state forms below the coherence temperature . at low temperatures
, @xmath21 typically varies as @xmath94 with a prefactor @xmath95 [ @xmath96 cm k@xmath34(mj / mol)@xmath97 @xmath98 that is consistent with the kadowaki - woods relation @xcite .
in contrast , @xmath21 of pros@xmath0sb@xmath1 @xcite exhibits typical metallic behavior with negative curvature at higher temperatures and a pronounced ` roll off ' below @xmath99 k before it vanishes abruptly when the compound becomes superconducting . the `
roll off ' in @xmath100 is consistent with a decrease in charge or spin dependent scattering of conduction electrons by the pr@xmath5 ions due to the decrease in population of the low lying first excited state ( @xmath7 ) as the temperature is lowered . the ` roll off ' below @xmath99 k and the negative curvature at higher temperatures in @xmath21 can be described reasonably well by calculations based on magnetic and aspherical coulomb scattering of conduction electrons by the pr@xmath5 ions with a low lying @xmath7 excited state separated from the @xmath6 ground state by @xmath75 k and excited @xmath9 triplet and @xmath10 singlet states at much higher energies comparable to those found from the analysis of the @xmath19 and ins measurements @xcite .
the @xmath21 data can be described by a temperature dependence of the form @xmath94 between @xmath99 k and @xmath101 k , but with a prefactor @xmath102cm / k@xmath34 that is nearly two orders of magnitude smaller than that expected from the kadowaki - woods relation ( @xmath103 cm / k@xmath34 for @xmath104 mj / mol k@xmath34 ) @xcite .
interestingly , @xmath21 is consistent with @xmath105 behavior with a value @xmath106 cm / k@xmath34 in fields of @xmath82 tesla @xcite in the high field ordered phase discussed in section @xmath107 .
the zero - field temperature dependence of @xmath21 is similar to that observed for the compound prinag@xmath12 , which also has a low value of the coefficient @xmath108 , an enormous @xmath44 of @xmath109 j / mol k@xmath34 , and a @xmath6 nonmagnetic doublet ground state @xcite .
another possible source of the enhanced effective mass in pros@xmath0sb@xmath1 may involve excitations from the ground state to the the low lying first excited state in the pr@xmath5 cef energy level scheme @xcite .
two studies of the nonlinear magnetic susceptibility have been performed in an attempt to determine the cef ground state of the pr@xmath5 ion in pros@xmath0sb@xmath1 @xcite .
the nonlinear susceptibility @xmath110 is the coefficient of the @xmath111 term in the expansion of the magnetization @xmath112 in a series of odd powers of @xmath113 ; i.e. , @xmath114 , where @xmath115 is the ordinary linear susceptibility .
in an ionic situation , @xmath110 is isotropic and varies as @xmath116 for a magnetic ground state , whereas @xmath110 is anisotropic and diverges at low temperatures for @xmath117 $ ] and approaches a constant for @xmath118 $ ] for a non - kramers @xmath6 doublet ground state @xcite .
this type of study was previously employed in an attempt to determine the ground state of u in the compound ube@xmath11 @xcite . in both studies of pros@xmath0sb@xmath1 , @xmath119 was found to behave similarly for @xmath117 $ ] and @xmath118 $ ] , exhibiting a minimum near @xmath83 k followed by a maximum near @xmath77 k and a negative divergence with decreasing temperature .
calculations based on the quadrupolar anderson - hamiltonian described the @xmath119 data reasonably well for @xmath117 $ ] , but not very well for @xmath118 $ ] .
it was concluded that the data were qualitatively consistent with a @xmath6 ground state , given the limitations of the experiment and the complexity of the theory .
the @xmath119 studies are difficult to interpret because of the curvature of @xmath120 and the complications that arise at lower temperatures @xmath121 and lower fields @xmath122 due to the superconductivity and at temperatures @xmath123 k and higher fields @xmath124 tesla by the onset of the high field ordered phase , discussed in section @xmath107 .
a number of features in the superconducting properties of pros@xmath0sb@xmath1 indicate that the superconductivity of this compound is unconventional in nature .
one of these features is the ` double - step ' structure in the jump in @xmath20 near @xmath41 in single crystals ( lower inset of fig . [ heat ] ) that suggests two distinct superconducting phases with different @xmath41 s : @xmath125 k and @xmath126 k @xcite .
this structure is not evident in the @xmath20 data taken on the pressed pellet of pros@xmath0sb@xmath1 shown in the upper inset of fig .
[ heat ] , possibly due to strains in the single crystals out of which the pressed pellet is comprised that broaden the transitions at @xmath127 and @xmath128 so that they overlap and become indistinguishable . however , at this point , we are unable to eliminate the possibility that the two apparent jumps in @xmath20 are due to sample inhomogeneity .
it is noteworthy that all of the single crystal specimens prepared in our laboratory and investigated by our group and our collaborators exhibit this ` double - step ' structure .
multiple superconducting transitions , apparently associated with distinct superconducting phases , have previously been observed in two other heavy fermion superconductors , upt@xmath129 @xcite and u@xmath130th@xmath131be@xmath132 @xcite .
measurements of the specific heat in magnetic fields reveal that the two superconducting features shift downward in temperature at nearly the same rate with increasing field , consistent with the smooth temperature dependence of the @xmath133 curve @xcite .
these two transitions have also been observed in thermal expansion measurements @xcite , which , from the ehrenfest relation , reveal that @xmath127 and @xmath128 have different pressure dependences , suggesting that they are associated with two distinct superconducting phases .
another feature is the power law t - dependence of @xmath134 , @xmath135 , after the schottky anomaly and @xmath30 lattice contributions have been subtracted from the @xmath20 data .
( as reported in ref .
@xcite , @xmath134 follows a power law with @xmath136 when the schottky anomaly is not subtracted . )
however , this dependence can only be established from @xmath41 down to @xmath137 , since it is not possible to reliably correct the @xmath20 data at lower temperatures for an enormous nuclear schottky anomaly .
power law t - dependences of the superconducting properties are generally attributed to nodes in the superconducting energy gap at points or lines on the fermi surface . among three recent experiments on pros@xmath0sb@xmath1 , described below , two yield evidence for an isotropic energy gap , while another provides evidence for two distinct superconducting phases in the h - t plane with different numbers of point nodes in the energy gap
. recent transverse field @xmath138sr @xcite and sb - nqr measurements @xcite on pros@xmath0sb@xmath1 are consistent with an isotropic energy gap . along with the specific heat ,
these measurements indicate strong coupling superconductivity .
these findings suggest an s - wave , or , perhaps , a balian - werthamer p - wave order parameter . on the other hand ,
the superconducting gap structure of pros@xmath0sb@xmath1 was investigated by means of thermal conductivity measurements in magnetic fields rotated relative to the crystallographic axes by izawa et al .
these measurements reveal two regions in the @xmath139 plane , a low field region in which @xmath140 has two point nodes , and a high field region where @xmath140 has six point nodes .
the line lying between the low and high field superconducting phases may be associated with the transition at @xmath128 , whereas the line between the high field superconducting phase and the normal phase , @xmath133 , converges with @xmath127 as @xmath141 .
clearly , more research will be required to further elucidate the nature of the superconductivity in pros@xmath0sb@xmath1 .
vs @xmath25 for laos@xmath0sb@xmath1 , based on resistance
@xmath142 vs temperature @xmath25 data shown in the inset.,width=302 ] it is noteworthy that the reference compound without 4f electrons , laos@xmath0sb@xmath1 , is also superconducting , but with @xmath41 @xmath143 1 k , considerably smaller than that of pros@xmath0sb@xmath1 .
this suggests that the pairing interaction is enhanced by the presence of the pr @xmath83f electrons , possibly through the interaction of the pr@xmath5 electric quadrupole moments with the conduction electrons .
in contrast , the values of @xmath41 of the other prt@xmath13x@xmath1 filled skutterudites are smaller than their la - based counterparts .
the values of @xmath41 in k for mt@xmath13x@xmath1 compounds with m = la or pr are listed in table [ tc ] . a plot of the upper critical field @xmath57 vs @xmath25 for laos@xmath0sb@xmath1 and the resistive @xmath144 transition curves upon which it is based are shown in fig .
[ uppercrit ] . .superconducting critical temperature @xmath41 of lnt@xmath13x@xmath1 compounds for ln = la , pr ;
t = fe , ru , os ; and x = p , as , sb . the @xmath41 values listed are in k and have been derived from references @xcite . the symbols and indicate that superconductivity has not been observed in those compounds above @xmath145 k and @xmath146 k , respectively . [ cols="^,^,^,^,^,^,^,^ " , ] [ tc ]
evidence for a high field ordered phase was first derived from magnetoresistance measurements in the temperature range @xmath147 mk @xmath148 k and magnetic fields up to @xmath27 tesla @xcite .
recently , the magnetoresistance measurements have been extended up to @xmath149 tesla for @xmath146 k @xmath148 k @xcite .
the @xmath139 phase diagram , depicting the superconducting region and the high field ordered phase , is shown in fig .
the line that intersects the high field ordered phase represents the inflection point of the ` roll - off ' in @xmath21 for @xmath150 tesla at low temperatures and is a measure of the splitting between the pr@xmath5 ground state and the first excited state , which decreases with field ( see fig .
[ magrho ] ) .
the high field ordered phase has also been observed by means of large peaks in the specific heat @xcite and thermal expansion @xcite and kinks in magnetization vs magnetic field curves @xcite in magnetic fields @xmath17 tesla and temperatures @xmath18 k. shown in fig .
[ magrho ] ( a ) and ( b ) are @xmath21 data for various magnetic fields up to @xmath149 tesla for pros@xmath0sb@xmath1 , which reveal drops in @xmath21 due to the superconductivity for @xmath151 tesla and features in @xmath21 associated with the onset of the high field ordered phase for @xmath124 tesla .
isotherms of electrical resistivity @xmath152 vs @xmath113 at various temperatures @xmath146 k @xmath153 k and fields @xmath154 tesla are shown in fig .
[ magrho](c ) .
the fields denoting the boundaries of the high field ordered phase , @xmath155 and @xmath156 , are indicated in the figure .
vs temperature @xmath25 at various magnetic fields @xmath113 up to 18 tesla ( t ) for a pros@xmath0sb@xmath1 single crystal .
( c ) @xmath152 vs @xmath113 at various temperatures between @xmath146 k and @xmath157 k .
the rapid drop in @xmath152 to zero for @xmath158 t is due to the superconducting transition , while the shoulder in @xmath21 at @xmath159 k above @xmath160 t and the kinks in @xmath161 ( indicated at @xmath155 and @xmath156 ) below @xmath162 k are due to a field induced phase ( high field ordered phase - hfop ) .
after refs .
experiments on the filled skutterudite compound pros@xmath0sb@xmath1 have revealed a number of extraordinary phenomena : a heavy fermion state characterized by an effective mass @xmath15 , unconventional superconductivity below @xmath14 k with two distinct superconducting phases , and a high field ordered phase , presumably associated with magnetic or quadrupolar order .
analysis of @xmath19 , @xmath20 , @xmath21 , and ins data indicate that pr@xmath5 has a nonmagnetic @xmath6 doublet ground state that carries an electric quadrupole moment , a low lying @xmath7 triplet excited state at @xmath8 k , and @xmath9 triplet and @xmath10 singlet excited states at much higher energies .
this suggests that the interaction between the quadrupole moments of the pr@xmath5 ions and the charges of the conduction electrons , as well as the excitations between the @xmath6 ground state and @xmath7 low lying excited state , may play an important role in generating the heavy fermion state and superconductivity in this compound .
the heavy fermion state , unconventional superconductivity , and high field ordered phase observed in pros@xmath0sb@xmath1 and reviewed herein present a significant challenge for theoretical description @xcite .
this research was supported by the us department of energy ( doe ) , grant no . de - fg03 - 86er-45230 ; the us national science foundation ( nsf ) , grant no .
dmr-00 - 72125 ; and the nedo international joint research program at ucsd .
los alamos national laboratory is supported by the nsf , the state of florida , and the us doe .
99 ho p - c , zapf v s , bauer e d , frederick n a , maple m b , geister g , rogl p , berger st , paul ch , and bauer e 2001 in _ physical phenomena at high magnetic fields - iv _ boebinger g , fisk z , gorkov l p , lacerda a , and schrieffer j r , eds .
( world scientific , singapore ) 98 - 103 . | superconductivity is observed in the filled skutterudite compound pros@xmath0sb@xmath1 below a critical temperature temperature @xmath2 k and appears to develop out of a nonmagnetic heavy fermi liquid with an effective mass @xmath3 , where @xmath4 is the free electron mass .
features associated with a cubic crystalline electric field are present in magnetic susceptibility , specific heat , electrical resistivity , and inelastic neutron scattering measurements , yielding a pr@xmath5 energy level scheme consisting of a @xmath6 nonmagnetic doublet ground state , a low lying @xmath7 triplet excitied state at @xmath8 k , and much higher temperature @xmath9 triplet and @xmath10 singlet excited states .
measurements also indicate that the superconducting state is unconventional and consists of two distinct superconducting phases . at high fields and low temperatures , an ordered phase of magnetic or quadrupolar origin
is observed , suggesting that the superconductivity may occur in the vicinity of a magnetic or quadrupolar quantum critical point . |
a main goal of future collider physics is to shed light into the electroweak symmetry breaking mechanism . within the standard model this mechanism
is generated by the higgs sector , leading to gauge invariant mass terms where needed . at the lagrangean level , this is a gauged linear @xmath6-symmetric sigma model , spontaneously broken to @xmath7 by a nonvanishing vacuum expectation value .
the mass of the higgs boson itself is directly related to the corresponding quartic coupling , which implies the nondecoupling property of the higgs boson . for large higgs mass
one expects large quantum corrections which blow up within the standard perturbative approach . within the perturbative approach
, a lot of work has been done in the recent years .
many quantum corrections for relevant observables which contain information on the higgs boson are available @xcite .
the radiative corrections of enhanced electroweak strength to the decay of the higgs boson were first derived at one - loop in @xcite , and at two - loop in @xcite . for low values of the higgs mass the theoretical uncertainty is thus small . for heavier masses the leading order and next - to - leading order results , start to deviate from the nnlo result , inducing theoretical ambiguities .
one expects large dependences on the renormalization scheme essentially coming from the truncation of the perturbative series which makes theoretical predictions unreliable .
whereas the perturbative region will be fully covered by lep , tevatron and the lhc , it is of special importance to have a good theoretical understanding of the heavy higgs regime which is also in reach of the lhc .
although the existing experimental data points presently in the direction of a higgs mass in the perturbative region , one should not forget that these bounds are logarithmically soft , and that the experimental data on @xmath8 still differ considerably in the high precision measurements at lep1 and slc @xcite .
moreover , certain two - loop quantities are presently known theoretically only as mass expansions , and given the high statistics of the experiments , this induces rather substantial shifts in the higgs mass prediction .
apart from the phenomenological motivation , it is a major challenge to develop and improve calculational methods beyond the standard perturbative approach .
a well - known nonperturbative approach are monte carlo studies on a lattice .
the results of lattice calculations indicate that for higgs masses beyond @xmath9 700 gev one expects large cutoff effects @xcite .
they stem from the fact that the higgs sector is not asymptotically free .
we advocate another nonperturbative approach , the @xmath0 expansion . there ,
the quantum correction are ordered by the number of field components @xcite .
the method was applied to the higgs sector to leading order a long time ago @xcite but the large deviation from standard perturbative results shaded doubts on its validity .
recently we performed a next - to - leading order calculation of the higgs propagator , showing at low coupling a remarkable agreement between the nnlo perturbative result and our result @xcite .
this proved the @xmath0 expansion to be a valuable tool for computing observables beyond the perturbative approach , and for avoiding the truncation of the perturbative series and thus eliminating any renormalization scale dependence . in this paper
we extend the results to the case of three - point functions , which results in a quantitative , nonperturbative understanding of the physical processes @xmath10 .
it was tried to make the computation as transparent as possible to convince the reader that the method is conceptually well settled and practicable , though some amount of numerical work is unavoidable , which on the other hand is common to the calculation of higher - order finite momentum green functions in massive theories .
after some preliminaries , where we remind the lo results and the issue of the tachyon of the @xmath0 approach , we show how one obtains a well defined , tachyon free representation of green functions .
an ultraviolet subtraction scheme is defined in terms of an order @xmath0 counterterm lagrangean .
then , the finite and tachyon free green functions are expressed in form of multiloop graphs made out of dressed propagators .
these can be reduced to two - dimensional integrals over dressed propagators and form factors .
then we describe how to perform the remaining integrations numerically , to obtain the final results for the two- and three - point functions .
we show and discuss our nonperturbative result for the higgs line shapes in two physical scattering processes , and compare to the perturbative result . after this discussion , we summarize the conclusions of the paper .
in this section we review the leading order result for fixing our notations , and discuss the structure of the counterterms which absorb the ultraviolet divergencies at leading and next - to - leading order . because we are interested in the effects of a heavy higgs boson
, we neglect all the gauge couplings and also the higher order corrections in the yukawa couplings .
thus , the starting point for the @xmath0 expansion in a scalar theory is given by an @xmath11-symmetric @xmath12 theory : @xmath13 here @xmath14 is a @xmath15-component vector and @xmath16 , @xmath17 are bare coupling constants indicated by the subscript zero .
because we want to focus on the spontaneously broken case , we rewrite it as @xmath18 @xmath19 is the bare vacuum expectation value of @xmath20 .
the symmetry of the ground state is broken down to @xmath21 , which amounts to the presence of @xmath22 massless goldstone bosons ( gbs ) , @xmath23 , and a higgs particle @xmath24 , of mass @xmath25 . in the case
@xmath26 , the gbs are related to the longitudinal degrees of freedom of the vector bosons @xmath27 , @xmath28 by the equivalence theorem . to overcome the combinatorial difficulties of multi - loop calculations
, we use a well - known auxiliary field formalism @xcite .
we add to the lagrangean the following expression , which becomes identically zero when one uses the equations of motion : @xmath29 ^ 2 .
\end{aligned}\ ] ] @xmath30 is an arbitrary lagrange multiplier .
one additionally has the freedom to replace @xmath31 by @xmath32 , where @xmath33 are some constants , without changing the equations of motion .
these constants correspond to a @xmath34 vacuum expectation value and a wave function renormalization .
we introduce renormalized quantities in the following way : @xmath35 from the discussion above it is clear that only @xmath36 , @xmath37 , @xmath38 and @xmath39 are renormalization constants in the usual sense .
the other can be chosen arbitrarily as long as the choice is consistent .
their introduction will turn out to be useful for the renormalization of functions which contain auxiliary particles as external legs .
the lagrangean is now of the form : @xmath40 note that a mixing between @xmath34 and @xmath41 takes place .
we see that the expression contains up to a trivial surface term @xmath42 all possible field combinations of dimension four which are built out of an @xmath11 vector and a dimension two singlet field allowed by lorentz invariance and the @xmath11 symmetry .
the vacuum expectation value of the higgs reduces the symmetry as usual to @xmath21 .
the constants @xmath43 are expressible in terms of the above renormalization constants .
@xmath44 these constants define a redundant counterterm structure of the theory , and are fixed by imposing conditions on one - particle irreducible green functions . whereas some are necessary to absorb the divergencies occurring in physical processes , other can be adjusted to make subtractions on green functions containing auxiliary fields as external legs .
clearly all these subtractions have to be consistent with each other . in order to determine the renormalization constants , the quantum corrections to the tree level green functions have to be calculated up to a given order , and have to be related to the counterterms by a renormalization scheme . in perturbation theory
the occurring renormalization scheme dependence is proportional to some power of the coupling constant , and namely one order higher than the loop order in which the process is calculated .
thus it is enhanced in the case of strong coupling .
it is a crucial point that this is not the case in the @xmath0 expansion . once the relations between a set of observables and the green functions are fixed , all further predictions are exact , up to unevaluated powers of @xmath0 .
a change of the ultraviolet subtraction point @xmath45 ( see below ) is compensated by a change in @xmath46 .
thus it does not effect any observable prediction , being merely a reparameterization of the theory . before describing the computation of the higgs propagator in next - to - leading order in @xmath0
, we would like to repeat the leading order results to set the stage for the next step . to order @xmath47 the quantum corrections are simply the sum of all graphs containing arbitrary numbers of external @xmath34 legs attached to a gb loop .
the sum can be carried out and a simple closed expression for the resulting effective potential can be found @xcite .
the only divergent graphs are a @xmath34 tadpole and a momentum dependent @xmath34 self - energy graph shown in figure 1 .
( 13,2.5 ) ( + 1.5,+1.25)(0,0)@xmath48 ( + 2.5,+0.25 ) ( + 6.5,+1.25)(0,0),@xmath49 ( + 8.0,+0.25 ) both divergencies can be absorbed into @xmath39 and @xmath50 , or in other words , into the coupling constants of the theory .
no other renormalization constant is involved . whereas the observable vev is defined by the ground state of the theory
resulting in an observable vector boson mass if one couples the model to a gauge theory the value of @xmath46 can be fixed by a measurement of the higgs mass . in terms of the introduced renormalization constants ,
one has : @xmath51 the integrals which correspond to the leading order corrections are trivial .
absorbing combinatorial factors in the expressions , one finds in dimensional regularization : @xmath52 defining an energy dependent higgs mass by @xmath53 , where @xmath54 , one gets an expression for the matrix of two - point functions .
its inverse defines the dressed propagators to leading order .
@xmath55 is the renormalization scale .
the resulting propagators are found to be : @xmath56 the gb propagator is not influenced at leading order .
clearly the same results can be obtained without the auxiliary field .
the dressed @xmath34 propagator corresponds to a geometric series of loop graphs @xcite .
note that the propagators contain an imaginary part for physical , timelike momenta , which describes higgs decay into goldstone bosons . to keep the pole on the correct riemann sheet one has to use the @xmath57 prescription in the logarithm of the @xmath58-dependent mass : @xmath59 for spacelike momenta
, a landau pole occurs at the scale : @xmath60 \quad .\ ] ] since @xmath61 is negative beyond the landau scale @xmath62 , all these propagators have an extra pole for a spacelike momentum @xmath63 which is defined by the equation : @xmath64 there is a tendency to discard non asymptotically free theories as fundamental theories .
the occurrence of the tachyon in the sigma model is widely believed to be an indication of an ill - defined quantum field theory .
however , to or knowledge there is so far no rigorous proof thereof .
we will show in the next section that the tachyon is technically an artifact of perturbation theory and can by no means be considered an inconsistency of the model .
apart from these issues , it is natural to ask for higher order effects . how to go beyond the leading order in @xmath0 is directly related to the tachyon problem , because on has to integrate over graphs built out of tachyon contaminated propagators .
a method to treat this is given below .
such a procedure is touching the aspects of the theory which can not be investigated by feynman diagrams , even summed up to all orders . in this sense
any tachyonic regularization is modeling either effects from a more fundamental theory containing the higgs sector as a low energy limit , or a certain type of nonperturbative effects within the higgs sector itself .
our nonperturbative treatment of the sigma model is independent of an interpretation either as an effective theory or as a fundamental theory , and we will make no assumption in this direction in this paper .
in this section we explain the necessary steps to calculate the nlo corrections to the processes @xmath65 in the @xmath0 expansion . in the case of longitudinal vector boson decay modes ,
we calculate the processes by using the equivalence theorem .
therefore we repace the longitudinal vector bosons by the corresponding would - be goldstone bosons @xmath66 and @xmath67 . to this end
one has to know the two- and three - point functions to that order . to get a physically meaningful quantity one has to define a perturbative renormalization procedure to deal with the usual divergencies . on the other hand one
has to define a scheme to treat the tachyon which is a perturbative artifact of the leading order calculation .
we have seen that the infinite series of feynman diagrams shows up in the propagators as a momentum dependent mass producing a tachyon pole .
the dressed propagators in eq .
[ propa ] are , together with the occurring vertices and the counterterms defined in eqn .
[ lagrenor ] , the building blocks of the diagrams which describe the next - to - leading order effects . to leading order
, only gb propagators appear as internal lines .
it is crucial for the simplicity of the calculation that , with the help of the auxiliary field @xmath34 , the next - to - leading order contributions of the @xmath0 expansion to the higgs propagator can be depicted by a small number of graphs . in the following
, a tachyonic regularization will be defined .
then the diagrammatical part of the calculation will be given together with the renormalization procedure .
the subtractions will be explained in detail for the relevant multi - loop graphs , and the finite , tachyon - free quantum corrections to the two- and three - point functions will be derived .
they will be represented as two - fold integrals .
finally the remaining two dimensional integration is performed numerically , which gives quantitatively the propagator and vertex corrections .
before starting the calculation , we introduce a tachyon - free representation for the occurring propagators @xcite . assuming that its occurrence is not due the failure of the theory under consideration , but of the intermediary expansion technique used @xcite , it is reasonable to circumvent the ill - defined part by an adequate subtraction of the tachyonic pole .
if the @xmath12 theory is only describing a low energy effective theory , some physics at a higher scale will cure the problem .
it is clear that _ a priori _ the tachyon regularization is not unique , and one has to pay the price in the form of an additional ambiguity .
we modify the propagators in eq .
[ propa ] by the following subtraction of the tachyonic pole , which is the minimal way of eliminating the tachyonic pole : @xmath68 @xmath69 is just the residuum of the tachyonic pole . to get a quantitative feeling for the size of the subtracted terms we give some values of the appearing constants in table [ tachsize ] .
it is clear from eq .
[ epspre ] that only a pair of parameters @xmath70 defines a physical situation .
we related @xmath55 to the position of the higgs pole in the complex plane , @xmath71 , by the relation @xmath72 @xcite .
its position is also shown in the table .
.[tachsize ] _ information on the physical and the tachyonic pole at leading order .
the parameter @xmath73 is related to the physical pole in the way explained in the text .
the dimensionful quantities are in tev . _ [ cols="^,^,^,^,^,^",options="header " , ] note that the sum of the @xmath0 graphs and counterterms for a given topology gives an infrared finite result .
thus , to derive the following formulae , it is necessary to introduce an infrared regulator for the individual terms which cancels in the final sum .
as stated above , we are using analytical expressions for the massless three- and four- point functions , @xmath74 , @xmath75 , which are given in an appendix for completeness .
@xmath76 \left[\frac{1}{(p+q)^2}-\frac{1}{p^2}-\frac{q^2}{(p^2)^2 } \bigl(1-\frac{4(p\cdot q)^2}{p^2q^2}\bigr)\right ] \nonumber\\ & = & \frac{1}{v^2 } \int dq \frac{f(q^2)}{q^2 ( 4\pi)^2 } \left [ 1-\frac{2(q\cdot k)^2}{q^2k^2 } -\frac{q^2}{k^2}\log\bigl(\frac{r^2}{q^2}\bigr ) \right ] \nonumber\\ a_4 & = & \frac{1}{v^2 } \int dq \frac{f(q^2)}{q^2 } \left ] ( q^2)^2 d_0(k , q ,-
k ) - \frac{1+\log\bigl(\frac{q^2}{k^2}\bigr)}{8\pi^2 } \right ] \nonumber\\ a_5 & = & \frac{1}{v^2 } \int dq \left [ f(r^2)f(q^2 ) c_0(k , q ) - f^2(q^2 ) \frac{2+\log\bigl(\frac{q^2}{k^2}\bigr)}{(4\pi)^2q^2 } \right ] \nonumber\\ a_6 & = & \frac{1}{v^4 } \int dq \left [ r^2 q^2f(r^2)f(q^2 ) c_0 ^ 2(k , q ) -
f^2(q^2 ) \left(\frac{2+\log\bigl(\frac{q^2}{k^2}\bigr)}{(4\pi)^2}\right)^2 \right ] \nonumber\\ c_2 & = & \frac{1}{v^2 } \int dq \left [ q^2 f(r^2)f(q^2 ) c_0(k , q ) - f^2(q^2 ) \frac{2+\log\bigl(\frac{q^2}{k^2}\bigr)}{(4\pi)^2 } \right ] \nonumber\\ f_3 & = & \frac{1}{v^4 } \int dq \left [ f(r^2)f(q^2 ) c_0(k , q ) \frac{r^2 q^2}{(q - l)^2 } - f^2(q^2 ) \frac{2+\log\bigl(\frac{q^2}{k^2}\bigr)}{(4\pi)^2 } \right]\end{aligned}\ ] ] we will need in the following sums of these expressions taking into account combinatorial factors and signs .
@xmath77 the functions @xmath78 are integrals over minkowski space , parameterized by @xmath79 . after performing the angular integration ,
the number of integrations can be reduced to two . choosing the timelike momentum as @xmath80 , it is obvious that the integrals which belong to two - point functions are isotropic . for the three - point function with on - shell gbs ,
one has a second timelike four - vector @xmath81 .
explicitly , one finds for the angular integration : @xmath82 \nonumber\\ \int \frac{d\omega}{4\pi } \frac{1}{(q - l)^2(q - l+k)^2 } & = & \frac{1}{4q_0\rho s}\log\left[\frac{(q^2)^2-s ( q_0-\rho)^2}{(q^2)^2-s ( q_0+\rho)^2}\right]\end{aligned}\ ] ] thus we end up with two - dimensional integral representations for the relevant graphs contributing to the processes we look for . in principle
the @xmath83 integration is also manageable by complex rotation , but one has to take great care because of the complicated singularity and cut structure of the integrands .
we proceed numerically in the following .
the numerical problems encountered in the calculation of the needed diagrams are twofold .
first of all , one faces convergence problems due to the physical particle poles .
although the massive propagators have a finite width in our case , one still has to use a wick rotation to get a fast numerical convergence .
to do the two dimensional numerical integration : @xmath84 one chooses the contours in the complex @xmath85 plane to avoid the cuts of the logarithms and to stay away from the poles .
they are shown in figure [ contours ] .
for the two - point functions , a simple rotation @xmath86 leads to a smooth integrand ( a ) , for the three point function one has to avoid also the cut of the logarithms present in eq .
[ angint ] ( b ) .
no singularity and cut is crossed for any @xmath87 .
the independence of the numerical result of @xmath88 serves as a test of the stability of the integration .
( 10,3 ) ( 1,0.0 ) ( 6,0.0 ) the procedure gives fast and accurate results as long as the integrand is well behaved at infinity .
this is the second problem of the numerical part of the calculation .
usually in loop calculations one is able to separate the uv divergent part by evaluating the feynman graphs in @xmath89 dimensions .
the regularization leads to manifestly finite integrands .
if , as in our calculation , the regularization has to be done by subtracting divergent counterterms from divergent graphs , the numerical situation is more delicate because of cancellations . to ensure numerical ultraviolet stability ,
it is necessary to arrange the integrand in such a way that as many cancellations as possible are manifest .
for the integrations we used adaptive , deterministic fortran routines .
to derive physical amplitudes , we collect our results for the two- and three- point functions . for the renormalized two - point functions one finds the following expressions : @xmath90 = \left[\begin{array}{cc } \frac{nv^2}{m^2(s ) } + a(s ) & -\sqrt{n}v\left(1-\frac{c(s)}{n}\right)\sqrt{\frac{z_h}{z_\phi } } \\ -\sqrt{n}v\left(1-\frac{c(s)}{n}\right)\sqrt{\frac{z_h}{z_\phi } } & \left ( s + \frac{v^2 b(s)}{n}\right)\frac{z_h}{z_\phi } \end{array}\right ] \nonumber \\
\gamma_{\pi_k\pi_l}(s ) & = & s \left [ 1 + \frac{v^2}{n } d(s ) \right ] \delta_{kl}\end{aligned}\ ] ] the leading order contribution is contained in the momentum dependent mass .
the gbs , labeled by @xmath91 , do not mix with the @xmath92 fields .
the wave function renormalization constants appear because of our definition of @xmath93 .
as mentioned above , we used only the wave function renormalization of the gbs , because we are interested in amplitudes without external higgs particles .
thus it is not necessary to compute @xmath37 . with the methods described in the previous section , it is easy to calculate @xmath94 .
the inverse of the two point function matrix gives the propagators . as explained earlier , the diagonal gb propagators still have their pole at @xmath95 .
one finds : @xmath96 \right\ } \nonumber\end{aligned}\ ] ] the other propagators are : @xmath97 \nonumber \\
\frac{d_{\chi\chi}(s)}{d_{\chi\chi}^0(s ) } & = & \frac{z_h}{z_\phi } \frac{d_{hh}(s)}{d_{hh}^0(s ) } - \frac{1}{n}\bigl[\frac{m^2(s)}{v^2}a(s)-\frac{v^2}{s}b(s)\bigr ] \end{aligned}\ ] ] the three - point functions are : @xmath98 with these ingredients , one can express the amplitudes of the reactions @xmath99 and @xmath100 .
note that for any @xmath101 vertex one gets a multiplicative factor @xmath102 .
the decomposition of the amplitudes into one - particle irreducible contributions is depicted in figure [ mffzzamp ] .
( 14,2.2 ) ( 0.1,0.0 ) ( 4.,.9)@xmath103 ( 4.7,0.0 ) ( 8.7,.9)@xmath104 ( 9.4,0.0 ) @xmath105\right\ } \end{aligned}\ ] ] the amplitudes can be measured directly at a future muon collider , and also can be related to higgs production by gluon fusion at the lhc .
the processes are available also to nnlo in usual perturbation theory , and thus are good quantities to compare the different theoretical approaches .
one has simply to put @xmath26 and @xmath106 gev in our formulae . in figure [ pawfiga ]
we plot the amplitudes of the two processes for several values of the coupling parameterized by @xmath73 , defined in eq .
[ epspre ] .
the yukawa coupling is set to one .
note how the higgs signal evolves from a sharp resonance to the damped regime , as the coupling is getting strong .
= 12.5 cm we compare our results with the perturbative ones in figure [ compare ] .
= 12.5 cm the peaks of the nlo @xmath0 and nnlo perturbation theory result are aligned to allow for a physically meaningful comparison between the two calculations . for on - shell higgs masses below @xmath107 gev the curves practically coincide . from this figure
it is obvious that the results of both approaches agree with each other up to the highest accessible energies of the lhc .
away from the resonance , the perturbative curves which we give are not expected to be a good approximation .
close to the vector boson pair production threshold , the equivalence theorem is not valid anymore , and the curves will be modified by contributions of order of the vector boson masses .
as discussed in earlier publications @xcite , one finds in both approaches a saturation of the peak position .
this is confirmed now also for the process @xmath100 .
the saturation effect is shown in figure [ satur ] .
= 12.5 cm it turns out that the nnlo perturbative result allows a maximal peak position of about 1100 gev , whereas for the @xmath0 result one finds 975 gev . in the case of
@xmath99 one finds 980 gev for the perturbative and 930 gev for the @xmath0 result . in this sense ,
the nonperturbative prediction is phenomenologically more restrictive .
we described in some detail the computation of higgs production processes in the nonperturbative @xmath0 expansion for processes which are relevant for future colliders . we want to stress that our calculation reproduces the perturbative results in the region up to around @xmath108 gev in a very precise manner , which shows the consistency between perturbative methods and the @xmath0 expansion including our tachyon subtraction scheme .
once the calculational tools are developed , the calculation is not more difficult than a two - loop calculation in the higgs sector . beyond @xmath109 gev ,
there are differences between the @xmath0 expansion and perturbation theory concerning the position of the peak maxima , but the implications for higgs - discovery physics are not too large up to @xmath110 tev .
this leads to the conclusion that the higgs sector is theoretically well understood for the entire kinematic region relevant for higgs searches at lhc @xcite .
a saturation of the peak position at strong coupling seems to be a general feature of a heavy higgs boson .
note that the @xmath0 approach disentangles the uncertainties from the renormalization scheme which is simply not present , from the cutoff effects due to the tachyonic pole .
the effect of the tachyonic regularization becomes more important as the coupling is increased , modeling nondecoupling effects from a more fundamental theory containing the higgs sector as an effective theory .
the following formulae @xcite are used to get the integral representations of the @xmath0 graphs of eq .
[ integralreps ] . with its help
the appearing two- and three - loop topologies can be cast into the form of one - loop integrals .
we use the notation @xmath113 for the mandelstam variables which are defined by the external momenta @xmath114 : @xmath115\biggr\ } \hspace{2 cm } \end{aligned}\ ] ] whereby , @xmath116 the cut of the logarithm lies along the negative real axis .
an infinitesimal replacement @xmath117 is included to stay on the correct riemann sheet if an argument vanishes .
both formulae allow for analytical continuation of an external momentum which is important for our numerical purposes .
this was checked by recalculating several known two - loop diagrams with the same methods as used for the computation of the 1/n graphs .
a. ghinculov and j.j .
van der bij , nucl . phys .
* b436 * ( 1995 ) 30 ; a. ghinculov , nucl . phys .
b455 ( 1995 ) 21 ; phys .
b337 ( 1994 ) 137 ; ( e ) b346 ( 1995 ) 426 .
p. n. maher , l. durand and k. riesselmann , phys .
d48 ( 1993 ) 1061 ; ( e ) d52 ( 1995 ) 553 ; l. durand , b.a .
kniehl and k. riesselmann , phys .
d51 ( 1995 ) 5007 ; phys .
72 ( 1994 ) 2534 ; ( e ) phys .
74 ( 1995 ) 1699 ; a. frink , b.a .
kniehl , d. kreimer , k. riesselmann , phys .
d54 ( 1996 ) 4548 . | we show in this paper that the @xmath0 expansion is a reliable tool to calculate the properties of a heavy higgs boson .
the 1/n expansion sums up all orders in perturbation theory , and therefore avoids the renormalization scheme dependence of the conventional perturbative approach .
it is explained how effects due to the landau pole of the higgs sector are isolated and subtracted , and how to perform actual calculations , by computing the higgs line shape for the processes @xmath1 at next - to - leading order in the @xmath0 expansion .
the results are compared to the perturbative results to show the agreement between the perturbative and the nonperturbative approach for higgs masses up to 1 tev .
we conclude that the theoretical predictions for higgs observables are well under control for the entire kinematical region of the lhc . = 14pt = 6.0 in = 8.5 in -0.25truein 0.30truein 0.30truein lapth 692 - 98 + um - th 98 - 12 + cern - th/98 - 244 + august 1998 * * * new developments in the @xmath0 expansion and nonperturbative higgs physics * + thomas binoth@xmath2 and adrian ghinculov@xmath3 + _
@xmath2laboratoire dannecy - le - vieux de physique thorique lapp , _ + _ chemin de bellevue , b.p . 110 , f-74941 , annecy - le - vieux , france _ + _ @xmath4randall laboratory of physics , university of michigan , _ + _ ann arbor , michigan 48109 - 1120 , usa _
+ _ @xmath5cern , 1211 geneva 23 , switzerland _ + |
in 1917 einstein introduced the cosmological constant into his field equations in order to obtain a static cosmological model since , as is well known , without the cosmological term his field equations admit only non - static solutions . after the discovery of the redshift of galaxies and its explanation as being due to the expansion of the universe , einstein regretted his introduction of the cosmological constant .
recently , there has been much interest in the cosmological term in context of quantum field theories , quantum gravity , super - gravity theories , kaluza - klein theories and the inflationary - universe scenario . shortly after einstein s general theory of relativity weyl , in 1918 , suggested the first so - called unified field theory based on a generalization of riemannian geometry .
in retrospect , it would seem more appropriate to call weyl s theory a geometrized theory of gravitation and electromagnetism ( just as the general theory was a geometrized theory of gravitation only ) , rather than a unified field theory .
it is not quite clear to what extent the two fields have been unified , even though they acquire ( different ) geometrical significances in the same geometry . the theory was never taken seriously because it was based on the concept of non integrability of length transfer , and , as pointed out by einstein , this implies that spectral frequencies of atoms depend on their past histories and therefore have no absolute significance .
nevertheless , weyl s geometry provides an interesting example of non - riemannian connections , and recently folland @xcite has given a global formulation of weyl manifolds thereby clarifying considerably many of weyl s basic ideas . in 1951
lyra @xcite proposed a modification of riemannian geometry by introducing a gauge function into the structureless manifold , as a result of which the cosmological constant arises naturally from the geometry .
this bears a remarkable resemblance to weyl s geometry .
but in lyra s geometry , unlike weyl s , the connection is metric preserving as in remannian ; in other words , length transfers are integrable .
lyra also introduced the notion of a gauge and in the `` normal '' gauge the curvature scalar in identical to that of weyl .
in consecutive investigations sen @xcite , sen and dunn @xcite proposed a new scalar - tensor theory of gravitation and constructed an analog of the einstein field equations based on lyra s geometry . it is thus possible @xcite to construct a geometrized theory of gravitation and electromagnetism much along the lines of weyl s `` unified '' field theory without , however , the inconvenience of non - integrability length transfer .
+ halford @xcite has pointed out that the constant vector displacement field @xmath1 in lyra s geometry plays the role of cosmological constant @xmath0 in the normal general relativistic treatment .
it is shown by halford @xcite that the scalar - tensor treatment based on lyra s geometry predicts the same effects , within observational limits , as the einstein s theory .
several authors sen and vanstone @xcite , bhamra @xcite , karade and borikar @xcite , kalyanshetti and wagmode @xcite , reddy and innaiah @xcite , beesham @xcite , reddy and venkateswarlu @xcite , soleng @xcite , have studied cosmological models based on lyra s manifold with a constant displacement field vector . however
, this restriction of the displacement field to be constant is merely one of convenience and there is no _ a priori _ reason for it .
beesham @xcite considered frw models with time dependent displacement field .
he has shown that by assuming the energy density of the universe to be equal to its critical value , the models have the @xmath2 geometry .
singh and singh @xcite , singh and desikan @xcite have studied bianchi - type i , iii , kantowaski - sachs and a new class of cosmological models with time dependent displacement field and have made a comparative study of robertson - walker models with constant deceleration parameter in einstein s theory with cosmological term and in the cosmological theory based on lyra s geometry .
soleng @xcite has pointed out that the cosmologies based on lyra s manifold with constant gauge vector @xmath3 will either include a creation field and be equal to hoyle s creation field cosmology @xcite or contain a special vacuum field which together with the gauge vector term may be considered as a cosmological term . in the latter case
the solutions are equal to the general relativistic cosmologies with a cosmological term .
the einstein s field equations are a coupled system of highly nonlinear differential equations and we seek physical solutions to the field equations for their applications in cosmology and astrophysics . in order to solve the field equations we normally assume a form for the matter content or that space - time admits killing vector symmetries @xcite .
solutions to the field equations may also be generated by applying a law of variation for hubble s parameter which was proposed by berman @xcite .
in simplest case the hubble law yields a constant value for the deceleration parameter .
it is worth observing that most of the well - known models of einstein s theory and brans - deke theory with curvature parameter @xmath4 , including inflationary models , are models with constant deceleration parameter . in earlier literature
cosmological models with a constant deceleration parameter have been studied by several authors @xcite .
but redshift magnitude test has had a chequered history . during the 1960s and the 1970s ,
it was used to draw very categorical conclusions .
the deceleration parameter @xmath5 was then claimed to lie between @xmath6 and @xmath7 and thus it was claimed that the universe is decelerating .
today s situation , we feel , is hardly different .
observations ( knop et al .
@xcite ; riess et al . , @xcite ) of type ia supernovae ( sne )
allow to probe the expansion history of the universe .
the main conclusion of these observations is that the expansion of the universe is accelerating .
so we can consider the cosmological models with variable cosmological term and deceleration parameter .
the readers are advised to see the papers by vishwakarma and narlikar @xcite and virey et al . @xcite and references therein for a review on the determination of the deceleration parameter from supernovae data .
recently pradhan and otarod @xcite have studied the universe with time dependent deceleration parameter and @xmath0-term in presence of perfect fluid . motivated with the situation discussed above , in this paper , we study a new class of lrs bianchi - i cosmological models in lyra geometry by considering a time dependent deceleration parameter in an expanding universe . this paper is organized as follows .
the metric and the field equations are presented in section 2 . in section 3
we deal with a general solution .
the sections 4 , 5 , and 6 contain the three different cases for the solutions in exponential , polynomial and sinusoidal forms respectively .
finally in section 7 concluding remarks will be given .
we consider lrs bianchi type i space - time @xmath8 where @xmath9 .
we take a perfect fluid form for the energy momentum tensor [ eq2 ] @xmath10 together with co - moving coordinates @xmath11 , where @xmath12 . + the field equations in normal gauge for lyra s manifold , as obtained by sen [ 3 ] are [ eq3 ] @xmath13 where @xmath3 is a time - like displacement field vector defined by @xmath14 and other symbols have their usual meaning as in riemannian geometry . here we want to mention the fact that the ansatz choosing the coordinate system with matter require the vector field happens to be in the required form exactly in the matter comoving coordinates . the essential difference between the cosmological theories based on lyra geometry and the riemannian geometry lies in the fact that constant vector displacement field @xmath15 arises naturally from the concept of gauge in lyra geometry where as cosmological constant @xmath0 was introduced in _ adhoc _
fashion in the usual treatment . now the field equations can be set up and one obtains @xmath16 @xmath17
@xmath18 @xmath19 the energy conservation equation is @xmath20 \left(\frac{\dot a}{a } + \frac{2 \dot b}{b}\right ) = 0\ ] ] where @xmath21 . here and in what follows
a prime and a dot indicate partial differentiation with respect to @xmath22 and @xmath23 respectively . equations ( [ eq4])-([eq7 ] ) are four equations in five unknowns @xmath24 and @xmath25 . for complete determinacy of the system one extra condition is needed .
one way is to use an equation of state .
the other alternative is a mathematical assumption on the space - time and then to discuss the physical nature of the universe . in this paper
we confine ourselves to assume an equation of state @xmath26
equation ( [ eq5 ] ) , after integration , yields @xmath27 where @xmath28 is an arbitrary function of @xmath22 .
equations ( [ eq4 ] ) and ( [ eq6 ] ) , with the use of equation ( [ eq10 ] ) , reduces to @xmath29 setting @xmath30 equation ( [ eq11 ] ) yields on integration @xmath31 where @xmath32 is an arbitrary function of @xmath23 . with the help of equation ( [ eq13 ] ) ,
equation ( [ eq10 ] ) becomes @xmath33 now the metric ( [ eq1 ] ) takes the form @xmath34,\ ] ] where @xmath35 .
equations ( [ eq4 ] ) and ( [ eq4 ] ) give @xmath36 @xmath37 using equation ( [ eq4 ] ) and eliminating @xmath38 from equations ( [ eq16 ] ) and ( [ eq17 ] ) we have @xmath39 now the expressions for the energy density and the pressure are given by @xmath40\ ] ] the function @xmath32 remains undetermined . to obtain its explicit dependence on @xmath23 , one may have to introduce additional assumptions .
accordingly we assume the deceleration parameter to be variable and set @xmath41 where @xmath42 is the hubble parameter .
the above equation may be rewritten as @xmath43 the general solution of eq .
( [ eq21 ] ) is given by @xmath44 where @xmath45 is an integrating constant . in order to solve the problem completely
, we have to choose @xmath46 in such a manner that eq .
( [ eq22 ] ) be integrable . without loss of generality , we consider @xmath47 which does not effect the nature of generality of solution . hence from eqs .
( [ eq22 ] ) and ( [ eq23 ] ) , we obtain @xmath48 of course the choice of @xmath49 , in eq .
( [ eq24 ] ) , is quite arbitrary but , since we are looking for physically viable models of the universe consistent with observations , we consider the following case :
let us consider @xmath50 , where @xmath51 is arbitrary constant . in this case on integration of eq .
( [ eq24 ] ) gives the exact solution @xmath52 where @xmath53 is an arbitrary constant .
using eqs .
( [ eq9 ] ) and ( [ eq25 ] ) in ( [ eq18 ] ) and ( [ eq19 ] ) , we obtain expressions for displacement field @xmath15 , pressure @xmath54 and energy density @xmath25 as @xmath55 @xmath56.\ ] ] vs time with parameters @xmath57 , @xmath58 , and @xmath59,width=302,height=302 ] from eq .
( [ eq25 ] ) , since scale factor can not be negative , we find @xmath32 is positive if @xmath60 . from figure 1 , it can be seen that in the early stages of the universe , i.e. , near @xmath61 , the scale factor of the universe had been approximately constant and had increased very slowly . at specific time
the universe had exploded suddenly and expanded to large scale .
this is consistent with big bang scenario . from eq .
( [ eq26 ] ) , it is observed that @xmath62 is a decreasing function of time .
as mentioned earlier the constant vector displacement field @xmath63 in lyra s geometry plays the role of cosmological constant @xmath0 in the normal general relativistic treatment and the scalar - tensor treatment based on lyra s geometry predicts the same effects , within observational limits , as the einstein s theory .
recent cosmological observations ( garnavich et al .
@xcite , perlmutter et al .
@xcite , riess et al .
@xcite , schmidt et al .
@xcite ) suggest the existence of a positive cosmological constant @xmath0 with the magnitude @xmath64 .
these observations on magnitude and red - shift of type ia supernova suggest that our universe may be an accelerating one with induced cosmological density through the cosmological @xmath0-term . in our model
, it is seen that @xmath15 plays the same role as cosmological constant and preserves the same character as @xmath0-term , in turn with recent observations .
the expressions for @xmath62 and @xmath25 can not be determined for the empty universe @xmath65 and stiff matter @xmath66 models . in this case , the ricci scalar is obtained as @xmath67 from above equation , we see that the ricci scalar remains positive for @xmath68 the expansion and shear scalar are given by @xmath69 the model represents uniform expansion as can be seen from eq .
( [ eq29 ] ) .
the flow of the fluid is geodetic with the acceleration vector @xmath70 .
let @xmath71 , where @xmath72 and @xmath73 are constants .
+ in this case , on integrating , eq .
( [ eq24 ] ) gives the exact solution @xmath74 where @xmath75 , @xmath76 and @xmath77 are arbitrary constants .
using eqs .
( [ eq9 ] ) and ( [ eq30 ] ) in ( [ eq18 ] ) and ( [ eq19 ] ) , we obtain the expressions for displacement field @xmath15 , pressure @xmath54 and energy density @xmath25 as @xmath78}{(1 - \gamma ) ( \alpha_{1}t^{2 } + \alpha_{2}t + \alpha_{3})^{2}},\ ] ] @xmath79}{(1 - \gamma ) ( \alpha_{1}t^{2 } + \alpha_{2}t + \alpha_{3})^{2}}.\ ] ] vs time with parameters @xmath80 , @xmath81 , @xmath82 and @xmath59,width=302,height=302 ] from eq .
( [ eq30 ] ) , we note that @xmath83 for @xmath84 if @xmath75 , @xmath76 and @xmath77 are positive constants .
figure @xmath85 , shows that the scale factor is a decreasing function of time , implying that our universe is expanding .
( [ eq31 ] ) shows that @xmath86 for all times as @xmath87 and is a decreasing function of time , characteristically similar to @xmath0 in einstein s theory of gravitation . in this model
, @xmath15 plays the role as cosmological constant and it preserves the same character as @xmath0-term .
this is consistent with recent observations ( garnavich et al .
@xcite , perlmutter et al .
@xcite , riess et al .
@xcite , schmidt et al .
@xcite ) . a negative cosmological constant adds to the attractive gravity of matter ; therefore , universe with a negative cosmological constant is invariably doomed to re - collapse .
a positive cosmological constant resists the attractive gravity of matter due to its negative pressure . for most of the time ,
the positive cosmological constant eventually dominates over the attraction of matter and drives the universe to expand exponentially .
the expressions for @xmath62 and @xmath25 can not be determined for the empty universe @xmath65 and stiff matter @xmath66 models . in this case , the ricci scalar is obtained as @xmath88 } { ( \alpha_{1}t^{2 } + \alpha_{2}t + \alpha_{3})^{2}}.\ ] ] from eq .
( [ eq33 ] ) we observe that ricci scalar remains positive if @xmath89 .
this condition also implies that @xmath90 .
the scalar of expansion is given by @xmath91 it is of the interest to note that all physical parameters in our models are defined at @xmath61 and we do not have any singularity .
let @xmath92 , where @xmath15 is constant .
+ in this case , on integrating , eq .
( [ eq24 ] ) gives the exact solution @xmath93 where @xmath94 , @xmath95 and @xmath96 are constants . using eqs .
( [ eq9 ] ) and ( [ eq35 ] ) in ( [ eq18 ] ) and ( [ eq19 ] ) , we obtain the expressions for displacement field @xmath15 , pressure @xmath54 and energy density @xmath25 as @xmath97}{3(1 - \gamma)q^{2}},\ ] ] @xmath98}{(1 - \gamma)q^{2}},\ ] ] where @xmath99 @xmath100 vs time with parameters @xmath101 , @xmath102 , @xmath103 , @xmath104 , and @xmath59,width=302,height=302 ] in this case the ricci scalar and the scalar of expansion are obtained as @xmath105 } { \left(m\sin(\beta t ) + n\cos(\beta t ) + \beta_{1}\right)^{2}},\ ] ] @xmath106 } { m\sin{\beta t } + n\cos{beta t } + \beta-{1}}.\ ] ] from the figure 3 , we note that at early stage of the universe , the scale of the universe increases gently and then decreases sharply , and after wards it will oscillate for ever .
we must mention here that the oscillation takes place in positive quadrant this has physical meaning . since , in this case , we have many alternatives for choosing values of @xmath94 , @xmath95 , @xmath15 , @xmath96
, it is just enough to look for suitable values of these parameters , such that the physical initial and boundary conditions are satisfied .
in this paper we have obtained exact solutions of sen equations in lyra geometry for time dependent deceleration parameter .
the nature of the displacement field @xmath107 and the energy density @xmath38 have been examined for three cases ( i ) exponential form ( ii ) polynomial form and ( iii ) sinusoidal form .
the solutions obtained in sections @xmath108 , @xmath109 and @xmath110 are to our knowledge quite new . here the displacement field @xmath15 plays the role of a variable cosmological term @xmath0 .
recently there is an upsurge of interest in scalar fields in general relativity and alternative theories of gravitation in the context of inflationary cosmology .
therefore the study of cosmological models in lyra geometry are relevant for inflationary models .
further the space dependence of the displacement field @xmath15 is important for inhomogeneous models in the early stage of the evolution of the universe .
besides , the implication of lyra s geometry for bodies of astrophysical interest is still an open question .
the problem of equations of motion and gravitational radiation need investigation .
one of the authors ( a. pradhan ) would like to thank the inter - university centre for astronomy and astrophysics , pune , india for providing hospitality where part of this work was carried out .
000 g. folland ( 1970 ) _ j. diff .
geom . _ * 4 * 145 . g. lyra ( 1951 ) _ math . z. _ * 54 * 52
. d.k . sen ( 1957 )
_ z. f@xmath111r phys . _ * 149 * 311 .
d.k . sen and k.a .
dunn ( 1971 ) _ j. math
* 12 * 578 .
w.d . halford ( 1970 ) _ austr .
j. phys . _
* 23 * 863 .
halford ( 1972 ) _
* 13 * 1699 .
d.k . sen and j.r .
vanstone ( 1972 ) _ j. math
. phys . _ * 13 * 990 .
bhamra ( 1974 ) _ austr .
j. phys . _
* 27 * 541 .
karade and s.m .
borikar ( 1978 ) _ gen .
gravit . _ * 9 * 431 .
kalyanshetti and b.b .
waghmode ( 1982 ) _ gen .
* 14 * 823 .
reddy and p. innaiah ( 1986 ) _ astrophys .
space sci .
_ * 123 * 49 .
a. beesham ( 1986 ) _ astrophys .
space sci .
_ * 127 * 189 .
reddy and r. venkateswarlu ( 1987 ) _ astrophys .
space sci . _ * 136 * 183 .
h.h . soleng ( 1987 ) _ gen . rel . gravit . _
* 19 * 1213 .
a. beesham ( 1988 ) _ austr .
_ * 41 * 833 .
t. singh and g.p .
singh ( 1991a ) _
phys . _ * 32 * 2456 ; + t. singh and g.p .
singh ( 1991b ) _ il .
nuovo cimento _ * b106 * 617 ; + t. singh and g.p .
singh ( 1992 ) _ int .
phys . _ * 31 * 1433 ; + t. singh , and g.p .
singh ( 1993 ) _ fortschr .
phys . _ * 41 * 737 .
g.p . singh and k. desikan ( 1997 ) _ pramana - journal of physics _ , * 49 * 205 . f. hoyle ( 1948 ) _ monthly notices roy .
. soc _ * 108 * 372 ; + f. hoyle and j.v .
narlikar ( 1964 ) _ proc .
ser . a mathematical & physical sciences , * 278 * 465 ; + f. hoyle and j.v . narlikar ( 1964 )
ser . a mathematical & physical sciences * 277 * 1 .
d. kramer et al .
( 1980 ) _ exact solutions of einstein s field equations _ , cambridge university press , cambridge .
berman ( 1983 ) _
nuovo cimento _
* b74 * 182 . m.s .
berman and f. m. gomide ( 1988 ) _ gen .
* 20 * 191 ; + v.b .
johri and k. desikan ( 1994 ) _ pramana - journal of physics _ * 42 * 473 ; + s.d .
maharaj and r. naidoo ( 1993 ) _ astrophys . space sc . _ * 208 * 261 ; + a. beesham ( 1993 ) _ phys .
rev . _ * d48 * 3539 ; + a. pradhan and a.k .
vishwakarma ( 2000 ) _ sujst _ * xii * sec .
b 42 ; + a. pradhan , v.k . yadav and i. chakrabarty ( 2001 )
* d10 * 339 ; + a. pradhan and i. aotemshi ( 2002 ) _ int . j. mod . phys . _ * d9 * 1419 ; + a. pradhan and a.k . vishwakarma ( 2002 )
_ * d8 * 1195 ; + a. pradhan and a.k .
vishwakarma ( 2004 ) _ j. geom
. phys . _ * 49 * 332 .
knop et al .
( 2003 ) _ astrophys .
j. _ * 598 * 102 . a.g .
riess et al . ( 2004 ) _ astrophys .
j. _ * 607 * 665 .
vishwakarma and j.v . narlikar ( 2005 )
_ * d14 * 345 .
virey et al .
( 2005 ) _ astro - ph/0502163_. a. pradhan and s. otarod ( 2006 ) _ astrophys .
space sci .
_ in press , _ gr - qc/0607129_. p.m. garnavich et al .
( 1998a ) _ astrophys .
j. _ * 493 * ; + p.m. garnavich et al . ( 1998b ) * 509 * 74 .
s. perlmutter et al . ( 1997 ) _ astrophys .
j. _ * 483 * 565 ; + s. perlmutter et al . ( 1998 ) _ nature _ * 391 * 51 ; + s. perlmutter et al . ( 1999 ) _ astrophys . j. _ * 517 * , 565 .
riess et al . ( 1998 ) _ astron .
j. _ * 116 * 1009 .
schmidt et al . ( 1998 ) _ astrophys .
j. _ * 507 * 46 . | lrs bianchi type - i models have been studied in the cosmological theory based on lyra s geometry .
a new class of exact solutions has been obtained by considering a time dependent displacement field for variable deceleration parameter models of the universe .
we have compared our models with those of einstein s field theory with the cosmological term @xmath0 .
our frame of reference is restricted to the recent ia observations of supernovae .
some physical behaviour of the models is also examined in the presence of perfect fluids .
+ + + + + key words : cosmology ;
l r s bianchi type - i models ; lyra geometry + pacs number : 98.80.-k , 98.80.jk |
the large promise of quantum - optical technologies to enable secure communication and novel computation architectures sets stringent targets for the quality of photon sources and detectors . in order to meet these demands many efforts
are currently devoted to realizing bright sources of single photons @xcite .
these developments at the same time require novel nanophotonic engineering designs around emitters to enhance light - matter interaction strength as well as indefinitely stable two - level systems that neither bleach , blink , nor spectrally jump . on the nanophotonic engineering side , many different systems have been proposed to control whereto , how fast and with which polarization an emitter emits provided that one manages to locate it exactly in the right location .
these systems include whispering - gallery - mode cavities @xcite , micropillars and cylindrical wires @xcite , photonic crystal microcavities @xcite , photonic crystal waveguides near cut - off @xcite , anderson localizing systems @xcite , as well as ultrabroadband plasmonic waveguides and antennas @xcite . essentially , all these techniques modify the photonic environment of an emitter via the local density of optical states ( ldos ) @xcite .
the ldos quantifies the light - matter interaction strength that appears in fermi s golden rule for spontaneous emission . placing the emitter at a position where the ldos is enhanced ,
first implies a much higher fluorescence decay rate .
second , if the ldos enhancement is due to a select set of designed modes as in a cavity or waveguide , the enhanced rate is accompanied by extraction of photons preferentially via these enhanced modes . thus , photonic engineering promises optimum brightness , dynamics , and collection efficiency . as regards the choice of emitter , a wide variety of systems have been used .
unfortunately , all choices appear to carry large disadvantages when going beyond pilot studies : most dye molecules photobleach @xcite , ii - vi quantum dots @xcite blink as well as bleach , and many systems , such as iii - v emitters , only have desirable properties when cooled to cryogenic temperatures @xcite . a promising candidate to provide an indefinitely stable source @xcite that furthermore allows room - temperature spin - control
@xcite is the nitrogen - vacancy ( nv ) color center in diamond . when one chooses diamond nv centers as emitters for quantum optics in nanophotonic devices , one can either aim to manipulate emission by fabricating photonic structures directly in diamond @xcite , or one can assemble diamond nanocrystals with photonic structures of a different material @xcite
recently , pick - and - place strategies @xcite were reported in which a single nanoparticle from a diluted powder of diamond nanocrystals dispersed on a substrate is selected and pushed to a desired location by a scanning probe , such as an atomic - force - microscopy tip , or a manipulator in a scanning electron microscope .
reports of this technique span from the coupling of nanodiamonds to photonic - crystal cavities @xcite to coupling to plasmonic antennas and nanowires @xcite . as a variation of the pick - and - place strategy ,
a few groups recently developed so called ` scanning - emitter ' near - field microscopes in which nanosources are not deposited irreversibly inside a nanostructure , but actually remain attached to a scanning- probe @xcite .
the advantage of such a scanning probe approach is that one can first construct a full map of the ldos using near - field lifetime imaging to determine where one should ultimately place the nanosource @xcite .
microscopy with a light - source as a near - field tip is interesting as a microscopy technique , but only viable if the nanosource is indefinitely stable and does not blink , for which nanodiamonds appear the sole candidate @xcite .
moreover nanodiamond scanning probes offer the possibility to create nanometer sized local magnetometry probes read out optically by nv spectroscopy @xcite . in view of the scanning probe microscopy and assembly efforts that seek to combine nanodiamonds and photonic structures
, one question stands out as of key importance : given an ensemble of individual nanodiamonds , how does one recognize the ideal nanocrystal ?
this question is especially relevant given that both fabrication of the photonic structure _ and _ the intended scanning - probe procedure is highly laborious @xcite .
naively , one might think that all nanocrystals of subwavelength size that contain a single nv center will be equally suited , since an nv center is a defect of uniquely defined composition and geometry in the diamond crystalline matrix .
however , many workers on diamond nanocrystals have established that nv centers in diamond nanocrystals actually show a distribution of photophysical properties , such as brightness , stability , and decay rates @xcite .
this distribution is usually ascribed to the fact that even though all nv centers are expected to have unit quantum efficiency and the same oscillator strength , these identical unit - quantum - efficiency sources are each differently placed inside their nanoscopic high - index diamond grains @xcite .
this is anticipated to cause different decay rates , due to the fact that the ldos even in an isolated nanoscopic object varies as a function of position and dipole orientation @xcite . in this paper
we address the question how to recognize the ideal nanocrystal from an ensemble of nanocrystals on basis of the requirement that the ideal nanocrystal must contain a single nv@xmath0 center that fluorescences with a high quantum yield , so that it can be useful as a reporting probe of ldos in nanophotonic systems .
we report that the commonly made assumption that nv centers are unit - efficiency dipoles randomly distributed in high index nanoscopic objects is not reconcilable with measured brightness and decay rate histograms data for nanocrystals in the frequently used size ranges around 2050 nm and 50150 nm . on basis of the wide distribution of brightness and rate that can not be explained by ldos variations between nanoparticles alone
we conclude that both the radiative and nonradiative rates are broadly distributed . to quantify this distribution ,
we have performed experiments in which we measure emission rate changes of individual nanodiamonds as we controllably vary the ldos of their environment . for small nanocrystals
we find a decay rate distribution indicative of low - quantum - efficiency emitters .
furthermore , consistent with this low quantum efficiency , we find no evidence that small single nv centers are responsive ldos probes when applying moderate ldos variations .
for the larger nanocrystals ( 100 nm diameter ) , we for the first time manage to induce reversible changes of up to 25% in the total decay rate of single nv centers using a calibrated ldos change induced by a nanomechanically moved mirror . on basis of these measurements
we argue that the apparent quantum efficiency of nanodiamonds in the size range 50150 nm ranges widely from about 10% to 90% .
this wide range of quantum efficiencies implies that prior to constructing a photonic structure , it is necessary to screen the nanocrystals using a calibrated quantum- efficiency measurement on the single nv center level , as shown in this paper .
this paper is organized in the following manner
. first we report on our experimental methods in section [ sec : setup ] .
next , we discuss the distributions of brightnesses and rates for large nanocrystals in section [ subsec : distriblarge ] and for small nanocrystals in section [ subsec : distribsmall ]
. finally we report on efforts to tune rates of nv centers in small diamonds by controlled ldos changes .
the two methods used are liquid tuning of ldos in section [ sec : liquidldos ] , and nanomechanically approaching a mirror as reported in section [ sec : mirrorldos ] .
since single nv centers in nanodiamonds are comparatively dim emitters that require high excitation powers , it is essential to avoid background fluorescence when performing fluorescence microscopy .
therefore , we use intrinsically low - fluorescent quartz coverslips ( esco products ) as sample substrates that were cleaned by @xmath1 sonication in water followed by a @xmath1 bath in base piranha ( nh@xmath2(aq.,@xmath3):h@xmath4o@xmath4(aq.,@xmath3):h@xmath4o mixed at ratio 1:1:5 , at @xmath5c ) . in our work it is essential that we can unambiguously pinpoint the position of nanodiamonds containing nv centers in the course of the measurement , so that we can revisit the same color center after , for instance , material - deposition steps . to this end
we define an array of markers on top of the coverslips using electron - beam lithography . to remove any organic residues after the lift - off process
, we treated the coverslips with a mild @xmath6-plasma descum ( oxford instruments plasmalab 80 + , using a 5mtorr , 25sccm @xmath6 plasma ) . as sources for nv centers , we use solutions of monocrystalline synthetic nanodiamonds ( microdiamant msy ) with a narrow size tolerance obtained from microdiamant ag , lengwil , switzerland .
we note that these nanoparticles have been used recently by a large number of groups in experiments that rely on fluorescence @xcite .
we use these nanodiamonds exactly in the manner as described by schietinger , i.e. without further washing or purification steps @xcite .
we prepared samples from diamond slurries with two different size distributions by spin - coating aqueous solutions of nanodiamonds that are diluted to a concentration of 1% of the as - received stock solutions on the cleaned and patterned coverslips .
the first type of sample , from here on referred to as ` 25 nm diamond sample ' was made from microdiamant msy 00.05 , which nominally has sizes from 0 to 50 nm .
these diamonds have a median diameter of 26 nm , with fewer than 1% of the diamonds above 50 nm in size , according to particle sizing performed on this batch of diamonds by the manufacturer .
the second type of sample ( ` 100 nm diamond sample ' ) was made from microdiamant msy 00.2 , which nominally has crystal sizes ranging from 0 to 200 nm .
this ensemble has crystals with median diameter 108 nm , and fewer than 1% of particles above 175 nm in size , again according to the size - distribution histogram supplied by the manufacturer . scanning electron microscope ( sem ) inspection indicated a size distribution of nanodiamonds consistent with these specifications .
the average density of spin - coated nanodiamonds on the coverslips as checked through the sem images was about 1 to 8@xmath7 depending on the sample , assuring that only a few nanodiamonds are illuminated at the diffraction - limited focus of the objective .
only a small fraction of the nanodiamonds are fluorescent , and an even smaller fraction fluoresces due to an nv center .
we identify a diamond as containing at least one nv center , if we can clearly identify the characteristic zero phonon line ( zpl ) @xcite in its emission spectrum according to the criterion we specify in section [ single_nv_charac ] . with this criterion , for 100 nm nanodiamonds at an average density of 1@xmath7
, we found on average one nv center in an area equal to @xmath8@xmath9 , while for 25 nm nanodiamonds at an average density of 5@xmath7 we found on average 2 nv centers in a @xmath10@xmath11 area .
these numbers translate to identifying fewer than 0.05% of the 100 nm nanodiamonds , and fewer than 0.001% of the 25 nm diamonds as containing a ( single ) nv center beyond doubt .
schietinger @xcite have reported a higher density of nv centers of about 1% for msy 00.05 nanodiamonds . a difference in reported nv center densities could be either due to a different degree of strictness in labeling a fluorescent emitter as an nv center beyond doubt ( see section [ single_nv_charac ] for our criteria ) , or alternatively to batch - to - batch variations in nanodiamond slurries .
dry objective .
the emitted fluorescence can be guided to spectrometer or to apds to measure the lifetime and second order correlation function @xmath12 of the emitted photons .
[ fig : setup ] , width=302 ] the optical setup , sketched in figure [ fig : setup ] , consists of an inverted confocal fluorescence microscope equipped with a sample scanning piezo stage .
nv centers are optically excited at 532 nm using either a frequency - doubled nd : yag pulsed laser ( time - bandwidth , 10mhz repetition rate ) or a continuous - wave ( cw ) diode laser ( cni ) .
the laser beam is focused through the coverslip to a diffraction - limited spot on top of the sample , using a @xmath13 dry objective with a numerical aperture of 0.9 ( nikon cfi plan fluor ) .
the same objective collects the luminescence and guides it through a long - pass filter ( 580 nm cut - off ) before it is imaged on a ccd camera ( nikon ds - qi1mc ) or confocally detected on one or two avalanche photodiodes ( apd ) with single - photon sensitivity ( both apds : i d quantique id100 - 20uln ) .
the apds in combination with a sub - nanosecond - resolution 16 channel correlator ( becker & hickl , dpc-230 ) allow for single - photon counting .
the correlator can perform time - correlated single - photon - counting ( tcspc ) lifetime measurements by correlation of detection events with laser pulse arrival times , or photon - photon correlations using multiple apds .
we use a second apd in a hanbury - brown and twiss ( hbt ) configuration to measure photon correlation statistics ( antibunching ) at cw excitation .
single - nanocrystal spectra are collected using an imaging spectrometer ( spectrapro 2300i ) equipped with a thermoelectrically cooled back - illuminated si ccd array detector ( princeton instruments pixis:100b ) . of an nv center obtained by cw excitation .
the zero dip in @xmath12 curve confirms the single nature of the center .
( d ) background - corrected decay traces of photons emitted from an nv center ( red circles , total acquisition time 20 s ) with a single - exponential fit ( black curve ) to extract the lifetime .
[ fig : fig_1 ] ] in order to find nanodiamonds containing nv centers , the sample is first pumped with wide - field laser illumination and the fluorescence is imaged on the ccd camera . under these conditions
, any potential nv center will appear as a diffraction - limited fluorescent spot .
subsequently , we switch to diffraction - limited illumination and position the potential candidate at the laser focus using a piezostage ( figure [ fig : fig_1](a ) ) .
the fluorescence intensity of the emitter is quantified by integration over a fixed ccd area which is indicated by the green line in figure [ fig : fig_1](a ) . at this stage , we spectrally resolve the fluorescence from the isolated emitter with the spectrometer to specifically find the characteristic zero phonon line of nv centers at a wavelength of about 638 nm ( figure [ fig : fig_1](b ) ) @xcite .
on basis of the zpl , we either discard the emitter ( absence of zpl ) , or identify the nanodiamond as having one or more nv centers ( zpl well above the detector and photon - counting noise ) .
for each identified nv center , we record a second - order correlation function @xmath14 to verify that the emission originates from a single center . to this end
, we pump the center with the cw laser at a power of about 1mw to achieve count rates on the order of about @xmath15 counts per second on each apd and integrate for about 1000s to obtain reasonable correlation statistics of around 30 coincidences per 658ps bin .
figure [ fig : fig_1](c ) shows a typical background - subtracted @xmath14 curve , with a clearly resolved minimum at zero delay @xmath16 that is well below 0.5 , indicating that emission is from a single nv center . here
, the background is subtracted via the procedure reported in @xcite , based on a measurement directly next to the nv center where we expect identical fluorescence background from the substrate .
the background measured only from the substrate shows a poissonian emission statistics .
the bunching in the @xmath14 in figure [ fig : fig_1](c ) at longer delay times ( visible at @xmath17 50ns ) is well known to be due to the presence of a shelving state @xcite . once we have identified a nanodiamond as containing a single nv center based on its spectrum and antibunching signature , we measure its fluorescence lifetime by switching to pulsed laser excitation .
the lifetime of each center is extracted by fitting a single - exponential decay incorporating a small constant background to the decay histogram of emitted photons as shown in figure [ fig : fig_1](d ) .
we note that the substrate itself has a weak background fluorescence that shows a time dependence .
we correct for this artifact by collecting decay traces from the sample pumped directly next to the nv center .
we parameterize the background by a tri - exponential fit ( with a dominant sub - nanosecond lifetime ) which we subtract from the nv center decay histogram prior to fitting .
the residual background is around 10 counts per bin for a total acquisition time of about 20s .
in this section we report on statistical distributions of the various fluorescence characteristics , i.e. brightness , @xmath12 and lifetime , that we collected on both nanodiamond size ranges .
we first discuss the ` 100 nm diamond ' sample , containing 0200 nm size nanodiamonds with 108 nm median diameter , and subsequently we discuss the ` 25 nm diamond ' sample with 050 nm size nanodiamonds and 26 nm median diameter . for each sample ,
our statistics is based on identifying 30 to 40 nv centers as described in section [ single_nv_charac ] . as a first step in the characterization process
, we have measured the brightness of nanodiamonds that show a clear zpl line .
figure [ fig : fig_2](a ) shows a histogram of measured brightnesses as quantified by intensity on the ccd at fixed illumination intensity ( solid bars ) .
the intensity on the ccd is obtained by summing all pixels within the diffraction - limited image of each diamond ( green line in figure [ fig : fig_1](a ) ) .
we observe a wide distribution of intensities , spanning from @xmath19 to @xmath20 adu s ( analog digital units ) on the ccd per 300ms exposure time . at an estimated photon - to - adu conversion factor of 6 for our ccd camera ,
these brightnesses correspond to @xmath21 to @xmath22 collected photons per second , at a pump power around @xmath23 supplied by the cw laser .
first , these numbers show , at least assuming that nv centers are reasonably efficient emitters , the absorption cross sections of nanodiamond nv centers at @xmath24 nm pump wavelength are an order of magnitude below those of dyes and ii - vi quantum dot nanocrystals .
second , the wide distribution of brightnesses of over a factor 3 to 4 from nominally identical emitters is highly surprising . as described in section [ single_nv_charac ] , some nanodiamonds may have multiple nv centers , especially in the case of big nanoparticles . to exclude
that brightness variations are due to multiple nv centers , we screen the nanodiamonds on basis of @xmath12 minimum values . figure [ fig : fig_2](b ) shows a correlation plot , plotting the fluorescence intensities of all identified nv centers on 100 nm nanodiamond sample versus the recorded value of @xmath12 at zero delay ( @xmath25 ) .
we find that not just the brightness , but also the minimum in @xmath12 is distributed with minima in @xmath12 ranging from 0 to 0.75 .
the minimum in @xmath18 is only weakly correlated with collected intensity , especially via an apparent stepwise increase in the fluorescence intensity as @xmath18 reaches above @xmath26 .
such a stepwise increase would be expected since the minimum in @xmath12 scales with the number of emitters @xmath27 as @xmath28 . for @xmath29
more than a single fluorescent center is involved in the emission , resulting in higher fluorescence intensities on average . in the remainder
we concentrate our analyses on single nv centers , i.e. those nanocrystals for which @xmath30 . in figure
[ fig : fig_2](a ) , we overplot a histogram of fluorescence intensities of the subset of centers with @xmath30 as patterned bars .
the histogram shows a wide fluorescence distribution , with a factor of three difference between brightest and dimmest nv centers , and with a relative distribution of about 30% around the most frequent value .
figure [ fig : fig_2](b ) shows that for the single nv centers the wide distribution in fluorescence intensities notably does not show any correlation to @xmath18 values .
previously , beveratos @xcite attributed the distribution of @xmath18 values to inseparable background emission from the nanodiamond crystal host .
the lack of correlation between brightness and @xmath18 suggests that this assertion is not valid , i.e. that differences in the nv emission itself causes the brightness distribution .
it should be noted that brightness differences between nv centers could occur due to the fact that the linearly polarized pump field projects differently on the nv center absorption dipole moment . according to ref .
@xcite , an nv center has two dipole moments of equal size in the plane perpendicular to the nv axis . for a random distribution of nv orientations @xmath31 , the dipole projection @xmath32 on a linearly polarized pump field @xmath33 would lead to a brightness distribution that is strongly asymmetric , skewed strongly to high brightness . in contradistinction
, the brightness histrogram we find for nv centers is not skewed to high brightness .
to conclude , the single nv centers display a variation in brightness over a factor 3 , which can not be attributed to either background emission of the crystal host or variations in absorption cross - section through dipole orientation . in ( b ) .
( b ) fluorescence intensity as a function of @xmath34 .
( c ) bars : histogram of decay rates of emitters with @xmath35 .
solid line : calculated distribution of @xmath36 according to nanodiamond size distribution , normalized to the rate of nv centers in bulk diamond .
dotted line : free scaling of the calculated ldos distribution ( solid line ) to match the experimental histogram ( bars ) .
( d ) fluorescence intensity as a function of decay rate for emitters with @xmath35 ( decay rate error bars are within symbol size ) .
error bars are comparable to plot symbol size.[fig : fig_2],width=566 ] on basis of the large brightness distribution that can neither be explained by dipole orientation nor by background fluorescence , we hypothesize that there is a substantial variation in quantum efficiency ( qe ) of the nv centers . to examine whether quantum efficiency effects may be at play
, we have measured fluorescence decay rates for each identified single nv center .
we plot a histogram of the measured decay rates in figure [ fig : fig_2](c ) .
we observe a very wide distribution , with the slowest emitters decaying almost 4 times more slowly than the fastest ones .
the most frequently occurring decay rate is around @xmath37ns@xmath38 ( corresponding to about 33ns ) .
the time constant of 33ns is appreciably slower than the fluorescence lifetime of nv centers in bulk diamond , for which the accepted literature value is 11.6 ns @xcite .
this much slower decay as well as the occurrence of a distribution of rates is in agreement with previously reported values @xcite .
if the hypothesis that quantum efficiency variations are responsible for the large variability in brightness in figure [ fig : fig_2](a ) is valid , one might expect a correlation between the brightness and decay rate of the emitters .
we plot the measured brightness as a function of decay rate for each nv center in figure [ fig : fig_2](d ) which , however , displays no clear correlation .
importantly , we note that the measured quantity here is the total decay rate @xmath39 which reflects both variations of radiative decay rate @xmath40 and nonradiative decay rate @xmath41 , whereas the quantum efficiency is given by @xmath42 .
based on the collection efficiency of our microscope objective ( @xmath43 % ) , the count rates in figure [ fig : fig_2 ] correspond to an intermediate ( most data ) to strong ( data point at and above 150 kadu/300 ms))excitation regime , where the photon emission rate is approximately 0.2 to 0.9 times the total decay rate @xmath44 .
both if one assumes to be in saturation , and below saturation , one expects the fastest decay to imply highest brightness if one assumes @xmath41 to be approximately constant but @xmath40 to be distributed .
this conclusion is not strongly supported by the data .
if conversely we assume @xmath40 to be approximately constant across nv centers while @xmath41 is distributed , one would expect highest brightness to correlate with the slowest decay rate ( lowest @xmath41 ) , a hypothesis also not strongly supported by the data .
specifically , we find that the subset of crystals around the most frequent brightness show the full spread of decay constants , while conversely also the subset of crystals that have decay constant around the most frequent rate , contain the full range of brightnesses .
we conclude that the common assumption that nv centers in such large nanocrystals as we study are near - unity quantum efficiency sources is not supported by the data . instead _
both _ the radiative and the nonradiative decay rates could be widely distributed .
many workers had already noticed that nanodiamond decay rates are widely distributed @xcite . the variation is commonly attributed to variations solely in the radiative rate @xmath40 due to a local - density - of - photonic - states effect , assuming zero nonradiative decay ( i.e. unit quantum efficiency ) .
the hypothesis , explained in detail by inam @xcite , is that variations are in large part due to the fact that the ldos experienced by an nv center is influenced by the nanoscale geometry of its environment , i.e. the fact that the source is situated inside , and close to the surface of , a very high index nano - object that is embedded in a low index environment .
a distribution in ldos can arise from the size distribution of nanodiamonds and from the fact that different nv centers have different dipole - moment orientation and positions within the crystals @xcite . here
we assess if this ldos hypothesis is quantitatively reasonable .
considering that the radiative decay rate @xmath45 is proportional to ldos , we evaluate the @xmath45 distribution by calculating the ldos distribution assuming spherical nanoparticles , for which the ldos is analytically known @xcite .
we modeled nanodiamond nv centers as point dipoles with randomly oriented dipole moments homogeneously distributed in position in a set of dielectric spheres with dielectric constant equal to the bulk diamond value ( @xmath46 ) .
we furthermore take into account the distribution of particle size as specified according to the nanodiamond size - distribution histogram provided by the manufacturer .
we histogram the occurrence of ldos values to find its probability distribution over all sphere sizes , dipole positions and dipole configurations .
the resulting ldos distribution can be converted into a distribution directly comparable to experimentally measured decay rates by scaling the ldos to the previously reported decay rate of ( 11.6 ns)@xmath47ns@xmath38 of nv centers in bulk diamond @xcite .
we note that this entire procedure involves no adjustable parameter or any fit to data .
the final result is plotted in figure [ fig : fig_2](c ) as the blue solid line .
the calculated histogram correctly predicts that emission is significantly decelerated compared to decay in bulk diamond , consistent with the fact that decay in small dielectric spheres is decelerated both compared to vacuum , and bulk dielectric .
however , we observe that the calculated histogram peak falls at much lower decay rate than the experimentally measured histogram peak , with a discrepancy amounting to a factor of 3 . a similar discrepancy between the calculated and experimentally measured decay rates of nanodiamond nv centers
was recently reported for samples with a much wider size distribution , centered at much larger median size , measured by inam @xcite .
we note that this discrepancy can not be attributed to the fact that we have taken particle shape to be simply spherical , and that we have neglected the presence of a substrate .
these effects cause only small changes in the expected decay rate histogram , as verified in fdtd simulations by inam @xcite .
one might further argue that the occurrence of a degenerate in - plane dipole moment @xcite could skew the histogram of expected decay rates towards higher values , if one assumes the dipole moment is free to diffuse prior to de - excitation .
however , if we just histogram the fastest rate instead of the average rate at each possible nv center position , the resulting histogram also does not lead to a consistent explanation ( not shown ) .
we conclude that ldos variations in nanodiamond under the hypothesis of unit quantum efficiency and a bulk decay rate of ( 11.6 ns)@xmath38 do not explain the variation in measured decay rates .
inam proposed that although calculated and experimental absolute decay values are inconsistent on basis of ldos theory and the bulk rate in diamond , the calculated and experimental results can be scaled onto each other . indeed ,
if we scale the reference rate that sets the rate axis for the calculated histogram peak not by 11.6ns ( rate constant in bulk diamond @xcite ) , but by a factor of 3 shorter , calculated and measured histogram coincide reasonably ( figure [ fig : fig_2](c ) black dotted line ) .
such a scaling would imply as hypothesis a unit quantum efficiency , together with an as yet hidden explanation that introduces a multiplicative correction factor in the calculated rate distribution .
a second hypothesis could be that no adjustment should be made of the bulk rate that enters the comparison , but that the quantum efficiency of nv centers , while unity in bulk , is not unity in nanocrystals .
indeed , an additive offset to the calculated histogram is introduced by nonradiative decay channels that do not occur for bulk diamond , but could occur for nanocrystals due to defects and the presence of large surface area that could contain quenching sites .
so far , significant nonradiative decay was evidenced only for very small nanocrystals ( 5 nm @xcite ) .
a distribution of nonradiative rates around @xmath48ns@xmath38 would shift the calculated histogram to the measured rates .
the magnitude of the required @xmath41 implies that quantum efficiencies should be around 30% to 50% for the slowest nanocrystals in the measured ensemble ( assuming unit efficiency in bulk ) .
a third hypothesis could be that the assumption that the bulk rate is entirely radiative to start with , is incorrect , i.e. that the quantum efficiency of nv centers in bulk is significantly below unity contrary to common assumption . assuming a bulk quantum efficiency of 70% instead of 100% would overlap the calculated histogram peak with the measured most frequent rate .
however , this explanation would severely underestimate the width of the decay rate distribution unless a distribution of @xmath41 is at play .
we conclude that ldos variations in nanodiamond alone do not explain the variation in measured decay rates , and that a distribution of radiative and nonradiative decay constants must be at play for nv centers in nanodiamond .
furthermore we conclude that an actual experimental calibration of quantum efficiency of individual nv centers is highly desired , which we will return to in sections [ sec : liquidldos ] and [ sec : mirrorldos ] .
( shown in the inset ) according to nanodiamonds size distribution , normalized to the rate of nv centers in bulk diamond . solid black line : free scaling of the calculated ldos distribution ( blue line )
so to match the experimental histogram ( bars ) .
[ fig : fig_3],width=566 ] for experiments in which nanodiamonds are intended as probes of , or sources to be embedded in , nanophotonic environments , a size smaller than that of the 100 nm nanodiamonds would be advantageous , as smaller size implies higher spatial resolution and a lower perturbative effect on the modes of the nanophotonic system @xcite . therefore , we repeated the brightness and decay rate statistical measurements for the 25 nm sample , i.e. the batch of crystals with median diameter 26 nm . for 25 nm nanodiamonds ,
background - subtracted @xmath12 measurements show a zero dip for all identified nv centers ( for a typical 25 nm nanodiamond @xmath12 see e.g. figure [ fig : fig_1](c ) ) .
the fact that we found no nanodiamonds with multiple nv centers for this sample is commensurate with the smaller average crystal size . for these single nv centers ,
figure [ fig : fig_3](a ) shows a histogram of the fluorescence intensities measured on the ccd using cw illumination at 1 mw/@xmath49m@xmath50 . as in the case of the 100 nm sample , the histogram exhibits a wide fluorescence intensity distribution , in this case spanning a factor of about two .
the most frequent brightness is approximately a factor two below that of the 100 nm sample .
the fact that the histogram is skewed towards lower intensities suggests that the intensity distribution is not due to the projection of the dipole moment orientation on the pump - field polarization , as this would cause a skewing in the opposite direction . as in the case of the 100 nm nanocrystals
, we have also measured the decay rate for all 25 nm nanocrystals that we identified .
the decay rate distribution , plotted as a histogram in figure [ fig : fig_3](c ) , again shows a very wide distribution , with a factor of 4 contrast in decay rate .
we find higher decay rates on average compared to 100 nm nanodiamonds , with a most frequent decay rate around @xmath51ns@xmath38 ( corresponding to about 28ns ) .
the most frequent decay rate is approximately 25% faster than for the 100 nm nanodiamonds .
exactly as in the case of 100 nm nanodiamonds , the wide distribution of fluorescence brightness and decay rates does not imply a correlation between the two . as confirmation , in figure [ fig : fig_3](b )
we plot the fluorescence intensities and decay rates of 25 nm diamond nv centers , where we find no clear correlation .
we hence conclude that as for the larger diamonds , a distribution in absorption cross - sections , radiative and nonradiative rates is likely at play . to assess whether the common hypothesis that the distribution in rate is due to a distribution in ldos is valid for small nanodiamonds , we calculated the ldos distribution ( plotted in the inset of figure [ fig : fig_3](c ) ) also for the 25 nm sample taking into account the size distribution histogram measured by the nanodiamond supplier .
the rate distribution expected from the calculated ldos scaled with the bulk rate ( blue solid line in figure [ fig : fig_3](c ) ) is first , considerably narrower than the experimental decay rate distribution , and second , at considerably reduced rate compared to the measured decay rates .
the magnitude of the discrepancy in decay rate is approximately a factor 6 , i.e. twice larger than for the 100 nm nanodiamonds .
a scaling of the calculated distribution by a multiplicative factor ( here a factor of about 6 ) as proposed by inam @xcite does not lead to a good correspondence as in the 100 nm case , as the relative width of the measured histogram far exceeds that of the scaled calculation ( black line in figure [ fig : fig_3](c ) ) . taking our data on both 100 nm and 25 nm nanodiamonds together , we hence do not find support for the hypothesis by inam that a hitherto hidden effect multiplies the radiative rate of nv centers in nanocrystals compared to bulk diamond .
a more likely explanation that does not involve a scaling of @xmath40 due to an unknown origin is that nv centers are subject to a distribution of nonradiative rates on top of the ldos - induced radiative rate distribution .
the measured distribution of 25 nm nanodiamond decay rates points at a wide distribution of nonradiative rates . according to this hypothesis
, those crystals with decay around the most - frequent decay rate of @xmath52ns@xmath38 must have a quantum efficiency below 15 - 20% , twice lower than for the 100 nm nanocrystals .
the overall lower quantum efficiency estimate is commensurate with the reduction in average brightness , and is also consistent with the fact that a larger sensitivity to nonradiative decay channels is potentially associated with the increased nanocrystal surface - to - volume ratio .
our data indicate that nv centers even in nanocrystals as large as 100 nm across do not have unit quantum efficiency , and that there is a distribution of nonradiative and radiative rates .
therefore , it is desirable to measure quantum efficiency , @xmath40 and @xmath41 independently on individual nv centers in order to decide if /which
nv centers in nanodiamonds are suitable for use as ldos probes for integration in nanophotonic devices . in literature , several reports have appeared that evidence lifetime changes of nv emission that are induced by placing nanocrystals in photonic environments of varying index and topology @xcite . to our knowledge
all these measurements of lifetime changes for nanodiamonds were performed by comparing the _ mean rate _ from an ensemble of single - center measurements in one system , to measurement on a different ensemble of single - centers in a second system @xcite .
ruijgrok @xcite in particular report changes for a system in which the induced ldos change is exactly known , and report changes in the _ mean _ rate consistent with those expected for unit - quantum - efficiency emitters .
however , we note that in these ensemble measurements , the change of the _ mean _ decay rate was far smaller than the width of the rate distribution .
therefore , those measurements do not allow to ascertain whether for _ any given _ single nv center the rate actually varies with varying ldos in a manner that is consistent with the expectations for efficient emitters .
we perform measurements in which we change the ldos of single nv centers in two ways .
first , we have immobilized the nanocrystals by evaporating a thin layer of sio@xmath4 , and then introduced liquids with different refractive indices to systematically modify the ldos around the nv centers ( section [ sec : liquidldos ] ) .
second , we have used nanomechanical tuning of ldos using a piezo - controlled mirror to effectuate a drexhage experiment @xcite on single nanodiamonds ( section [ sec : mirrorldos ] ) .
the nanomechanical method is preferable since it rapidly allows continuous variation of ldos . due to their smaller apparent brightness , however , we were unable to apply the technique also to 25 nm diamond nanocrystals as the much higher pump energy required to reach appreciable count rates caused thermal breakdown of the micro - mirror
. therefore we applied the liquid tuning method ( section [ sec : liquidldos ] ) to the small nanodiamonds , and report on the nanomechanical method for the large nanodiamonds ( section [ sec : mirrorldos ] ) . for the small nanodiamonds
only the liquid tuning method could be applied due to experimental constraints .
the idea and first implementation of this method was pioneered by snoeks @xcite , and later adopted by e.g. @xcite .
sources are placed in close proximity to a planar interface between a dielectric and a half - space that can be filled with liquids of different refractive index .
the advantage of using liquid tuning of refractive index for sources near an interface is that the ldos changes near an interface are nearly independent of emission frequency , and are known to be excellently described by the theory explained in full by urbach and rikken @xcite . in view of the broad emission spectrum of nv centers in diamond , it is important to apply a broadband ldos change when seeking to measure lifetime changes .
a narrowband ldos variation as obtained with a high - q microcavity that would for instance be tuned to the zero phonon line , would not necessarily affect the rate , but rather only the branching ratio between the zero phonon line and the rest of the spectrum .
we prepared the sample by evaporating a 60 nm thick layer of sio@xmath53 on one of the 25 nm nanodiamond samples on which we had identified single nv centers .
this step immobilizes the nanodiamonds , and ensures that liquid application to tune ldos does not add new chemically induced nonradiative decay channels .
next , we defined a 3 mm deep liquid reservoir on top of the sample using a ring - shaped enclosure cut from polydimethylsiloxane ( pdms ) bonded to the substrate .
figure [ fig : fig_4](a ) depicts an schematic of the sample . in order to investigate the decay dynamics of the nv centers in response to the ldos variations , we randomly selected six nv centers from the ensemble of figure [ fig : fig_3 ] and measured their lifetimes in three conditions : before adding any liquids where the top half space of nanodiamonds consists of air ( @xmath54 ) , after adding water with refractive index of @xmath55 , and after adding isopropyl alcohol ( ipa ) with refractive index of @xmath56 to the liquid bath on top of the nanodiamonds . in the lower panel of figure [ fig : fig_4](b )
we plot the decay rate of each nv center as a black bar measured when the sample is dry .
the color bars in the upper panel of figure [ fig : fig_4](b ) show the difference between decay rates of each nv center after and before introducing a liquid .
we observe that for most centers , the decay rate either barely varies , or varies non - monotonically with applied index , although generally , one expects to observe an increase in the decay rate of a dipole emitter in a homogenous medium by increasing the refractive index of the environment due to an increase of the ldos @xcite . for comparison to the data we calculate the normalized decay rate of a dipole positioned in the middle of a 60 nm thick sio@xmath4 slab ( @xmath57 ) which is sandwiched between a semi - infinite slab of quartz with @xmath58 at the bottom and a semi - infinite slab with varying refractive index at the top .
figure [ fig : fig_4](c ) shows the calculated decay rate of the dipole as a function of the refractive index of the top layer for different dipole - moment orientations .
the calculation shows that as the refractive index increases the decay rate change is expected to be monotonic and increasing and is the largest for a dipole oriented perpendicular to the interface while the change is slight for a dipole - moment parallel to the interface . comparing the calculated rate changes to the measured decay rates ( figure [ fig : fig_4](b ) )
, we find no systematic rate variation for the nv centers within the error bars of our experimental data .
the error bar for measurements on individual nv centers is composed of two contributions .
first , the error bar contains the uncertainty in the fit to the fluorescence decay trace .
second , the experiment is hampered by the fact that the decay rates of some of the single nanodiamonds was observed to jump slightly between observations , even if no change in environment was applied .
these jumps amount to approximately 5% , and occur on time scales of hours to days . to exclude this effect
, we have measured the rate for each liquid and a reference in air on the same day for each nv center to generate the plot of rate changes in figure [ fig : fig_4](b ) .
, width=566 ] we find that there is no evidence that the rate of _ individual _ nv centers in 25 nm diamond nanocrystals actually varies in accordance with the known applied ldos changes .
this observation appears to be at variance with earlier work that evidenced a small change in the mean decay rate of an ensemble of nv centers in a liquid - tuning experiment @xcite . as two caveats , we note that first , in that previous work it was not verified if any such change occurred for each source individually , and second , that a shortcoming of our experiment for quantititave interpretation is that for dipoles located close to an interface ( as is the case here ) , the decay rate enhancement significantly depends on the unknown dipole - moment orientation of the emitter with respect to the interface . while the fact that we do not know the dipole orientations for each nv center makes it impossible to conclude with certainty that an ldos effect is absent , two hypotheses for absence of an ldos effect could be proposed .
first , that the quantum efficiency of 25 nm diamond nanocrystals is low , or second , that the very high index diamond shell around the nv centers intrinsically prevents the nv center from responding to ( moderate ) changes in the ldos of the environment it is supposed to probe .
we exclude the latter explanation on basis of a set of finite element ( comsol ) simulations , in which we calculate the decay rate of a point dipole randomly located in a diamond nanosphere , which in turn is placed at varying distances from a planar interface with materials of various dielectric constants .
we find that in all cases , the decay rate of the source simply follows the theory of urbach and rikken @xcite for a source without dielectric shell , multiplied by a pre - factor that is essentially a quasistatic local field correction factor due to the diamond shell and is independent of the varying ldos .
to conclude , our observation that no nv center responds to ldos changes together with our earlier correlation plots of brightness versus decay rate ( figures [ fig : fig_2 ] and [ fig : fig_3 ] ) means that the quantum efficiencies of 25 nm diamond nanocrystals are low .
a second important conclusion is that single nv centers of 25 nm size appear unsuitable to measure ldos changes due to the slight jumps in lifetime , unless the entire measurement scheme ( including reference measurements ) takes less than a few hours . in this section
we report on the results of a second method to assess if single nv centers are suited for ldos measurements , and to calibrate emission rates .
this second method has the advantage that we can apply calibrated ldos changes rapidly so that jumps in the intrinsic rate constants can be avoided that hamper experiments where macroscopic sample changes are required to modify ldos , as in a liquid immersion experiment .
the scheme is similar to a measurement procedure reported by buchler @xcite , based on a drexhage experiment @xcite . in a drexhage experiment ,
a mirror is used to impose a large ldos change , and rates are measured as a function of the emitter - mirror separation . while this method is usually implemented by creating a set of macroscopic samples where mirrors are coated with spacers of calibrated thicknesses @xcite or by creating a single sloping wedge between a mirror and an ensemble of emitters @xcite , buchler realized a nanomechanical version that can be applied to a single emitter .
buchler used a silver - coated curved fiber end as a mirror attached to a piezo - stage to precisely tune the mirror distance to an underlying single emitter on the sample @xcite . here , we use a similar method in a shear - force - feedback near - field microscope .
however , instead of a vertical mirror displacement , we use a lateral displacement of the curved mirror while staying in shear - force feedback to keep the mirror and sample substrate in near - contact .
as the contact point moves sideways , the spherical mirror shape ensures that the mirror - to - emitter distance varies ( figure [ fig : fig_5](a ) ) . to fabricate the curved mirror ,
we glued 45@xmath49 m polystyrene beads ( polysciences , inc . ) to the end of cleaved optical fibers and coated the beads with about 200 nm of silver . ) as a function of mirror lateral displacement ( c ) corresponding decay rate map of the nv center as a function of mirror lateral displacement .
decay rates are extracted from a single - exponential fit to the time - trace data for each pixel .
( d ) dots : decay rate as a function of the emitter distance to mirror , extracted from the decay rate map shown in ( c ) . solid line : fitted ldos on the measured rate oscillations with fit parameters @xmath40 , @xmath41 and a set of dipole moment orientations .
the shown qe is the most - likely value .
( e ) and ( f ) decay rate as a function of emitter - mirror distance for two other nv centers ( number 4 and 5 in table [ table : table_1 ] , respectively ) and the fitted rate from ldos calculation .
[ fig : fig_5],width=566 ] we have applied the scanning - mirror ldos changing technique to 100 nm nanodiamonds containing single nv centers , prepared as described in section [ subsec : smaple_prep ] , and subsequently embedded in a 200 nm thick layer of planarizing spin - on glass ( fox-14 , dow corning ) .
the spin - on glass immobilizes the nanocrystals , so that they are not moved during shear - force scanning . in order to vary the distance between the mirror and the emitter , we scan the mirror bead laterally on top of an identified nv center . for each position of the mirror bead
, we collect the fluorescence emission of the nv center , positioned at the focus of the pump laser , through the confocal microscope setup described in section [ subsec : exp_setup ] . by scanning the mirror bead ,
we obtain a confocal fluorescence intensity map as shown in figure [ fig : fig_5](b ) . here
, each pixel represents the relative position of the mirror bead with respect to the nv center with the false color representing the collected fluorescence intensity .
we clearly observe interference rings in the fluorescence intensity map .
these rings mainly stem from the fact that the mirror imposes a standing - wave pattern on the 532 nm pump field , which subsequently results in a modulation of the fluorescence intensity . for each pixel in the fluorescence intensity map
, we stored absolute photon arrival times , as well as laser pulse arrival times , allowing us to extract the decay dynamics .
we use a single - exponential fit to obtain the corresponding decay rate of the emitter for each mirror position , as plotted in figure [ fig : fig_5](c ) . here
, the color scale represents the fitted decay rate corresponding to a defined mirror position .
interestingly , we observe a radial modulation of the decay rates varying between about 0.02ns@xmath38 to 0.04ns@xmath38 , which we attribute to the varying ldos in front of the mirror . to quantify the variation as a function of the distance from emitter to mirror , we extracted the decay rates as a function of the lateral distance to the mirror central position . to this end
, we bin pixels in concentric rings of equal lateral distance to the mirror center , and concatenate the photon - correlation time traces of all pixels in each bin to obtain a single fluorescence decay trace per radial distance . using the spherical form factor of the bead ,
we convert lateral position of the mirror to normal distance of the mirror to emitter .
the decay rate fitted to the fluorescence decay for each mirror - emitter separation is depicted in figure [ fig : fig_5](d ) ( black dots ) .
we observe a distinct oscillation of the rate around 0.028ns@xmath38 , with a 15% amplitude . to our knowledge , this is the first report of a reversible change in the decay rate of a single nv center in a calibrated ldos experiment . due to the large radius of the mirror compared to its distance to emitters ,
it is reasonable to consider the experimental configuration as a planar glass - air - mirror system in which we vary the air thickness . for this system
, the ldos is exactly known @xcite for any dipole orientation and position .
we fit the experimental decay rate data using the theoretical ldos @xmath59 according to @xmath60 , where the fit parameters are the nonradiative decay rate @xmath61 , the radiative decay rate @xmath62 , and the dipole orientation @xmath63 relative to the normal to the plane .
finally a small offset @xmath64 appears that is due to the unknown distance - of - closest approach in shear - force microscopy , of the order of 15 nm .
care must be taken that a correlation exists between the dipole orientation @xmath63 , and the apparent quantum efficiency @xmath65 returned by the fit routine , due to the fact that the ldos for different dipole - moment orientations is similar in qualitative @xmath66-dependence , but different in oscillation contrast . to overcome this dependency , we examine fits of the rate oscillations for a set of dipole orientations and evaluate the goodness of fits individually .
the goodness of fit is established by measuring in how far the residuals of the fit ( point - by - point deviation between data and fit function ) are within the data point error bar .
this gives a range of quantum efficiency values for which the fit is consistent with the data . in this paper
we report the range of quantum efficiency values consistent with the data given that the dipole orientation should be treated as an unknown parameter . in case the dipole moment orientation of the nv center
is known , which could in principle be realized using fourier microscopy @xcite , one can extract a more precise value for the quantum efficiency .
for the particular data set shown in figure [ fig : fig_5](d ) , we find a most likely quantum efficiency of 26% and a dipole moment orientation within @xmath67 along the sample plane .
the range of quantum efficiencies consistent with the data for this nv center is bounded from below by 26% , and from above by 50% . to our knowledge
this is the first experimental calibration of the quantum efficiency of a single nv center in a diamond nanocrystal . in order to investigate the typical quantum efficiencies of nv centers , we selected five random nv centers and examined them with the moving mirror experiment as explained above .
each nanodiamond was selected to be a single nv center containing nanocrystal according to the criteria we outlined in section [ single_nv_charac ] , without further post - selection for inclusion in the drexhage experiment .
table [ table : table_1 ] summarizes the confidence intervals for the fitted values of the quantum efficiencies , given that we do not know the dipole orientation a priori .
the total decay rate @xmath44 , i.e. the sum of @xmath40 and @xmath41 varies almost over a factor 2 , and can be fitted accurately . in table
[ table : table_1 ] we also report radiative and nonradiative decay rates .
values are reported as fitted to the data while fixing the dipole orientation at its most likely value , in which case the data can be fit with small error bars on @xmath40 and @xmath41 .
the wide range of quantum efficiencies consistent with the data of course imply that in a completely free fit @xmath40 and @xmath41 both have a large error bar , though their sum @xmath44 does not .
we find that the different nv centers we probed , while having reasonably comparable brightness , in fact have widely different quantum efficiencies .
for instance , nv center 4 ( figure [ fig : fig_5](e ) ) has an efficiency certainly below 14% , while nv center 5 ( figure [ fig : fig_5](f ) ) has a quantum efficiency certainly above 58% , in the range 58% to 90% .
furthermore , we note that the fits neither result in the conclusion of a fixed nonradiative rate at varying radiative rate , nor conversely in the conclusion that the radiative rate is a constant while the nonradiative rate varies . instead
, both the radiative and the nonradiative decay rates are distributed . even with the wide error bars on quantum efficiencies , we establish that both distributions span at least around a factor two in range . our measurements hence show that the common tacit assumption of near - unit quantum efficiency and nearly identical emission characteristics barring those due to depolarization factors in the nanodiamonds should be discarded . .extracted quantum efficiencies , rates and brightnesses for five randomly chosen nv centers in 100 nm nanodiamonds .
quantum efficiencies are quoted as a range of values , taking into account that the dipole orientation is a free parameter . in this fit procedure ,
the proper fit parameters are the quantum efficiency and the total decay rate , which has small error bar . for completeness
we also show the two derived quantities @xmath40 and @xmath41 . these rates and their error bars
are given as fitted when fixing the dipole orientation to its most likely value .
fitted brightnesses were converted to ccd response units to be comparable with fig .
3[table : table_1 ] [ cols="<,^,^,^,^,^",options="header " , ]
in order to assess the suitability of nanodiamonds as photonic ldos probes and as building blocks for hybrid photonic systems , we have investigated the brightness , decay rate , and quantum efficiency of single nv centers in nanodiamonds of 25 nm , and of 100 nm median size . for both size
ranges a wide distribution in brightnesses and rates is found , consistent with reports by earlier workers .
we conclude that the wide distribution of rates is due to a distribution in radiative rates , nonradiative rates and quantum efficiencies .
this conclusion contradicts earlier work , that interpreted the wide distribution of decay rates as mainly due to a photonic effect that causes a distribution in radiative decay rates via the ldos . instead
, our measurements show that even nv centers in large nanocrystals show a wide range of quantum efficiencies when probed in a controlled ldos experiment . for nv centers in 100 nm nanocrystals
we find quantum efficiencies distributed between 10% and 90% , while for the smaller nv centers the quantum efficiency for fluorescence is likely a factor of two smaller on average . for applications in nanophotonic experiments using single nv centers
, a highly problematic property is that quantum efficiency does not correlate with brightness , due to the fact that both the radiative and nonradiative rate are distributed . in other words , screening nanodiamonds to pick the best one to probe ldos or for incorporation in a photonic device can not rely on a simple metric such as spectrum or brightness .
instead , we argue that future work in the hybrid assembly of nanodiamonds in plasmonics and photonic crystals to realize accelerated spontaneous emission decay always requires an experimental protocol in which nanodiamonds are first individually calibrated in terms of quantum efficiency . while we have shown a method for such calibration , this is a highly tedious procedure that is not easily implementable and requires dedicated near - field manipulation equipment . in this work ,
we have not speculated on the origin of the apparent low quantum efficiency of single nv centers in nanodiamonds .
we close with two remarks on the origin of the low apparent quantum efficiency .
firstly , we have treated the nv center as a quasi two - level system .
our experiments thus address the question what the apparent quantum efficiency is when attempting to use an nv center as a two - level ldos probe . from the nv center spectrum it is clear that the spectrum is very wide with a large vibrational broadening spectrum .
moreover it is well known that the nv@xmath68 center is not a two level system . instead , the nv@xmath68 has different spin sublevels , and may experience spin - flip intersystem crossing between allowed spin transition manifolds .
the rates for these transitions were recently characterized in detail for nv centers in bulk diamond in ref .
a further complication is that the nv@xmath68 defect may infrequently transition to an uncharged nv center that also luminesces , yet at different rate and efficiency .
the many rates involved in these transitions can further vary between nv centers due to , e.g. , variations in crystal strain .
a full treatment of the response of nv centers to ldos changes would hence have to treat the full rate equations in which radiative transition rates are , and intersystem crossing rates are not , affected by ldos .
an important step beyond our work will be to perform the measurements we have described here but employing the spin selective techniques reported by robledo @xcite to establish what the nonradiative and radiative rates are for each transition separately , instead of lumping rates into effective two - level parameters .
while the ldos changes we have applied using a mirror are broadband ldos changes that modify radiative transitions roughly equally across the emission spectrum , ldos changes that have strong spectral features could be used advantageously to enhance or suppress the importance of intersystem crossing .
having established that further work is required to separate the quantum efficiencies reported here into parameters per transition in a more complete level scheme , we turn to possible reasons for the below - unity values of quantum efficiency that we find .
the lower brightness of the smaller nv centers , as well as their lack of response to ldos changes suggest that the surface , i.e. , surface contamination with carbon , or surface defects may play a role in providing quenching sites . indeed , it has already been suggested for very small ( 5 nm )
nanodiamonds that nv centers may suffer quenching due to graphite shells on the diamonds . while the nanodiamonds we used have been employed by several groups in spontaneous emission control experiments in untreated form , exactly as in our experiments , additional surface treatments such as by prolonger immersion in boiling sulphuric acid , or cleaning in kningswasser have been proposed by several workers .
whether or not such treatments actually affect quantum efficiency is as yet unclear , as is whether quenching can be completely suppressed .
we propose that the quantum efficiency calibration method that we demonstrated will be an indispensable tool to evaluate such cleaning methods , as well as to screen other color centers in diamond for advantageous fluorescence properties .
we thank s. schietinger for communications regarding sample preparation methods , m. frimmer for experimental help and discussions , and c. osorio for suggestions to improve the manuscript .
this work is part of the research program of the `` foundation for fundamental research on matter ( fom ) '' , which is financially supported by the `` the netherlands organization for scientific research ( nwo ) '' .
afk gratefully acknowledges a nwo - vidi grant for financial support .
99 santori c , fattal d and yamamoto y 2010 _ single - photon devices and applications _ ( john wiley & sons ) sprik r , van tiggelen b a and lagendijk a 1996 _ europhys .
lett . _ * 35 * 265270 brouri r , beveratos a , poizat j p , and grangier p 2000 _ opt .
lett . _ * 25 * 1294 kurtsiefer c , mayer s , zarda p , and weinfurter h 2000 _ phys . rev
. lett . _ * 85 * 290 beveratos a , brouri r , gacoin t , poizat j p , and grangier p 2001 _ phys . rev . a _ * 64 * 061802 tisler j , balasubramanian g , naydenov b , kolesov r , grotz b , reuter r , boudou j p , curmi p a , sennour m , thorel a , borsch m , aulenbacher k , erdmann r , hemmer p r , jelezko f , and wrachtrup j 2009 _ acs nano _ * 3 * 1959 babinec t m , hausmann j m , khan m , zhang y , maze y r , hemmer p r , and lonar 2010 _ nature nanotechn . _ * 5 * 195 michaelis j , hettich c , mlynek j and sandoghdar v 2000 _ nature _ * 405 * 325328 cuch a , drezet a , sonnefraud y , faklaris o , treussart f , roch j f and huant s 2009 _ opt
. express _ * 17 * 1996919980 cuch a , mollet o , drezet a and huant s 2010 _ nano lett .
_ * 10 * 45664570 degen c. l. 2008 _ appl .
lett . _ * 92 * , 243111 taylor j m , cappellaro p , childress l , jiang l , budker d , hemmer p r , yacoby a , walsworth r and lukin m d 2008 _ nature phys _ * 4 * 810816 balasubramanian g , chan i y , kolesov r , al - hmoud m , shin c , kim c , wojcik a , hemmer p r , krueger a , hanke t , leitenstorfer a , bratschitsch r , jelezko f and wrachtrup j 2008 _ nature _ * 455 * , 648651 maletinsky p , hong s , grinolds m s , hausmann b , lukin m d , walsworth r l , loncar m , and yacoba a 2012 , _ nature nanotechnology _ * 7 * , 320 - 324 inam f a , gaebel t , bradac c , stewart l , withford m j , dawes j m , rabeau j r , and steel m j 2011 _ new j. phys .
_ * 13 * 073012 ruijgrok p v , wuest r , rebane a a , renn a , and sandoghdar v 2010 _ opt .
exp . _ * 18 * 6360 van der sar t , hagemeier j , pfaff w , heeres e , thon s , kim h , petroff p , tjerk o , bouwmeester d , and hanson r 2012 _ j. opt . soc .
b _ * 29*(4 ) 698703 drexhage k h 1970 _ j. lumin . _
* 1 - 2 * 693 buchler b c , kalkbrenner t , hettich c , and sandoghdar v 2005 _ phys
lett . _ * 95 * 063003 | nv centers in diamond are generally recognized as highly promising as indefinitely stable highly efficient single - photon sources .
we report an experimental quantification of the brightness , radiative decay rate , nonradiative decay rate and quantum efficiency of single nv centers in diamond nanocrystals .
our experiments show that the commonly observed large spread in fluorescence decay rates of nv centers in nanodiamond is inconsistent with the common explanation of large nanophotonic mode - density variations in the ultra - small high - index crystals at near - unity quantum efficiency .
we report that nv centers in 25 nm nanocrystals are essentially insensitive to local density of states ( ldos ) variations that we induce at a dielectric interface by using liquids to vary the refractive index , and propose that quantum efficiencies in such nanocrystals are widely distributed between 0% and 20% . for single nv centers in larger 100 nm nanocrystals ,
we show that decay rate changes can be reversibly induced by nanomechanically approaching a mirror to change the ldos . using this scanning mirror method , for the first time
we report calibrated quantum efficiencies of nv centers , and show that different but nominally identical nanocrystals have widely distributed quantum efficiencies between 10% and 90% .
our measurements imply that nanocrystals that are to be assembled into hybrid photonic structures for cavity qed should first be individually screened to assess fluorescence properties in detail . |
experimentally , the accmor collaboration in 1981 observed a @xmath5 , @xmath6 structure at 1850 mev in the @xmath7 @xmath8-wave with a width of about 240 mev@xcite .
subsequently , the ves collaboration reported a @xmath6 threshold enhancement in the @xmath9 channel with a mass of about 1840 mev and width of about 210 mev in their @xmath10 data@xcite and also in @xmath11 data where @xmath12 ; they also observed a strong peak at the same mass in the @xmath7 @xmath8-wave in their @xmath13 data@xcite . in 2001 , anisovich et al .
reported a @xmath5 , @xmath6 resonance with a mass of about 1880 mev and a width of about 255 mev in the @xmath9 and @xmath7 @xmath8-wave@xcite .
more recently , a similar @xmath3 resonance was observed by the e852 collaboration in the @xmath14@xcite , @xmath15@xcite , and @xmath9@xcite channels , respectively .
it has been established that these observations in different channels refer to a single state @xmath0@xcite . in the meson summary table of the pdg2008 , the mass and width of the @xmath0
are quoted to be @xmath16 mev and @xmath17 mev , respectively@xcite . as for the @xmath0 nature , after a @xmath3 hybrid conjecture [ @xmath18 was first proposed by anisovich et al.@xcite , several groups also claimed the @xmath0 being a viable non - exotic hybrid candidate@xcite . with the @xmath19 as the well - established @xmath20 @xmath4 state@xcite , the @xmath0 looks like the @xmath21 rather than the @xmath1 isovector @xmath4 state [ @xmath22 $ ] based on its mass , because the observed mass of the @xmath0 just overlaps the flux - tube model prediction of about @xmath23 gev for the light @xmath3
hybrid@xcite , but is about 200 mev lower than the godfrey - isgur ( gi ) quark model prediction of about 2.1 gev for the @xmath1 nonstrange @xmath4 state@xcite .
however , comparing the experimental evidence for the @xmath0 with the strong decay properties of the @xmath21 expected by the model of hybrid meson decay developed by page , swanson , and szczepaniak(pss ) based on the heavy quark expansion of qcd and the strong coupling flux tube picture of nonperturbative glue@xcite(see the table ii of ref.@xcite ) , one can find the following features of the @xmath0 casting doubt over the hybrid interpretation for the @xmath0 : \i ) the observation in the @xmath15 channel of the @xmath0 is inconsistent with the hybrid interpretation where the coupling of the @xmath21 to @xmath15 is expected to vanish .
\ii ) the observation in the @xmath7 @xmath8-wave of the @xmath24 is also inconsistent with the hybrid interpretation where the @xmath7 @xmath8-wave is strongly suppressed and the @xmath25-wave is dominant .
\iii ) the measured width of the @xmath0 , @xmath17 mev , is again inconsistent with the hybrid interpretation where the @xmath21 width is less than 100 mev .
therefore , the claims for the @xmath0 can be accepted as a resonance state of exotic nature may be premature .
in fact , it is important to exhaust possible conventional @xmath4 description of the @xmath0 before resorting to more exotic interpretation such as a hybrid . in this work
, we shall discuss the possibility of the @xmath0 being the @xmath26 . as mentioned above
, a problem with identifying the @xmath0 with the @xmath26 is that its mass is about 200 mev lower than the expectation from the gi quark model .
notice that the @xmath27 and @xmath28 , both about 100 - 200 mev lower in mass than the gi quark model anticipated@xcite , turn out the excellent candidates for radial excitations@xcite , which indicates that gi quark model maybe overestimate the masses of the higher-@xmath29 radially excited mesons by about 100 - 200 mev , and therefore the @xmath26 with a mass about 1.9 gev is not implausible .
also , the mass of the @xmath26 in the spectrum integral equation@xcite is expected to be about 1.937 gev , very close to the @xmath0 mass .
therefore , the assignment of the @xmath0 as the @xmath26 seems also possible based on its mass .
however , only the @xmath0 mass information is insufficient to identify its nature , further studies of its decay dynamics are needed . the main purpose of this work is to discuss whether the @xmath26 interpretation for the @xmath0 is reasonable or not by investigating its strong decay properties in two models , the @xmath2 model and the flux - tube model .
the organization of this paper is as follows . in sec .
ii , the decay properties of the @xmath0 as the @xmath26 within the @xmath2 model and the flux - tube model are presented .
the discussions are presented in sec .
iii , and the summary and conclusion are given in secs .
the @xmath2 model and the flux - tube model which are the standard models for strong decays at least for mesons in the initial state , have been widely used to evaluate the strong decays of hadrons@xcite , since they give a good description of many of the observed decay amplitudes and partial widths of the hadrons . in this work
, we shall employ the @xmath2 model and the flux - tube model with simple harmonic oscillator ( sho ) wave functions to evaluate the two - body open - flavor strong decay widths of the initial state .
since there exists exhaustive literature on these two models , we just list the numerical values of the partial decay widths of the @xmath0 as the @xmath26 in table 1 . in our calculation , the sho wave function scale parameter @xmath30 , the pair production strength parameter @xmath31 in the @xmath2 model , the pair - creation constant @xmath32 and the string tension @xmath33 in the flux - tube model , and the constituent quark mass @xmath34 are is higher than that used by ref.@xcite ( 0.505 ) by a factor of @xmath35 due to different field conventions , constant factor in @xmath36 , etc .
the calculated results of the widths are , of course , unaffected . ]
@xmath37 gev , @xmath38 , @xmath39 , @xmath40 gev@xmath41 , @xmath42 gev , and @xmath43 gev@xcite , respectively .
the meson masses used to determine the phase space and final state momenta are is the ground scalar meson as refs.@xcite . ]
@xmath44 mev , @xmath45 mev , @xmath46 mev , @xmath47 mev , @xmath48 mev , @xmath49 mev , @xmath50 mev , @xmath51 mev , @xmath52 mev , @xmath53 mev , @xmath54 mev , @xmath55 mev , and @xmath56 mev .
.decays of the @xmath0 as the @xmath26 in the @xmath2 model and the flux - tube model .
the initial state mass is set to @xmath57 mev . [ cols="^,^,^",options="header " , ] it is clear from table 1 that the numerical results in the @xmath2 model are similar to those in the flux - tube model .
very characteristic differences between the @xmath26 and @xmath21 assignments for the @xmath0 are evident when we compare our results with the expectations from the pss model for the @xmath21@xcite .
the total width of the @xmath26 is expected to be about 223 mev in the @xmath2 model or about 233 mev in the flux - tube model , both in good agreement with the @xmath0 width ; however , the total width of the @xmath21 is expected to be less than 100 mev , at least 100 mev lower than the experiment . the partial width of the @xmath58 is significantly large , consistent with the observation in the @xmath15 channel of the @xmath0 ; whereas the @xmath59 is expected to vanish .
the @xmath60 is dominant in the @xmath8-wave and the @xmath8-wave width is significantly large , and therefore the @xmath26 should be readily observable in the @xmath7 @xmath8-wave , consistent with the observation of the @xmath0 in the @xmath7 @xmath8-wave ; while the @xmath61 is strongly suppressed in the @xmath8-wave and dominant in the @xmath25-wave .
also , for the @xmath26 , the partial width of the @xmath62 mode , the @xmath63-wave widths of the @xmath64 and @xmath65 modes , and the @xmath66-wave width of the @xmath7 mode are not zero , especially the @xmath67 @xmath63-wave width is significantly large ; whereas for the @xmath21 , all these widths vanish exactly .
the further experimental study on these decay modes are also important to examine whether the @xmath0 is the @xmath26 or the @xmath21 . from these remarkable discriminants between the @xmath26 and the @xmath21 ,
it is clear that the available experimental evidence for the @xmath0 is consistent with it being the @xmath26 rather than the @xmath21 for the @xmath26 is about 4 , inconsistent with the measured value of @xmath68@xcite which agrees with the pss model prediction of about 23 for the @xmath69@xcite .
notice that the systematic error is not estimated in this measured datum , and this datum is not used for averages , fits , limits , etc . by pdg2008@xcite .
the further confirmation of this ratio is needed .
] , assuming the @xmath2 model and the flux - tube model are accurate . in order to test the robustness of our results ,
the dependence of the predicted results on the initial state mass @xmath70 and the sho function scale parameter @xmath30 is studied .
we show the variation of the total width of the @xmath26 with @xmath70 and @xmath30 in fig .
1 . in both the @xmath2 model and the flux - tube model ,
the total width of the @xmath26 becomes large with the increase of the @xmath70 , and it always lies in the width range of the @xmath0 . when the @xmath30 varies from 300 to 500 mev , in the flux - tube model the @xmath26 width varies from about 230 to 260 mev , depending weakly on the @xmath30 , while in the @xmath2 model it varies dramatically with the @xmath30 . in order to reproduce the @xmath0 width in the @xmath2 model
, it requires @xmath71 mev , overlapping the typical value of about @xmath72 mev used in the computation of the light meson decays for the sho wave functions with a common @xmath30@xcite .
the partial widths of the @xmath26 versus the @xmath70 are shown in fig .
2 , where the variations of the partial widths of the @xmath26 with the @xmath70 in the @xmath2 model are similar to those in the flux - tube model . the partial widths increase when the @xmath70 increases and the dominant decay modes are @xmath67 , @xmath15 , @xmath7 , @xmath73 , @xmath65 and @xmath9 .
the partial widths of the @xmath26 versus the @xmath30 are shown in fig .
we see from fig .
3 that in two models , @xmath67 , @xmath15 , @xmath7 , @xmath73 , @xmath65 and @xmath9 are still the dominant decay modes when the @xmath30 varies . for small @xmath30 ( @xmath74 mev )
, @xmath73 dominates @xmath9 , however for large @xmath30 ( @xmath75 mev ) , @xmath9 dominates @xmath73 . in the vicinity of @xmath76 mev , @xmath77 .
the similar behavior also exists for the modes @xmath14 and @xmath78 .
the measurement of the @xmath79 and @xmath80 for the @xmath0 would be useful for the reasonable choice for the @xmath30 .
we note that for the @xmath26 , the @xmath7 @xmath8-wave dominates the @xmath25-wave and the @xmath67 @xmath63-wave dominates the @xmath81-wave , which is unusual because in most cases the lower partial waves are dominant . as mentioned above , these results are remarkably different with the expectations from the pss model@xcite for the @xmath21 .
we find that in both the @xmath2 model and the flux - tube model , the @xmath63-wave always dominates the @xmath81-wave for the @xmath82 and the @xmath8-wave always dominates the @xmath25-wave for the @xmath60 when the @xmath70 varies in the mass range of the @xmath0 and @xmath30 varies in the range 370 - 420 mev , the required range for reproducing the @xmath0 width in the @xmath2 model mentioned above . determining these partial widths ratios experimentally is very important to distinguish the @xmath26 interpretation from the @xmath21 assignment for the @xmath0 .
generally speaking , the pure @xmath26 can mix with the pure @xmath21 to produce the physical state @xmath0 .
the available experimental evidence for the @xmath0 is in favor of the @xmath26 interpretation for the @xmath0 based on the remarkably different decay patterns of the @xmath26 and @xmath21 , but it is insufficient to quantitatively determine the @xmath4-hybrid content of the @xmath0 being the mixture of the @xmath26 and @xmath21 , the measured partial widths of the @xmath0 are needed to determine the hybrid - quarkonium content of the @xmath0 quantitatively . ] , which is essential to confirm or refute that the possibility of the hybrid admixture in the @xmath0 .
therefore , the possibility of the @xmath0 being in fact a mixture of the @xmath26 and @xmath21 might exist at present time .
we can qualitatively estimate the hybrid component of the @xmath0 would be small based on its available experimental information .
the @xmath83 is expected to vanish from the pss model@xcite , therefore , the observation of the @xmath0 in the @xmath15 channel@xcite makes that the substantial hybrid admixture in the @xmath0 seems impossible .
however , it should be noted that the @xmath0 signal in the @xmath15 channel was observed only by the e852 collaboration@xcite , and even it is not clear whether the unitarity conserving fit to the @xmath15 mass distributions would need to have the @xmath0 decaying to @xmath15 or not .
therefore , further evidence is needed to confirm whether the hybrid component of the @xmath0 is small or not . fortunately , as mentioned in sec .
i , the @xmath0 has been observed by three different groups in the @xmath7 @xmath8-wave , which implies that the @xmath21 component of the @xmath0 would be small because the @xmath61 is strongly suppressed in the @xmath8-wave .
similarly , the further experimental information of the @xmath0 in the @xmath78 and @xmath84_{l=3}$ ] channels would be useful to shed light on this issue . finally , our predicted @xmath85 for the @xmath26 inconsistent with the measurement of the e852 collaboration@xcite may be a hint for the @xmath0 being in fact a mixture of the @xmath26 and @xmath21 , and the small hybrid admixture in the @xmath0 might make this measured ratio shift from the predicted value for the pure @xmath26 .
the strong decays of the @xmath0 as the @xmath26 are investigated in both the @xmath2 model and the flux - tube model .
the overall behaviors of the decay modes in the @xmath2 model are similar to those in the flux - tube model .
the decay properties of the @xmath26 and the @xmath21 are remarkably different .
the decay modes @xmath67 , @xmath15 , @xmath7 , @xmath86 , @xmath62 and @xmath65 are crucial for distinguishing the conventional quarkonium interpretation from the hybrid assignment of the @xmath0 .
the available experimental evidence for the @xmath0 is consistent with it being the conventional @xmath1 @xmath4 meson rather than the light @xmath3 hybrid .
the possibility of the small hybrid admixture in the @xmath0 might exist .
further experimental study on the partial widths of the @xmath0 is desirable .
we tend to conclude that the @xmath0 is the convincing @xmath1 @xmath4 state or the @xmath1 @xmath4 with small hybrid admixture .
this work is supported in part by hancet under contract no .
2006hancet-02 , and by the program for youthful teachers in university of henan province . n. isgur and j. paton , phys .
d * 31 * , 2910 ( 1985 ) ; t. barnes , f. e. close , and e. s. swanson , phys . rev .
d * 52 * , 5242 ( 1995 ) .
s. godfrey and n. isgur , phys .
d * 32 * , 189 ( 1985 ) .
p. r. page , e. s. swanson , and a. p. szczepaniak , phys .
d * 59 * , 034016 ( 1999 ) .
t. barnes , f. e. close , p. r. page and e. s. swanson , phys .
d * 55 * , 4157 ( 1997 ) .
v. v. anisovich , l. g. dakhno , m. a. matveev , v. a. nikonov , and a. v. sarantsev , phys .
nucl . * 70 * , 450 ( 2007 ) .
a. le yaouanc , l. oliver , o. pene , and j - c .
raynal , phys .
2223 ( 1973 ) ; phys . rev .
d * 9 * , 1415 ( 1974 ) ; phys .
d * 11 * , 1272 ( 1975 ) ; phys .
b * 71 * , 397 ( 1977 ) ; phys .
b * 72*,57 ( 1977 ) .
a. le yaouanc , l. oliver , o. pene , and j - c .
raynal , hadron transitons in the quark model ( gordon and breach science publishers , new york , 1988 ) .
w. roberts and b. silvestr - brac , few - body syst . * 11 * , 171 ( 1992 ) .
h. g. blundel , hep - ph/9608473 .
r. kokoski and n. isgur , phys .
d * 35 * , 907 ( 1987 ) .
p. geiger and e. s. swanson , phys .
d * 50 * , 6855 ( 1994 ) .
blundell and s. godfrey , phys .
d * 53*,3700 ( 1996 ) .
e. s. ackleh , t. barnes and e. s. swanson , phys .
d * 54 * , 6811 ( 1996 ) .
t. barnes , n. black and p. r. page , phys .
d * 68 * , 054014 ( 2003 ) .
f. e. close and e. s. swanson , phys .
d * 72 * , 094004 ( 2005 ) .
s. capstick and n. isgur , phys .
d * 34 * , 2809 ( 1986 ) ; s. capstick and w. roberts , phys .
d * 49 * 4570 ( 1994 ) ; h. g. blundell , s. godfrey , and b. phelps , phys .
d * 53 * , 3712 ( 1996 ) ; l. burakovsky and p. r. page , phys .
d * 62 * , 014011 ( 2000 ) ; h. q. zhou , r. g. ping , and b. s. zou , phys .
b * 611 * , 123 ( 2005 ) ; t. barnes , s. godfrey and e. s. swanson , phys .
d * 72 * , 054026 ( 2005 ) ; j. lu , w. z. deng , x. l. chen , and s. l. zhu , phys .
d * 73 * , 054012 ( 2006 ) ; b. zhang , x. liu , w. z. deng , and s. l. zhu , eur .
j. c * 50 * , 617 ( 2007 ) ; f. e. close , c. e. thomas , o. lakhina , and e. s. swanson , phys .
b * 647 * , 159 ( 2007 ) ; c. chen , x. l. chen , x. liu , w. z. deng , and s. l. zhu , phys .
d * 75 * , 094017 ( 2007 ) ; g. j. ding and m. l. yan , phys .
b * 657 * , 49 ( 2007 )
. g. j. ding , j. j. zhu , and m. l. yan , phys .
d * 77 * , 014033 ( 2008 ) ; d. m. li and b. ma , phys .
d * 77 * , 074004 ( 2008 ) ; d. m. li and b. ma , phys .
rev , d * 77 * , 094021 ( 2008 ) ; d. m. li and s. zhou , phys .
d * 78 * , 054013 ( 2008 ) . | the strong decays of the @xmath0 as the @xmath1 quark - antiquark state are investigated in the @xmath2 model and the flux - tube model , respectively .
the results are similar in the two models .
it is found that the decay patterns of the conventional @xmath1 meson and the @xmath3 light hybrid are very different , and the experimental evidence for the @xmath0 is consistent with it being the conventional @xmath1 meson rather than the @xmath3 light hybrid .
the possibility of the @xmath0 being a mixture of the conventional @xmath4 and the hybrid is discussed . -1 cm mesons , @xmath2 model , flux - tube model , decays * pacs numbers:*14.40 .
cs , 12.39.-x , 13.25.gv |
wave scattering from irregular surfaces continues to present formidable theoretical and computational challenges @xcite , especially with regard to analytical treatment of statistics , and numerical solution for wave incidence at low grazing angles @xcite , where the insonified / illuminated region may become very large .
computationally , the cost of the necessary matrix inversion scales badly with wavelength and domain size and can rapidly become prohibitive ; this is compounded by the large number of green s function evaluations , whose overall cost is therefore sensitive to the form which this function takes . under the assumption of purely forward - scattering ,
a successful approach has been the parabolic integral equation method ( pie)@xcite .
this makes use of a ` one - way ' parabolic equation ( pe ) green s function , leading to the replacement of the helmholtz integral equations by their small - angle analogue . for 2d problems this green s function takes a particularly tractable form ;
this , together with the volterra ( one - sided ) form of the governing integral operator , affords the key advantage of high numerical efficiency , and in the perturbation regime allows derivation of analytical results @xcite .
nevertheless , the method yields no information about the field scattered back towards the source . on the other hand , where backscatter is required ,
operator series solution methods such as left - right splitting and method of ordered multiple interactions @xcite have proved highly versatile , in both 2 and 3 dimensions .
these use the full free - space green s function and proceed by expanding the surface fields about the dominant ` forward - going ' component , and thereby circumvent the difficulties of tackling the full helmholtz equations . in this paper
we combine these approaches , extending the standard pie description to a ` two - way ' method , thus allowing for both left- and right - travelling waves .
this is obtained in the obvious way by replacing the parabolic equation green s function by a form symmetrical in range
. the integral operator can be split into left- and right - going parts ; under the assumption that forward scattering dominates , the solution can then be written as a series and truncated .
every term of this series is a product of volterra operators and is therefore treated as efficiently as the standard pie method , which corresponds approximately to truncation at the first term . in the second part of the paper
we impose the additional restriction to the perturbation regime of small surface height @xmath0 , within which analytical expressions for the mean field and autocorrelation function are obtained .
this extends the corresponding results@xcite derived under the pie method .
the approach there was first to obtain the scattered field to second order in @xmath0 at the mean surface plane , and find the far - field under the assumption that propagation outwards from the surface is governed by the full helmholtz equation . in the standard pe case
, this modification allows a small amount of backscatter , but precludes any backscatter enhancement which can be thought of as due to coherent addition of reversible paths@xcite , because interactions at the surface are assumed to be take place in the forward direction only .
the formulation presented here allows one to remove this restriction , and separate the forward and backward going interactions to various orders , although this aspect is not explored in detail here .
in particular this method produces a correction term , whose statistics can be obtained in the perturbation regime .
the paper is organised as follows : the standard parabolic integral equation method and preliminary results are given in section [ i ] .
in section [ ii ] the full two - way parabolic integral equation method is set out , and the iterative solution explained .
analytical results for the statistics under the extended method are derived in section [ iii ] .
we consider the problem of a scalar time - harmonic wave field @xmath1 scattered from a one - dimensional rough surface @xmath2 with a pressure release boundary condition .
( equivalently , @xmath1 is an electromagnetic @xmath3 or @xmath4 polarised wave and @xmath5 is a perfectly conducting corrugated surface whose generator is in the plane of incidence . )
the wavefield has wavenumber @xmath6 and is governed by the wave equation @xmath7 .
the coordinate axes are @xmath8 and @xmath9 where @xmath8 is the horizontal and @xmath9 is the vertical , directed out of the medium ( see fig .
[ fig.1 ] ) .
angles of incidence and scatter are assumed to be small with respect to the positive @xmath8-direction .
it will be assumed that the surface is statistically stationary to second order , i.e. its mean and autocorrelation function are translationally invariant .
we may choose coordinates so that @xmath2 has mean zero .
the autocorrelation function @xmath10 is denoted by @xmath11 , and we assume that @xmath12 at large separations @xmath13 .
( the angled brackets here denote the ensemble average . )
then @xmath14 is the variance of surface height , so that the surface roughness is of order @xmath15 .
1 true cm schematic view of scattering geometry , title="fig:",height=226 ] since the field components propagate predominantly around the @xmath8-direction , we can define a slowly - varying part @xmath16 by ( x , z ) = p(x , z ) ( -ikx ) . slowly varying incident and scattered components @xmath17 and @xmath18
are defined similarly , so that @xmath19 .
it may be assumed that @xmath20 for @xmath21 , so that the area of surface insonification is restricted , as it would be for example in the case of a directed gaussian beam .
the governing equations for the standard parabolic equation method @xcite are then _ i ( * r*_s ) = - _ 0^x g_p ( * r*_s ; * r * ) dx [ 1 ] where both @xmath22 , @xmath23 lie on the surface ; and now , equations ( [ 1 ] ) and ( [ 2 ] ) do not apply to plane wave scattering at small or negative @xmath8 because of the truncated lower limit of integration , equivalent to the restricted surface insonification .
nevertheless , we can formally apply the integral equation to a plane wave , to obtain a solution which will be physically meaningful and asymptotically accurate at large values of @xmath8 .
this procedure has been used@xcite to derive the field statistics ; where necessary we will assume that @xmath8 is sufficiently large for this to hold .
consider an incident plane wave @xmath30)$ ] , where @xmath31 is the angle with respect to the vertical .
the grazing angle is then denoted @xmath32 ( see fig .
[ fig.1 ] ) .
this plane wave has slowly - varying component @xmath33)$ ] , where s = - 1 , [ 5 ] which we refer to as the reduced plane wave .
in this section the two - way version of the pie method will be described , and the iterative solution will be given .
this provides an efficient means of calculating the back - scattered component at small angles of scatter . the governing equations ( [ 1 ] ) , ( [ 2 ] )
must first be modified to take into account scattering from the right . to do this , we simply replace @xmath26 by its symmetrical analogue @xmath34 .
this form arises if we apply the small angle approximation described in section [ i ] to the full free space green s function without requiring @xmath35 to vanish when @xmath36 .
we thus obtain g(x , z ; x , z ) \ { lr & = , x < x & = x x [ 6 ] .
the factor @xmath37 $ ] arises for @xmath38 because we are solving for the reduced wave @xmath16 . applying this green s function to the reduced wave @xmath16 we obtain _ s ( x , z ) = _ 0^g(, ) ( )z dx [ 7 ] where @xmath39 , @xmath40 .
this is the analogue of equation ( 2 ) , effectively containing a back - scatter correction .
taking the limit of ( 7 ) as @xmath41 yields an integral equation relating the incident field to the scattered field at the surface : ( x , h(x ) ) = -_0^g(_s, ) ( )z dx [ 8 ] where now @xmath42 , @xmath40 both lie on the surface .
( note that the addition of a correction to the parabolic equation is closely related to a method proposed by thorsos@xcite . )
equations ( 7 ) , ( 8) can be written in operator notation : _ s ( x , z ) = -(l+r)z [ 9 ] ( x , h(x ) ) = ( l+r)z [ 10 ] where @xmath43 , @xmath44 are defined by lf(x , z)=_0^x g(, ) f(x ) dx , rf(x , z)=_x^g(, ) f(x ) dx and @xmath39 , @xmath40 . these integral operators and their inverses
are volterra , or ` one - sided ' in an obvious sense .
the main computational task in any such boundary integral method is the inversion of the integral equation ( [ 10 ] ) .
one of the principal advantages of the standard forward - going pie method ( equations ( [ 1])-([2 ] ) ) is that its one - way form allows gaussian elimination to be used , so that inversion is highly efficient . in the above two - way formulation this advantage
is initially lost , since direct inversion of @xmath45 in eq .
( [ 10 ] ) offers no benefit compared with solving the full helmholtz equations .
however , the computational advantage can be regained by forming an iterative series solution , in which each term is a product of volterra integral operators .
integral equation ( [ 10 ] ) has formal solution = ( l+r)^-1[11 ] which can be expanded in a series = [ 12 ] under the assumption that @xmath44 is small in the following sense the series ( [ 12 ] ) is convergent , as is already required implicitly for the standard pie solution ; the series can then be truncated after finitely many terms . by ` small ' we mean that @xmath46 is small for all terms @xmath47 in the series .
it can be shown that this assumption is indeed justified at low grazing angles for surfaces whose slopes are not too large , since the kernel of @xmath44 oscillates rapidly especially at small wavelengths .
it is nevertheless difficult to give this a precise range of validity , and we will not attempt to do so here .
solution for the field can therefore be obtained by truncating the series ( [ 12 ] ) and substituting into the integral ( [ 7 ] ) .
the _ first term _
@xmath48 in series ( [ 12 ] ) corresponds to the solution for @xmath29 under the standard pie method ( e.g. @xcite ) .
denote this first approximation by @xmath49 , i.e. = l^-1 . [ 13 ] note however that the integral ( [ 7 ] ) allows for outgoing components scattered to the left , unlike its pie analogue ( [ 2 ] ) , so even this lowest order truncation gives backscatter .
this can be considered the _
direct backscatter _ component .
truncation of ( [ 12 ] ) at the _ second term _ gives : + c [ 14 ] where @xmath50 is a correction term , c= l^-1rl^-1_i .
[ 15 ] the above expression will be used in section [ iii ] to obtain some statistical measure of the backscattered component in the perturbation regime of small surface height .
we remark that this is the lowest - order truncation consistent with reversible ray paths . the general term of ( [ 12 ] )
is a product of the operators @xmath51 and @xmath44 .
evaluation of the integral @xmath44 is straightforward . for computational purposes
we assume that the incident wave insonifies only a finite region of the rough surface ; the source may for example be a gaussian beam . a finite upper limit of integration @xmath52 , say ,
may then be assumed .
numerical inversion of @xmath43 is also highly efficient since discretization of @xmath43 gives rise to a lower - triangular matrix .
this has been described elsewhere ( e.g. @xcite ) and will only be summarized here .
consider the equation @xmath53 obtained by truncating ( [ 12 ] ) at the first term .
this equation is discretized with respect to range @xmath8 using , say , @xmath54 equally spaced points @xmath55 .
this then yields a matrix equation @xmath56 in which the matrix @xmath57 is lower - triangular .
numerical inversion of this expression is carried out by gaussian elimination , requiring @xmath58 operations , which compares with @xmath59 operations required to treat the full helmholtz integral equation .
the solution is thereby obtained for the first term , @xmath49 .
typically only one further term , @xmath60 , will be required .
the simplest way to obtain this is to discretize the integral @xmath44 , evaluate @xmath61 numerically , and then to solve l^-1r= z - by gaussian elimination as before .
the evaluation of the integral @xmath44 also requires @xmath58 operations .
subsequent terms in the series may be obtained similarly .
the computation can be simplified further in the perturbation regime of small scaled surface height @xmath62 , if the operators @xmath43 and @xmath44 are replaced by the flat surface forms in the calculation of the correction term @xmath50 .
this is described in the section [ iii ] .
the mean field and higher moments based on the standard parabolic equation approximation were obtained elsewhere@xcite to second order in surface height in the case of pure forward scattering . in this section the statistics of the backscatter correction ( eq .
( [ 14 ] ) ) due to the two - way pie method will be derived .
suppose that a reduced plane wave @xmath63)$ ] is incident on the rough surface at an angle @xmath31 measured from the normal .
we first summarize the perturbational calculation used to obtain the scattered field statistics previously .
suppose that a plane @xmath64 , say , can be chosen ` close ' to every point on the surface .
the scattered field is obtained to second order in surface height along this plane , for a given incident plane wave , and the statistics are found from this .
statistical results obtained in this way do not depend on the choice of @xmath65 so for convenience we may set @xmath66 .
an expression is thus found for the scattered field _
s(x,0 ) = -_i^(x , h ) - h - 12 h^2 ^2 _ i(x,0 ) z^2 + o(^3 ) .
[ 16 ] the only term here which is not known _ a priori _ is @xmath29 .
the standard pie solution @xmath49 for @xmath29 is given@xcite to second order in @xmath0 by : = - 1 d dx_0^x _ i^(x,h(x ) ) dx .
[ 17 ] this arises from ( [ 13 ] ) by substitution of the flat surface form of @xmath43 ( see ( [ 20 ] ) below ) .
denote by @xmath67 the approximation to @xmath18 obtained by substituting ( [ 17 ] ) in ( [ 16 ] ) , so that
_ s(x,0 ) - _ i^(x , h ) & + h + & - h^22 ^ 2_i^(x , 0)z^2 . [ 18 ] we wish to calculate the backscatter correction to this expression due to the replacement of @xmath29 in ( [ 16 ] ) by the corrected two - way pe solution @xmath68 ( equations ( [ 14 ] ) , ( [ 15 ] ) ) .
we therefore repeat the above derivation replacing ( [ 13 ] ) by ( [ 14 ] ) , to obtain _ s ( x,0 ) = _ s(x,0 ) + h(x ) c(x ) .
[ 19 ] since the correction term @xmath50 appears here with a factor @xmath5 , it is necessary to evaluate it only to order @xmath15 . expanding @xmath43 and @xmath44 ( eqs .
( [ 9])-([10 ] ) ) in surface height @xmath2 , it is seen that @xmath69 , @xmath70 , where @xmath71 , @xmath72 denote the deterministic ( i.e. flat surface ) forms of the operators @xmath43 and @xmath44 respectively : l_0 = _
0^x 1 dx , r_0 = _
x^ dx [ 20 ] in evaluating @xmath50 ( eq . ( [ 15 ] ) ) to order @xmath15 we may thus ignore fluctuating parts of the operators , and replace @xmath43 , @xmath44 by @xmath71 , @xmath72 respectively .
we can therefore write c = l_0 ^ -1 r_0 + o(^2 ) .
[ 21 ] an expression of the form @xmath73 is abel s integral equation , which has the well - known solution@xcite g(x ) = 1ddx _ 0^x 1 f(y ) dy . now to first order in @xmath5 , @xmath49 in ( 21 ) is given@xcite by ( r ) - . [ 22 ] where , for large @xmath74 , @xmath75 takes the form ( see eq .
( [ 15 ] ) of @xcite ) d_(r ) ~-2ik e^iksr [ 23 ] and @xmath76 is an integral i(r ) = _ 0^r ikh(r ) dr .
[ 24 ] therefore @xmath75 and @xmath77 are @xmath78 and @xmath79 respectively , so that in eq .
( [ 21 ] ) @xmath50 becomes c(x ) = 1 ^ 2 ddx .
[ 25 ] to second order in surface height the scattered field @xmath80 at the mean surface is therefore described by eq .
( [ 19 ] ) , with @xmath50 given by ( [ 25 ] ) .
the effect of the correction term @xmath50 on the scattered field statistics can now be examined .
we first find the mean field @xmath81 .
it is sufficient to obtain this quantity on the mean surface plane @xmath82 , using equation ( [ 19 ] ) , i.e. = < _ s(x ) > + < h(x)c(x ) > .
the solution for @xmath83 has been obtained previously@xcite , and we can restrict attention to finding the correction @xmath84 to this . denote the correlation @xmath85 by @xmath86 for any @xmath87 , @xmath8 , i.e. ( x , x)=<h(x)c(x ) > .
consider first the function @xmath88 . since @xmath89 vanishes , eq .
( [ 22 ] ) gives = - .
[ 26 ] now from eq .
( [ 25 ] ) ( x , x ) = . the term @xmath90 can be taken under the integral signs as part of the operand of @xmath91 .
the order of integration and averaging can then be reversed so that , by ( 26 ) , ( x , x ) = -1 ^ 2 .
[ 27 ] consider the term @xmath92 in the inner integrand . by (
[ 24 ] ) , = & + = & ik [ 28 ] this may be substituted into ( [ 27 ] ) to give an analytical expression for the correlation @xmath85 .
we can simplify this expression by evaluating the derivatives explicitly .
the term @xmath93 is independent of @xmath74 , so writing = f(r , r ) [ 29 ] the expression ( [ 28 ] ) becomes = ik @xmath94 \label{30 } \\ = & ik\cos\theta ~\lim_{\epsilon\rightarrow 0 } { 1\over\epsilon } \left[k_1 + k_2 - k_3 \right ] \nonumber\end{aligned}\ ] ] where @xmath95 consider these three integrals in detail .
the first gives k_1= 1 _
0^ f(r+,r)(x - r)dr ( x)_0^ 1 dr -0.8 cm @xmath96 \label{32 } \\
\cong & { \rho(x ) \over \alpha\sqrt{r } } \nonumber\end{aligned}\ ] ] using a taylor expansion in @xmath97 . changing variables , @xmath98 in can be written _ ^r+ f(r+,r)(x - r)dr = _ 0^r
f(r+,r+ ) ( x - r- ) dr . [ 33 ] now f(r+,r+)=e^iks(r+ ) = e^iksf(r , r so from ( [ 33 ] ) _ ^r+ f(r+,r)(x - r)dr = _
0^r e^iks f(r , r)(x - r- ) dr .
[ 34 ] thus the difference @xmath99 in ( [ 30 ] ) becomes & _
0^r f(r , r ) dr + & _
0^r f(r , r ) dr [ 35 ] where @xmath100 , which may be assumed to be differentiable , has been expanded to leading order in @xmath97 . substituting ( [ 32 ] ) and
( [ 35 ] ) in ( [ 28 ] ) , we obtain = ik\ { ( x ) +
_ 0^r e^iksr dr } .
[ 36 ] this removes the derivative with respect to @xmath8 in ( [ 27 ] ) , and indeed for several important autocorrelation functions
( [ 36 ] ) can be written in closed form .
the term @xmath101 is an artifact of the finite lower bound of integration and can be dropped , as we can assume the range variable @xmath87 to be large .
equation ( [ 27 ] ) therefore becomes ( x , x ) = -ik^3 [ 37 ] where r(x , r)=iks(x - r)-d(x - r)dx .
[ 38 ] the derivative with respect to @xmath8 in ( [ 37 ] ) can be evaluated similarly , and after further manipulation ( see appendix ) the required expression can be written , setting @xmath102 , = -ik^3 & _ x = x [ 39 ] where = \{(1+ik)r(x , r ) + drdr } . [ 40 ] the main quantity of interest is the angular spectrum of intensity , which may be defined as the fourier transform of the autocorrelation function ( i.e. the second moment ) of the scattered field .
this remains essentially unchanged with distance from the surface , so that we may again concentrate on obtaining the form on the mean surface plane , @xmath82 .
denote the second moment m_2(x , y)= where @xmath103 indicates the complex conjugate , and denote its approximation using the standard parabolic equation method by ( x , y ) .
the perturbational solution of @xmath104 was obtained in @xcite .
it is relatively straightforward to express @xmath105 , to second order in surface height under the present two - way pie method , as the sum of @xmath104 and correction terms .
these additional terms , which are expected to be small , represent the ` indirect ' contribution to the backscatter . from ( [ 19 ] )
we have _ s ( x ) _
s^*(y ) = ( x ) ^*(y ) + ( x ) h(y ) c^*(y ) + ^*(y ) h(x ) c(x ) + h(x)h(y)c(x)c^*(y ) .
[ 41 ] we can write @xmath106 and @xmath50 to zero and first order in surface height , = _ 0 + _ 1 + o(^2 ) where @xcite _ 0 ( x ) & = - e^iksx + _ 1 ( x ) & = - 2ik h(x ) e^iksx h(x ) d_(x ) , [ 42 ] and c = c_0+c_1 [ 43 ] where @xmath107 therefore to @xmath108 the second moment can be written & m_2(x , y ) = ( x , y ) + _ 0(x ) + c_0^*(y ) + & + _ 0^*(y ) + c_0(x ) + ( x - y ) c_0(x)c_0^*(y ) . [ 44 ] since @xmath109 , equation ( [ 44 ] ) can be expressed as & m_2(x , y ) = ( x , y ) + _
0(x ) ^*(y ) + _ 0^*(y ) ( x ) + & + ( x - y ) c_0(x)c_0^*(y ) + c_0^*(y ) + c_0(x ) .
[ 45 ] in this equation , only the last two terms remain to be determined . from ( [ 42 ] ) , @xmath110 is just = ( x - y ) d_[46 ] and similarly for @xmath111 so that ( [ 45 ] ) becomes & m_2(x , y ) = ( x , y ) + _ 0(x ) ^*(y ) + _ 0^*(y ) ( x ) + & + ( ) [ 47 ] where @xmath112 .
the parabolic integral equation method has been extended here to allow the calculation of backscatter of due to a scalar wave impinging on a rough surface at low grazing angles .
the solution is written in terms of a series of volterra operators , each of which is easily evaluated , and which allows examination of multiple scattering resulting from increasing orders of surface interaction .
truncation at the first term the leading forward- and back - scattered components ; higher - order multiple scattering are available from subsequent terms .
the parabolic green s function is applicable for wave components at low angles of incidence and scatter , which imply small surface slopes , but without restriction on surface heights . with the additional assumption of small surface heights , analytical solutions
have then been obtained , to second order in height , for the mean field and its autocorrelation .
these provide backscatter corrections to the solutions given in the purely forward - scattered case@xcite with the potential for further insight into the role of different orders of multiple scattering .
( small height perturbation theory derived directly from helmholtz equation has of course been well established for many years and yields particularly simple single scattering results .
the results here are from a different perspective ; the first term already includes multiple - forward - scattering , and subsequent terms incorporate back- and forward - scatter contributions systematically at higher orders . ) in the context of long - range propagation at low grazing angles , parabolic equation methods remain very widely used . in this regime
the form of the green s function together with the series decomposition provide computational efficiency and the means to extend existing pe methods to include backscatter , in addition to yielding tractable analytical results for statistical moments .
these benefits should , nevertheless , be put in context .
the computational advantages of the pe green s function over the full free space green s function are lost in fully 3-dimensional problems ( since evaluation of the 3d pe green s function is computational expensive ) , or those for which wide - angle scatter needs to be taken into account . on the other hand there
remains a need for further theoretical understanding of the mechanisms of enhanced and multiple backscatter , and the approach here may be applied in a more general setting .
computational and theoretical results in application to long - range propagation over rough sea surfaces will appear in a separate paper .
we can write the expression ( [ 34 ] ) as ( x , x)= -ik^3 ddx_0^x g(x , y ) h(x , y ) dy [ a.1 ] where g(x , y ) = 1 , [ a.2 ] h(x , y)= _
_ 0^r e^iksr r(x , r ) dr dr , [ a.3 ] and @xmath44 is given by ( [ 35 ] ) .
differentiation with respect to @xmath8 is carried out as for the @xmath74-derivative ( equations ( [ 27])-([33 ] ) ) : the @xmath8-derivative is thus expressed as a limit of a finite difference , and the integral split into three parts , ( x , x ) = -ik^3 _ 0 1 where @xmath113 we thereby obtain _ 0^x g(x ,
y ) h(x , y ) dy = 2 h(x , y ) + _
0^x g(x , y ) dy .[a.4 ] the term @xmath114 is then = ddy_y^a(y , r ) j(x , r ) dr [ a.5 ] where a(y , r ) = e^ikr , [ a.6 ] j(x , r)= _ 0^r e^iksr r(x , r ) dr [ a.7 ] treating the derivative as before gives = - _ y^a(y , r ) ( dj(x , r)dr + ikj(x , r ) ) dr .
[ a.8 ] finally , = ddr _ 0^r e^iksr r(x , r ) dr [ a.9 ] from which we similarly get = r(x,0 ) + _
0^r e^iksr \ { iks r(x , r ) + drdr } dr .
[ a.10 ] as before ( see ( [ 34 ] ) ) the expression @xmath115 vanishes for large @xmath87 and can be dropped . successively substituting ( [ a.8 ] ) , ( [ a.10 ] ) , ( [ a.3 ] ) and ( [ a.4 ] ) into ( [ a.1 ] ) , we eventually obtain = -ik^3 @xmath116 \label{a.11 } \end{aligned}\ ] ] where = \{(1+ik[1+s])r(x , r ) + drdr } .
[ a.12 ] in this expression , @xmath44 is given by ( [ 35 ] ) , so that = iksd(x - r)dr - d^2(x - r)dx^2 .
[ a.13 ] it is clear then that the correction term introduces a higher - order dependence on the correlation function .
i. simonsen , a.a .
maradudin , & t.a .
leskova , scattering of electromagnetic waves from two - dimensional randomly rough perfectly conducting surfaces : the full angular intensity distribution , _ physical review a _ , * 81 * 013806 ( 2010 ) 1 - 13 m .- j .
kim , h.m .
berenyi & r.e .
burge , scattering of scalar waves by two - dimensional gratings of arbitrary shape ; application to rough surfaces near grazing incidence , _ proc .
a _ * 446 * ( 1994 ) 1 - 20 . c. qi , z. zhao , w. yang , zp nie , & g. chen , electromagnetic scattering and doppler analysis of three - dimensional breaking wave crests at low - grazing angles , _ progress in electromagnetics _ , * 119 * ( 2011 ) 239 - 252 .
p. tran , calculation of the scattering of electromagnetic waves from a two - dimensional perfectly conducting surface using the method of ordered multiple interaction , _ waves in random media _ , * 7 * , 295 - 302 ( 1997 ) .
pino , l. landesa , j.l .
rodriguez , f. obelleiro , & r.j .
burkholder , the generalized forward - backward method for analyzing the scattering from targets on ocean - like rough surfaces _ ieee trans .
_ , * 47 * , 961 - 969 ( 1999 )
. a. ishimaru , j.s .
chen , p. phu , & k. yoshitomi , numerical , analytical , and experimental studies of scattering from very rough surfaces and backscattering enhancement , _ waves in random media _ * 1 * ( 1991 ) s91-s107 . | this paper extends the parabolic integral equation method , which is very effective for forward scattering from rough surfaces , to include backscatter .
this is done by applying left - right splitting to a modified two - way governing integral operator , to express the solution as a series of volterra operators ; this series describes successively higher - order surface interactions between forward and backward going components , and allows highly efficient numerical evaluation .
this and equivalent methods such as ordered multiple interactions have been developed for the full helmholtz integral equations , but not previously applied to the parabolic green s function .
in addition , the form of this green s function allows the mean field and autocorrelation to be found analytically to second order in surface height .
these may be regarded as backscatter corrections to the standard parabolic integral equation method .
= cmr8 1 true cm department of applied mathematics and theoretical physics , the university of cambridge cb3 0wa , uk 1.5 true cm |
there has been enormous progress over the past decade in discovering galaxies and qsos at increasingly high redshifts .
we are now probing far enough back in time that the universe at these early epochs was fundamentally different from its predominantly ionized state today .
observations of @xmath5 qsos ( becker et al .
2001 , fan et al .
2002 ) show near - complete absorption of flux at wavelengths short - ward of lyman-@xmath2 ( gunn & peterson 1965 ) , indicating that the universe is optically thick to this line , and that the neutral fraction of hydrogen is much greater than at lower redshifts .
_ wmap _ results from the cosmic microwave background indicate that the universe was completely neutral at redshifts of @xmath6 ( spergel et al .
there is an ongoing debate as to what reionized the universe at @xmath3 : is it agn or ionizing photons ( produced in hot , massive , short - lived stars ) escaping from star forming galaxies ?
agn appear to be very under - abundant at these epochs ( e.g. , dijkstra , haiman & loeb 2004 ) , so to address the reionization issue it is crucial to know the global star formation rate at high redshifts , along with the escape fraction for these ionizing photons . in this review talk
i will outline the work done in identifying star - forming galaxies within the first billion years of the big bang through deep imaging in the optical and near - infrared with the _ hubble space telescope ( hst ) _ and spectroscopy with large telescopes from the ground .
i will also discuss what can be learned about the stellar masses and formation epochs of these galaxies , and what role they may have played in the reionization of the universe .
the standard `` concordance '' cosmology is used throughout ( @xmath7 , @xmath8 , and use @xmath9 ) .
all magnitudes are on the @xmath10 system .
to address the contribution of star - forming galaxies to reionization , we need to reliably identify galaxies at @xmath11 and measure their integrated rest - frame uv flux .
magnitude - limited spectroscopic surveys have shown that the high - redshift tail is small , so simply taking spectra of everything in a deep field is an expensive way of discovering a handful of high - redshift galaxies .
some way of isolating candidate high - redshift galaxies from the foreground population is required to provide targets for follow - up spectroscopic confirmation .
one technique is to use narrow - band imaging to identify galaxies where highly - redshifted line emission falls within the passband .
after many years of fruitless searches in the 1980s and early 1990s , this method is now yielding great results , although the lyman-@xmath2 line typically targetted is a poor measure of the star formation rate . in the past decade ,
deep imaging in several different broad - bands ( typically spanning the optical and near - infrared ) has also been used to estimate the redshifts of galaxies on the basis of their observed colours essentially using the few photometric datapoints at different wavelengths as a crude low - dispersion spectrum . in this article
i will focus on a particular variant of these photometric redshifts " , which relies on a strong spectral break arising from absorption by neutral hydrogen in intervening clouds along the line of sight ( the lyman-@xmath2 forest ) .
this `` lyman break technique '' was originally pioneered at @xmath12 by steidel and co - workers ( steidel , pettini & hamilton 1995 ; steidel et al .
1996 ) who identified @xmath13-band drop - outs . at @xmath12
the technique involves the use of three filters : the @xmath13-band below the lyman limit ( @xmath14 ) ; the @xmath15- or @xmath16-band in the lyman forest region and a third filter , the @xmath17-band , long - ward of the lyman-@xmath2 line ( @xmath18 ) . at @xmath0
, we can efficiently use only two filters , above ( @xmath19-band 9000 ) and below ( @xmath1-band 8000 ) the continuum break at the lyman-@xmath2 line ( @xmath20 ) ; since the integrated optical depth of the lyman-@xmath2 forest at @xmath0 is @xmath21 ( see figure 1 ) , there is essentially no flux at shorter wavelengths , rendering the shortest - wavelength filter below the lyman limit redundant .
the key issue is to work at a sufficiently - high signal - to - noise ratio that lyman break `` drop - out '' galaxies can be safely identified through detection in a single redder band ( i.e. , @xmath19-band ) .
this approach has been demonstrated to be effective by the sdss collaboration in the detection of @xmath0 qsos using the @xmath1- and @xmath19-bands alone ( fan et al .
2002 ) , using wide - area but shallow ground - based imaging , which is sensitive to bright but rare quasars .
several deep imaging programmes with the advanced camera for surveys ( acs ) on _ hst _ use the sharp - sided sdss f775w ( @xmath1 ) and f850lp ( @xmath19 ) filters , which can be used to locate the``@xmath22-drop '' candidate @xmath0 galaxies ( fainter but more numerous than the qsos ) . figures 1&2 illustrates how a colour cut of @xmath23 can be effective in selecting sources with @xmath24 .
the acs images of the _ chandra _ deep field south from the _ hst _ treasury `` great observatory origins deep survey '' ( goods ; giavalisco & dickinson 2003 ) have been used to discover this @xmath1-drop population of @xmath0 galaxies ( stanway , bunker & mcmahon 2003 ; dickinson _ et al . _ 2004 ; giavalisco _ et al .
stanway , bunker & mcmahon ( 2003 ) searched 150arcmin@xmath25 for @xmath26 @xmath1-drop objects , of which six were probable @xmath27 galaxies brighter than @xmath28 . to address potential cosmic variance issues , a similar analysis in the goods - north field , which yielded a consistent estimate of the surface density of @xmath29 star forming sources ( stanway et al .
since these initial results , larger numbers of @xmath1-drops have been uncovered in additional _
hst_/acs pointings ( e.g. , bouwens et al . 2006 ) and from the ground ( e.g. , the subaru suprime - cam iwata et al .
2007 ) as well as deeper imaging of the goods fields , particularly the _ hubble ultra deep field _
( bunker et al . 2004 ) within goods - south .
[ fig : tracks ]
the effectiveness of the lyman - break technique has been demonstrated at lower redshifts through spectroscopic confirmation of hunderds of @xmath13-band dropouts at @xmath30 and @xmath16-band drop - outs at @xmath31 ( steidel et al .
however , it is important to demonstrate that the @xmath1-drop selection is similarly isolating galaxies at @xmath0 , if we are to use the surface density of @xmath1-drops to draw global inferences about galaxies within the first billion years .
a known possible contaminant is the extremely red object ( ero ) population of evolved galaxies at @xmath32 which can produce large @xmath33 colours : deep near - infrared imaging should identify eros , and this possible contamination is considered in stanway , mcmahon & bunker ( 2005 ) from the _ hst_/nicmos survey of the hudf .
low - mass galactic stars ( m / l / t dwarfs ) are another interloper population .
deep spectroscopy with keck / deimos of the brightest @xmath1-drop identified by stanway , bunker & mcmahon ( 2003 ) in goods - south revealed the brightest one , galaxy sbm03#3 ( @xmath34mag ) , to be at @xmath35 ( bunker et al . 2003 ; dickinson et al .
2004 ) and another @xmath1-drop ( sbm03#1 , @xmath36 ) was found to lie at a similar redshift of @xmath37 ( stanway et al .
2004ab ; dickinson et al . 2004 ) .
the @xmath1-drop spectra typically show a single emission line , with no significant continuum detected at moderate spectral resolution ( @xmath38 ) ; spectrally - resolved profiles of the emission lines are asymmetric ( as high-@xmath39 lyman-@xmath2 tends to be ) with a p - cygni - like profile and a sharp cut - off on the blue wing ( figure 2b ) .
the line fluxes of these brighter @xmath1-drops are @xmath40 .
the equivalent widths are @xmath41 using the @xmath19-band photometry from hst / acs , which is within the range seen in high-@xmath39 star - forming galaxies at @xmath42 .
the velocity widths of the lyman-@xmath2 are comparatively narrow ( @xmath43 ) and we do not detect the high - ionization line n v@xmath441240 , which strongly support the view that this line emission is powered by star formation rather than an agn .
subsequent spectroscopic programmes , including using fors2 on the eso vlt ( vanzella et al .
2006 ) and gmos `` nod&shuffle '' spectroscopy on gemini ( glare , stanway et al . 2004b , 2007 ) have confirmed many more @xmath0 galaxies in goods - south .
slitless spectroscopy of the hudf with _
hst_/acs ( grapes , malhotra et al .
2005 ) has also confirmed the spectral breaks of the @xmath1-drop galaxies , although the spectral resolution of @xmath45 is usually insufficient to see the lyman-@xmath2 emission lines in all but the most extreme equivalent widths .
although such spectroscopy has proven that the @xmath1-drop selection technique does select star - forming galaxies at @xmath0 , the efforts to date have been limited despite pushing the limits of 10-m class telescopes from the ground . only a fraction of the @xmath46 @xmath1-drops imaged with _ hst _ ( see the compilation of bouwens et al .
2006 ) have spectroscopic redshifts .
it seems likely that lyman-@xmath2 line emission does not emerge from some galaxies , through resonant scattering and absorption by dust ( a well - known result at lower redshifts , e.g. , steidel et al .
1999 ) , although the fraction of galaxies at @xmath4 with large lyman-@xmath2 equivalent widths seems greater than at @xmath12 ( stanway et al .
2007 ) , possibly implying low - metallicity , a top - heavy initial mass function , or extremely young ages for the starbursts . with its multi - object spectroscopic capability using micro - shutter technology , nirspec on _ jwst _ ( gardner et al .
2006 ) will have the capability to determine redshifts for most of the @xmath1-drop @xmath4 candidates identified , even to the depth of the hudf . beyond simply extending ground - based lyman-@xmath2 spectroscopy at @xmath4 to fainter galaxies
, nirspec will have the sensitivity to detect the rest - frame uv and optical continuum .
studies of interstellar medium ( ism ) and stellar photospheric lines of @xmath13-drop galaxies at @xmath12 in the rest - uv have provided clues to the initial mass function , metallicity and outflow velocities in these galaxies ( e.g. , shapley et al . 2003 ) . only in the case of
the brightest @xmath1-drop have such ism metal absorption lines been seen ( dow - hygelund et al .
2005 ) , but nirspec will enable the whole population to be studied with its intermediate - resolution and high - resolution modes ( @xmath47 and @xmath48 ) .
perhaps most importantly , nirspec will not be restricted to the rest - uv at @xmath4 , which is the case with ground - based optical / near - ir spectroscopy .
nirspec has a coverage from @xmath49 m , unhampered by the oh sky lines and large thermal background seen from ground - based observatories .
hence , for the @xmath4 population , we can target the full range of emission lines from lyman-@xmath21216 to h@xmath26563 .
emission line diagnostics can then be used to estimate the metallicities and the extinction due to dust .
[ fig : madau_plot ]
estimates of the star formation rate density at the highest redshifts have relied on the intensity of the rest - frame uv continuum emission , redshifted into the @xmath19-band at @xmath0 ; this is dominated by the light from the most massive stars ( @xmath50 which are the hottest and bluest stars ) , which are the shortest lived , and hence a tracer of the instantaneous star formation rate . to go from the number of massive stars formed to the total star formation rate requires an assumption about the stellar initial mass function ( imf ) . in the absence of dust obscuration ,
the relation between the flux density in the rest - uv around @xmath51 and the star formation rate ( @xmath52 in @xmath53 ) is given by @xmath54 from madau , pozzetti & dickinson ( 1998 ) for a salpeter ( 1955 ) imf with @xmath55 .
the relatively bright magnitude cut of @xmath56 for the original @xmath1-drop selection from the goodsv0.5 data ( stanway , bunker & mcmahon 2003 ; stanway et al .
2004a ) corresponds to an unobscured star formation rate of @xmath57 at @xmath58 ( the luminosity - weighted average redshift ) , equivalent to @xmath59 for the @xmath13-band dropout population at @xmath30 .
a tantalizing result from our work so far is that at @xmath4 there are far fewer uv - luminous star forming galaxies than would have been predicted if there was no evolution , based on a comparison to the well - studied @xmath42 lyman break population ( steidel et al.1999 ) .
in fact , the volume averaged ( comoving ) star formation rate in galaxies with @xmath60 is @xmath61 _ less _ at @xmath0 than at @xmath30 .
so it would appear that at this crucial epoch , where abundant sources of uv photons are needed to reionize the universe , there is in fact a deficit of luminous star forming galaxies .
the bright end of luminosity function has evolved greatly from @xmath4 to @xmath12 .
other groups have claimed less dramatic evolution or even no evolution in the total volume - averaged star formation rate , based on the same fields ( giavalisco et al .
2004 ; dickinson et al .
2004 ) and similar _
hst_/acs data sets ( bouwens et al . 2003 ; yan , windhorst & cohen 2003 ) , but these groups work closer to the detection limit of the images and introduce large completeness corrections for the faint source counts . to do a complete inventory of the uv light from star formation we must address the contribution of low - luminosity star - forming galaxies to the ionizing flux .
the public availability of the _ hubble ultra deep field _
( hudf ; beckwith , somerville & stiavelli ) can address this puzzle by pushing down the luminosity function at @xmath0 to well below the equivalent of @xmath62 for the @xmath30 population .
the _ hubble ultra deep field _ ( hudf ) was a cycle 12 stsci director s discretionary time programme executed over september 2003 january 2004 , comprising 400 orbits in 4 broad - band filters ( including f775w @xmath1-band for 144 orbits ; f850lp @xmath19-band for 144 orbits ) . as the hudf represents the deepest set of images yet taken , significantly deeper than the @xmath63-band exposures of the _ hubble deep fields _ ( williams et al .
1996 ) , and adds the longer - wavelength @xmath19-band , it is uniquely suited to the goals of our program .
we performed the first analysis of @xmath1-drops in the udf ( bunker et al .
2004 ) , presenting details of 54 candidate star forming galaxies at @xmath0 in a preprint the day after the release of the udf data .
our analysis was subsequently independently repeated by yan & windhorst ( 2004 ) , with agreement at the 98% level ( see bunker & stanway 2004 for a comparison ) .
we take our magnitude limit as @xmath64 ( a @xmath65 cut).we measure a star formation density of @xmath66 at @xmath0 from galaxies in the udf with sfrs@xmath67 ( equivalent to @xmath68 at @xmath30 ) .
this confirmed the previous discovery of a relatively low co - moving star formation rate density at @xmath0 compared to the more recent past at @xmath12 ( figure 3 ) .
at the relatively bright cut of @xmath56 used in stanway , bunker & mcmahon ( 2003 ) from the goodsv0.5 survey , the udf data is 98% complete for sources as extended as @xmath69arcsec .
interestingly , we detect no extended ( low surface brightness ) @xmath1-drops to this magnitude limit in addition to the targetted @xmath1-drop sbm03#1 in the deeper udf data .
this supports our assertion that the @xmath1-drop population is predominantly compact and there can not be a large completeness correction arising from extended objects ( c.f .
lanzetta et al . 2002 ) .
the acs imaging is of course picking out hii star forming regions , and these uv - bright knots of star formation are typically @xmath70kpc ( @xmath71arcsec at @xmath0 ) even within large galaxies at low redshift .
it is interesting that the level of stellar contamination in the udf @xmath1-drops is only 2% , compared with about one in three at the bright end ( @xmath56 ; stanway , bunker & mcmahon 2003 ; stanway et al.2004 )
. this may be because we are seeing through the galactic disk at these faint limiting magnitudes to a regime where there are no stars ( see also pirzkal et al .
we compare our @xmath1-drop luminosity function with the work of madau , haardt & rees ( 1999 ) for the density of star formation required for reionization .
we have updated their equation 27 for the more recent concordance cosmology estimate of the baryon density of spergel et al .
( 2007 ) , and for the predicted mean redshift of our sample ( @xmath72 ) : @xmath73 this relation is based on the same salpeter initial mass function as we have used in deriving our volume - averaged star formation rate .
@xmath74 is the concentration factor of neutral hydrogen , @xmath75 .
simulations suggest @xmath76 ( gnedin & ostriker 1997 ) .
even if we take @xmath77 ( no absorption by h i ) this estimate of the star formation density required is a factor of @xmath78 higher than our measured star formation density of @xmath66 at @xmath0 from galaxies in the udf with sfrs@xmath67 . for faint end slopes of @xmath79 galaxies with @xmath80 account for @xmath81% of the total luminosity , so would fall short of the required density of lyman continuum photons required to reionize the universe .
if the faint - end slope is as steep as @xmath82 then there would just be enough uv lyman continuum photons generated in star forming galaxies at @xmath0 ( assuming a salpeter imf ) , but the required escape fraction for complete reionization would still have to be implausibly high ( @xmath83 , whereas all high-@xmath39 measurements to date indicate @xmath84 : fernnadez - soto , lanzetta & chen 2003 ; steidel , adelberger & pettini 2001 ) .
if star forming galaxies at redshifts close to @xmath85 were responsible for the bulk of reionization , then a very different initial mass function would be required , or the calculations of the clumping factor of neutral gas would have to be significantly over - estimated ( see also stiavelli , fall & panagia 2004 ) . alternatively another low - luminosity population ( e.g. , forming globular clusters ; ricotti 2002 ) could be invoked to provide some of the shortfall in ionizing photons .
it is also plausible that the bulk of reionization occurred at redshifts well beyond @xmath85 : the wmap polarization data indicate @xmath86 ( spergel et al .
2007 ) , and it is possible that the gunn - peterson troughs seen in @xmath87 qsos ( becker et al . 2001 ; fan et al .
2002 ) mark the very last period of a neutral igm .
if some of these @xmath1-drop galaxies can be shown to harbour stellar populations with ages of a few hundred myr , then this pushes their formation epoch to @xmath6 .
measurements of the stellar masses of individual @xmath4 galaxies can also constrain structure formation paradigms ; in a simple hierarchical model , massive galaxies assemble at later times through merging , so it might be expected that in this scenario the number density of massive evolved galaxies in the first gyr would be low . obtaining
such observational evidence of old stars at high redshift is challenging : the initial work on @xmath4 galaxies was based on optical and near - infrared imaging and explored the rest - frame ultraviolet ( uv ) in these galaxies , which is completely dominated by recent or ongoing star formation .
the addition of imaging at wavelengths @xmath88 m with the irac camera on the _ spitzer space telescope _ extends the coverage to the rest - frame optical at @xmath4 , beyond the age - sensitive balmer/4000 break and where the light is dominated by stars that form the bulk of the stellar mass . with the addition of this _
spitzer_/irac imaging , spectral energy distributions from the multi - wavelength broad - band photometry can be fit to stellar population synthesis models to constrain the stellar masses and ages , and hence the preceding star formation history and the formation epochs can be explored . in eyles
( 2005 ) we identified two @xmath1-drops with spectroscopic redshifts of @xmath89 which exhibited significant balmer/4000 spectral breaks ( figure 4 ) , implying stellar ages of @xmath90myr , and formation redshifts of @xmath91 .
large stellar masses in the range @xmath92 were inferred .
we then studied the whole lyman - break population in the goods - south field with irac/_spitzer _ at @xmath0 ( the @xmath1-drops , eyles et al .
2007 ) and @xmath93 ( the @xmath94-drops , stark et al . 2007 ) .
half of these are confused with foreground sources at _ spitzer _ resolution , but from those with clean photometry we find that a surprisingly large fraction ( 40% ) have evidence for substantial balmer/4000 spectral breaks .
this indicates the presence of old underlying stellar populations that dominate the stellar masses .
analysis of those @xmath1-drops that are undetected by irac at 3.6@xmath95 m indicates that a subset of the population are younger , considerably less massive systems .
we calculate that emission line contamination should not severely affect our photometry or derived results . using sed fits out to @xmath96 m
, we find little evidence for substantial intrinsic dust reddening in our sources .
using the constrained properties of our @xmath1-drop sample , we were able to calculate a value for the @xmath4 stellar mass density of @xmath97 ( figure 5 ) , correcting for those objects eliminated from our analysis due to their untreatable irac - confusion and those lacking goods - music photometric redshifts .
using a somewhat uncertain correction in order to account for the stellar mass in objects below our @xmath19-band magnitude selection limit , this value could perhaps be @xmath98 .
any post - starburst and dust - obscured @xmath4 sources would not be found using the @xmath1-drop selection technique , and hence our @xmath4 stellar mass density value is necessarily a lower limit , and is consistent with the estimates of yan et al .
( 2006 ) . exploring the previous star formation histories of our @xmath1-drops ,
as inferred from their sed fitting , we suggest that the global star formation of these sources may have been substantially higher prior to the epoch of observation ( figure 5 ) , and the resultant uv flux at @xmath99 may have played an important role in reionizing the universe . | the quest to discover the most distant galaxies has developed rapidly in the last decade .
we are now exploring redshifts of 6 and beyond , when the universe was less than a billion years old , an epoch when the previously - neutral intergalactic medium was reionized .
the continuing discovery of galaxies at progressively higher and higher redshifts has been driven by the availability of large telescopes on the ground and in space , improvements in detector technology , and new search strategies . over the past 4 years
, the lyman break technique has been shown to be effective in isolating @xmath0 star - forming @xmath1-drop galaxies through spectroscopic confirmation with large ground - based telescopes ( keck , gemini and the eso vlts ) .
narrow - band imaging , notably with the wide field of the subaru telescope , has also produced samples of lyman-@xmath2 emitters at these redshifts .
analysis of the _ hubble ultra deep field _ ( hudf the deepest images obtained so far , and likely to remain so until the _ james webb space telescope , jwst _ ) , has enabled us to explore the faint end of the luminosity function , which may contribute the bulk of the total star formation .
the discovery of this @xmath1-drop galaxy population has been used to infer the global star formation rate density at this epoch ( @xmath0 ) , and we are now begining to constrain the contribution to reionization of the uv flux from these galaxies . infrared data from the _ spitzer space telescope _ has been used to determine the spectral energy distributions ( seds ) from the rest - frame uv to the optical of some @xmath1-drops , and constrain the previous star formation histories , masses and age of these sources .
the indications are that much of the stellar mass of these galaxies might have formed in vigorous bursts at @xmath3 .
the next big advances would be to test the population synthesis modelling of these @xmath4 galaxies through spectroscopy of the rest - frame optical ( rather than crude broad - band seds ) , and also to push the observational horizon for galaxies further to directly explore star formation during the reionization epoch .
_ jwst _ is likely to have a profound impact on realising these goals . |
in this paper we investigate the problem of optimal attitude control of a 3-dimensional body which can be rotated around two fixed axes .
the problem goes back to euler @xcite who proved in 1776 that an arbitrary rotation @xmath0 of a 3-dimensional body may be factored as @xmath1 where @xmath2 ( resp .
@xmath3 ) is a rotation in angle @xmath4 around @xmath5-axis ( resp .
@xmath6-axis ) .
the parameters @xmath7 are called the euler s angles .
we could allow decompositions for @xmath0 with more factors : @xmath8 clearly we will get infinitely many such decompositions for a fixed element @xmath0 in the group @xmath9 of rotations . thus it is natural to pose the question of finding a decomposition ( [ eutwo ] ) that minimizes the total angle of rotation @xmath10 .
it can happen that decompositions with more factors have a smaller total angle of rotation than the euler s decomposition ( [ eu ] ) .
it turns out that this problem is not well - posed : for some @xmath0 an optimal decomposition ( [ eutwo ] ) does not exist . instead
, the infimum of the total angle of rotation is attained as a limit on a sequence of decompositions ( [ eutwo ] ) with @xmath11 .
we can overcome this difficulty by noting that @xmath12 hence it is natural to extend the set of controls from @xmath13 to @xmath14 @xmath15 . implementing a rotation @xmath16 corresponds to carrying out rotations around axes @xmath5 and @xmath6 simultaneously with the ratio @xmath17 of angular velocities .
once we extend the control set , our optimization problem becomes well - posed and for every @xmath18 there is an optimal decomposition with a finite number of factors .
we study this problem in a more general setting , where we allow an arbitrary angle @xmath19 between the axes @xmath5 and @xmath6 .
we also introduce a more general cost function to be minimized , where a rotation in angle @xmath4 around @xmath6-axis has the same cost as a rotation around @xmath5-axis in angle @xmath20 , @xmath21 .
we solve the optimization problem in this greater generality and determine possible patterns for the optimal decompositions .
each of these patterns has ( at most ) 3 independent time parameters , and it is fairly easy to find numerically the decompositions of a given element @xmath18 according to each pattern .
this produces a finite number of decompositions and we can immediately see which one of them is optimal .
it happens that our optimization problem has a bifurcation at @xmath22 .
for the cases @xmath23 and @xmath24 we get different lists of optimal patterns .
there are also special cases when @xmath25 or @xmath26 .
let us present the list of optimal patterns in case when the axes @xmath5 and @xmath6 are perpendicular to each other and @xmath27 .
since in this case the problem is symmetric with respect to the dihedral group of order @xmath28 , generated by transformations @xmath29 , @xmath30 , @xmath31 , the list of patterns will also be symmetric with respect to this group .
we denote this group of @xmath28 symmetries by @xmath32 and use it to present the list of patterns in a more compact form .
let the angle between the axes @xmath5 and @xmath6 be @xmath33 and let @xmath27 . for an element @xmath18
there is an optimal decomposition with @xmath34 of one of the following types : @xmath35 and symmetric to these under the group of transformations @xmath32 .
suppose we would like to decompose a rotation @xmath36 as a product of rotations around @xmath5- and @xmath6-axes , where @xmath37 is the standard orthogonal basis of @xmath38 .
the pattern for the optimal decompositions will depend on the value of @xmath4 . if @xmath39 then the following decomposition realizes the minimum of the total rotation angle : @xmath40 for @xmath41 the euler s decomposition ( [ eu ] ) becomes optimal : @xmath42 when @xmath43 both patterns are optimal .
for @xmath44 the optimal decompositions may be obtained by switching @xmath5 with @xmath6 in the above expressions . in case when the axes @xmath5 and @xmath6 are perpendicular to each other and @xmath25 ( meaning that rotations around @xmath6-axis have zero cost ) ,
the optimal decompositions are precisely those described by euler ( [ eu ] ) . in 2009
nasa launched a space telescope kepler with a mission of finding planets outside the solar system .
this spacecraft was placed in an orbit around the sun . to take images of stars ,
the telescope needs to be pointed in the target direction , with its solar panels facing the sun .
the attitude control of kepler is done with reaction wheels , which are heavy disks mounted on electric motors .
once the reaction wheel is turned , the spacecraft will turn around the same axis in the opposite direction due to the angular momentum conservation law .
if we have three reaction wheels with linearly independent axes , by rotating them simultaneously with appropriate relative angular velocities , we can implement a continuous rotation of the spacecraft around an arbitrary axis . for redundancy ,
kepler was equipped with four reaction wheels with their axes in a tetrahedral configuration , so that any three of them could provide an efficient attitude control .
however by may 2013 , two of the four reaction wheels failed , leaving kepler with just two available axes of rotation @xcite .
the results of our paper provide optimal methods for attitude control with two rotation axes , like in situation with the kepler space telescope .
this paper builds on our previous work @xcite , where we studied a similar problem for @xmath45 , also with two available controls , but with a restriction that only a positive time evolution is allowed . that paper was motivated by the applications to quantum control in a 1-qubit system .
in the present paper we use the geometric control theory @xcite , which is an adaptation of the pontryagin s maximum principle to the setting of lie groups .
the maximum principle provides only necessary conditions for optimality , which need not be sufficient . in section 3
we identify decompositions that satisfy the necessary conditions of the pontryagin s maximum principle .
then we go into a more detailed analysis in section 4 by showing that decompositions with a large number of factors are not optimal , even when they satisfy the conditions of the maximum principle .
our main results are stated in theorems [ czero ] [ meq ] at the end of the next section .
i thank cornelius dennehy , ken lebsock , eric stoneking and alex teutsch for the stimulating discussions .
support from the natural sciences and engineering research council of canada is gratefully acknowledged .
for a unit vector @xmath46 denote by @xmath47 an operator of rotation of @xmath38 in angle @xmath4 around @xmath5 , with the plane perpendicular to @xmath5 turning counterclockwise when viewed from the endpoint of @xmath5 . as a @xmath48 matrix , @xmath47
is given by the formula : @xmath49 where for @xmath50 with @xmath51 , @xmath52 is the adjoint matrix of @xmath5 with respect to the cross product , so that @xmath53 : @xmath54 for @xmath46 with @xmath55 we set @xmath56 , where @xmath57 and @xmath58 .
the set of all rotations of @xmath38 forms the group @xmath9 . for a fixed @xmath5
the set @xmath59 is a 1-parametric subgroup in @xmath9 .
it is well - known that for any two non - proportional unit vectors @xmath60 , the corresponding @xmath61-parametric subgroups together generate the whole group @xmath9 .
this means that every element @xmath18 may be decomposed into a product @xmath62 with @xmath63 .
decomposition ( [ dec ] ) is of course not unique .
it is then natural to consider the optimization problem of finding the infimum of @xmath64 over all decompositions ( [ dec ] ) with fixed @xmath18 .
more generally , we may assign cost to each generator @xmath60 and minimize the total cost in ( [ dec ] ) .
introduction of the cost parameters may be warranted in case when the body that we control has unequal momenta of inertia with respect to the axes @xmath5 and @xmath6 , thus making it easier to rotate it around one of the axes . without loss of generality ,
we assume that @xmath65 and renormalize the cost function by fixing @xmath66 , @xmath67 with @xmath21 . for the rest of the paper we fix two non - proportional vectors @xmath68 with @xmath69 .
an important parameter is the angle @xmath70 between these vectors . without loss of generality
we assume @xmath71 , otherwise we can replace @xmath6 with @xmath72 . throughout the paper we will the use parameter @xmath73 , @xmath74 .
let @xmath75 be a vector perpendicular to @xmath5 and @xmath6 , @xmath76 , @xmath77 .
it could happen that the infimum of cost is not attained on any particular decomposition ( [ dec ] ) , but rather as a limit on a sequence of such decompositions with @xmath78 .
it turns out that we can overcome this difficulty by enlarging the set of generators to be @xmath79 note that rotations corresponding to elements of @xmath80 can be realized as limits of products of rotations with axes @xmath81 : @xmath82 from the point of view of the attitude control , this corresponds to turning on controls @xmath5 and @xmath6 simultaneously with intensities @xmath83 and @xmath84 respectively .
we extend the definition of the cost function in such a way that the cost of both sides of ( [ explim ] ) is the same : @xmath85 our goal is to solve the following * problem 1 .
* for a given @xmath18 find a decomposition @xmath86 with @xmath87 , @xmath88 , realizing the the infimum of @xmath89 it was shown in @xcite , theorem 1.4 , that the infimum cost in this problem problem is the same as for its more restricted version where the set of controls is taken to be @xmath90 instead of @xmath80 .
in fact , we shall see that we would not need the whole set @xmath80 , but require in addition to controls @xmath90 only the elements @xmath91 , where @xmath92 ( resp .
@xmath93 ) is a linear combination of @xmath5 and @xmath6 , which is orthogonal to @xmath94 ( resp .
@xmath95 ) .
since the cost of @xmath96 and @xmath97 is the same , we can rescale the generators without changing the cost of decompositions .
we can thus drop the requirement @xmath98 for the generators @xmath99 .
we fix @xmath100 and @xmath101 .
taking into account that @xmath102 and @xmath103 , it is easy to check that @xmath104 and @xmath105
. now we can state the main results of the paper .
it turns out that the problem we consider has a bifurcation at @xmath106 , and we need to consider the cases @xmath107 and @xmath108 separately .
there will be also a special case when @xmath25
. we will give the solution of the above optimal control problem by specifying the patterns of optimal decomposition ( [ dec ] ) .
we begin with some elementary observations .
obviously we may restrict all angles of rotation to be less or equal to @xmath109 .
if @xmath86 is an optimal decomposition then a decomposition @xmath110 with @xmath111 , @xmath112 , @xmath113 , is also optimal .
we call ( [ sub ] ) a _ subword _ of @xmath114
. we shall present the optimal decompositions as subwords of certain patterns .
since the number of patterns can be fairly large , we shall use various symmetries in order to group several patterns together .
for example , if we have an optimal decomposition @xmath115 with @xmath116 , then @xmath117 is also an optimal decomposition ( for a different element of @xmath9 ) .
this follows from the fact that multiplication of controls by @xmath118 is an automorphism of our problem .
we denote this symmetry transformation on the set of patterns by @xmath119 . whereas the set of optimal patterns is always invariant with respect to the symmetry @xmath119 , other types of symmetries that we shall consider are not universal and are present only for some patterns .
if we make the following schematic representation of the controls , all symmetries that we consider will be elements of the dihedral group of symmetries of a square : consider a transformation @xmath120 , @xmath121 , @xmath122 , @xmath123 .
together with the symmetry @xmath119 this generates a set of @xmath124 transformations .
we denote this set of symmetries by @xmath125 .
we assume that all symmetries we consider are compatible with multiplication by @xmath118 , even though they are not linear in general .
we also consider a transformation @xmath126 , @xmath127 , @xmath128 , @xmath129 .
together with @xmath119 , this generates a set of @xmath124 transformations , which we denote by @xmath130 . finally ,
if we consider all of the above transformations together , we generate a full set of @xmath28 symmetries of the square in fig.1 , which we denote by @xmath131 .
[ czero ] let @xmath132 , @xmath133 .
for an element @xmath18 the infimum of the optimization problem 1 is attained on a subword of one of the following patterns : \(i ) @xmath134 where @xmath135 , @xmath136 , and symmetric to it under @xmath131 .
\(ii ) @xmath137 , with @xmath138 , and symmetric to it under @xmath131 .
\(iii ) @xmath139 , with @xmath138 , and symmetric to it under @xmath131 .
when we apply symmetry transformations , e.g. @xmath126 , @xmath127 , we change the parameters @xmath140 , @xmath141 accordingly , but the relation @xmath135 in ( i ) is preserved . under this symmetry transformation , the pattern ( i ) takes form @xmath142 with @xmath135
. set @xmath143 [ mgt ] let @xmath144 .
for an element @xmath18 the infimum of the optimization problem 1 is attained on a subword of either pattern ( i ) or one of the following : \(iv ) @xmath145 , with @xmath138 , and symmetric to it under @xmath130 .
\(v ) @xmath146 , with @xmath138 , and symmetric to it under @xmath130 .
\(vi ) @xmath147 , with @xmath138 , and symmetric to it under @xmath130 .
\(vii ) @xmath148 , with @xmath138 , and symmetric to it under @xmath130 .
[ mlt ] let @xmath108 . for an element
@xmath18 the infimum of the optimization problem 1 is attained on a subword of either patterns ( i ) , ( iv ) , ( v ) given above , or the following pattern \(viii ) @xmath149 , with @xmath150 , and symmetric to it under @xmath151 .
[ meq ] let @xmath25 , @xmath152 .
for an element @xmath18 the infimum of the optimization problem 1 is attained on a subword of one of the following two patterns \(ix ) @xmath153 , with @xmath138 , and symmetric to it under @xmath154 .
\(x ) @xmath155 , with @xmath138 , and symmetric to it under @xmath154 . when @xmath156 we have @xmath93 proportional to @xmath5 , and the lists of patterns in theorems [ mgt ] and [ mlt ] become equivalent .
in this section we will review the geometric optimization theory following @xcite , and apply it to our optimization problem .
the lie algebra @xmath157 of the lie group @xmath9 is the tangent space to @xmath9 at identity and consists of skew - symmetric @xmath48 matrices .
the lie bracket of two matrices in @xmath157 is @xmath158 = ab - ba$ ] .
we may identify the space @xmath157 with @xmath38 via the map ( [ cross ] ) @xmath159 .
the corresponding lie bracket of two vectors in @xmath38 is the cross product .
fix @xmath18 .
a curve leading to @xmath0 is an absolutely continuous function @xmath160 \rightarrow so(3)$ ] such that @xmath161 and @xmath162 .
an absolutely continuous function has a measurable derivative @xmath163 \rightarrow so(3)$ ] such that @xmath164 for almost all @xmath4 .
the derivative @xmath165 is lebesgue integrable @xcite .
let us formulate a differential version of our optimization problem .
* problem 2 . * for an element @xmath18 find the infimum of @xmath166 over all absolutely continuous curves @xmath160 \rightarrow so(3)$ ] leading to @xmath0 , satisfying @xmath167 for almost all @xmath4 .
the parameter @xmath168 in problem 2 is not fixed and when taking the infimum we consider the curves with all @xmath169 .
it is clear that the restriction to the case of piecewise constant controls @xmath170 gives precisely problem 1 .
on the other hand we shall see that the solutions of problem 2 indeed have piecewise constant controls , which implies equivalence of problems 1 and 2 .
[ existence ] for any @xmath18 there exists an absolutely continuous optimal solution @xmath171 for problem 2 .
the proof of this proposition is based on the observation that the cost assigned to a curve @xmath172 \rightarrow so(3)$ ] is independent of the choice of its parametrization . to prove this
, we first note the cost function ( [ costf ] ) satisfies @xmath173 for @xmath174 .
consider an absolutely continuous increasing surjective reparametrization @xmath175 \rightarrow [ 0 , t_0]$ ] and the corresponding reparametrized curve @xmath176
. then @xmath177 and @xmath171 have the same cost : @xmath178 this computation shows that rescaling of the set of controls @xmath80 does not change the cost of a curve @xmath171 leading to @xmath0 with @xmath179 .
let us modify problem 2 by replacing the set @xmath80 with its convex hull @xmath180 once the control set is convex , we can apply theorem 4.10 from @xcite to obtain the existence of an absolutely continuous optimal solution @xmath172 \rightarrow so(3)$ ] for the modified problem . to go back to the setting of problem 2
, we note that every absolutely continuous curve admits a parametrization by the arc length , i.e. , the natural parametrization ( see for example section 5.3 in @xcite ) .
then it is easy to see that the curve @xmath171 may also be reparametrized with @xmath179 .
since reparametrization does not change the cost , we see that an optimal solution of the modified problem with the set of controls @xmath181 yields an optimal solution for problem 2 .
our optimization problem induces a left - invariant metric on @xmath9 .
it is possible to see that this metric does not correspond to any riemannian structure on this lie group .
the hamiltonian function for problem 2 is @xmath182 which involves a parameter @xmath183 ( see section 11.2.2 in @xcite for details ) . for each @xmath184
we define the maximal hamiltonian @xmath185 [ pmp ] ( pontryagin s maximum principle , @xcite ) let @xmath171 be an optimal curve leading to @xmath18 for problem 2 .
then there exists an absolutely continuous function @xmath186 \rightarrow { { \mathbb r}}^3 = so(3)$ ] and a constant @xmath183 such that for almost all @xmath187 $ ] the following equations hold : @xmath188 and @xmath189 if @xmath190 then @xmath191 is non - zero for almost all @xmath187 $ ] .
[ cons ] the quantity @xmath192 is conserved .
@xmath193 note that for our problem the parameter @xmath194 can not be zero , otherwise condition ( i ) implies that @xmath195 , hence @xmath191 is proportional to @xmath75 for almost all @xmath4 , and so is @xmath196 .
however ( ii ) implies that @xmath197 and thus @xmath198 and @xmath191 is a constant multiple of @xmath75 . inspecting ( ii ) again , we conclude that @xmath191 must be zero for almost all @xmath4 , which contradicts the last claim of the theorem . in case
when the parameter @xmath194 is non - zero , it can be rescaled to any negative value .
a convenient choice for us is @xmath199 .
consider a second basis @xmath200 of the @xmath201-plane , where @xmath202 then @xmath203 , @xmath204 .
in this basis @xmath205 and @xmath206 . according to theorem [ pmp ] , the value of @xmath191 determines the value of @xmath207 via ( i ) , while by ( ii ) the value of @xmath207 determines the evolution of @xmath191 .
let us analyze ( i ) to see which values of @xmath191 are admissible , and what are the corresponding controls @xmath165 .
let us write @xmath208 and @xmath209 .
then @xmath210 since the set @xmath80 is closed under symmetry @xmath211 , @xmath212 , we see that the maximum in @xmath213 of @xmath214 is attained when @xmath83 has the same sign as @xmath215 and @xmath84 has the same sign as @xmath216 .
hence @xmath217 by property ( i ) of the theorem , @xmath218 , thus the admissible values of @xmath191 satisfy either @xmath219 , @xmath220 or @xmath221 , @xmath222 .
we summarize this in the following lemma , which describes controls in the resulting regions : [ reg ] ( a ) let @xmath223 .
\(i ) if @xmath224 , @xmath225 then @xmath226 , @xmath227 , the control is @xmath228 ; \(ii ) if @xmath229 , @xmath225 then @xmath230 , @xmath227 , the control is @xmath231 ; \(iii ) if @xmath232 , @xmath233 then @xmath234 , @xmath235 , the control is @xmath236 ; \(iv ) if @xmath237 , @xmath233 then @xmath234 , @xmath238 , the control is @xmath239 .
\(b ) if @xmath25 then @xmath240 , @xmath222 . when @xmath241 and @xmath233 we could have either control @xmath236 or @xmath239 . at the points where two regions meet ,
the whole segment joining the corresponding two controls is allowed .
for example , when @xmath224 and @xmath242 we could have any control @xmath208 with @xmath243 , @xmath244 .
we will call such values of @xmath245 _
critical_. if the curve @xmath191 reaches a critical point , one of three things could happen : the curve @xmath191 could cross the boundary of a region , in which case the control will switch ; the curve @xmath191 could return to the same region where it came from without a switch of control ; or the curve @xmath191 may stay inside the critical boundary for some positive time .
let us describe evolution of @xmath191 inside the critical boundary .
[ critreg ] ( a ) suppose @xmath246 for @xmath247 $ ]
. then @xmath248 and @xmath249 for @xmath247 $ ] .
\(b ) suppose @xmath250 for @xmath247 $ ] and @xmath223 .
then @xmath251 , @xmath252 and @xmath249 for @xmath247 $ ] .
cases @xmath253 , @xmath254 and @xmath253 , @xmath255 are analogous , the controls are @xmath256 and @xmath257 respectively and @xmath249 . to prove ( a ) consider equation ( ii ) in theorem [ pmp ] .
we get @xmath258 taking into account that @xmath259 we get that @xmath260 since @xmath261 and @xmath262 are constant , this implies @xmath249 for @xmath247 $ ]
. then we get @xmath263 and @xmath165 is proportional to @xmath92 .
case ( b ) is analogous , except that for @xmath264 the segment joining @xmath5 and @xmath6 does not contain a vector proportional to @xmath93 . 3 in .,title="fig : " ] 3 in .,title="fig : " ] + 3 in .,title="fig : " ] 3 in .,title="fig : " ] an optimal solution of problem 2 could only involve controls @xmath265 , @xmath266 , @xmath267 and @xmath268 .
moreover , controls @xmath268 do not occur if @xmath269 .
note that when @xmath106 we get @xmath93 proportional to @xmath5 .
when @xmath25 we get @xmath270 .
next , let us study evolution of @xmath191 under controls @xmath265 and @xmath266 .
as we have seen in lemma [ reg ] , control @xmath5 corresponds to the region @xmath224 , @xmath271 . let @xmath272 . by part ( ii ) of theorem [ pmp ] , evolution of @xmath191
is given by @xmath273 from this we get @xmath274 setting @xmath275 , we get the equations of the harmonic oscillator @xmath276 with solutions @xmath277 , @xmath278 .
we plot the trajectories in @xmath279-plane in fig.[ex ] and [ fx ] .
similarly , we plot the trajectories for the other regions described in lemma [ reg ] .
this gives us the trajectories that satisfy the conditions of theorem [ pmp ] .
for example , the path corresponds to the decomposition @xmath280 when a trajectory reaches a critical point , for example , it could continue from either using evolution with controls @xmath5 , @xmath6 or remain at this critical point for some positive time using control @xmath93 .
the conservation law of lemma [ cons ] ensures that for the trajectory the points and have equal @xmath75-coordinates .
the same property holds in other similar cases , and in particular the trajectory that starts at a critical point and goes to will reach the critical point .
it follows that for the trajectory evolution times for the parts and are the same , since the corresponding arcs are symmetric to each other .
next we establish the relations between the time parameters in these trajectories ( cf .
proposition 2.1 in @xcite ) .
[ times ] ( a ) let @xmath140 be the @xmath5-evolution time , and @xmath141 be @xmath72-evolution time for the trajectory to 85pt -18pt @xmath61 @xmath281 -8pt -18pt @xmath28 @xmath281 -8pt -22pt @xmath282 .
then @xmath283 the same relation holds for the trajectories to 88pt -21pt @xmath282 -1pt @xmath281 -8pt -21pt @xmath284 -1pt @xmath281 -8pt -18pt @xmath61 , to 88pt -18pt @xmath285 @xmath281 -8pt -21pt @xmath286 -1pt @xmath281 -8pt -21pt @xmath287 , to 89pt -21pt @xmath286 -1pt @xmath281 -8pt -21pt @xmath287 -1pt @xmath281 -8pt -21pt @xmath288 , etc .
, with @xmath140 being the time parameter for @xmath265-evolution and @xmath141 for @xmath266-evolution .
\(b ) let @xmath289 be the time of evolution for the trajectories involving critical points , to 60pt -18pt @xmath290 @xmath281 -8pt -18pt @xmath291 , to 60pt -18pt @xmath292 @xmath281 -8pt -18pt @xmath290 , to 60pt -21pt @xmath293 @xmath281 -8pt -21pt @xmath294 or to 60pt -21pt @xmath294 -1pt @xmath281 -8pt -21pt @xmath295 .
let @xmath296 be the time of evolution for the trajectories to 60 pt -21pt @xmath294 -1pt @xmath281 -8pt -18pt @xmath292 , to 60pt -18pt @xmath291 @xmath281 -8pt -21pt @xmath294 , to 60pt -21pt @xmath295 -1pt @xmath281 -8pt -18pt @xmath290 or to 60pt -18pt @xmath290 @xmath281 -8pt -21pt @xmath293 .
then @xmath297 consider the trajectory .
let @xmath298 and @xmath299 be @xmath75-coordinates of the points , and respectively . then @xmath298 is also the @xmath75-coordinate of the point . 3 in .,title="fig : " ] 3 in .,title="fig : " ] + 3 in .,title="fig : " ] 3 in .,title="fig : " ] since the points -19pt @xmath61 and -19pt @xmath28 lie on a circle with the center at @xmath300 , @xmath301 , they satisfy the equation @xmath302 let @xmath84 be the base of the isosceles triangle with vertices at -19pt @xmath61 , -19pt @xmath28 and the center of the circle , and let @xmath303 be the altitude in this triangle
. then @xmath304 and @xmath305 since @xmath140 is the angle at the vertex of this triangle , we have @xmath306 similarly , @xmath307 since @xmath308 , in order to establish claim ( a ) , we need to show that @xmath309 .
this equality however follows from ( [ cir ] ) : @xmath310 the proof for the other cases in ( a ) is analogous . 3 in .,title="fig : " ] 3 in .,title="fig : " ] let us prove part ( b ) .
consider the trajectory .
the time parameter @xmath289 is the angle corresponding to this arc of the circle with center at @xmath300 , @xmath301 and radius @xmath311 . taking the projection to @xmath312-axis we get @xmath313
the derivation of the formula for @xmath296 is analogous .
the patterns listed in theorem [ mgt ] can be traced on the diagrams in fig.[kgtc ] , while the patterns of theorem [ mlt ] can be seen on fig .
for example , the pattern given in part ( i ) of theorem [ mgt ] corresponds to the trajectory .
the above proposition describes the relations between the time parameters of the evolution . to complete the proofs of theorems [ czero ] [ meq ] we need to show that decompositions with a large number of switches can not be optimal .
we defer this to the next section .
in this section we are going to show that certain decompositions are not optimal , even though they satisfy the necessary conditions of the pontryagin s maximum principle .
this will give us constraints on the number of control switches in optimal decompositions .
let us recall the construction of @xmath45 based on the quaternions .
the algebra of quaternions @xmath315 has a basis @xmath316 and relations @xmath317 , @xmath318 , @xmath319 , @xmath320 .
similar to the complex numbers , we have the conjugation on @xmath315 , given by @xmath321 , @xmath322 , @xmath323 , @xmath324 , and the norm : @xmath325 .
every non - zero element of @xmath326 has a multiplicative inverse given by @xmath327 .
the lie algebra @xmath329 of the group @xmath45 is the tangent space at identity , it is a 3-dimensional subspace in @xmath315 spanned by @xmath330 .
we are going to identify this lie algebra with @xmath38 via @xmath331 , @xmath332 , @xmath333 , where @xmath334 is the standard basis of @xmath38 . since @xmath335 = ij - ji = 2k$ ] , @xmath336 = jk - kj = 2i$ ] , @xmath337 = ki - ik = 2j$ ]
, we see that two lie algebra structures on @xmath38 coming from @xmath157 and @xmath329 differ by a factor of @xmath292 . for this reason there is a factor of @xmath292 in the formula for the homomorphism @xmath338 :
@xmath339 here for a vector @xmath340 the exponential is computed in the algebra of quaternions @xmath341 .
note that the rotation operator @xmath342 is also an exponential : @xmath343 .
pontryagin s maximum principle that we use above is essentially a local first derivative test . in order to obtain stronger results ,
we need to either apply non - local transformations ( those that do not come from a small variation of parameters ) or use higher derivatives . in proposition [ epsfive ] we will be using the second derivative in order to show that certain decompositions are not optimal .
an example of a non - local transformation is the identity @xmath348 where @xmath349 .
this trivial observation may be generalized in the following way .
suppose @xmath350 is a rotation in angle @xmath109 .
then we get a relation @xmath351 .
note that both sides of this equality have the same cost .
this non - local relation and its consequences will be quite useful for our analysis .
clearly , @xmath354 is a rotation in angle @xmath109 if and only if @xmath355 is the identity matrix , but @xmath354 is not identity .
this is equivalent to @xmath356 , @xmath357 in @xmath45 .
it is easy to see that the only solutions to @xmath358 are @xmath359 .
thus the preimages of rotations in angle @xmath109 are precisely @xmath360 with @xmath361 , or equivalently , @xmath362 .
since @xmath363 , this becomes @xmath364 . for @xmath365
this is equivalent to @xmath353 .
the group @xmath45 acts on its lie algebra @xmath329 by conjugation , and its center @xmath376 acts trivially .
this gives the action of @xmath9 on @xmath329 , which is the natural action of @xmath9 on @xmath38 .
since this action is transitive on pairs of unit vectors with a given angle between them , we may set without loss of generality @xmath377 , @xmath378 .
we complete this to a basis of @xmath329 by setting @xmath379 = k \sin \alpha$ ] .
we can easily verify that @xmath380 = 2 y - 2 \cos \alpha x , \quad & \quad [ z , y ] = - 2 x + 2 \cos \alpha y , \\ xzx = z , \quad & \quad yzy = z . \end{split}\ ] ] we also note that @xmath381 and @xmath382 and likewise for @xmath6 .
we may assume without loss of generality that @xmath396 .
let us begin with the case of @xmath392 .
we take its preimage under @xmath338 : @xmath397 , where @xmath398 , @xmath399 , @xmath400 , @xmath401 .
we claim that the decomposition @xmath402 given by the previous proposition will have a lower cost .
since @xmath403 we need to show that @xmath404 , where @xmath405 we have @xmath406 and @xmath407 . if @xmath408 then @xmath409 and we get that the new cost is lower . if @xmath410 then @xmath411 since @xmath412 and @xmath413 , we get that @xmath414 , so the new cost is again lower .
we now apply the same approach to @xmath394 .
we again take its preimage @xmath415 in @xmath45 and transform it into @xmath416 using proposition [ flip ] . here
the values of @xmath418 , @xmath419 are still given by ( [ sprime ] ) with @xmath412 .
we have @xmath407 and consider the sign of @xmath418 .
the case @xmath408 is treated in the same way as before . when @xmath410 we consider two subcases : @xmath420 and @xmath23 .
if @xmath420 , the claim of the proposition follows from the observation that on the diagrams ( a ) , ( c ) in fig .
[ kltc ] the arcs , , and correspond to an angle not exceeding @xmath421 .
let us assume @xmath23 . to show that the transformed expression has a lower cost , we need to prove that @xmath422 .
however @xmath423 . since @xmath424 , we get @xmath422 , which completes the proof of the proposition .
let us assume by contradiction that the given decomposition is optimal .
as before , we take a preimage @xmath429 , where @xmath430 , @xmath431 , @xmath432 .
we shall express the given decomposition in the following way : @xmath433 we are going to solve for @xmath434 , @xmath435 , @xmath436 and @xmath437 in terms of @xmath438 , and show that the new decomposition has a lower cost .
note that the parameters @xmath435 and @xmath436 are bound by the relation @xmath439 .
we will use the campbell - hausdorff formula ( [ ch ] ) to rewrite both sides of ( [ twoexp ] ) in the form @xmath440 we shall calculate @xmath441 up to the second order in @xmath438 . applying ( [ ch ] ) to the left hand side of ( [ twoexp ] ) , we get that @xmath442 + { { o}}(\ep^2),\ ] ] where @xmath443 and @xmath444 let us carry out the detailed calculations
. we shall use the basis @xmath445 and relations ( [ rel ] ) as in the proof of the proposition [ flip ] .
@xmath446 next , @xmath447 doing the same calculations for the right hand side of ( [ twoexp ] ) , we get @xmath448 + [ l_3 , l_5 ] + [ l_3 , l_6 ] + [ l_4 , l_5 ] + [ l_4 , l_6 ] + [ l_5 , l_6 ] \right ) + { { o}}(\ep^2),\ ] ] where @xmath449 we begin by solving ( [ twoexp ] ) to the first order in @xmath438 . equating ( [ firstl ] ) with ( [ secondl ] )
we get : @xmath450 we divide both sides of this equation by @xmath451 , which allows us to express everything in terms of @xmath452 . using the relation @xmath453 ,
we further eliminate @xmath454 . to make the equations more compact we denote @xmath455 by @xmath456 . by proposition [ epsthree ]
we have @xmath457 .
since we also have the relation @xmath458 , we use the taylor expansion to find the relation between @xmath435 and @xmath436 to the first order : @xmath459 expressing this in terms of @xmath456 , we get @xmath460 equating the coefficients at @xmath461 , we get a system of equations @xmath462 the determinant of this system is the jacobian of ( [ twoexp ] ) and equals @xmath463 since @xmath457 , @xmath133 , @xmath74 , we see that the only case when the jacobian vanishes is @xmath464 , @xmath465 , @xmath132 .
we will consider this case separately below . in all other cases
the jacobian is non - zero , hence by the implicit function theorem , equation ( [ twoexp ] ) has a unique solution for small @xmath438 .
this completes the proof of the proposition for the decomposition @xmath471 .
the cases of the decompositions obtained from this one by applying symmetries @xmath32 are analogous .
for example , in the case of @xmath472 , we use the transformation @xmath473 the cost of the right hand side is @xmath474 which is lower than the cost @xmath475 of the left hand side .
it follows from proposition [ epsfive ] that for the optimal decompositions corresponding to the trajectory , and symmetric to it , the number of factors is at most @xmath124 .
this corresponds to pattern ( i ) in theorems [ czero ] [ mlt ] .
suppose @xmath476 .
for the trajectory , and symmetric to it , the number of factors is bounded by @xmath292 , since the evolution times and exceed @xmath109 , and optimal decompositions can not have such time parameters .
the case of two factors is incorporated in patterns ( iii ) and ( vii ) with @xmath477 in theorems [ czero ] and [ mgt ] . in the case
@xmath478 , the decompositions corresponding to the trajectory , could have up to @xmath479 factors , since @xmath6-evolution time exceeds @xmath109 , but @xmath5-evolution time does not exceed @xmath480 , which is less than @xmath109 .
this corresponds to pattern ( viii ) in theorem [ mlt ] . to complete the proof of theorems [ czero ] [ meq ] , we need to establish a bound on the number of factors for the trajectories that pass through the critical points . for the critical points
@xmath268 with @xmath481 the existence of such a bound immediately follows from the diagrams in fig .
[ kgtc ] , since the trajectories connected to these points are full circles and require time evolution of @xmath482 to complete the circle , while any evolution with time exceeding @xmath109 is not optimal .
let us consider a preimage @xmath488 in @xmath45 for @xmath489 . here
@xmath490 , @xmath491 .
it follows from ( [ tim ] ) that @xmath492 then @xmath493 by lemma [ dz ] , the image of @xmath303 in @xmath9 is a rotation in angle @xmath109 and the first equality in part ( b ) holds .
it is easy to see that @xmath303 is orthogonal to @xmath92 : @xmath494 which implies that the axis of rotation corresponding to @xmath303 is orthogonal to @xmath92 and also that @xmath495 . taking the exponential of both sides
, we get that @xmath496 , from which the third claim of part ( b ) follows .
first we consider the case @xmath481 and @xmath426 .
we have pointed out above that the factor @xmath499 may only appear when @xmath500 and there will be only one such a factor in that case . in case when @xmath501 , we have that @xmath93 is proportional to @xmath5 , and we do not need to consider the factors of the form @xmath499 at all .
consider an optimal decomposition with factors @xmath502 . without loss of generality
assume that the time parameter in the first such factor is positive
. then it will necessarily have the form @xmath503 where @xmath504 and for @xmath505 @xmath506 by proposition [ wconj ] we can combine all factors of type @xmath498 into one without changing the cost : @xmath507 next suppose @xmath25 .
using proposition [ wconj](c ) and applying the above argument we see that there is an optimal decomposition with at most one factor of type @xmath508 .
this completes the proof of theorem [ meq ] .
now let us consider the case @xmath132 .
here we could have a decomposition that contains factors of both types , @xmath509 and @xmath499 .
note that @xmath510 .
suppose that an optimal decomposition of @xmath18 contains a factor @xmath498 with @xmath511 .
this factor will be followed by either an @xmath5-evolution or @xmath72-evolution .
let us assume it is @xmath5-evolution that follows .
if the time parameter for @xmath5-evolution is less than @xmath109 , that will be the last factor in the decomposition , as the control switch can not occur .
otherwise , we get @xmath498 followed by a factor @xmath512 . but this will imply optimality of the expression @xmath513 , which gives a contradiction since @xmath514 can not be optimal since it does not satisfy the necessary conditions for optimality of theorem [ pmp ] .
all other cases are analogous and we conclude that in case @xmath515 factors @xmath516 in optimal decompositions may be preceded or followed by just a single factor @xmath517 or @xmath518 with @xmath519 , thus completing the proof of theorem [ czero ] .
[ fivew ] let @xmath520 , @xmath133 .
suppose an optimal decomposition for @xmath18 contains a factor @xmath498 with @xmath511 .
then there exists an optimal decomposition of @xmath0 , which is a subword in one of the following : @xmath521 let us show that in an optimal decomposition of @xmath0 the number of factors following @xmath498 is at most two . indeed ,
if it is followed by three or more factors , such evolution must begin with either @xmath489 or @xmath522 . by proposition [ wconj ] , @xmath523 .
this will be followed by evolution with control @xmath524 or @xmath6 .
this would imply optimality of either @xmath525 or @xmath526 for small @xmath425 .
let us show that these decompositions are not optimal .
consider a preimage @xmath527 in @xmath45 for @xmath525 . here
@xmath528 , @xmath529 .
since @xmath530 we get that @xmath531 and we can assume that @xmath532 .
choose @xmath533 such that @xmath534 .
since @xmath535 , we conclude that @xmath536 .
then by proposition [ flip](a ) we get @xmath537 however the latter decomposition is not optimal since it does not correspond to a trajectory in fig .
[ kgtc ] , [ kltc ] , yet both sides in the above equality have the same cost .
this implies that @xmath525 is not optimal .
the argument for @xmath526 is analogous .
this completes the proof of proposition [ fivew ] and theorems [ mgt ] and [ mlt ] .
nasa , ( 2013 ) .
nasa ends attempts to fully recover kepler spacecraft , potential new missions considered .
[ online ] available at : http://www.nasa.gov/content/nasa-ends-attempts-to-fully-recover-kepler-spacecraft-potential-new-missions-considered [ accessed 14 aug .
2014 ] .
p. petersen , riemannian geometry , graduate texts in mathematics * 171 * , springer , 2006 .
g. v. smirnov , introduction to the theory of differential inclusions , graduate studies in mathematics * 41 * , amer .
soc . , 2002 . | euler proved that every rotation of a 3-dimensional body can be realized as a sequence of three rotations around two given axes .
if we allow sequences of an arbitrary length , such a decomposition will not be unique . in this paper
we solve an optimal control problem minimizing the total angle of rotation for such sequences .
we determine the list of possible optimal patterns that give a decomposition of an arbitrary rotation .
our results may be applied to the attitude control of a spacecraft with two available axes of rotation . |
fractal transformations are mappings between pairs of attractors of iterated function systems .
they are defined with the aid of code space structures , and can be quite simple to handle and compute .
they can be applied to digital images when the attractors are rectangular subsets of @xmath0 .
they are termed `` fractal '' because they can change the box - counting , hausdorff , and other dimensions of sets and measures upon which they act . in this paper
we substantially generalize and develop the theory and we illustrate how it may be applied to digital imaging . previous work was restricted to fractal transformations defined using fractal tops .
fractal tops were introduced in @xcite and further developed in @xcite .
the main idea is this : given an iterated function system with a coding map and an attractor , a _ section _ of the coding map , called a tops function , can be defined using the `` top '' addresses of points on the attractor . given two iterated function systems each with an attractor , a coding map , and a common code space , a mapping from one attractor to the other can be constructed by composing the tops function , for the first iterated function system , with the coding map for the second system .
under various conditions the composed map , from one attractor to the other , is continuous or a homeomorphism . in the cases of affine and projective iterated function systems , practical methods based on the chaos game algorithm @xcite are feasible for the approximate digital computation of such transformations .
fractal tops have applications to information theory and to computer graphics .
they have been applied to the production of artwork , as discussed for example in @xcite , and to real - time image synthesis @xcite . in the present paper
we extend the theory and applications .
much of the material in this paper is new .
the underlying new idea is that diverse sections of a coding map may be defined quite generally , but specifically enough to be useful , by associating certain dynamical systems with the iterated function system .
these sections provide novel collections of fractal transformations ; by their means we generalize the theory and applications of fractal tops .
we establish properties of fractal transformations , including conditions under which they are continuous .
the properties are illustrated by examples related to digital imaging . a notable result , theorem [ goldthm ] , states the existence of nontrivial fractal homeomorphisms between attractors of some affine overlapping iterated function systems .
the proof explains how to construct them .
an example of one of these new homeomorphisms , applied to a picture of lena , is illustrated in figure [ goldenlenna ] .
goldenlenna.png in section [ ifssec ] we review briefly the key definitions and results concerning point - fibred iterated function systems on compact hausdorff spaces .
since this material is not well - known , it is of independent interest .
the main result is theorem [ mainifsthm ]
. this can be viewed as a restatement of some ideas in @xcite ; it describes the relationship between the coding map and the attractor of a point - fibred iterated function system
. in section [ topsec ] we define , and establish some general properties of , fractal transformations constructed using sections of coding maps . in theorem [ sectionthm ] we present some general properties of coding maps
. then we use coding maps to define fractal transformations and , in theorem [ ctyfractaltransthm ] , we provide sufficient conditions for a fractal transformation to be continuous or homeomorphic . in section [ masksec ]
we define two different types of section of a coding map : ( i ) with the aid of a _ masked _ dynamical system ; and ( ii ) with the aid of fractal tops .
theorem [ maskthm ] establishes the connection between the masked dynamical system and a _ masked section _ of the coding map .
theorem [ maskbranchthm ] includes a statement concerning the relationship between the masked section to the coding map and the shift map .
here we also establish the relationship between fractal tops and masked systems .
a key result , theorem [ l : disjoint ] , gives a condition under which the ranges of different masked sections intersect in a set of measure zero .
this enables the approximate storage of multiple images in a single image , as illustrated in figure [ allmasked ] .
in section [ appsec ] we apply and illustrate the theoretical structures of sections 2,3 , and 4 , in the context of digital imaging .
our goal is to illustate the diversity of imaging techniques that are made feasible by our techniques , and to suggest that fractal transformations have a potentially valuable role to play in digital imaging . in section [ appsec1 ]
we illustrate how fractal transformations may be applied to image synthesis , that is to making artificial interesting and even beautiful pictures .
specifically we explain how the technique of color - stealing @xcite extends to masked systems . in section [ fhomsec ]
we apply fractal homeomorphisms , using theorem [ ctyfractaltransthm ] , to transform digital images , for image beautification , roughening , and special effects ; in particular , we present and illustrate theorem [ goldthm ] which extends the set of known affine fractal homeomophisms .
in section [ filtsec ] we consider the idea of composing a fractal transformation , discretization , and the inverse of the transformation to make idempotent image filters . in section [ packsec ]
we apply theorem [ l : disjoint ] to the approximate storage or encryption of multiple images in a single image . in section [ meassec ]
we provide a second technique for combining several images in one : it combines invariant measures of a single iterated function system with several sets of probabilities , to make a single `` encoded '' ] [ ifssec]point - fibred iterated function systems ---------------------------------------------- let @xmath1 be a nonempty compact hausdorff space , and let @xmath2 be the set of nonempty compact subsets of @xmath1 .
it is known that @xmath2 endowed with the vietoris topology is a compact hausdorff space , see for example ( * ? ? ?
* theorem 2.3.5 , p.17 ) .
this encompasses the well - known fact that if @xmath1 is a compact metric space then @xmath2 endowed with the hausdorff metric is a compact metric space .
let @xmath3 be a finite index set with the discrete topology .
let @xmath4 be a sequence of continuous functions .
following @xcite@xmath5 @xmath6 is called an _ iterated function system _ over @xmath1 . following ( * ? ? ?
* definition 4.1.4 , p.84 ) we define a map @xmath7 for all sequences @xmath8 belonging to @xmath9 .
the map is well - defined because @xmath10 is the intersection of a nested sequence of nonempty compact sets .
the following definition is based on ( * ? ? ?
* definition 4.3.6 , p.97 ) .
let @xmath11 be an iterated function system over a compact hausdorff space @xmath1 .
if @xmath10 is a singleton for all @xmath12 then @xmath13 is said to be * * point - fibred**@xmath5 and the * coding map * of @xmath13 is defined by@xmath14 where @xmath15 denotes the range of @xmath16 .
theorem [ kieningerthm ] , due to kieninger , plays a central role in this paper .
it generalizes a classical result of hutchinson @xcite that applies when @xmath1 is a compact metric space and each @xmath17 is a contraction .
[ kieningerthm]let @xmath9 have the product topology .
if @xmath13 is a point - fibred iterated function system on a compact hausdorff space @xmath1 then the coding map @xmath18 is continuous .
this follows from ( * ? ? ?
* proposition 4.3.22 , p.105 ) .
we define @xmath19 by slight abuse of notation we use the same symbol @xmath13 for the iterated function system , the maps that it comprises , and the latter function .
we define @xmath20 , the identity map on @xmath1 , and @xmath21 for @xmath22 .
the following definition is a natural generalization of the notion of an attractor of a contractive iterated function system , see for example ( * ? ? ?
* definition on p.1193 and theorem 11.1 , p.1206 ) , @xcite , and also @xcite .
[ attractordef]let @xmath13 be an iterated function system on a compact hausdorff space @xmath1 .
an * attractor * of @xmath13 is @xmath23 @xmath2 with these properties : ( i ) @xmath24 ; ( ii ) there exists an open set @xmath25 such that @xmath26 and @xmath27 for all @xmath28 with @xmath29 .
( the limit is with respect to the vietoris topology on @xmath2 . )
the largest open set @xmath30 such that equation [ attractoreq ] holds for all @xmath31 with @xmath32 is called the * basin * of @xmath33 . the relationship between coding maps and attractors is provided by noting that @xmath34 .
this leads to the following theorem .
[ mainifsthm]if @xmath13 is a point - fibred iterated function system on a compact hausdorff space @xmath1 then \(i ) @xmath35 has a unique fixed - point @xmath36 i.e. @xmath37 ; \(ii ) @xmath33 is the unique attractor of @xmath13 ; \(iii ) @xmath33 is equal to the range of the coding map @xmath16 , namely@xmath38 \(iv ) the basin of @xmath33 is @xmath39 \(v ) if @xmath29 then @xmath40 for all @xmath41 this follows from ( * ? ? ?
* proposition 4.4.2 , p.107 , see also proposition 3.4.4 , p.77 ) .
the following remark tells us that if an iterated function system possesses an attractor then it is point - fibred when restricted to a certain neighborhood of the attractor .
necessary and sufficient conditions for an iterated function system of projective transformations to possess an attractor are given in @xcite .
let @xmath33 be an attractor of an iterated function system @xmath13 on a compact hausdorff space @xmath1 , and let @xmath42 be the basin of @xmath43 then , following @xcite , @xmath13 defines an iterated closed relation @xmath44 , and @xmath33 is an attractor of @xmath45 . by (
* theorem 7.2 , p.1193 ) there exists a compact neighborhood @xmath46 of @xmath33 such that @xmath47 and @xmath48 ( @xmath49 denotes the interior of the set @xmath50 in the subspace topology induced on @xmath51 by @xmath1 . ) it follows that @xmath52 is a point - fibred iterated function system on a compact hausdorff space .
the attractor of @xmath53 is @xmath43 let @xmath13 be a point - fibred iterated function system on a compact hausdorff space .
the set @xmath9 is called the * code space * of @xmath13 .
a point @xmath12 is called an * address * of @xmath54 . in the rest of this paper
the underlying space @xmath1 is a compact hausdorff space .
also in the rest of this paper the symbols @xmath55 denote point - fibred iterated function systems on compact hausdorff spaces .
we will say that an iterated function system is _ injective _ when all of the maps that it comprises are injective .
we will say that an iterated function system is _ open _ when all of the maps that it comprises are open .
here we present a generalized theory of fractal transformations and establish some continuity properties .
fractal transformations are defined using sections of coding maps .
we are concerned with continuity properties ; for example , theorem [ ctyfractaltransthm ] ( i ) provides a sufficient condition for a fractal transformation to be continuous .
[ branchdef]let @xmath18 be the coding map of @xmath13 .
a subset @xmath56 is called an * address space * for @xmath13 if @xmath57 and @xmath58 is one - to - one .
the corresponding map @xmath59 is called a * section * of * * * * @xmath16 .
theorem [ sectionthm ] summarises the properties of sections of @xmath60 [ sectionthm]let @xmath61 be a point - fibred iterated function system on a compact hausdorff space , with attractor @xmath33 , code space @xmath9 , and coding map @xmath62 if @xmath63 is a section of @xmath16 then \(i ) @xmath63 is bijective ; \(ii ) @xmath64 @xmath33 is continuous ; \(iii ) @xmath65 , the identity map on @xmath66 and @xmath67 , the identity map on @xmath68 ; \(iv ) if @xmath13 is injective and @xmath69 for all@xmath70 with @xmath71 , then @xmath72 ; \(v ) if @xmath68 is closed then @xmath63 is a homeomorphism ; \(vi ) if @xmath33 is connected and @xmath33 is not a singleton , then @xmath73 is not continuous .
\(i ) by definition [ branchdef ] @xmath74 is bijective , so @xmath75 @xmath76 is bijective .
\(ii ) by theorem [ kieningerthm ] @xmath18 is continuous .
it follows that @xmath77 is continuous .
\(iii ) if @xmath78 then @xmath79 also @xmath80 .
\(iv ) suppose @xmath81 .
then there are @xmath82 , @xmath83 , such that @xmath84 .
we show that this is impossible . if @xmath84 then theorem [ mainifsthm ] ( v ) implies @xmath85 .
then @xmath86 and @xmath87 .
since @xmath13 is injective , each @xmath17 is injective , which implies that @xmath88 is injective .
since @xmath86 , it now follows that @xmath89 . since @xmath90 and
@xmath91 it now follows that @xmath92 .
\(v ) if @xmath56 is closed then it is compact , because @xmath9 is compact .
it follows that @xmath77 is a continuous bijective mapping from a compact space @xmath68 onto a hausdorff space @xmath33 . by (
* theorem 5.6 , p.167 ) it follows that @xmath93 is a homeomorphism .
it follows that @xmath63 is a homeomorphism .
\(vi ) suppose that @xmath33 is connected and @xmath33 is not a singleton .
it follows that @xmath68 is not a singleton .
it follows that @xmath68 is not connected .
( since @xmath68 contains more than one point and is a subset of @xmath94 which is totally disconnected when it contains more than one point , it follows that @xmath68 is not connected . )
now suppose @xmath63 is continuous .
then @xmath95 is a homeomorphism .
it follows that @xmath33 is not connected . but
@xmath33 is connected .
so @xmath63 is not continuous . in definition [ ftransdef ]
we define a type of transformation between attractors of iterated function systems by composing sections of coding maps with coding maps .
we call these transformations `` fractal '' because they can be very rough ; specific examples demonstrate that the graphs of these transformations , between compact manifolds , can possess a non - integer hausdorff - besicovitch dimension .
[ ftransdef]let @xmath61 be a point - fibred iterated function system over a compact hausdorff space @xmath1 .
let @xmath96 be the attractor of @xmath13 .
let @xmath97 be the coding map of @xmath13 .
let @xmath98 be a branch of @xmath99 .
let @xmath100 be a point - fibred iterated function system over a compact hausdorff space @xmath101 let @xmath102 be the attractor of @xmath103 .
let @xmath104 be the coding map of @xmath103 .
the corresponding * fractal transformation * is defined to be@xmath105 in theorem [ ctyfractaltransthm ] we describe some continuity properties of fractal transformations .
these properties make fractal transformations interesting for applications to digital imaging .
[ ctyfractaltransthm]let @xmath13 and @xmath103 be point - fibred iterated function systems as in definition [ ftransdef ] .
let @xmath106 be the corresponding fractal transformation .
\(i ) if , whenever @xmath107 , @xmath108 @xmath109 @xmath110 then @xmath111 is continuous .
\(ii ) if @xmath112 is an address space for @xmath103 , and if , whenever @xmath113 , @xmath108 @xmath114 @xmath110 then @xmath115 is a homeomorphism and @xmath116 .
\(i ) assume that , whenever @xmath107 , @xmath108 @xmath109 @xmath117 .
we begin by showing that @xmath118 is the same as @xmath119 let @xmath120 .
then there is @xmath121 such that @xmath122 because @xmath123 is a code space for @xmath124 hence @xmath125 ( using theorem [ sectionthm ] ( iii ) ) .
hence @xmath126 , where we have used our initial assumption .
it follows that @xmath127 .
it follows that @xmath128 is a continuous map from a compact space @xmath129 to a compact hausdorff space @xmath130 but @xmath131 is the composition of a continuous mapping @xmath132 from a compact hausdorff space @xmath129 onto a hausdorff space @xmath133 , with a mapping @xmath134 from @xmath133 into a hausdorff space @xmath102 .
it follows by a well - known theorem in topology , see for example ( * ? ? ?
* proposition 7.4 , p. 195)@xmath5 that @xmath135 @xmath134 is continuous .
\(ii ) assume that @xmath136 is an address space for @xmath103 , and that , whenever @xmath137 , @xmath108 @xmath114 @xmath117 .
then by ( i ) both of the mappings @xmath138 and @xmath139 are continuous . using the fact that the range of @xmath140 is @xmath123 it is readily checked that @xmath141 @xmath142 and @xmath143 @xmath144 hence @xmath115 is a homeomorphism and @xmath145 .
suppose that the equivalence relations @xmath146 and @xmath147on @xmath148 , induced by @xmath149 and @xmath150 respectively , are the same .
then it is well known that , using compactness , the quotient topological space @xmath151 is homeomorphic to both @xmath133 and @xmath102 .
see @xcite for discussion of relationships between the topology of an attractor and the equivalence class structure induced by a coding map .
in order to construct a fractal transformation we need to specify a section of @xmath16 . in order to construct a section of @xmath16
we have to construct an address space @xmath68 ; that is , we need to specify one element from each of the sets in the collection @xmath152 . to do this in a general way
seems to be difficult ; for example , if @xmath153 contains a nonempty open set @xmath154 , then @xmath155 is non - denumerable for all @xmath156 .
however there are two particular related methods .
these methods yield interesting structure ; for example , in both cases the resulting address space @xmath56 is mapped into itself by the shift operator , see theorem [ maskbranchthm ] ( ii ) .
they are as follows .
\(a ) ( masked iterated function system method . )
this method requires that @xmath13 is injective .
define a dynamical system on @xmath157 with the aid of inverses of the functions in @xmath13 .
follow orbits of @xmath158 to define @xmath159 ; in effect one uses a markov partition associated with @xmath158 to define @xmath123 .
\(b ) ( fractal tops method . )
use the dictionary order relation on @xmath9 to select a unique element of @xmath155 for each @xmath160 .
this method applies when @xmath13 is not required to be injective .
when @xmath13 is injective , it is a special case of ( a ) . to date
, the fractal tops method seems to be the easiest to convert to computational algorithms and applications .
definition [ maskdef ] introduces a special partition of an attractor .
[ maskdef]let @xmath13 be a point - fibred iterated function system on a compact hausdorff space .
let @xmath33 be the attractor of @xmath13 . a finite sequence of sets @xmath161
is called a * mask * for @xmath13 if 1 .
@xmath162 , @xmath163 ; 2 .
@xmath164 , @xmath165 , @xmath71 ; 3 .
@xmath166 . note that for any @xmath78 there exists a unique @xmath163 such that @xmath167 .
this enable us , in definition [ dynsysdef ] , to define an associated dynamical system on the attractor . [ dynsysdef ]
let @xmath168 be a mask for an injective point - fibred iterated function system @xmath13 with attractor @xmath43 the associated * masked dynamical system * for @xmath13 is @xmath169 theorem [ maskthm ] associates a unique section of @xmath16 with a masked dynamical system .
[ maskthm]let @xmath13 be an injective point - fibred iterated function system with attractor @xmath43 let @xmath170 be a masked dynamical system for @xmath13 , associated with mask @xmath171 .
let @xmath78 and let @xmath172 be the orbit of @xmath173 under @xmath158 ; that is , @xmath174 and @xmath175 for @xmath176 .
let @xmath177 be the unique symbol such that @xmath178 , for @xmath179 .
then @xmath180 is an address space for @xmath13 .
let @xmath78 .
we begin by proving that , for all @xmath181 and @xmath182 we have@xmath183 fix @xmath184 we use induction on @xmath185 since @xmath186 it follows that
@xmath187 so @xmath188 .
it follows that equation [ inductioneq ] is true for @xmath189 .
suppose that equation [ inductioneq ] is true for @xmath190 .
it follows that @xmath191 .
we also have @xmath192 so @xmath193 which implies @xmath194 hence equation [ inductioneq ] is true for @xmath195 .
this completes the induction on @xmath196 .
it follows that @xmath197 for all @xmath184 it follows that @xmath198 .
it follows that @xmath199 for all @xmath200 it follows that @xmath201 ; that is , @xmath202 is injective , suppose @xmath203 for some @xmath204 then for some @xmath205 we have @xmath206 which implies @xmath207 . hence @xmath208 it follows that @xmath209 .
let @xmath13 be an injective point - fibred iterated function system .
the address space @xmath210 provided by theorem [ maskthm ] is called a * masked address space * for @xmath13 .
the corresponding section of @xmath16 , say @xmath211 , is called a * masked section * of @xmath60 masked address spaces and masked sections have all of the properies of address spaces and sections , such as those in theorem [ sectionthm ] , and associated fractal transformations have the properties in theorem [ ctyfractaltransthm ] .
but these objects have additional properties that derive from the existence and structure of the masked dynamical system .
some of these additional properties are described in theorem [ maskbranchthm ] .
[ maskbranchthm]if @xmath61 is an injective point - fibred iterated function system , with attractor @xmath33 , code space @xmath9 , coding map @xmath18 , mask @xmath212 , masked address space @xmath213 and masked section @xmath211 , then \(i ) if @xmath13 is open then @xmath211 is continuous at @xmath78 if and only if @xmath214 for all @xmath22 ; \(ii ) the shift map @xmath215 is well - defined , with @xmath216 ; \(iii ) the following diagram commutes @xmath217{ccc}a & \overset{t}{\rightarrow } & a\\ \tau\downarrow\text{\ \ \ \ } & & \text { \ \ \ } \downarrow\tau\\ \omega_{\mathcal{m } } & \overset{s}{\rightarrow } & \omega_{\mathcal{m}}\end{array } \label{commutediagram}\ ] ] \(iv ) if there is @xmath163 such that @xmath218 then @xmath219 \(i ) suppose that @xmath214 for all @xmath220 .
let @xmath221 converge to @xmath222 let @xmath223 and let @xmath224 . since @xmath225 there is an integer @xmath226 such that @xmath227 for all @xmath228 .
since , for all @xmath163 , @xmath229 is continuous ( because @xmath13 is invertible and open ) we have @xmath230converges to @xmath231 as @xmath232 .
similarly , for any given @xmath233 , there is @xmath234 such that @xmath235 for all @xmath236 and all @xmath237 .
it follows that , for any given @xmath233 , @xmath238 for all @xmath236 and all @xmath237 . it follows that @xmath239 converges to @xmath240 , i.e. @xmath241 converges to @xmath242 it follows that @xmath243 is continuous at @xmath173 . to prove the converse we assume that it is not true that @xmath244 for all @xmath22 .
it follows that there is some @xmath245 such that @xmath246 here as in the first part of the proof , we write @xmath247 .
it follows that there is a sequence @xmath248 that converges to @xmath249 with @xmath250 for all @xmath196 .
( any neighborhood of @xmath249 must contain a point that is in @xmath251 . )
it follows that @xmath252 converges to @xmath253 since @xmath254 is continuous .
( we define @xmath255 . ) but @xmath256 while @xmath257 because @xmath258 .
it follows that @xmath243 is not continuous .
the desired conclusion follows at once .
( ii)&(iii ) let @xmath259 let @xmath260 .
then @xmath261 , whence @xmath262 .
it follows that the shift map @xmath215 is well - defined , with @xmath216 .
\(iv ) let @xmath263 and let @xmath163 be such that @xmath218 .
then @xmath264 whence @xmath265 . by ( iii ) we have @xmath266 so @xmath267 .
we remark that masked dynamical systems are related to markov partitions in the theory of dynamical systems .
see for example ( * ? ? ?
* proposition 18.7.8 , p.595 ) .
in general a masked dynamical system @xmath170 depends on the mask @xmath268 . by suitable choice of mask
we can sometimes obtain a dynamical system with a desired feature such as continuity , or which relates the iterated function system to a known dynamical system , as illustrated in the following example .
[ e : example1 ] consider the ifs @xmath269 where @xmath270 is a parameter .
the attractor of @xmath271.$ ] let @xmath272,\ m_{2}=(\frac{1}{2},1]$ ] .
then the masked transformation @xmath273 is continuous .
this is the well - known one - parameter tent map dynamical system , see for example ( * ? ? ?
* exercise 2.4.1 , p.78 ) .
we note that , for any @xmath274 there exist a positive integer @xmath275 such that @xmath276 $ ] for all @xmath277 ( see fig .
[ example1 ] ) . as a consequence ,
if @xmath278 denotes the set of masked addresses of points in @xmath279 $ ] then the masked address space for @xmath13 is @xmath280 example1r.png theorem [ l : disjoint ] concerns the relationship between masked address spaces corresponding to distinct masks .
it has an application to packing multiple images into a single image , as illustrated in example [ expack3 ] .
[ l : disjoint ] let @xmath13 be a point - fibred iterated function system with attractor @xmath33 . let @xmath170 be a masked dynamical system for @xmath13 corresponding to mask @xmath268 .
let @xmath281 be a measure on @xmath33 .
let @xmath282 be a mask for @xmath13 such that @xmath283 let @xmath284 and @xmath285 be the masked sections of @xmath16 corresponding to @xmath268 and @xmath282 .
then @xmath286 for @xmath287almost all @xmath78 .
if @xmath288 it follows that @xmath289 for all @xmath205 , where @xmath290 is the masked dynamical system corresponding to @xmath282 .
then @xmath291 for all @xmath205 . but
this is impossible for @xmath292almost all @xmath78 .
`` fractal tops '' is the name we use to refer to the mathematics of tops functions , tops code spaces , tops dynamical systems , and associated fractal transformations ; see for example @xcite . here
we show that , in the case where @xmath13 is injective , fractal tops arise as a special case of masked iterated function systems .
specifically , a _ tops code space _ is a special case of _ masked address space , a tops function _ is a special case of a _ section of _ @xmath293 and a _ tops dynamical system _ is a special case of a _ _ masked dynamical system__@xmath294 the computations associated with fractal tops tend to be less complicated than those for masked systems .
define a dictionary ordering ( see for example @xcite ) on @xmath9 as follows : if @xmath295 , @xmath296 then@xmath297 where @xmath205 is the least index for which @xmath298 , for all @xmath295 .
with this ordering , every subset of @xmath299 possesses a greatest lower bound and a least upper bound . since @xmath300 is continuous and @xmath9 is compact , @xmath301 possesses a unique largest element , @xmath302 , for each @xmath78 .
this allows us to define an address space @xmath56 for @xmath13 by@xmath303 the corresponding section of @xmath16 is@xmath304 if @xmath13 is injective then the tops dynamical system is a masked dynamical system , corresponding to the mask defined by@xmath305 the fact that @xmath306 can be computed from the set @xmath307 without reference to other points on the orbit of @xmath173 simplifies the computation of @xmath306 in applications , see for example @xcite .
note that we can use the orbits of a tops dynamical system to calculate the top address of any point @xmath78 according to @xmath308 where@xmath309
here we generalize the technique of color - stealing , introduced in @xcite and implemented for example in @xcite .
define a _ picture _ to be a function of the form @xmath310 where @xmath311 is a color space .
the set @xmath312 is the _ domain _ of the picture .
we are concerned with situations where @xmath312 is a subset of an attractor of an iterated function system .
let @xmath13 be an injective iterated function system with attractor @xmath313 .
let @xmath314 be an iterated function system with attractor @xmath315 and the same code space @xmath9 as for @xmath13 .
let @xmath268 be a mask for @xmath13 and let @xmath284 be the corresponding section of @xmath316 let @xmath317 be the coding map for @xmath314 .
then we can define a mapping @xmath318 from the space of pictures on @xmath319 into the space of pictures on @xmath33 according to @xmath320 we refer to this procedure as color - stealing because colors from the picture @xmath321 are mapped onto the attractor @xmath33 to define the new picture @xmath322 .
it follows from theorem [ maskbranchthm ] ( i ) that the transformation @xmath323 is continuous at all points @xmath78 whose orbits lie in @xmath324 @xmath325 where @xmath326 denotes the boundary of @xmath327 . in some cases , such as those in example [ stealingex ] , @xmath328 is a set of lebesgue measure zero , so the transformation is continuous almost everywhere .
this explains patches of similar colors tends to exist in pictures that are obtained by color - stealing from real world photos , where patches of similar colors occur for physical reasons .
[ stealingex ] figure [ combined ] illustrates color - stealing using ( i ) fractal tops ( left ) , and ( ii ) a masked iterated function system ( right ) that is not a tops system .
the picture @xmath329 from which colors are stolen is lena embedded in a black surround , shown in the middle panel . in ( i ) @xmath314 is an affine iterated function system whose attractor @xmath319 is a filled square , the domain of @xmath329 , such that @xmath330 is a set of tiles that tile @xmath319 by rectangles . in ( i ) @xmath13 is the projective iterated function system @xmath331 , where@xmath332@xmath333{|c|c|c|c|c|c|c|c|c|c|}\hline $ n$ & $ a_{n}$ & $ b_{n}$ & $ c_{n}$ & $ d_{n}$ & $ e_{n}$ & $ k_{n}$ & $ g_{n}$ & $ h_{n}$ & $ j_{n}$\\\hline $ 1 $ & $ 19.05 $ & $ 0.72 $ & $ 1.86 $ & $ -0.15 $ & $ 16.9 $ & $ -0.28 $ & $ 5.63 $ & $ 2.01 $ & $ 20.0$\\ $ 2 $ & $ 0.2 $ & $ 4.4 $ & $ 7.5 $ & $ -0.3 $ & $ -4.4 $ & $ -10.4 $ & $ 0.2 $ & $ 8.8 $ & $ 15.4$\\ $ 3 $ & $ 96.5 $ & $ 35.2 $ & $ 5.8 $ & $ -131.4 $ & $ -6.5 $ & $ 19.1 $ & $ 134.8 $ & $ 30.7 $ & $ 7.5$\\ $ 4 $ & $ -32.5 $ & $ 5.81 $ & $ -2.9 $ & $ 122.9 $ & $ -0.1 $ & $ -19.9 $ & $ -128.1 $ & $ -24.3 $ & $ -5.8$\\\hline \end{tabular } .\ ] ] and @xmath314 is an affine ifs whose attractor is a filled square ; see also @xcite . in ( ii ) the iterated function system @xmath13 and mask @xmath268 are the same as in example [ goldenlennaex ] , while @xmath314 is a perturbed version of @xmath103 in example [ goldenlennaex ] .
combined.png under the conditions of theorem [ ctyfractaltransthm ] ( ii ) the fractal transformation @xmath115 is a homeomorphism@xmath294 such homeomorphisms can be applied to pictures to yield new pictures that have the same topological properties as the original .
for example the connectivity properties of the set defined by a particular color will be preserved , as will be the property that certain colors lie in an arbitrary neighborhood of a point . but
geometrical properties , such as hausdorff dimension and collinearity , may not be preserved and , indeed , may be significantly changed . techniques for constructing and computing fractal homeomorphisms using fractal tops , with projective , affine , and bilinear ifss , have been discussed in @xcite .
families of fractal homeomorphisms , built from such transformations in @xmath0 , may be established by using code space arguments .
typically , the attractors of the iterated function systems in question are non - overlapping , which simplifies the proofs : in some situations one only needs to show that the equivalence classes of addresses agree on certain straight line segments .
the resulting families of transformations are described by a finite sets of real parameters .
these parameters may be adjusted to achieve desired effects such as increased roughness , or continuous ( but non - differentiable ) transformation from a meaningless picture into a meaningful one .
both of these effects are illustrated in example [ roughex ] .
[ roughex ] figure [ transforms ] illustrates three fractal homeomorphisms of the unit square applied to lena .
all the iterated function systems involved are constructed using bilinear functions defined as follows .
let @xmath334^{2}\subset\mathbb{r}^{2}$ ] denote the unit square , with vertices @xmath335 let @xmath336 denote , in cyclic order , the successive vertices of a possibly degenerate quadrilateral .
we uniquely define a bilinear function @xmath337 such that @xmath338 by@xmath339 this transformation acts affinely on any straight line that is parallel to either the @xmath173-axis or the @xmath340-axis . for example , if @xmath341 is the restriction to @xmath342 of @xmath42 and if @xmath343 is the affine function defined by @xmath344 then @xmath345 .
sufficient conditions for a bilinear iterated function system to be point - fibred are given in @xcite .
each homeomorphism in figure [ transforms ] is generated using a pair of iterated function systems of the form in figure [ bilinear2 ] ; each such pair has the same address structure .
bilinear2.png the following theorem provides practical sufficient conditions for a bilinear ifs to be hyperbolic .
transforms.png it is more difficult to establish conditions under which pairs of masked overlapping iterated function systems , built from geometrical functions such as affines on @xmath346 yield homeomorphisms . in order to establish that a pair of masked iterated function systems provides a fractal transformation that is a homeomorphism , it is necessary to establish that the two masked address spaces agree .
interesting non - trivial cases involve overlapping iterated function systems , and can not be reformulated as fractal tops .
a beautiful family of such examples is provided by theorem [ goldthm ] .
[ goldthm ] let @xmath347,f_{1}(x)=ax , f_{2}(x)=by+(1-b)),$ ] where @xmath348 and @xmath349 have mask @xmath350 where @xmath351 $ ] , @xmath352,$ ] and @xmath353 $ ] .
let @xmath354,g_{1}(x)=bx , g_{2}(x)=ay+(1-a))$ ] have mask @xmath355 .
then there exists @xmath356 $ ] such that the corresponding fractal transformation @xmath357\rightarrow\lbrack0,1]$ ] is a homeomorphism when @xmath358 the inverse of this homeomorphism is @xmath359\rightarrow\lbrack0,1]$ ] defined by associating the mask @xmath360 with @xmath103 .
this is an outline .
let @xmath334^{2}\subset\mathbb{r}^{2}$ ] .
define affine transformations by @xmath361 let @xmath362 and @xmath363 .
we consider the dynamical system@xmath364{c}w_{-}^{-1}(x , y)\text { if } ( x , y)\in s_{-},\\ w_{+}^{-1}(x , y)\text { if } ( x , y)\in s_{+}. \end{array } \right.\ ] ] this possesses a `` repeller '' , a compact set @xmath365 such that @xmath366 in order to define @xmath367 , we define @xmath368 and @xmath369 @xmath367 is well - defined because it is the intersection of a decreasing sequence of nonempty compact sets .
( it is quite easy to see that @xmath370 is the graph of a monotone function from @xmath371 $ ] onto @xmath371.$ ] ) using symmetry about the line @xmath372 and the contractivity of @xmath373 and @xmath374 in both the @xmath173 and @xmath340 directions it can be proved that @xmath367 has the following properties .
\(i ) @xmath375 \(ii ) @xmath367 is symmetrical about the line @xmath376
\(iii ) @xmath377,$ ] where @xmath378 denotes the projection in the @xmath173 direction and @xmath379 denotes projection in the @xmath340 direction ; \(iv ) there is a continuous , monotone strictly increasing function @xmath380\rightarrow\lbrack0,1],$ ] such that @xmath381 , @xmath382 , @xmath383 for all @xmath384,$ ] and @xmath385\}$ ] ; \(v ) there is a unique @xmath356 $ ] such that @xmath386 \(vi ) there is a continuous , monotone strictly increasing function , @xmath387\rightarrow\lbrack0,1],$ ] such that @xmath388 , @xmath389 , @xmath390 for all @xmath391 $ ] , and @xmath392\}$ ] ; \(vii ) @xmath393 \(viii ) @xmath394 for all @xmath391 $ ] ; \(ix ) if @xmath395 , then the masked dynamical system @xmath396\rightarrow\lbrack0,1]$ ] obeys@xmath397;\ ] ] \(x ) if @xmath395 , then the masked dynamical system @xmath398\rightarrow\lbrack0,1]$ ] obeys@xmath399.\ ] ] these statements imply the theorem .
clearly essentially the same result and proof applies for any analogous pair of overlapping strictly increasing functions on @xmath371 $ ] .
figure [ goldenimage ] illustrates the `` repeller '' .
it is a subset of the attractor of the iterated function system @xmath400 and may be used to compute @xmath401 as illustrated in figure [ figure ] .
goldenimage.png an example of a fractal transformation , arising from a masked pair of overlapping affine iterated function systems , is given in example [ goldenlennaex ] .
the resulting homeomorphism with @xmath402 and @xmath403 yields a picture of lena with extra large eyes , figure [ goldenlenna ] .
[ goldenlennaex ] let @xmath404 be the family of affine iterated function systems defined by @xmath405 @xmath406 @xmath407 let @xmath408 and , for @xmath409,$ ] let @xmath410 be the mask@xmath411 where @xmath412 let @xmath413 then , by theorem [ goldthm ] , there is a value of @xmath414 @xmath415 such that the condition in theorem [ ctyfractaltransthm ] ( ii ) holds , and if @xmath395 then @xmath416 then is a homeomorphism . the value of @xmath401 is@xmath417 where @xmath418 is the attractor of @xmath419 illustrated in figure [ figure ] .
experimentally we find @xmath420 which is used to compute figure [ goldenlenna]@xmath294 the original lena was overlayed on a black background , as in figure [ combined ] .
the transformed picture also had a black background that has been omitted here .
figure.png here we restrict attention to iterated function systems defined on @xmath421^{2}\subset$ ] @xmath0 .
we are concerned with fractal transformations from @xmath371^{2}$ ] to itself , associated with a pair of iterated function systems of the form@xmath422^{2},f_{1},f_{2}, ... ,f_{n})\text { , } \mathcal{g}=([0,1]^{2},g_{1},g_{2}, ... ,g_{n}).\ ] ] we suppose that @xmath423^{2}$ ] . in applications to digital imaging , @xmath371^{2}$ ]
is discretized . here
we refer to the the locations of pixels as discretized coordinates .
we suppose that discretized versions of @xmath133 and @xmath102 have resolutions @xmath424 and @xmath425 respectively .
let @xmath426 be a projection operator , namely a function with the property @xmath427 for all @xmath160 .
for example , @xmath428^{2}\text{.}\ ] ] if @xmath111 is a homeomorphism between the non - discretized spaces then @xmath429 is also a projection operator because @xmath430 the identity on @xmath431 whence @xmath432 . a trivial but instructive example is provided by choosing @xmath13 and @xmath103 to be the same , with @xmath433 and@xmath434 if @xmath435 then @xmath436 is the filter that corresponds to downsampling followed by doubling the width and height of each pixel .
[ lenaex]let @xmath405 @xmath437 , @xmath438 and @xmath439 the family of ifss @xmath440 has attractor @xmath1 and address structure that is independent of @xmath441 if we set @xmath442 and @xmath443 then @xmath444 is a homeomorphism .
the result of applying the fractal homeomorphism @xmath115 to a digital ( @xmath445 ) picture of lena ( left ) is illustrated in the middle panel of figure [ lena ] . in effect
the middle image is obtained by composing a projection @xmath446 , onto a @xmath445 pixel grid , with @xmath115 .
the image on the right is the result of applying @xmath447 to the middle image .
that is , the right - hand image is the result of applying the projection operator , that we may refer to as a `` fractal filter '' ,
@xmath448 @xmath115 to the original lena .
untitled-1.png suppose we have a collection of masks @xmath449 on @xmath33 such that the conditions of lemma [ l : disjoint ] are holds true for any pair of masks .
let the second iterated function system @xmath314 be such that almost all points of the attractor @xmath319 have a unique address .
thus by lemma [ l : disjoint ] @xmath450 for almost all @xmath78 and @xmath451 .
therefore we can consider the collection of pictures @xmath452 as an almost disjoint fractal decomposition of the picture @xmath329 .
this leads us to the following trick . by means of fractal transformations we map different pictures to different almost disjoint components of @xmath329 . by inverting the transformations we retrieve ( approximations to ) the original pictures , as illustrated in the following example .
[ expack3 ] see figure [ allmasked ] .
let @xmath453 and @xmath454 we define a family of masks for @xmath13 by @xmath455.\ ] ] @xmath456 is the attractor of both systems .
we denote the corresponding masked fractal transformation @xmath457 by @xmath458 for @xmath459 $ ]
. then @xmath460 is injective and invertible on its range , @xmath461 .
moreover , from theorem [ l : disjoint ] it follows that @xmath462 for all @xmath463 , where @xmath464 is lebesgue measure .
let @xmath465 and @xmath466 be two pictures , each supported on @xmath456 . then @xmath467 is a picture supported on @xmath461 and @xmath468 is a picture supported on @xmath469 .
we choose @xmath470 @xmath471 @xmath472lena and @xmath473inverted - lena , where both pictures are @xmath445 . the digitized combined picture @xmath474 , also of resolution @xmath445 ,
is shown in the middle panel of figure [ allmasked ] .
pixels which correspond to points in @xmath461 both @xmath469 are colored white .
the left - hand panel in figure [ allmasked ] illustrates the picture @xmath475 and the right - hand panel shows @xmath476 .
hence we can `` store '' the two pictureswe have that for each distinct choice of @xmath459 $ ] the ranges of the @xmath477 and @xmath478 respectively .
allmasked.png here we describe a different method for encoding several images within a single image .
the method strives to create a single image that is simultaneously `` homeomorphic '' , under different fractal transformations , to several different given images .
an example of such a single `` encoded '' image is shown in the middle panel of figure [ all ] ; two different fractal homeomorphisms applied to it ( but in the digital realm ) yield the images on the right and the left .
the underlying idea is that different invariant measures , belonging to the same iterated function system , but associated with different probabilities , are mutually `` singular continuous '' , and are concentrated on different sets , albeit they have the same support .
the method uses the chaos game algorithm , both to encode and decode .
consider for example the ifs @xmath347;f_{1}(x)=0.5x , f_{2}(x)=0.5x+0.5)$ ] .
if we associate probabilities @xmath479 and @xmath480 with @xmath481 and @xmath482 respectively , then the associated markov operator has invariant probability measure @xmath281 equal to uniform lebesgue measure supported on @xmath371 $ ] , concentrated on the set of points whose binary expansions contain equal proportions of zeros and ones . on the other hand , if we associate probabilities @xmath483 and @xmath484 then the associated markov operator has invariant probability measure @xmath485 that is also supported on @xmath371 $ ] , but singular continuous and concentrated on the set of points whose binary expansions contain nine times as many ones as zeros . hence ,
if the chaos game is applied to @xmath13 using the first set of probabilities , the points of the resulting random orbit will tend to be disjoint from those obtained by of applying the chaos game using the second set of probabilities .
the precise manner in which the orbits of the two systems concentrate is governed by convergence rates associated with the central limit theorem .
the same idea can be applied to `` store '' several images in a single image e : a stored image is retrieved by applying the appropriate fractal transformation to e. all.png we describe the method by means of an example .
[ expack4 ] we use three iterated function systems:@xmath486 then @xmath13 is associated with probabilies @xmath487 @xmath488 similarly the iterated function system @xmath103 is associated with probabilities @xmath489 the goal is to `` store '' two standard digital color images , pepper and lena , each @xmath490 in a single color image @xmath491 also @xmath445 . the image @xmath492 is supported on @xmath456 and associated with two probability measures , @xmath493 and @xmath494 invariant under @xmath495 with probabilities @xmath496 and @xmath497 respectively .
\(1 ) in order to `` encode '' pepper , we run a coupled chaos game algorithm with @xmath13 associated with pepper , supported on a copy of @xmath498 and @xmath495 associated with @xmath491 with ( supposedly ) i.i.d .
probabilities @xmath496 ; that is , we compute a random sequence of points @xmath499 where @xmath500 where @xmath501 ( associated with pepper ) and @xmath502 ( associated with @xmath492 ) and @xmath503 where @xmath504 with probability @xmath505 .
at each step the pixel containing @xmath506 in the ( initially blank ) image @xmath492 is plotted in the color of pepper at the pixel containing @xmath507 . at the end of this process
@xmath492 consists of an `` encoded '' version of pepper .
\(2 ) in order to `` encode '' lena , we again run a coupled chaos game algorithm with @xmath103 associated with lena , supported on a copy of @xmath498 and @xmath495 associated with @xmath492 ( already `` painted '' with an encoding of pepper ) with probabilities @xmath497 ; that is , we compute a sequence of points @xmath508 where @xmath509 where @xmath510 ( associated with lena ) and @xmath502 ( associated with @xmath492 ) and @xmath511 where @xmath504 with probability @xmath512 . at each step
the pixel containing @xmath513 in the image @xmath492 is plotted in the color of lena at the pixel containing @xmath514 .
( the pixel is overwritten by the latest colour value . )
we have used half as many iterations in the encoding of lena as we did for pepper , because a proportion of the points that correspond to pepper are overwritten by points corresponding to lena .
an image @xmath492 that is a realization of steps ( 1 ) and ( 2 ) is shown in the central panel of figure [ all ] .
the image @xmath492 is approximately homeomorphic to both of the images pepper and lena , under the fractal transformations @xmath515 and @xmath516 respectively , that is@xmath517 in practice , to obtain the decoded images , shown on the left and right hand sides of figure [ all ] , we use the chaos game algorithm again , as follows .
\(3 ) in order to `` decode '' pepper , we run a coupled chaos game algorithm , with probabilities @xmath518 with @xmath13 associated with an image ( initially blank ) , supported on a copy of @xmath498 and @xmath495 associated with @xmath492 ( now encoding both pepper and lena ) .
that is , we compute a random sequence of points @xmath519 where @xmath500 where @xmath501 ( associated with pepper ) and @xmath502 ( associated with @xmath492 ) and @xmath503 where @xmath504 with probability @xmath505 . at each step
the pixel containing @xmath507 in the ( initially blank ) copy of @xmath456 is plotted in the color of @xmath492 at the pixel containing @xmath506 .
the result of such a decoding , starting from the encoded @xmath492 illustrated in the middle panel , is shown in the left panel in figure [ all ] .
\(4 ) in order to decode lena , we run a coupled chaos game algorithm , with probabilities @xmath520 with @xmath103 associated with an image ( initially blank but to become the decoded image ) , supported on a copy of @xmath498 and @xmath495 associated with @xmath492 .
that is , we compute a random orbit @xmath508 where @xmath500 where @xmath510 ( associated with pepper ) and @xmath521 ( associated with @xmath492 ) and @xmath522 where @xmath504 with probability @xmath512 . at each step
the pixel containing @xmath514 in the ( initially blank ) copy of @xmath456 is plotted in the color of @xmath492 at the pixel containing @xmath513 .
the result of following this decoding algorithm , starting from the encoded @xmath492 in the middle panel , is shown in the left panel in figure [ all ] .
we thank louisa barnsley for help with the illustrations . | we present a general theory of fractal transformations and show how it leads to new type of method for filtering and transforming digital images .
this work substantially generalizes earlier work on fractal tops .
the approach involves fractal geometry , chaotic dynamics , and an interplay between discrete and continuous representations .
the underlying mathematics is established and applications to digital imaging are described and exemplified .
= 1 iterated function systems , dynamical systems , fractal transformations . 37b10 , 54h20 , 68u10 |
let @xmath0 be a smooth projective toric variety .
the variety @xmath0 can be explicitly written as the symplectic reduction of the hermitian space @xmath2 by a hamiltonian action of a torus @xmath3 , where @xmath4 is the picard number of @xmath0 .
let @xmath5 denote the classes in @xmath6 poincar dual to the toric divisors .
let @xmath7 denote the coordinates in @xmath6 with respect to an integral , nef basis @xmath8 , and let @xmath9 be the exponential coordinates .
recall that the mirror theorem of givental @xcite states that if @xmath10 is semipositive ( nef ) , then the cohomology valued function ( of the b - model ) @xmath11 determines the j - function @xmath12 ( of the the @xmath13-model or gromov - witten theory ) @xmath14 via a change of coordinates @xmath15 , @xmath16 , in such a way that @xmath17 . here
the variables @xmath18 of the b - model are called _ mirror coordinates _ and this change of variables is called _ mirror transformation _ ( or _ mirror map _ ) .
this relation can be used to show that the small quantum cohomology ring @xmath19 differs from the original presentation suggested by batyrev @xcite only by this change of coordinates .
we refer to givental @xcite and the text by cox and katz @xcite for further details on this discussion .
let @xmath20 denote a symplectic manifold .
for a loop @xmath21 in the group of hamiltonian symplectomorphisms on @xmath22 , seidel @xcite , assuming @xmath22 monotone , constructed an invertible element @xmath23 in quantum cohomology counting sections of the associated clutched hamiltonian fibration @xmath24 with fibre @xmath22 .
the _ seidel element _
@xmath23 defines an element in @xmath25 via quantum multiplication , and the association @xmath26 a representation of @xmath27 on @xmath28 .
this construction was later extended for all symplectic manifolds , see for instance mcduff and tolman @xcite where seidel s construction was used to study the underlying symplectic topology in toric manifolds . in the case where the loop @xmath21 is a ( relatively simple ) circle action ,
the asymptotic form of @xmath23 can be written explicitly in terms of geometric and morse theoretic information ( * ? ? ?
* theorem 1.10 ) .
regarding @xmath0 as a hamiltonian space , they consider the seidel element is a variant of the seidel element @xmath29 given by @xmath30 , where @xmath31 is the variable corresponding to the maximal section class of the associated bundle @xmath32 . ]
@xmath33 associated to an action @xmath34 that fixes the toric divisor @xmath35 ( see section [ subsec : action_fixing_divisor ] ) .
it is proved that @xmath33 is a series of the form @xmath36 if @xmath1 is nef , and @xmath37 in the fano case ( @xmath1 is ample ) .
moreover , it is shown that these elements satisfy the following batyrev s relations : @xmath38 where @xmath39 .
this shows that @xmath19 is abstractly isomorphic to the batyrev presentation ( * ? ? ?
* proposition 5.2 ) . based on this
, one can conjecture that seidel elements should be closely related or even equivalent to the mirror transformation . in a recent paper fukaya , oh ,
ohta and ono @xcite have used seidel elements in the mirror symmetry context as well .
we begin with a calculation of seidel elements .
it turns out that @xmath33 appears as a coefficient of the j - function @xmath40 of the associated bundle @xmath32 ( proposition [ p : seidelfromj ] ) , and thus one can use the mirror transformation for @xmath32 ( which is toric as well ) to calculate @xmath33 . let @xmath41 be the matrix of toric divisors such that @xmath42 . as mirror analogues of seidel elements , we introduce the _ batyrev elements _
@xmath43 by @xmath44 these elements satisfy batyrev s relations with @xmath45 there replaced by @xmath46 ( proposition [ p : bat - rel ] ) .
we can calculate them as explicit hypergeometric series in @xmath47 ( lemma [ l : tdj ] ) .
note that the batyrev elements and the seidel elements satisfy the _
same _ relation in _ different _ coordinates .
we see that they only differ by a function multiple .
let @xmath0 be a smooth projective toric variety with @xmath1 nef .
the seidel element associated to the toric divisor @xmath35 is given by @xmath48 the correction term @xmath49 here is an explicit hypergeometric series in @xmath47 .
next we ask the converse : whether the seidel elements determine the mirror transformation .
one can see from the definition that the batyrev elements @xmath50 can determine the jacobian of the mirror map , in particular , the mirror map itself .
therefore , it is important to know when @xmath49 vanishes .
we show that @xmath49 vanishes if and only if @xmath1 is * big * on the toric divisor @xmath35 , i.e. @xmath51 . in terms of the fan of @xmath0 ,
this is also equivalent to the primitive generator @xmath52 of the corresponding 1-cone being a vertex of the fan polytope ( proposition [ p : vanishing ] ) .
another important fact is that the batyrev elements @xmath53 satisfy the same linear relations as the toric divisors do : @xmath54 the seidel elements @xmath33 do not necessarily satisfy the linear relations .
the partial vanishing of @xmath49 and these linear relations are enough to reconstruct the batyrev elements from the seidel elements .
[ t : intro3 ] for a smooth projective toric variety @xmath0 with @xmath1 nef , the seidel elements @xmath33 entirely determine the batyrev elements @xmath53 and in particular the mirror transformation .
more precisely the batyrev elements @xmath55 , @xmath56 are uniquely characterized by the following conditions : 1 .
@xmath57 for some @xmath58 ; 2 .
@xmath59 if @xmath51 ; 3 .
@xmath53 satisfy the linear relations .
we add a remark on mirror symmetry .
the mirror coordinates @xmath18 here represent the complex moduli of the mirror landau - ginzburg model .
there are no preferred choices of coordinates on the complex moduli space and both @xmath60 and @xmath47 serve as local coordinates .
however the @xmath47 coordinates are considered to represent a global algebraic structure of the complex moduli space .
therefore our result suggests that gromov - witten theory itself ( via torus action and seidel elements ) can reconstruct a global algebraization of the khler moduli space which is a priori a formal germ at @xmath61 .
* acknowledgments .
* we thank david cox , dusa mcduff , kaoru ono and chris woodward for useful discussions regarding an early draft of the paper .
we also thank changzheng li and kwokwai chan for helpful comments on an earlier version of the paper .
the authors are grateful for the hospitality at the centre international de rencontres mathmatiques in luminy , where this project was started .
e.g. is supported by nsf grant dms-1104670 and h.i .
is supported by grant - in - aid for young scientists ( b ) 22740042 .
we introduce the main notation and constructions of seidel element and explain its relation to the j - function .
we begin with the definition of the seidel element .
let @xmath0 be a smooth projective variety , equipped with a @xmath62 action .
each @xmath62-orbit in @xmath0 contains a fixed point in its closure , and thus the associated @xmath63-action on @xmath0 is hamiltonian by frankel s theorem @xcite . in this paper
we will restrict to this case , however the construction works more generally in the symplectic category .
the original definition is due to seidel @xcite for monotone symplectic manifolds .
the reader can consult @xcite for a detailed exposition and @xcite for the construction in general case . for relations with the computation of small quantum cohomology rings see @xcite .
[ eq : a - bundle ] the * associated bundle * of the @xmath62-action is the @xmath0-bundle over @xmath64 @xmath65 where @xmath62 acts with the standard diagonal action @xmath66 .
let @xmath67 denote a basis for the rational cohomology @xmath68 . by abuse of notation , every time we omit coefficients we mean rational cohomology
let @xmath69 denote the dual basis , that is @xmath70 where @xmath71 is the usual pairing in cohomology .
there is a ( non - canonical ) splitting @xcite @xmath72 we let @xmath73 denote a basis for @xmath74 , and let @xmath75 denote the dual basis . there is a unique @xmath62-fixed component @xmath76 such that the normal bundle of @xmath77 has only negative @xmath62-weights . when we take a hamiltonian function for the @xmath63-action
, @xmath77 is the maximum set of the hamiltonian .
each fixed point @xmath78 defines a section @xmath79 of @xmath80 .
we denote by @xmath81 the section associated to a fixed point in @xmath77 .
this maximal section defines a splitting @xmath82 \oplus ( h_2(x,{{\ensuremath{\mathbb{z}}}})/\text{tors})\ ] ] let @xmath83 denote the mori cone , that is the cone generated by effective curves and set @xmath84 .
we introduce @xmath85 and @xmath86 similarly .
[ l : split ] @xmath87 + { { \operatorname}{ne}}(x)_{{\ensuremath{\mathbb{z}}}}$ ] . the associated bundle @xmath80 has the @xmath88 action defined by @xmath89 = [ t_1 x , ( z_1 , t_2 z_2)]$ ] . by the @xmath90-action
, every curve can be deformed to a sum of @xmath90-invariant curves in the same homology class .
a @xmath90-invariant curve is either contained in the fibres at @xmath91 , @xmath92 or in the subspace @xmath93 for some fixed component @xmath94 .
therefore it suffices to show that the section class @xmath95 $ ] associated to a fixed point @xmath78 is of the form @xmath96 + d$ ] for some @xmath97 .
take a nontrivial @xmath62-orbit @xmath98 in @xmath0 and consider its closure @xmath99 .
this gives an embedding of the hirzebruch surface @xmath100 into @xmath80 , where @xmath101 is the order of the stabilizer at @xmath98 .
it is well known that the maximal section class in @xmath102 is the sum of the minimal section class and @xmath101 times the fibre class . by joining @xmath103 with the maximal fixed component @xmath77 by a chain of @xmath62-orbits
, we obtain a relation @xmath95= [ \sigma_0 ] + d$ ] with @xmath104 a sum of @xmath62-invariant curves in @xmath0 .
let @xmath105 denote the affine subspace of @xmath106 which consists of section classes , that is the classes that project to the positive generator of @xmath107 .
we set @xmath108 .
the above lemma shows that @xmath109 + { { \operatorname}{ne}}(x)_{{\ensuremath{\mathbb{z}}}}.\ ] ] let @xmath4 be the rank of @xmath6 .
we choose an integral basis @xmath110 of @xmath6 which pairs non - negatively with @xmath111 ( i.e. @xmath112 is nef ) .
there are unique lifts of @xmath113 in @xmath114 which vanish on @xmath96 $ ] .
we denote these lifts by the same symbols @xmath113 . by the above lemma
, these lifts are also nef .
let @xmath115 denote the pull - back of the positive generator of @xmath116 .
then @xmath117 forms an integral basis of @xmath114 .
let @xmath118 denote the novikov variables of @xmath0 dual to the basis @xmath113 .
similarly let @xmath119 denote the novikov variables of @xmath80 dual to @xmath120 .
we take @xmath121 to be the novikov rings of @xmath0 and @xmath80 respectively .
we write @xmath122 the small quantum cohomology ring @xmath123 is defined over the novikov ring @xmath124 .
let @xmath125 ( resp .
@xmath126 ) denote the genus @xmath127 , degree @xmath104 gromov - witten invariant of @xmath0 ( resp .
@xmath80 ) with @xmath101 insertions .
we refer the reader to @xcite and references therein for the definition of algebraic gromov - witten invariants .
since the proof of givental s mirror theorem @xcite is based on algebraic geometry , we will work with algebraic gromov - witten invariants .
[ def : seidel ] the * seidel element * of @xmath0 is the class @xmath128 in @xmath129 . here
@xmath130 denotes the inclusion of a fibre . by equation
, the seidel element can be factorized as @xmath131 with @xmath132 . in general
, one can define the seidel element @xmath23 for a loop @xmath21 in the group @xmath133 of hamiltonian diffeomorphisms and one gets a representation of @xmath134 on @xmath19 via the quantum multiplication by @xmath23 . in our simple situation ,
this fact can be stated as follows .
suppose we have two commuting @xmath62-actions @xmath135 .
let @xmath136 be the composite @xmath62-action .
let @xmath137 , @xmath138 , @xmath139 be the @xmath0-bundle and the seidel element associated to @xmath140 .
the two commuting @xmath62-actions define the associated @xmath0-bundle @xmath141 over @xmath142 such that the restriction to @xmath143 ( resp .
@xmath144 ) is isomorphic to the bundle @xmath145 ( resp .
@xmath146 ) .
then @xmath147 can be obtained as the restriction of @xmath141 to the diagonal in @xmath142 . from this geometry
we have a natural map is isomorphic to the fibre sum @xmath148 . ]
@xmath149 under the multiplication of novikov variables induced by this map , we have @xmath150 by considering the inverse @xmath62-action , we find that the seidel element is invertible in @xmath19 since the trivial @xmath62-action gives rise to the trivial seidel element @xmath151 .
[ d : je ] the ( small ) * j - function * of @xmath80 is the cohomology valued function @xmath152 where @xmath153 is a formal variable , and @xmath154 is the first chern class of the universal cotangent line bundle @xmath155 at the marked point .
the fraction @xmath156 in the correlator should be expanded in the series @xmath157 .
the j - function @xmath12 of @xmath0 is defined similarly ( see equation ) . in order to see the relation with the seidel elements , we expand @xmath158 in terms of powers of @xmath153 as follows .
@xmath159 where the functions @xmath160 are power series with values in @xmath74 .
[ p : seidelfromj ] the seidel element of the action is given by @xmath161 . from definition
[ d : je ] we find @xmath162 using the duality identity @xmath163 we get @xmath164 the conclusion follows from equations and .
we now fix some notation on toric geometry for this paper . for more details see @xcite . for this paper a * toric manifold * @xmath0 is a projective smooth toric variety , as constructed from the following data .
1 . an integral lattice @xmath165 and its dual @xmath166 .
we denote by @xmath167 the natural pairing between @xmath168 and @xmath169 .
2 . a fan @xmath170 in @xmath171 consisting of a collection of strongly convex rational polyhedral cones @xmath172 , which is closed under intersections and taking faces .
we shall assume that the fan @xmath170 is complete and regular .
let @xmath173 denote the set of @xmath174-cones ( rays ) in @xmath170 , and we let @xmath175 denote the set of integral primitive generators of the @xmath174-cones .
the group @xmath168 is the lattice of the torus @xmath176 and thus @xmath169 is the lattice of characters in @xmath176 .
the _ fan sequence _ of @xmath0 is the exact sequence @xmath177 where the second map takes the canonical basis to the primitive generators @xmath178 and @xmath179 is defined to be the kernel of the second map .
this in turn defines a torus @xmath180 , with character and weight lattices @xmath181 respectively and a sequence @xmath182 the dual of the sequence is the _ divisor sequence _
@xmath183 the first arrow takes @xmath184 into the tuple @xmath185 .
the images of the canonical basis under the second map will be denoted by @xmath186 , @xmath187 .
the weights @xmath186 give an homomorphism @xmath188 , and we let the torus @xmath189 act on @xmath2 via this homomorphism .
the combinatorics of the fan defines a stability condition of this action as follows .
let @xmath190 denote the union @xmath191 where @xmath192 is the collection of anti - cones , that is the subsets of indices that do not yield a cone in the fan @xmath193 the toric variety @xmath0 is defined as the quotient @xmath194;\ \quad { \mathcal{u}}:={{\ensuremath{\mathbb{c}}}}^m \setminus z(\sigma).\ ] ] each character @xmath195 defines a line bundle @xmath196 the correspondence @xmath197 yields an identification of the picard group with the character group of @xmath189 .
thus , we have @xmath198 the poincar dual of the prime toric divisor @xmath199 is the image of @xmath186 in @xmath200 . by abuse of notation
, @xmath186 denotes both the divisor @xmath201 itself and its class in @xmath202 .
we note that @xmath203 .
the khler cone @xmath204 of @xmath0 , the cone consisting of khler classes , is given by @xmath205 we assume that @xmath204 is nonempty so that @xmath0 is projective .
we will need later the following notation .
as before @xmath206 denote a nef integral basis , that is an integral basis such that @xmath207 then we write the toric divisors as @xmath208 for some @xmath209 .
the mori cone @xmath83 is the dual of the cone @xmath210 . as before @xmath111 denotes the semi - group @xmath211 .
we shall now explain the symplectic structure of @xmath0 .
take @xmath212 to be maximal compact in @xmath189 .
the @xmath212-action on @xmath2 is generated by the hamiltonian @xmath213 taking a khler class @xmath214 , we have an homeomorphism ( cf .
@xcite ) @xmath215 which induces a symplectic structure ( still denoted by ) @xmath216 on @xmath0 .
this fact and the equivalence of the algebraic and symplectic gromov - witten invariants @xcite yield the following expected result .
the seidel element of the symplectic toric manifold @xmath217 as defined in @xcite coincides with the one in equation . for each divisor @xmath35 we take a @xmath62-action on @xmath0 rotating around @xmath35 and describe the geometry of its associated bundle .
consider the action of @xmath62 on @xmath2 given by @xmath218 and the induced action on @xmath219 .
the toric divisor @xmath220 is the maximal fixed component of this action .
we extend this to the diagonal @xmath62-action on @xmath221 by @xmath222 the associated bundle @xmath32 of the @xmath62-action on @xmath0 is given by @xmath223 therefore @xmath32 is also a toric variety .
we can identify @xmath224 with the lattice of characters of @xmath225 : @xmath226 this is dual to the splitting in equation . in light of this splitting ,
the @xmath227 weights of @xmath225 defining @xmath32 are just given by @xmath228 this in turn yields the divisor sequence @xmath229 the fan of @xmath32 is contained in @xmath230 .
the following generators of the 1-cones @xmath231 yield the fan sequence for @xmath32 @xmath232 which is dual to the divisor sequence above .
we set @xmath233 under the splitting , a nef integral basis @xmath110 of @xmath6 can be lifted to a nef integral basis @xmath117 of @xmath234 .
we have @xmath235 the following result is immediate .
[ lem : ej - nef ] if @xmath1 is nef then for all @xmath236 , @xmath237 is nef .
we now recall givental s mirror theorem ( * ? ? ?
* theorem 0.1 ) .
let @xmath238 be a nef basis of @xmath234 as above .
the i - function of @xmath0 is the @xmath239-valued function : @xmath240 note that all but finitely many factors in the infinite products cancel . here
similarly the i - function of @xmath32 is the @xmath242-valued function @xmath243 where @xmath244 .
[ t : mirror ] let @xmath0 be a toric manifold with @xmath1 nef .
then we have @xmath245 under an invertible change of variables of the form @xmath246 where @xmath247 is a power series in @xmath18 which is homogeneous of degree zero with respect to the degree @xmath248 and @xmath249 .
the coordinates @xmath18 are called the * mirror coordinates * of @xmath0 .
they are asymptotically the same as @xmath118 as @xmath250 , in the sense that @xmath251 . because @xmath252 is nef , we can apply this mirror theorem to @xmath32 .
hence we have @xmath253 under a change of variables @xmath254 [ l : mirrormaps ] the function @xmath255 does not depend on @xmath256 .
moreover we have @xmath257 this means that the mirror maps for @xmath0 and @xmath32 coincide for @xmath118 . the functions @xmath255 appear as the coefficients of @xmath258 in the expansion of @xmath259 ( see @xcite ) : @xmath260 the functions @xmath261 are determined by @xmath262 similarly .
using @xmath263 , we can see that @xmath264 note that the restriction to @xmath265 in the left - hand side is well defined because @xmath266 .
these facts imply that @xmath267 on the other hand , @xmath268 and all other monomials appearing in @xmath255 have nonnegative degree ( since @xmath252 is nef ) .
thus the homogeneous series @xmath255 of degree zero does not depend on @xmath256 .
let @xmath18 be the mirror coordinates of a toric manifold @xmath0 with @xmath1 nef .
set @xmath269={{\ensuremath{\mathbb{q}}}}[y_1^\pm , \dots , y_r^\pm]$ ] .
batyrev s quantum ring is a @xmath270$]-algebra generated by the variables @xmath271 corresponding to the toric divisors @xmath272 subject to the following two types of relations : @xmath273 where @xmath274 .
we refer to these relations as * batyrev relations*. by the divisor sequence , the linear relations can be written in the form : @xmath275 [ re : inv ] since @xmath0 is compact , there exist positive integers @xmath276 such that @xmath277
. then by the fan sequence we have @xmath39 such that @xmath278 .
this gives a relation @xmath279 .
therefore the variables @xmath280 are invertible in the batyrev ring .
[ def : batelem ] let @xmath0 be a toric manifold with @xmath1 nef
. we can regard @xmath281 as corresponding to the logarithmic vector field @xmath282 .
we introduce an element @xmath283 which corresponds to @xmath284 as @xmath285 recall from equation that @xmath42 .
we define the * batyrev element * associated to @xmath35 as @xmath286 [ p : bat - rel ] the batyrev elements @xmath50 satisfy both the multiplicative and linear batyrev relations for @xmath287 .
we use the fact @xcite that if the j - function satisfies the differential equation @xmath288 , then we have a relation @xmath289 in the small quantum cohomology ring .
we can easily show that the i - function of @xmath0 satisfy the differential equation @xmath290 for @xmath39 where @xmath291 and @xmath292 . from the mirror theorem , we know that the j - function satisfies the corresponding differential equation under the mirror change of coordinates .
the multiplicative batyrev relations follow from this and the fact above .
it is obvious that @xmath53 s satisfy the linear relations .
seidel elements satisfy the multiplicative relations in the @xmath60-coordinates , as proved in ( * ? ? ?
* proposition 5.2 ) , but do not necessarily satisfy the linear relations . in this paragraph
we will assume that @xmath1 is nef .
since @xmath237 is also nef , we can expand the i - function of @xmath32 in @xmath258 as follows ( cf . ) .
@xmath293 where @xmath294 is the mirror map of @xmath32 in equation .
the coefficients @xmath295 are power series in @xmath18 taking values in @xmath296 respectively . under the coordinate change @xmath297
, we can rewrite @xmath298 as @xmath299 [ l : seidelmirror ] the seidel element @xmath300 associated to the toric divisor @xmath35 is given by @xmath301 under the mirror transformation for @xmath32 . by proposition [ p : seidelfromj ] , the seidel element @xmath33 is the coefficient of @xmath302 in @xmath303 .
the result follows from @xmath304 . we now digress to reinterpret @xmath305 .
a straightforward computation shows the following lemma .
the i - function of @xmath32 satisfies the differential equation @xmath306 where @xmath41 is the matrix appearing in equation . using the expansion for @xmath259 in equation
we obtain @xmath307 where we used @xmath308 . on the other hand @xmath309 where @xmath53 is the batyrev element of @xmath35 . here
@xmath311 are related by the mirror map for @xmath32 , but we know by lemma [ l : mirrormaps ] that the mirror maps of @xmath0 and @xmath32 coincide for @xmath118 . comparing and we get the following result .
[ l : aux2 ] @xmath312 .
noting @xmath30 and the effect of the mirror map @xmath313 , we obtain the desired expression of seidel elements from lemma [ l : aux2 ] and lemma [ l : seidelmirror ] .
[ th : batsei ] the seidel element @xmath33 and batyrev element @xmath53 are related by @xmath314 under the mirror transformation of @xmath0 .
the theorem above and remark [ re : inv ] show the invertibility of the seidel elements , obtaining the result of seidel in our particular toric setting .
we shall calculate @xmath49 and @xmath53 as explicit hypergeometric series in the mirror coordinates @xmath47 .
it is not hard to see the following lemma .
[ lem : productfactor - asymptotics ] about the product factors appearing in the i - function @xmath259 , we have @xmath315 where h.o.t .
means higher order terms in @xmath258 and @xmath316 [ l : g0 ] the coefficient @xmath49 is given by @xmath317 we want to investigate the coefficient of @xmath258 in the power series expansion of @xmath259 . by lemma [ lem : productfactor - asymptotics ]
, the summand indexed by @xmath318 contributes to the coefficient of @xmath258 if @xmath319 .
since @xmath320 is nef ( lemma [ lem : ej - nef ] ) , this happens only in the following three cases : * @xmath321 and @xmath322 ; * @xmath323 and @xmath324 ; * @xmath321 and @xmath325 . in the first case
, we have @xmath326 for all @xmath327 and so @xmath328 .
this contributes nothing to the coefficient of @xmath258 .
the second case does not happen because in this case @xmath329 has to satisfy @xmath330 except for one @xmath327 , and this implies @xmath328 . in the third case
, @xmath329 has to be a fibre class from @xmath111 ( i.e. @xmath331 ) because @xmath332 and @xmath1 , @xmath333 are nef . therefore the coefficient of @xmath258 in @xmath259 is the sum of @xmath334 over all the fibre classes @xmath335 such that @xmath336 and @xmath337 for exactly one @xmath327 from @xmath338 .
( note that @xmath339 . )
now @xmath49 is the coefficient corresponding to @xmath333 . among the divisors @xmath340
, @xmath341 is the only one which contains @xmath333 .
therefore the terms of the form which contribute to @xmath49 are those with @xmath335 for which @xmath342 , @xmath343 for @xmath344 and @xmath345 .
the batyrev element @xmath53 is given by [ l : tdj ] @xmath346 by the same calculation leading to equation , we find @xmath347 the conclusion follows from a calculation similar to the previous lemma .
let @xmath348 be the second hirzebruch surface .
let @xmath349 be the nef basis of @xmath6 that are poincar dual to the fibre and the infinity section .
the divisor matrix is @xmath350,\ ] ] that is @xmath351 the @xmath60 coordinates are @xmath352 the mirror transformation of @xmath0 is well known ( cf .
@xcite ) ; it is given by @xmath353 the classes @xmath354 , @xmath355 corresponding to the vector fields @xmath284 are @xmath356 the batyrev elements are @xmath357 hence @xmath358 the correction term @xmath49 appears only for @xmath359 . by theorem [ th : batsei ] , the seidel element associated to @xmath360 is just given by @xmath361 where the term @xmath362 is the sum over all effective classes @xmath104 such that @xmath363 , @xmath364 as in equation .
these are just the classes @xmath365 , for @xmath366 .
so we have @xmath367 and @xmath368 .
thus @xmath369 ( this has already appeared in other places e.g. @xcite or @xcite . ) using the formulas above , it is not hard to see that @xmath370 and thus @xmath371 .
all other seidel elements @xmath372 agree with the batyrev elements @xmath373 .
therefore we have , @xmath374 this computation agrees with the one in mcduff - tolman @xcite . it is easy to check that the batyrev relations are compatible in the two coordinate systems @xmath375
we continue to assume that @xmath0 is a toric manifold with @xmath1 nef .
let @xmath170 denote its fan .
if the toric divisor @xmath35 is nef , then a direct computation from equation shows that the correction coefficient @xmath49 is trivial . in such case
the seidel element and batyrev element agree .
however the nef condition is too restrictive .
we show that @xmath49 vanishes if and only if the restriction of @xmath1 to @xmath35 is big , i.e. the image of @xmath35 under the map @xmath376 has the same dimension as @xmath35 for sufficiently big @xmath377 . for each cone
@xmath382 we let @xmath383 denote the minimal face of the fan polytope @xmath380 which contains the collection @xmath384 of primitive generators @xmath385 . by abuse of notation
we write @xmath386 to denote the face @xmath387 .
take a support function @xmath394 such that @xmath380 is contained in the half - space @xmath395 and such that @xmath396 .
we have the following relation @xmath397 by evaluating @xmath398 we have @xmath399 where the second inequality follows from the fact that @xmath400 and @xmath401 , if @xmath402 . because of the zeroes in each hand - side , the inequality is an equality . therefore @xmath403 , if @xmath404 .
this in turn implies that @xmath405 , if @xmath404 .
we first prove the equivalence of ( ii ) and ( iii ) . by (
* lemma 9.3.9 ) , the bigness of @xmath406 translates to the fact that the face @xmath407 of the dual polytope @xmath408 has the maximal dimension @xmath409 , where @xmath410 this is equivalent to @xmath52 being a vertex of @xmath380 .
next we prove the equivalence of ( i ) and ( ii ) .
suppose @xmath52 is not a vertex of @xmath380 .
then @xmath52 is in the relative interior of @xmath411 .
since @xmath411 is convex , there exist @xmath412 on @xmath411 and nonnegative constants @xmath413 such that such that @xmath414 for all @xmath415 and @xmath416 by the fan sequence , the relation gives an element in @xmath417 such that @xmath418 . by equation , @xmath419 .
such @xmath104 contributes to the sum in equation and @xmath420 .
suppose that @xmath52 is a vertex of @xmath380 and @xmath421 .
then by equation there exists @xmath392 such that @xmath422 , @xmath423 and @xmath419 . by lemma [ l : aux4 ] , we know that @xmath424 , if @xmath425 .
however , @xmath52 is a vertex of @xmath380 , which means that @xmath411 contains only @xmath52 .
therefore @xmath424 for all @xmath344 .
since the divisors @xmath426 span @xmath427 it follows that @xmath428 , which contradicts @xmath429 .
let @xmath430 be the dimension of @xmath0 and @xmath4 be the picard number . out of @xmath431 correction terms @xmath49 , at least @xmath432 vanish .
in other words , there are at most @xmath433 non - vanishing correction terms . by proposition [ p : bat - rel ] the elements @xmath53 satisfy both the multiplicative and the linear batyrev relations .
however seidel elements only satisfy the multiplicative relations .
the reconstruction of the mirror transformation is based on this observation .
we now have all we need to establish the proof of theorem [ t : intro3 ] . by all our statements above ,
batyrev elements satisfy the conditions ( i ) ( iii ) .
so we only need to show the uniqueness property .
the linear relations ( iii ) are equivalent to the identity ( cf .
equation ) @xmath434 in the tensor product @xmath435 .
substituting @xmath53 with @xmath436 and using the property ( ii ) , we have @xmath437 here we used the fact that @xmath51 is equivalent to @xmath52 being a vertex of the fan polytope ( proposition [ p : vanishing ] ) .
since @xmath438 , to see the uniqueness of @xmath439 , it suffices to show that @xmath440 is linearly independent .
suppose we have a linear relation @xmath441 if @xmath52 is not a vertex , the face @xmath411 of the fan polytope @xmath380 contains @xmath52 in its relative interior .
since @xmath411 is a convex hull of vertices @xmath442 of @xmath380 , we have a relation @xmath443 for some rational numbers @xmath444 .
this relation gives an element @xmath389 such that @xmath445 , @xmath446 and @xmath447 when @xmath448 is none of @xmath449 .
contracting equation with @xmath450 , we get @xmath451 thus @xmath452 for all @xmath236 for which @xmath52 is not a vertex .
this completes the proof .
we revisit the example in section [ subsec : batseiex ] .
the fan of @xmath453 is given by the following primitive vectors of 1-cones : @xmath454 here @xmath455 are vertices of the fan polytope and thus @xmath456 for @xmath457 .
the mirror coordinates @xmath458 can be reconstructed from these seidel elements , via the formulas @xmath459 let @xmath461 be the @xmath64-bundle over @xmath462 .
collapsing the zero and the infinity section to @xmath64 yields @xmath463 , where @xmath464 acts on @xmath465 by @xmath466\mapsto [ z_1:z_2:-z_3:-z_4].\ ] ] the toric variety @xmath463 has transversal @xmath467 singularities along two @xmath64 s and @xmath0 is its crepant resolution .
the fan of @xmath0 is given by @xmath468 the vertices of the fan polytope are @xmath469 .
the mori cone is spanned by the three homology classes @xmath470 with the intersection matrix @xmath471 where each row vector gives a relation amongst @xmath472 .
the classes @xmath473 define a @xmath474-basis for @xmath475 .
let @xmath476 denote the dual basis and let @xmath477 denote the corresponding novikov variables .
* mirror coordinates . *
@xmath478 * the batyrev and seidel elements . *
@xmath479 * reconstruction from seidel elements . *
the reconstruction in this case becomes a little more involved .
since @xmath480 have no correction terms , we have the relations @xmath481 but these are not enough to determine the mirror coordinates @xmath482 .
we also need to require that the mirror map is homogeneous , i.e. the euler vector field is preserved @xmath483 the equations , can reconstruct the mirror coordinates @xmath482 .
[ [ crepant - resolution - of - ensuremathmathbbp1times - ensuremathmathbbp1ensuremathmathbbz_2 ] ] crepant resolution of @xmath485 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ let @xmath464 act on @xmath462 by @xmath486 , [ w_1,w_2 ] ) \mapsto ( [ z_1,-z_2 ] , [ w_1,-w_2]).\ ] ] the quotient @xmath485 has four isolated singular points of type @xmath467 .
let @xmath0 denote the minimal resolution of @xmath485 .
it is given by the complete regular fan ( figure ) spanned by @xmath487 the vertices of the fan polytope are @xmath488 .
define @xmath489 to be the @xmath490-basis of @xmath491 with the following intersection matrix : @xmath492 again each row vector gives a relation of @xmath493 . in this case
the mori cone @xmath494 is not simplicial , but is contained in the cone spanned by @xmath495 .
let @xmath496 be the dual basis of @xmath495 and let @xmath497 be the corresponding novikov variables .
note that @xmath498 and @xmath499 .
* mirror coordinates .
* @xmath500 from this we get the elements @xmath501 * the batyrev and seidel elements . *
@xmath502 * reconstruction from the seidel elements . * in this case , the divisors @xmath503 and @xmath484 do not span @xmath6 and the method in the previous example does no apply .
assume that we know the seidel elements @xmath504 given above .
we will reconstruct the batyrev elements from these .
we set @xmath505 for @xmath506 and @xmath507 for @xmath508 for some @xmath509 .
the linear relation @xmath510 gives the equation @xmath511 from the computation above , we have @xmath512 since @xmath513 and @xmath514 are linearly independent , it follows that @xmath515 , @xmath516 .
similarly one can solve for @xmath517 .
consider the @xmath464 action on the weighted projective space @xmath519 given by @xmath520 \mapsto [ -z_1,z_2 , -z_3].\ ] ] the quotient @xmath518 has three singular points , two of which are type @xmath467 and the other is of type @xmath521 .
let @xmath0 be the minimal resolution of @xmath518 .
the fan of @xmath0 is given by ( see figure [ fig : count ] ) @xmath522 the picard number of @xmath0 is 6 .
the vectors @xmath523 are vertices of the fan polytope .
let @xmath524 be the basis of @xmath525 such that @xmath526 the mori cone @xmath494 is not simplicial but is contained in the simplicial cone generated by @xmath527 .
( here @xmath528 for @xmath529 , but @xmath530 .
note also that @xmath527 do not form an integral basis . ) set the novikov variable @xmath531 .
let @xmath532 be the dual basis of @xmath533 .
* mirror coordinates . *
the mirror coordinates for surface @xmath534 singularity resolutions are calculated in ( * ? ? ?
* appendix a , proposition a.6 ) .
the same method applies here , as the mirror map for the @xmath521 singularity resolution appears as part of the mirror map in this case .
we only state the final result here .
@xmath538 @xmath539 @xmath540 @xmath541 @xmath542 @xmath543 @xmath544 * reconstruction of batyrev from seidel . *
again the divisors @xmath545 and @xmath484 do not span @xmath6 in this case .
we have the following linear relations : @xmath546 suppose we know only the seidel elements @xmath547 given above .
we can check ( assisted by computer ) that there exist unique functions @xmath548 of @xmath549 s which solves the linear equations : @xmath550 here @xmath551 are given by ( as expected ) @xmath552 the batyrev elements are given as : @xmath553 for @xmath554 and @xmath555 , @xmath556 , @xmath557 , @xmath558 , @xmath559 .
then the batyrev elements determine the mirror co - ordinates @xmath560 .
consider the toric variety @xmath0 given by the fan spanned by the following vectors ( see figure [ fig : cubic ] ) : @xmath562 only @xmath563 are the vertices of the fan polytope
. a cubic surface in @xmath465 can degenerate to the singular toric variety @xmath561 ( with 3 nodes of type @xmath564 ) and @xmath0 is a minimal resolution of @xmath561 where the @xmath565-action on @xmath566 is given by @xmath567 \mapsto [ z_1,\omega z_2 , \omega^2 z_3]$ ] .
we can construct @xmath0 as a 6 times blowup of @xmath566 at infinitely near points .
thus @xmath0 is deformation equivalent to a del pezzo surface of degree 3 ( i.e. cubic surface ) .
we take a basis @xmath568 of @xmath525 such that the intersection matrix becomes @xmath569 the mori cone @xmath494 is contained in the simplicial cone generated by @xmath570 s .
note that @xmath571 are not a @xmath474-basis and that @xmath494 itself is not simplicial .
we have @xmath528 , @xmath572 and @xmath573 . as usual , we take @xmath574 to be the dual basis of @xmath568 and @xmath531 to be the novikov variable .
* reconstruction of batyrev from seidel .
* we have the following linear relations : @xmath586 suppose we only know the seidel elements @xmath587
. we can check that the following linear equation for @xmath588 has a unique solution : @xmath589 the batyrev elements are given as @xmath590 , @xmath591 , @xmath592 , @xmath593 , @xmath594 , @xmath595 and @xmath553 for @xmath596 .
the mirror co - ordinates @xmath597 are determined by these .
it is interesting to compare the givental - hori - vafa mirrors of @xmath0 and a cubic surface .
the mirror of @xmath0 is a landau - ginzburg model defined by the function @xmath598 on the torus @xmath599 with coordinates @xmath600 @xmath601 where the coefficients @xmath602 are determined by the relation @xmath603 on the other hand , the mirror of a ( generic ) cubic surface @xmath22 is @xcite @xmath604 under the specialization @xmath605 and @xmath606 , we have @xmath607 * a givental * , _ a mirror theorem for toric complete intersections _ , from : `` topological field theory , primitive forms and related topics ( kyoto , 1996 ) '' , progr . math . 160 , birkhuser boston , boston , ma ( 1998 ) 141175 | the goal of this article is to give a precise relation between the mirror symmetry transformation of givental and the seidel elements for a smooth projective toric variety @xmath0 with @xmath1 nef .
we show that the seidel elements entirely determine the mirror transformation and mirror coordinates . |
borides , carbides , and nitrides were among the early compound superconductors discovered in the first half of the previous century@xcite ( see , e.g. , ref .
for a recent review ) . however , it was only in 2001 , with the discovery of superconductivity in mgb@xmath1 at 39k,@xcite that researchers intensified the search for superconductivity in other diborides .
based on a large number of studies , mgb@xmath1 was identified as a two - band two - gap superconductor .
its peculiar fermi surface exhibits two - dimensional hole - like cylinders from the @xmath7 bands , a hole - like tubular network due to the bonding @xmath8 bands , and an electron - like tubular network due to the antibonding @xmath8 bands.@xcite due to this electronic configuration and to a distinct anisotropy of the electron - phonon interaction strength , the electronic excitation spectrum of mgb@xmath1 adopts two gaps in the superconducting phase : a large gap of 7.2mev in the @xmath7 ( @xmath9 ) bands , and a small gap of 2.8mev in the bonding and antibonding @xmath8 ( @xmath10 ) bands . at the same time , an upper critical field anisotropy has been observed , with @xmath11 t and @xmath12 t at zero temperature.@xcite besides alkaline - earth metals ( such as mg ) , diborides of other elements have been proposed to be checked for superconductivity . because of
the presence of partially filled 3@xmath13 , 4@xmath13 , and 5@xmath13 orbitals , considered as promising for superconductivity , these new attempts involved icosagens ( al ) and various @xmath13-type transition metals ( t ) .
the latter ( tb@xmath1 ) , which are claimed to combine average coupling constants with comparable phonon frequencies to mgb@xmath1@xcite ( due to the presence of light boron atoms ) , were the natural candidates in this search .
unfortunately , these renewed efforts proved unsuccessful and to date there are no reports of superconductivity for the majority of tb@xmath1 materials .
zrb@xmath1 and hfb@xmath1 are two such non - superconducting refractory materials with melting points of @xmath14k , behaving essentially as pauli paramagnets down to low temperatures .
the electrons in the 4@xmath13 shell of zirconium ( zr ) and those in the 5@xmath13 shell of hafnium ( hf ) are less localized than those of the 3@xmath13 row . in a recent study,@xcite it was found that by replacing small amounts of zr or hf with v , the resulting compounds zr@xmath15v@xmath16b@xmath1 and hf@xmath15v@xmath16b@xmath1 are superconductors .
maximum superconducting temperatures @xmath17k and 7.31k were reached in zr@xmath3v@xmath4b@xmath1 and hf@xmath5v@xmath6b@xmath1 , respectively , at the upper solubility limit of v ( @xmath18 ) .
x - ray powder diffraction ( xrd ) patterns@xcite indicate that an increase in v doping does not change the in - plane lattice parameter @xmath19 , while it reduces the inter - layer distance @xmath20 . at the same time
, it has been shown that the structural and electronic properties of these compounds are influenced by the presence of b vacancies.@xcite in ref .
, the properties of superconducting zr@xmath3v@xmath4b@xmath1 were investigated by means of macroscopic techniques.@xcite in this work we aimed at combining macro- and microscopic techniques ( including magnetometry , nmr , resistivity , and preliminary @xmath2sr experiments ) on zr - based diborides and extend our study to include the hf - based compound . since we succeeded in synthesizing samples with less magnetic impurities ( below 10 ppm ) with respect to the previous ones,@xcite whenever relevant
, a comparison between the two batches is included .
our extensive data sets allowed us to unravel clear analogies and differences between the t - diborides zr@xmath3v@xmath4b@xmath1 and hf@xmath5v@xmath6b@xmath1 and the well - known mgb@xmath1 . in sec .
[ ssec : nmr ] we show that spin - lattice relaxation processes in t - diborides are two orders of magnitude slower that in mgb@xmath1 , indicating significantly different electronic structures . combined magnetometry ( sec .
[ ssec : squid ] ) and resistivity ( sec .
[ ssec : res ] ) measurements were performed to evaluate the upper and lower critical fields , respectively .
these compounds prove to be extreme type - ii superconductors , as reflected by the high values ( @xmath21 100 ) of the ginzburg - landau @xmath22 parameter .
the london penetration depth @xmath23 was evaluated through @xmath2sr ( sec .
[ ssec : musr ] ) experiments and the coherence length @xmath24 via upper critical field measurements ( sec .
[ ssec : res ] ) .
thanks to the efficient complementarity of these four techniques , we argue that , besides the qualitatively different fermi surfaces of mgb@xmath1 and transition - metal diborides , in both cases we are dealing with @xmath0-wave superconductors . while the rather high @xmath25 of mgb@xmath1 is understood as a consequence of a favorable electronic structure and electron - phonon interaction , the drastic effect of v - doping in the t - diborides is still rather surprising .
alb@xmath1 structure of the transition - metal diborides highlighting the graphene - like b layers ( magenta atoms ) and the hexagonal metal layers ( blue atoms).,scaledwidth=33.0% ]
both pure and v - doped compounds crystallize in the layered alb@xmath1 structure with a @xmath26 hexagonal space group , where the zr ( or hf ) atoms and the b atoms occupy , respectively , the @xmath27 ( 0,0,0 ) and @xmath28 ( 1/3 , 2/3 , 1/2 ) positions . as in mgb@xmath1 ,
their crystal lattices are characterized by hexagonal metal layers alternating with graphite - like b layers ( see fig .
[ fig : zrb2_structure ] ) .
previous results of band - structure calculations@xcite and fermi surface ( fs ) representations@xcite are available in the literature for both zrb@xmath1 and hfb@xmath1 . since , however , the printed version of the published fs of zrb@xmath1 suffers from low quality , we chose to present the result of our own calculation in fig .
[ fig : fs_zb ] , intended to serve for a comparison with the published fs of mg@xmath29,@xcite well known for its amazingly high critical temperature @xmath25 of the order of 40k for the onset of superconductivity .
our density - functional calculations use an lda approach@xcite based on the dmol@xmath30 band - structure modeling program.@xcite the self - consistent field ( scf ) was accomplished using the standard dnp@xcite variational basis set and a @xmath31-centered @xmath32 @xmath33-mesh in the reciprocal cell . pseudized scalar relativistic corrections@xcite were applied .
the calculations are based on the experimental geometries published in ref . .
the fs plot was generated using the xcrysden program@xcite based on the dmol@xmath30 output for a @xmath34 @xmath33-mesh .
the zrb@xmath1 fs consists of four short cylindrical hole - type pockets around the @xmath35 point and four tripod - shaped electron - type pockets , each consisting of a triangular ring around the @xmath36 point and three elliptic extensions near the @xmath37 plane .
since the fs of hfb@xmath1 exhibits essentially the same features , we refrained from presenting it in a separate figure .
these features are really quite different from those of the fs of mgb@xmath1 .
most obvious is the reduction of the two almost two - dimensional hole sheets centered near the @xmath35 point for mgb@xmath1 to one much smaller and 3-d type pocket for zr- and hf - diboride . likewise
the 3-d parts are quite different in shape for mgb@xmath1 .
it is therefore not surprising that the two borides investigated here are less favorably conditioned for superconductivity than mgb@xmath1 and indeed for both pure compounds , no superconductivity was detected above 1k .
the band - structure calculations for the t - borides considered here indicate that in both cases the density of electronic states at the fermi energy @xmath38 is located in a region where @xmath39 exhibits a pseudo - gap , i.e. , is much reduced .
it is thus remarkable that a very small ( 45% ) v - for - zr or v - for - hf substitution results in onsets of superconductivity up to approximately 8k .
more details on the crystal structure and defect - induced phase instabilities of these types of compounds are discussed in ref . .
fermi surface of zrb@xmath1.,scaledwidth=35.0% ]
polycrystals of zrb@xmath1 , hfb@xmath1 , zr@xmath3v@xmath4b@xmath1 , and hf@xmath5v@xmath6b@xmath1 were synthesized via boron carbide reduction and structurally characterized as described in ref . . for our magnetic measurements , the samples in form of fine powders , with masses between 50 and 100 mg , were sealed in teflon ( ptfe ) tubes .
the nmr investigations including line - shape and spin - lattice relaxation measurements were performed in an applied magnetic field of 3.505 t , since higher fields would reduce the possibility to resolve the quadrupolar effects . in our case ,
the @xmath40b nucleus ( spin @xmath41 ) proved to be the most suitable one , since it allows a direct comparison between the four samples and it is four times more abundant than @xmath42b .
the nmr signals were monitored by means of standard spin - echo sequences , consisting in @xmath43 and @xmath10 pulses of 2 and 4@xmath44s , respectively , with recycle delays ranging from 10 to 100s , depending on the temperature ranging between 4 and 295k .
the nmr line - shapes were obtained via the fast fourier transform ( fft ) of the echo signal which , due to the high sensitivity of @xmath40b , could be acquired using relatively few scans ( from 4 to 2048 ) .
the spin - lattice relaxation times @xmath45 were measured with the inversion recovery method , using a @xmath10-@xmath43-@xmath10 pulse sequence .
the magnetometry measurements were performed by using a commercial magnetic property measurement system ( mpms xl ) from quantum design , equipped with a 7-t magnet and covering the temperature range from 2 to 400k . for the resistivity measurements ,
the samples were densely packed in cylindrical pellets with a diameter of 1.4 mm and a thickness of 5 mm , produced by applying high external pressures .
the electrical contacts with a four - probe configuration were made by means of an electrically - conducting silver epoxy .
preliminary @xmath2sr measurements were made at the gps spectrometer of the s@xmath44s facility of paul scherrer institute ( psi ) in villigen , switzerland .
the available sample mass ( 300 mg ) was sufficient to stop the 4-mev muons without additional degraders and with a minimal background signal . to avoid pinning effects , well known for distorting the vortex lattice ( vl ) in the superconducting phase of mgb@xmath1,@xcite the transverse - field ( tf ) muon - spin rotation measurements
were made at the highest field available ( 0.6 t ) .
the @xmath46b nmr lines of all the samples ( both pure and v - doped ) were measured from 5 to 295k ; typical data are shown in fig .
[ fig : diborides_nmr_lines_comparison ] . the reference @xmath46b nmr frequency in an applied magnetic field of 3.505 t was evaluated to be @xmath47mhz . in our case ,
the nmr lines exhibit peaks which are very close to the reference , with an absolute positive shift of only about 6khz , corresponding to a knight shift of 120ppm . in the covered temperature range between 5 and 295k
the @xmath46b nmr lines practically coincide ( see fig .
[ fig : diborides_nmr_lines_comparison ] ) , implying temperature - independent knight shifts for all the measured samples , compatible with the @xmath48 plateaux observed in the magnetometry data , measured under zero - field cooled conditions ( see fig . [
fig : chi_vs_t_10oe_allsamples ] ) .
the trend of the knight shift in the superconducting phase could not be resolved because of the appreciable width of the resonance signal .
as the line position , also the full width at half maximum ( fwhm ) is practically constant upon varying the temperature . for the zr - based samples its value is 13khz ( with 1khz of additional broadening below @xmath25 ) , while for those containing hf the width is 14khz ( + 1khz at low temperatures ) .
the typical quadrupole splitting @xmath49 in zrb@xmath1 and hfb@xmath1 is approximately the same , i.e. 54khz .
this relatively small value implies a rather small electric - field gradient ( efg ) , especially if compared with mgb@xmath1 , for which @xmath50 khz,@xcite which is a signature of a different electronic charge distribution , confirmed by the different orbitals involved in the bonds , i.e. only @xmath0 and @xmath51 orbitals for mgb@xmath1 and also @xmath13 orbitals for t - borides .
@xmath46b nmr lines in the four tb@xmath1 compounds , respectively - from bottom to top , zrb@xmath1 , zr@xmath3v@xmath4b@xmath1 , hfb@xmath1 , and hf@xmath5v@xmath6b@xmath1 . the measurements were performed at 3.505 t and for each sample the lines measured at 5 and at 295k are superposed , showing no relevant change .
the dashed black line superposed on the zrb@xmath1 spectra represents a simulation of a s = @xmath52 powder pattern with small quadrupole splitting ( reflected by the two shoulders around the central transition).,scaledwidth=45.0% ] for typical powder spectra of a @xmath53 nucleus with a small quadrupole splitting , an analytical expression for the line - shape can be derived by considering the quadrupole term as a first - order perturbation in the main zeeman hamiltonian.@xcite as shown in fig .
[ fig : diborides_nmr_lines_comparison ] , the experimental spectrum and the simulated line - shape ( dashed line ) for zrb@xmath1 agree quite well .
the simulation of the powder spectrum , employing a matlab code , performs the integration according to euler s method over all the possible orientations of the nmr line - shape factor@xcite @xmath54 , i.e. , the integration of the transition frequencies , derived from the quadrupolar theory and weighted using a gaussian broadening function.@xcite from a quantitative analysis , the electric - field gradient ( efg ) tensor can be evaluated using magnetic susceptibility data [ @xmath48 in si units ] at 1mt for zrb@xmath1 , hfb@xmath1 , zr@xmath3v@xmath4b@xmath1 , and hf@xmath5v@xmath6b@xmath1 .
the undoped samples ( and the doped ones above @xmath25 ) exhibit diamagnetic behavior ( plateaux ) .
small zrb@xmath55 impurities in zrb@xmath1 induce a superconducting transition at @xmath56k .
zr@xmath3v@xmath4b@xmath1 and hf@xmath5v@xmath6b@xmath1 become superconductors at @xmath57k and 7.31k , respectively.,scaledwidth=45.0% ] @xmath58 with @xmath59 the largest efg component ( parallel to the applied magnetic field ) , @xmath60 the nuclear spin , and @xmath61 the quadrupole moment of the nucleus .
by considering the hexagonal symmetry of the alb@xmath1 structure , we can assume that the in - plane anisotropy parameter @xmath62 and , therefore , @xmath63 . since the @xmath64 tensor is traceless , by evaluating @xmath65 from the simulated @xmath49 data ( see eq . [ eq : efg ] ) , we get @xmath66vm@xmath67 and @xmath68vm@xmath67 .
@xmath46b nmr @xmath69 data at 3.505 t for all the investigated samples , compared with mgb@xmath1 and alb@xmath1 data ( from ref . ) . from a standard fit , assuming only one spin - lattice relaxation time,@xcite we obtain a linear trend , with korringa constants as reported in table [ tab : summary ] .
materials with transition metal cations , such as zr and hf , exhibit slow spin - lattice relaxation processes , two orders of magnitude slower than in mgb@xmath1 , whereas alb@xmath1 is an intermediate case.,scaledwidth=45.0% ] as summarized in table [ tab : summary ] , the knight shift values are of the same order of magnitude as in mgb@xmath1,@xcite but we note a difference of two orders of magnitude ( ! ) in the spin - lattice relaxation rates .
the measured @xmath70 value for mgb@xmath1 is @xmath71sk which , by considering its 70ppm knight shift , implies an experimental korringa constant @xmath72sk , approximately 3 times smaller than the theoretical value @xmath73sk . upon v - doping ,
the very slow relaxation processes in t - diborides are non significantly altered but , nevertheless , this apparently insignificant doping induces superconductivity at relatively high critical temperatures .
furthermore , due to strong covalent bonds , the largest contribution to the electronic density of states ( dos ) at the fermi level @xmath74 is due to itinerant electrons in the boron layers . with respect to alb@xmath1 and even more so to mgb@xmath1 , @xmath74 of our compounds is drastically reduced , as clearly confirmed by the spin - lattice relaxation data , in turn in good agreement with our theoretical calculations on the electronic structure of zrb@xmath1 and hfb@xmath1 .
.[tab : summary ] key nmr parameters for the investigated samples compared with those of alb@xmath1 and mgb@xmath1 . [ cols="<,^,^,^",options="header " , ] the raw data , i.e. , @xmath46b nmr @xmath45 inverse saturation recovery curves , for the temperatures above @xmath25 were fitted by assuming the standard magnetization recovery formula for a single spin - lattice relaxation time.@xcite however , due to the expected high anisotropy of @xmath75 , and by analogy with mgb@xmath1,@xcite in powder samples we expect grains with different orientations , i.e. , where the applied magnetic field lies in the @xmath76 plane or is parallel to the @xmath20 axis . _ a priori _ the orthogonal and parallel magnetic susceptibilities and the upper critical field depend on the direction of the field .
the existence of nonequivalent grains , due to the anisotropy of the upper critical field , is also confirmed by magnetometry ( fig .
[ fig : chi_vs_t_10oe_allsamples ] ) and resistivity ( fig .
[ fig : diborides_r_vs_t_all ] ) measurements , which show a superconducting width transition @xmath77k , hence suggesting the persistence of inhomogenous domains . in a first approximation and following a procedure employed in the mgb@xmath1 case,@xcite , we can fit the data , by assuming two relaxation processes ( see eq .
[ eq : irfit ] ) , related to the normal phase and to the superconducting phase , respectively : @xmath78 here @xmath79 is the time delay in the nmr pulse sequence , @xmath80 and @xmath81 the spin - lattice relaxation times of the superconducting and normal grains , respectively , @xmath82 the superconducting volume fraction and @xmath83 the stretching parameter ( close to 1 in this case ) .
we assume that @xmath82 is a temperature - dependent fit parameter , ranging from @xmath84 ( in case of maximum superconducting fraction , as evaluated from magnetometry data ) , down to 0 . in the transition region
the two plateaux are connected with a sigmoidal function .
the @xmath69 spin - lattice relaxation data above @xmath25 follow the linear behavior in @xmath85 of a simple metal ( see fig . [
fig : diborides_t1_compar ] ) , where nuclear relaxation occurs mostly via interactions with the conduction electrons . on the other hand , below @xmath25
, the superconducting grains may exhibit two trends : a power - law with an integer exponent , typical of anisotropic superconductors , or an exponential trend , as expected for @xmath0-wave superconductors .
if the sample is not perfectly homogeneous , a second relaxation component could persist as a linear trend associated to normal grains , as shown in fig .
[ fig : diborides_gap ] .
the hypothesis of the two relaxation times is justified by the good fit results that we obtain for the evaluation of spin - lattice relaxation rates @xmath86 .
a quantitative analysis is however hampered by a contribution to the relaxation from the flux vortex lines ( i.e. , their thermal motion).@xcite since we performed a field cooling ( fc ) measurement , the formation of a flux line lattice ( fll ) with a regular arrangement of vortices , most likely with hexagonal symmetry , is expected .
the geometric parameter of this lattice is the intervortex spacing @xmath87 , which implies @xmath88 nm in our case .
this value is approximately 6 times the diameter of the vortices @xmath24 , as evaluated in sec .
[ ssec : res ] , implying that the measured t@xmath89 values consist of the sum of a slow contribution from outside the vortex cores and a faster contribution from the normal region within the vortices.@xcite we speculate that the considerable anisotropy of the upper critical field of mgb@xmath1,@xcite is also a characteristic of our materials
. a rigorous confirmation would be obtained by relevant experiments on single crystals as , e.g. , thermal conductivity measurements . in general , @xmath0-wave superconductors exhibit an exponential decrease of @xmath69 well below @xmath25 , from which the gap value @xmath90 can be extracted .
the appearance of a hebel - slichter coherence peak is usually interpreted as confirming the cooper - pairing with spherical symmetry . in our case
, the absence of a coherence peak does not rule out an @xmath0-wave parity , since the size of the peak can be significantly reduced by the pair - breaking mechanism in the presence of high fields.@xcite due to the above - mentioned complexity of the relaxation processes and the quality of our data , it is impossible to extract the gap value @xmath90 from the exponential decrease of the spin - lattice relaxation time . in any case a first evidence for @xmath0-wave superconductivity is the increasing deviation from a power - law behavior with an exponent 3 towards lower temperatures ( see fig .
[ fig : diborides_gap ] ) . to justify the similarities between the phonon - mediated @xmath0-wave superconductivity mechanism in mgb@xmath1 and our v - doped samples we note that vanadium , given its @xmath91 orbital ,
has one more electron , if compared to zr ( @xmath92 ) and hf ( @xmath93 ) .
an analogous electron doping is confirmed in the case of mgb@xmath1 , where the @xmath0-states of mg are pushed up by the b @xmath8 orbitals and , therefore , fully donate their electrons to the boron - derived conduction band.@xcite this doping mechanism occurs also in the opposite direction ( reduction of the @xmath25 value ) in mgb@xmath1 . in this case , both the substitution of mg with li ( hole doping ) and of boron with carbon or al ( electron doping ) reduce the @xmath25 of the material.@xcite in this case , it is claimed that the electrons fill the @xmath9 band and holes occupy the @xmath10 band , therefore making charge compensation impossible .
furthermore , a recent paper@xcite supports the hypothesis of two gaps in zr@xmath3v@xmath4b@xmath1 from critical current density @xmath94 measurements in different fields .
in fact , @xmath94 can suitably be fitted by the sum of two contributions @xmath95 and @xmath96 , respectively , related to the first and the second gap , following an exponential trend as a function of the applied magnetic field .
@xmath46b nmr @xmath97 data at 3.505 t for zr@xmath3v@xmath4b@xmath1 ( main plot ) and hf@xmath5v@xmath6b@xmath1 ( inset ) , representing the two relaxation times of eq .
[ eq : irfit ] , which result from fitting the inversion recovery curves .
both superconducting samples exhibit a conventional metallic trend ( red points ) and a superconducting dropdown ( magenta points ) ; the blue lines are power laws with critical exponent 3 , that would be a signature of @xmath13-wave superconductivity . in both the samples we argue that the faster relaxation rates below 5.5k with respect to the power - law support the hypothesis of @xmath0-wave pairing.,scaledwidth=45.0% ] the temperature dependence of @xmath98 exhibits a negative curvature .
by using parabolic fits ( dashed lines ) the approximate @xmath99 values are obtained : 2.5 mt for zr@xmath3v@xmath4b@xmath1 and 3.6 mt for hf@xmath5v@xmath6b@xmath1 .
inset : the @xmath100 vs. @xmath101 plots ( zero field cooling ) in zr@xmath3v@xmath4b@xmath1 at 3 , 5 , 9 , and 12k , scaledwidth=45.0% ] squid magnetometry measurements were made on all the samples ( zrb@xmath1 , hfb@xmath1 , zr@xmath3v@xmath4b@xmath1 , and hf@xmath5v@xmath6b@xmath1 ) . the high sensitivity ( 10@xmath102am@xmath103 ) of the reciprocating sample option ( rso ) of the mpms magnetometer allowed us to detect small impurities .
in particular , in the zrb@xmath1 and zr@xmath3v@xmath4b@xmath1 samples a small mass fraction of about 0.5% exhibits a superconducting transition at 5.5 k , the typical @xmath25 of zrb@xmath55 impurities .
smaller impurity contributions are also visible from the @xmath48 plots at fields between 0.1 and 7 t , exhibiting a steady increase below 20k . therefore , an accurate measurements of @xmath104 was achieved by applying a small magnetic field of 1mt .
the obtained values of @xmath25 are 8.33k for zr@xmath3v@xmath4b@xmath1 and 7.31k for hf@xmath5v@xmath6b@xmath1 . as shown in the inset of fig .
[ fig : hc1_trend ] , the typical type - ii sc cycles can be observed below @xmath25 ; above @xmath25 , a clear diamagnetic trend is confirmed with a typical @xmath105m@xmath30mol@xmath106 , a value approximately 5 times smaller than bismuth .
the low - field region ( from 0.2 to 6mt ) , exhibits an initial linear trend in @xmath107 .
it is possible to extract an approximate value of @xmath98 , defining it as the field at which the deviation from the linear trend ( called _ meissner line _ ) exceeds the sensitivity of the instrument . performing this analysis , for each of the @xmath100 vs. @xmath101 curves , we obtain @xmath108 , as shown in fig .
[ fig : hc1_trend ] for both zr@xmath3v@xmath4b@xmath1 and hf@xmath5v@xmath6b@xmath1 .
according to the two - band ginzburg - landau theory applied to mgb@xmath1 , the lower critical field exhibits a change in concavity ( from negative to positive , upon cooling ) at @xmath109 0.5.@xcite since this change is scarcely distinct and the model depends upon the interband mixing of the two order parameters and of their gradients , it is difficult to interpret the @xmath108 trend which , as reported in the literature for mgb@xmath1,@xcite can be fitted even with a line . on the other hand , the previously reported zr@xmath3v@xmath4b@xmath1 data,@xcite show a pronounced upturn , which is not present in our case .
the reason for this discrepancy is unclear at the moment .
systematic resistivity measurements were performed in zero field on all the samples ( zrb@xmath1 , hfb@xmath1 , zr@xmath3v@xmath4b@xmath1 , and hf@xmath5v@xmath6b@xmath1 ) , as shown in fig .
[ fig : diborides_r_vs_t_all ] , and on the superconducting samples zr@xmath3v@xmath4b@xmath1 ( see the inset of fig .
[ fig : diborides_r_vs_t_all ] ) and hf@xmath5v@xmath6b@xmath1 in magnetic fields up to 7 t .
@xmath110 vs. @xmath85 data for all the samples ( zrb@xmath1 , hfb@xmath1 , zr@xmath3v@xmath4b@xmath1 , and hf@xmath5v@xmath6b@xmath1 ) in zero field .
the superconducting dropdown in zr@xmath3v@xmath4b@xmath1 and hf@xmath5v@xmath6b@xmath1 are evident and confirm the squid magnetometry measurement . on the other hand , undoped samples exhibit a constant plateau , as reflected also by the constance of knight shift ( see sec . [ ssec : nmr ] ) .
we attribute the small dropdown of zrb@xmath1 to zrb@xmath55 impurities .
the inset shows the resistivity data of zr@xmath3v@xmath4b@xmath1 between 2 and 35k has been measured at different fields ( from 0 to 7 t ) , confirming a reduced steepness towards higher fields ( @xmath111 ranging from 2.5 to 3.5k towards higher fields ) and the expected negative shift.,scaledwidth=45.0% ] each resistance measurements in zero field ( zf ) was performed from 2 to 310k , while in field we focused our attention on the superconducting transition region ( range from 2 to 10k ) .
due to the small resistivity values , i.e. , @xmath112 0.7 @xmath113 cm for zr@xmath3v@xmath4b@xmath1 and 0.8 @xmath113 cm for hf@xmath5v@xmath6b@xmath1 , we argue that we can analyze our data within the clean limit approximation . based on the data at different fields ,
the @xmath114 values were evaluated . according to the theory of werthamer , helfand , and hohenberg ( whh),@xcite in the clean limit@xcite and for small spin - orbit couplings,@xcite
we have @xmath115 within this approximation , @xmath116 can be fitted with a parabolic curve : @xmath117\ ] ] @xmath116 plots for zr@xmath3v@xmath4b@xmath1 ( filled orange circles ) and hf@xmath5v@xmath6b@xmath1 ( filled red triangles ) are derived from resistivity measurements at different field from 0 to 7 t . via the whh model , that implies a parabolic fit ( green and blue continuous lines ) , we can evaluate @xmath118 , i.e. 16.4 t for zr@xmath3v@xmath4b@xmath1 and 10.8 t for hf@xmath5v@xmath6b@xmath1 .
since hf is heavier than zr , we argue that the lower value in the upper critical field is related to the lower average phonon frequency . the old data ( ref . , empty squares ) and the new ones ( circles ) are in good agreement.,scaledwidth=45.0% ] from the ginzburg - landau formula @xmath119^{1/2}$ ] , our estimate of the coherence lengths @xmath120 in both zr@xmath3v@xmath4b@xmath1 and hf@xmath5v@xmath6b@xmath1 are 4.5(1 ) and 5.5(1)nm , respectively .
by numerically solving @xmath121 we finally obtain the ginzburg - landau @xmath122 parameter , with a value of 125 for zr@xmath3v@xmath4b@xmath1 and 80 for hf@xmath5v@xmath6b@xmath1 , respectively indicating the strong type ii nature of these superconductors . from @xmath123 ,
the london penetration depth @xmath124 in the two materials , is 570 and 445 nm , respectively .
values of similar magnitude for the magnetic field penetration depth were obtained from preliminary muon - spin rotation ( @xmath44sr ) experiments on zr@xmath3v@xmath4b@xmath1 ( see fig . [
fig : zrvb2_tf_musr ] ) . once implanted in matter , spin - polarized muons act as microscopic probes of magnetism , which upon decay emit positrons preferentially along the muon - spin direction . from the spatial anisotropy of the emitted positrons ( i.e. , the asymmetry signal )
one can reveal the distribution of the local magnetic fields .
@xcite in our case , by applying of 0.6 t , a regular flux - line lattice ( fll ) develops in the superconducting phase below @xmath25 . by uniformly sampling the fll muons experience an additional relaxation @xmath125 , which is related to the absolute magnetic penetration depth @xmath23 via:@xcite @xmath126 here
@xmath127t@xmath44m@xmath103 is the magnetic flux quantum and @xmath128mhz / t , the muon gyromagnetic ratio .
transverse - field @xmath129 data in the superconducting phase of zr@xmath3v@xmath4b@xmath1 in a magnetic field @xmath130t.,scaledwidth=45.0% ] figure [ fig : zrvb2_tf_musr ] shows the temperature dependence of @xmath131 , proportional to the superfluid density ( @xmath132 ) , together with a numerical fit with an average - field model @xmath133 $ ] , which gives @xmath134@xmath44m@xmath67 and @xmath135 .
subsequently , by using the relation @xmath136 , we estimate the in - plane magnetic penetration depth @xmath137 nm .
this value is close to the one determined via macroscopic methods ( see above ) , but it is very different from @xmath138 nm , also measured via @xmath44sr in mgb@xmath1.@xcite this difference can be accounted for by considering the rather small electronic density of states in zr@xmath3v@xmath4b@xmath1 compared to that in mgb@xmath1 , compatible with the very different nmr relaxation - rate values reported in table [ tab : summary ] . as a final note
, we recall that the choice of the applied transverse field is crucial for the correct determination of the field penetration depth . indeed , detailed studies of the magnetic field dependence of the muon - spin depolarization rate in mgb@xmath1 ( see , e.g. , figs . 1 and 2 in ref . )
have shown strong pinning effects in low applied fields ( below 0.3 t ) .
these imply a considerably distorted vortex lattice leading to a strong decay of the muon asymmetry and hence underestimated @xmath139 values .
however , since in fields exceeding 0.3 t , only weak or no pinning effects were observed , we are confident that by applying a transverse field of 0.6 t , our results reflect the penetration depth .
from squid magnetometry , nmr , resistivity , and preliminary @xmath44sr experiment data on zrb@xmath1 , hfb@xmath1 , zr@xmath3v@xmath4b@xmath1 , and hf@xmath5v@xmath6b@xmath1 , we argue that the latter two samples are @xmath0-wave superconductors , resulting from electron doping via @xmath13 orbitals of vanadium . on the other hand , in mgb@xmath1 the peculiar self - doping exists
thanks to the boron - like electrons at the fermi level ; at the same time also the @xmath0 states of mg donate their electrons to the boron - derived conduction bands ( _ metallic _ b ) .
it is also worth mentioning that b vacancies ( a common defect occurring in diborides ) enhance the dos even further ( more than twice for doping of about 0.5%).@xcite besides the many similarities with mgb@xmath1 , the lower @xmath25 values ( 7.31 and 8.33k instead of 39k ) are accounted for by the differences in the electronic structure and fermi surface .
the dos per unit cell of mgb@xmath1 at the fermi level ( 0.719states / ev ) is dominated ( about 60% ) by the @xmath51 orbitals of b atoms , while in the case of the undoped transition - metal diborides ( zrb@xmath1 , for instance ) 80% of the dos ( 0.130states / ev)@xcite derives from the @xmath13 orbitals of the cation . furthermore , our data support the hypothesis that we are dealing with new two - band two - gap @xmath0-wave superconductors , since the coexistence of a normal and a superconducting phase below @xmath25 suggests an anisotropy of the upper critical field , as well documented in the paradigmatic case of mgb@xmath1 .
more direct evidence for this assumption could be obtained by various experiments probing single crystals of zr@xmath3v@xmath4b@xmath1 and hf@xmath5v@xmath6b@xmath1 .
42ifxundefined [ 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 10.1007/bf01501127 [ * * , ( ) ] @noop `` , '' ( ) , , link:\doibase 10.1038/35065039 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.86.4656 [ * * , ( ) ] link:\doibase 10.1103/physrevb.65.180505 [ * * , ( ) ] link:\doibase 10.1103/physrevb.67.180510 [ * * , ( ) ] link:\doibase 10.1103/physrevb.87.174502 [ * * , ( ) ] http://stacks.iop.org/0953-8984/27/i=43/a=435702 [ * * , ( ) ] link:\doibase 10.1134/1.1514768 [ * * , ( ) ] link:\doibase 10.1016/j.jallcom.2009.03.123 [ * * , ( ) ] link:\doibase 10.1016/s0921 - 4534(02)02299 - 2 [ * * , ( ) ] link:\doibase 10.1103/physrevb.45.13244 [ * * , ( ) ] link:\doibase 10.1063/1.458452 [ * * , ( ) ] link:\doibase 10.1063/1.1316015 [ * * , ( ) ] \doibase 10.1002/(sici)1097 - 461x(1998)69:3<423::aid - qua19>3.0.co;2 - 2 [ * * , ( ) ] link:\doibase 10.1016/0022 - 5088(74)90169 - 6 [ * * , ( ) ] link:\doibase 10.1016/s1093 - 3263(99)00028 - 5 [ * * , ( ) ] link:\doibase 10.1103/physrevb.65.094512 [ * * , ( ) ] link:\doibase 10.1103/physrevb.65.012510 [ * * , ( ) ] link:\doibase 10.1002/pssb.2221400125 [ * * , ( ) ] link:\doibase 10.1103/physrev.96.1278 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.450211 [ * * , ( ) ] \doibase http://dx.doi.org/10.1016/s0079-6565(99)00019-9 [ * * , ( ) ] link:\doibase 10.1103/physrevb.66.104510 [ * * , ( ) ] link:\doibase 10.1006/jmra.1995.1087 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.87.127001 [ * * , ( ) ] http://stacks.iop.org/0953-2048/16/i=2/a=303 [ * * , ( ) ] link:\doibase 10.1103/physrevb.64.012514 [ * * , ( ) ] http://stacks.iop.org/0034-4885/61/i=10/a=002 [ * * , ( ) ] link:\doibase 10.1143/jpsj.26.309 [ * * , ( ) ] link:\doibase 10.1103/physrevb.77.214507 [ * * , ( ) ] link:\doibase 10.1063/1.4821283 [ * * , ( ) ] http://stacks.iop.org/0953-2048/15/i=2/a=102 [ * * , ( ) ] link:\doibase 10.1103/physrevb.63.220508 [ * * , ( ) ] link:\doibase 10.1103/physrevb.64.094522 [ * * , ( ) ] link:\doibase 10.1103/physrev.147.288 [ * * , ( ) ] link:\doibase 10.1103/physrevb.84.104502 [ * * , ( ) ] link:\doibase 10.1103/physrev.147.295 [ * * , ( ) ] link:\doibase 10.1080/001075199181521 [ * * , ( ) ] @noop _ _ ( , , ) link:\doibase 10.1103/physrevb.37.2349 [ * * , ( ) ] link:\doibase 10.1103/physrevb.68.054506 [ * * , ( ) ] | unlike the widely studied @xmath0-type two - gap superconductor mgb@xmath1 , the chemically similar compounds zrb@xmath1 and hfb@xmath1 do not superconduct above 1k . yet , it has been shown that small amounts of self- or extrinsic doping ( in particular with vanadium ) , can induce superconductivity in these materials . based on results of different macro- and microscopic measurements , including magnetometry , nuclear magnetic resonance ( nmr ) , resistivity , and muon - spin rotation ( @xmath2sr )
, we present a comparative study of zr@xmath3v@xmath4b@xmath1 and hf@xmath5v@xmath6b@xmath1 .
their key magnetic and superconducting features are determined and the results are considered within the theoretical framework of multiband superconductivity proposed for mgb@xmath1 .
detailed fermi surface ( fs ) and electronic structure calculations reveal the difference between mgb@xmath1 and transition - metal diborides . |
diversity of dark matter candidates requires the balanced experimental program in direct / indirect detection , astroparticle probes and collider searches .
often these approaches make heavy use of numerical tools and computer simulations , and especially there is a variety of useful programs for the purpose of collider searches . in the case of dark matter , there are a few tools available but practically micromegas @xcite is the only tool that works for any arbitrary model for physics beyond the standard model .
it is based on calchep and includes various components such as relic abundance , direct and indirect detection capabilities . in this talk
we introduce the updated version of maddm @xcite , a new dark matter tool based on a well known collider package , madgraph5@xmath0amc@nlo framework @xcite .
new version incorporates direct detection capability in addition to relic abundance calculation @xcite .
short description on the implementation of relevant effective operators and validations against existing results are presented in the following sections .
maddm is available for download from launchpad web site , https://launchpad.net/maddm and no separate installation is required .
however it uses python routines in madgraph5@xmath0amc@nlo and maddm needs to be copied under the main directory of madgraph5@xmath0amc@nlo .
new version of maddm inherits the already established structure of maddm v1 .
a python module prepares relevant amplitudes for relic abundance and direct detection ( with aloha @xcite ) , while a fortran module deals with the numerical calculations .
maddm is compatible with any ufo ( universal feynrules output @xcite ) model that contains a dark matter candidate and can be easily linked to any width or mass spectrum generator which can produce a parameter card in les houches format . for direct detection
, maddm follows the procedure described in ref .
@xcite . at the quark level , quark - dm interaction at the low energy
is written in terms of a set of effective operators in table [ efftab ] .
they are further classified as odd and even operators under the interchanges of quark and anti - quarks .
projection over the effective operators select either spin - independent ( si ) and spin - dependent ( sd ) contributions . finally the corresponding cross sections are computed with the user s choice of target material and several options such as exposure time , size of the target material , threshold cut , detector resolution etc .
we have made various consistency checks against existing results in literature .
first we have considered all possible simplified models with different mediators ( scalar , fermion and vector ) and different dm candidate ( scalar , fermion and vector ) , which are shown in table [ efftab ] and verified that si and sd cross sections from maddm are consistent with results from micromegas . as an illustration of validation , here we consider two examples : scalar dm ( @xmath1 ) with higgs ( @xmath2 ) as a mediator ( @xmath3 ) and dirac fermion dm ( @xmath4 ) with a vector ( @xmath5 ) mediator ( @xmath6 ) .
si and sd elastic scattering cross sections are shown in fig .
[ simpmodels1 ] and fig .
[ simpmodels2 ] , respectively , together with results obtained using micromegas . in both cases
, we find an excellent agreement between maddm and micromegas with the differences at @xmath7 level .
similar results are obtained with all operators in the table [ efftab ] .
.,title="fig : " ] .,title="fig : " ] .,title="fig : " ] .,title="fig : " ] as for more complete dark matter models , we have considered minimal universal extra dimensions ( mued ) and minimal supersymmetric standard model ( mssm ) with sps1a mass spectrum . for mssm , two cross sections from micromegas and maddm agree at the @xmath8% level .
we have checked that this is true if we vary the neutralino mass around the sps1a point as well . for mued ,
[ ued ] shows excellent agreement between maddm cross sections and results in literature @xcite .
.,title="fig : " ] .,title="fig : " ] upon validating the dm - nucleon scattering cross section in maddm , we calculated the recoil rates for dm scattering off a target nucleus .
we performed a simple , model independent validation of the recoil rate calculation , where we simply assumed that the dm - nucleon cross section @xmath9 pb , chosen for the purpose of comparison with the results from ref .
@xcite . to reproduce the si
recoil rates as a function of energy / angle as in ref .
@xcite , we employed the differential recoil spectrum integrated over time and angle / energy . fig .
[ dndetarek ] shows the spin - independent recoil rates as a function of recoil energy ( left ) and recoil angle ( right ) .
we find that both distributions are in a very good agreement with the results found in ref .
@xcite , over a wide range of target materials .
[ dndetarek],title="fig : " ] pb . [ dndetarek],title="fig : " ] as a final validation of the maddm , we have reproduced the exclusion limit on the dm - nucleon scattering cross - section as a function of dm mass as in the lux 2013 paper @xcite .
we assumed the efficiency function of nuclear recoils displayed in the black curve of fig . 1 in ref .
exclusion curves shown in fig .
[ lux_rep ] are obtained assuming 2.3 events , which coincide with the number of events at 90% confidence as required by the feldman - cousins confidence intervals .
we find a good agreement between the lux data and limits from maddm .
the exclusion curves can be obtained in maddm by using ` lux_exclusion ` routine found in the test routines part in ` maddm.f ` . the routine multiplies @xmath10 by the efficiency obtained from ref .
@xcite , which is is then weighted by a 50% acceptance rate for nuclear recoils as stated in the lux analysis . from the recoil spectrum weighted by the efficiency and the acceptance rate ,
the function then calculates the total number of expected events .
the default value for the detector efficiency is 100% , and can be easily replaced with a user defined function .
( right panel ) .
limits are obtained from maddm for 2 kev ( black solid ) , 3 kev ( red dashed ) and 4 kev ( green dot - dashed ) and lux limits are shown in blue curve with circular data points.,title="fig : " ] ( right panel ) .
limits are obtained from maddm for 2 kev ( black solid ) , 3 kev ( red dashed ) and 4 kev ( green dot - dashed ) and lux limits are shown in blue curve with circular data points.,title="fig : " ]
future version of maddm would include various features such as improvement on precision and speed in computation of relic abundance , relic abundance beyond the leading order , direct detection with loop - induced operators , indirect detection , web version , etc . furthermore various dark matter models with more complicated dark sector such as one in ref .
@xcite will be treated properly .
we thank the center for theoretical underground physics and related areas ( cetup * 2015 ) for hospitality and partial support during dark matter workshop and ppc2015 conference .
especially we are grateful to barbara szczerbinska for all the arrangements and encouragement .
this work is supported in part by the belgian federal science policy office through the interuniversity attraction pole p7/37 , by the national research foundation of south africa under grant no .
88614 , and by the u.s .
doe under grant no .
de - fg02 - 12er41809 , and by durham international junior research fellowships .
m. backovic , k. kong , a. martini , o. mattelaer and g. mohlabeng , arxiv:1505.04190 [ hep - ph ] .
j. alwall , m. herquet , f. maltoni , o. mattelaer and t. stelzer , jhep * 1106 * , 128 ( 2011 ) [ arxiv:1106.0522 [ hep - ph ] ] . m. backovic , k. kong and m. mccaskey , physics of the dark universe * 5 - 6 * , 18 ( 2014 ) [ arxiv:1308.4955 [ hep - ph ] ] .
p. de aquino , w. link , f. maltoni , o. mattelaer and t. stelzer , comput .
commun . *
183 * , 2254 ( 2012 ) [ arxiv:1108.2041 [ hep - ph ] ] . c. degrande , c. duhr , b. fuks , d. grellscheid , o. mattelaer and t. reiter , comput .
commun . * 183 * , 1201 ( 2012 ) [ arxiv:1108.2040 [ hep - ph ] ] .
s. arrenberg , l. baudis , k. kong , k. t. matchev and j. yoo , phys .
d * 78 * , 056002 ( 2008 ) [ arxiv:0805.4210 [ hep - ph ] ] .
h. c. cheng , j. l. feng and k. t. matchev , phys .
lett . * 89 * , 211301 ( 2002 ) [ hep - ph/0207125 ] . | we present the updated version of maddm , a new dark matter tool based on madgraph5@xmath0amc@nlo framework .
new version includes direct detection capability in addition to relic abundance computation . in this article ,
we provide short description of the implementation of relevant effective operators and validations against existing results in literature . |
in 1949 schubert @xcite proved that long knots in @xmath10 have a unique decomposition into prime knots .
a concrete statement of his theorem is that there is a homotopy - associative pairing @xmath17 called the connect - sum operation which turns @xmath18 ( the isotopy classes of long knots ) into a free commutative monoid .
the generators are called prime knots .
the idea for why @xmath18 is commutative is summarized in the diagram below .
[ tl][tl][1][0]@xmath19 [ tl][tl][1][0]@xmath20 [ tl][tl][1][0]@xmath21 [ tl][tl][1][0]@xmath22 @xmath23{csloop.eps}\ ] ] ` little cubes and long knots ' @xcite can be viewed as a space - level generalization of schubert s work .
schubert s homotopy - associative connect - sum mapping @xmath24 is enhanced to an action of the operad of @xmath12-cubes @xmath25 on @xmath11 , giving an explicit operadic parametrization of the kinds of isotopies depicted above .
the main theorem of @xcite is that @xmath11 is free as an algebra over the @xmath12-cubes operad @xmath26 , which when we apply @xmath27 recovers schubert s result , since @xmath28 .
schubert went on to further decompose knots using what he called _
satellite operations _ in his massive paper _ knoten und vollringe _ @xcite . as schubert noticed
, there are many ways to construct the same knot via distinct satellite operations . in hindsight
we know this was partially an accident of notation , as schubert s notion of satellite operation was too linearly presented to see the symmetries inherent in the process of constructing satellite knots .
further , satellite constructions produce knots with incompressible tori in their complements , so the uniqueness statement must be tied to the jsj - decomposition of 3-manifolds .
the uniqueness statement for the jsj decomposition is quite delicate and in some sense its delicate nature was a key factor in it being difficult to find .
it has been pointed out several times since and in several different contexts @xcite that when reinterpreted via larry siebenmann s less linearly - ordered notion of _ splicing _
@xcite there is a unique decomposition theorem for satellite knots .
the primary point of this paper is to do for splicing what ` little cubes and long knots ' @xcite did for the connect - sum operation .
an operadic space - level encoding of splicing is given in proposition [ def2 ] .
theorem [ freenessthm ] shows @xmath11 to be a free algebra over the splicing operad @xmath9 , with free generating subspace the torus and hyperbolic knots @xmath29 , i.e. @xmath30 .
this provides a pleasant linkage between the low - dimensional topologists view of knots ( that torus and hyperbolic knots are in some sense the most essential ) , with the algebraic topologist s language of operads .
further , it forms a link between the usage of trees in the study of operads to depict iterated composites of the structure maps with trees in 3-manifold theory , used to depict the structure of the jsj - decomposition of a knot or link complement in @xmath6 .
this is closely related to the somewhat unsatisfactory recursive structure of the homology of the long knot space @xmath11 viewed as an algebra over the operad of @xmath12-cubes @xcite .
the main result of @xcite is that @xmath11 as an algebra over the operad of little @xmath12-cubes is free , where the free generating subspace is the space @xmath13 of prime long knots , i.e. @xmath26 . as was observed in @xcite and @xcite , the homology of @xmath31 has a deeper structure coming from the splicing decomposition of knots , forcing @xmath32 to reappear in shifted degrees inside @xmath33 in many ways .
the non - operadic nature of the description of @xmath11 given in @xcite is non - uniform and somewhat frustrating .
these complications largely disappear when @xmath11 is viewed through the lens of the splicing operad @xmath9 .
theorem [ operadstruc ] shows the splicing operad @xmath9 to be a free product ( in the category of @xmath16-operads ) of @xmath15 ( not the framed discs operad , but a different semi - direct product ) and various free operads .
the other free summands of @xmath9 correspond to cabling operations and hyperbolic satellite operations .
moreover , all these summands with the sole exception of @xmath15 are free operads , freely generated on certain @xmath34-spaces , whose equivariant homotopy - type is identified in theorem [ operadstruc ] .
a secondary point of this paper is that these techniques extend beyond the realm of classical knots .
there are splicing operads that act on a wide class of spaces of self - embeddings @xmath0 , for @xmath1 a compact manifold .
this includes the spaces @xmath2 and @xmath35 @xcite of self - embeddings @xmath3 with support contained in @xmath36^j \times m$ ] and @xmath37 respectively , but the definition of the splicing operad applies to more general self - embedding spaces , some are discussed briefly in section [ futuredir ] . in particular , the splicing operad for the ` cubically supported embedding spaces ' @xmath2 is generally richer than the action of the corresponding action of the @xmath5-cubes operad on @xmath2 .
the splicing operad differs significantly from the operad of cubes , in that the splicing operad is an infinite - dimensional frecht manifold , i.e. it is ` big ' when compared to many traditional operads , which tend to be levelwise finite - dimensional .
another large - scale difference is that while the operad of @xmath5-cubes acts on the space @xmath2 for all compact manifolds @xmath38 , there are distinct splicing operads for @xmath2 and @xmath39 provided @xmath38 and @xmath1 are distinct .
perhaps this new operad will lead to new insights into the homotopy - types of these embedding spaces .
this paper was influenced by conversations with jim mcclure , paolo salvatore and allen hatcher .
thanks to birs hosting allen hatcher s 65th birthday party where i had the opportunity to run these ideas past the participants .
thanks to the university of rome , tor vergata , for hosting me in the summer of 2009 where these ideas indirectly started fermenting . thanks
also to toshitake kohno , the university of tokyo and the institute for the physics and the mathematics of the universe ( ipmu ) for hosting me in the winter of 2010 and 2012 .
thanks to victor turchin and tom goodwillie for comments on the initial drafts of this manuscript .
the point of this section is to provide a motivating result , vaguely this is a ` flattening ' of the operad of little @xmath41-cubes to an equivalent operad called the operad of overlapping @xmath40-cubes .
none of the main results of this paper depend significantly on this section .
these results are provided as context , as part of the train of thought leading up to the construction in section [ splicing ] , which might otherwise seem as uninspired .
the point of this construction is that the operad of overlapping @xmath40-cubes has a more natural action on embeddings spaces , equivalent the the action of the operad of little @xmath41-cubes on @xmath42 .
[ operaddef ] a topological @xmath43-operad is a collection of right @xmath44-spaces @xmath45 for @xmath46 and maps @xmath47 satisfying an ( 1 ) associativity , ( 2 ) symmetry and ( 3 ) identity axiom . given @xmath48 and @xmath49 for @xmath50
denote the image of @xmath51 under the above map by @xmath52 . 1 .
the associativity condition is that @xmath53 whenever this makes sense , i.e. @xmath54 with each @xmath55 belonging to the operad @xmath56 .
the symmetry axiom is that @xmath57 .
we interpret @xmath58 as a @xmath59-tuple @xmath60 , so the left action of @xmath61 on @xmath58 is @xmath62 .
@xmath63 is the associated block permutation to @xmath64 .
similarly there is a symmetry condition @xmath65 provided @xmath66 with @xmath67 for all @xmath68 .
the identity axiom is that there is an element @xmath69 such that @xmath70 for all @xmath71 , and that @xmath72 for all @xmath73 .
an action of the operad @xmath74 on a space @xmath75 is a sequence of maps @xmath76 for @xmath77 satisfying an ( 1 ) associativity , ( 2 ) symmetry and ( 3 ) identity axiom .
as above , let @xmath48 , and @xmath49 for @xmath78 . 1 .
the associativity condition demands that @xmath79 provided @xmath80 2 .
the symmetry condition demands that @xmath81 where the left action of @xmath64 on @xmath82 is given by @xmath83 .
the identity condition demands that if @xmath69 is the identity of @xmath74 , then @xmath84 for all @xmath85 .
operads were originally designed as a category theoretic analogue of universal algebras .
the above definition immediately generalizes to operads in symmetric monoidal categories , see @xcite for example .
the space @xmath86 will be called the _ base _ of the operad ( sometimes called the @xmath87-th operadic grading , or the constants of the operad ) .
notice that the structure maps of @xmath74 restrict to an action of @xmath74 on the base @xmath88 , this will be called the _ augmentation action_. if the base consists of a single point , the operad is said to be _
unitial_. if @xmath89 consists of a single point the operad is said to be _
reduced_. some authors include as part of their definition that the base is empty @xcite , although this is not a uniform requirement among authors .
the first operad discussed in the literature is the operad of little cubes , which appears with both an unbased and unitial variant . in this paper
the cubes operad ( definition [ overlappingcubes ] ) is unitial . for
an operad with non - empty base the structure maps give _ degeneracy maps _
@xmath90 , which when restricted to @xmath91 gives maps @xmath92 , here @xmath93 is a choice of base - point .
[ overlappingcubes ] an increasing affine - linear function @xmath36 \to [ -1,1]$ ] is a _
little interval_. a product of little intervals @xmath36^n \to [ -1,1]^n$ ] is a _
little @xmath40-cube_. the space @xmath94 is the collection of @xmath95-tuples of little @xmath40-cubes whose images are required to have disjoint interiors , @xmath96 is the empty cube .
the collection @xmath97 is the operad of little @xmath40-cubes , it is a @xmath43-operad with structure maps @xmath98 defined by @xmath99 and @xmath100 given by @xmath101 .
we take @xmath102 throughout the paper .
sometimes we will further think of @xmath103 as the subgroup of @xmath104 that fix @xmath87 , but in this case @xmath103 will be denoted @xmath105 . a collection of _
@xmath95 overlapping @xmath40-cubes _ is an equivalence class of pairs @xmath106 where @xmath107 , each @xmath108 is a little @xmath40-cube and @xmath109 .
two collections of @xmath95 overlapping @xmath40-cubes @xmath110 and @xmath111 are taken to be _ equivalent _ provided @xmath112 and whenever the interiors of @xmath108 and @xmath113 intersect @xmath114 . given @xmath95 overlapping @xmath40-cubes @xmath115 we say the @xmath116-th cube @xmath108 is at _ height _ @xmath117 .
@xmath118 is the index of the _ bottom _ cube , and @xmath119 is the index of the _ top _ cube .
let @xmath120 be the space of all @xmath95 overlapping @xmath40-cubes , with the quotient topology induced by the equivalence relation .
the structure map @xmath121 is defined by @xmath122 the permutation @xmath123 is given for @xmath124 , @xmath125 @xmath126 this permutation is obtained by taking the lexicographical order on the set @xmath127 and then identifying with @xmath128 in the order - preserving way .
next we will adapt the action of @xmath129 on @xmath2 to be an action of @xmath130 on @xmath2 .
first a reminder of the definition and geometric context for the action of @xmath129 on @xmath2 .
[ longdef ] a ( thin ) long knot is a smooth embedding @xmath131 which agrees with the standard embedding @xmath132 outside of the cube @xmath133^j$ ] .
the space of thin long knots is denoted @xmath134 . in various situations
one might want to replace @xmath135 in this definition by @xmath136 .
we distinguish between these definitions by saying the knot has _ cubical _ support versus being _ supported on a disc_. it s an elementary rescaling argument that the inclusion @xmath137 is a homotopy - equivalence .
[ tl][tl][1.2][0]@xmath138 [ tl][tl][0.8][0]@xmath139 [ tl][tl][0.8][0]@xmath140 @xmath141{modelspace.eps}\ ] ] a ( fat ) long knot is an embedding @xmath142 such that @xmath143 .
the space of fat long knots is denoted @xmath2 .
the restriction map @xmath144 given by @xmath145 is a fibration whose fibre has the homotopy - type of @xmath146 .
so typically @xmath147 is called the space of _ framed _ long knots , as it consists of knots together with an explicit trivialization of a tubular neighbourhood .
the notation @xmath148 is meant to indicate ` embeddings with cubical support . '
@xmath149 has the homotopy - type of @xmath150 since the fibration @xmath151 splits at the fibre , with splitting given by the linking - number of @xmath152 and @xmath153 .
thus @xmath11 has the homotopy - type of @xmath154 and @xmath155 , where @xmath156 is the subspace of @xmath149 consisting of knots @xmath21 where the above linking number is zero .
the homotopy - equivalence @xmath157 is the restriction map @xcite . as with long knots , if one replaces every occurrence of @xmath135 by @xmath158 one gets a homotopy - equivalent space @xmath35 , the inclusion @xmath159 being a homotopy - equivalence .
the choice of usage of discs or cubes in the definitions of @xmath134 , @xmath35 and @xmath2 becomes important when one wants to study group actions on these spaces .
for example , @xmath160 admits an action of @xmath161 ( by conjugation ) , while @xmath162 does not .
further , the family of spaces @xmath162 fits into a pseudoisotopy fibration sequence ( see @xcite ) , while the family @xmath160 does not .
we assemble the ingredients of the action of @xmath129 on @xmath2 . given a little @xmath95-cube @xmath58 and @xmath163 the _ rescaling _ of @xmath21 by @xmath58 is @xmath164 . for this to make sense ,
reinterpret @xmath58 as its unique affine - linear extension @xmath165 .
given a @xmath5-cube @xmath58 , write it as a product @xmath166 where @xmath167 is a @xmath95-cube and @xmath168 is a @xmath139-cube .
let @xmath169 . given @xmath40 little @xmath5-cubes , @xmath170 define the @xmath40-tuple of ( non - disjoint ) little @xmath95-cubes @xmath171 .
similarly define @xmath172 by @xmath173 .
the action of @xmath129 on @xmath2 @xcite was defined as @xmath174 for @xmath175 which is given by @xmath176 where @xmath177 is any permutation such that @xmath178 .
notice that the action of @xmath129 on @xmath2 has a rather coarse dependence on the cubes @xmath58 , in that only the relative ordering specified by @xmath64 matters , much of the information given by @xmath168 is irrelevant .
this will be made precise in proposition [ flataction ] .
[ tl][tl][0.6][0]@xmath140 [ tl][tl][0.6][0]@xmath139 [ tl][tl][1][0 ] [ tl][tl][1][0]@xmath21 [ tl][tl][0.8][0]@xmath58 [ tl][tl][0.9][0]@xmath179 [ tl][tl][1][0]@xmath180 [ tl][tl][1][0]@xmath181 [ tl][tl][0.9][0]@xmath182 [ tl][tl][0.9][0]@xmath167 [ tl][tl][1][0]@xmath183 @xmath184{caut_action.eps } \hskip 1 cm \includegraphics[width=5cm]{proj.eps}\ ] ] rescaling @xmath21 by @xmath58 in @xmath149 projection @xmath185 and @xmath182 [ tl][tl][1][0]@xmath186 [ tl][tl][1][0]@xmath187 [ tl][tl][0.7][0]@xmath139 [ tl][tl][0.7][0]@xmath140 [ tl][tl][1][0]@xmath188 [ tl][tl][1][0]@xmath189 [ tl][tl][1][0 ] , [ tl][tl][1][0]@xmath190 [ tl][tl][1][0]@xmath191 [ tl][tl][1][0]@xmath192 @xmath193{action_2ndeg.eps}\ ] ] @xmath194 so @xmath195 and @xmath196 [ tl][tl][1][0]@xmath186 [ tl][tl][0.7][0]@xmath187 [ tl][tl][1][0]@xmath197 [ tl][tl][0.7][0]@xmath139 [ tl][tl][0.7][0]@xmath140 [ tl][tl][1][0 ] , [ tl][tl][1][0]@xmath198 [ tl][tl][1][0]@xmath199 [ tl][tl][1][0]@xmath188 [ tl][tl][1][0]@xmath189 [ tl][tl][1][0]@xmath200 [ tl][tl][1][0]@xmath191 [ tl][tl][1][0]@xmath192 [ tl][tl][1][0]@xmath200 @xmath193{action_eg.eps}\ ] ] @xmath201
so @xmath202 and @xmath203 [ flataction ] @xmath130 is a multiplicative @xmath43-operad and the projection map @xmath204 given by @xmath205 as defined above is an operad map which is also a homotopy equivalence .
the maps @xmath206 given by @xmath207 define an action of the operad @xmath130 on @xmath2 , and there is a commutative diagram @xmath208 ^ -{\kappa_n } \ar[d ] & { { \mathrm{ec}({j , m } ) } } \\
{ { \mathcal{c}}}_j'(n ) \times { { \mathrm{ec}({j , m})}}^n \ar[ur]_-{\kappa'_n } & .}\ ] ] to show @xmath130 is an operad , that @xmath209 is an action of the operad on @xmath2 and that the above diagram commutes is mechanical , compare to the proof of theorem 5 in @xcite . to see that the projection map @xmath210 is a homotopy - equivalence , notice that the fibre over any point in @xmath211 is a convex polyhedron , the affine structure being given by the top and bottom coordinates of @xmath168 .
the statement that @xmath130 is a multiplicative operad means that @xmath130 contains the associative operad as a sub - operad .
this is elementary , as @xmath212^j } , \cdots , id_{[-1,1]^j } , id_{\{1,\cdots , k\ } } ) : k\in { { \mathbb
n } } \ } \subset { { \mathcal{c}}}_j'$ ] is isomorphic to the associative operad
. there are ` overlapping ' variants of operads of balls , operads of framed discs and the operads of conformal balls @xcite .
for example , the operad of overlapping @xmath40-balls is equivalent to the operad of @xmath41-balls , but is also multiplicative . the operad of overlapping conformal @xmath40-balls is cyclic and multiplicative but it is not equivalent to the operad of conformal @xmath41-balls .
it fibers over the operad of overlapping @xmath40-balls but the fibre consists of products of @xmath213 .
for knots in @xmath6 , splicing has a particularly physical nature . splicing s role is to create new knots from old .
if a knot is sitting in front of you , with your hands reach out and ` grab ' the knot . in this grabbed position , each hand forms a loop around a collection of strands of the knot . in
abstract , we represent this ` grabbed position ' by a knot together with a disjoint trivial link ( it would be a @xmath12-component trivial link in the case of a single @xmath12-handed person grabbing the knot ) .
the second step involves isolating the strands grasped inside an individual hand , and performing a local modification on the knot . the rough idea for how to perform
the local modification is to cut the strands that pass through an individual hand , and perform a local knotting operation on those loose ends , before re - gluing the strands together .
the important aspect of this heuristic is that splicing involves two steps , ( 1 ) the ` grabbing ' of the knot , represented in definition [ splicedef ] by a _ knot generating link _ ( kgl ) and ( 2 ) the local operation on the ` grabbed ' knot , which is definition [ def1 ] , the splicing operation .
the notion of ` splicing ' was first described by siebenmann @xcite in his work on the jsj - decompositions of homology spheres .
splicing has its roots in schubert s satellite operations @xcite , but only came to prominence with the jsj - decomposition of 3-manifolds .
in 1987 bonahon and siebenmann went on to explain splicing for knots and links in 3-manifolds in some detail , together with the jsj - decomposition of the @xmath214-cyclic branched cover of links in 3-manifolds @xcite although their preprint has been out of distribution until recently .
eisenbud and neumann s book @xcite describes the splice decomposition of graph homology spheres in detail . the refinement of splicing adapted specifically to knots and links in @xmath6 was given in @xcite , of which some elements are sketched in this section .
the main point of this section is the construction of an operad @xmath215 which acts on @xmath2 ( and @xmath216 acting on @xmath35 respectively ) for which the @xmath8 and @xmath7 case the operad s action is splicing in the sense of @xcite , while it is closely related to splicing in the senses of @xcite .
section [ futuredir ] sketches some further generalizations of these operads .
[ splicedef ] a knot - generating link ( kgl ) @xcite is an @xmath41-tuple @xmath217 where @xmath218 is a thin long knot , @xmath219 \times d^2 $ ] is an embedding for @xmath220 such that @xmath217 are disjoint and @xmath221 represents the @xmath40-component unlink .
we require @xmath40 to be non - negative @xmath222 .
a _ splicing diagram _ is an enhanced or ` fattened ' kgl , allowing for a canonical definition of splicing .
while kgl s were developed for the embedding space @xmath11 @xcite , splicing diagrams will make sense for any embedding space of the form @xmath2 or @xmath35 .
a splicing diagram for @xmath2 is an equivalence class of @xmath223-tuple @xmath224 where @xmath225 is a permutation , @xmath226 , and @xmath227^j \times m \to [ -1,1]^j \times m$ ] is an embedding for all @xmath228 .
the equivalence relation is given by @xmath229 together with the relation that if @xmath230^j)^\circ \times m ) \cap l_j(([-1,1]^j)^\circ \times m ) \neq \emptyset$ ] then @xmath231 , where @xmath232 .
there is a further _ continuity constraint _ on a splicing diagram , that whenever @xmath233 , we require @xmath234^j \times m ) ) \setminus l_{k}\left([-1,1]^j \times m\right ) } \cap l_{k}\left(([-1,1]^j)^\circ \times \partial m\right ) = \emptyset$ ] , for any @xmath235 . for the purposes of the continuity constraint
, we use the convention @xmath236 ( i.e. every @xmath237 satisfies @xmath238 ) .
let @xmath239 , with the quotient topology induced by the equivalence relation @xmath240 .
above we use the convention that if @xmath75 is a manifold with boundary @xmath241 denotes the interior @xmath242 .
comments on choices made in the above definition : * if one wants to avoid manifolds - with - corners in the definition of a splicing diagram ( as in definition [ longdef ] ) replace all occurrences of @xmath36^j$ ] in definition [ splicedef ] with @xmath158 , similarly replace @xmath2 by @xmath35
. there are situations in which either formalism appears to be the more appropriate , cubes for pseudo - isotopy fibrations @xcite and discs when interested in symmetry .
let @xmath216 denote the splicing operad using the discs formalism .
notice this makes no difference in the @xmath7 case , i.e. @xmath243 and @xmath244 always . + a splicing diagram .
[ tl][tl][1.1][0]@xmath245 [ tl][tl][0.8][0]@xmath139 [ tl][tl][0.8][0]@xmath140 [ tl][tl][1.1][0]@xmath246 [ tl][tl][0.8][0]@xmath186 [ tl][tl][0.8][0]@xmath187 [ tl][tl][0.8][0]@xmath197 @xmath141{spliceddiag2col.eps}\ ] ] notice that @xmath247 is the only restriction on @xmath248 coming from definition [ splicedef ] , since the image of @xmath187 is partially contained in the image of @xmath197 .
the order ca nt be reversed since @xmath249\times s^1)$ ] intersects the interior of the image of @xmath197 .
* to make sense of the continuity constraint some terminology is useful .
given an element @xmath250 , out of analogy with the @xmath251 case call the embeddings @xmath252 for @xmath253 _ hockey pucks_. @xmath246 is the _ long knot _ associated with @xmath58 and @xmath64 is the mapping ( only well - defined modulo the equivalence relation on splicing diagrams ) from the _ relative heights _ of the pucks to their _ indices _ , i.e. @xmath108 has height @xmath117 .
@xmath254 is a bottom - most puck , @xmath255 is a top - most puck . * * hockey pucks allow for the construction of re - embedding maps . given a hockey puck @xmath108 and @xmath163 , the function @xmath256 is defined on the image of @xmath108 but we extend the definition of @xmath257 to be the map @xmath3 which is the identity on @xmath258^j\times m)$ ] . notice that this function can only fail to be smooth on the set @xmath259^j)^\circ \times \partial m)$ ] , and generally this is precisely the set of points where @xmath257 fails to be differentiable .
splicing diagrams give rise to a _ splicing operation _
( definition [ def1 ] ) and the continuity constraint allows for this to be a smooth map . * * a benefit of the continuity constraint is that it makes splicing diagrams into objects that are similar to links .
for example , given @xmath260 , generically @xmath261 will be a kgl . given @xmath262 , @xmath263 is generically a link with one component ` long ' .
there are certain circumstances where these embeddings will not be disjoint .
these are rare yet important cases , see propositions [ cubesinclusion ] and [ shrinkingprop ] . * * the definition of a splicing diagram does not explicitly state that @xmath264 is a trivial link when @xmath265 , but it follows by a simple induction argument by design the bottom - most hockey puck is disjoint from the other link components .
theorem [ afflin ] can be seen as an enhanced version of this observation .
* for the sake of defining a single splicing operation , disjointness of the pucks is perfectly acceptable .
but there are isotopies between spliced knots ( coming from diagrams with disjoint pucks ) that can not be realized as splices with the pucks disjoint throughout . by keeping track of the permutation @xmath64 and allowing non - disjointness of pucks ,
the definition of splicing diagrams allows the splicing operad , as a space , to capture natural isotopies that happen in spaces of knots .
meaning , the splicing operad more accurately reflects the homotopy - type of embedding spaces .
an example of the action of @xmath266 on @xmath267 from definition [ def1 ] .
[ tl][tl][0.8][0]@xmath191 [ tl][tl][0.8][0]@xmath192 [ tl][tl][0.8][0]@xmath139 [ tl][tl][0.8][0]@xmath140 [ tl][tl][1][0]@xmath246 [ tl][tl][0.8][0]@xmath186 [ tl][tl][0.8][0]@xmath187 [ tl][tl][1][0]@xmath268 @xmath23{splice_eg1.eps}\ ] ] in this example we are thinking of the figure-8 and trefoil knots as normalized to be in @xmath156 , which explains the @xmath269-fold twisting seen in the bottom long knot , as the trefoil s ` blackboard framing ' disagrees with its ` homological framing ' by three twists , while both framings are the same for the figure-8 knot .
[ def1 ] let @xmath270 and @xmath271 .
@xmath272 where @xmath273 and we use the convention that @xmath274 is defined to be the identity outside of the image of @xmath108 .
@xmath275 is called the splicing operation of @xmath58 on @xmath276 .
the remainder of this section is devoted to showing that the space of splicing diagrams forms an operad , and the splicing operation defined above becomes an operad action on @xmath2 . given a collection of composable functions @xmath277^{f_1 } & a_1 \ar[r]^{f_2 } & a_2 \ar[r]^{f_3 } & \cdots \ar[r]^{f_{n-1 } } & a_{n-1 } \ar[r]^{f_n } & a_n } \ ] ] their composite will be denoted @xmath278 [ def2 ] the collection @xmath279 is a multiplicative @xmath43-operad . with definition [ def1 ] ,
@xmath280 acts on @xmath2 .
the operad s structure map has the form @xmath281 ( superscripts @xmath38 and subscripts @xmath95 suppressed ) and is defined below .
let @xmath282 and @xmath283 for @xmath284 , then @xmath285 has @xmath87-th entry @xmath286 the @xmath287-th coordinate entry for @xmath288 and @xmath289 is given by @xmath290 as with definition [ overlappingcubes ] we identify the pairs @xmath127 with the set @xmath291 via the lexicographical ordering . the permutation associated to @xmath52 is the natural one induced by the permutations @xmath292 as in definition [ overlappingcubes ] .
the right action of @xmath44 on @xmath293 is given by @xmath294 \(1 ) associativity .
for this we need to show @xmath53 .
let @xmath295 .
notice the @xmath296-th entry of @xmath297 is given by @xmath298 @xmath299 while the @xmath296-th entry of @xmath300 is given by @xmath301 @xmath302 in this latter composite there are many occurrences of adjacent maps that are the inverses of each other .
cancelling these maps we see the above two expressions for the @xmath296-th term of @xmath300 and @xmath297 are identical . showing
the @xmath87-th entries agree is similar .
\(2 ) symmetry / equivariance .
there are two types , the ` internal ' equivarance , and the ` external ' one . for the internal equivariance
, we need to show that if @xmath303 and if @xmath304 for all @xmath305 with @xmath60 then whenever @xmath306 @xmath307 where @xmath308 , and @xmath63 is the associated block permutation .
this is immediate .
for the external equivarance , we need to show that if @xmath309 , @xmath310 , which is also immediate .
\(3 ) identity / unit .
the identity element in @xmath311 is @xmath312 where @xmath313 is the identity element .
given @xmath314 the identity axiom requires @xmath70 and @xmath315 , which are both satisfied .
that definition [ splicedef ] gives an action of @xmath316 on @xmath2 is a special case of the above arguments , since the structure maps for @xmath316 , @xmath317 _ is _ the action of @xmath280 on @xmath2 , as @xmath318 .
a multiplicative operad is one that contains the associative operad as a sub - operad . for @xmath280 ,
the suboperad is @xmath319^j\times m } , \cdots , id_{[-1,1]^j \times m } , id_{\{1,2,\cdots , k\ } } ) : k \in { { \mathbb n } } \}\subset { { \mathcal{sc}}}_j^m$ ] . an example of the structure map of @xmath266 , in pictures .
[ tl][tl][0.8][0]@xmath320 [ tl][tl][0.8][0]@xmath321 [ tl][tl][0.8][0]@xmath322 [ tl][tl][0.8][0]@xmath323 [ tl][tl][0.8][0]@xmath324 [ tl][tl][0.8][0]@xmath325 [ tl][tl][0.8][0]@xmath326 [ tl][tl][0.8][0]@xmath327 [ tl][tl][0.8][0]@xmath139 [ tl][tl][0.8][0]@xmath140 [ tl][tl][1][0]@xmath52 @xmath23{splice_eg2.eps}\ ] ] [ cubesinclusion ] there is an inclusion of operads @xmath328 given by the maps @xmath329 which have the form @xmath330 .
moreover , the action of @xmath280 on @xmath2 restricts to the action of proposition [ flataction ] .
the inclusion @xmath331 in a picture .
[ tl][tl][0.8][0]@xmath139 [ tl][tl][0.8][0]@xmath140 [ tl][tl][1][0 ] @xmath184{incl_cubes.eps}\ ] ] we visualize the overlapping nature of the intervals as an infinitesimal separation in an orthogonal direction .
similarly for elements of @xmath332 , although we have run out of extra dimensions , so we depict the relative order as if one cylinder were a thin film over the other(s ) .
it is appealing to think of the operad @xmath333 as an enhanced space of @xmath334-component links where the @xmath87-th component is ` long . '
the next proposition makes this a little more concrete in the case that @xmath38 is connected with non - empty boundary .
[ shrinkingprop ] let @xmath38 be a compact connected manifold with @xmath335 non - empty .
@xmath336 then @xmath337 is a suboperad without identity of @xmath280 , moreover the inclusion @xmath338 is a homotopy - equivalence .
the proof is by constructing a homotopy - inverse of the inclusion @xmath338 .
since it will be useful in theorem [ afflin ] we develop the case @xmath339 explicitly .
let @xmath340 be a @xmath341-smooth function such that @xmath342 , @xmath343 for all @xmath344 , @xmath345 for all @xmath346 , @xmath347 for all @xmath348 and @xmath349 for @xmath350 .
the _ standard shrinking map _ for @xmath351 is the family @xmath352 \times { { \mathbb r } } ^j \times d^n \to { { \mathbb r } } ^j \times d^n $ ] given by @xmath353 .
we let @xmath354 denote @xmath355 .
notice that when @xmath356 $ ] , @xmath357 .
given @xmath358 and @xmath356 $ ] , let @xmath359 denote the element where the @xmath87-th entry has the form @xmath360 the @xmath361-th element has the form @xmath362 @xmath363 is our desired homotopy - inverse .
the general case @xmath364 proceeds similarly , using a collar neighbourhood of @xmath365 as a replacement for the linear structure on @xmath366 .
notice that the identity element of @xmath332 is mapped via @xmath367 to the ` hopf link ' in @xmath368 .
the discovery of the operad @xmath369 came about fairly naturally .
individual splicing diagrams analogous to elements of @xmath370 first appear in @xcite as a formally convenient way to encode splicing . as a topological space
something similar to @xmath370 appears in @xcite when describing the homotopy - type of various components of @xmath11 .
thus ideas similar to definition [ def1 ] have been present for some time .
now consider making definition [ def1 ] satisfy an associativity law for a hypothetical operad structure on @xmath280 .
since the associativity law for an operad action uses the structure map of an operad _ only once _ and the action of the operad on @xmath2 _ three times _ , one could use the associativity condition together with a hypothetical action in an attempt to intuit an operad structure map @xmath371 .
this works and is precisely how the author was led to define the operad structure maps for @xmath369 .
we denote the wreath product of a group @xmath372 and @xmath44 be @xmath373 .
the main purpose of the wreath product for this paper is that it is the appropriate group that extends two natural group actions . if @xmath372 acts on @xmath75 , @xmath374 acts on @xmath375 via the product action and @xmath44 acts on @xmath375 via the regular representation .
@xmath373 fits into a short exact sequence @xmath376 . moreover , @xmath373 acts on @xmath375 and its action is equivariant with respect to this short exact sequence . @xmath373 is the group @xmath377 , i.e. the semi - direct product of @xmath374 and @xmath44 where @xmath44 acts on @xmath374 by the regular representation .
we will use the notation @xmath378 to denote @xmath379 .
@xmath380 should be thought of as the above wreath product construction but with the identification @xmath381 , i.e. @xmath382 .
we denote the sequence of groups @xmath383 by @xmath384 . since a preferred copy of @xmath372 splits off @xmath384 , if @xmath75 is a space with an action of @xmath384 , when @xmath385 and @xmath85 , @xmath386 will refer to the action of @xmath372 on @xmath75 coming from this preferred factor .
a @xmath384-operad @xmath74 is a sequence of spaces @xmath45 for @xmath387 together with group actions of @xmath380 on @xmath45 for all @xmath387 satisfying an ( 1 ) associativity axiom , a ( 2 ) symmetry axiom and an ( 3 ) identity axiom .
the ( 1 ) associativity and ( 3 ) identity axioms are exactly as in the definition of a @xmath43-operad .
the symmetry axiom ( 2 ) has two parts , an ` inner ' equivariance , together with an ` outer ' equivariance .
the ` inner ' equivariance condition can be expressed as @xmath388 , where @xmath389 , @xmath48 and @xmath390 .
if we write @xmath391 and @xmath390 as @xmath392 then @xmath393 similarly , if @xmath394 then @xmath395 where @xmath396 is the block permutation associated to @xmath397 .
the outer equivariance can be expressed as @xmath398 whenever @xmath399 , @xmath400 .
note we do not allow @xmath401 for this condition .
the next proposition will investigate further equivariance properties of the splicing operads and their actions .
let @xmath402 and denote the group of diffeomorphisms of @xmath403 that restrict to diffeomorphisms of @xmath404 , where @xmath405 $ ] .
similarly , let @xmath406 be the diffeomorphisms of @xmath37 that restrict to diffeomorphisms of @xmath407 .
[ equivarianceprop ] @xmath2 is taken to be a @xmath402-space , where the action is by conjugation .
similarly , @xmath35 is a @xmath406-space .
there is an action of @xmath408 on @xmath280 making @xmath280 into a @xmath408-operad .
similarly , there is an action of @xmath409 on @xmath216 making @xmath280 into a @xmath409-operad . @xmath280 and @xmath216 act on @xmath2 and @xmath35 in the sense of @xmath384-operad actions .
the right action of @xmath410 on @xmath411 is given by : @xmath412 @xmath413 abbreviate @xmath414 and @xmath415 .
let @xmath416 , and write @xmath417 .
then @xmath418 , whose @xmath87-th entry is @xmath419 and whose @xmath287-th entry ( before lexicographically ordering ) is @xmath420 cancelling inverse maps , these two expressions reduce to @xmath421 and @xmath422 respectively , which are the entries of @xmath423 .
the ` outer ' equivariance condition is immediate .
the next theorem should be thought of as a semi - linear ordering enhancement of cerf s homotopy - classification of spaces of tubular neighbourhoods @xcite .
[ afflin ] let @xmath424 be the subspace where the embeddings @xmath425 are affine linear for @xmath426 .
then the inclusion @xmath427 is a homotopy - equivalence for all @xmath428 .
recall the standard shrinking map from the proof of proposition [ shrinkingprop ] . given @xmath429 and @xmath356 $ ] ,
let @xmath430 denote the element where the @xmath87-th entry has the form @xmath360 the @xmath361-th element has the form @xmath362 the idea of the proof is to shrink elements @xmath429 to the point where we can apply a linearization process .
the linearization process @xmath431
\times d^j \times d^n \to { { \mathbb r } } ^{j+n}$ ] applied to @xmath108 for @xmath305 is given by @xmath432 if we think of this as a time - varying family of maps @xmath433 , we can make some observations on the family .
* for all @xmath348 the map @xmath434 is an embedding , thus the family is an isotopy of @xmath108 . *
@xmath108 and @xmath435 are uniformly close , moreover , an upper bound on their @xmath436-distance is given by the maximum of the norm of the hessian of @xmath108 .
* under the shrinking map the 2nd derivative of @xmath108 goes to zero at an order of magnitude faster than the 1st derivative
. given any @xmath429 , we can apply the shrinking map until linearization can be applied to the @xmath392 part of the family . via linearization
we can ensure @xmath437 are disjoint .
apply isotopy extension to the isotopy @xmath437 allows us to construct the family @xmath438 .
this gives us a path in @xmath439 that begins at @xmath58 and ends in @xmath440 .
moreover , we choose how long to run the shrinking map based on the maximum of the 2nd derivative of @xmath58 , which varies continuously on @xmath439 .
similarly the isotopy extension , since it is a solution to an ode varies continuously with the input isotopy .
this gives us a homotopy of the identity map on @xmath439 to a map @xmath441 , which is a homotopy - inverse to the inclusion @xmath424 .
there is a related theorem of brendle and hatcher @xcite , who have shown that in dimension @xmath269 the space of unlinks has the homotopy - type of the subspace of round unlinks .
their proof is analogous , one key difference is their step where they add spanning discs to their trivial links
this is via an application of the theorem that @xmath442 . to make the analogy a little more explicit
, the shrinking construction above supplies a homotopy - equivalence between @xmath439 and a subspace of @xmath439 where each @xmath58 has a unique semi - linear ordering @xmath306 up to equivalence ( this is essentially the ` separated ' subspace in @xcite ) .
this subspace of @xmath439 is therefore a genuine embedding space and therefore has the homotopy - type of a cw - complex @xcite .
the point of this section is to show how the splicing operad is in some sense a more natural operad than cubes operads for the purposes of describing the homotopy - type of embedding spaces .
this is largely done by example , for the splicing operad s action on the space @xmath11 .
we start by refining the splicing operad @xmath266 , throwing away the parts that contain redundant information from the point of view of the action on @xmath11 , to produce the irreducible splicing operad @xmath9 .
we then show @xmath11 to be free over @xmath9 .
further we show @xmath9 to be a free product of @xmath443 and a free operad over a @xmath444-space , which we identify .
let @xmath445 be the subspace with zero homological framing from definition [ longdef ] .
@xmath266 acts on @xmath267 but notice that it does _ not _ restrict to an action on @xmath156 since it does not preserve the homological framing of the knot .
moreover , not every element of @xmath266 results in a useful splicing construction think for example of an element @xmath446 where @xmath186 is disjoint from @xmath246 .
below we define a suitable suboperad of @xmath266 that acts on @xmath156 in a useful way .
the @xmath269-dimensional irreducible splicing operad @xmath9 is the subset of @xmath266 where @xmath447 is an element of @xmath448 provided all of the following conditions are satisfied : * @xmath449 , i.e. this is an operad with empty base .
* we demand that @xmath108 is an orientation - preserving embedding for each @xmath78 .
* @xmath450 , meaning that the linking numbers of @xmath451 and @xmath452 are zero . *
the link corresponding to @xmath58 is irreducible . *
every incompressible torus in the complement of the link associated to @xmath58 separates components of @xmath58
. condition ( 4 ) above uses _ irreducible _ in the sense of knot theory , that one can not separate components of the link @xmath453 by embedded co - dimension zero balls .
it can be restated as saying that the path - component of @xmath110 in @xmath454 does not contain a representative @xmath111 such that for some @xmath455 @xmath456 is disjoint from @xmath457 .
conditions ( 1 ) and ( 5 ) can be restated as saying the jsj - decomposition of the complement of @xmath58 contains no knot complements ( only link complements with two or more components are permitted in the jsj - decomposition ) .
note also that condition ( 4 ) forces condition ( 1 ) , since if the base of the operad were non - empty , the resulting degeneracy maps ( see the comments following definition [ operaddef ] ) could produce reducible links , as in the case of the borromean rings thought of as an element of @xmath458 .
it s interesting to consider how one might want to generalize the irreducible splicing operad to an appropriate irreducible splicing operad @xmath459 for all @xmath40 and @xmath460 .
there appears to be no high - dimensional analogue of ( 3 ) .
condition ( 4 ) immediately generalizes , although it s not clear when splicing preserves ( 4 ) .
the natural generalization of ( 5 ) would be to talk about incompressible @xmath461 manifolds in the link complement , presumably where incompressible means not bounding a @xmath462 , although perhaps a more flexible definition would be desireable . by the work of hatcher @xcite
, @xmath463 has the homotopy - type of its linear subgroup @xmath464 .
the subgroup that preserves the orientation of @xmath465 is isomorphic to @xmath466 , so we can consider @xmath9 to be a @xmath467-operad and @xmath156 as a space with an @xmath466-action given by conjugation . note that with this action action of @xmath466 on @xmath465 , mirror reflections reverse the orientations of both @xmath468 and @xmath469 factors .
[ complexitydef ] given @xmath470 , let @xmath471 denote the associated link in @xmath6 .
the idea is to consider @xmath472 as the one - point compactification of @xmath473 .
@xmath474 has @xmath334-components @xmath475 is the one - point compactification of @xmath476 .
@xmath477 is the image of @xmath478\times d^2 \subset { { \mathbb r } } ^3 \subset s^3 $ ] .
given @xmath470 we say it is seifert or hyperbolic respectively if the associated link @xmath471 has seifert - fibred or hyperbolic complement , respectively .
given a @xmath269-manifold @xmath38 let @xmath479 denote the number of components of @xmath38 split along its canonical ( geometric ) decomposition .
we ignore the compression - body decomposition .
so for a knot @xmath480 in @xmath6 , the complexity of its complement @xmath481 is @xmath87 if and only if it is the unknot ( since after compression the manifold is empty ) , @xmath139 if and only if it is a torus or hyperbolic knot .
similarly for @xmath58 an irreducible kgl , @xmath482 if and only if @xmath58 is the unknot , @xmath483 if and only if @xmath58 is hyperbolic or seifert . given a link @xmath58 in @xmath6 , the symmetry group of the link is denoted @xmath484 , i.e. the mapping class group of the pair @xmath485 . given @xmath486 , the symmetry group @xmath487 of @xmath58 is the defined to be a subgroup of @xmath488 , where we put the additional restriction that the action on @xmath6 is by orientation - preserving diffeomorphisms and we require that the @xmath475 component is preserved .
@xcite the splicing map @xmath489 satisfies @xmath490 except in the two possible degenerate cases : * @xmath491 is a hopf link , or @xmath477 is a hopf link for some @xmath116 .
* @xmath491 contains two parallel components , @xmath492 and @xmath493 for @xmath494 , i.e. @xmath492 and @xmath493 bound an untwisted embedded annulus disjoint from @xmath495 , and either @xmath496 or @xmath497 are not prime with respect to connect - sum along the @xmath87-th strand .
if @xmath498 we call @xmath52 a _ non - redundant splice_. @xmath499 type redundant splices are the only ones possible in the splicing operad @xmath9 @xcite . in the larger operad @xmath500 redundant splices of the form @xmath501 are possible , but this requires one of @xmath502 to be the unknot . every ( isotopy class of ) element of @xmath503 and @xmath9 can be expressed as an iterated non - redundant splice of objects from @xmath503 and @xmath9 whose complements @xmath38 satisfy @xmath504 .
moreover , up to isotopy and the action of @xmath444 on @xmath9 , this decomposition is unique @xcite .
this should be thought of as the analogous unique decomposition theorem to schubert s prime factorization of knots , but for satellite operations .
theorems [ freenessthm ] and [ operadstruc ] give the generalization of the above to a statement about the homotopy - type of spaces of knots .
[ freenessthm ] let @xmath505 be the subspace consisting of knots which are either non - trivial torus knots , or hyperbolic knots .
then the action of @xmath9 on @xmath156 induces an @xmath466-equivariant homotopy - equivalence @xmath506 the action of @xmath466 on @xmath507 is induced by the outer action of @xmath466 on @xmath508 .
the action of @xmath509 on @xmath508 is given by the inner action ( see proposition [ equivarianceprop ] ) .
further , the components of @xmath29 have two possible homotopy - types : * a torus knot component of @xmath29 has the homotopy - type of @xmath510 .
if @xmath511 is a torus knot there is an @xmath466-equivariant homotopy - equivalence @xmath512 .
the action of @xmath466 on @xmath510 is standard .
@xmath513 denotes the path - component of @xmath156 containing @xmath21 .
* a hyperbolic knot component of @xmath29 has the homotopy - type of @xmath514 . if @xmath511 is a hyperbolic knot , the @xmath466-action preserves @xmath513
if and only if the knot is invertible .
if the knot is invertible , there is an @xmath466-equivariant homotopy - equivalence @xmath515 .
the action of @xmath466 on @xmath514 is given by @xmath516 where @xmath517 . here
@xmath518 and @xmath519 is the standard linear action of @xmath466 on @xmath510 . if the knot is not invertible , the component of the knot @xmath21 and its inverse @xmath520 has the homotopy - type of @xmath521 and there is an @xmath466-equivariant homotopy - equivalence @xmath522 where the action of @xmath466 on @xmath521 is given by @xmath523 , where @xmath524 .
both brendle - hatcher @xcite and theorem [ afflin ] , have a central shrinking and linearization argument that assert that certain spaces of unlinks have the homotopy - type of the subspace consisting of linear embeddings .
both arguments are highly analogous .
although the brendle - hatcher argument is about the space of @xmath59-component unlinks ( denoted by them as @xmath525 ) , it applies equally well to the space of @xmath334-component kgls ( see definition [ splicedef ] ) , since the space of @xmath334-component kgls fibre over the space of @xmath59-component unlinks .
one of the key theorems of brendle - hatcher is that @xmath525 has the homotopy - type of @xmath526 ( the subspace where all the circles are round i.e. geometric circles ) , moreover this space has a homotopy - equivalent subspace @xmath527 where the circles are ` separated ' . the point being that elements of @xmath527 have a well - defined semi - linear ordering .
one component @xmath528 is ` lower ' than another @xmath529 if the shell @xmath530 bounds a ball containing @xmath531 . since the space of @xmath334-component kgls fibres over the space of @xmath59-component unlinks , it is therefore homotopy - equivalent to the subspace where the underlying @xmath59-component unlink is separated .
the proof of proposition [ afflin ] similarly gives a homotopy - equivalence between @xmath448 and the subspace where @xmath532 are separated .
thus @xmath448 has the homotopy - type of the space of @xmath334-component kgls .
given @xmath511 , let @xmath533 denote the path - component of @xmath503 containing @xmath21 .
let @xmath534 be the complement of an open tubular neighbourhood of the associated closed knot in @xmath6 .
let @xmath535 denote the group of diffeomorphisms of @xmath534 which restrict to the identity on the boundary , then @xmath536 .
this is a fairly standard argument based on the fact that the group of diffeomorphisms of the @xmath269-ball that fix the boundary point - wise , @xmath537 , is contractible @xcite ( see @xcite or @xcite for details on the homotopy - equivalence ) . let @xmath538 be the tori of the jsj - decomposition of @xmath534 .
one can think of @xmath539 as defining a rooted tree ( the ` jsj - tree ' @xcite ) where the vertices are the path - components of @xmath534 split along @xmath539 , and the edges are the path - components of @xmath539 .
the root of the tree is the component of @xmath534 split along @xmath539 containing @xmath540 .
let @xmath541 consist of @xmath534 with the submanifold of @xmath534 corresponding to the leaves of the jsj - tree removed .
the complement of @xmath541 in @xmath534 is the union of disjoint non - trivial knot complements @xmath542 , where @xmath543 .
an observation that goes back to schubert @xcite ( reproven in @xcite ) is that disjoint non - trivial knot complements in @xmath6 can be separated by disjoint embedded @xmath269-balls in @xmath6 .
the operation of ` unknotting ' @xmath191 through @xmath544 gives a new embedding of @xmath541 in @xmath6 as the complement of an @xmath334-component link @xmath471 corresponding to some @xmath486 .
the construction of @xmath58 can be made into a unique decomposition for @xmath21 provided we assert that @xmath21 is obtained by splicing i.e. @xmath21 is isotopic to @xmath545 .
let @xmath546 denote the subgroup of @xmath535 which preserves @xmath541 .
the inclusion @xmath547 is known to be a homotopy - equivalence @xcite ( see @xcite for details ) .
so we have a locally - trivial fibre bundle of topological groups @xmath548 .
we use ` locally trivial ' in the sense common in the study of embedding spaces , that fibres can vary as one moves from component to component in the base , in particular they can be empty .
the non - empty fibres can be identified with @xmath549 .
let @xmath550 denote the path - component of @xmath9 corresponding to @xmath58 .
the re - embedding diffeomorphism @xmath551 allows us to identify @xmath552 with @xmath553 .
let @xmath554 be the maximal subgroup of @xmath555 preserving the component @xmath556 for the action of @xmath555 on @xmath557 ( see proposition [ equivarianceprop ] ) . applying the classifying - space functor to the locally - trivial fibre bundle of groups @xmath558 gives a locally trivial fibre bundle with connected base space @xmath559 by design the knots @xmath560 for all @xmath116 ( see definition [ complexitydef ] ) .
the action of @xmath9 on @xmath503 gives us the central vertical map in a commuting diagram of onto fibrations @xmath561 \ar[d ] & { { \mathcal{sp}}}_{3,1}(l ) \times_{a_f } \prod_{i=1}^k \hat{{\mathcal{k}}}_{3,1}(f_i ) \ar[r ] \ar[d ] & { { \mathcal{sp}}}_{3,1}(l)/{a_f } \ar[d ] \\
\prod_{i=1}^k \hat{{\mathcal{k}}}_{3,1}(f_i ) \ar[r ] & \hat { { \mathcal{k}}}_{3,1}(f ) \ar[r ] & { { \mathcal{sp}}}_{3,1}(l ) / { a_f } } \ ] ] since the left and rightmost vertical arrows are homotopy - equivalences , the central vertical arrow is as well .
for the claims describing the @xmath466-action on @xmath29 , the key argument is to find suitable maximal - symmetry positions for the closed versions of the knot in @xmath6 .
the equivariant maps to @xmath156 are given by a stereographic projection construction which appears in detail in the proof of theorem [ operadstruc ] .
consider whether or not the splicing construction is an @xmath466-equivariant homotopy - equivalence . by the @xmath372-whitehead theorem @xcite , it would suffice to show that the map is a weak equivalence of @xmath466-spaces , meaning for every closed subgroup @xmath562 , the splicing map is a homotopy - equivalence when restricted to the subspace fixed by @xmath563 .
if a group @xmath372 acts on a space @xmath75 we denote the @xmath372-fixed point subspace of @xmath75 by @xmath564 . for @xmath563 any non - trivial closed subgroup of @xmath565 this is immediate as only the linearly - embedded unknot is fixed by a non - trivial element of @xmath565 .
the only interesting case remaining is @xmath566 , a subgroup whose fixed points @xmath567 are knots in strong inversion positions . stated another way , showing the splicing map from theorem [ freenessthm ] is an @xmath466-equivariant homotopy - equivalence amounts to showing that for strongly - invertible knots @xmath21 , the space of strong inversion positions of @xmath21 , @xmath568 is homotopy - equivalent to @xmath569 , and the splicing map is such a homotopy - equivalence . since 3-manifolds have equivariant jsj - decompositions @xcite and an equivariant loop theorem @xcite , the proof of proposition 2.1 from @xcite extends , giving the result that if a knot is isotopic to a non - trivial splice , and if that knot is strongly invertible , then one can put the knot into a position where it is simultaneouly strongly invertible and in the image of the splicing map .
thus splicing gives an onto map @xmath570 by the @xmath214-equivariant isotopy extension theorem @xcite , there is a fibre bundle @xmath571 . by a standard cut - and - paste argument using @xcite and @xcite
, @xmath572 is contractible .
thus the space of strong invertibility positions for a knot has the homotopy - type of @xmath573 . repeating the above argument that @xmath535 is a bundle over @xmath553 with fibre @xmath574 in this context , gives the homotopy - equivalence @xmath575 to make the statement of theorem [ operadstruc ] more compact , we introduce some terminology .
denote the @xmath444-suboperad of @xmath9 generated by the inclusion @xmath576 from proposition [ cubesinclusion ] be denoted @xmath577 . given a link @xmath474 in @xmath6
call it _ totally prime _ if the jsj - decomposition of @xmath578 contains no manifolds diffeomorphic to the product of a circle and a punctured disc .
let @xmath579 be the subspace of @xmath580 such that @xmath474 is totally prime .
[ overlinec1prime ] as a @xmath444-operad @xmath577 is equivalent to @xmath443 .
the action of @xmath466 on @xmath581 factors through the homomorphism @xmath582 , @xmath214 acting by mirror reflection on @xmath36 $ ] . by design , @xmath583 , as the act of taking the @xmath444-operadic closure of @xmath584 amounts to adding all linear reparametrizations of the pucks .
the result follows . before proceeding to theorem [ operadstruc ] ,
we record some useful facts about cyclic and dihedral groups acting on @xmath6 . for the next definition we will think of @xmath585 as being the @xmath40-th roots of unity .
given @xmath586 with @xmath587 , the @xmath588-embedding of @xmath589 in @xmath590 is given by the action @xmath591 where @xmath592 .
the _ standard involution _ of @xmath6 is the map @xmath593 .
[ cyclicaction ] let @xmath372 be a finite subgroup of the group of orientation - preserving diffeomorphisms of @xmath6 .
then @xmath372 is conjugate to a subgroup of @xmath594 .
if @xmath595 is cyclic then it is conjugate to a @xmath588-action for some @xmath596 with @xmath587 .
there is only one extension ( up to conjugacy ) of a @xmath588-action of @xmath589 on @xmath6 to an an action of @xmath597 on @xmath6 . if @xmath598 one of the involutions can be taken to be the standard involution .
when @xmath599 the extension of the @xmath600-action is by the antipodal map , as @xmath601 is abelian .
the fact that @xmath372 is conjugate to a subgroup of @xmath590 is the ` linearization ' part of the elliptization conjecture i.e. elliptization modulo the poincar conjecture .
if @xmath372 acts freely , see @xcite .
if the action is not free , see @xcite .
the remainder of this lemma can be derived by considering the eigenspaces of elements of @xmath372 .
notice that the part of @xmath6 on which @xmath372 does not act freely has a rather simple structure . in the case that @xmath372 is cyclic it acts freely on @xmath6 if and only if @xmath602 . if @xmath603 but @xmath604 there is the singular set @xmath605 , which is a trivial knot . if both @xmath606 and @xmath604 then the singular set is @xmath607 , a hopf link . in the case
that @xmath372 is dihedral there are also the circles fixed by the involutions . [ maxsym ] let @xmath608 be a hyperbolic link in @xmath6 .
then it has a maximal symmetry position with respect to the action of @xmath484 , meaning one can isotope @xmath58 into a position where the maps @xmath609 are isomorphisms . * @xmath610 is the group of isometries of @xmath6 that preserve @xmath58 as a set there may or may not be fixed points on @xmath58 .
* @xmath611 is the link symmetry group i.e. the mapping class group of the pair @xmath485 .
* @xmath612 is the group of hyperbolic isometries of the complement of @xmath58 which preserve meridional homology classes
i.e. isometries of the link complement which admit continuous extensions @xmath613 . *
the map @xmath614 is induced by mostow rigidity
i.e. restrict a diffeomorphism of the pair @xmath485 to @xmath615 and choose the unique hyperbolic isometry in that map s homotopy - class .
if we demand that @xmath616 is the trivial link , let @xmath487 denote the subgroup of @xmath484 that preserves @xmath246 and the orientation of @xmath6
. then one can isotope @xmath58 to ensure @xmath487 acts on @xmath485 by isometries of @xmath6 , and where @xmath108 are _ round circles _ for all @xmath50 , that is , the intersection of an affine 2-dimensional subspace of @xmath617 with @xmath6 . for the borromean rings @xmath58 on the left
, @xmath484 is the full octahedral group , having order @xmath618 .
an approximate maximal symmetry position on the left , and the maximal symmetry position for @xmath619 and @xmath186 , with @xmath187 round is on the right . in this picture
the dotted blue circle is the singular set of the action of @xmath620 on @xmath6 . in the language of lemma [ cyclicaction ] , this is the @xmath621-action of @xmath622 on @xmath6 and the dotted blue circle is @xmath623 .
the dotted blue circle intersects both @xmath186 and @xmath187 in two points each , but does not intersect @xmath246 .
[ tl][tl][1][0]@xmath246 [ tl][tl][0.8][0]@xmath186 [ tl][tl][0.8][0]@xmath187 @xmath624{bosym.eps } \hskip 2 cm \includegraphics[width=4cm]{bosym2.eps}\ ] ] the existence of maximal symmetry positions is a standard amalgamation of several major theorems : * the group @xmath612 is finite , since isometry groups of complete finite volume hyperbolic 3-manifolds are finite . by definition , @xmath612 preserves the longitudinal homology classes of @xmath58 so the action extends to an action of @xmath612 on @xmath6 giving an injective homomorphism @xmath625 . * due to the elliptisation theorem @xcite , the action of @xmath612 on @xmath6 is conjugate to a linear action , i.e. there exists a diffeomorphism of @xmath6 , @xmath626 such that the diagram commutes @xmath627_{i \times h } \ar[r ] & s^3 \\
isom_{\mathcal{h}^3}(s^3 \setminus h(l ) ) \times s^3 \ar[ur ] & } \ ] ] where the top horizontal arrow is the action of @xmath612 on @xmath6 and the diagonal arrow is a linear action of @xmath628 on @xmath6 .
@xmath629 is the ` maximal symmetry position ' for @xmath58 .
it is isotopic to @xmath58 since we can assume @xmath630 is orientation preserving .
orientation - preserving diffeomorphisms of @xmath6 are isotopic to the identity @xcite . * to complete the claim one uses work of hatcher and waldhausen that implies @xmath631 is a homotopy - equivalence , and by mostow rigidity that @xmath632 is homotopy - equivalence ,
see proposition 3.2 from @xcite for details . for the remainder of the proof
we replace @xmath58 with @xmath629 . to construct the maximal symmetry position for @xmath487 , apply the equivariant sphere theorem @xcite of jaco and rubinstein to the @xmath487-manifold @xmath633 , where @xmath634 indicates an open tubular neighbourhood of @xmath635 in @xmath6 .
this gives us an equivariant collection @xmath636 of embedded @xmath637 s in @xmath633 which separate the manifold into a collection of punctured spheres ( @xmath6 ) and punctured unknot complements ( @xmath638 ) .
think of @xmath487 as being a group of automorphisms of a rooted tree , the tree s vertices being the path - components of @xmath639 , and edges the path - components of @xmath636 . since finite groups acting on trees either fix a vertex or the centre of an edge , by replacing a sphere from @xmath636 with the boundary of its equivariant tubular neighbourhood in @xmath6 , we can arrange for there to be a vertex fixed by the action of @xmath487 , i.e. some component of @xmath639 is preserved by @xmath487 . by lemma [ cyclicaction ] we have models for the action of the various stabilizers in @xmath487 on the components of @xmath639 .
the components of @xmath639 are punctured spheres so the action is the restriction of some @xmath588-embedding of a dihedral group in @xmath590 , in particular the action is linear . consider a component @xmath640 of @xmath639 corresponding to a leaf of the tree , this is a 3-ball containing a single component of @xmath635 .
the subgroup of @xmath487 preserving @xmath640 , if not trivial has singular set either an unknotted arc in @xmath640 or two unknotted arcs meeting at a central vertex . thus if @xmath108 is in @xmath640 , @xmath108 either hopf links the singular set or meets the singular set in two points . either way , via a shrinking construction we can equivariantly linearize @xmath108 in @xmath640 to a round circle .
this allows us to equivariantly shrink @xmath640 to the point that it is a small round ball .
inductively , we can work from the leaves to the root of the tree associated to @xmath641 and assume all the spheres and link components @xmath532 are round . by equivariant isotopy extension
@xcite we can isotope @xmath58 into a position such that @xmath532 are round circles .
theorem [ operadstruc ] describes the equivariant homotopy - type of the operad @xmath9 .
a key step in the argument is the construction of finite - dimensional subspaces of @xmath9 where the equivariant homotopy type is explicitly understood .
the most elaborate case consists of the components of @xmath9 containing hyperbolic links . using the maximal symmetry positions of hyperbolic links , via a stereographic projection construction
we will create these finite - dimensional families in @xmath9 , allowing us to understand the @xmath642-equivariant homotopy - type of @xmath448 .
will need some conventions relating the group @xmath642 to the geometry of the link @xmath643 .
[ symconventions ] define @xmath644 where @xmath645 is the unit tangent bundle to @xmath477 , and @xmath646 is the frame bundle of @xmath246 , meaning @xmath647 @xmath648 should be thought of as the minimal data to uniquely describe : * a constant - speed diffeomorphism @xmath649 , * a constant - speed diffeomorphism @xmath650 * a unit - length normal vector field to @xmath475 for which its covariant derivative is parallel along @xmath475 , moreover we demand this normal vector field does not homologically link @xmath475 . here ` parallel ' means with respect to the connection on the normal bundle induced by orthogonal projection . by design
there is a left action of @xmath487 on @xmath648 given by post - composition of these parametrizations with an isometry of @xmath6 .
there is also a right action of @xmath651 on @xmath652 given by pre - composition with an isometry of the parametrizing domain @xmath653 , moreover these two actions on @xmath648 commute .
we use the convention that @xmath654 is the trivial @xmath510-bundle over @xmath510 , and @xmath655 is automorphisms of the bundle that are orientation - preserving on the total space .
since any two parametrizations differ by precomposition with an element of @xmath651 , @xmath648 is an @xmath651-torsor .
this induces a canonical injection @xmath656 .
the composition with the projection @xmath657 is an embedding of groups . a hyperbolic link @xmath58 with @xmath658 ( dihedral group of triangle ) . [ tl][tl][1][0]@xmath246 [ tl][tl][0.8][0]@xmath186 [ tl][tl][0.8][0]@xmath187 @xmath624{trans_action.eps}\ ] ]
a key geometric construction in theorem [ operadstruc ] is a map from @xmath648 to the space of kgl s ( see definition [ splicedef ] ) .
given a point of @xmath659 , it determines a point of @xmath660 .
this is a point on @xmath246 together with a unit tangent vector and a unit normal vector .
we think of @xmath6 as the unit vectors in @xmath617 . since all three vectors are orthogonal in @xmath617 , if we think of them as column vectors of a @xmath661-matrix , they extend uniquely to an element @xmath662 .
let @xmath663 be the permutation specified by @xmath659 .
the whitehead link in its maximal symmetry position in @xmath6 with @xmath664 a round circle , together with a sampling of stereographic projections along @xmath475 .
[ tl][tl][1.1][0]@xmath475 [ tl][tl][0.8][0]@xmath664 [ tl][tl][0.8][0]@xmath187 [ tl][tl][0.8][0]@xmath197 @xmath665{geom_whit.eps}\ ] ] let @xmath666 be the uniquely - prescribed constant - speed parametrization specified by @xmath667 , i.e. such that the derivative of @xmath668 at @xmath139 in the counter - clockwise direction is ( up to a positive multiple ) the unit tangent vector of @xmath669 specified by @xmath667 .
consider stereographic projection to be a map @xmath670 for any @xmath671 .
conjugation of @xmath672 by stereographic projection @xmath673^{a_w^{-1}f_{0,w } } \ar[d]^{p_1 } & s^3 \ar[d]^{p_1 } \\
{ { \mathbb r } } \ar[r ] & { { \mathbb r } } ^3 } \ ] ] produces a map @xmath674 which is ` almost ' an element of @xmath11 .
similarly , composing @xmath675 with @xmath676 produces the collection @xmath677 which is ` almost ' a kgl .
this collection is an embedding @xmath678 which fails to be a kgl precisely when @xmath679 fails to be linear in a sufficiently large neighbourhood of @xmath139 , or if @xmath680 get too close to @xmath475 in the sense that their stereographic projections may not be contained in @xmath681 ( see definition [ splicedef ] ) .
this is not a serious obstacle in that we can equivariantly linearize @xmath679 near @xmath139 and suitably rescale via a hyperbolic transformation of @xmath6 at @xmath682 , at which point stereographic projection will give an actual kgl .
a key point in this argument is that since stereographic projection preserves round circles , the stereographic projections of @xmath680 are round circles in @xmath10 , so they bound canonical flat discs which can be fattened into hockey pucks .
[ operadstruc ] as an @xmath467-operad , @xmath9 is the free product of @xmath577 and @xmath683 .
moreover , @xmath683 is a free @xmath444-operad , freely generated by the subspace @xmath684 of @xmath9 consisting of : 1 .
@xmath685 , these are the @xmath12-component seifert links ( hopf link not included ) in @xmath686 .
once we close these links to be links in @xmath6 , these are the links @xmath687 from @xcite with @xmath688 , @xmath587 and @xmath689 , i.e. the seifert link of type @xmath588 is a @xmath12-component link in @xmath6 consisting of two fibres in a @xmath588-seifert fibring of @xmath6 , one fibre singular , the other not .
hyperbolic links @xmath691 , meaning that @xmath486 belongs to @xmath692 if and only if the complement of the corresponding closed link @xmath474 in @xmath6 has a complete hyperbolic structure of finite - volume . restating the above in a different formalism ,
if we restrict the structure map @xmath693 to the appropriate path - components of the domain and range respectively corresponding to a non - redundant splice , then it is an @xmath694-equivariant homotopy - equivalence between those components .
* @xmath685 has the homotopy - type of a disjoint union of countably - many tori @xmath514 , two for every seifert link @xmath687 .
let @xmath695 denote the path - component of @xmath9 corresponding to the @xmath588-seifert link @xmath687 .
only an index two subgroup of @xmath696 preserves @xmath695 , so let @xmath697 be the union of the path - components of @xmath9 containing the link corresponding to @xmath687 and its image under the action of @xmath698 .
there is @xmath699-equivariant homotopy - equivalence @xmath700 , where the right - action of @xmath699 on @xmath521 given by @xmath701 .
* @xmath692 has the homotopy - type of a disjoint union of a countable collection of tori of the form @xmath702 .
notice that the action of @xmath642 may permute path - components of @xmath692 , depending on the symmetry properties of @xmath703 .
so we will describe the homotopy type of not just one path component , but the union of all path components containing @xmath58 and its image under the action of @xmath642 .
denote this subspace of @xmath692 by @xmath704 .
let @xmath474 be the associated closed link in @xmath6 , in its maximal symmetry position ( proposition [ maxsym ] ) .
there is a @xmath642-equivariant homotopy - equivalence @xmath705 where the action of @xmath642 on @xmath652 is given by considering @xmath648 as an @xmath651-torsor , see definition [ symconventions ] .
@xmath706 is a subgroup of @xmath707 .
think of @xmath708 as the subgroup of @xmath709 corresponding to pure translational reparametrizations of @xmath475 , so the inclusion @xmath710 corresponds to reversing the tangent vector to @xmath475 and rotating the normal vector .
our inclusion @xmath711 is induced by this inclusion .
the up - to - isotopy uniqueness statement for the splice decomposition was given in @xcite .
that the splicing map restricts to an equivariant homotopy - equivalence for non - redundant splices , this argument is essentially the same as the proof of theorem [ freenessthm ] , with little modification beyond what is explained below .
the homotopy - types of the spaces @xmath685 and @xmath692 are described in @xcite , although the maps provided in that paper do not respect the @xmath444-action .
below we give a short summary of how the @xmath444-equivariant homotopy - type of each component of @xmath9 are determined .
the component of @xmath712 corresponding to @xmath58 has the homotopy - type of the classifying space of a group of diffeomorphisms of a manifold @xmath713 , denoted @xmath553 .
@xmath713 is the complement of an open tubular neighbourhood of @xmath474 in @xmath6 .
@xmath553 denotes the group of diffeomorphisms of @xmath713 which restrict to the identity on the boundary - component of @xmath713 corresponding to @xmath475 .
we also require the diffeomorphisms to preserve the homology classes ( up to sign ) of the set of meridians corresponding to @xmath680 respectively , as this ensures the diffeomorphisms of @xmath713 extend to diffeomorphisms of @xmath6 . in the case
@xmath713 is seifert - fibred , the diffeomorphism group has the homotopy - type of the fibre - preserving subgroup @xcite .
* for a keychain link this group has the homotopy - type of the braid group on @xmath59 strands .
* for a seifert link , @xmath714 and it has the homotopy - type of @xmath715 .
the generator is a meridional dehn twist about a torus of @xmath716 corresponding to @xmath475 . in the hyperbolic case ,
proposition [ maxsym ] demonstrates that the full group of diffeomorphisms of @xmath713 has the homotopy - type of the group of hyperbolic isometries of @xmath717 .
the subgroup of @xmath718 that preserves the @xmath475 cusp acts faithfully that cusp , so the restriction map from the diffeomorphism group of @xmath713 that preserves the boundary torus corresponding to @xmath475 to the diffeomorphism group of that torus gives us an extension @xmath719 where @xmath276 is a finite cyclic group with at most one generator by the ` no bad monodromy ' result @xcite .
@xmath276 can be understood as the translational symmetries of @xmath475 induced by elements of @xmath718 from the perspective of definition [ symconventions ] .
the @xmath720 kernel consists of all the dehn twists about a torus in the interior of @xmath713 which are parallel to the boundary torus corresponding to @xmath475 .
the extension is non - split provided @xmath276 is non - trivial .
this is because the solution to the extension problem are ` fractional dehn twists ' @xcite .
this means that a diffeomorphism of @xmath713 that fixes the boundary torus ( corresponding to @xmath475 ) pointwise , can be isotoped to agree with an isometry of @xmath578 away from a collar neighbourhood of the fixed torus .
inside that collar neighbourhood the diffeomorphism is free to be arbitrary translations of the torus fibers .
@xmath721 is therefore free abelian of rank two .
since @xmath448 fibres over the space of kgls , the remainder of the proof is devoted to constructing an equivariant lift of the stereographic projection construction following definition [ symconventions ] to a @xmath34-equivariant map @xmath722 which descends to a homotopy - equivalence @xmath723 . to begin , we need to ` fatten ' @xmath475 , i.e. choose a @xmath487-equivariant tubular neighbourhood @xmath724 of @xmath475 in @xmath6 @xcite .
let @xmath725 for any @xmath726 , considering it to be the total - space of bundle over @xmath510 via the projection map @xmath727 .
trivialize the @xmath487-equivariant tubular neighbourhood explicitly , considering the trivialization to be a fibre - preserving diffeomorphism @xmath728 .
if @xmath729 is sufficiently small , we can ensure @xmath730 is conformal - linear along @xmath731 , and by choosing the constant - speed parametrization of @xmath475 we can ensure the conformal factor is constant .
the partial derivative of @xmath732 at @xmath733 in the direction of @xmath734 is a normal vector field along @xmath475 , and as in definition [ symconventions ] , we can choose it so that its covariant derivative is parallel along @xmath475 , and it does not homologically link with @xmath475 .
this reduces our choice of @xmath732 to the choice of @xmath729 and the initial data in @xmath660 .
let @xmath735 , @xmath736 , @xmath737 and @xmath738 be projection maps for @xmath305 , for the product @xmath739 . using the conventions of definition [ symconventions ]
, we will choose to think of @xmath740 as a map @xmath741 .
given @xmath659 let @xmath742 be the precomposition of @xmath732 with the uniquely - determined rigid motion @xmath743 so that @xmath744 agrees with some positive multiple of @xmath745 to first order at @xmath746 .
let @xmath747 be the unique hyperbolic conformal transformation of @xmath6 fixing @xmath748 such that @xmath749 is an isometry , therefore equal to @xmath745 .
@xmath750 fixes @xmath751 and its derivative is the identity on @xmath752 .
next we will apply a local linearization process to the embedding @xmath750 at @xmath751 . before that , a small digression into two standard linearization processes and how they can be related .
given a diffeomorphism @xmath753 where @xmath754 are open subsets of euclidean space such that @xmath755 , @xmath756 and @xmath757 is the identity @xmath758 , the _ rescaling linearization process _ means the homotopy @xmath759 .
notice that at time @xmath348 , the domain of @xmath760 is @xmath761 , and the image of @xmath760 is @xmath762 .
we can extend @xmath276 to @xmath763 by @xmath764 for all @xmath765 .
this is a variant of what is sometimes called the _
alexander trick_. notice that if @xmath766 has a lipschitz constant @xmath767 for @xmath768 , then the lipschitz constant for @xmath769 is @xmath770 .
similarly , the lipschitz constant for the hessian of @xmath760 is @xmath771 times the lipschitz constant for the hessian of @xmath21 .
the second linearization process we consider is the _ straight - line homotopy_. more precisely , consider the problem of asking when the straight - line homotopy @xmath772 between @xmath21 and @xmath773 is an isotopy : @xmath774 .
one can check , in order for this to be an isotopy , it is sufficient for the lipschitz constants for @xmath766 and the hessian of @xmath21 to be sufficiently small over the domain of @xmath21 . combining the two linearization processes , we can say that given a ball neighbourhood @xmath775 of @xmath755 such that @xmath776 , there is some @xmath729 , @xmath726 such that the straight - line homotopy between @xmath777 on @xmath775 and the identity map @xmath778 is an isotopy .
moreover , @xmath729 can be chosen to depend smoothly on the @xmath779-norm of @xmath21 .
our linearization process for @xmath750 is similar to what we did with @xmath753 .
multiplication by @xmath780 does not make sense on @xmath6 , so we conjugate by the hyperbolic conformal transformations that fix the point @xmath781 .
these conformal transformation conjugate via stereographic projection @xmath781 to multiplication by @xmath780 in @xmath752 so they are completely analogous . specifically , given @xmath782 and @xmath783
the map @xmath784 is multiplication by @xmath348 in @xmath785 conjugated by stereographic projection at @xmath786 to be a map @xmath787 fixing @xmath786 and @xmath788 .
@xmath789^{m_{p , t } } \ar@{}[d]| { { { \xy * { \xy ( 0,-3);(0,-2 ) * * \dir{- } \endxy } * \cir<3pt>{d^u } \endxy } } } & s^n \ar@{}[d]| { { { \xy * { \xy ( 0,-3);(0,-2 ) * * \dir{- } \endxy } * \cir<3pt>{d^u } \endxy } } } \\ q \ar@{|->}[r ] & \frac{\left((t-1)^2(q \cdot p)+t^2 - 1\right)p + 2tq}{(t^2 - 1)(q \cdot p)+t^2 + 1 } } \ ] ] @xmath790 denotes the standard euclidean inner product @xmath791 .
our linearization process will start by considering the family of embeddings for @xmath792 @xmath793 notice when @xmath794 , @xmath795 .
the domain of @xmath796 is @xmath797 , which for @xmath798 contains the _ right hemi - sphere _ @xmath799 .
moreover , for @xmath348 sufficiently close to @xmath139 , @xmath796 approximates the identity map on the right hemi - sphere , _ uniformly _ in the @xmath779-topology .
thus for @xmath348 sufficiently large we can ensure the geodesic / straight - line homotopy in @xmath6 from @xmath800 to the identity map on the right hemi - sphere is an isotopy .
the equivariant isotopy extension theorem @xcite allows us to extend this linearization of @xmath801 to and isotopy of @xmath796 .
let @xmath802 denote the resulting embedding which is linear on @xmath803 .
it is almost never the case @xmath804 is contained in the domain of @xmath802 , but using the convensions of definition [ symconventions ] , @xmath805 is defined on a neighbourhood of @xmath806 in @xmath10 , moreover , it is the identity on a neighbourhood of @xmath807 .
so the composite @xmath808 is defined and is an element of @xmath156 for some @xmath809 $ ] ( see proposition [ shrinkingprop ] for the definition of @xmath810 ) .
moreover , we can choose @xmath630 to vary continuously with @xmath659 .
denote this element @xmath811 .
given @xmath50 let @xmath812 be the constant - speed parametrization of @xmath477 such that @xmath813 , and the derivative of @xmath814 at @xmath139 in the counter - clockwise direction is a positive multiple of @xmath815 .
notice that @xmath816 is a parametrization of a round circle in @xmath36 \times d^2 $ ] and so it bounds a flat disc .
we choose a thickening of that disc and define it to be @xmath817\times d^2 \to [ -1,1]\times d^2 $ ] .
putting these all together , we have a map @xmath722 which is _ by design _ @xmath642-equivariant .
since our action of @xmath487 on @xmath652 commutes with the action of @xmath642 , this map descends to a @xmath642-equivariant map @xmath818 for which we can check is an equivariant homotopy - equivalence . given @xmath819 ,
@xmath745 is a framed point in @xmath475 , denote this point by @xmath820 .
notice that the action of @xmath642 leave the points @xmath820 fixed .
thus , the only fixed points of the action of @xmath34 on @xmath821 come from isometries of @xmath6 which _ reverse _ the orientation of @xmath475 . so as in the proof of theorem [ freenessthm ] we can use the fact that both @xmath822 and @xmath823 are principal @xmath555-bundles .
the remaining @xmath565-actions are free on both , there is only various order two mirror reflection subgroups of @xmath466 that have fixed - points , if the link is strongly invertible ( for hyperbolic links , strong invertibility is implied by invertibility ) . the homotopy - type of the space of strongly - invertible positions in @xmath824 is computed just as in theorem [ freenessthm ] , where we see immediately it has the homotopy - type of a product of two circles and our stereographic projection construction is by design an equivariant homotopy - equivalence .
@xmath9 contains a homotopy - equivalent suboperad such that each component is finite - dimensional .
the finite - dimensional suboperad of @xmath9 is of course the operad freely - generated by the suboperad @xmath825 and the images of the maps @xmath826 for the hyperbolic links @xmath58 in the splicing operad .
theorems [ operadstruc ] and [ freenessthm ] say that the @xmath466-equivariant homotopy - type of @xmath827 , and the @xmath444-equivariant homotopy - type of @xmath828 is completely prescribed by the action of @xmath487 on @xmath652 for the hyperbolic links @xmath58 in the splicing operad . determining which such representations arise is called the _ realization problem _ for the space of knots @xmath11 .
if one was only interested in the homotopy - type of @xmath11 and @xmath828 respectively , one could take @xmath27 of these actions , and consider it to be a homomorphism @xmath829 , and ask which such representations arise ? this is another variant of the realization problem .
the next proposition points out that proposition [ maxsym ] gives new restrictions on which such representations can occur . for the purpose of the realization problem
a representation @xmath830 is only interesting up to conjugacy .
conjugacy classes in the symmetric group are traditionally specified by cycle decompositions , which are essentially partitions of the set @xmath831 .
the group @xmath832 should be thought of as the signed permutation group , and conjugacy classes have a _ signed cycle decomposition_. a signed cycle that preserves all signs is denoted @xmath833 .
let ` @xmath834 ' denote the signed cycle type @xmath835 , meaning all signs are preserved except the last one , which reverses sign .
thus @xmath834 has order @xmath836 , while the sign - preserving cycle @xmath833 has order @xmath95 .
let @xmath837 be the subgroup of @xmath487 that preserves the orientation of @xmath475 , where @xmath838 is a @xmath334-component hyperbolic link in @xmath6 such that @xmath839 is the trivial link . since @xmath840 acts on @xmath475 by translations , it is a cyclic group .
let @xmath40 be the order of the cyclic group @xmath840 .
the representation @xmath841 is conjugate to a product of ( signed ) cycles and there are at most @xmath842 different cycle types can realized in the cycle decomposition of this action .
using the conventions from proposition [ maxsym ] , the action of @xmath840 on @xmath6 is conjugate to a @xmath588-action for some pair of integers @xmath688 with @xmath587 .
the possible cycles that can appear can have : 1 .
length @xmath40 , preserving sign .
these correspond to components of @xmath58 which can be separated from the singular set of the action of @xmath740 on @xmath6 by round balls .
2 . if @xmath604 cycles of length @xmath843 , preserving sign .
these are represented by components of @xmath58 which hopf link the singular set @xmath844 .
3 . if @xmath845 cycles of length @xmath846 , preserving sign .
these are represented by components of @xmath58 which hopf link the singular set @xmath847 .
4 . if @xmath848 , cycles of length @xmath849 , reversing sign .
these are represented by components of @xmath58 which intersect the singular set @xmath847 in two points .
if @xmath604 , there can be a cycle of length @xmath139 , preserving sign .
this corresponds to a single component of @xmath850 coinciding with a component of the singular set of the action , @xmath844 .
moreover , ( 5 ) and ( 2 ) are exclusive .
thus if ( 5 ) holds , @xmath851 is a non - negative integer - linear combination of @xmath40 and @xmath846 .
if ( 5 ) does not hold , @xmath59 is a non - negative integer - linear combination of @xmath40 , @xmath843 and @xmath846 .
sakuma s example where @xmath852 .
@xmath840 is cyclic of order @xmath853 acting on @xmath6 via a @xmath854-action . @xmath855
( with indicated orientations , taking the generator of @xmath840 to be counter - clockwise rotation in the plane of the figure by @xmath856 and rotation by @xmath857 in the direction of the axis orthogonal to the plane ) is the cycle @xmath858 .
[ tl][tl][1][0]@xmath246 [ tl][tl][0.8][0]@xmath186 [ tl][tl][0.8][0]@xmath187 [ tl][tl][0.8][0]@xmath197 [ tl][tl][0.8][0]@xmath859 [ tl][tl][0.8][0]@xmath860 @xmath861{sakumaeg.eps}\ ] ] a hyperbolic example where @xmath862 giving a @xmath863-action on @xmath6 .
moreover one of the components of the link coincides with a singular circle of the action of @xmath487 on @xmath6 .
[ tl][tl][1][0]@xmath246 [ tl][tl][0.8][0]@xmath186 [ tl][tl][0.8][0]@xmath187 [ tl][tl][0.8][0]@xmath197 [ tl][tl][0.8][0]@xmath859 [ tl][tl][0.8][0]@xmath860 [ tl][tl][0.8][0]@xmath864 [ tl][tl][0.8][0]@xmath865 @xmath861{d6symaxis.eps}\ ] ] @xmath840 is cyclic of order @xmath866 , @xmath867 having cycle - type @xmath868 .
this section points out some lines of inquiry that may be productive .
there is a wider class of embedding space that admits a ` splicing operad ' action . given a manifold @xmath1 with a co - dimension zero submanifold @xmath541 ,
denote the space of embeddings @xmath0 with support contained in @xmath541 by @xmath874 .
@xmath35 would be the case @xmath875 and @xmath876 .
assume that @xmath541 is a manifold with co - dimension @xmath12 cubical corners .
moreover , assume @xmath877 is partitioned into two smooth manifolds with a common boundary @xmath878 , @xmath879 the co - dimension @xmath12 corner stratum .
we assume @xmath880 and @xmath881 is properly embedded in @xmath1 .
the associated operad to @xmath874 would consist of equivalence classes @xmath882-tuples @xmath883 with @xmath884 and @xmath885 a self - embedding of @xmath541 , just as in the definition of @xmath216 . call this construction the operad of self - embeddings for the pair @xmath886 .
possibly interesting operads of this type would be when @xmath1 the total - space of a fibre bundle over a closed manifold ( @xmath887 ) with @xmath888 , @xmath889 a co - dimension @xmath87 submanifold .
are operads of self - embeddings ` interesting ' outside of the @xmath216 and @xmath280 cases ?
do they fit into larger structures are there more general higher algebraic structures encoding the basic structure of diffeomorphism groups of manifolds ?
an important difference between the descriptions of @xmath11 as an algebra over the operads @xmath581 and @xmath9 respectively is that , although they are both free , the description of @xmath11 over @xmath581 involves thinking of @xmath581 as an operad with non - empty ` base ' @xmath891 , while by design @xmath449 .
the augmentation maps for @xmath581 consist of ` deleting an interval ' . @xmath892 is non - empty , but in this case the augmentation maps consist of operations including , among others , puck - deletion .
if one deletes a component of the borromean rings , one gets a reducible link , thus @xmath9 has to have empty base . to get puck - deletion as the augmentation map for @xmath266 one has to choose @xmath893 as the base - point of @xmath892 .
if one chooses a non - trivially framed unknot , the augmentation maps become ` twist maps ' along the puck that is deleted .
thus the operad @xmath266 encodes a basic form of knot diagrammatics , and therefore must have a very rich homotopy - type , while @xmath9 is much more rigid and has a relatively simple homotopy - type , as an operad .
is there a useful spaces - of - knots level description of further knot diagrammatics ?
for example , is there an operadic or suitable higher - algebraic formalism for rational tangle decompositions of links @xcite , or d. thurston s knotted trivalent graph constructions @xcite ?
further afield , perhaps the complexes describing spaces of connect - sum decompositions of manifolds @xcite have an enlightening operadic formalism .
l. siebenmann , _ on vanishing of the rohlin invariant and nonfinitely amphicheiral homology @xmath269-spheres .
siegen , siegen , 1979 , pp .
172222 , lecture notes in math . ,
* 788 * , springer , berlin , 1980 . | a new topological operad is introduced , called the splicing operad .
this operad acts on a broad class of spaces of self - embeddings @xmath0 where @xmath1 is a manifold .
the action of this operad on @xmath2 ( self embeddings @xmath3 with support in @xmath4 ) is an extension of the action of the operad of @xmath5-cubes on this space defined in @xcite . moreover
the action of the splicing operad encodes a version of larry siebenmann s @xcite splicing construction for knots in @xmath6 in the @xmath7 , @xmath8 case , for which we denote the splicing operad @xmath9 .
the space of long knots in @xmath10 ( denoted @xmath11 ) was shown to be a free algebra over the @xmath12-cubes operad with free generating subspace @xmath13 , the subspace of long knots that are prime with respect to the connect - sum operation @xcite .
one of the main results of this paper is that @xmath11 is free with respect to the splicing operad @xmath9 action , but the free generating space is the significantly smaller space of torus and hyperbolic knots @xmath14 .
moreover , we show that @xmath9 is a free product of two operads .
the first free summand of @xmath9 is a semi - direct product @xmath15 operad which is _ not _ equivalent to the framed discs operad .
the second free summand of @xmath9 is a free @xmath16-operad , free on @xmath16-spaces which encode cabling and hyperbolic satellite operations , moreover the @xmath16-homotopy - type of these spaces is determined by finding adapted maximal symmetry positions for hyperbolic links in @xmath6 .
this is an in - principle explicit description of the homotopy - type of the space of knots in @xmath6 , and modulo the rather difficult problem of determining the symmetry groups of a class of hyperbolic links and their actions on the cusps , this is a closed form description of the homotopy - type . |
quasicrystals are a class of ordered materials possessing quasiperiodic positional order _ and _ long - range orientational order .
the intriguing property of quasicrystals is that their symmetry does not belong to the traditional crystallographic point groups . as such
, the discovery of quasicrystals has led to a redefinition of the term crystal as any ordered materials having an essentially discrete diffraction pattern . for a traditional crystal or periodic crystal ,
the discrete diffraction pots form an periodic array on their reciprocal lattice . for a quasicrystal ,
the discrete diffraction spots densely fill the reciprocal space although , in practice , only the most intense reflections will be observed . since the discovery of quasicrystals by shechtman@xcite in a rapidly - quenched al - mn alloy thirty years ago
, materials with quasicrystalline order have been found in more than a hundred different metallic alloys@xcite . besides the large number of quasicrystals found in metallic alloys ,
quasicrystalline order has been observed in many soft condensed matter systems , including micelle - forming liquid crystals@xcite , block copolymers@xcite , colloidal suspensions@xcite and binary mixtures of nanoparticles@xcite .
in contrast to the metallic quasicrystals , the building blocks of soft - matter quasicrystals are on a much larger length - scale , e.g. tens to hundreds of nanometers .
soft quasicrystals provide an interesting platform for the study of fundamentals of quasiperiodic long - range orders , as well as potential candidates for advanced applications based on their unique electronic or photonic properties . due to a large number of studies since the discovery of quasicrystals , their structure and symmetry are now well understood@xcite .
in particular , quasicrystalline order can be described as the projection from a higher - dimensional periodic lattice@xcite . on the other hand
, the study of the thermodynamic stability of quasicrystals requires the examination of their free energy , which remains a challenge@xcite .
theoretical approaches to investigating the origin and stability of an ordered phase , including periodic crystals and quasicrystals , often involve minimizing an appropriate free energy functional of the system , and comparing the free energies of different candidate structures@xcite .
therefore a systematic examination of the stability of quasicrystals requires the availability of suitable free energy functionals and accurate methods to compute the free energy of phases with quasicrystalline order .
several microscopic models have been developed over the years to explore ordered structures , which have yielded surprisingly rich phase diagrams . in some cases even stable quasicrystals
are found@xcite .
parallel to the development of microscopic models , phenomenological theories based on coarse - grained free energy functionals have been widely used to study phases and phase transitions of ordered systems .
the utilization of such landau - type theories provides an effective method to investigate the phase behaviour of physical systems exhibiting ordered phases . in the case of block copolymers , a free energy functional developed by leibler@xcite
has provided useful insight and a rather accurate description of the ordered phases of block copolymers .
similarly , for soft quasicrystals , a number of coarse - grained model free energy functionals have been proposed to explore the quasicrystalline order arising from model systems with more than one characteristic length - scales@xcite . among these models , the free energy functional proposed by lifshitz and petrich , or the lifshitz - petrich ( lp ) model@xcite ,
may be the simplest model with two length - scales .
initially the lp model was developed to study parametrically - excited surface waves ( faraday wave@xcite ) . using a two - mode approximation ,
lifshitz and petrich obtained stable two - dimensional patterns with two- , six- and twelve - fold symmetries for the lp model . recently
the lp model has been used to explain the source of stability of soft quasicrystals with dodecagonal symmetry . in particular ,
the stability of soft quasicrystals is attributed to the existence of two length - scales and three - body interactions in the free energy functional@xcite .
nevertheless , the existence and stability of other quasiperiodic structures within the lp model remain an open question .
besides a proper free energy functional of the system , the study of thermodynamic stability requires methods to compute the free energy of the different ordered phases accurately . due to the spatial periodicity
, the computation of periodic crystals can be carried out within a unit cell with periodic boundary conditions . on the contrary
, it is not possible to reduce the structure of a quasicrystal to unit cells because quasicrystals are space - filling ordered structure without spatial periodicity . in the literature , a common method to overcome
this difficulty is to utilize periodic structures with large unit cells to approximate quasicrystals@xcite ; their free energies are also approximated by that of these periodic orders as well .
however , it is not immediately obvious whether such approximation of the free energy is accurate .
indeed , jiang and zhang@xcite recently showed that the free energy density of dodecagonal quasicrystal is _ always _ lower than its approximant evaluated by this approximated approach . an alternative approach to calculate the free energy of quasicrystals is based on the observation that quasiperiodic lattices can be generated by a cut - and - project method from higher - dimensional periodic lattices@xcite .
it follows that the density and free energy of quasicrystals can be obtained using the quasiperiodic lattices derived from the higher - dimensional periodic structure . a previous approach along this line is the gaussian method , in which the density profile of a quasicrystal is assumed to be given by a sum of gaussian functions centered at the lattice points of a predetermined quasicrystalline lattice@xcite .
the width of the gaussian function is treated as a variational parameter , which is optimized to minimize the free energy of the system .
more recently , jiang and zhang@xcite have proposed a generalized spectral method for characterizing the density profile and computing the free energy of quasiperiodic structures . instead of working in the real space
, jiang and zhang utilized the fact that the fourier spectrum of quasicrystalline structures can be lifted into that of a higher - dimensional periodic structure , so that the fourier spectrum of the quasiperiodic pattern can be obtained by projecting the higher - dimensional reciprocal lattice vectors onto the original fourier space through a projection matrix .
details of the lifting and projection depend on the symmetry and rank of the desired structures .
as a special case , the projection method can also be used to investigate periodic crystals by setting the projection matrix as an identity matrix . from this perspective
the projection method provides a unified computational framework for the study of periodic crystals and quasicrystals .
in particular , the free energy of periodic crystals and quasicrystals can be obtained with the same accuracy . in the current work
, we apply the projection method to study the relative stability of different ordered phases of the lp model .
the main objective is to investigate the existence and relative stability of periodic structures and quasicrystals beyond the 12-fold symmetric patterns .
specifically , two - dimensional quasiperiodic patterns with 12- , 10- and 8-fold symmetries as well as several periodic crystal structures including the three - dimensional body - centered - cubic ( bcc ) phase ( with space group @xmath0 ) have been obtained as possible phases of the lp model .
a comparison of the free energies of these candidate phases leads to the construction of a phase diagram of the system , thus extending the earlier results of lifshitz and petrich@xcite .
in particular , the theoretical study predicts that the decagonal quasicrystal can become a stable phase .
since the discovery of quasicrystals , a large number of theoretical studies have been carried out to investigate their symmetry , structure characterization and thermodynamic properties . in general , the formation of quasicrystalline phases is characterized by two or more length - scales .
one possibility of realizing two length - scales is to have a pairwise interaction that depends on two length - scales explicitly .
for example , a recent monte carlo simulation study by dotera _
et al._@xcite demonstrated that a simple step - like interaction potential with two length - scales could lead to the formation of various two - dimensional quasicrystals .
another possibility of realizing two length - scales is to construct a landau - type free energy functional with two or more characteristic length - scales@xcite . in particular , lifshitz and petrich@xcite
have proposed a landau theory with a scale order parameter and with a second - order term containing two length - scales explicitly .
it has been demonstrated that the lp model exhibits dodecagonal quasicrystal .
in what follows we will use the lp theory as a model free energy functional to examine the relative stability of various two - dimensional quasicrystals . in order to examine the relative stability of different ordered phases , the free energies of the candidate structures have to be evaluated accurately .
the candidate phases correspond to free energy local minima in the free energy landscape of the system .
therefore the computation of their free energies corresponds to finding the solutions of the euler - lagrange equation of the free energy functional . for quasiperiodic phases , a direct computation in the real space
can only be carried out approximately using periodic approxmants or other truncation methods . on the other hand ,
the computation can be carried out in reciprocal or fourier space using projection method as described recently by jiang and zhang@xcite . with appropriate choice of the fourier basis ,
the projection method provides a unified numerical framework to study periodic crystals and quasicrystals with the same accuracy . in this work
we apply the projection method to the lp model and determine the relative stability of different phases including two - dimensional dodecagonal , decagonal and octagonal quasicrystals .
the rest of this section gives a brief introduction to the lp model and the projection method for ordered phases .
the lifshitz - petrich model is a generic landau theory for a physical system with two length - scales@xcite .
the original lp model was motivated by experiments on the parametrically - excited surface waves , which exhibits dodecagonal quasiperiodic order@xcite .
lifshitz and petrich@xcite introduced a spatial - varying scale order parameter , @xmath1 , and constructed a free energy functional , @xmath2 $ ] , which governs the pattern - forming dynamics of the system .
the essential characteristic of the lp model is that the disordered phase , corresponding to @xmath3 , undergoes an instability simultaneously at two wave - lengths .
specifically the lifshitz - petrich model is defined by the free energy functional , @xmath4 = \,\frac{1}{v}\int d\mathbf{r}\,\big\{\frac{c}{2}&[(\nabla^2 + 1)(\nabla^2+q^2)\phi]^2 \\ & -\frac{\varepsilon}{2}\phi^2-\frac{\alpha}{3}\phi^3+\frac{1}{4}\phi^4\big\ } , \label{eqn : lp}\end{aligned}\ ] ] where @xmath5 is the volume of the system , @xmath6 is an energy penalty factor , @xmath7 is a temperature - like controlling parameter , and @xmath8 is a parameter characterizing the third - order interactions .
the order parameter @xmath1 corresponds to the density profile for a soft matter system . comparing with the usual landau theory of phase transition ,
the crucial feature of the lp model is the occurrence of the second length - scale specified by the parameter @xmath9
. it should be noticed that the basic length scale of the model is take as the unit wavelength ( as specified by the @xmath10 term ) and the second length - scale is specified by a wavevector , @xmath9 , which is the ratio of the two characteristic length - scales . in the case of the surface waves ,
the origin of the two length - scales is the two - frequency parametric excitation of the fluid surface@xcite . in the case of generic soft matter system
, the two length - scales could be from the pairwise interactions of the soft particles . in the current work the parameter @xmath9 is taken as one of the controlling parameters of the systems . in principle , free energy minimization should be carried out with respect to @xmath9 as well . however , carrying out such minimization is a formidable task .
instead , we study the relative stability of different patterns with a limited number of @xmath9-values . in their original study@xcite ,
lifshitz and petrich used a two - mode approximation to describe the different ordered phases .
they demonstrated that the simple lp model exhibits various two - dimensional ordered patterns with 2- , 4- , 6- and 12-fold symmetries , and obtained a phase diagram for the lp model based on the two - mode approximation .
subsequent studies@xcite have demonstrated that such free energy functionals with two length - scales could be considered as a generic model for self - assembling soft matter systems , providing insights into the essential features of soft quasicrystals forming in dendrimers , tetrablock terpolymers and micelle - forming liquid crystals .
furthermore , the theory with two length - scales has been used to study the soft - particle systems@xcite .
it is also noticed that three - body interactions characterized by the cubic term in the lp free energy functional plays an important role in stabilizing quasicrystals . for a given set of parameters , the candidate phases are local minima of the free energy functional , corresponding to solutions of the euler - lagrange equation , @xmath11 for the lp model , the euler - lagrange equation ( eq . [ eqn : el ] ) corresponds to a eighth - order nonlinear partial differential equation , which can be solved using a number of numerical methods . for structures with quasicrystalline order ,
an efficient method is the projection method formulated in the fourier space .
it is based on the observation that the fourier spectrum of a @xmath12-dimensional aperiodic structure , such as a quasicrystal , consists of bragg peaks on a , @xmath13-dimensional lattice , where @xmath14 . in other words , the fourier spectrum of the quasiperiodic structure
can be lifted into an @xmath13-dimensional periodic lattice .
the @xmath13-dimensional reciprocal vectors can be spanned by a set of bases @xmath15 , which are the primitive reciprocal vectors in the @xmath13-dimensional reciprocal space , with integer coefficients , _
@xmath13-dimensional reciprocal vector @xmath16 can be written as @xmath17 , @xmath18 , @xmath19 @xcite .
the physical @xmath12-dimensional wavevector @xmath20 is then obtained from the @xmath13-dimensional vector @xmath16 by a projection , @xmath21 , where @xmath22 is a projection matrix of @xmath23-order .
the dimensionality @xmath13 and the specific form of the projection matrix is determined by the structure of the ordered phases@xcite .
the expression of @xmath22 is not unique and relies on the symmetry of the quasicrystals .
more precisely , it depends on the choice of the basis vectors . using the @xmath13-dimensional vectors and the projection operator , the fourier expansion of any quasiperiodic function @xmath1 can be written in the form , @xmath24 in the above expression , @xmath25 , and @xmath26 .
one simple observation is that the expansion ( eq . [ eqn : pm ] ) allows us to treat the @xmath12-dimensional quasiperiodic structure as a slice of an @xmath13-dimensional periodic structure whose orientation is determined by @xmath22 . expanding the order parameter @xmath1 in the form of eq .
[ eqn : pm ] and inserting it into eq .
[ eqn : lp ] , the free energy functional can be written in terms of the fourier coefficients @xmath27 .
for a given structure of interest , the reciprocal lattice vectors are determined by its symmetry , and the optimal coefficients are obtained by minimizing the free energy functional . in our previous work@xcite
, it has been shown that , when using the projection method , it suffices to have a free energy functional defined in the lower ( physical ) @xmath12-dimensional space .
therefore the computations are implemented in the @xmath13-dimensional space , while the final results represent the @xmath12-dimensional structures through ( [ eqn : pm ] ) .
the computation in the @xmath13-dimensional space is carried out on a regular periodic grid since the reciprocal lattice in this space is periodic . as a special case of quasiperiodic structures , a @xmath12-dimensional periodic structure can be described within the projection method by setting the projection matrix as a @xmath28 identity matrix . in this case
the projection method is reduced to the commonly used fourier - spectral method .
this dose not provide any computational advantage when it comes to periodic crystals .
however , this view provides a unified computational scheme of the periodic crystals and quasicrystals . in practice
, we adopt a dissipative method to obtain solutions of the euler - lagrange equation of the free energy functional . inserting the generalized fourier expansion ( eq . [ eqn : pm ] ) into the lp model , the free energy functional ( eq . [ eqn : lp ] )
becomes a function of the fourier coefficients , @xmath29 where @xmath30 , @xmath31 , @xmath32 are the components of the projection matrix @xmath22 , and @xmath33 is the @xmath34-th component of the primitive reciprocal vector @xmath35 . instead of solving the nonlinear euler - lagrange equation directly , we adopt a relaxation method to solve the minimization problem .
specifically , the fourier coefficients are iterated according to the following relaxation equation , @xmath36 it should be pointed out that the variable @xmath37 here is not time , but a parameter controlling the iteration steps . in this expression the quadratic term and cubic term
are given by , @xmath38 from these expressions it is obvious that the nonlinear ( quadratic and cubic ) terms in eq .
[ eqn : lp : pm ] are @xmath13-dimensional convolutions in the reciprocal space
. a direct evaluation of these nonlinear terms will be computationally expensive . instead
, these terms are simple multiplication in the @xmath13-dimensional positional space and the computation of these nonlinear terms in the position space is straightforward .
the pseudospectral method takes advantage of this observation by evaluating the gradient terms in the fourier space and the nonlinear terms in the position space , thus providing an efficient technique to find solutions of the euler - lagrange equation .
the pseudospectral method requires access to the density function in real and reciprocal spaces .
the transformation between the real - space and reciprocal space was done by performing fast fourier transformation ( fft ) in the @xmath13-dimensional space . starting from an initial configuration with a specified symmetry , a steady state solution of eq .
[ eqn : lp : pm ] , corresponding to a local minimum of the free energy functional , is obtained . using initial configurations with different symmetries
leads to different ordered structures as solutions of the minimization problem .
the ordered structures corresponding to these solutions are taken as candidate phases of the problem .
the free energies of these candidate structures are then compared and used to construct phase diagrams of the system .
using the projection method outline above , we will obtain possible ordered phases , and examine their relative stability , of the lp model . due to the computational demand , in the current study ,
the quasicrystals of interest are restricted to two - dimensional quasiperiodic patterns whose point group symmetries can be realized by periodic lattices in 4-dimensional space , namely 5- , 8- , 10- and 12-fold symmetric structures@xcite . therefore the dimensions of the physical space and the reciprocal space are @xmath39 and @xmath40 .
besides , one- , two- and three - dimensional periodic patterns , corresponding to the commonly observed lamellar , cylindrical and spherical phases , are included in our study . in practice , the @xmath13-dimensional fourier space is discretized using @xmath41 basis functions along each direction .
the total number of variables is thus @xmath42 .
we remark that the projection method works equally well if we use different numbers of basis functions along each dimension , although we will not do that here for simplicity . a semi - implicit scheme is adopted to solve eq .
[ eqn : lp : pm ] until the relative change of @xmath43 between consecutive iteration steps is smaller than @xmath44 . for the cases of periodic crystals and quasicrystals ,
the computation starts with initial configurations with the desired symmetries .
the choice of initial configurations will speed up the computations by reducing the number of iterations .
if the initial symmetric structure is a local minimum of the free energy functional for a given set of model parameters , the calculation will lead to a converged solution with the prescribed symmetry . in the case
that the chosen symmetry of the initial configuration does not correspond to a local minimum , the iteration procedure will lead to other phases , or more commonly , to the trivial solution @xmath3 corresponding to the disordered phase .
it should be noticed that the homogeneous phase is always a solution of the euler - lagrange equation . in this subsection
we enumarate the ordered phases used in our study .
all these ordered phases are solutions of the euler - lagrange equation , corresponding to local minima of the lp free enegy functional .
these ordered phases are used as candidate structures for the construction of the phase diagram . at a given point of phase space
only one of these phases could become stable whereas all the others are metastable .
order parameter profiles @xmath1 and @xmath45 ( with @xmath46 removed ) obtained by minimizing free energy density ( eq . [ eqn : lp ] ) with different choices of @xmath9 . for ( a)-(c ) , the upper panels are density profiles , while the below ones are the corresponding fourier transforms .
( a ) 12-fold symmetric pattern for @xmath47 , ( b ) 10-fold symmetric pattern for @xmath48 , ( c ) 8-fold symmetric pattern for @xmath49 , ( d ) 6-fold symmetric crystal with two different length scales , located at @xmath50 and @xmath51 ( @xmath52 ) . ] with proper choice of the second length - scale characterized by @xmath9 , solutions corresponding to periodic and quasiperiodic patterns are obtained as local minima of the lp free energy functional . in particular
, the choice of @xmath47 , @xmath53 and @xmath54 leads to three two - dimensional quasicrystals , with 12- , 10- and 8-fold symmetries , as the potential stable phases in the model .
the order parameter or density profiles @xmath1 , and the corresponding profiles in the fourier space @xmath55 , of these ordered structures are shown in fig.[fig : phases](a)-(c ) . besides these quasicrystalline phases , several periodic structures , corresponding to two - dimensional patterns with 2- , 4- and 6-fold symmetries and a three - dimensional periodic phase of spheres packed on a body - centered - cubic ( bcc ) lattice ,
have been observed as local minima of the lp free energy functional .
fig.[fig : phases](d ) presents the order parameter profiles of the 6-fold symmetric pattern . because the energy penalty factor @xmath6 is finite , more nonzero fourier modes which do not set on two rings of radius @xmath56 and @xmath9 appear as fig.[fig : phases ] shows .
before we examine the stability of these ordered structures , it is useful to verify the symmetries of these quasicrystalline patterns .
_ dodecagonal quasicrystal ( ddqc ) _ the 12-fold symmetric dodecagonal quasicrystal is obtained by choosing structural parameter @xmath47 . the corresponding projection matrix is given by , @xmath57 as is shown in fig.[fig : phases ] ( a ) , the fourier spectrum of this structure consists of two stars of reciprocal vectors as its principal fourier components ( red dots in fig.[fig : phases ] ( a ) ) .
each star contains 12 vectors separated evenly by an angle of @xmath58 .
the reciprocal vectors on the @xmath59 circle are sums of two neighboring reciprocal vectors on the @xmath60 circle .
the real space and fourier space profiles of this structure clearly exhibit 12-fold rotational symmetry .
the structural parameter of the dodecagonal quasicrystal is in agreement with that of lifshitz and petrich@xcite , where the dodecagonal pattern was obtained with the same choice of @xmath9 .
_ decagonal quasirystal ( dqc ) _ the 10-fold symmetric quasicrystal is obtained when @xmath9 is set at @xmath48 , with a projection matrix specified by , @xmath61 fig.[fig : phases ] ( b ) gives the order parameter profiles of this phase in the real space and fourier space , both of which exhibit the 10-fold rotational symmetry .
the principal fourier components ( red dots in fig.[fig : phases ] ( b ) ) contain twenty reciprocal vectors , with ten vectors located on the circle of @xmath60 , and the other ten on the circle of @xmath59 . unlike the dodecagonal quasicrystal , the reciprocal vectors of the decagonal quasicrystal located on the @xmath59 and @xmath60 circles are collinear .
it should be noticed that the decagonal quasicrystal has the same diffraction pattern as the pentagonal quasicrystal@xcite .
therefore identification of the decagonal quasicrystal requires a detailed examination of the fourier coefficients . for the lp free energy functional
, it has been found that the fourier coefficients on a given circle are real and equal to each other ( _ e.g. _ @xmath62 , @xmath63 when @xmath64 , @xmath65 and @xmath66 ) . from this observation
it can be concluded that the structure obtained from the calculations has a 10-fold symmetry , corresponding to decagonal quasicrystal .
_ octagonal quasicrystal ( oqc ) _ the third aperiodic structure is the 8-fold symmetric octagonal quasicrystal that is found by choosing @xmath49 and a projection matrix , @xmath67 as shown in fig.[fig : phases ] ( c ) , the principal fourier coefficients of this structure compose of two 8-fold symmetric stars of wavevectors on the @xmath60 and @xmath59 circles .
similar to the case of ddqc , the wavevectors located on @xmath59 circle are sums of the neighboring wavevectors located on the @xmath60 circle
. _ periodic crystals _ the projection method is used to obtain periodic structures as local minima of the lp free energy functional by setting the projection matrix as a unit matrix . due to the existence of two characteristic length - scales in the lp model , two stable periodic structures with the same symmetry but
different lattice spacings can be obtained@xcite .
these structures are termed _ sibling periodic crystals_. when @xmath52 , a number of periodic phases with their sibling periodic crystals , including two - dimensional 2- , 4- and 6-fold symmetric patterns and a three - dimensional bcc phase , have been obtained from our calculations .
the basic fourier vectors of two sibling phases are located at the circles with radii @xmath68 , and @xmath59 , respectively .
for instance , fig.[fig : phases ] ( d ) presents the real densities and the distinct fourier spectra corresponding to two hexagonal sibling patterns .
note that we only show one copy of the real density , since those associated with sibling phases only differ in scale .
it should be noticed that the candidate structures obtained from our calculations differ from that obtained by lifshitz and petrich@xcite .
specifically , lifshitz and petrich reported that the ddqc could become an equilibrium phase with @xmath47 , whereas the oqc or dqc were not obtained for any choice of @xmath9 .
this difference can be attributed to the different numerical methods used in the calculations . in lifshitz and petrich s simulations ,
a pseudospectral algorithm is applied to a large square computational box with periodic boundary condition . for a given quasicrystal
, the periodic computational domain is strictly determined by the non - crystallographic rotational symmetry , yet still having some approximate error@xcite .
on the other hand , the projection method employed in the current study respects the symmetry of the quasicrystals strictly , leading to a robust method to obtain quasiperiodic structures numerically .
for the lp free energy functional , three quasicrystals and four periodic crystals ( and their siblings ) are obtained as the candidate phases for the model system .
the relative stability of these phases is obtained by comparing their free energies . a phase diagram in the @xmath7-@xmath8 plane
can be constructed from the free energy comparison . in this section
we will examine factors influencing the relative stability of the different phases first , and then present the phase diagram of the system .
the value of @xmath6 will affect the stable areas of different phases .
our extensive calculations for cases with different values of @xmath6 demonstrate that the boundaries between different phases do change with @xmath6 .
however the topological of the stable regions of the different phases is insensitive to the @xmath6-values .
therefore in what follows we will use @xmath64 as an example of finite @xmath6 , and examine the phase behaviour of the system as a function of @xmath7 and @xmath8 . as a initial step in the study of the phase behaviour of the system
, we will examine the relative stability of the sibling phases first .
results for the other ordered phases are presented subsequently .
and the small hexagonal pattern @xmath69 as a function of @xmath8 for fixed @xmath66 , @xmath48 , and @xmath64 .
] _ relative stability of sibling periodic crystals _ sibling periodic crystals are obtained by imposing the same symmetry , but with different length - scales , on the initial configurations of @xmath1 . as an example , fig.[fig : phases ] ( d ) shows the real - space density profile and fourier spectra of the hexagonal sibling phases .
for convenience , the hexagonal structure with larger lattice spacing in the real space is termed as @xmath70 ( the middle figure in fig.[fig : phases ] ( d ) ) , and the one with the smaller lattice spacing as @xmath71 ( the right figure in fig.[fig : phases ] ( d ) ) .
fig.[fig : energy6fold10xi0.5 ] shows the free energy density ( computed from eq .
[ eqn : lp ] ) of the hexagonal structures for @xmath66 , @xmath48 as a function of @xmath72 $ ] .
it is observed that the free energy of @xmath70 is always lower than that of @xmath71 .
our extensive calculations show that for other choices of @xmath7 , @xmath73 and @xmath52 , the free energy of @xmath70 is always lower than that of @xmath71 . as @xmath6 increases , the free energy differences between the two hexagonal sibling phases decrease . when @xmath74 , the fourier modes with nonzero coefficients are constrained to be on the circle with radius @xmath56 or @xmath9 . in this case
the free energies of the two hexagonal sibling crystals become equal .
the comparison of the free energy of the two sibling hex phases clearly demonstrate that the @xmath70 structure always has lower free energy than the @xmath71 phase over the range of model parameters used in our calculations . for other sibling periodic crystals , such as the two - dimensional patterns with 2-fold symmetry and the three - dimensional bcc phases
, we also find that the phases with larger lattice spacing in the physical space have lower free energy than the one with smaller lattice spacings .
therefore , only the periodic phases with larger lattice spacing may become equilibrium phases on the phase diagram . in the following
we will only consider the phases with larger lattice spacing in the phase diagram .
in particular , we will term the hexagonal phase @xmath70 in the phase diagram as the hex phase . on the phase path of fixed @xmath75 and @xmath64 . ]
_ relative stability of quasicrystals and periodic crystals _
we now examine the relative stability of quasicrystals and periodic crystals by comparing their free energies in the @xmath7-@xmath8 plane . in what follows
we will treat @xmath9 as one of controlling parameters of the system , dependent on the symmetry of ordered phases , that is chosen a priori . for the quasiperiodic phases ,
the choice of @xmath9 for the three two - dimensional quasicrystals of interest have been given before .
for the periodic crystals , different values of @xmath9 may affect the free energy of the sibling periodic crystals .
however , it will not significantly change the relative stability of quasicrystals and periodic crystals , merely affects the stability region .
therefore , in what follows we set @xmath48 for periodic crystals . from our calculations
, it is found that the 4-fold symmetric pattern is always metastable .
therefore , this structure is not included in the following analysis .
the candidate structures are therefore composed of the three quasicrystals ( ddqc , dqc and oqc ) , the hex phase and the bcc phase .
using the free energy of the hex phase as the baseline , the free energies of the ddqc , dqc , oqc and bcc phases as a function of @xmath8 are shown in fig.[fig : energyplotxi-0.1 ] with @xmath75 .
from the results shown in fig.[fig : energyplotxi-0.1 ] , it is obvious that the oqc has higher free energy than the other four candidate structures .
thus the oqc structure is a metastable phase of the system with the given set of parameters . on the other hand , the other four candidate structures ( ddqc , dqc , hex and bcc ) can become equilibrium phases with the lowest free energy . specifically ,
when the value of @xmath8 is increased , the phase transition sequence is predicted to be ddqc to bcc to dqc to hex .
the corresponding stable regions of these phases are @xmath76 ( ddqc ) , @xmath77 ( bcc ) , @xmath78 ( dqc ) , and @xmath79 ( hex ) .
this predicted phase behaviour for finite @xmath6 is very different from that of infinite @xmath6 as given by lifshitz and petrich@xcite . for infinite @xmath6
, it was predicted that only the ddqc could become an equilibrium quasicrystalline phase .
this behaviour is justified by the fact that when @xmath80 , the fourier wavevectors of ddqc can form the most triangles among the ddqc , dqc and oqc structures , thus decreasing the free energy of the ddqc structure . however , when @xmath6 is finite , more nonzero fourier modes arise away from the circles with radii @xmath56 and @xmath9 .
the higher - harmonic contributions to the free energy could lower the free energy of the other quasicrystalline structures , resulting in the predicted phase behaviour from the current calculations . for fixed @xmath75 and @xmath64 .
the oqc is a metastable along this phase path , therefore , its energy difference is not shown .
, title="fig : " ] for fixed @xmath75 and @xmath64 .
the oqc is a metastable along this phase path , therefore , its energy difference is not shown . , title="fig : " ] in order to analyze contributions to the free energy from the different fourier modes , it is informative to divide the fourier modes into two parts , _
i.e. _ fundamental modes and higher - harmonics .
the fundamental modes are those fourier modes with nonzero coefficients lying on circles with radius @xmath56 and @xmath9 , and the higher - harmonics are the other fourier modes with nonzero coefficients . the fundamental part of energy is defined by the contribution of the fundamental fourier modes to the free energy , while the higher - harmonic part of energy is the remainder when subtracting the fundamental part of energy from the total energy .
the fundamental part of energy can be calculated analytically using the two - mode approximation method@xcite .
the analytic expressions of the fundamental part of energy corresponding to various patterns is given in the appendix .
fig.[fig : energyplotxi-0.1part ] shows the fundamental and higher - harmonic contributions to the free energy for the ddqc , dqc and bcc phases using the hex phase as the baseline .
the free energy of the oqc structure is not included here since the oqc is always a metastable phase of the model system ( see fig.[fig : energyplotxi-0.1 ] ) . from fig.[fig : energyplotxi-0.1part ] ( a ) , it can be concluded that the ddqc is favoured by the fundamental modes because it has the largest number , @xmath41 , of nonzero fourier modes located on the circles with radii of @xmath56 and @xmath9 , thus forming the most triangles in the fourier space . on the other hand ,
since the energy penalty factor @xmath6 is finite , the higher - harmonics can not be ignored .
as fig.[fig : energyplotxi-0.1part ] ( b ) shows , these higher - harmonics have large impacts on the free energy of the system .
hex has the lowest higher - harmonics contribution to the free energy , since its fourier modes with nonzero coefficients form the least tetragonal interaction .
it can be observed from eq .
( [ eqn : lpqcfour ] ) that the four - body interaction or the quartic term increases the free energy
. meanwhile , the differential term is no longer zero , which increases the energy . due to the competition of the fundamental modes and higher - harmonics , the stable phases are ddqc , bcc , dqc and hex patterns as @xmath8 increases . to further understand the influence of the higher - harmonics on the stability region of various phases , we set @xmath66 and examine the contributions to the free energy as a function of @xmath8 . in this case , the quadratic and cubic terms in eq .
[ eqn : lpqcfour ] decrease the free energy since @xmath81 , while the quartic terms still increase the free energy .
the free energy differences as a function of @xmath82 $ ] are plotted in fig.[fig : energyplotxi0.5 ] . for fixed @xmath66 and @xmath64 .
] for this set of parameters , the hex , bcc and dqc structures form the possible equilibrium phases , whereas the ddqc and oqc structures become metastable .
the stability regions of the bcc and dqc phases are @xmath83 and @xmath84 , respectively . compared to the previous discussion for the case of @xmath85 ,
the stability region of the bcc structure becomes smaller .
an interesting phenomenon is the reentrance transition of the hex phases , in which its stability regions , @xmath86 and @xmath87 , are separated by the bcc and dqc phases .
moreover , the hex phase in these two regions share the same lattice spacing .
similar to the case of @xmath85 , the contributions to the free energy from different fourier modes can be used to understand the phase behavior ( see fig.[fig : energyplotxi0.5part ] ) . for fixed @xmath66 and @xmath64 .
the oqc is a metastable along this phase path , therefore , its energy difference is not shown .
, title="fig : " ] for fixed @xmath66 and @xmath64 .
the oqc is a metastable along this phase path , therefore , its energy difference is not shown .
, title="fig : " ] as is noted before , since the fundamental fourier modes of ddqc and bcc structures can form more triangles than that of the hex phase , the ddqc and bcc phases have lower free energy from the fundamental modes , as shown fig.[fig : energyplotxi0.5part ] ( a ) . however , the free energy differences in both fundamental and higher - harmonic parts between the dqc and hex phases exhibit non - monotonic behaviour . specifically , when @xmath8 is small , hex has lower energy value in the fundamental part .
as @xmath8 increases , dqc becomes more favoured .
when @xmath8 is large enough , hex has lower free energy again . from eqs . and
, we find that dqc has more two- , three- and four - body interactions in the fundamental part of energy than the hex does .
more precisely , when @xmath8 is small , the sum of the cubic terms in the energy expression of dqc is smaller than that of the hex , and the sums of the quadratic and quartic terms are larger than the counterparts for hex .
therefore , when @xmath8 is small , the quadratic term dominates the free energy ; hex thus has lower free energy than the dqc . as @xmath8 increases , the cubic term becomes important in @xmath43 , and thus dqc has lower energy in the fundamental part .
when @xmath8 is sufficiently large , the value of cubic and quartic terms of dqc becomes comparable with the hex ( results not shown ) .
thus the difference in the quadratic terms dominates the difference of the free energy in this case .
therefore , hex becomes favored in energy again .
the energy differences in the higher - harmonic part are given in fig.[fig : energyplotxi0.5part ] ( b ) .
the ddqc and bcc phases have relatively high energy as before .
the energy difference between dqc and hex has the similar behavior with the fundamental part , which can be explained in a similar way .
therefore , as @xmath8 increases , the phase transition occurs in the sequence of hex , bcc , dqc , and hex .
the phase transition sequence for other values of @xmath7 can be obtained by repeating the free energy comparison among the candidate structures .
the results of the phase transition sequences can be summarized in terms of phase diagrams in the @xmath7-@xmath8 plane . for a fixed value of @xmath64 , the phase diagram in the range of @xmath88 and @xmath89
is presented in fig.[fig : phasediagram ] for the lp free energy functional . .
lam and d represent the 2-fold pattern and disordered phase , respectively . for ddqc and dqc
, the structural parameter @xmath9 equals to @xmath90 and @xmath53 , respectively . for periodic crystals , @xmath48 . ] besides the ddqc , dqc , oqc , hex and bcc phases discussed above , two more phases , the 2-fold symmetric pattern , or the lamellar phase ( lam ) , and the disordered phase ( d ) are included in the phase diagram .
the regions of stability of the different phases are obtained by comparing the free energy of these seven candidate structures .
the phase boundaries are determined by calculating the cross over point of the free energies of the two neighbouring phases . because the contribution from the higher - harmonics becomes significant for the cases with finite values of @xmath6 , the predicted phase diagram from the current study includes more features when compared with the previous phase diagram obtained from the analytic calculation using the two - mode approximation@xcite . in the previous work , lifshitz and petrich
used the two - mode approximation method@xcite to analytically analyze the phase behavior of the lp model when @xmath6 is sufficiently large .
it is found that when @xmath47 and @xmath80 , the lam phase is stable for @xmath91 ; the hex phase has the lowest energy for @xmath92 ; and the ddqc is stable for @xmath93 . while the generic feature of the phase diagrams is maintained ,
there are significant differences between the analytic phase diagram and the numerical result presented in the current work , which contains more features by taking higher - harmonics into account .
the first noticeable difference is that the current phase diagram includes the dqc as a new equilibrium quasicrystal phase , whereas the previous phase diagram only contains the ddqc as the unique stable quasicrystalline phase .
it is interesting to notice that the dqc - hex phase boundary possess a special turning point at @xmath94 .
increasing @xmath8 at @xmath95 leads to a reentrance transition of the hex phase .
another interesting feature is the prediction of a triple point at @xmath96 , at which the bcc , dqc and hex phases coexist .
the oqc is only observed as a metastable phase , which is in agreement with the analytic result by two - mode approximation .
moreover , a three - dimensional phase with spheres packed on a body - centred - cubic lattice ( bcc ) is included in the current phase diagram .
in summary , we have applied the projection method to obtain quasicrystalline structures as possible local minima of the lp free energy functional .
the projection method enables accurate numerical calculations of the free energy of the quasicrystals , resulting in a phase diagram with more accurate phase boundaries and , most importantly , more ordered phases as compared with the previous study .
specifically , three two - dimensional quasiperiodic patterns , namely the oqc , dqc and ddqc structures , and several periodic phases ( lam , hex and bcc ) emerge from our calculations as candidate structures for the construction of the phase diagram . by comparing the free energy of the candidate structures , a phase diagram for the case of @xmath97
is obtained in the @xmath7-@xmath8 place .
it is predicted that for the lp free energy function given in eq.[eqn : lp ] , the 10-fold ( dqc ) and 12-fold ( ddqc ) symmetric quasicrystals can become equilibrium phases in the phase diagram . on the other hand the 8-fold ( oqc ) symmetric quasicrystal is found to be a metastable phase of the model system .
the prediction of a stable dqc phase within the lp free energy functional is a new result .
the present study complements and extends the previous work of lifshitz and petrich@xcite .
the two - mode approximation provides an accurate description of the lp model at the limit of @xmath98 , at which the wavevectors are forced to be on two circles defined by @xmath60 and @xmath59 . a finite value of @xmath6 , as exemplified by the case of @xmath64 , allows the occurrence of higher - harmonics other than the modes with @xmath60 and @xmath59 .
the additional fourier modes lead to phase diagram at finite @xmath6 that is significantly different from the phase behavior of the case of @xmath98 .
in particular , the quasicrystalline phases dqc , ddqc , the lamellar phase , hex and bcc phase are predicted to be potentially stable phases for the lp free energy functional .
these results provide a good understanding of the rich phase behaviour contained in the simple lp model .
the numerical methods and insights obtained from them can be helpful for further studying more complex soft matter systems such as block copolymers and soft particles with multi - length - scale interactions .
the work is supported by the national science foundation of china ( grant no.11421110001 , 11421101 , 21274005 , and 11401504 ) .
kj would like to thank the financial supported by the hunan science foundation of china ( grant no .
2015jj3127 ) .
acs acknowledges the support from the natural sciences and engineering research council ( nserc ) of canada .
in the two - mode approximation valid in the limit of @xmath98 , the fourier modes are restricted to the two circles defined by @xmath60 and @xmath59 . keeping only the fourier coefficients at @xmath60 and @xmath59
, the free energy of the system can be obtained analytically .
the analytic expressions of fundamental energy ( which has been defined in section [ subsec : relative stability of ordered phases and phase diagram ] ) of different patterns are given as follows .
@xmath99 @xmath100 @xmath101 @xmath102 @xmath103 where @xmath104 stand for the fourier coefficients on the circles of @xmath56 and @xmath9 , respectively . | the relative stability of two - dimensional soft quasicrystals is examined using a recently developed projection method which provides a unified numerical framework to compute the free energy of periodic crystal and quasicrystals .
accurate free energies of numerous ordered phases , including dodecagonal , decagonal and octagonal quasicrystals , are obtained for a simple model , _
i.e. _ the lifshitz - petrich free energy functional , of soft quasicrystals with two length - scales .
the availability of the free energy allows us to construct phase diagrams of the system , demonstrating that , for the lifshitz - petrich model , the dodecagonal and decagonal quasicrystals can become stable phases , whereas the octagonal quasicrystal stays as a metastable phase . |
fast development of experimental techniques allowing for miniaturization of electronic devices led to renewed interest in theoretical research in the area of quantum transport .
in particular in molecular electronics community focus of the research was shifting from elastic ( landauer ) to inelastic charge transport through molecular junction , to its noise characteristics and heating .
spin flux as an alternative to charge current in electronic devices , where magnetic field provides an additional ( to bias ) mean of control , is studied by spintronics.@xcite quantum ratchets@xcite and electric potential@xcite were proposed as additional controls of the spin flux .
schemes for optical control of a spin trapped in quantum dots were reported in the literature.@xcite recently a combination of spintronics with molecular electronics started to reveal itself as molecular spintronics.@xcite theoretical schemes for spin pumps were considered in refs . , and
inelastic effects of spin transport through the junctions were reported for spin valves@xcite and for tunneling through a junction with embedded spin.@xcite shot noise of spin current was considered in ref . .
here we present a model for a molecular junction , consisting of a molecule between two normal metallic contacts , where pure spin current is controlled by an external electric field .
the spin field - effect transistor is a generalization of a spin pump introduced in ref . .
directed spin current seems to be a more convenient element for molecular spintronics device .
as in the case of the spin pump , only pure spin current participates in the transport .
we consider inelastic effects of the spin current revealed in the transport properties of the device .
this consideration takes place as a part of a general approach of full counting statistics for multi - level molecular systems .
in particular , we discuss current and zero frequency noise of the junction .
flexibility of molecules , and as a result well - pronounced inelastic features in transport properties of molecular junctions , makes heating of a device an important issue in molecular electronics .
spin field - effect transistors ( fet ) inherit this problem of the usual fets , since spin current is also caused by electrons crossing molecule - contact interface which results in heating molecular device . within the model
we discuss spin current through molecular junction , and propose external electric and magnetic fields as possible controls capable of tuning molecular structure to diminish heating of the device .
section [ model ] presents a model and outlines the method .
section [ fcs ] is devoted to full counting statistics of multi - level molecular systems .
we describe general approach and introduce junction characteristics spin and charge current and noise .
section [ heat ] is devoted to heating in spin field - effect transistors . in section [ numres ]
we present results of numerical simulations .
section [ conclude ] concludes .
we consider extension of the spin pump proposed in ref . to the junction situation . the spin pump of ref . is a model of a two - level system coupled to one electrode .
application of magnetic field generates spin current in the contact due to spin - flip process taking place at the molecule .
we propose a generalization where the spin flux is generated across the junction ( spin current between two contacts ) .
note that the spin flux in our model is optically controlled , i.e. external electric rather than magnetic field controls spin flux .
the junction is composed of two molecules ( represented by single levels @xmath0 and @xmath1 coupled to molecular vibrations @xmath2 and @xmath3 ) attached to left @xmath4 and right @xmath5 normal metal electrodes , respectively .
each molecule is subject to a _
dc _ and _ ac _ pair of magnetic fields @xmath6 and @xmath7 ( @xmath8 ) .
note that level dependent g - factors in nanowire quantum dots were recently reported in the literature,@xcite something that , here , is represented by the site dependent field .
the molecules are weakly coupled by external source of light of particular frequency @xmath9 .
the light can be used as a handle to switch on and off spin transport through the junction .
the hamiltonian of the system is @xmath10 where @xmath11\hat n_{i\sigma } \nonumber \\ & + \omega_i \hat a_i^\dagger\hat a_i -g\mu_b b^{(ac)}_i \bigl(\hat d_{i\uparrow}^\dagger\hat d_{i\downarrow}e^ { i\omega^{(b)}_i t } + h.c .
\bigr)\biggr ) \nonumber \\ & + \sum_{k\in l;\sigma}\bigr ( v_{1k}\hat d_{1\sigma}^\dagger\hat c_{k\sigma}+h.c.\bigr ) \nonumber\\ & + \sum_{k\in r;\sigma}\bigl ( v_{2k}\hat d_{2\sigma}^\dagger\hat c_{k\sigma } + h.c.\bigl),\end{aligned}\ ] ] @xmath12 where @xmath13 ( @xmath14 ) and @xmath15 ( @xmath16 ) are creation ( annihilation ) operators for corresponding state in the molecule(s ) and in the contact(s ) , @xmath17 and @xmath18 , @xmath19 ( @xmath20 ) are creation ( annihilation ) operator for vibration quanta of molecule @xmath21 , @xmath22 is direction of spin projection , and where @xmath23 we start by transforming the hamiltonian ( [ h ] ) into the rotating frames of the magnetic field @xcite @xmath24 this transformation eliminates the time - dependence from the ac magnetic field terms , shifts the positions of molecular levels , and induces spin biases in the contacts , according to @xmath25 the perturbation ( [ v ] ) takes the form @xmath26 in what follows we will put @xmath27 to create a spin bias across the junction ( see fig .
[ scheme ] for a sketch ) .
similarly , we put @xmath28 .
the interaction with the external optical field is taken into account within perturbation theory .
expansion of the evolution operator on the keldysh contour up to second order in ( [ vbar ] ) in the rotating - wave approximation ( rwa ) leads to electronic self - energy due to interaction with external optical field in the form ( see appendix [ appa ] for details ) @xmath29 .
\nonumber\end{aligned}\ ] ] as a result of the transformation to the rotating frames and the rwa we obtain a time - independent ( steady - state ) description .
( color online ) sketch of an optically controlled spin field - effect transistor . ] the coupling between electronic and vibrational degrees of freedom is treated below within two different approaches .
the transport through the junction is considered in a basis obtained by a small polaron transformation@xcite @xmath30 this transformation decouples the electronic and vibrational degrees of freedom on the molecule and dresses the molecular fermion operators @xmath31.\ ] ] we take the vibrational shift operators @xmath32 into account within the usual born - oppenheimer - like approximation , which allows us to introduce the franck - condon factors@xcite multiplying the electronic gf @xmath33 in the description of the junction heating , we utilize the born approximation@xcite instead , which allows us to keep the consideration simple . as usual we implement a non - crossing approximation , i.e. diagrams for electron transfer and interaction with vibrations do not cross ( the processes do not happen simultaneously).@xcite in this case green function is obtained from the dyson equation @xmath34 where @xmath35 is green function in the absence of electron - phonon interaction ( @xmath36 ) , and @xmath37 is self - energy due to electron - phonon interaction @xmath38 here @xmath39 is phonon green function . we treat it within a quasi - particle approximation.@xcite
the theoretical concept of full counting statistics was originally proposed by levitov and lesovik.@xcite the approach was applied to the non - equilibrium anderson impurity model in ref . .
measurements of shot noise in molecular junction@xcite prove possibility of experimental observation of moments beyond average current .
this together with the flexibility of molecules , i.e. importance of inelastic effects in transport through molecular junctions , recently caused several theoretical investigations devoted to study of resonant level coupled to single vibration model.@xcite here we discuss a generalization of the result derived by gogolin and komnik@xcite to a multilevel situation , and apply the expression to calculate the first ( current ) and second ( noise ) cumulants of the distribution .
following the derivation of ref .
for the case of multilevel molecule we obtain an expression for the derivative of the adiabatic potential in the form@xcite @xmath40^{-1 } \left(e-\mathbf{h}_0+\mathbf{\sigma}^{++}(e)\right ) + \mathbf{\sigma}_{\lambda}^{<}(e ) \right]^{-1 } \right .
\\ & \left.\ \ - \left[\left(e-\mathbf{h}_0+\mathbf{\sigma}^{++}(e)\right ) \left[\mathbf{\sigma}_{\lambda}^{<}(e)\right]^{-1 } \left(e-\mathbf{h}_0-\mathbf{\sigma}^{--}(e)\right ) + \mathbf{\sigma}_{\lambda}^{>}(e ) \right]^{-1 } \mathbf{\sigma}_k^{>}e^{-i\lambda_k } \right\}. \nonumber\end{aligned}\ ] ] here , @xmath41 ( @xmath42 ) is a counting field for the interface between the molecule and contact @xmath43 on the forward ( backward ) branch of the keldysh contour , @xmath44 . `
@xmath45 ' , ` @xmath46 ' , ` @xmath47 ' , and ` @xmath48 ' are causal , anti - causal , lesser , and greater projections . the trace runs over molecular degrees of freedom .
the electronic self - energy due to the coupling to contact @xmath43 is denoted by @xmath49 , and @xmath50 is the total electronic self - energy dressed with the counting field @xmath51 . in particular , within the non - crossing approximation@xcite @xmath52 where @xmath53 is the electronic self - energy due to interactions dressed with the counting field @xmath51 .
it seems difficult to obtain an expression for the logarithm of the generating function ( integral of ( [ du ] ) over @xmath41 ) in the multi - level case .
however , eq . ( [ du ] ) itself can be used to calculate cumulants .
the time - averaged charge cumulant of order @xmath54 due to charge transport through interfaces @xmath55
( @xmath56 ) is @xmath57 here and below @xmath58 .
first cumulant yields the well - know expression for steady - state current@xcite @xmath59 .
\nonumber\end{aligned}\ ] ] since no spin - flip events are allowed on the metal - molecule interface , the electronic self - energy due to coupling to the contacts is diagonal in spin space , that is , @xmath60 where @xmath61 for @xmath62 and @xmath63 for @xmath64 , whereas @xmath65 is the green function ( gf ) of a free electron . as a result , the charge current of electrons with spin @xmath66 at interface @xmath43 is given by @xmath67 .
\nonumber\end{aligned}\ ] ] the spin and charge currents at interface @xmath43 are @xmath68 respectively .
the second cumulant yields an expression for the zero - frequency noise . in the non - interacting model , it reads@xcite
@xmath69 -\mathbf{i}_{k_1}(e)\,\mathbf{i}_{k_2}(e ) \right .
\nonumber \\ &
\qquad\qquad\qquad + \mathbf{\sigma}_{k_1}^{<}(e)\,\mathbf{g}^{++}(e)\ , \mathbf{\sigma}_{k_2}^{>}(e)\,\mathbf{g}^{--}(e ) + \mathbf{\sigma}_{k_1}^{>}(e)\,\mathbf{g}^{--}(e)\ , \mathbf{\sigma}_{k_2}^{<}(e)\,\mathbf{g}^{++}(e ) \\ & \left.\qquad\qquad\qquad -\mathbf{\sigma}_{k_1}^{<}(e)\,\mathbf{g}^{>}(e)\ , \mathbf{\sigma}_{k_2}^{>}(e)\,\mathbf{g}^{<}(e ) -\mathbf{\sigma}_{k_1}^{>}(e)\,\mathbf{g}^{<}(e)\ , \mathbf{\sigma}_{k_2}^{<}(e)\,\mathbf{g}^{>}(e ) \right\ } , \nonumber\end{aligned}\ ] ] where @xmath70 is a matrix of the energy - resolved current operator at interface @xmath43 @xmath71 the spin and charge noise at interface @xmath43 are expressed by @xmath72 where the spin resolved elements of the noise @xmath73 can be obtained from ( [ sk1k2 ] ) by generalizing the counting field to include spin .
higher order cumulants for a multi - level molecule can be similarly derived from ( [ du ] ) in a , cumbersome but , straightforward manner .
within the model ( fig . [ scheme ] ) , the charge current is compensated at each interface so that only spins are transferred through the junction .
nevertheless , since electrons cross the molecule - contact interfaces also in the spin field - effect transistor , the question of heating of the molecular vibrations is still important . to estimate the _ temperature _ of the vibrations we extend an approximate scheme used by one of us in a previous publication.@xcite
the essence of the approach is quasi - particle assumption used to describe vibrational degrees of freedom ( phonons ) . in this case
the only relevant characteristic of the vibration ( besides frequency ) is its average population , while actual nonequilibrium distribution may be disregarded ( density of states is a delta - function ) . at steady - state influx of energy through the molecule - contact interfaces , @xmath74 ,
is compensated by outflux , @xmath75 , so that energy of the molecule does not change @xmath76 .
since energy is carried by both electrons and phonons , we get @xmath77 within the quasi - particle approximation the latter for a vibration @xmath78 coupled to a thermal bath is@xcite @xmath79,\ ] ] where @xmath78 and @xmath80 ( @xmath8 ) is a vibration and the corresponding average population of molecule @xmath21 , respectively , whereas @xmath81 is the spectral function of the bath , and @xmath82^{-1}$ ] is the bose - einstein distribution .
the electronic energy flux for a system coupled to a set of baths @xmath83 can be written as @xmath84,\end{aligned}\ ] ] where @xmath85 is the electronic self - energy due to the coupling to bath @xmath83 . in our model
we want to estimate the _ temperature _ of two vibrations , thus , following the approximate procedure introduced in ref . , we split the system into two parts at the molecule - molecule interface . in this way we can speak about electronic energy flux in two parts of our system .
each part has two interfaces : one between the molecule and corresponding contact and the other between the molecules .
the electronic self - energy due to the coupling to the contact is given by the standard expression ( [ sk ] ) .
the self - energy on the molecule - molecule interface is given by ( [ se ] ) .
we obtain the electronic flux in the molecule @xmath21 part of the system through the following procedure .
we begin with the hamiltonian @xmath86 which is transformed to the rotating reference frames of the magnetic fields , eq.([rf ] ) , however , without performing the small polaron transformation .
we , then , treat the electron - vibration coupling in the born approximation@xcite . within the non - crossing and quasi - particle approximations
we accordingly obtain the electronic flux @xmath87i^{(-)}_i - n_ii^{(+)}_i\right)\ ] ] where @xmath88 using ( [ jph ] ) and ( [ je_born ] ) in ( [ zero ] ) yields @xmath89 the vibrational _ temperature _ is obtained from ( [ ni ] ) under assumption that the vibrations are populated according to the bose - einstein distribution .
we note that a more physically motivated procedure to introduce the _ temperature _ of the molecular vibrations at non - equilibrium can be used,@xcite however , for demonstration purposes the simpler procedure described above suffices .
atomic cooling caused by sub - resonance optical excitation is a well - known effect.@xcite in molecular junction the analogous detuning may lead to cooling of the molecular vibration . here , we utilize the sub - resonance detuning of the electric and magnetic field frequencies from the energy difference of the inter - molecular and contact - molecule electronic transitions as a mechanism to pump energy out of the molecular vibration . naturally , cooling is most efficient when the detuning frequency coincides with the frequency of the vibration .
additional possibilities to cool the current carrying molecular junction are discussed in ref . .
here we present numerical results in order to illustrate the transport properties of our model for the spin field - effect transistor sketched in fig .
[ scheme ] .
we use the level width @xmath90 due to coupling to a contact as unit of energy . unless explicitly specified otherwise , the parameters for calculations are temperature in the leads @xmath91 , the positions of the levels in absence of external fields @xmath92 , escape rates to the contacts ( wide - band approximation is employed ) @xmath93 , strength of coupling to external _
dc _ magnetic field @xmath94 , strength of the coupling to the external _ ac _ magnetic field @xmath95 and its frequency @xmath96 , strength of the coupling to the external electric field @xmath97 .
the parameters for the molecular vibrations are indicated in each case separately .
( color online ) elastic transport .
spin current vs. ( a ) _ ac _ magnetic field frequency @xmath98 ( solid line , blue ) and ( b ) electric field frequency @xmath9 .
the latter shows results for @xmath99 ( dotted line , black ) , @xmath100 ( dashed line , magenta ) , and @xmath101 ( solid line , red ) at @xmath96 .
also shown curve for @xmath97 at @xmath102 ( dash - dotted line , blue ) .
dashed line ( black ) in ( a ) shows charge current . see text for other parameters . ]
( color online ) inelastic transport .
spin current vs. ( a ) _ ac _ magnetic field frequency @xmath98 and ( b ) electric field frequency @xmath9 .
the latter shows results for symmetric @xmath103 ( dashed line , red ) and asymmetric @xmath104 , @xmath105 ( solid line , blue ) cases .
electron - vibration coupling is @xmath106 .
other parameters are as in fig .
[ elast ] . ]
figure [ elast ] shows the spin current @xmath107 in the elastic transport regime . in the calculations ,
the shift of the levels due to changes in @xmath98 is assumed to be compensated by the _ dc _
magnetic field , so that levels are set as in fig .
[ scheme ] and do not move .
the spin current ( solid line ) vs. the external _ ac _ field frequency ( spin bias ) is plotted in fig .
[ elast]a .
note that the electric field is an additional source of energy in the model , so that one will observe spin current even without spin bias in the contacts .
the sign of the spin current shows the direction of the spin - up flux ( direction from left to right is chosen as positive ) .
the flux switches around @xmath108 , as is shown in fig .
[ elast]a .
the point of sign change is defined by a competition between the spin bias and the electric field mediated inter - molecular transfer . at @xmath102
the lower spin levels are occupied , while the higher levels are empty ( see fig . [ scheme ] ) . in this regime
the electric field facilitates the spin - up flux from right to left , which defines a negative sign of the spin flux .
the spin bias grows with the frequency @xmath98 , and the value @xmath108 the two processes cancel each other .
when the spin bias grows further , the population of the levels changes .
the higher levels become populated due to the increased bias , while the population of lower levels diminishes ( the level goes above the corresponding spin - resolved chemical potential , see fig .
[ scheme ] ) .
the positions of the peaks are defined by the resonance condition for inter - molecule electron transfer : the position of the molecular levels is kept fixed by the _ dc _ field adjustment , while the frequency of the transition changes with both electric field frequency @xmath9 and _ ac _ magnetic field frequency @xmath98 , see eq.([vbar ] ) .
the condition for the resonance is @xmath109 .
we will return to the question of the role the spin bias plays in our discussion of the noise and heating properties .
note also , that the charge current ( dashed line in fig .
[ elast]a ) is identically zero ( the flux of spin up electrons from left to right is compensated by the flux of spin down electrons from right to left ) .
[ elast]b shows the dependence of the spin current on frequency of the external electric field .
naturally , the dependence has a maximum at resonance . shown
are plots for magnetic field frequency @xmath98 in the first ( dash - dotted line ) and second ( solid line ) maximum of fig .
[ elast]a . also shown
are curves for smaller ( dashed line ) and zero coupling to the electric field ( dotted ) . in the absence of coupling no spin current
is observed .
this indicates that , for the parameter range chosen , we are not in the regime of the spin pump described in ref . .
( color online ) noise properties .
( a ) zero - frequency charge and spin noise for @xmath110 ( dashed line , red ) and @xmath111 ( solid line , blue ) . vs. electric field frequency @xmath9 .
( b ) difference between spin and charge zero - frequency noise for @xmath95 ( dash - dotted line , red ) , @xmath112 ( solid line , magenta ) , and @xmath113 ( dashed line , blue ) .
see text for parameters . ]
figure [ inelast ] is inelastic analog of results presented in fig . [ elast ] .
molecular vibrations are taken into account employing small polaron transformation , and assuming separation of timescales , so that introduction of the franck - condon factors becomes possible .
[ inelast]a shows spin current vs. frequency of _ ac _ magnetic field .
parameters for molecular vibrations are @xmath114 , electron - vibration coupling is @xmath115 .
other parameters are as in fig .
[ elast ] .
an unusual form of the vibration sidebands is caused by resonance condition for intermolecular electron transfer rather than resonance situation at molecule - contact interface .
this makes vibrational characteristic different from the one presented in ref . .
at the same time , vibrational structure observed in spin current vs. electric field frequency resembles such for charge current vs. bias plots .
figure [ inelast]b shows two such characteristics : for symmetric @xmath103 ( dashed line ) and asymmetric @xmath116 ( @xmath104 and @xmath105 , solid line ) cases .
presence of a higher frequency naturally leads to observation of less vibrational sidebands .
figure [ noise ] displays the noise properties of the junction .
we show the results for elastic transport only , leaving the study of inelastic noise properties for a future publication .
[ noise]a shows the spin ( or charge ) zero - frequency noise as function of the external electric field frequency . note
that while the charge current is identically zero , charge noise does exist .
shown are the results for two situations : the spin levels of each molecule are well resolved ( dashed line , red ) vs. essential overlap between the two ( solid line , blue ) .
the first case corresponds to a situation with well pronounced resonance behavior . in such situations , the transmission probability at resonance approaches unity , which leads to suppression of the noise.@xcite the latter case corresponds to a situation when the transmission probability is lower . in this case
no noise suppression is observed at resonance .
the effect is similar to the behavior of a molecular junction in symmetric vs. asymmetric couplings at the two sides of the junction.@xcite in contrast to charge noise ,
the spin noise strongly depends on spin - flip events within the system.@xcite fig .
[ noise]b shows the difference between the spin and charge zero - frequency noise at several magnitudes of the coupling to the external _ ac _ magnetic field . note that at resonance , where the electron transport through the junction becomes pronounced , the difference between the spin and charge components diminishes .
( color online ) ` temperature ' of the first @xmath2 ( dashed line , red ) and second @xmath3 ( solid line , blue ) molecular vibration vs. ( a ) _ ac _ magnetic field frequency @xmath98 and ( b ) external electric field frequency @xmath9 .
see text for parameters . ] as was mentioned already earlier , there is no net charge transport between the leads .
nevertheless , charge transfer between contacts and molecule does occur .
thus , the spin field - effect transistor may be heated . as was discussed in the literature ( see e.g. refs . ) , charge transfer by itself does not necessarily lead to heating of the molecular device . here , we demonstrate one such possibility .
cooling of a molecular vibration is caused by tuning the external field frequency out of resonance ( an analog of atomic optical cooling ) .
figure [ tv]a shows the _ temperatures _ of the first @xmath2 ( dashed line ) and second @xmath3 ( solid line ) molecular vibrations as functions of the spin bias .
the coupling to the _ dc _ magnetic field is @xmath117 , whereas the frequency of the electric field is @xmath118 .
the molecular vibrations are taken to be @xmath119 and @xmath120 with electron - vibration couplings @xmath121 .
other parameters of the calculation are introduced at the start of the section .
the parameters ( electric and magnetic field frequencies ) are chosen in such way that the most probable ( resonant ) electronic tunneling through the junction has to go uphill in energy at the contact - molecule and inter - molecule transition steps . in this case , the tunneling energy difference required for resonant electron transition is taken from the molecular vibration , which leads to cooling of the device .
[ tv]b shows similar behavior when the frequency of the electric field is used as a control instead .
the molecular vibrations are @xmath122 and @xmath123 .
here cooling takes place at the molecule - molecule interface only and is not as effective .
however , one can achieve a stronger signal in this case .
the efficiency of the cooling at maximum spin current is an interesting question for future studies .
spintronics is a quickly developing field of research ( both experimental and theoretical ) . here , we studied the spin current transport properties within a model for an all - electric controlled spin field - effect transistor .
the model provides pure spin currents through a junction which consists of two spin - pumps ( molecules under influence of external _ dc _ magnetic fields)@xcite each attached to its normal metal contact .
the two parts are coupled by an external electric field , which serves as a control for the spin flux . within a full counting statistics
approach to multi - level systems , we discuss elastic and inelastic spin flux and noise properties of the junction .
external electric and magnetic fields are indicated as possible controls of the spin current through the junction .
the charge and spin zero - frequency noises are shown to be different when spin - flip processes within the junction become pronounced .
zero - frequency spin noise as function of the external electric field shows the same single to double peak structure transition when the tunneling probability approaches unity .
similar behavior was observed earlier for a model of charge field - effect transistor.@xcite problem of molecular junction heating recently discussed in the literature in connection to charge transport , retains its importance also for spin molecular devices . within the model , we discuss possibility to use external fields slightly de - tuned from the molecular resonances for cooling the molecular vibrations .
this process is similar to optical cooling of atoms .
we find that external _
ac _ magnetic fields may be effective cooling media .
external electric fields also provides the effect .
the efficiency of the cooling at maximum spin flux is an interesting question for future studies .
thanks ucsd for its hospitality during his visit in july 2009 , and the swedish research council and the royal swedish academy of sciences for financial support .
m.g . gratefully acknowledges support by the ucsd ( startup funds ) , the uc academic senate ( research grant ) , and the u.s .- israel binational science foundation .
here we derive an expression ( [ se ] ) for the self - energy due to the coupling to the external electric field within the rotating - wave approximation .
we start by partitioning the total hamiltonian of the system into a zero - order hamiltonian @xmath124 and perturbation @xmath125 , eq.([vbar ] ) .
the first is given by an analog of eq.([h0 ] ) after transformation to rotating frames .
it will be similar to ( [ h0 ] ) with molecular and contact states energies renormalized according to ( [ eibar ] ) and ( [ ekbar ] ) , respectively , and without the time dependence of magnetic field terms . for the moment
, we disregard the coupling to molecular vibrations .
hamiltonian @xmath124 defines the zero - order gfs of the system @xmath126 .
we treat the interaction @xmath125 by perturbation theory , i.e expanding the evolution operator on the keldysh contour ( interaction representation ) @xmath127 $ ] to second order in @xmath128 .
this leads to a dyson type equation for the green function @xmath129 with matrix elements for the self - energy given by @xmath130 \nonumber\end{aligned}\ ] ] where @xmath131 numerate molecule in the junction , whereas @xmath132 means opposite of @xmath133 .
application of the rotating wave approximation leaves only terms proportional to @xmath134 and enforces @xmath135 , in ( [ se1 ] ) .
the resulting expression is presented in eq.([se ] ) . in the case of the small polaron transformation ,
the coupling to molecular vibrations is treated by dressing the expression ( [ se ] ) by franck - condon factors @xmath136 when perturbation theory is employed , the gf in eq .
( [ g1 ] ) is considered as a zero - order gf of the born approximation .
99 i.uti ' c , j.fabian , and s. das sarma .
phys . * 76 * , 323 ( 2004 ) .
r.hanson , l.p.kouwenhoven , j.r.petta , s.tarucha , and l.m.k.vandersypen .
79 * , 1217 ( 2007 ) .
m.e.flatt ' e. nature physics * 4 * , 587 ( 2008 ) .
m.braun and g.burkard .
* 101 * , 036802 ( 2008 ) . j.j.krich and b.i.halperin .
b * 78 * , 035338 ( 2008 ) .
s.amasha , k.maclean , i.p.radu , d.m.zumbuhl , m.a.kastner , m.p.hanson , and a.c.gossard .
* 100 * , 046803 ( 2008 ) .
p.debray , s.m.s.rahman , j.wan , r.s.newrock , m.cahay , a.t.ngo , s.e.ulloa , s.t.herbert , m.muhammad , and m.johnson .
nature nanotechnology * asap * ( 2009 ) .
j.berezovsky , m.h.mikkelsen , n.g.stoltz , l.a.coldren , and d.d.awschalom .
science * 320 * , 349 ( 2008 ) .
f.kuemmeth , s.ilani , d.c.ralph , and p.l.mceuen .
nature * 452 * , 448 ( 2008 ) .
d.press , t.d.ladd , b.y.zhang , and y.yamamoto nature * 456 * , 218 ( 2008 ) .
j.r.hauptmann , j.paaske , and p.e.lindelof .
nature physics * 4 * , 373 ( 2008 ) .
c.galland and a.imamoglu .
* 101 * , 157404 ( 2008 ) .
f.dubin , m.combescot , g.k.brennen , r.melet .
lett * 101 * , 217403 ( 2008 ) .
j.r.petta , s.k.slater , and d.c.ralph .
lett . * 93 * , 136601 ( 2004 ) .
a.r.rocha , v.m.garcia-suarez , s.w.bailey , c.j.lambert , j.ferrer , and s.sanvito .
nature materials * 4 * , 335 ( 2005 ) .
s.sanvito and a.r.rocha .
nanoscience * 3 * , 624 ( 2006 ) . l.bogani and w.wernsdorfer .
nature materials * 7 * , 179 ( 2008 ) .
b.wang , j.wang , and h.guo .
b * 67 * , 092408 ( 2003 ) . j.peng and z.chen .
physics lett .
a * 365 * , 505 ( 2007 ) .
j.fransson and j .- x.zhu .
b * 78 * , 113307 ( 2008 ) .
n.jean and s.sanvito .
b * 73 * , 094433 ( 2006 ) . j.fransson
. nano lett . *
9 * , 2414 ( 2009 ) .
b.wang , j.wang , and h.guo .
b * 69 * , 153301 ( 2004 ) .
h.a.nilsson , p.caroff , c.thelander , m.larsson , j.b.wagner , l .- e.wernersson , l.samuelson , and h.q.xu .
nano lett .
* 9 * , 3151 ( 2009 ) .
p.zhang , q .- k.xue , and x.c.xie .
* 91 * , 196602 ( 2003 ) .
_ many - particle physics_. kluwer academic / plenum publishers ( 2000 ) .
m.galperin , a.nitzan , and m.a.ratner .
b * 76 * , 035301 ( 2007 ) .
m.galperin , m.a.ratner , and a.nitzan . j.
* 121 * , 11965 ( 2004 ) .
l.p.kadanoff and g.baym .
_ quantum statistical mechanics _
w. a. benjamin , inc .
( 1962 ) . l.s.levitov and g.b.lesovik .
jetp lett .
* 58 * , 230 ( 1993 ) .
l.s.levitov , h.lee , and g.b.lesovik . j. math
37 * , 4845 ( 1996 ) .
yu.v.nazarov . ann .
( leipzig ) * 16 * , 720 ( 2007 ) .
a.o.gogolin and a.komnik .
b * 73 * , 195301 ( 2006 ) .
d.djukic and j.m .
van ruitenbeek .
nano lett .
* 6 * , 789 ( 2006 ) . t.l.schmidt and a.komnik .
b * 80 * , 041307(r ) ( 2009 ) . r.avriller and a. levy yeyati . phys .
b * 80 * , 041309(r ) ( 2009 ) .
f.haupt , t.novotny , and w.belzig .
phys . rev .
* 103 * , 136601 ( 2009 ) .
note that eq.([du ] ) is equivalent to eq.(1 ) of ref . .
y.meir and n.s.wingreen .
* 68 * , 2512 ( 1992 ) .
h.haug and a .- p.jauho .
_ quantum kinetics in transport and optics of semiconductors_. springer ( 2008 ) . deriving ( [ sk1k2 ] ) from ( [ du ] )
we use mathematical matrix identity @xmath137 m.galperin , a.nitzan , and m.a.ratner .
b * 74 * , 075326 ( 2006 ) .
m.galperin , m.a.ratner , and a.nitzan . j.
phys . * 130 * , 144109 ( 2009 ) . m.galperin and a.nitzan . j.
phys . * 124 * , 234709 ( 2006 ) .
m.galperin , a.nitzan , and m.a.ratner .
b * 75 * , 155312 ( 2007 )
. d.leibfried , r.blatt , c.monroe , and d.windeland .
phys . * 75 * , 281 ( 2003 ) .
m.galperin , k.saito , a.v.balatsky , and a.nitzan .
b * 80 * , 115427 ( 2009 ) .
y.m.blanter and m.buttiker .
physics reports * 336 * , 1 ( 2000 ) .
m.zareyan and w.belzig .
. lett . * 70 * , 817 ( 2005 ) .
z.huang , f.chen , r.dagosta , p.a.bennett , m. di ventra , and n. tao , nat
. nanotechnol .
* 2 * , 698 ( 2007 ) . | we consider a model for a spin field - effect molecular transistor , where a directed pure spin current is controlled by an external electric field .
inelastic scattering effects of such molecular device are discussed within a framework of full counting statistics for a multi - level molecular system .
we propose that the heating of the molecular junction can be controlled by external electric and magnetic fields .
characteristic features of the model are demonstrated by numerical calculations . |
the advent of modern x - ray observatories such as _ chandra _ and _ xmm - newton _ has produced a spectacular increase in both the quantity and the quality of the observations of type ia supernova remnants ( snrs ) . yet , these excellent observations have led only to a modest improvement in our knowledge of the physics of type ia supernovae ( sne ) .
important issues such as the nature of the progenitor systems , the last stages of their evolution prior to the sn explosion or the physical mechanism behind the explosion itself still remain obscure ( see * ? ? ?
* ; * ? ? ?
* for reviews ) .
in a recent paper ( * ? ? ? * henceforth paper i ) , we examined the prospects for the identification of the explosion mechanism in type ia sne through the analysis of the x - ray spectra of young snrs .
we assembled a grid of type ia sn explosion models , simulated their interaction with a uniform ambient medium ( am ) and calculated the predicted x - ray spectra from the ensuing snrs .
the calculated x - ray snr spectra varied dramatically from model to model , demonstrating that it is possible to use young type ia snrs to probe the details of the type ia sn explosion mechanism . in this paper ,
we expand the results that were introduced in paper i. our objective is to examine the relationship between type ia sn explosions and the x - ray spectra of their snrs within the framework of hydrodynamic , ionization , and spectral simulations . by comparing our models with observations , we aim at improving our understanding of both type ia sne and young , ejecta - dominated snrs . in [ sec : techniques ] , we review the simulation scheme used in paper i , and we discuss the influence of two important parameters which we had not hitherto explored : the amount of collisionless electron heating at the reverse shock and the density of the uniform am that interacts with the ejecta . in [ sec : expanding ] we examine the performance of recent 3d type ia sn explosion models in the context of snrs , and we discuss the ability of these 3d models to reproduce the fundamental properties of the x - ray spectra of type ia snrs .
our conclusions are presented in [ sec : discussion ] . in order to facilitate the comparison between our models and x - ray observations of snrs , we have generated a library of synthetic spectra .
this library is presented and discussed in the appendix . in a forthcoming paper ( badenes et al .
, in preparation ) , we will make a detailed comparison between our models and the x - ray spectrum of the tycho snr .
although the x - ray spectra of young type ia snrs contain much information about the structure and composition of the material ejected by the sne that originated them , this information is generally difficult to extract .
the ejecta material consists almost entirely of heavy elements that are impulsively heated to x - ray emitting temperatures as the reverse shock propagates inwards in the reference frame of the expanding ejecta .
the propagation of the reverse shock is in turn intimately related to the density structure of the ejecta , which results in an intricate dynamical behavior of the snr early in its evolution ( see * ? ? ?
if there is a significant degree of stratification in the elemental composition of the ejecta , different chemical elements are shocked at different evolutionary times , after different periods of free expansion , and therefore emit x - rays under different physical conditions .
this results in a very complex spectrum , which is hard to model and interpret .
the approach taken in paper i was based on a grid of 1d type ia sn explosion models that included all the mechanisms currently under debate for the single degenerate type ia sn scenario : deflagrations , delayed detonations , pulsating delayed detonations and sub - chandrasekhar explosions .
the dynamics of the interaction of each explosion model with a uniform ambient medium ( am ) of density @xmath0 was followed with a 1d hydrodynamic code .
the dynamic evolution of each fluid element in the shocked ejecta ( i.e. , the time evolution of density @xmath1 and specific internal energy per unit mass @xmath2 ) , together with its chemical composition as determined by the sn explosion model , were used as the input to ionization calculations .
these calculations included the interactions between ions and electrons in the shocked plasma , and they provided time - dependent nonequilibrium ionization ( nei ) states and electron temperature . using these ionization states and electron temperatures , spatially
integrated synthetic x - ray spectra were generated with a spectral code by adding the weighted contributions from each fluid element in the shocked ejecta . for a more detailed explanation ,
see paper i and the references therein . within this simulation scheme , the x - ray spectrum from the shocked ejecta is determined by : ( 1 ) the density and chemical composition profiles of the sn ejecta from the explosion model , ( 2 ) the age of the snr , ( 3 ) the amount of collisionless electron heating at the reverse shock , and ( 4 ) the density of the uniform am . in paper
i , we analyzed the importance of ( 1 ) and ( 2 ) ; the impact of ( 3 ) and ( 4 ) is the focus of this section .
the unknown efficiency of collisionless electron heating in snr shocks is one of the main uncertainties affecting the calculated x - ray spectra of snrs .
direct application of the rankine - hugoniot relations at the shock front yields @xmath3 for each population of particles @xmath4 , where @xmath5 is the particle mass , @xmath6 is the shock velocity and @xmath7 is boltzmann s constant . because of the large difference between electron and ion masses ,
the electrons are expected to be much colder than the ions , and the quotient of postshock specific internal energies defined as @xmath8 is expected to be close to 0 , where @xmath9 is the preshock ionization state ( i.e. , the number of free electrons per ion in the unshocked ejecta ) .
however , @xcite argued that plasma waves can redistribute energy among cold electrons and hot ions at the shock , bringing the value of @xmath10 close to @xmath9 ( for a discussion of collisionless electron heating see * ? ? ?
so far , the observational evidence hints at a decreasing level of thermal equilibration with increasing shock speeds or mach numbers in the forward shocks of snrs ( see * ? ? ? * and references therein ) . in the forward shock of tycho
, @xcite found a value of @xmath11 by analyzing the optical balmer emission , while @xcite estimated a much lower value at the forward shock of sn1006 from x - ray observations .
the only constraint on the value of @xmath10 in the reverse shock of a young type ia snr comes from the absorbed spectrum of the schweizer and middleditch star behind sn1006 , where the amount of thermal energy deposited in the electrons was found to be negligible @xcite .
the model spectra presented in paper i were calculated assuming no collisionless electron heating at the reverse shock , effectively setting @xmath10 to the lowest possible value , @xmath12 , where @xmath13 is the average ion mass in a fluid element . it is clear from the works cited above that , while full thermal equilibration between ions and electrons at the shock ( i.e. , @xmath14 ) is not compatible with the observations , values of @xmath10 larger than @xmath15 can not be excluded .
the effect of varying amounts of collisionless electron heating at the forward shock on the x - ray spectrum emitted by the shocked am of snrs in the sedov stage was discussed in @xcite ; here , we shall analyze the impact of a small ( but nonzero ) amount of collisionless electron heating at the reverse shock on the properties of the shocked ejecta .
we will illustrate this effect using a delayed detonation model as an example .
of all the classes of one - dimensional type ia explosion models , delayed detonations have been the most successful in reproducing the light curves and spectra of type ia sne @xcite , and therefore it is of much interest to analyze the details of the x - ray emission that these models predict for type ia snrs . within the delayed detonations , we chose model ddte because it has the largest amount of intermediate mass elements ( si , s , etc . ) in the ejecta .
this should make it easier to estimate the effects that collisionless electron heating at the reverse shock has on the prominent x - ray lines from these elements .
figure [ fig-1 ] shows the shocked ejecta of model ddte for a uniform am of density @xmath0 , 430 yr after the explosion ( the age of the tycho snr ) .
the electron heating and plasma ionization processes in the shocked ejecta have been calculated for @xmath16 , @xmath17 , and @xmath18 .
the unshocked ejecta were assumed to be singly ionized in all cases .
as discussed in paper i , the interaction of ejecta with the am leads to the formation of density structures within the shocked ejecta , which strongly affect the distributions of ionization states ( represented here by the average ion charge @xmath19 ) , electron temperatures @xmath20 , and ionization timescales ( @xmath21 ) . together with the chemical composition profile of the ejecta
, these distribution functions determine the spectral properties of each element , and ultimately the shape of the emitted x - ray spectrum
. an important feature of the shocked ejecta is the pronounced density peak towards the contact discontinuity ( cd ) that appears in all type ia snr models ( see paper i and * ? ? ?
in the case with no collisionless heating ( @xmath16 ) , the electron temperature profile rises monotonically from the reverse shock to the cd , as internal energy is gradually redistributed from the hot ions to the cold electrons through coulomb collisions .
the electron temperature profile peaks at the cd , where the fluid elements have been shocked for the longest time and have the highest density ( the rate at which the ion and electron temperatures equilibrate in the shocked ejecta scales with @xmath1 , see eq .
1 in paper i ) .
increasing the value of @xmath10 makes the electrons just behind the reverse shock hotter , but the electron temperature drops as numerous cold electrons are liberated in the ongoing ionization process and the total internal energy in the electrons is redistributed among more particles . for @xmath22 ,
the electron temperature profile eventually relaxes to the profile obtained without any collisionless heating at the reverse shock , but in the @xmath23 case there is a significant residual temperature excess even in the outermost ejecta layers .
the average ionization state and ionization timescale become severely affected only for @xmath24 , when the electrons reach extremely high(@xmath25 k ) temperatures behind the reverse shock .
the ionization process is less efficient at these extreme temperatures , leading to lower mean ion charges and ionization timescales in the shocked ejecta .
the significance of this modification in the electron temperature profile is better understood when viewed in the context of the stratification that is inherent to 1d type ia sn explosion models . in the case of ddte , as in all delayed detonation models , the inner ejecta layers are dominated by fe and ni , surrounded by a region rich in intermediate mass elements ( mostly si and s , but also ar , ca , and others ) , with o dominating the outermost ejecta layers . in figure
[ fig-1 ] , this stratified structure has been represented schematically in panel b. in this example , the increase in @xmath20 caused by collisionless electron heating at the reverse shock affects primarily the fe - rich ejecta layers at this age . in terms of the emission measure - averaged electron temperature for each element
x , @xmath26 , increasing the value of @xmath10 effectively reverses the approximate ordering in @xmath20 of the ejecta elements that is maintained throughout the evolution of the snr .
this is illustrated in figure [ fig-2 ] , which can be contrasted with figure 5 in paper i. in all the calculations presented here and in paper i , @xmath9 has been set to 1 , which is generally a good approximation for nei plasmas in snrs .
photoionization by uv starlight or by x - rays emitted by the shocked material in the snr could raise @xmath9 , but only by a factor of 34 ( see * ? ? ?
after the shock , the heavy element plasma ionizes rapidly , and the values of @xmath27 in fluid elements that started with different @xmath9 will converge over time .
due to this , the postshock x - ray emission is generally insensitive to moderate variations in the preshock ionization . in the presence of collisionless electron heating ,
increasing @xmath9 for a fixed value of @xmath10 raises the number of hot electrons and decreases their temperature in the same proportion .
x - ray spectra , however , are sensitive mostly to the total internal energy transferred to the electrons at the shock and to the final electron temperature , and not so much to how the internal energy is distributed among the electrons . for all the values of @xmath9 ,
the final electron temperature is similar because the final value of @xmath19 is very similar . in view of this
, we do not expect significant deviations from the results presented here for @xmath28 at a fixed value of @xmath10 .
the density of the am affects the spectral properties of the elements in the shocked ejecta in a dramatic way .
this is due to two closely related effects : the acceleration of all the collisional plasma processes in denser media on one hand and the scaling of the hydrodynamic models with @xmath29 on the other hand ( see section 4.1 in paper i for a discussion of the hydrodynamic scaling ) .
these effects are illustrated in figures [ fig-3 ] and [ fig-4 ] , which display the structure of the shocked ejecta of model ddte at an age of 430 yr after the explosion , for a uniform am of @xmath30 and @xmath31 , respectively .
these simulations are also shown for @xmath16 , @xmath17 , and @xmath18 to facilitate comparison with figure [ fig-1 ] . as a result of the scaling laws mentioned in paper i ,
a snr in a denser am will be in a more evolved evolutionary stage at any given time , and vice versa note how the reverse shock has not reached the fe - dominated region of the ejecta at @xmath32 yr for @xmath31 .
the mean ionization state in the shocked ejecta , which peaks at @xmath33 in the outermost si - dominated layers for @xmath34 , rises as high as @xmath35 for @xmath30 in the same region , but only reaches @xmath36 for @xmath31 .
these differences in the mean ionization state correspond to differences in the ionization timescales of roughly an order of magnitude throughout the shocked ejecta for each factor 5 increase in @xmath29 .
the electron temperature profiles are affected as well , although only by a factor of @xmath37 . the higher densities and faster ionization rates tend to mitigate the effect of collisionless electron heating , favoring the convergence to the canonical @xmath16 case ( compare panels c of figures [ fig-3 ] and [ fig-4 ] ) . varying the value of @xmath29 has an immediate impact on the emission measures and emission measure averaged quantities , mainly through the hydrodynamic scaling laws mentioned in paper i. the approximate scaling of @xmath38 and @xmath39 is given by eqs .
( 3)-(5 ) in paper i : @xmath40 for the t axis , @xmath29 for @xmath38 and @xmath41 for @xmath39 .
these approximate scalings are accurate within a factor of 2 for @xmath42 , but they might break down for values of @xmath29 outside this range .
the effect of a change of @xmath29 on the electron temperatures is more complex , and difficult to approximate with sufficient accuracy in view of the sensitivity of x - ray spectra to electron temperature . whereas an accuracy within a factor of 2 is reasonable for @xmath43 and @xmath44 , which span several orders of magnitude , changes by a factor of 2 are too large compared with the more modest ( but still up to 2 orders of magnitude ) range in @xmath20 . in figure
[ fig-5 ] , the temporal evolution of the spectra from the shocked ejecta of model ddte is presented for three values of @xmath29 ( @xmath45 , @xmath46 , and @xmath47 ) and two values of @xmath10 ( @xmath15 and @xmath18 ) .
a preliminary inspection reveals that variations in @xmath29 have profound effects on the calculated spectra .
as expected from the scaling law for @xmath48 , the plasma ionization state varies greatly , and the presence of different ions results in emission of different ionic lines . at @xmath30 , for instance , the more advanced ionization state of fe leads to a higher flux in the fe l complex , which blends with o ly@xmath49 and mg he@xmath49 emission at the _ xmm - newton _ ccd spectral resolution .
the increase in the fe k@xmath49 line , on the other hand , is due to the higher temperatures in the ejecta .
the prominent o he@xmath49 line at 0.56 kev , seen at early times for @xmath34 , disappears at higher values of @xmath29 , because he - like o is ionized more rapidly .
the overall higher ionization state of the plasma also leads to an increase in the flux of the ly@xmath49 lines of si and s , and a shift towards higher energies of the ca k@xmath49 line . the shape and flux of the continuum emission also change . at lower densities ,
these effects are reversed .
the fe k@xmath49 line and fe l complex virtually disappear , revealing the underlying ne he@xmath49 and ne ly@xmath49 lines at 0.9 and 1.0 kev . the o he@xmath49 line becomes more important than o ly@xmath49 , and the ly@xmath49 and he@xmath10 lines of si and s vanish almost completely , as well as the ca k@xmath49 line .
the continuum flattens and the emitted flux is generally lower at all energies .
in contrast with the global effects of variations of @xmath29 , changes in the amount of collisionless heating at the reverse shock have a different impact on different elements in a model with stratified ejecta like ddte . for @xmath34 , the flux in the fe k@xmath49 line , which probes material at higher @xmath20 and lower @xmath50 than the fe l complex ,
is increased by almost two orders of magnitude for @xmath51 .
none of the other elements seems to be affected at this am density , although model ddte has a significant amount of s , si and ca in the inner ejecta .
this increase in the fe k@xmath49 flux becomes less pronounced with time , and is accompanied by a slight change in the shape of the continuum . for @xmath52 , the continuum is unaffected and the increase of the fe k@xmath49 line flux is reduced to less than an order of magnitude at early times , disappearing completely at late times . at @xmath31 , however , the collisionless electron heating has a more noticeable effect .
the shape of the spectrum is not changed at low energies , but the flux is somewhat lower at early times for @xmath23 . at high energies ,
the level of continuum rises and the flux in the fe k@xmath49 line flux greatly increases .
the effects of collisional electron heating can be clearly seen even at ccd spectral resolution . with a higher spectral resolution such as provided by _
astro - e2 _ , the predicted large temperature increases caused by collisionless heating ( fig .
[ fig-2 ] ) should be detectable through various temperature - sensitive line diagnostics for a number of different chemical elements within the shocked ejecta .
we emphasize that model ddte is presented here just as an illustrative example . for obvious reasons , it is not practical to present the effects of @xmath29 and @xmath10 on the spectra of all the models in our grid .
although the details may vary , the general trends identified here for ddte can be applied to most of the other models ( for a discussion of other delayed detonation models , see * ? ? ?
* ) . to conclude
, we note that collisionless electron heating at the reverse shock can have interesting effects on the spatially resolved x - ray emission . in particular , the enhanced flux in the fe k@xmath49 line discussed above would come mainly from the hotter regions of fe - rich ejecta close to the reverse shock ( see the shape of the electron temperature profile in figures [ fig-1 ] , [ fig-3 ] , and [ fig-4 ] for values of @xmath10 above 0.01 ) .
this scenario is compatible , at least qualitatively , with the finding that the fe k@xmath49 emission peaks at a smaller radius than the fe l and si he@xmath49 emission in the x - ray ccd images of both the tycho @xcite and kepler @xcite snrs .
collisionless electron heating provides a simpler explanation to the rise of the electron temperature profile towards the reverse shock than the relic of an interaction with a circumstellar medium invoked by @xcite for the tycho snr .
in paper i , we introduced a grid of eight one dimensional type ia sn explosion models that included examples of all the paradigms currently under debate : sub - chandrasekhar explosions , deflagrations , delayed detonations , and pulsating delayed detonations .
this reduced grid is just a representative sample of a more extensive grid of 19 models , which constituted the base for the study of type ia snrs conducted by @xcite .
the remaining 11 grid models are intermediate cases obtained by varying the parameters involved in the calculation of each explosion paradigm .
we have included these models in the appendix , both for reference in future works and for the convenience of those readers who want to use our synthetic snr spectra for their own research .
this grid is one representation of our current understanding of one dimensional type ia explosion models , upon which most of our knowledge of the physics of type ia sne is based . in view of the recent developments in the field ,
however , it has become clear that 1d calculations will soon be superseded by the three dimensional models that have begun to appear in the literature @xcite .
these works have focused on pure deflagrations in 3d , proving that they are capable of producing robust explosions , but the ability of these models to explain the observations of type ia sne has not been fully established yet .
a common feature in all 3d deflagration models , and the most remarkable difference with respect to 1d models , is the uniform mixing of unburnt c and o material with @xmath53ni and the other products of nuclear burning throughout the ejecta .
this mixing is due to the deformation of the flame front caused by rayleigh - taylor instabilities , an effect which seems unavoidable in 3d deflagrations .
there has been some concern that the presence of large amounts of c and o in the inner layers of ejecta would lead to a spectral evolution inconsistent with optical observations @xcite , but complex spectral simulations are required to verify this claim @xcite .
alternatives to the 3d deflagration scenario are being explored right now , including delayed detonations in 3d @xcite and two new explosion paradigms : gravitationally confined detonation @xcite and pulsating reverse detonation @xcite , but none of these models has been completely understood yet . here , we study 3d deflagration models in the context of the x - ray spectra of young snrs to provide an independent method of assessing their viability for type ia explosions .
we use a one dimensional average of model b30u , a 3d deflagration from @xcite , to illustrate what can be expected from this class of models .
the chemical composition and density profile of this model are presented in the appendix , and they are very similar to those of the models obtained by @xcite and @xcite , even though the computational techniques and the resolution of the calculations are different in each case ( see table 1 in * ? ? ? * for a more detailed comparison of these works ) .
this shows that three dimensional deflagrations are relatively well understood , and supports our use of model b30u as a representative example of this class .
the evolution of the emission measures and emission measure - averaged ionization timescales and electron temperatures of the principal elements in the ejecta of b30u are shown in fig .
[ fig-6 ] for an interaction with @xmath34 and @xmath16 .
interpretation of differences between these plots and figures 4 , 5 , and 6 of paper i is not straightforward , because the 3d calculations are not fully self - consistent with the 1d models of the grid ( the effect of the energy deposited by the decay of @xmath53ni on the density profile , for instance , has not been taken into account ) .
nevertheless , the main features of the evolution of the shocked ejecta do not depend on such details .
the most striking property of this model is the similarity in the spectral properties ( @xmath54 and @xmath55 ) of fe and si throughout the evolution of the snr .
this is in marked contrast to 1d models , where the stratification of the ejecta leads to significant differences between fe and si .
the abrupt changes in several plots seen at @xmath56 s are due to the impact of the reverse shock on a remnant of unburnt white dwarf material formed in model b30u ( for details , see * ? ? ?
* ) . in figure [ fig-7 ]
, we show the ejecta spectra of model b30u at the same values of @xmath57 , @xmath29 and @xmath10 as in figure [ fig-5 ] for model ddte .
the most remarkable properties of these spectra are the high fe l - shell flux and the presence of the prominent ni k@xmath49 line at @xmath58 kev ( except at the lowest am densities ) .
this is due to the large amounts of fe and ni that are found in the outermost layers of b30u , where the density of the shocked ejecta is highest .
the results are a long fe ionization timescale , which leads to the enhanced fe l - shell flux , and a high ni emission measure leading to a strong ni k@xmath49 emission .
such a strong ni k@xmath49 line has never been observed in thermal x - ray spectra of snrs .
another interesting feature is the relative weakness of the si and s k@xmath49 lines .
the reason for this is twofold : first , 3d deflagrations produce smaller amounts of si , s and other intermediate mass elements than the conventional 1d delayed detonations ; second , equivalent widths of the si and s lines are smaller because of the strong continuum produced by the large amount of c and o that is present throughout the ejecta .
these spectral characteristics exhibited by model b30u are common to all 3d deflagrations with well mixed ejecta .
we compare the results of our simulations for the 3d deflagration model b30u with the basic properties of type ia snrs .
the prediction of similar emission measure - averaged electron temperatures and ionization timescales for si and fe can be easily tested by examining x - ray observations .
we have searched the literature for young snrs with published good - quality x - ray spectra that have been classified as type ia .
six objects meet these requirements : the historical remnants of tycho , kepler and sn1006 , and three large magellanic cloud snrs : n103b , 0509 - 67.5 , and dem l71 .
we note that the classification of kepler s snr as type ia is controversial ( e.g. , * ? ? ?
the remnant of sn1006 is not suitable for our purposes because it lacks strong fe emission @xcite . in the case of dem l71 , although the x - ray spectrum of this snr has been analyzed in some detail ( see * ? ? ?
* ; * ? ? ?
* ) , we found no published estimates of temperatures and ionization timescales for fe and si in the ejecta .
the results of the analysis of the integrated spectrum for the other four snrs are summarized in table [ tab-1 ] .
the spectral properties of all the snrs considered here show that an important fraction of the fe in the shocked ejecta is emitting under conditions different from those of the si .
the authors of the works referenced in table [ tab-1 ] accounted for this by either adding a spectral component made of pure fe to their fits or by using plane - parallel models that allowed to treat fe and si separately by assigning different values of @xmath59 and @xmath60 to each element .
since the analysis techniques , models , and data quality were different in each case , these results can only be compared either with our models or among themselves in a qualitative way .
nevertheless , a clear trend can be observed in all four snrs considered here : the fe component was always hotter than the si component by at least a factor of 2 .
the fe was at a lower ionization timescale in three out of four objects : tycho , kepler , and n103b . in 0509 - 67.5 , however , the fe component has a higher ionization timescale than si . in this case , the statistics of the fe k@xmath49 line were poor , and the ionization timescale of fe was constrained mostly by fitting the fe l complex .
@xcite note that their fit to the fe l complex emission was not complete , because a strong line had to be added by hand . improved atomic physics and
higher resolution data would be highly desirable to confirm this result for 0509 - 67.5 .
since the emission measure averaged ionization timescales and electron temperatures of si and fe do not differ by more than 30% in model b30u ( see figure [ fig-6 ] ) , we conclude that this model is in conflict with the observations listed on table [ tab-1 ] , at least within the limitations of our simulations .
this conclusion is extensible to any model in which fe and si are well mixed throughout the ejecta , and therefore can be applied to all the 3d deflagration models for type ia sne discussed in the previous section . as we have seen in [
sec : techniques ] , a plasma state with @xmath61 and @xmath62 arises naturally in type ia sn models with stratified ejecta , such as 1d delayed detonations or pulsating delayed detonations , that undergo a moderate amount of collisionless electron heating at the reverse shock .
the ionization timescales of fe and si in 0509 - 67.5 are clearly incompatible with this scenario , but in this case @xcite found a very low amount of fe in the shocked ejecta , with fe to si abundance ratios below 0.07 .
while this is very difficult to interpret in the context of well - mixed type ia sn ejecta , it could be more easily explained if the reverse shock were just entering the fe - dominated region in stratified ejecta .
a detailed comparison of our models with this snr would be required to confirm this hypothesis .
we emphasize that our simulations based on one dimensional averages are too simple to rule out well - mixed 3d type ia sn explosion models .
we do not account for a number of processes that might result in the fe and si in the ejecta emitting under different conditions , like the ni bubble effect @xcite or the formation of clumps in the ejecta @xcite .
nevertheless , we find that the observations of type ia snrs seem easier to explain in the light of type ia sn explosion models with stratified ejecta .
in this paper , we have examined several important aspects of the x - ray spectral models for the ejecta in type ia snrs that were introduced in @xcite .
we have explored the impact of the amount of collisionless electron heating at the reverse shock , @xmath10 , and the density of the ambient medium , @xmath29 , on the integrated x - ray emission from the ejecta in type ia snr models of different ages .
we found that even small amounts of collisionless electron heating can modify the electron temperature profile inside the ejecta in a significant way , leading to a region of hot material at low ionization timescales close to the reverse shock . in the context of type ia
sn explosion models with stratified ejecta , this modified temperature profile can affect the emission from the inner layers rich in fe for a broad range of dynamical ages , increasing the flux in the fe k@xmath49 complex .
this could explain why the fe k@xmath49 emission peaks at smaller radii than fe l in both the tycho @xcite and kepler @xcite snrs .
the density of the am also has a strong impact on the x - ray emission from the ejecta . for higher values of @xmath29 ,
the snr is in a more advanced evolutionary stage at a given age , and the ionization timescale of the shocked ejecta increases significantly . at lower values of @xmath29 ,
the ionization timescales decrease and the evolutionary stage is less advanced .
we have provided approximate scaling laws to estimate these effects , and discussed their impact on specific emission lines and line complexes through an example .
we have also reviewed the fundamental properties of the recent deflagration models for type ia sne calculated in 3d , and their performance in the context of the x - ray spectra of snrs . using our 1d simulation scheme ,
we have shown that the mixing of fuel and ashes throughout the ejecta , which is a common feature of these 3d explosion models , results in all the elements in the shocked ejecta of the snr having very similar spectral characteristics .
in particular , the emission measure averaged ionization timescales and electron temperatures of elements like fe and si are always very close to each other .
this is in conflict with the observations of type ia snrs in our galaxy and the magellanic clouds , where the fe and si in the shocked ejecta are found to be emitting under different physical conditions . within the limitations of our 1d simulation scheme , these observations are easier to explain with type ia explosion models that have stratified ejecta than with models that have well mixed ejecta like 3d deflagrations .
we believe that our models represent a significant improvement over current methods of analyzing and interpreting the x - ray emission from the shocked ejecta in snrs . in order to facilitate the comparison between our models and observations , we have built a library of synthetic spectra , which is available from the authors upon request .
this library is presented in the appendix , where more type ia sn explosion models are introduced , and some aspects relevant to the comparison between the synthetic spectra and observations are discussed .
a detailed example of this kind of comparison in the framework of the ejecta emission from the tycho snr will be the subject of a forthcoming paper ( badenes et al .
, in preparation ) .
we wish to thank jack hughes and jessica warren for detailed discussions concerning 0509 - 67.5 .
we also acknowledge conversations with una hwang and martin laming on several aspects of the research presented here .
we are grateful to the anonymous referee for suggestions that helped to improve the quality of this manuscript .
this research has been partially supported by the cirit and mcyt in spain , through grants aya2000 - 1785 , aya2001 - 2360 , and aya2002 - 04094-c03 .
cb would like to acknowledge support from gencat ( grant 2000fi 00376 ) and ieec in barcelona , and from grant go3 - 4066x from sao at rutgers .
kjb is supported by nasa grant nag 5 - 7153 .
in this appendix , we introduce our library of synthetic spectra for the ejecta emission in type ia snrs .
the objective of this library is to provide observers with a complete set of synthetic spectra for the ejecta emission in snrs , calculated from an extensive grid of type ia sn explosion models , at different values of @xmath57 , @xmath29 , and @xmath10 . at present , our library includes more than 800 synthetic spectra in sequences like those presented in figures [ fig-5 ] and [ fig-7 ] for models ddte and b30u . for each model ,
we have generated synthetic spectra for several values in the ranges @xmath63 yr ; @xmath64 ; and @xmath65 . in
[ subsec : rationale ] , we discuss these synthetic spectra in the context of the tools that are currently used for the analysis of ejecta emission in snrs .
in [ subsec : comparing ] we comment on potential applications for our models . in
[ subsec : caveats ] , some important caveats and limitations of the models are discussed . finally , in [ subsec : grid ]
, we introduce a number of type ia sn explosion models that , together with those presented in paper i , complete the exploration of the parameter space for thermonuclear supernovae .
the spectral analysis of the ejecta emission in young snrs is a complex problem . despite the spectacular increase in the quality of the observations
, it has proved very difficult to extract the relevant physical parameters from these observations in a reliable way with the available tools .
a frequent approach involves the fitting of several more or less sophisticated nei components with varying abundances , electron temperatures , and ionization timescales to the observed x - ray spectra ( several examples have been cited in
[ subsec : can3ddoit ? ] ) .
the results of applying this approach are not unique , and frequently very hard to interpret , because average parameters ( like @xmath20 or @xmath59 ) are assigned to a plasma whose physical properties have an enormous dynamic range , and where different chemical elements often emit under different conditions ( see figures [ fig-1 ] , [ fig-3 ] , and [ fig-4 ] ) .
the determination of elemental abundances in the entire volume of shocked plasma , which is crucial for establishing the connection between the snrs and the sn explosions that originated them , is particularly unreliable when it is based on this approach .
often , nei models just provide estimates for the emission measures of the elements , under the assumption of a homogeneous composition , and the difference between the ratios of these emission measures and the true abundance ratios in the plasma can be several orders of magnitude ( see section 4.2 of paper i ) .
the synthetic spectra presented in paper i and in the present work open new possibilities for the interpretation of x - ray observations of type ia snrs . without claiming to include all the complex physical processes at play in young snrs ( see [ subsec : caveats ] ) , these synthetic spectra provide a much more accurate representation of the state of the shocked ejecta in young type ia snrs than the simple nei models currently available within software packages like xspec .
moreover , since the synthetic spectra are calculated from realistic sn explosion models , the connection between the observed spectrum and quantities like the explosion energy or the amount of each element present in the ejecta are easy to establish .
the trade off is that the comparison between our synthetic spectra and x - ray observations is not necessarily a straightforward procedure .
several strategies with varying degrees of sophistication can be followed to compare our models to observations .
a somewhat crude possibility is to focus on derived quantities like @xmath66 and @xmath67 , as we have done in
[ subsec : can3ddoit ? ] .
while this can lead to interesting results , it is far better to perform more direct spectral comparisons using the library of synthetic spectra that we present here .
the most effective way to apply this library will depend on the specific observational constraints for the snr under study .
in some cases , like the historical galactic snrs , the age will be known accurately , but the distance ( and hence the total integrated x - ray flux and the radius of the forward shock ) will be more uncertain .
in other cases , like the snrs in the magellanic clouds , the distance will be known , but the age will not .
reliable independent estimates for @xmath29 may or may not be available
. in each case , there will be more than one way to reduce the dimensionality of the problem . rather than providing a recipe
which may not be adequate for a specific case , we make here some suggestions which might prove useful in a more general context .
first , it is important to note that each synthetic spectrum is based on an underlying hydrodynamic model , so quantities such as the radius of the forward and the reverse shocks and their expansion parameters are available for each spectral model ( see figure 3 and eqns . ( 3 ) , ( 4 ) , and ( 5 ) in paper i ) . in principle , it is possible to reverse the problem , find out which hydrodynamic models agree better with the observations and thus reduce the number of synthetic spectra to consider . in doing so ,
however , the limitations of 1d adiabatic hydrodynamics must be considered ( see the following section ) .
second , the selection of a particular synthetic spectrum from our library to represent the ejecta component in an observed x - ray spectrum may not be trivial . the substantial uncertainties in the atomic data and the relative simplicity of the models with respect to real snrs
will probably make it impossible to attain a statistically valid fit .
synthetic spectra like ours are more vulnerable to these factors , because there is little room for self - adjustment , in contrast to conventional nei models with variable abundances .
if the emissivity of a particular line is underestimated in the spectral code used to generate our library , for instance , this can not be compensated by artificially enhancing the abundance of that particular element , as in a conventional nei model . yet , even if some specific details of the observed spectrum can not be reproduced , it is often possible to find a model whose overall characteristics are in reasonable agreement with the observations . under these circumstances ,
a procedure needs to be devised in order to measure the degree of success of a specific synthetic spectrum .
an example shall be provided in a forthcoming paper on the tycho snr ( badenes et al .
, in preparation ) .
our models are just a simplified representation of the complexity of young snrs , and their limitations have to be considered when making comparison with observations .
the crucial approximations were reviewed in sections 3.5 and 5 of paper i , but it is important to revisit several issues here .
the most important simplification is certainly the assumption of spherical symmetry .
any description of young snrs in the framework of 1-d models is necessarily incomplete , because it does not include important processes such as ejecta clumping and dynamic instabilities at the contact discontinuity between shocked ejecta and shocked am @xcite .
the degree of ejecta clumping is crucial , and it is clear that our 1d models ( and in particular , the distribution of @xmath50 and @xmath20 for each element ) would be invalidated if clumps with a large density contrast like those proposed by @xcite were to dominate the emission measure of the shocked ejecta in type ia snrs . in this case ,
gross inconsistencies are expected to emerge from comparison of 1-d models with observations .
the degree of ejecta clumping strongly affects the morphology of the x - ray emission , and examination of this morphology in type ia snrs should shed light on this issue .
multi - dimensional hydrodynamical simulations coupled with x - ray emission calculations could prove useful for this .
another important issue whose impact on the x - ray spectra is hard to estimate is the effect of cosmic ray acceleration at the shocks .
there is some indication that this process might affect the dynamics and x - ray spectra of the shocked am without significantly modifying those of the shocked ejecta @xcite , but more detailed simulations are needed to shed light on this question ( see * ? ? ? * for a discussion ) . to conclude this section , we comment on the importance of radiative losses , which have received some attention lately in the work of blinnikov et al .
( see * ? ? ?
* and references therein ) . in section 4 of @xcite
, it was shown that radiative losses always lead to catastrophic cooling in heavy element plasmas , driving the shocked material to infrared- and optically - emitting temperatures . because no optically emitting knots with a composition dominated by heavy elements
have been observed in kepler , tycho , or sn1006 ( see * ? ? ?
* ; * ? ? ?
* ) , we conclude that radiative losses are not dynamically important in young type ia snrs under usual conditions .
radiative losses are not included in our models in a self - consistent way , but we have extended the _ a posteriori _ monitoring of radiative losses described in section 3.5 of paper i to the more unfavorable case of @xmath30 .
our previous conclusions have been verified : radiative losses only affect the outermost layers of the models with the steepest ejecta density profiles .
the values of @xmath68 , as defined in paper i ( the time when the calculated losses exceed 10% of the specific internal energy in at least 5% of the ejecta mass ) for models with @xmath695,000 yr are provided in table [ tab-2 ] . in these models ,
our calculations for the properties ( density , electron temperature , ionization state , and x - ray emission ) of the layers that undergo radiative losses are not reliable close to or beyond @xmath68 , and some amount of infrared or optical emission should be expected from this region of the ejecta . the fact that such emission is not observed in kepler , tycho or sn1006 suggests that models which predict substantial radiative losses in the ejecta are in conflict with observations of these historical snrs . among the eight one dimensional type ia sn explosion models introduced in paper i , one was a sub - chandrasekhar explosion ( sch ) , one was a pure detonation ( det ) , two were pure deflagrations ( defa and deff ) , two were delayed detonations ( ddta and ddte ) , and two were pulsating delayed detonations ( pdda and pdde ) .
these explosion paradigms , and the details involved in the calculation of the models , are described in paper i ( section 2 and appendix ) . for the deflagrations , delayed detonations and pulsating delayed detonations , the models presented in paper i were extreme cases , obtained by considering the highest and lowest reasonable values of the parameters involved in each calculation . in the case of the deflagration models , the relevant parameter is @xmath70 , which controls the propagation velocity of the subsonic flame . for the delayed detonation and pulsating delayed detonation , the parameters are @xmath71 , which determines the density at which the transition from deflagration to detonation occurs , and @xmath72 , which determines the flame velocity in the deflagration stage .
all these parameters are defined in the appendix of paper i. by varying these parameters , we have generated four more deflagrations ( defb , defc , defd , and defe ) , four more delayed detonations ( ddtb , ddtbb , ddtc , and ddtd ) , and three more pulsating delayed detonations ( pddb , pddc , and pddd ) . in table
[ tab-3 ] and figure [ fig-8 ] , we present the nucleosynthetic output , chemical composition profiles and density profiles of these intermediate models that complete the exploration of the parameter space .
a brief discussion on the state of the art in 3d calculations of thermonuclear sn explosions can be found in [ subsec:3dmodels ] of this work ; for a review see @xcite . without going into the details of how these 3d models are calculated , here we present four one - dimensional mappings of 3d models that are representative of the current trends .
their main characteristics are given in figure [ fig-8 ] and table [ tab-4 ] .
model b30u is a 3d deflagration from @xcite , very similar to the models by @xcite and @xcite ( see discussion in [
sec : expanding ] ) .
model ddt3da is a 3d version of the delayed detonation paradigm @xcite . in this model ,
a detonation was artificially inducted in those regions were the flame resulting from the turbulent deflagration phase was well described by a fractal surface of dimension larger than 2.5 .
we note that this particular model also results in very well mixed ejecta , and in fact the properties of its x - ray emission in the snr phase are very similar to those of model b30u .
other delayed detonations in 3d calculated with different assumptions for the induction of the detonation result in more stratified ejecta @xcite . for a comparison between these two kinds of three dimensional delayed detonations , see table 1 and the accompanying text in @xcite .
finally , two 3d sub - chandrasekhar models from @xcite have also been included in the grid .
model sch3dop is a sub - chandrasekhar explosion calculated in 3d where the layer of degenerate he was ignited at one single point , while in sch3dmp the ignition happened at five different points .
it is worth noting that none of the 3d models has been followed for a sufficient time to account for the effects of the decay of @xmath53ni on the density profiles .
blinnikov , s. , bakalanov , p. , a.v.kozyrcva , & sorokina , e. 2004 , in 1604 - 2004 : supernovae as cosmological lighthouses , ed .
m. turatto , l. zampieri , s. benetti , & w. shea .
( asp conference series ) , in press [ astro - ph/0409396 ] bravo , e. , badenes , c. & garca - senz , d. 2005 , in interacting binaries : accretion , evolution and outcomes , ed . l.a .
antonelli , l. burderi , f. dantona , t. di salvo , g.l .
israel , l. piersanti , o. straniero & a. tornamb .
( aip conference series ) , in press , [ astro - ph/0412155 ] ( top left ) , @xmath73 ( top right ) and @xmath74 ( bottom ) for the principal elements in the shocked ejecta of model b30u , interacting with a uniform am of @xmath34 .
the crosses in the @xmath75 plot represent the total emission measure of the shocked ejecta .
see the on - line edition for a color version of this figure .
[ fig-6],title="fig : " ] ( top left ) , @xmath73 ( top right ) and @xmath74 ( bottom ) for the principal elements in the shocked ejecta of model b30u , interacting with a uniform am of @xmath34 .
the crosses in the @xmath75 plot represent the total emission measure of the shocked ejecta .
see the on - line edition for a color version of this figure .
[ fig-6],title="fig : " ] ( top left ) , @xmath73 ( top right ) and @xmath74 ( bottom ) for the principal elements in the shocked ejecta of model b30u , interacting with a uniform am of @xmath34 .
the crosses in the @xmath75 plot represent the total emission measure of the shocked ejecta .
see the on - line edition for a color version of this figure .
[ fig-6 ] ] ) are represented at @xmath76 s after the sn explosion , and they have been normalized by @xmath77 .
see the on - line edition for a color version of this figure.[fig-8],title="fig : " ] ) are represented at @xmath76 s after the sn explosion , and they have been normalized by @xmath77 .
see the on - line edition for a color version of this figure.[fig-8],title="fig : " ] ccccccccc tycho & @xcite & 432 & nei , single @xmath20 , & @xmath78 & @xmath79 & nei , single @xmath20 , & @xmath80 & @xmath81 + & & & single @xmath59 & & & single @xmath59 & & + kepler & @xcite & 404 & nei , single @xmath20 , & @xmath82 & @xmath83 & nei , single @xmath20 , & @xmath84 & @xmath85 + & & & single @xmath59 & & & single @xmath59 & & + 0509 - 67.5 & @xcite & @xmath86 & plane - parallel & @xmath87 & @xmath88 & plane - parallel & @xmath89 & @xmath90 + & & & nei shock & & & nei shock & & + n103b & @xcite & @xmath91 & plane - parallel & @xmath92 & @xmath93 & plane - parallel & @xmath94 & @xmath95 + & & & nei shock & & & nei shock & & + ccccccccccccc defb & 0.08 & & 1.37 & 0.64 & -19.14 & 0.94 & 0.61 & 0.61 & 0.025 & 0.017 & 0.0040 & 0.0043 + defc & 0.10 & & 1.37 & 0.74 & -19.29 & 0.99 & 0.68 & 0.55 & 0.021 & 0.014 & 0.0032 & 0.0032 + defd & 0.12 & & 1.37 & 0.80 & -19.34 & 1.02 & 0.71 & 0.52 & 0.021 & 0.014 & 0.0032 & 0.0034 + defe & 0.14 & & 1.37 & 0.81 & -19.29 & 0.98 & 0.73 & 0.49 & 0.021 & 0.013 & 0.0029 & 0.0028 + ddtb & 0.03 & @xmath103 & 1.37 & 1.36 & -19.67 & 1.11 & 0.98 & 0.05 & 0.10 & 0.084 & 0.022 & 0.027 + ddtbb & 0.01 & @xmath104 & 1.37 & 1.31 & -19.66 & 1.12 & 0.99 & 0.05 & 0.10 & 0.084 & 0.022 & 0.027 + ddtc & 0.03 & @xmath105 & 1.37 & 1.16 & -19.51 & 1.11 & 0.80 & 0.12 & 0.17 & 0.13 & 0.033 & 0.038 + ddtd & 0.03 & @xmath106 & 1.37 & 1.08 & -19.30 & 0.94 & 0.72 & 0.14 & 0.20 & 0.15 & 0.037 & 0.043 + pddb & 0.03 & @xmath105 & 1.37 & 1.36 & -19.72 & 1.14 & 1.04 & 0.03 & 0.085 & 0.070 & 0.018 & 0.022 + pddc & 0.03 & @xmath106 & 1.37 & 1.25 & -19.64 & 1.11 & 0.98 & 0.04 & 0.11 & 0.093 & 0.024 & 0.029 + pddd & 0.03 & @xmath107 & 1.37 & 1.24 & -19.53 & 1.04 & 0.89 & 0.05 & 0.15 & 0.13 & 0.034 & 0.041 + ccccccccc b30u & 1.37 & 0.42 & 0.53 & 0.66 & 0.045 & 0.011 & 0.0019 & 0.0017 + ddt3da & 1.37 & 0.78 & 0.76 & 0.38 & 0.063 & 0.027 & 0.0066 & 0.0072 + sch3dop & 1.02 & 1.14 & 0.58 & 0.23 & 0.064 & 0.035 & 0.0093 & 0.0077 + sch3dmp & 1.02 & 1.19 & 0.67 & 0.07 & 0.081 & 0.054 & 0.019 & 0.017 + | the supernova remnants left behind by type ia supernovae provide an excellent opportunity for the study of these enigmatic objects . in a previous work
, we showed that it is possible to use the x - ray spectra of young type ia supernova remnants to explore the physics of type ia supernovae and identify the relevant mechanism underlying these explosions .
our simulation technique is based on hydrodynamic and nonequilibrium ionization calculations of the interaction of a grid of type ia explosion models with the surrounding ambient medium , coupled to an x - ray spectral code . in this work
we explore the influence of two key parameters on the shape of the x - ray spectrum of the ejecta : the density of the ambient medium around the supernova progenitor and the efficiency of collisionless electron heating at the reverse shock .
we also discuss the performance of recent 3d simulations of type ia sn explosions in the context of the x - ray spectra of young snrs .
we find a better agreement with the observations for type ia supernova models with stratified ejecta than for 3d deflagration models with well mixed ejecta .
we conclude that our grid of type ia supernova remnant models can improve our understanding of these objects and their relationship to the supernovae that originated them . |
it is well known that many of the observed properties of galaxies correlate with their environment .
the most famous of these correlations is the morphology - density relation .
davis & geller ( 1976 ) showed that elliptical galaxies are more strongly clustered on the sky than spirals , and dressler ( 1980 ) demonstrated that the elliptical fraction in galaxy clusters is an increasing function of local density .
the dependence of the clustering of galaxies on luminosity has been much more controversial , but recent analyses of the largest available redshift surveys confirm that l@xmath3 and brighter galaxies have a higher clustering amplitude than low - luminosity galaxies ( park et al 1994 ; loveday et al 1995 ; benoist et al 1996 ; valotto & lambas 1997 ) .
low surface brightness galaxies are also distributed differently from their high surface brightness counterparts .
they are less clustered on all scales ( mo , mcgaugh & bothun 1994 ) and are particularly isolated on scales smaller than a few mpc ( bothun et al 1993 ) .
finally , the star formation histories and the gas fractions of galaxies of nominally the same hubble type are influenced by their local environments .
spiral galaxies in dense environments have redder colours and lower star formation rates that spirals in the field , and often also exhibit truncated hi disks ( e.g. kennicutt 1983 ; cayette et al 1994 ) .
one popular hypothesis for the origin of clustering differences between galaxies of different types is that mergers , tidal encounters or interactions with a surrounding gaseous medium modify galaxy properties .
such interactions are more probable in high - density environments . it should be noted , however , that differences in the clustering amplitude of galaxies of different morphologies , luminosities and surface brightnesses persist out to scales where the crossing time is much larger than the hubble time ( mo et al 1992 ; mo , mcgaugh & bothun 1994 ) , so it does appear that galaxies bear some imprint of the initial field of density fluctuations in the universe . according to the standard cosmological paradigm , structure in the universe
is built up through a process of hierarchical clustering .
small - scale fluctuations in the initial density field are the first to collapse to form bound , virialized objects ( or _ dark matter halos _ ) and these merge together over time to form more and more massive systems .
galaxies form when gas cools , settles , and turns into stars at the centres of the halos ( white & rees 1978 ; white & frenk 1991 ) .
galaxy formation models of this type have met with considerable success in explaining many of the trends seen in the properties of galaxies , both at present day and at high redshift ( e.g. lacey et al 1993 ; kauffmann , white & guiderdoni 1993 ; cole et al 1994 ) .
their most serious weakness arises from the fact that star formation and supernova feedback processes are poorly understood and hence difficult to model in a convincing way
. it should be noted , however , that one _ fundamental _ aspect of this picture is that the structure and evolutionary history of a galaxy is _ fully determined _ by the structure and evolutionary history of its surrounding dark matter halo .
the merging history of the halo determines the rate at which gas will cool and become available for star formation , as well as the frequency of merging events and the mass distribution of accreted objects .
the density profile of the halo , its shape and its distribution of angular momentum determine the structure , the size and the rotation curve of the galaxy that forms at its centre ( dalcanton , spergel & summers 1997 ; mo , mao & white 1997 ) .
if the properties of galaxies are observed to vary as a function of environment , it follows that the properties of their surrounding halos must also vary with environment . by understanding exactly which halo properties
can be affected by local conditions , one can hope to gain a deeper understanding of the origin of the clustering differences discussed above . in this paper
, we use numerical simulations of gravitational clustering to study the properties of dark matter halos as a function of local density . in order to achieve both good statistics and an accurate treatment of the formation history and internal structure of halos ,
high resolution simulations are needed , which nevertheless contain a fair sample of the universe , thus accounting correctly for the influence of large - scale structure on the formation of the halos .
we look for correlations between a set of present - day halo properties , including their masses , formation redshifts , concentrations , shapes and spins , with the overdensity of their local environment .
we also look for correlations of these quantities with the surrounding tidal field .
our conclusion is extremely simple .
only the mass distribution of dark halos varies as a function of environment .
the variation in the mass function is described extremely well by a simple analytic formula based on the conditional press - schechter theory derived by mo & white ( 1996 ) .
we find no significant dependence of any other halo property on environment .
our results do not depend on our choice of cosmology .
this leads to the conclusion that the dependences of galaxy morphology , luminosity , surface brightness and star formation rate on environment , must _ all _ arise because galaxies are preferentially found in higher mass halos in overdense environments and in lower mass halos in underdense environments .
the gif project is a joint effort of astrophysicists from germany and israel .
its primary goal is to study the formation and evolution of galaxies in a cosmological context using semi - analytical galaxy formation models embedded in large high - resolution @xmath4-body simulations .
this is done by constructing merger trees of particle halos from dark - matter only simulations and placing galaxies into them using a phenomenological modelling ( for a detailed description of this procedure as well as results cf .
kauffmann et al .
1997 ) .
the code used for the gif simulations is called hydra .
it is a parallel adaptive particle - particle particle - mesh ( ap@xmath5 m ) code ( for details on the code cf .
couchman , thomas , & pearce 1995 ; pearce & couchman 1997 ) .
the current version was developed as part of the virgo supercomputing project and was kindly made available by them for the gif project .
the simulations were started on the cray t3d at the computer centre of the max - planck society in garching ( rzg ) on 128 processors .
once the clustering strength required an even larger amount of total memory , they were transferred to the t3d at the edinburgh parallel computer centre ( epcc ) and finished on 256 processors .
a set of four simulations with @xmath6 and with different cosmological parameters was run . in this paper
we focus on a variant of a cold dark matter model , @xmath7cdm , with @xmath8 , @xmath9 , @xmath10 and shape parameter @xmath11 . a value of @xmath11 is usually preferred by analyses of galaxy clustering , cf . peacock & dodds ( 1994 ) .
this is achieved in the @xmath7cdm model despite @xmath8 and @xmath9 by assuming that a massive neutrino ( usually taken to be the @xmath7 neutrino ) had existed during the very early evolution of the universe and came to dominate the energy density for a short period .
it then decayed into lighter neutrinos which are still relativistic , thus delaying the epoch when matter again started to dominate over radiation .
the neutrino mass and lifetime are chosen such that @xmath11 . for a detailed description of such a model see white , gelmini , & silk ( 1995 ) .
the normalization was chosen so as to match the abundance of rich clusters ( white et al 1993 ) .
the simulation box size was 85 h@xmath12 mpc and the particle mass was @xmath13 . in order to be sure that our results are not sensitive to our choice of cosmological initial conditions ,
we have also analyzed a low - density cdm model with @xmath14 , @xmath15 , @xmath11 , h=0.7 and @xmath16 ( @xmath17cdm ) .
this simulation has a somewhat larger box size ( 141 h@xmath12 mpc ) , but the same particle mass .
halos were selected from the @xmath18 output times in the simulations as follows .
first , we searched for high - density regions using a standard friends - of - friends groupfinder with a linking length of @xmath19 times the mean interparticle separation .
we then searched for the particle with the lowest potential energy and adopted its position as the halo centre .
the distances of all the particles to the centre were ordered and the radius of the largest sphere with an overdensity @xmath20 was defined to be the _ virial radius _ of the halo , and the mass contained within this radius was defined to be the _
virial mass_. halos were only included in our analysis if the friends - of - friends mass exceeded 70 particles and the virial mass was between 35% and 95% of the friends - of - friends mass .
the upper bound served as a check that the majority of particles at overdensities greater than @xmath21 were located by the groupfinder .
the lower bound was chosen to ensure that enough particles were located within the virialized region so that the various halo quantities could be calculated reliably .
the following quantities were evaluated for each halo : 1 . the spin parameter @xmath22 , calculated for the particles within the virial radius , where @xmath23 and @xmath24 are the angular momentum and the thermal kinetic energy of the halo .
the formation redshift @xmath25 , defined as the redshift at which the most massive halo progenitor was half the present - day mass of the halo .
3 . concentration indices @xmath26 and
@xmath27 , where @xmath28 and @xmath29 are the radii enclosing one tenth and one twentieth of the virial mass respectively .
these parameters provide a coarse measure of the shape of the density profile of the halo .
the shape of the halo was determined by diagonalizing the moment of inertia tensor of halo particles within the virial radius .
this gives the principal axes @xmath30 .
the properties of the local environment around each halo were determined as follows .
we evaluated the overdensity @xmath31 in spheres of various radii surrounding the halo .
we also calculated overdensities in shells around the halo , thereby excluding the contribution of the halo mass itself . in order to obtain a measure of the higher order harmonics of the surrounding dark matter distribution , a density field on a 128@xmath32 grid ( cell size 0.66 h@xmath12 mpc )
was obtained from the simulation using cloud - in - cell ( cic ) interpolation .
the resulting field was smoothed using a top - hat smoothing window of 10 h@xmath12 mpc and the first and second - order spatial derivatives of the density and potential fields were determined using fast fourier transforms in k - space .
we made use of the following k - space window functions : * density dipole : @xmath33 * density shear field : @xmath34 * potential dipole field ( acceleration ) : @xmath35 * potential shear field ( tidal field ) : @xmath36 the resulting fields were interpolated to the positions of the halos using the cic method .
the shear fields were diagonalized and the eigenvalues ordered .
in an extension of the press - schechter theory , bond et al ( 1991 ) and bower ( 1991 ) derive an expression for the fraction of mass in a region of initial radius @xmath37 and linear overdensity @xmath38 , which at redshift @xmath39 is contained in halos of mass @xmath40 .
if one assumes that the region @xmath37 evolves in size and overdensity according to the spherical top - hat collapse model , it is simple to derive a formula for the mass function of halos in a present - day region of radius @xmath41 and overdensity @xmath31 .
this was first done by mo & white ( 1996 ) .
they tested their analytic formula against results from n - body simulations containing 128@xmath32 particles . in figure 1 , we show how the shape of the halo mass function changes as a function of local overdensity @xmath31 evaluated within spheres of r= 10 h@xmath12 mpc .
the thick solid line shows results derived from the @xmath7cdm simulation at @xmath18 .
the thin solid line shows the analytic prediction of mo & white ( 1996 ) . for comparison ,
the thick dashed line is the _ average _ halo mass function evaluated from the simulation , multiplied by a factor @xmath42 .
the thin dashed line is the same quantity calculated from the press - schechter theory . for clarity ,
the simulation and the analytic curves have been offset by 1 decade in the y - direction .
although it is not apparent as a result of the way we have chosen to present our results , it should be noted that the halo abundances predicted by the analytic theory exceed those derived from the simulation by a factor @xmath43 over the range of halo masses shown in figure 1 .
the magnitude of this offset is in rough agreement with that found by lacey & cole ( 1994 ) in their tests of the press - schechter formalism .
nevertheless , as can be seen in the plot , the theory does predict the change in shape of the halo mass function with @xmath31 remarkably well .
as can be seen , in low density regions , high - mass halos are underepresented , whereas in high - density regions the situation is reversed and high - mass halos are overabundant .
our principal result is that we find _ no correlation _ between the environment of halos and their formation times , spins , concentrations and shapes in either the @xmath7cdm or the @xmath17cdm simulations . out of a large number of possible scatterplots
, we have selected the following few to illustrate this conclusion .
results are shown only for the @xmath7cdm model , but plots for @xmath17cdm are virtually identical . in figure 2
we plot the value of the spin parameter @xmath44 against the overdensity in top - hat spheres of radius 10 h@xmath12 mpc for halos in four different mass bins .
this figure shows that the mean value of @xmath44 varies at most very weakly with overdensity or with mass .
there does appear to be a slight tendency for @xmath44 to increase with overdensity .
these results do not change if we adopt a different value of the smoothing radius . in figure 3
we plot formation redshift against the overdensity in a shell between 2 h@xmath12 and 5 h@xmath12 mpc surrounding each halo .
no dependence of mean formation redshift on overdensity is seen .
higher mass halos form somewhat later on average , as expected ( lacey & cole 1993 ) . figures 2 and 3 demonstrate that the _ average _ spin and formation time of halos are not correlated with their environments .
the next step is to demonstrate that the full probability distributions are also independent of local overdensity .
this is shown in figure 4 , where we plot @xmath45 and @xmath46 , for three different ranges in overdensity @xmath31 . as can be seen ,
the curves are almost indistinguishable from each other .
again , there is perhaps a slight shift towards larger spins in overdense regions .
the distribution of @xmath44 is consistent with that found in earlier work ( e.g. barnes & efstathiou 1987 ) .
finally , figure 5 is a mosaic illustrating a further subset of the possibilities we have explored . in the two left panels , we show the concentration index @xmath47 versus the overdensity evaluated in spheres of radius 10 h@xmath12 mpc surrounding each halo , and the shape axis ratio @xmath48 versus the tidal field axis ratio ( @xmath49 ) . in the two right panels
, we plot the spin parameter and the formation redshift versus principal axes ratios of the tidal field and the density shear field respectively .
( note that the apparent gaps in the distribution of @xmath50 and @xmath51 at small positive values of @xmath49 result from the ordering of the eigenvalues and the values that the @xmath52 assume in practice . )
once again , no correlations are found , either in the mean values of the halo quantities , or in their full probability distributions .
it is interesting that neither the spins , nor the distribution of shapes of dark halos , depend on the surrounding tidal field .
in this paper , we have demonstrated that mass is the only halo property that correlates significantly with local environment .
the dependence of the halo mass function on environment can be understood in terms of a simple extension of the press - schechter theory .
it is interesting to note that the fact that halo formation histories do not depend on local overdensity , is a prediction of the extended press - schechter theory in the excursion set formulation of bond et al ( 1991 ) . in this approach ,
the standard formulae are obtained by assuming that the mass of a halo to which any given mass element belongs , can be followed by studying the behaviour of the initial linear density field as it is smoothed with a succession of sharp k - space filters . in this model ,
the history of each mass element of a halo of mass @xmath53 ( and thus the formation history of the halo itself ) is statistically independent of the future of the element and thus of the halo s environment .
navarro , frenk & white ( 1996 ) have demonstrated that the density profile of dark matter halos in cdm - like cosmologies has a `` universal '' form , and that the characteristic density of a halo is related simply to its formation time .
it is thus also not surprising that our concentration parameters @xmath47 and @xmath54 turn out to be independent of environment .
there has also been substantial analytic work concerned with the angular momentum generated by tidal torques acting on local density maxima in the linear density field .
it has been argued ( blumenthal et al 1984 ; hoffman 1986 ) that the height of a peak is anticorrelated with its angular momentum , on the basis that high peaks collapse early , so there is not much time for tidal torques to act . if this was the case , overdense regions of the universe , in which high peaks are more probable , should harbor halos with systematically lower angular momentum .
heavens & peacock ( 1988 ) pointed out that this effect is counterbalanced by the fact that higher peaks experience greater tidal torques .
for realistic power spectra , they estimate that the two effects should nearly cancel out .
steinmetz & bartelmann ( 1995 ) obtain a similar result by modelling the formation of a halo as a collapse of a homogeneous ellipsoid acted upon by tidal shear from the surrounding matter .
both analyses are consistent with the very weak environment - dependence of spin in our data .
the shapes of dark matter halos have proven more difficult to calculate using analytic methods .
dubinski ( 1992 ) has investigated halo shapes in high - resolution n - body simulations and finds that there is no relationship between the shape of the initial density peak and the shape of the final collapsed halo . on the other hand , west , villumsen & dekel ( 1991 ) and tormen ( 1997 ) have demonstrated a clear tendency for the major axis of cluster - sized halos to align with larger - scale structures .
the main astrophysical implication of our conclusion is that clustering differences between galaxies of different types must arise purely because different galaxies sample different mass halos .
semi - analytic models of galaxy formation in hierarchical cosmologies demonstrate that it is possible to explain many of the clustering trends seen in the data in this way ( kauffmann , nusser & steinmetz 1997 ) . in these models ,
elliptical galaxies form when disk galaxies of comparable mass merge with each other .
these mergers occur preferentially in groups at redshifts @xmath55 .
the groups then coalesce to form clusters and superclusters .
as shown by kauffmann , white & guiderdoni ( 1993 ) and baugh , cole & frenk(1996 ) , the fraction of ellipticals increases strongly with halo mass in this picture and ellipticals thus end up more clustered than spirals .
in addition , the luminosity of the central galaxy in a halo scales in proportion to the halo mass , simply because more gas is able to cool and form stars in more massive halos .
luminosity - dependent clustering is thus a natural outcome of the models .
finally , it should also be noted that galaxies are assumed to be stripped of their surrounding dark matter once they have been accreted by a larger system , such as a group or cluster .
they thus lose their supply of new cold gas , their star formation rates decline , and their stellar populations redden and fade .
red , gas - poor galaxies are thus found predominantly in high mass halos and are again predicted to be more clustered .
the semi - analytic models at present provide no explanation for the clustering differences between galaxies of different surface brightnesses .
one might speculate that low surface - brightness disks are more fragile and are more easily destroyed in high - density environments .
more detailed modelling is necessary , however , before a conclusion can be drawn as to whether this will work in practice . | we use large n - body simulations of dissipationless gravitational clustering in cold dark matter ( cdm ) cosmologies to study whether the properties of dark matter halos are affected by their environment .
we look for correlations between the masses , formation redshifts , concentrations , shapes and spins of halos and the overdensity of their local environment .
we also look for correlations of these quantities with the local tidal field .
our conclusion is extremely simple .
only the mass distribution varies as a function of environment .
this variation is well described by a simple analytic formula based on the conditional press - schechter theory .
we find no significant dependence of any other halo property on environment .
our results do not depend on our choice of cosmology . according to current hierarchical models ,
the structure and evolutionary history of a galaxy is fully determined by the structure and evolutionary history of the dark halo in which it is embedded .
if these models are correct , clustering variations between galaxies of differing morphological types , luminosities , colours and surface brightnesses , must all arise because the halo mass function is skewed towards high mass objects in overdense regions of the universe and towards low mass objects in underdense regions .
-0.55 cm -0.55 cm 0.1 mm 16.75 cm environmental influences on dark matter halos and consequences for the galaxies within them + gerard lemson@xmath0 & guinevere kauffmann@xmath1 + + _ @xmath2max - planck institut fr astrophysik , d-85740 garching , germany _ + keywords : galaxies : haloes ; galaxies : formation ; galaxies : evolution ; cosmology : theory ; cosmology : dark matter |
recently , ghirlanda , ghisellini and lazzati ( 2004 , ggl04 thereafter ) found a surprisingly tight correlation between the peak of the @xmath7ray spectrum @xmath8 ( in a @xmath9 plot ) and the collimation corrected energy emitted in @xmath7rays @xmath10 in long gamma - ray bursts ( grbs ) .
the latter is related to the isotropically equivalent energy @xmath11 by the value of the jet aperture angle @xmath12 , by @xmath13 .
the scatter around this correlation is tight enough to wonder if the correlation itself can be used for a reliable estimate of @xmath11 , making grbs distance indicators , and therefore probes for the determination of the cosmological , parameters , and for the exploration of the matter to vacuum dominance transition .
this issue is similar to the case of sn ia : they are not perfect standard candles ( i.e. their luminosities are not all the same ) , nevertheless the luminosity of a specific supernova can be found through the correlation of their luminosity and the speed of the decay of their light curve ( i.e. the slower the brighter , phillips 1993 ; riess , press & kirshner 1995 ) .
it is the existence of this correlation among sn ia which made possible their cosmological use ( riess et al .
2004 , hereafter r04 ; perlmutter et al . 1999 ; schmidt et al . 1998 ) . very recently
, this problem has been explored by dai , liang & xu ( 2004 ) , which found tight constraints on and using the correlation found by ggl04 .
their result is however based on a strong assumption : they assume as universal the correlation measured in a particular cosmology ( without errors on its slope ) , and use it to derive the cosmology itself .
actually , the best fit correlation depends on the cosmology adopted to derive burst luminosities and the correlation should be re calibrated for each cosmology . in this letter
we demonstrate that a correct approach leads to less tight constraints on the cosmological parameters using grbs alone . on the other hand
, a more interesting cosmological result can be acheived if a combined fit with sn ia is performed .
as in the case of sn ia , the use of a class of objects as cosmological `` rulers '' requires them being standard candles . the luminosity of grbs , calculated assuming isotropy , spans @xmath14 orders of magnitude ( frail et al .
2001 ) , but strong observational evidence ( i.e. the achromatic break in the afterglow light curve ) indicates that the burst emission is collimated into a cone / jet of some aperture angle @xmath12 ( levinson & eichler 1993 ; rhoads 1997 ; sari , piran & halpern 1999 ; fruchter et al .
the corresponding energy emitted in @xmath7rays , corrected by the collimation factor @xmath15 , clusters around @xmath16 erg , with a small dispersion ( 0.5 dex ) , yet not small enough for a cosmological use ( bloom et al .
2003 ) .
ggl04 found a tight correlation between @xmath10 and the ( rest frame ) peak energy @xmath8 of the @xmath9 prompt emission spectrum : @xmath17 .
the exact value of @xmath18 depends on the assumed cosmology .
using @xmath19 and @xmath20 we have @xmath21 the scatter of the data points around the correlation is of the order of 0.1 dex .
this allows to reconstruct the value of @xmath10 by measuring @xmath8 .
this is analogous to sn ia , for which there is a tight relation between their peak luminosity and the stretching factor of their optical light curve ( phillips 1993 ; goldhaber et al .
2001 ) , with less luminous supernovae showing a faster post maximum light curve decay ( reiss et al . 1995 ) .
the proper modelling of this effect ( hamuy et al .
1996 ; perlmutter et al . 1999 ) improves the determination of the sn ia luminosity and consequently reduces the scatter in the hubble diagram , yielding constraints on the cosmological parameters ( see r04 using sn ia with redshift up to @xmath11.75 ) .
the @xmath22 correlation for grbs makes them a new class of `` rulers '' for observational cosmology and combining grbs and sn ia can further reduce the region of allowed values in the cosmological parameter space .
furthermore , since grbs are detectable at larger @xmath23 , they are a powerful tool to explore in more detail the cosmic kinematics . the difference between the standard candle assumption and the use of the intrinsic correlations , for both grbs and sn ia , is shown in fig .
1 ( top and bottom panel , respectively ) through the hubble diagram in the form of luminosity distance vs redshift . in the upper panel
we assume that grbs and sn ia are standard candles with a unique energy ( @xmath24 erg ) for grbs and luminosity ( @xmath25=-21.1 ) for sn .
the derivation of the luminosity distance @xmath26 for sn follows straightforwardly from their distance modulus ( r04 ) .
for grbs we have @xmath27 $ ] , where @xmath28 is the @xmath7ray fluence ( i.e. the time integrated @xmath7ray flux ) .
note that the determination of @xmath12 requires the knowledge of the isotropic energy ( see e.g. eq . 1 in frail et al .
2001 ) , in turn requiring specific values of ( , ) . in the bottom panel
we plot sn ia and grbs after correcting for the stretching
luminosity and the @xmath29@xmath8 relations , respectively . in this case
the isotropic energy @xmath11 of grbs has been estimated from their measured @xmath8 through the @xmath10@xmath8 correlation and the error on the slope of this correlation has been properly included in the @xmath26 total uncertainty .
also in this case we must fix a given ( , ) cosmology both for the derivation of @xmath12 and for finding the best @xmath29@xmath8 relation .
as in the sn ia case , the luminosity distance of grbs derived from their @xmath10@xmath8 correlation ( bottom panel ) highly reduces the scatter around possible different cosmologies ( solid , dashed and dotted lines ) .
moreover , grbs populate the @xmath30 region , where @xmath31 is rather sensitive on ( , ) .
the correlation found by ggl04 _ assumes _ = 0.3 and = 0.7 , and @xmath32 .
changes on and induce a change on the normalization and slope of eq .
[ corr ] , together with a _ different scatter _ of the data points around the best fit line .
we can then ask what is the pair of cosmological values , which produces the `` minimum scatter '' around the fit , performed using the very same , pair . to this aim
we use all the 15 bursts of known redshifts , @xmath8 and jet break time @xmath33 listed in tab . 1 and tab . 2 of ggl04 . the difference with the study of dai
, liang & xu ( 2004 ) lies mainly in this point : they assumed that the @xmath10-@xmath8 correlation is exact and cosmology independent , while it is not .
erg ( i.e. a @xmath02.6% and @xmath016% change in slope and normalization with respect to eq .
1 ) . with = 1 , = 1 we obtain @xmath34 erg ( i.e. a @xmath09% and @xmath030% change in slope and normalization with respect to eq .
additional differences concern : i ) the estimate of the errors on the density of the interstellar medium when it is unknown
( they assume @xmath35 @xmath36 while we allow @xmath37 to cover the entire 110 @xmath36 range ) ; ii ) we do not exclude grb 990510 and grb 030226 from the analysis ; iii ) we include grb 030429 , for which a jet break time was recently found by jakobsson et al .
( 2004 ) ; iv ) we always use @xmath15 ( instead of the @xmath38 approximation ) as the collimation correction factor ( also when estimating the error on @xmath10 ) .
we also consider the 156 sn ia of the `` gold '' sample of r04 , finding the corresponding , contours using the distance moduli and corresponding errors listed in their tab .
2 shows our results .
grbs alone ( red lines ) are almost insensitive to but limit to lie within @xmath00.05 and 0.22 ( 68% confidence level ) .
we also show the region pinpointed by the wmap experiment ( spergel et al .
2003 ) , which is only marginally consistent with the allowed region from sn ia alone ( blue lines in fig .
2 ) . the combined grbs+sn ia fit ( filled regions in fig . 2 ) selects a region which is more consistent with the cosmic microwave background ( cmb ) results .
the minimum ( with a reduced @xmath39=1.146 ) is for = 0.37@xmath40 and = 0.87@xmath41 ( 1@xmath42 ) . assuming a flat universe yields = 0.29@xmath43 and = 0.71@xmath44 .
if we use the classical " hubble diagram method , we compare the @xmath26 values given by estimating @xmath10 through the actual correlation found in each point of the , plane with the luminosity distance calculated through e.g. eq .
11 of r04 ( see also carrol et al . 1992 ) .
then , by a @xmath45 statistics , we find the confidence regions in the , plane , which are plotted as dashed line on fig .
this classical method is very similar to the previous one , since it uses the same available information .
the small difference ( contours slightly larger ) is due to the fact that with the minimum scatter " method we calculate the distance of the data points from the correlation ( i.e. perpendicular to the fitting line ) , while , using the classical " hubble diagram method , we are using the distance between the @xmath10 data point and the corresponding @xmath10 by the correlation
. we can further constrain , with the combined grb and sn samples , the dark energy component which is parametrized by its equation of state @xmath46 . furthermore , @xmath47 could be varying , and one possible parametrization is @xmath48 ( see e.g. r04 ) . adopting this law
, we compute the luminosity distance according to eq .
14 of r04 , assuming a flat cosmology with = 0.27 . in this case
the fit is performed in the @xmath49@xmath50 plane for grbs , sn and for the combined samples .
as before , we recompute the @xmath10@xmath8 relation for each @xmath51 , @xmath50 pair . , @xmath50 , consider that , for @xmath52 and @xmath53 the correlation becomes @xmath54 erg ( i.e. a @xmath03.4% and @xmath015% change in slope and normalization with respect to eq .
1 ) . ] fig .
3 reports the corresponding confidence intervals . besides making the confidence region smaller than what derived for sn alone ,
the effect of grbs is to include within the 68% contour of the joint sn+grb sample ( filled region ) the @xmath55 , @xmath56 point , corresponding to the classical cosmological constant .
grbs can now be used as cosmological rulers " , bridging the gap between the relatively nearby type ia supernovae and the cosmic microwave background . the swift satellite ( gehrels et al .
2004 ) , designed for the fast localization of grbs , is expected to find about one hundred of grbs per year : it can open up a new era for the accurate measurements of the geometry and kinematics of our universe ( for a more extended discussion see firmani et al . 2004 ) .
we stress that , besides finding high redshift bursts , which are of course very important for finding tighter constraints , it is crucial to find _ low redshift _
grbs , to determine the @xmath10@xmath8 correlation in a redshift range which is not affected by the , values .
this would allow to use the resulting correlation unchanged for all values of , , strongly reducing the associated errors . in turns , this will allow to constrain cosmological parameters independently from sn ia .
this is important since grbs are unaffected by dust extinction and it is very unlikely that two completely different classes of objects would have similar evolutions to mimic a consistent set of cosmological parameters . in fig .
4 we show an illustrative example of what can be done _ if a given correlation were known to be valid independently of the cosmological parameters_. for this we have chosen the correlation given by eq .
it can be seen that even the limited sample of our bursts can strongly influence the grb+sn confidence contours , making them more in agreement with the wmap results ( not unexpectedly , since we have used just the correlation appropriate for = 0.3 and = 0.7 . a similar consideration concerns
the dai et al .
2004 result ) .
we would like to stress that in order to use grbs to find the cosmological parameters , we need a set of well measured data , and especially a well measured jet break time @xmath33 , necessary to find the collimation angle @xmath12 , and a good spectral determination of the prompt emission .
bloom , j. s. , frail , d. a. , kulkarni , s. r. , 2003 , apj , 594 , 674 carroll , s. m. ; press , w. h. ; turner , e. l. , 1992 , ara&a , 30 , 499 dai , z. g. , liang , e. w. & xu d. , 2004 , apj in press ( astro - ph/0407497 ) frail , d. a. , kulkarni , s. r. , sari , r. et al .
, 2001 , apj , 562 , l55 firmani , c. et al .
, 2004 , in preparation fruchter , a. et al . 1999 ,
apj , 519 , l13 gehrels , n. , chincarini , g. , giommi , p. , et al . , 2004 ,
apj , in press ( astro ph/0405233 ) ghirlanda , g. , ghisellini , g. & lazzati , d. , 2004 , apj , in press ( astro ph/0405602 ) ( ggl04 ) goldhaber , g. , groom , d. e. , kim , a. , aldering , g. , et al .
, 2001 , apj , 558 , 359 jakobsson , p. , hjorth , j. , fynbo , j. p. u. , et al .
, subm . to a&a ( astro
ph/0407439 ) hamuy , m. , phillips , m. m. , suntzeff , n.b .
et al . , 1996 , aj , 112 , 2398 levinson , a. , eichler , d. , 1993 , apj , 418 , 386 perlmutter , s. , aldering , g. , goldhaber g. , et al .
, 1999 , apj , 517 , 565 phillips , m. m. , 1993 , apj , 413 , l105 riess , a. g. , press , w.h . &
kirshner r.p . , 1995 ,
apj , 438 , l17 riess , a. g. , strolger , l g .
, tonry h. , et al . , 2004 , apj , 607 , 665 ( r04 ) rhoads , j. h. , 1997 , apj , 487l sari r. , piran , t. , halpern , j. , 1999 , apj , 524 , l43 schmidt , b. p. , suntzeff , n. b. , phillips , m. m. , et al . , 1998 ,
apj , 507 , 46 spergel , d. n. , verde , l. , peiris , h. v. , et al . , 2003 ,
apjs , 148 , 175 | the best measure of the universe should be done using a standard ruler " at any redshift .
type ia supernovae
( sn ia ) probe the universe up to [email protected] , while the cosmic microwave background ( cmb ) primary anisotropies concern basically @xmath11000 .
apparently , gamma ray bursts ( grbs ) are all but standard candles .
however , their emission is collimated and the collimation corrected energy correlates tightly with the frequency at which most of the radiation of the prompt is emitted , as found by ghirlanda et al .
( 2004 ) . through this correlation
we can infer the burst energy accurately enough to probe the intermediate redshift ( @xmath2 ) universe . using the best known 15 grbs
we find very encouraging results that emphasize the cosmological grb role
. a combined fit with sn ia yields @xmath3 and @xmath4 . assuming in addition a flat universe , the parameters are constrained to be @xmath5 and @xmath6 .
grbs accomplish the role of missing link " between sn ia and cmb primary anisotropies
. they can provide a new insight on the cosmic effects of dark energy , complementary to the one supplied by cmb secondary anisotropies through the integrated sachs wolfe effect .
the unexpected standard candle cosmological role of grbs motivates the most optimistic hopes for what can be obtained when the grb - dedicated satellite , swift , will be launched . |
in the past decade the study of cosmology has taken its first major steps as a precise empirical science , combining concepts and tools from astrophysics and particle physics .
the most recent of these results have already brought surprises .
the universe s expansion is apparently accelerating rather than decelerating as expected due to the gravitational attraction of matter .
this implies that the flat , matter - dominated model for the universe does not apply and that our current fundamental physics understanding of particles , forces , and fields is likely to be incomplete .
this evidence for a vacuum energy , or more generally a negative - pressure component called `` dark energy '' , comes from the supernova measurements of changes in the universe s expansion rate that directly show the acceleration .
figure [ confcmbclust ] shows the results of @xcite ( see also @xcite ) which compare the standardized brightnesses of ground and space - observed high - redshift ( @xmath10 ) type ia supernovae ( sne ia ) with a large sample of low - redshift sne ia @xcite .
the data implies that for a flat universe a fraction @xmath11 of the critical density would reside in a cosmological constant .
more generally , these measurements indicate the presence of a new , unknown energy component with energy density @xmath12 that can cause acceleration , hence having an equation of state ( ratio of pressure to energy density @xcite ) with @xmath13 .
the evidence for dark energy has received strong corroboration from cosmic microwave background ( cmb ) results @xcite which are sensitive to the total energy density , combined with galaxy power spectrum and cluster abundance measurements @xcite which probe the matter density @xmath14 , or with an @xmath15 prior ( see figure [ confcmbclust ] ) .
two of these three independent measurements would have to be in error to make dark energy unnecessary in the cosmological model .
the dark energy might be due to the cosmological constant term in einstein s equations , which implies vacuum energy with @xmath16 .
alternatively , it could be due to a dynamical scalar field with @xmath17 and/or time - varying @xmath18 .
the fundamental importance of a universal vacuum energy has sparked a flurry of activity in theoretical physics with several classes of models being proposed ( e.g. quintessence @xcite , pseudo - nambu - goldstone boson ( pngb ) models @xcite , cosmic defects @xcite , and modified gravity @xcite .
all these models explaining the accelerating universe have observational consequences for the supernova hubble diagram @xcite .
the luminosity distance as a function of redshift can for convenience be parameterized to first order with effective dark - energy parameters , e.g. @xmath19 , or @xmath6 , @xmath20 .
a constant @xmath19 can represent dark - energy models such as the cosmological constant while @xmath21 is successful at describing a wide variety of both scalar field and more general models @xcite .
predictions of dark - energy models can be either compared directly to supernova hubble diagrams or to the dark - energy parameters as fit by the data .
placing some constraints on possible dark - energy models , @xcite find that for a flat universe , the data are consistent with a cosmological - constant equation of state and @xmath22 ( figure [ wconf ] ) , or generally @xmath23 at 95% confidence level .
when combining supernova with other results , such as from a variety of cmb experiments and large - scale - structure measurements@xcite , @xcite find @xmath24 ( @xmath25 systematics ) and @xcite get @xmath26 .
cosmic strings ( @xmath27 ) are already ruled out as dark energy , while the domain walls ( @xmath28 ) and tracking quintessence models ( @xmath29 ) are disfavored .
current and upcoming ground - based supernova experiments will continue to improve measurements of the dynamical effects of the dark energy .
for example , the nearby supernova factory @xcite will produce a photometrically calibrated spectral time - series for hundreds of @xmath30 supernovae , providing a rich data set for refining our understanding of sn ia behavior .
the snls and essence projects will discover and follow hundreds of supernovae to @xmath31 mag accuracy in the range @xmath32 , giving significantly tighter constraints on the value of @xmath19 .
unfortunately once the assumption that @xmath18 is constant is relaxed , these experiments are not able to provide strong constraints on @xmath6 nor @xmath9 @xcite . testing the nature of dark energy will thus be difficult with these experiments .
a new experiment is then necessary to study the properties and possible models for the dark energy , and refine the measurement of the matter and dark energy densities of the universe . since the discovery of cosmic acceleration in 1998 , many observational approaches to dark energy have been proposed ( see @xcite for recent reviews ) .
the most well developed and understood method of constructing the expansion history of the universe is the same one that led to the discovery of dark energy : the measurement of the supernova hubble diagram .
the systematic and statistical uncertainties in current sn ia measurements are on the same order of magnitude : any new experiment to perform precision cosmology must eliminate or significantly reduce the influence of the systematic uncertainties while discovering and measuring high - quality light curves and spectra of a statistically large number of supernovae . in this paper
we describe the supernova / acceleration probe ( snap ) , a dedicated satellite mission designed to measure the properties of dark energy .
originally conceived in 1999 @xcite , the snap concept has progressed through a series of refinements all starting with science requirements and then deriving observational , instrumental , and mission requirements . the snap reference mission is presented in this paper in the form of a spacecraft , orbit , telescope , imager , spectrograph , and observation strategy .
this mission serves as a reference from which future optimization studies will be performed and is a demonstration of a design that can realistically accomplish the science goals .
it is self - consistent , technically feasible , reduces risk , and is specifically designed to provide data that strictly control statistical and systematic uncertainties for supernova and weak gravitational lensing science .
snap can run as a complete supernova experiment providing discovery , and photometric and spectroscopic followup of a full sample of supernovae over the target redshifts .
snap is also an excellent platform for performing deep wide - field surveys with space - quality imaging providing crucial complementary constraints to study dark energy from weak lensing and other studies .
snap s full potential is realized when used in conjunction with complementary results that are anticipated by the time snap launches : a large sample of well - measured low - redshift sne ia and precise measurements of the distance to the surface of last scattering from the planck satellite . in [
dataset : sec ] we describe the precision cosmology that can be achieved using type ia supernovae .
the needed supernova data set is based on the identification of known or plausible systematic uncertainties that fundamentally limit measurement accuracy , and on the number and redshift range of supernovae necessary to probe precisely the dark - energy parameters .
the power of weak gravitational lensing surveys in measuring cosmological parameters is discussed in [ lensing : sec ] and the added complementarity of lensing and other dark - energy probes made possible with snap data is described in [ otherde : sec ] .
we present in [ proposed_experiment ] the observing strategies and instrumentation suite tailored to provide the data that satisfy both the statistical and systematic requirements for the supernova and lensing surveys .
the snap calibration program is described in
[ calib : sec ] .
the results of detailed simulations of the snap mission are given in
[ sim : sec ] . in [
ancsci : sec ] we discuss the general properties of the snap surveys and the science resources they will provide .
type ia sne have already proved to be excellent distance indicators for probing the dynamics of the universe .
however , the current supernova cosmology measurements are limited by systematic uncertainties ( which occur for all cosmological probes ) . in order to move toward the era of precision cosmology ,
a new data set must both address potential systematic uncertainties and provide sufficient statistical power to accurately determine dark - energy properties , especially the time variation of the equation of state . because control of systematic uncertainties is so central we begin with aspects of the data set necessary for their control .
we proceed by listing the sources of systematic uncertainty , strategies to reduce their effects , and the resulting improvements that can be obtained in measuring the supernova hubble diagram .
we then determine which supernova observations give a statistical uncertainty comparable to the systematic uncertainty , leading to requirements on the number of supernovae we need to find , their distribution in redshift , and how precisely we need to determine each one s peak brightness .
high - redshift supernova searches have been proceeding since the late 1980 s @xcite . particularly since the discovery of the accelerating expansion of the universe , the high - redshift supernova methodology for measuring cosmological parameters has been critically scrutinized for sources of systematic uncertainty .
below are identified systematic effects which any experiment aiming to make maximal use of the supernova technique will need to recognize and control .
we also give a rough estimate of the expected _ systematic _ residual in supernova magnitude after statistical correction for such effects with the snap data set or from independent data sets from other astrophysical studies .
_ extinction by host - galaxy `` normal '' dust : _ extinction from host - galaxy dust can significantly reduce the observed brightness of a discovered supernova although typically type ia supernovae suffer only modest extinction @xcite .
cross - wavelength flux - calibrated spectra and multi - band photometry will identify the properties of the obscuring dust and gas and the amount of extinction suffered by individual supernovae .
there will remain a residual uncertainty proportional to the calculated extinction due to dust - model dependence .
this can be minimized by excluding highly - extincted supernovae from the analysis .
stringent requirements on the snap calibration system are needed to control uncertainties in the photometric - system zeropoints , particularly since the uncertainties are correlated from supernova to supernova .
our calibration program is being designed to control extinction uncertainty to 1% .
_ gravitational lensing by clumped mass : _ inhomogeneities along the supernova line of sight can gravitationally magnify or demagnify the supernova flux and shift the mode of the supernova magnitude distribution by @xmath331
10% depending on redshift @xcite .
since flux is conserved in this process , averaging large numbers of supernovae per redshift bin will give the correct mean brightness .
deviations from a gaussian distribution for the brightness of supernovae at similar @xmath34 can be used to determine the clumpiness of the intervening mass distribution @xcite .
the large snap supernova sample will reduce the distance modulus uncertainty per redshift bin to @xmath35% . snap weak gravitational lensing measurements , and micro - lensing studies can further help distinguish whether or not the matter is in compact objects . _ k - correction and cross - filter calibration : _ broadband photometry of supernovae at different redshifts is sensitive to differing supernova - rest - frame spectral regions .
k - corrections are used to put these differing photometry measurements onto a consistent rest - frame passband @xcite . applying k - corrections with only today s methods and calibration sets
could produce errors of several percent . using the currently available spectra
, we tune the snap filter - set specifications to give the target systematic uncertainty @xmath36 mag .
( the situation will improve even further as the spectral library of supernovae grows and snap itself will add to this library . ) furthermore , strict requirements are placed on the snap calibration program to ensure accurate transformations of fluxes between filters . _ galactic extinction : _ extinction maps of our own galaxy are uncertain by @xmath33 110% depending on direction @xcite .
supernova fields can be chosen toward the low extinction galactic caps .
future spitzer observations will allow an improved mapping between color excesses ( e.g. of galactic halo subdwarfs in the snap field ) and galactic extinction by dust .
galactic - extinction uncertainty can then be controlled to @xmath37% in brightness .
_ non - sn ia contamination : _ other supernova types are on average fainter than sne ia and their contamination could bias their hubble diagram . observed supernovae must be positively identified as sn ia .
as some type ib and ic supernovae have spectra and brightnesses that otherwise mimic those of sne ia , a spectrum covering the defining rest frame si ii 6150 feature for every supernova at maximum will provide a pure sample .
this systematic would then be eliminated .
_ malmquist bias : _ a flux - limited sample preferentially detects the intrinsically brighter members of any population of sources .
the amount of magnitude bias depends on details of the search but can reach the level of the intrinsic magnitude dispersion of a standard candle . directly correcting this bias would rely on knowledge of the sn ia luminosity function , which may change with lookback time .
a detection threshold fainter than peak by at least five times the intrinsic sn ia luminosity dispersion ensures sample completeness with respect to intrinsic supernova brightness , eliminating this bias @xcite . a systematics - limited experiment must account for speculative but reasonable sources of uncertainty .
the following are sources of systematic uncertainty for which there is no direct evidence but which can not yet be discounted .
as opposed to normal dust , gray dust is postulated to produce wavelength - independent absorption in optical bands @xcite . although models for gray - dust grains dim blue and red optical light equally , the near - infrared ( nir ) light ( @xmath331 - 2 @xmath38 m ) is less affected .
cross - wavelength calibrated spectra extending to wavelength regions where `` gray '' dust is no longer gray will characterize the hypothetical large - grain dust s absorption properties .
armed with the extinction color excess properties of the gray dust , broadband near - infrared colors can provide `` gray '' dust extinction corrections for supernovae out to @xmath39 ; @xcite have explored the observations necessary to be sensitive to 0.02 mag gray - dust extinction .
gray dust will re - emit absorbed starlight and thus contribute to the far - infrared background ; current observations indicate that the fir flux is attributable to point sources @xcite .
quasar colors in the sdss sample as a function of redshift give an upper limit to the possible dimming due to gray dust at the level of 0.2 mag ( in the restframe b - band ) at @xmath40 @xcite .
deeper scuba and spitzer observations should further tighten the constraints on the amount of gray dust allowed .
supernova behavior itself may have systematic variations depending , for example , on properties of the progenitor systems .
the distribution of these stellar properties is likely to change over time``evolve''in a given galaxy , and over a population of galaxies .
supernova heterogeneity manifests itself through observed signatures in their spectral light emission ( the bulk emitted in optical wavelengths ) , and the time - evolution of that emission .
signs of supernova diversity are thus captured in observations of multi - band light curves and/or spectral time - sequences that begin shortly after explosion and extend to the nebular phase of the event .
this has been done using optical measurements of nearby sne ia drawn from a wide range of galactic environments that provide an observed evolutionary range of sne ia @xcite .
the photometric and spectral differences that have been identified in these data are well calibrated by the sn ia light curve width - luminosity relation , leaving a 10% intrinsic peak - brightness dispersion .
there is currently no evidence for systematic residuals after correction , for example with galaxy type or supernova location @xcite .
high signal - to - noise ( @xmath41 ) multi - band light curves and spectra over optical wavelengths can provide precise control over additional general supernova variability .
efforts are underway to expose additional effects using larger , more precise and systematic , low - redshift supernova surveys @xcite
. these high - quality data are also necessary to control possible evolution effects when performing precision cosmology with high - redshift supernovae .
theoretical models can identify observables that are expected to display heterogeneity .
these key features , indicative of the underlying initial conditions and physical mechanisms controlling the supernova , will be measured with snap , allowing statistical correction for what would otherwise be a systematic uncertainty .
the state of empirical understanding of these observables at the time snap launches will be explicitly tested by snap measurements @xcite . at present
, we perform fisher matrix analyses on model spectra and light curves to estimate the statistical measurement requirements , and to ensure that we have necessary sensitivity to use subsamples to test for residual systematics at better than the 2% level .
this approach reveals the main effects of as well as the covariance between the following observables : _ rise time from explosion to peak : _ this is an indicator of opacity , fused @xmath42ni mass , and possible differences in the @xmath42ni distribution .
a 0.1 day uncertainty corresponds to a 1% brightness constraint at peak @xcite .
achieving such accuracy requires discovery within @xmath332 days of explosion , on average , @xmath3330@xmath43 fainter than peak brightness .
current constraints on rise - time differences are two days @xcite .
_ plateau level 45 days past peak : _ the light - curve plateau level that begins @xmath3345 days past and more than 10@xmath43 fainter than peak is an important indicator of the c / o ratio of the progenitor star , and fused @xmath42ni . in models ,
a 5% constraint on this plateau brightness corresponds to a 1% constraint on the peak brightness @xcite .
_ overall light - curve timescale : _ the `` stretch factor '' ( @xcite , see also @xcite ) that parameterizes the light - curve timescale @xcite is affected by almost all of the aforementioned parameters since it tracks the sn ia s light - curve development from early to late times .
it is correlated with rise time and plateau level and it ties snap s controls for systematics to the controls used in the current ground - based work .
a 0.5% uncertainty in the stretch factor measurement corresponds to a @xmath331% uncertainty at peak @xcite .
stretch distributions are quite consistent between the current sets of low and high - redshift supernovae @xcite .
_ spectral line velocities : _ the velocities of several spectral features throughout the uv and visible make an excellent diagnostic of the kinetic - energy injection in sne ia . in models , velocities constrained to @xmath33250 km s@xmath44 constrain the peak luminosity to @xmath331% @xcite , given a typical sne ia expansion velocity of 15,000 km s@xmath44 .
current data show smooth velocity development for a given supernova , but also clear differences between supernovae which have not yet been found to correlate with supernova luminosity @xcite . _
spectral features : _ the positions of various spectral features in the restframe uv are strong metallicity indicators of the sne ia . by achieving a sufficient @xmath41 and resolution on such features @xcite snap will be able to constrain the metallicity of the progenitor to 0.1 dex @xcite .
this spectral region has only recently begun to be explored with uv spectroscopy of nearby supernovae with hst .
spectral features in the restframe optical ( ca ii h&k and si ii at 6150 ) provide additional constraints on the opacity and luminosity of the sn ia @xcite . figures [ syslc : fig ] and [ sysspec : fig ] show the particular light - curve and spectral parameters that can serve as indicators of supernova evolution . by measuring all of the above features for each supernova we can tightly constrain the physical conditions of the explosion .
this makes it possible to recognize subsets of supernovae with matching initial conditions , ensuring a small luminosity range for each subset .
in addition to these features of the supernovae themselves , we will also study the host galaxy of the supernova .
we can measure the host galaxy luminosity , colors , morphology , and the location of the supernova within the galaxy , even at redshifts @xmath45 .
the last two observations are difficult or impossible from the ground for high - redshift galaxies .
these galaxy properties can provide clues to any shifts with redshift in the supernova progenitor population .
the behavior of systematic uncertainties are complicated and the determination of their propagated effects on cosmology requires rigorous analysis @xcite .
particularly important systematic effects for supernova - cosmology experiments are those that put a floor on how well the distance modulus can be measured in a certain redshift bin .
for example , supernovae at similar redshifts whose rest - frame flux regions lie in common observer passbands can share completely correlated calibration uncertainty .
an increase in the statistical sample of supernovae can not decrease this uncertainty .
in addition , these uncertainties could be correlated across redshift bins . again taking calibration as an example , a temperature uncertainty in a blackbody fundamental calibrator propagates into correlated flux and magnitude zeropoint uncertainties .
correlated errors across redshift bins can lead to a bias in the determination of the cosmological and dark - energy parameters .
we do not further consider correlated bias in this paper as @xcite find that the systematic bias in @xmath6 and @xmath9 lies well within the 68%-error contours due to comparably - sized irreducible systematic uncertainties . for convenience
we adopt in this paper a systematic - uncertainty model of @xmath46 magnitude irreducible uncertainty per redshift bin of width 0.1 as proposed by @xcite , where @xmath47 represents the depth of the survey . the systematic uncertainty introduced in the comparison of high - redshift supernovae with the local sample
is an increasing function of redshift .
systematic uncertainty will exist at some level even for the local sample , due for example to uncertainty in the galactic - extinction zeropoint and spatial correlations in @xmath48 maps .
the subtyping strategy employed for evolution based on observables tied to the explosion characteristics can be more generally applied .
the data ( supernova risetime , early detection , light - curve peak - to - tail ratio , identification of the type ia - defining si ii spectral feature , separation of supernova light from host - galaxy light , and identification of host - galaxy morphology , etc . ) make it possible to study each individual supernova and measure enough of its physical properties to recognize deviations from standard brightness subtypes
. only the change in brightness as a function of the parameters classifying a subtype is needed , not any intrinsic brightness .
supernovae can not change their brightness in one measured wavelength range without affecting brightness somewhere else in the spectral time series an effect that is well - captured by expanding atmosphere computer models . as a precision experiment snap
is designed to distinguish supernovae `` demographics '' by detecting subtle variations in light - curve and spectral parameters . by matching like to like among the supernova subtypes
we can construct independent hubble diagrams , each with systematic biases reduced to the level of 1% in distance .
this procedure is described and tested in @xcite .
comparison within a redshift bin reveals possible systematics , while comparison within a subtype over the redshift range cleanly probes the cosmology . the targeted residual systematic uncertainty from effects such as malmquist bias , @xmath49-correction , etc .
total @xmath332% .
thus , a subset , or `` like vs. like '' , analysis based on physical conditions should group supernovae to within @xmath502% in luminosity .
this is in addition to a purely statistical correction for any second - parameter effects beyond the stretch factor .
the statistical supernova sample appropriate for experiments with the presence of systematic uncertainty has been considered in the fisher - matrix analysis of @xcite .
( their results have been validated with our independent monte carlo analysis . )
they assign a statistical uncertainty of 0.15 magnitudes to each supernova , which combines statistical measurement uncertainty and intrinsic supernova dispersion . the systematic - error model described in [ syserrmod : sec ] is included .
the importance of using sne ia over the full redshift range out to @xmath51 for measuring the cosmological parameters is demonstrated in figure [ sci_deltaw.fig ] , which shows the uncertainty in measuring the equation of state parameter variation , @xmath9 , as a function of maximum redshift probed in distance surveys @xcite .
( note that the results look very similar whether one defines @xmath9 in terms of a linear expansion of @xmath5 in redshift or in scale factor , and whether one uses a prior on matter density or on distance to the cmb last scattering surface . )
this calculation considers 2000 sne ia measured in the range @xmath52 , along with 300 low - redshift sne ia from the nearby supernova factory @xcite .
a flat universe is assumed .
three types of experiments were considered .
the first is an idealized experiment subject only to statistical uncertainties , free of any systematics , and with extremely tight prior knowledge of the matter density .
the second two are more realistic models which assume both statistical and systematic uncertainties and different priors on @xmath14 . from this figure
we conclude that a sn ia sample extending to redshifts of @xmath53 is crucial for realistic experiments in which some systematic uncertainties remain after all statistical corrections are applied .
ignoring systematic uncertainties can lead to claims that are too optimistic .
similar conclusions apply for the other dark - energy parameters , including @xmath6 and an assumed constant @xmath18 @xcite .
the sample of @xmath332000 sne ia is enough to provide systematic - uncertainty limited measures of the dark - energy parameter @xmath9 .
these supernovae represent a subset of sne ia that can provide robust cosmological measurements ( e.g. non - peculiar , low - extinction ) , culled from an even larger set of discovered supernovae .
this large sample is necessary to allow model - independent checks for any residual systematics or refined standardization parameters , since the sample will have to be subdivided in a multidimensional parameter space of redshift , light curve - width , host properties , etc .
the large number of lower signal - to - noise light curves from supernovae at @xmath54 will provide an extended redshift bin in which to check for consistency with the core sample .
expected parameter measurement precisions for the snap supernova program with @xmath55 are summarized in table [ sci_errors.tab ] .
we determine precisions for two fiducial flat universes with @xmath56 , one in which the dark energy is attributed to a cosmological constant and the other to a sugra - inspired dark - energy model .
the parameter precisions then depend on the choice of data set , priors from other experiments , assumptions on the flatness of the universe , and the model for the behavior of @xmath18 .
( parameter determination from monte carlo analysis of simulated data from the reference snap mission is given in
[ sim : sec ] . )
cccccc fiducial universe : flat , @xmath56 , cosmological constant dark energy + + snap sn ; @xmath16 & 0.02 & 0.05 & & + snap sn + wl ; @xmath16 & 0.01 & 0.01 & & + snap sn;@xmath57 prior ; flat ; @xmath58 & & @xmath59 & @xmath60 & @xmath61 + snap sn;planck prior ; flat ; @xmath58 & & @xmath62 & @xmath60 & @xmath63 + snap sn + wl ; flat ; @xmath58 & & @xmath64 & @xmath65&@xmath66 + + fiducial universe : flat , @xmath56 , sugra - inspired dark energy + + snap sn ; @xmath57 prior ; flat ; @xmath58 & & @xmath59 & @xmath67&@xmath68 + snap sn ; planck prior ; flat ; @xmath58 & & @xmath69 & @xmath60&@xmath70 + snap sn + wl ; flat ; @xmath58 & & @xmath64 & @xmath59&@xmath71 + if the universe is taken to be spatially flat , the sn data ( with planck prior ) determine @xmath14 to 0.01 .
the present value of the dark energy equation of state is constrained within 0.09 and the physically crucial dynamical clue of the equation of state time variation is bounded to within 0.19 .
note that these constraints are _ systematics limited _ ; much tighter constraints would be obtained if systematics were ignored .
( in [ otherde : sec ] we discuss the tighter constraints obtained by combining the supernova results with the other snap measurements . ) the precise , homogeneous , deep supernova data set allows for robust interpretation in more general terms than @xmath6 and @xmath9 .
the function @xmath5 can be implemented in a nonparametric , uncorrelated bin , or eigenmode method @xcite .
one can also obtain the expansion history @xmath72 itself , e.g.@xcite .
we can also obtain constraints on flatness of the universe , rather than assuming it .
current data are mostly sensitive to one linear combination of @xmath14 and @xmath73 ( roughly their difference ) .
snap is designed to obtain sufficient brightness - redshift data for a wide range of redshifts ( @xmath74 ) , and to enable an extraordinarily accurate measurement of @xmath14 and @xmath73 separately ( see figure [ confcmbclust ] ) , the culmination of an approach originally proposed by @xcite .
assuming that the dark energy is the cosmological constant , the supernova experiment can simultaneously determine the mass density @xmath14 to an accuracy of @xmath69 , the cosmological constant energy density @xmath73 to @xmath65 .
propagating this through one obtains a 0.06 measurement uncertainty of the curvature of the universe , @xmath75 .
gravitational weak lensing provides an independent and complementary measurement of the dark - energy parameters through the mapping of galaxy - shape distortions induced by mass inhomogeneities in the universe .
weak gravitational lensing of background galaxies by foreground dark matter ( large - scale structure ) provides a direct measurement of the amount and distribution of dark matter ( see [ science : sec ] for details on the impact snap can have on these measurements ) .
the growth of these dark matter structures with cosmic time is determined by the dark energy .
thus , precise weak lensing observations can provide significant and independent constraints on the properties of dark energy . for a review of the techniques used to measure weak lensing and the current status of weak - lensing measurements ,
see @xcite .
dark energy modifies the weak - lensing observables by altering the distance - redshift relation as well as the matter power spectrum .
these two , in turn , determine the weak - lensing shear power spectrum @xcite .
specifically , dark energy will modify the normalization and nonlinear part of the matter power spectrum and redshift evolution .
weak lensing measurements with a wide - field space telescope such as snap can thus achieve an accurate measurement of @xmath12 , and the dark - energy parameters @xcite . in figure [ power :
fig ] we show the expected measurement of the shear power spectrum possible with 100 million galaxies over 300 square degrees divided into 2 redshift bins .
the upper curve represents a measurement using resolved source galaxies with @xmath76 while the lower curve represents @xmath77 galaxies .
the power of space - based observations is most apparent at small scales ( high @xmath78 ) , where the shot noise is small due to the large surface density of resolved galaxies . in figure
[ snapconst : fig ] we show the joint constraints on @xmath14 and assumed - constant @xmath18 using 10 million galaxies , 100 million galaxies , and 100 million galaxies divided into two redshift bins with additional skewness information . note that the constraints from weak lensing are largely orthogonal to the constraints from supernovae .
thus , weak lensing complements the supernova technique in deriving constraints on @xmath18 .
@xcite have demonstrated that by combining additional information on non - gaussianity from the lensing bispectrum in the moderately nonlinear regime with the power spectrum , cosmological constraints can be improved significantly ( about a factor of three over the results shown in figure [ snapconst : fig ] ) . at present the ability to exploit this information
is limited by computationally intensive numerical predictions for non - linear growth , but the gravitational physics involved is well understood and work on obtaining accurate predictions is already underway . at still smaller ( arcminute ) angular scales , the lensing signal is very strong , but the non - linear growth predictions become very difficult .
new statistical treatments of weak - lensing data have recently been proposed @xcite which measure the cosmic geometry while canceling uncertainties in the non - linear growth .
this is achieved by cross - correlating the lensing signal with an estimate of the foreground mass distribution .
figure [ bj03fig ] presents an estimate by @xcite of the statistical uncertainties on @xmath6 and @xmath20 ( or @xmath9 , recall @xmath79 at @xmath80 ) that could be obtained by the snap surveys using this lensing cross - correlation technique .
the quality of the measurement is comparable to that of the type ia sn measurements , as long as the systematic uncertainties are kept below the statistical uncertainties .
several conditions are necessary to make possible high signal - to - noise weak - lensing measurements for these types of analyses
. the shapes of millions of background galaxies must be measured to overcome cosmic variance and provide a precise measure of the dark matter power spectrum .
it is particularly important to be able to resolve many small , distant galaxies beyond a redshift of @xmath81 ; the broad redshift range of source galaxies provides a powerful lever arm on the constraint of @xmath18 by providing information on the growth of structure .
high spatial resolution is necessary to extract statistically powerful shape information from each individual galaxy and decrease the size of the smallest resolved galaxies in a survey . a stable point spread function
( psf ) and a minimal level of internal optical distortions are essential for the precise measurement of galaxy shapes .
the high spatial density of resolved background galaxies afforded by space - based observations allows cross - correlation measurements and gives measurements of the mass power spectrum at scales where non - linearities are most apparent .
accurate photometric redshifts are required for the lensing analyses , particularly the lensing cross - correlation , where any systematic biases in photo - z s above @xmath330.001 in @xmath82 would dominate the error budget .
catastrophic photo - z errors would similarly need to be kept below 1 per 1000 .
it may be possible use internal analyses of lensing data to trade precision on the dark energy parameters for looser constraints on photo - z biases .
reducing the photo - z errors , biases , and catastrophic errors requires high - accuracy photometry in bands spanning the visible and near - ir .
atmospheric emission lines make it impractical to acquire these data in the near - ir and near certain bright visible lines .
satisfying all of these requirements simultaneously requires a wide - field space based imaging telescope in a thermally stable orbit . with a high - throughput , space - based 2 meter telescope , an exposure time of 2000s
will provide the depth to resolve significant numbers of galaxies in the scientifically interesting range beyond @xmath83 .
unhindered by atmospheric emission and absorption , nir observations from space provide significant redshift depth and improved photometric redshift estimates .
avoiding the earth s atmosphere in a thermally stable orbit gives the fine resolution and stable psf .
the psf and depth give a high galaxy surface density . a ground - based survey of tens of thousands of square degrees ,
exploring the mass power spectrum on large angular scales with small statistical errors would be complementary .
in particular , such large - angular - scale surveys probe the linear part of the matter power spectrum while deeper but somewhat less wide surveys probe the nonlinear region .
( for completeness , we discuss the possibility of a panoramic , wide - field survey with snap in
[ panoramic : sec ] . )
the two complementary cosmological surveys that constitute the leading components of the snap primary mission together combined give constraints on the properties of dark energy significantly stronger than either one by itself .
table [ sci_errors.tab ] gives the expected dark - energy parameter uncertainties from the joint results from the deep supernova survey and the combination of power spectrum , bispectrum , and cross - correlation cosmography techniques from a 1000 square - degree lensing program with @xmath84 arcmin@xmath85 .
a very conservative estimate of the systematics floor on shear calibration of 3% is used . for a flat universe
the present dark energy equation of state @xmath6 can be measured to @xmath86 , and the time variation can be bounded within 0.11 . without assuming a flat universe , the mass density @xmath14 , the cosmological constant energy density @xmath73 , and curvature @xmath87 can be determined to @xmath88 .
this poses an interesting test of the cosmological framework by cross - checking the curvature measured at @xmath89 with current measurements out to @xmath90 of @xmath91 @xcite .
note that _ no priors _ are required when we consider the full snap mission of supernovae plus weak lensing .
these are significant improvements over the results from supernovae alone .
the constraints are subject to improvement as practical understanding of future weak lensing data develops .
universes with more dark energy ( than the fiducial @xmath92 assumed here ) , dark energy possessing a more positive equation of state or a stronger time variation would lead to a measurement with more sensitivity to dark - energy properties .
we have been conservative in taking the cosmological constant as fiducial , the least sensitive of the canonical quintessence models .
for example , in a supergravity - inspired model of dark energy @xcite , @xmath9 would be more tightly constrained to @xmath93 . with strong constraints on @xmath9
we will be able to differentiate between the cosmological constant and a range of dynamical scalar - field ( `` quintessence '' ) particle - physics models ( @xcite . )
the snap reference mission focuses on the established methods of type ia supernova hubble diagrams and weak gravitational lensing to study dark energy .
other less - tested and more model - dependent methods for measuring cosmological parameters are now being explored as additional probes of dark energy with snap .
these include ( but are not limited to ) using type ii supernovae as distance indicators @xcite , baryon oscillations as a standard ruler in both transverse and radial directions @xcite , redshift - space correlations for the alcock - paczynski test @xcite , strong lensing time - delays and statistics @xcite , and cluster - counting statistics @xcite . attacking dark energy with measurements from independent techniques that are subject to differing , but tightly controlled systematic uncertainties will give increased confidence in the ultimate conclusions made about its nature . since these techniques generally constrain different combinations of cosmological parameters , it is important that each `` stand on its own two feet '' with systematics control so that they can be combined for studies of dark energy .
snap can naturally study dark energy using many of these other probes .
the necessary measurements can be made with the snap instrumentation suite as is , or in coordination with ground or other space - based telescopes .
a snap panoramic - survey observing mode that covers a significant fraction of sky can provide large number statistics , redshift depth , wavelength coverage , discovery of rare objects , and measures on large angular scales ; these are exactly the characteristics of the data needed for many of these complementary methods , with all the advantages of observations above the distorting and absorbing atmosphere .
discovery and study of a larger number and of more distant supernovae ( or any probe ) is by itself insufficient to successfully accomplish a rigorous investigation of the cosmological model . as shown in
[ exec_uncertain ] we must address each of the systematic concerns while making precise supernova measurements , thus requiring a major leap forward in measurement techniques . the requirements placed on the snap instrument derive directly from the science goals .
the primary requirement is to obtain the corrected peak brightness vs. redshift of at least 2000 type ia sne out to a redshift of @xmath94 .
identification of type ia sne requires the measurement of characteristic spectral features near peak luminosity .
host - galaxy redshift is also determined spectroscopically .
the corrected peak magnitudes are derived from supernova rest - frame optical light curves and spectra .
we require a statistical measurement uncertainty for the peak magnitude corrected for extinction and shape inhomogeneity roughly equal to the current intrinsic magnitude dispersion of supernovae , @xmath95 mag , and a systematic uncertainty almost an order of magnitude smaller .
to meet the scientific requirements , the snap reference mission has a large , 0.7 square degree instrumented field of view and an observing cadence of 4 days , commensurate with the timescale over which early - epoch supernova light curves change .
the imager s large field - of - view gives a significant multiplex advantage ; each exposure contains @xmath96 type ia supernovae within 5.4 observer - frame months from the date of explosion ( corresponding to 2 rest - frame months for @xmath94 supernovae ) .
observations in multiple filters yield multi - color rest - frame optical light curves . a spectrometer optimized for supernova spectra
is allocated observation time for supernova type - identification , feature measurements , and spectral - template building .
the telescope aperture is constrained by the photometric and spectroscopic @xmath41 requirements and the focal plane must accommodate the large field of view .
the derived requirements for the snap reference mission described here serve as a starting point ; they are subject to change as the mission is further refined . as other subsidiary approaches to measure dark energy
are developed as discussed in
[ otherde : sec ] we can also consider minor detector additions or replacements on the focal plane with either extended broad - band filters or with dispersive elements .
for example , a band in between the reference filter bands could sharpen the photometric redshift accuracy for a part of the survey field . as a general rule ,
broadband photometry is significantly faster than slitless spectroscopy by the ratio @xmath97 for equivalent @xmath41 , however , for specific applications a detector with a small dispersive element in place of a filter may be useful
. it could provide spectral time series of core - collapse supernovae that fall in the limited field enabling the determination of their distance .
it could also provide precise redshifts of field sources that can calibrate photometric redshifts especially useful for tomographic weak lensing , or for the baryon - oscillation method .
the primary snap science program consists of a deep survey and a wide - field survey with individually - designed observing schedules .
the repeated scans of the deep - survey field provide supernova discovery and light curves , and when coadded give a high surface density of background galaxies for lensing surveys .
the wide - field survey is designed for dark - energy measurements using weak lensing and other @xmath18-sensitive techniques .
in addition , we explore a possible panoramic survey that extends the weak - lensing data and provides the general astronomical community a `` legacy '' survey of space - quality optical - to - nir images of a significant fraction of the sky .
the programs are described here and summarized in table [ obs : tab ] .
the surveys are described by the solid - angle of sky monitored , the exposure time for each scan of the field , and the number of times the survey region is scanned .
cccccc deep north & @xmath98 & optical : @xmath99 & 4 & 120 + & & nir : @xmath100 & & + wide - field & @xmath101 & optical : @xmath102 & & 1 + & & nir : @xmath103 & & + deep south & @xmath98 & optical : @xmath99 & 4 & 120 + & & nir : @xmath100 & & + panoramic & @xmath104 & optical : @xmath105 & & 1 + & & nir : @xmath106 & & + a simple , predetermined observing strategy repeatedly steps through a 7.5 square degree northern zone with a 4 day cadence continuing for 16 months .
later in the mission this strategy is repeated for a southern survey region .
the fields are selected to be perpendicular to the sun , earth , and moon and where natural zodiacal light is near minimum , toward the north and south ecliptic caps .
this deep - survey strategy allows discovery and automatic followup for sne ia that explode in those regions .
every field will be visited every four days for 16 months , with sufficiently long exposures that all relevant sne ia in the snap survey regions will have significant signal within a few restframe days of explosion .
( supernovae at much higher redshifts on average will be found slightly later in their light curve rise times although their prior history is in the data . )
the periodic observation of fixed fields ensures that every supernova at @xmath107 will have its light curve followed for at least several months in the rest frame as it brightens and fades .
sixty percent of the snap deep supernova survey will be spent for the pure photometric scan ; about 120 exposures are possible with a four day cadence to cover a scan area of 7.5 square degrees .
running this survey again in the south yields another 7.5 square degrees .
the zodiacal light will be the dominant source of background given snap s fields , orbit , and shielding of sun and earthshine .
the cosmic - ray flux contamination of 4.5 particles/@xmath108/s will make multiple exposures necessary to eliminate significant contamination of the images .
forty percent of the snap deep survey will be spent doing targeted spectroscopy of supernovae with parallel imaging ( that is , imaging occurs simultaneously during these spectroscopic observations ) .
the resulting images will cover random positions and orientations within the snap field and will be used to increase the depth of the survey and help with photometric and astrometric cross - calibration between detectors .
this prearranged observing program will provide a uniform , standardized , calibrated data set for each supernova , allowing for the first time comprehensive comparisons across complete sets of sne ia .
the following reference strategies and measurements will address , and often eliminate , the statistical and systematic uncertainties described in [ exec_uncertain ] .
* blind , multiplexed searching .
* sne ia at @xmath109 . * spectrum for every supernova at maximum covering the rest frame uv and si ii 6150 feature .
* spectral time series of representative sn ia , with cross - wavelength relative flux calibration . * light curves sampled at frequent , standardized intervals starting shortly after explosion and extending to 80 restframe days after explosion to obtain a light - curve - width- and extinction - corrected peak rest - frame @xmath110 brightness to 10% .
this requires percent - level peak - magnitude determinations in the supernova - frame optical passbands . * multiple color measurements in a filter set consisting of 9 bands approximating rest - frame @xmath110 at different redshifts .
* final images and spectra after the supernova is no longer present to enable clean subtraction of host galaxy light .
the quality of these measurements is as important as the time and wavelength coverage , so we require control over @xmath41 for these photometry and spectroscopy measurements to give high statistical significance for supernovae over the entire range of redshifts .
we also require control over calibration for these photometry and spectroscopy measurements , by collecting monitoring data to measure cross - instrument and cross - wavelength calibration .
note that to date no single sn ia has ever been observed with this complete set of calibrated measurements , either from the ground or in space , and only a handful have a data set that is comparably thorough . with the observing strategy described here , _ every one _ of @xmath111 followed sn
ia will have this complete set of measurements .
the requirements on spectroscopic exposure times and signal - to - noise are described in detail in @xcite . using an exposure time of 8.5 hours for @xmath94 supernovae will provide the desired spectral - parameter measurements given the snap spectrograph parameters described in [ spec : sec ] .
exposure times for supernovae at lower redshifts down to @xmath89 obey a scaling factor of @xmath112 .
the exact exposure time necessary will be refined as the library of uv optical supernova spectra continues to grow in coming years .
the number of supernovae that snap is expected to discover is based on the supernova rate per unit volume of @xcite assumed independent of redshift , which when extrapolated is roughly consistent with the discoveries of the goods / acs search and subaru supernova search .
this is a conservative assumption compared to rates calculated incorporating the cosmic star formation rate @xcite .
snap will be able to spectroscopically observe almost all the supernovae with deep and good temporal photometric coverage , @xmath113 supernovae in the 32-month survey .
@xcite find that in a resource - limited survey with systematic uncertainties taken into account , using a relatively uniform distribution at high redshift is optimal for providing cosmological parameter measurements . as a result
, snap will not provide spectroscopic followup for a fraction of the highest redshift supernovae , which are more plentiful than the lower redshift supernovae .
for the cosmology analysis , we expect to cull the sample of spectroscopically observed supernovae to @xmath114 events that satisfy strict homogeneity and extinction cuts to yield a flat distribution in cosmic time as shown in figure [ sndist : fig ] .
the strengths that make snap excellent for supernova observations apply to lensing as well ; a wide - field imager in space with stable and narrow point - spread - functions can provide large survey areas , accurate shape measurements , and high galaxy angular - surface densities .
the nine optical and near infrared filters are necessary for accurate photometric redshift determinations .
the snap deep - survey fields in which the supernovae are found will thus serve as a deep lensing field .
the snap wide - field survey has been tailored to resolve the shapes of @xmath4 galaxies per square arcminute .
over one year @xmath115 million galaxies will be resolved .
for source galaxies with an intrinsic ellipticity distribution @xmath116 @xcite , this will achieve a signal - to - noise of unity on the shear measurement in a 1 square arcminute cell , over 1000 square degrees .
the reference snap deep survey will exceed even this density requirement by a factor of three @xcite .
current plans for the wide - field survey call for each field to be observed for 1200 seconds in each optical filter , and twice as long in each nir filter .
each pointing will be divided up into 6 individual 200 second exposures which are needed to eliminate cosmic rays and cover the spatial gaps between the detectors .
we are engaged in determining the systematic uncertainties intrinsic to the snap weak - lensing program to find the optimal survey area and allotted time for the wide - field survey ; initial estimates indicate 300 - 1000 square degree field ( 516 months ) are appropriate for the wide - field survey .
the on - board storage space is matched to the telemetry and these two constraints limit how many images can be stored and subsequently sent to earth . if the telemetry constraints can be eased , either through a higher download rate ( and greater on - board storage ) or through data compression , the exposure times can be made shorter , thus increasing the area of the weak lensing survey .
a full description of the snap weak - lensing program can be found in a separate series of papers @xcite .
a extremely wide - field `` panoramic '' imaging survey is potentially useful for weak lensing and other probes of dark energy , cosmology , and astronomy in general .
snap will provide a unique means for performing wide - field optical imaging from space .
we thus explore the depth obtained from a three - year snap survey covering 700010000 square degrees
. we do not address specific scientific programs made possible from this survey , considering it as a possible legacy dataset that can serve the astronomical community at large .
the reference mission calls for a halo orbit about the second earth - sun lagrange point ( l2 ) .
the l2 point is about 1.5 million kilometers from the earth along the sun - earth line and has a very stable thermal environment .
there are no eclipses and heat from the earth is constant and negligible .
there are ample communication contact opportunities from the ground to snap .
a ground station would communicate from 2 - 4 hours per day with the spacecraft .
this orbit is highly advantageous from the standpoint of achieving passive detector cooling ; a three square meter radiator that operates at 130k provides 45 watts of gross cooling capacity .
having the snap fields near the ecliptic caps places the sun at nearly right angles to the viewing direction throughout the year . in an l2 orbit
, the earth and moon will always be nearly at right angles to the viewing direction and on the sun side of the spacecraft .
we utilize this viewing geometry in several ways .
first , the solar panels can be rigidly body - mounted on the sunward side of the spacecraft , avoiding the cost , failure modes , and flexure of deployed panels .
second , the passive cooling radiator can be rigidly located in permanent shadow on the antisunward side of the spacecraft .
third , the stray light baffling can be optimized for a limited range of solar roll and elevation angles , and for a limited range of earth elevation angles .
we plan to have the spacecraft perform 90 degree roll maneuvers every 3 months during the mission , to keep up with the mean ecliptic longitude of the sun .
the detector layout in the focal plane has a 90 degree roll symmetry that allows photometric scans with identical filter coverage from season to season ( see [ imager : sec ] ) .
the requirements placed on the snap telescope derive directly from the science goals and the mission constraints .
the wavelength coverage is determined by the need to observe a number of filter bands across the visible and nir wavelength range spanning over 0.35 @xmath38 m to 1.7 @xmath38 m , and to conduct low - resolution spectroscopy of each supernova near maximum light .
this requirement effectively rules out refracting optical trains and drives the telescope design toward all - reflective optics .
the light gathering power is set by the need to discover supernovae early in their expansion phases and to permit photometry and low - resolution spectroscopy near maximum light .
these requirements can be met with a minimum aperture of about two meters .
note that for a fixed signal - to - noise ratio ( @xmath41 ) the exposure time for an isolated point source is proportional to the telescope aperture to the fourth power with background - limited noise and diffraction - limited optics .
image quality is also a factor in determining @xmath41 because of the effects of natural zodiacal light and detector noise .
we intend to achieve angular resolution near the diffraction limit for wavelengths longward of one micron . for a two - meter aperture and one - micron wavelength
the airy disk size is 0.11 fwhm . to match this diffraction spot size to the ccd pixel size ( @xmath117 @xmath38
m ) one must adopt an effective focal length of about 20 meters .
this focal length is also matched in the nir where hgcdte detectors with 1820 @xmath38 m pixels will observe out to a photon wavelength of 17000 .
a large field of view is needed to obtain multiplexing advantage : a single exposure can contain tens of different supernovae that are in important phases of their light curves .
the snap reference mission has a field of view of the order one square degree , of which about 0.7 square degrees will be instrumented by detector pixels .
the ratio of working field area to diffraction patch area is about 800 million , comparable to the total number of detector pixels .
an excellent optical quality and low - level of distortion must be maintained over the working focal plane .
the image quality of the telescope is driven in part by the @xmath41 requirement , and also by the potential systematic supernova spectrum contamination by unwanted light from the supernova host galaxy
. a system strehl ratio of 0.90 at one micron wavelength is baselined , which corresponds to an rms wavefront error ( wfe ) of 0.05@xmath38 m . at 0.633 @xmath38 m
the strehl ratio is 0.77 .
( the strehl ratio is the peak monochromatic image irradiance divided by the theoretical peak irradiance for the ideal diffraction limited image . )
prospective launch vehicles ( delta iv , atlas v , sealaunch ) and payload fairing dimensions impose limits on the telescope size and its mass . an overall payload length of about 6 meters and a payload diameter of about 2.5 meters will accommodate the snap observatory . through a series of packaging exercises we have explored ways to fit the maximum length stray light
baffle into available launch fairings , and find that with a short optical package , @xmath118 m in length , and a tall outer baffle , the required stray light rejection can be achieved ( [ stray : sec ] ) . to accommodate dimensional limitations and wide - field optical quality ,
the three mirror anastigmat described by @xcite is used .
a schematic view is shown in figure [ optics : fig ] .
details of the reference snap optical configuration have been determined by an iterative process involving exploring various alternative choices for focal length , working field coverage , and packaging constraints @xcite .
the optimized optical parameters are summarized in table [ tel : tab ] .
the overall length of the optical train is 3.3 meters .
compared with the 21.66 meter effective focal length , this system has an effective telephoto advantage of about 6.5 .
the mirrors are pure conic sections of revolution having no polynomial terms .
the use of higher polynomial terms has not yet been explored .
the location of the vertex of each element is listed in a cartesian ( x , z ) coordinate system whose origin is the vertex of the primary mirror and z is the optical axis .
ccccccc primary & 2.00 & 0.5 & -0.2037466 & -0.981128 & 0 & 0 + secondary & 0.45 & none & -0.9099607 & -1.847493 & 0 & -2.00 + folding flat & 0.80 @xmath43 0.48 & 0.20 @xmath43 0.12 & 0 & 0 & 0 & 0.91 + tertiary & 0.69 & none & -0.7112388 & -0.599000 & -0.87 & 0.91 + focal plane & 0.577 & 0.258 & 0 & 0 & 0.9 & 0.91 the launch environment imposes both stiffness and strength requirements on the payload .
vehicle aeroelastic stability concerns prescribe the needed payload stiffness in terms of minimum structural frequencies in the axial ( @xmath119 hz ) and lateral ( @xmath117 hz ) directions .
the launch environment includes both quasi - steady and random acceleration events that are combined to establish the peak loads , or strength requirements for the payload .
for a space mission it is vital to create a mechanical configuration which provides an extremely stable metering structure that maintains the optical element alignment during ground testing , launch , and orbit operations .
the concept adopted for snap is to create three structural components that will be brought together during spacecraft / payload integration : a stiff low - precision outer baffle cylinder carrying the exterior solar panels and extensive thermal insulation ; a stiff low - precision spacecraft bus structure that carries antennas , batteries , and other major spacecraft support components ; and a stiff high - precision telescope structure comprising carbon - fiber metering elements , the kinematically - mounted mirrors , the instrumentation suite , and its own thermal control system .
figure [ cut : fig ] shows the overall payload and spacecraft layout , while figure [ meter : fig ] shows details of the secondary and tertiary metering structures .
space - proven optical mirror technology is largely based on two approaches : open - back schott zerodur glass ceramic composite material and corning ultra - low expansion ule glass honeycomb structure . for snap either technology
has sufficiently low coefficient of thermal expansion and sufficiently well proven manufacturing techniques .
studies are underway exploring the detailed fabrication and test flows using each process .
the metering structure will utilize a low - cte carbon - fiber construction .
in particular , the secondary support tripod will have to maintain the primary to secondary spacing accurate to a few microns .
this tripod , and the other major metering components , will certainly require a dedicated active thermal control system with passive cooling .
we anticipate the need for five - axis motorized adjustment for the secondary mirror during ground integration , on - orbit observatory commissioning , and occasionally during science operations .
for this reason we plan to include a hexapod or other multi - axis positioner into the secondary - mirror support structure .
the single highest priority bearing on the choice of mirror coating is the system throughput at the longest wavelengths where rest - frame optical photons from the highest - redshift supernova are the most distant and photons are the most precious .
a secondary consideration is to establish a low thermal emissivity for the mirrors so that operating them at approximately 290k will not seriously impair our astronomical sensitivity in the nir bands .
the most common coating for astronomical mirrors at visible wavelengths is sio overcoated aluminum .
it offers outstanding durability and unmatched reflectance throughout the visible band , 0.4 to 0.7 @xmath38 m .
a less common choice is protected silver , which is less efficient in the blue but more efficient in the red and nir .
the snap optical system has four reflections so its throughput varies as the fourth power of the mirror coating reflectivity .
we have assumed here the use of protected silver rather than protected aluminum owing to its substantially higher total system throughput at the photon - starved wavelengths . in the reference
snap optical configuration , the center of the focal plane is severely vignetted by the necessary hole in the folding flat which lies near the cassegrain focus . in optimizing the optical performance of the system , the useful imaging surface on the focal plane
is taken to have an annular shape .
the optical performance of our reference telescope is fundamentally limited by aberrations and manufacturing errors at short wavelengths , and by diffraction at long wavelengths .
accordingly , our expected performance figures divide into two areas : geometrical ray traces that quantify the aberrations and pupil diffraction studies .
we summarize the key performance items in table [ opt : tab ] and figure [ spot : fig ] .
cc focal length & 21.66 meters + aperture & 2.0 meters + final focal ratio & f/10.83 + field & annular , 6 to 13 mrad ; 1.37 sq deg + rms geometric blur & 2.8 @xmath38 m , average 1 dimension + central obstruction & 16% area when fully baffled + vane obstruction & 4% area , tripod + from spot diagrams at various off - axis angles , we have compiled the statistics on the mean radial centroid of ray hits in the focal plane , and the second moments of the spot distributions .
these are listed in table [ moment : tab ] in the form of the two orthogonal rms breadths ( radial rms and tangential rms ) in columns 2 and 3 .
the mean breadth is listed in column 4 below as microns on the focal plane , and in column 5 as an angular size on the sky in milliarcseconds . at the inner and outer radii of the image annulus ,
the fwhm becomes as large as 60 milliarcsec , although in the midrange of the annulus it is smaller .
cccccc 0.006 & 3.32 & 1.60 & 2.62 & 24.9 & -838 + 0.007 & 3.33 & 1.60 & 2.62 & 24.9 & -782 + 0.008 & 3.18 & 1.59 & 2.53 & 24.1 & -630 + 0.009 & 2.83 & 1.51 & 2.28 & 21.7 & -373 + 0.010 & 2.28 & 1.37 & 1.89 & 18.0 & 0 + 0.011 & 1.57 & 1.35 & 1.47 & 14.0 & 509 + 0.012 & 1.18 & 1.89 & 1.58 & 15.0 & 1165 + 0.013 & 2.09 & 3.23 & 2.73 & 26.0 & 1983 + & & average= & 2.22 & 21.0 + distortion is another fundamental optical aberration , but it does not impact the @xmath41 nor does it directly impact the detection of supernovae unlike the other seidel aberrations distortion .
it does however cause the loci of scanned field objects to depart from parallel tracks in the focal plane , and does complicate the weak lensing science . in our reference design ,
we have disregarded distortion as a driver , in order to use all available design variables to maximize the working field of view and minimize the net geometrical blur .
it is nonetheless important to explore the resulting distortion quantitatively .
the telescope mirror assembly ( tma ) distortion is axisymmetric owing to the symmetry of the unfolded ( powered ) optical train , and in polar coordinates any off - axis angle maps onto a single focal plane radius independent of azimuth angle .
the distortion is therefore purely radial .
column 6 of table [ moment : tab ] lists the radial distance of an off - axis field point as a function of the sine of the off - axis angle , and the departure from proportionality to the sine of that angle .
the tma distortion is of the pincushion type , having increased magnification toward the extremity of the field .
compared to a linear mapping of @xmath120 onto focal plane radius , the distortion amounts to about two percent . for a star at infinity and a telescope focused at infinity ,
the pupil diffraction pattern is computed using the fraunhofer formalism .
the focal plane irradiance is simply the square of the modulus of the two - dimensional fourier transform of the pupil .
we have computed this irradiance function and diffracted light background for a variety of prospective metering structures .
a comprehensive stray light control plan has been developed for snap .
the goal is to keep all stray light sources far below the natural zodiacal irradiance level as seen at the focal plane .
the primary concern is sunlight scattered past the forward edge of the outer light baffle .
this requires a minimum of two successive forward edges , since the light diffracted past a single edge would exceed the allowable irradiance at the primary mirror , assuming typical mirror scattering values .
figure [ baffle : fig ] shows our treatment of the outer baffle interior vane arrangement .
the mission reference l2 orbit simplifies the stray light design by eliminating the possibility of sun and full earth light entering the main baffle from opposite directions .
analysis efforts are underway to quantify the effects of both out of band light rejected by the focal plane filters , as well as ghosting and internal reflections within the filters .
a systems engineering approach is being pursued , along with development of a detailed stray light budget for both solar and stellar contamination sources .
the departure of any surface from its nominal mathematical conic section or the misplacement or misorientation of any of the surfaces causes a wavefront error and a degraded image quality .
one measure of this degradation is the telescope s strehl ratio .
the strehl ratio can be converted into rms wavefront error ( rms wfe ) through marechal s relation . to achieve a system strehl ratio of 0.77 at 0.633 @xmath38 m wavelength ,
the total wfe must not exceed 0.05 nm rms .
this allowed wfe will be apportioned into individual contributions as part of the detailed telescope design .
a tolerance budget has been developed based on a group of exploratory studies of the sensitivity of the geometrical spot size to variations in element curvature , shape , location , and orientation .
these calculations show that by far the single most critical parameters are the primary mirror curvature and the spacing between primary and secondary mirrors expected given the fast ( @xmath121 ) primary mirror . a two - micron displacement of the secondary piston , or a two - micron displacement in the virtual image created by the primary mirror is found to increase the rms geometrical blur by about 3 @xmath38 m .
similarly , a 15-@xmath38 m lateral displacement or a 15-microradian tilt of the secondary mirror causes a 3-@xmath38 m growth in the rms geometrical blur .
the reference model snap telescope includes on - orbit mechanical adjustments that permit the relocation and reorientation of the secondary mirror , and possibly the tertiary mirror as well , to optimize image quality . by means of these adjustments we anticipate accommodating small shifts in any of the optical elements locations and orientations , allowing for correction of geometrical blur .
the wide field of view of the snap imager allows simultaneous batch discovery and building of supernova rest - frame optical light curves , and over the mission lifetime will yield several thousands of supernovae at @xmath74 with the proposed photometric accuracy .
more distant , less precisely measured supernovae will also be available in our data set .
the multiplexing advantage with this large field reduces by more than an order of magnitude the total exposure times necessary to follow all the supernovae .
figure [ syslc : fig ] shows critical points on the light curve and the desired measurement accuracy that the snap imager must furnish .
we note that the stated @xmath41 need not be achieved with a single measurement but can be synthesized from multiple measurements , taking advantage of the substantial time dilation for high - redshift supernovae
. weak lensing by large - scale structure is a statistical study that requires the accurate measurement of the shapes of many galaxies . in order to efficiently survey a large area in a finite time ,
a large field of view covered by detectors with high quantum efficiency is necessary . given a fixed number of pixels in the focal plane , there are competing desires to have better resolution , which requires smaller pixel scales and wider sky coverage , which requires larger pixel scales .
the theoretical optimum of this trade - off requires that the pixel size be roughly matched to the size of the psf @xcite ; further plate scale and pixel size studies using both simulated and real data are currently in progress for optimizing photometric and shape - measurement accuracy while maintaining a large field of view . for the case of the state - of - the - art snap ccds in which the psf size is dominated by diffusion in the silicon at @xmath122 ,
the theoretical optimum requires a pixel size of roughly 0.1 arcsecond .
( diffusion would be worse in a conventional ccd that has a field - free region . ) in order to make precise , bias - free photometric redshift measurements , wide spectral coverage for the optical to near infrared is needed .
such photo - z measurements are crucial to interpreting the cosmological implications of the wide - field survey .
the aforementioned capabilities are necessary to measure the shapes of a significant number of faint galaxies beyond @xmath40 in order to study the evolution of structure .
crucially important systematic errors in shape measurements arising from unstable anisotropic psfs and systematic errors / biases in photometric redshift determination are significantly reduced in a space - based imager such as snap when compared to ground - based facilities .
the snap imager addresses the above requirements using two detector technologies to efficiently cover the wavelength range of 3500 to 17000 .
the visible region ( 3500 to 10000 ) is measured with lawrence berkeley national laboratory new - technology @xmath123-channel high - resistivity ccds @xcite which have high ( @xmath3380% ) quantum efficiency for wavelengths between 3500 and 10000 .
extensive radiation testing shows that these ccds will suffer little or no performance degradation over the lifetime of snap @xcite . in the reference design a pixel size of 10.5
@xmath38 m has been matched to the telescope focal length and diffraction limit at 10000 of 0.1 arcsec .
improved shape measurements and robust psf photometry come from a dithered observation strategy .
undersampling , dithering , cosmic ray hits , and many other effects are included in the exposure time calculator developed by @xcite .
the nir range ( 9000 to 17000 ) is measured with commercially available hgcdte arrays .
current large area ( 2048 @xmath43 2048 pixel ) hgcdte detector devices have a pixel sizes in the range of 18 - 20 @xmath38 m .
the telescope optics are designed to give an angular pixel size of 0.17 arcsec , matched to the telescope focal length and diffraction limit at 17000 .
the snap reference performance specifications require low read noise and dark current ( @xmath50 6 - 8 e@xmath124 for multiple reads and @xmath50 0.1 e@xmath124/sec / pix , respectively ) , high quantum efficiency ( @xmath12560% ) , and uniform pixel response .
reference imager detector specifications are given in table [ ref : tab ] .
the evolution of the snap focal - plane design is described in @xcite .
ccc field of view ( deg@xmath126 ) & 0.34 & 0.34 + plate scale ( arcsec ) & 0.10 & 0.17 + wavelength ( ) & 350010000 & 9000 17000 + @xmath127 ( % ) & 80 & 60 + read noise ( multiple reads ) ( e@xmath124/pixel ) & 4 & 68 ( multiple read ) + dark current ( e@xmath124/s / pixel ) & 0.002 & 0.1 + diffusion ( @xmath38 m ) & 4 & 5 + number of fixed filters & 6 & 3 + the minimum filter set required is primarily determined by the precision needed for the cross - filter @xmath49-correction , that is necessary when reconstructing the primary restframe band ( @xmath110-band in the reference design ) light from a set of laboratory - frame filter measurements .
six visible filters and three nir filters are sufficient if they are derived from a b - band filter with logarithmic @xmath128 scaling of their wavelength centers and widths .
the reference set that we consider here consists of nine redshifted @xmath110 filters logarithmically distributed in wavelength with effective wavelengths at @xmath129 for @xmath130 .
further optimization of the filter set for supernova science and photometric - redshifts is in progress . in particular
, the ultimate snap `` @xmath110 '' filter will not necessarily be a johnson @xmath110 although it is expected to have a similar central wavelength and width .
a fixed - filter design is preferred over a large filter wheel which would be a single point failure risk .
figure [ fplane : fig ] shows an array of visible and nir filters in the snap focal plane . to enhance the amount of nir light that is integrated , the individual nir filters have twice the area in the focal plane of the individual visible filters .
the constraint that the satellite be rotated in 90@xmath131 increments requires the filter pattern to remain symmetric with respect to two orthogonal axes .
consider the filter arrays in figure [ fplane : fig ] .
note that the arrays can be scanned though an observation field left - to - right , right - to - left , top - to - bottom , and bottom - to - top , and that a given star will be measured with each filter bandpass but not necessarily the same physical filter .
note that any 90@xmath131 rotation of the filter array can still measure the star field in all filter types .
as shown in figure [ fplane : fig ] , underlying each nir filter is one 2k @xmath43 2k , 18 @xmath38 m hgcdte device , 36 in total .
underlying each 2@xmath432 array of visible filters is one 3.5k @xmath43 3.5k , 10.5 @xmath38 m ccd , with 36 devices in all .
the total number of pixels is @xmath33600 million .
the snap focal plane will be passively cooled to operate at 140k , a habitable temperature for the ccds and sufficiently cold to achieve the noise requirements from the 1.7@xmath38m - cutoff hgcdte devices .
short flex cables penetrate the focal plane to bring the signals to the electronics located on the backside .
the present reference design envisions an asic - based readout that would also operate at 140k ; this avoids routing low - level analog signals long distances and reduces the size of the cable plant between the cold focal plane and the warm data acquisition electronics located to the side .
a four - channel custom chip will read out each ccd @xcite .
it consists of a single - ended to differential converter followed by a correlated double sampler and a novel multi - slope integrator .
this circuit is designed to operate at room temperature for test purposes and at 140 k. the readout speed is 100 khz .
the 16-bit dynamic range is covered using 3 gains each with a 12 bit signal to noise ratio .
a 12-bit pipeline adc digitizes the analog data . in the deep survey ,
multi - band exposures of a point of sky will be achieved by shift - and - stare observations in which the pointings are shifted by the 2.9 arcmin width of the optical filters .
each pointing will consist of four 300-second exposures ; the multiple exposures are for cosmic - ray rejection and dithering of our undersampled pixels .
the fixed filters prevent independent tuning of exposure times for different filters . within a scan
, over one hundred pointings are required to cover the 7.5 square degrees in all filters to the desired depth .
a scan of the north ( south ) field will be repeated every four days for 16 months for a total 120 scans . in the wide ( and panoramic ) surveys , long adjacent strips
will be observed using the shift and stare method .
six rather than four dithers will be made per pointing so as to fill the gaps between the detectors and provide robust galaxy - shape measurements .
observed patches of sky will not be revisited .
the exact shapes of the fields are under review .
the spectrometer is used to make a positive identification of type ia supernovae by observing the characteristic siii feature at 6150 , measure the features associated with supernova heterogeneity , determine the redshift of the underlying host galaxies and supernovae , build a template spectral library of sne ia , and link standard stars with the snap photometric system .
the specific signature of type ia supernovae is the siii line at @xmath132 ( rest frame ) . in models ,
the strongest sensitivity to the metallicity in the progenitor system lies in the rest - frame uv band .
the need for both supernova - frame uv and si features defines a broad wavelength range , 4000 - 17000 that must be covered by the instrument to observe @xmath133 supernovae .
the broad features of the supernova spectra and the non - negligible detector noise contribution for the faintest objects make a low - resolution spectrograph optimal : studies @xcite find an optimal resolving power @xmath134 at fwhm and 1 pixel per fwhm sampling , with constant resolving power in the 0.6 - 1.7@xmath38 m range .
other types of supernovae , such as ii or ib , have lines of h or he in the same wavelength range with similar equivalent widths , allowing supernova classification of all possible candidates with one spectrograph . the silicon feature and other features in the spectrum of a typical type ia supernovae , shown in figure [ sysspec : fig ] , are to be carefully observed . their characteristics ( position , width , height , etc . )
are related to the supernova peak magnitude through physical parameters such as temperature , velocity and progenitor metallicity .
the field of view must include the underlying galaxy in order to determine its spectrum during the same exposure .
this is necessary for subtraction of the host spectrum from the spectrum in the supernovae region , and for an accurate determination of supernova redshift . to avoid single - point failure in the spectrograph we desire two detectors in each focal plane .
the detectors cover 3 @xmath43 6 split into 40 slices that can easily cover the mean size of galaxies at redshift 1 - 2 .
the large field of view allows the pointing requirements to be relaxed .
the main spectrograph specifications are summarized in table [ spec : tab ] .
see @xcite for details on the design process .
ccc wavelength coverage ( @xmath38 m ) & 0.35 - 0.98 & 0.98 - 1.70 + field of view & 3.0 @xmath43 6.0 & 3.0 @xmath43 6.0 + spatial resolution element ( arcsec ) & 0.15 & 0.15 + number of slices & 40 & 40 + spectral resolution , @xmath135 & 100 & 100 + the requirement for simultaneous acquisition of supernova and host spectra , the ability to subtract off the host - galaxy spectrum after the supernova has faded , spectrophotometry for calibration purposes , and the high object acquisition precision that would be needed for a traditional long slit spectrograph , make a 3d spectrograph an attractive option .
a 3d spectrograph reconstructs the data cube consisting of the two spatial directions x and y plus the wavelength direction . for each spatial pixel ,
the spectrum is reconstructed .
two principal techniques are indicated for 3d spectroscopy : first , the use of a fourier transform spectrometer ( fts ) , and second , the use of integral field spectroscopy .
trade studies indicate that integral - field spectroscopy using the image slicer technique is preferred for snap .
this technique , developed since 1938 in order to minimize slit losses , is very powerful @xcite .
the new generation of image slicers improves the efficiency and the compactness of the system and appears to be a very well adapted solution for the snap mission . figure [ slicer : fig ] shows the principle of this technique .
the field of view is sliced along @xmath136 ( in the drawing @xmath137 , for snap @xmath138 ) strips on a slicing mirror , consisting of a stack of @xmath136 plates where the active surface is on an edge .
each of the @xmath136 slices re - images the telescope pupil , creating @xmath136 telescope pupil images in the pupil plane .
thanks to a tilt adapted to each individual slice , the @xmath136 pupil images lie along a line . in the pupil plane , a line of `` pupil mirrors ''
is arranged .
each pupil mirror is placed on a pupil image and reimages the field strip .
these images are arranged along a line and form a `` pseudo - slit '' . at this stage , therefore , there is an image of each of the @xmath136 strips of the field of view .
the pseudo - slit is placed in the entrance plane of the spectrograph , acting as the entrance slit .
a last line of mirrors is placed on the pseudo - slit .
this line adapts the output pupil of the slicer into the input pupil of the spectrograph .
the spectrograph components are summarized in the block diagram shown in figure [ sblock : fig ] with the principal components described below .
this unit is the interface between the telescope beam and the instrument .
some telescope aberrations can be corrected within this optical system .
a simple three - mirror configuration should be sufficient to allow picking off the beam at a point most convenient for the spectrometer .
the slicer unit acts as a field reformatting system .
as described above , the principle is to slice a 2d field of view into long strips and optically align all the strips to form a long spectrograph entrance slit .
the slicing mirror is comprised of a stack of slicers .
each slicer has an optically active spherical surface on one edge .
a line of pupil mirrors does the reformatting .
each pupil mirror sends the beam to a slit mirror , which adapts the pupil to the entrance of the spectrograph .
the long thin active surface of each individual slicer will produce a large diffraction effect . in order to minimize flux losses to a few percent , the spectrograph entrance pupil must be oversized .
a combined theoretical and experimental approach is underway at laboratoire dastrophysique de marseille to define the optimum entrance pupil for a jwst application ( in the infrared bands 1 - 5 @xmath38 m ) .
this will be directly adapted for the snap concept .
the reference requirements on the slicer unit are an accuracy of @xmath139 rms on the optical surfaces and a surface roughness of @xmath140 rms .
existing prototypes fully meet these specifications .
thanks to the moderate beam aperture and field of view , the spectrograph optics will be straightforward .
the reference is a classical dichroic spectrograph : one collimator mirror , one prism with a dichroic crystal , and two camera mirrors are required .
using spherical shapes for all the mirrors would provide an adequately sharp image , but using aspheric mirrors will make it possible to have a very compact spectrograph .
the prism solution is well matched to the requirement of a constant resolution over the whole wavelength range .
the dichroic crystal allows the coverage of two channels simultaneously : one for the visible ( e.g .. 0.350.98 @xmath38 m ) and one for the infrared ( 0.981.70 @xmath38 m ) . in the visible ,
the main goals are high quantum efficiency and very low noise . given concerns over degradation due to radiation exposure and the poor performance of conventional thinned ccds in the red part of the visible ,
the fully - depleted lbnl ccds are the detectors of choice
. thinned , back - side illuminated , deep depletion , low - noise conventional ccds of 1024 @xmath43 1024 pixels are an alternative option . in the nir ,
some factors constrain the detector technologies .
the operating temperature will be dictated by the detector dark current .
hgcdte arrays with cutoff wavelength of 1.7 @xmath38 m are currently under consideration which allow operation in the 130 - 140 k range .
a detailed list of the performance specifications for the detectors is provided in table [ specdec : tab ] . to achieve the listed performance in read noise and dark current
, a multiple sampling technique is required . to optimize exposure time
, the impact of the rate of cosmic rays on the spectrum quality is under study .
ccc detector size & 1k @xmath43 1k & 1k @xmath43 1k + pixel size ( @xmath38 m ) & 10 - 20 & 18 - 20 + detector temperature(k ) & 140 & 130 - 140 + @xmath141(% ) & @xmath12580 & @xmath12560 + read noise(e@xmath124/pixel ) & 2 & 5 ( multiple read ) + dark current(e@xmath124/pixel / s ) & 0.001&0.02 + simulations of the efficiency of the instrument show a cumulative efficiency of the instrument from the relay optic and including the detector at a level better than 50% in the visible and 40% in the infrared .
this excellent performance is due to high slicer efficiency and is based on a conservative value for the efficiency of individual mirrors ( 98% ) , much lower than is expected from a silver coating at these wavelengths .
the principal losses are from the prism ( also conservatively estimated at 80% ) , and from the detector . during the reference deep survey , the observation time of snap
is partitioned roughly 60% to preprogrammed , stepped photometry and 40% to targeted spectroscopy with parallel photometry .
the former is divided into a sequence of 300s exposures followed by 30s of sensor readout .
the latter is comprised of exposures varying from a few seconds to a thousand seconds . for the longer exposure times , the spectrograph nir detector is continuously read in up - the - ramp mode to allow cosmic - ray rejection .
imaging is possible during spectroscopic exposures ; the imager is read as in photometry mode except that the exposure times vary , determined by the time to achieve a required @xmath41 in the spectrograph for a particular supernova . for each imaging exposure , the ccd devices produce a single frame of correlated double sampled data while the nir devices produce two frames ( each containing the average of multiple reads ) , one for post reset values and the other for post integration values so that the digital correlated double sampling can be done on the ground . the operating concept for processing the raw data is to perform only lossless compression of the frames in the satellite .
a factor of two is easily done ; a greater compression factor is probably achievable .
the primary motivations are to be able to look retrospectively into the data to better determine a supernova explosion time and to be able to co - add many months of data for weak lensing science to extract weak signals from the per frame noise . with this approach
we avoid developing flight software for automated acquisition and processing of reference , dark and flat frames and applying them irrecoverably to data frames .
the consequence is that in l2 snap will require about 133 gb of flight data storage for a 24 hour contact period through the deep space network capacity .
we anticipate that dsn will have the ability to receive 300 mbit s@xmath44 ka - band data by the time snap launches .
the current reference weak lensing survey @xcite calls for a 1-year 1000 square degree survey consisting entirely of imaging . at l2 ,
6 dithered exposures per optical filter would be taken for a total of 1900 s. such a survey would require a daily four hour download of data and would use only @xmath142 gb of on - board storage .
a wider survey could be taken if more time were given to the survey or if the exposure time per filter were lowered .
the latter option would require that the data download time be accordingly raised to compensate for more exposures or that the on - board compression of the data be increased beyond the factor of two that can be achieved through lossless compression .
an important feature of the supernova - cosmology analysis is its dependence on the _ relative _ brightness of type ia supernovae .
cosmological and dark - energy parameters are determined from the shape , and not the absolute normalization , of the hubble brightness - redshift relationship . in order to constrain the possible effects of population evolution , snap will use homogeneous subsets of sn ia .
identification of these sn ia subsets depends in part on the brightness and color evolution of supernovae .
each supernova redshift @xmath34 is plotted against its restframe b band flux which depends on the k - correction and the extinction correction , which in turn depend on the color of the supernovae . as a result ,
the analysis relies on the _ color _ , where we use this term to mean the flux ratios of calibrated passbands .
the snap calibration program must thus precisely control the absolute _ color _ calibration , not necessarily the absolute flux calibration , in the @xmath143 @xmath38 m observing window
. the snap calibration program will provide fluxes in physical units associated with the same restframe passband for a homogeneous set of supernovae distributed out to @xmath94 .
calibration provides the transformation of the measured signal in a given passband into a flux associated with a calibrated passband .
calibration affects flux - correction for extinction due to the milky way , the sn host galaxy , and the intergalactic medium , and also the k - corrections which provide the transformation between fluxes in the observed and restframe passbands . in the snap cosmological - parameter determination
, calibration enters early in the analysis and consequently the calibration uncertainty propagates non - trivially throughout . depending on its nature and size
, the calibration uncertainty can contribute significantly to the @xmath144 error budget .
the propagation of these uncertainties has been explored in @xcite for a simplified blackbody calibration model .
realistic calibration uncertainty will be more complex and the snap calibration methodology is being crafted to ensure that calibration does not impede achievement of the science goals .
the snap reference calibration concept consists of several programs which are subject to ongoing work examining modifications , extensions , and alternative strategies .
a set of astronomical sources with known physical fluxes ( or more importantly known physical colors ) are established as `` fundamental standards '' .
a network of fainter primary and secondary standard stars , accessible to snap and larger ground - based telescopes , are established with respect to the fundamental standards .
the snap observatory has an observation and analysis program that takes standard star observations and derives a transformation between snap signals and fluxes in the instrument passbands .
these programs are described in detail in the following subsections . in the ideal situation , target supernovae
would be calibrated directly against a true irradiance standard observed through the same optical chain . however , because of the unavailability of convenient irradiance standards observable through the same optical path as the supernovae , a suitable , calibrated proxy ( or proxies ) must be identified .
consequently snap must use stars as irradiance and color standards ( standard candles ) .
these selected stars may be established as standards using calculations of model stellar atmospheres as a substitute for a nist - traceable irradiance standard .
a few stars may be unreddened and easy to model , so that all or part of the standard candle flux distributions can be replaced by a suitably normalized model atmosphere calculation .
traditionally in optical astronomy the bright , @xmath145 star vega has been used as the nist stand - in . however , vega itself is far too bright to be used for most astronomical programs , so it has been used to calibrate fainter stars e.g. the landolt standards , oke standards etc .
the snap calibration program has identified `` fundamental standard stars '' to be bright , @xmath146 , stars whose spectral energy distributions are independently determined . at @xmath147 ,
the fundamental stars may be observed with the snap spectrograph .
the current hubble space telescope fundamental flux standards are a set of three wd stars and could be a good choice for the basis of the snap flux calibration .
these stars have pure hydrogen atmospheres and are too hot for molecular constituents , which means that the physics of the model calculations are relatively straightforward .
the three models are normalized near 5500 using landolt v band magnitudes .
the hst program ( @xcite also provides three primary solar analog standards , whose optical flux has been measured on the wd flux scale and whose ir flux shape is solar .
the stellar model approach is a tested method that yields a precision of @xmath148% in the uv and optical , while the near infrared relative fluxes are less well tested .
the systematic flux errors in the wd model technique have been estimated by the difference between the two extreme models that still fit the observations .
this uncertainty reaches a maximum in the snap wavelength range of 1.8% at 1.7@xmath38 m , meeting the goals of the reference mission .
one possibility for further reducing the uncertainties is to verify that the wd model @xmath149 and @xmath150 values derived from the balmer line profiles fit uniformly for the lyman and paschen lines , as well .
a second verification in progress with nicmos is to confirm to @xmath151% the relative ir fluxes of the models and the solar analogs .
the direct method of determining stellar fluxes is to compare observations of stars , particularly in the near infrared , with a lamp source . as stated in @xcite
`` in order to improve the accuracy of the standard star ir flux distributions , some standard stars should be compared to an actual standard ir lamp to define a proper absolute flux calibration , just as was done for vega .
such fundamental data are lacking ''
these authors normalize the flux scale at one bandpass once the spectral shape has been determined , while a lamp comparison determines both the absolute level and the shape of the spectral flux distribution .
the absolute calibration of vega @xcite exemplifies a direct method of establishing a fundamental calibrator .
the need and approach for such an experiment is under study by the snap calibration team . in order to minimize the need for on - orbit calibration time ,
a ground - based standard star network consisting of the fundamental standards linked to primary and secondary standards will be established .
this network will provide the preliminary calibration data for the spacecraft by setting the spectrophotometric zeropoints .
further , the network will enable parallel and follow - on observation from the ground to support on - orbit science observations .
the network will establish an all - sky photometric system covering the spectrum from 3500 to 17000 and spanning approximately 6 magnitudes ( 12 to 18 mag ) of brightness .
primary standards are fainter ( @xmath152 ) than the fundamental standards and will be calibrated with respect to the fundamental standards through spectrophotometry and photometry .
they will be used to transfer the calibration to an `` all sky '' grid .
some will lie in the snap science fields and others around the celestial equator to tie the northern and southern snap fields together and provide standards distributed over 24 hours .
the primary standards will be tied to the secondary standards through spectrophotometry and photometry .
secondary standards will be in the snap fields .
they will span a wide color and magnitude ( @xmath153 ) range , and hence spectral type .
uncertainties will be minimized by multiple observations of the fields . once in orbit
, we will verify the zeropoints for both the primary and the secondary standard stars .
the planned repeating observation cadences will allow typically 140 observations of every star in the snap - n and snap - s fields over the course of the mission , leading to refinement of the brighter secondaries and establishment and verification of the fainter secondaries .
snap is developing a flux - based ( rather than magnitude - based ) plan for transferring the fundamental standard star calibration to the primary and secondary standards , and ultimately to supernovae through a combination of spectroscopy and photometry .
there are two common choices of calibration passbands .
traditionally in observational astronomy a set of passbands is associated with a specific magnitude system .
these ( virtual ) passbands , when used to observe the system s primary standard stars , reproduce exactly the tabulated magnitudes .
@xcite has produced throughput realizations of the johnson - cousins magnitude system .
another approach is to use observations to infer the actual throughput of the observers optical system ( including atmosphere ) .
snap can use both approaches in its calibrations .
further , as the demand for precise calibration is fairly stringent , snap needs to be able to account for changes in the instrument performance .
for example , we intend to re - determine the transmission properties of the photometry filters during the mission by taking advantage of the multiple observations of stars in the snap fields provided by the 4-day observing cadence .
internal calibrated , stable illumination sources may also be utilized .
the design of the snap reference mission has been guided by the fisher - matrix analysis described in [ asprobe ] .
a detailed simulation of the reference mission now provides validation of the scientific reach of the snap mission and provides a tool for further mission refinement and optimization .
we have developed a detailed parameterized simulator of supernova cosmology missions .
we describe here its use in projecting the scientific yield from the snap supernova reference mission described in this paper .
a similar but more complete pixel - level simulation will be used to test the mission - analysis software . for the telescope / camera configuration and observing plan described in the previous sections ,
we generate a list of type ia supernovae and their associated host galaxies in the observed field of view during the course of the survey .
the incident flux at the telescope is calculated for each supernova , accounting for the underlying cosmology , magnification from gravitational lensing , and host and galactic dust .
the photometric observations of the supernovae from the snap scanning strategy are simulated , resulting in multi - band light curves for each event .
the rest - frame @xmath110 and @xmath154 light curves are fit to sn ia - class templates while the light curves in other filters are fit to a polynomial , in order to measure the magnitude , color , and light - curve parameters of each supernova .
triggered spectroscopic observations are simulated through fisher - matrix techniques @xcite which predict the quality of the spectroscopic parameter measurements .
supernova observables are related to the intrinsic peak absolute magnitude using the models described in @xcite . in this manner ,
supernova s light - curve and spectral parameters are then used to determine its distance modulus , along with host - galaxy extinction ( @xmath155 and @xmath156 of the @xcite dust model ) and its individual absolute magnitude .
given the reference snap supernova distribution and expected distance moduli folded with the expected supernovae from the nearby supernova factory , we fit the cosmological parameters . while this can be done for a general set of parameters , we concentrate on the dark - energy parameters @xmath6 and @xmath9 assuming a flat universe and a fiducial model with @xmath157 and @xmath158 ( recall we define @xmath159 ) .
the standard systematic - error model is adopted .
marginalizing over all other parameters , we get uncertainties in @xmath6 of 0.07 and @xmath9 of @xmath160 for the prior 0.03 uncertainty in @xmath14 . if we take no prior in @xmath14 but assume cosmological constraints expected from planck , the @xmath9 uncertainty improves to @xmath161 .
these results are somewhat better than those determined in the fisher analysis shown in table [ sci_errors.tab ] where all supernovae are assumed to have identical distance - modulus uncertainty . by setting exposure times based on light - curve requirements for the highest redshift supernovae ,
the lower redshift events have better @xmath41 with the fixed filter design . to explore how snap
will be able to distinguish between different dark - energy models , we simulate constraints on @xmath6 and @xmath9 choosing as fiducial models instantiations of two postulated theories that account for an accelerating universe : a cosmological constant and a supergravity - inspired model .
using monte carlo we generate 68% confidence contours for snap supernovae , in conjunction with the data anticipated from the planck mission , and snap supernovae and weak lensing .
( figure [ w0w:fig ] ) .
we marginalize over the absolute magnitude and @xmath14 with no prior imposed on either . even with just snap supernovae
, the chosen models are distinct , with the fiducial values lying well outside the 68% contours of the other model .
inclusion of snap weak - lensing power spectrum , bispectrum , and cross - correlation cosmography information provides improved sensitivity .
snap and complementary data will provide important guidance to the true nature of dark energy . we have done the most detailed simulation of a weak lensing survey to date by identifying instrumental aspects of snap that will affect the psf , applying this psf to simulated images , and calculating the lensing efficiency and subsequent constraints on cosmological parameters from the snap weak lensing survey .
the results of these studies are in @xcite .
the cosmological reach of the snap weak - lensing program presented in this paper is based on these simulations .
we are continuing to develop the image simulations for more accurate results and to better identify instrumental factors that might be tweaked for gains in sensitivity to weak lensing .
we also are exploring the impact of new analysis techniques , e.g. bispectrum measurements of the mass distribution and cross - correlation cosmography , for measuring dark - energy parameters @xcite .
the snap surveys will have an unprecedented combination of depth , solid - angle , angular resolution , temporal sampling , and wavelength coverage from the optical to the nir . the wealth of potential science from deep and wide - field surveys has already been demonstrated in such surveys as the hubble deep fields @xcite and the sdss @xcite . in [ otherde : sec ] we summarized how the snap data set will contribute to our understanding of dark energy .
we here explore the properties of the surveys and their potential scientific yield in other areas . based on the telescope and camera characteristics ,
we quantify the expected depth , solid - angle , and time resolution of the snap supernova and weak lensing surveys .
the telescope and camera properties of snap have been modeled and incorporated into an advanced exposure - time calculator ( etc ) @xcite .
besides having all the usual capabilities of a standard etc , our etc includes unique handling of the pixel response function , undersampling , dithering , and probabilistic cosmic - ray rejection . as mentioned in [ imager : sec ] , snap will use dithering to obtain spatial resolution below the pixel scales .
the high cosmic - ray flux in a space environment produces a non - trivial reduction of effective exposure times ; pixels from a single exposure that are contaminated by a cosmic ray are recognized through median and other filtering and dropped in the dithered reconstruction .
the expected contamination probability of a point source is 0.0002 per second .
short individual exposure ( plus readout ) times limit the contamination : four 300 second exposures give better than a 99% probability that there will be no cosmic - ray contamination at a given position for at least one of the dithers that make up a pointing .
the magnitude depths for individual scans and co - added images of the snap supernova fields are calculated for each filter .
table [ depth : tab ] shows the limiting @xmath162 ab magnitude for each filter in the surveys , for point sources that are not contaminated with cosmic rays . as with the hst ,
the limiting magnitude for any given point is probabilistic , due to the random occurrence of cosmic rays .
the probability for a given level of cosmic contamination ( and approximate degradation in signal - to - noise ) can be determined using the binomial distribution . in a single scan of the deep survey 77% of point sources
will have no cosmic contamination in any of the four dithered images while 98% will have one or fewer cosmic - contaminated image , only slightly reducing signal - to - noise .
cccccccc 1 & 4400 & 1000 & 27.9 & 30.6 & 28.3 & 26.7 + 2 & 5060 & 1150 & 27.9 & 30.6 & 28.3 & 26.8 + 3 & 5819 & 1322 & 27.8 & 30.5 & 28.2 & 26.8 + 4 & 6691 & 1520 & 27.8 & 30.4 & 28.1 & 26.8 + 5 & 7695 & 1749 & 27.8 & 30.4 & 28.1 & 26.8 + 6 & 8849 & 2011 & 27.7 & 30.3 & 28.0 & 26.7 + 7 & 10177 & 2313 & 27.5 & 30.1 & 27.8 & 26.6 + 8 & 11704 & 2660 & 27.5 & 30.1 & 27.8 & 26.6 + 9 & 13459 & 3059 & 27.4 & 30.0 & 27.7 & 26.5 + the snap observing strategy provides remarkably even depth over the range of filters . for a given filter , _ individual scans _ of the deep and wide - field surveys
are only @xmath163 magnitudes shallower than the hubble deep fields ( hdfs ) while the deep fields co - added over time are @xmath164 magnitudes deeper than the hdfs @xcite and up to 0.5 magnitudes deeper than the udf depending on the filter .
snap has the additional advantage of having nine filters observing to this depth , compared to the four filters of the hdfs , and 9000 times the area in the deep survey ; when these data from all filters are combined , the limiting magnitude increases by another 0.6 magnitudes .
the wide - field survey has @xmath165 times the area of the hdfs .
snap fields will contain many faint diffuse galaxies whose detection is important for the weak - lensing studies , and for other potential science projects .
the limiting magnitudes for gaussian - aperture photometry of an exponential - disk galaxy with fwhm=0.12 are shown in table [ depth_diff : tab ] .
cccccccc 1 & 4400 & 1000 & 26.4 & 29.0 & 26.8 & 25.2 + 2 & 5060 & 1150 & 26.3 & 28.9 & 26.7 & 25.3 + 3 & 5819 & 1322 & 26.3 & 28.9 & 26.6 & 25.3 + 4 & 6691 & 1520 & 26.2 & 28.8 & 26.6 & 25.3 + 5 & 7695 & 1749 & 26.3 & 28.9 & 26.6 & 25.3 + 6 & 8849 & 2011 & 26.2 & 28.8 & 26.5 & 25.3 + 7 & 10177 & 2313 & 26.2 & 28.8 & 26.5 & 25.3 + 8 & 11704 & 2660 & 26.2 & 28.8 & 26.5 & 25.3 + 9 & 13459 & 3059 & 26.1 & 28.7 & 26.5 & 25.3 in this section we give a brief outline of additional science topics likely to benefit from snap survey data .
this list is incomplete , and includes only a few of the most obvious ancillary science projects .
the sdss @xcite , hubble deep fields ( hdfs ) @xcite , and ultra deep field demonstrate the vast range of science enabled by wide , deep , and precise multi - band surveys .
snap will carry this tradition forward as the first wide field survey in space .
snap deep - survey fields will cover more than 9000 times the area of the hdfs , go even deeper , include 9 color observations , and provide significant time domain information .
the snap wide - survey field is comparable in size to the sloan southern survey and cfht legacy survey but includes diffraction limited imaging , and probes many magnitudes fainter in more filters .
a panoramic survey would cover almost a quarter of the full sky at slightly shallower depth .
the combination of depth , temporal coverage , wavelength range , diffraction - limited seeing , and wide field make snap imaging surveys uniquely powerful in the study of a wide range of objects and phenomena , which we outline below .
* galaxies within the 15 square degree deep survey area , snap will make accurate ( @xmath166 ) photometric redshift estimates for at least @xmath167 galaxies from redshift 0 to 3.5 , spanning more than 90% of the age of the universe .
statistical studies are possible with such a large sample , e.g. the determination of the galaxy luminosity function and color distributions as a function of redshift .
photometric redshifts can be estimated from the 4000 break for galaxies out to about @xmath168 . for galaxies at
still higher redshift , the simplest indicator is the lyman break . for snap ,
the lyman break enters the optical imager around redshift 3 . in principle
it can be followed using snap data beyond redshift 10 , allowing identification of extremely high - redshift galaxies .
the magnitude depth also allows discovery of low - surface - brightness and very high - redshift galaxies .
high - resolution images will provide a view of the internal structure of galaxies and their interactions with each other .
this data set , which will include morphological information for every object , will provide a unique opportunity to study the evolution of galaxies .
the flood of galaxy evolution papers based on the hubble deep fields points the way to what will be possible with the snap imaging data set .
* galaxy structure formation the wide - field surveys from snap offer an excellent opportunity to study galaxy clustering and its evolution at high redshifts ( @xmath169 3 ) .
this is a regime which is both especially interesting and rather poorly understood .
there are some measurements of galaxy clustering at high redshift @xcite , but these are severely limited by sample variance @xcite .
snap , increasing the size of high redshift surveys by several orders of magnitude , will be substantially less sensitive to cosmic variance .
photometric redshifts greatly enhance this work , allowing detailed study of the evolution of galaxy clustering .
* galaxy clusters galaxy clusters , the most massive bound objects in the universe , provide important probes of our understanding of structure formation @xcite .
constraining their formation and evolution is an important observational goal for the coming decade .
recent advances have overcome earlier limitations of optically selected cluster samples , essentially by using photometric redshift information to eliminate projection effects @xcite .
hubble volume @xmath170cdm simulations @xcite predict that the snap surveys will provide detailed information on tens of thousands of galaxy clusters with masses above @xmath171 m@xmath172 . with nir sensitivity
, snap will allow sensitive detection of galaxy clusters all the way back to their earliest appearance around @xmath173 .
the overall number of galaxy clusters is exponentially sensitive to the amplitude of initial density fluctuations . *
quasars quasars are identified in multi - color imaging surveys by their non - stellar colors .
this method has been shown by the sloan digital sky survey to be extremely effective at identifying quasars to redshift 6 and beyond .
sdss quasar discovery is limited to redshift 6.4 by the ccd sensitivity cutoff at 1.0 @xmath38 m @xcite .
the most distant sdss quasar , at redshift 6.42 @xcite , has a @xmath34-band magnitude of 20 . by probing to wavelengths 1.7 times greater , and
to depths 9 magnitudes fainter , snap will be able to detect quasars to redshift 10 , and to probe the quasar luminosity function to 100 times fainter than the brightest quasars .
snap s ability to identify diffuse objects associated with quasars may present interesting opportunities for the study of galaxy formation .
* gamma - ray burst afterglows current evidence suggests that gamma - ray bursts are associated with the collapse of massive stars which live short lives and die where they are born . as a result , grb s may trace the cosmic star formation rate . if so , there should be grb s essentially coincident with the epoch of formation of the first stars .
the most distant grb known occurred at redshift of 4.5 @xcite .
snap will be able to identify grb afterglows , and the orphan afterglows predicted by some models of beaming in grb s to @xmath174 .
such orphan afterglows may even be detected as backgrounds to the sne search .
* reionization history the universe became neutral at the time of recombination , around @xmath175 , and the thermal radiation from that epoch travels to us undisturbed as the cosmic microwave background radiation .
the lack of a gunn - peterson effect in the spectra of most quasars demonstrates that the universe was reionized at some time between @xmath175 and @xmath176 .
the source of the ionizing radiation is the subject of substantial speculation .
the recent discovery of an apparent gunn - peterson trough in the spectra of several @xmath177 sdss quasars may provide the first glimpse of the epoch of reionization @xcite . by identifying many quasars and galaxies to @xmath178 , snap will set the stage for mapping the epoch of reionization in unprecedented detail . in combination with ground based and jwst spectroscopy
, it will enable measurements of the proximity effect and studies of the spatial structure of reionization .
* transients / variables the discovery and observation of sne ia are the primary goals of snap , but transient `` backgrounds '' are interesting in their own right : quasars , active - galactic - nuclei , gamma - ray - burst optical counterparts , supernovae of other types , variable stars , and eclipsing binaries . of particular interest
to cosmology are time - delay measurements of the large expected number of strongly lensed variables @xcite .
gravitational microlensing studies of stars and quasars as probes of compact dark matter are also possible . * stars faint limiting magnitudes and excellent star - galaxy separation will yield faint dwarf and halo stars .
proper motion can be detected with high - resolution and a long time reference .
snap s accurate colors will yield excellent photometric parallaxes to all stars in the field .
the geometry and substructure of the galactic halo and disk in the direction of the snap fields can be mapped .
of particular interest would be a census of low - mass l and t stars and brown dwarfs throughout the milky way disk @xcite .
* solar - system objects the peculiar motion and parallax in the time - series data will facilitate the identification of local objects such as asteroids and kuiper - belt objects . snap time series data
will provide an excellent probe of faint , red objects in the kuiper belt and beyond .
a 2 - 3 month dedicated ecliptic survey with the snap survey would detect 1050,000 kuiper belt objects down to the size of the comet halley s nucleus . * weak gravitational lensing
the importance of weak gravitational lensing for dark energy is described in [ lensing : sec ] .
weak gravitational lensing is also a powerful tool in mapping the distribution of mass in the universe .
maps reconstructed from the observed shear field are sensitive to any mass along a given line of sight , regardless of its nature or state allowing dark matter mapping and the unbiased detection of galaxy clusters and their mass .
the high surface density of resolved background galaxies enables the construction of both two- and three - dimensional maps of the foreground dark matter distribution .
in addition to providing further constraints upon cosmological parameters , these maps trace the growth of structure in the universe and can be use to study the bias between mass and light through comparison with the galaxy distribution .
+ in recent years many groups have measured the shape distortions of field galaxies due to weak lensing and used these shear measurements to set constraints on cosmological parameters . in a number of weak - lensing surveys using ground and space - based data the second moment of the shear field
has provided a constraint on the matter power spectrum , in particular on the quantity @xmath179 , that is comparable to the constraints set by more traditional methods such as the abundance of x - ray clusters .
current weak - lensing surveys seeking to measure higher order moments of the shear field and the magnification of background galaxies promise to provide an independent measure of @xmath14 , thus breaking the degeneracy between @xmath14 and @xmath180 , where @xmath180 is the amplitude of mass fluctuations on the scale @xmath181 mpc .
+ precise photometric redshifts enable the making of 2-dimensional mass maps in multiple redshift slices , and even 3-dimensional mass reconstructions .
this redshift tomography is much less effective from the ground because of the lower surface density of resolved galaxies , and can not be extended as far .
space - based imaging is needed to resolve sufficiently many source galaxies at @xmath76 and follow the growth of structure at those redshifts .
furthermore , high - resolution images coupled with photometric redshifts permit the use of a recently formulated approach to 3-d mapping @xcite which recreates the full 3-d mass distribution by simultaneously considering both the shear estimator and the photometric redshift of a galaxy . applied to the snap deep survey ,
this technique will detect mass overdensities with a @xmath182 sensitivity of @xmath183 ( 1/5 the mass of the virgo cluster ) at @xmath184 . * strong gravitational lensing the high spatial resolution of snap nir observations will enable the discovery of a large number of new strong lenses @xcite .
the statistics from such a large , uniformly selected sample would be ideal to impose constraints on the density profiles of galaxies and clusters , as well as to constrain cosmological parameters beyond those of dark energy . the nir observations , which are much less sensitive to dust extinction within the lens galaxy , are especially important in regard to strong lensing observations .
@xcite find that assuming a flat universe , the statistical uncertainty on the mass density is @xmath185 , although uncertainty of the lens modeling are likely to dominate .
the sdss demonstrates clearly the way in which a wide - field survey can produce scientific yield well beyond its primary design goals .
individual objects found on sdss images are routinely observed spectroscopically at the largest telescopes in the world , fulfilling the historical trend of wide - field small - aperture telescope imaging feeding targets for large - aperture telescope spectroscopy .
the snap surveys will provide a similar opportunity in working with jwst and the next generation of ground - based extremely large telescopes .
the discoveries of recent years make this a fascinating new era of empirical cosmology , in which we are developing new technologies and instrumentation capable of addressing fundamental questions .
the extraordinary challenge of the precision and accuracy necessary to explore the properties of dark energy requires a new stringent data set in order to improve statistical uncertainties and control systematic uncertainties ; large numbers of supernovae over an extended redshift range with light curves and spectra measured to high signal - to - noise and deep , wide fields full of weak - lensed galaxies with high - resolution psfs with unprecedented stability .
based on these data requirements , we have developed a reference mission called the supernova / acceleration probe , consisting of a detailed observing program carried out with an optimized wide - field telescope and attendant imaging and spectrograph instruments .
the optical , mechanical , and thermal studies and analyses conducted to date indicate that the reference model is manufacturable and testable using proven techniques .
the imager will provide psf - controlled wide - field images of galaxies with efficient multiplexed supernova discovery and light - curve building while the spectrograph satisfies the performance requirements for measuring key spectral features of the faintest supernovae .
the reference model presented in this paper addresses the many requirements necessary for a high - precision supernova experiment , a psf - controlled weak - lensing experiment , and makes possible a range of further complementary dark - energy studies .
we use numerical simulations and trade studies to refine the mission concept and quantify the expected science yield of snap .
this paper will be followed up with further results of these studies and the progress in the optimization of snap .
although snap is designed as a self - contained experiment , its surveys can be used in conjunction with complementary ground - based and jwst - based studies that can further strengthen the reference snap mission .
snap presents a unique opportunity to probe the dark energy and advance our understanding of the universe .
moreover , the principal survey missions of snap will produce a cornucopia of observations capable of revolutionizing other areas of astrophysics and cosmology .
after completion of these primary missions , a guest survey program is envisioned filling the remaining satellite lifetime to allow the full potential of snap to be realized .
we thank masahiro takada for making detailed results of his work available .
this work was supported by the director , office of science , of the u.s .
department of energy under contract no .
de - ac03 - 76sf00098 . | the supernova / acceleration probe ( snap ) is a proposed space - based experiment designed to study the dark energy and alternative explanations of the acceleration of the universe s expansion by performing a series of complementary systematics - controlled astrophysical measurements .
we here describe a self - consistent reference mission design that can accomplish this goal with the two leading measurement approaches being the type ia supernova hubble diagram and a wide - area weak gravitational lensing survey .
this design has been optimized to first order and is now under study for further modification and optimization .
a 2-m three - mirror anastigmat wide - field telescope feeds a focal plane consisting of a 0.7 square - degree imager tiled with equal areas of optical ccds and near infrared sensors , and a high - efficiency low - resolution integral field spectrograph .
the instrumentation suite provides simultaneous discovery and light - curve measurements of supernovae and then can target individual objects for detailed spectral characterization .
the snap mission will discover thousands of type ia supernovae out to @xmath0 and will obtain high - signal - to - noise calibrated light - curves and spectra for a subset of @xmath1 supernovae at redshifts between @xmath2 and @xmath3 in a northern field and in a southern field .
a wide - field survey covering one thousand square degrees in both northern and southern fields resolves @xmath4 galaxies per square arcminute , or a total of more than 300 million galaxies . with the psf stability afforded by a space observatory
, snap will provide precise and accurate measurements of gravitational lensing .
the high - quality data available in space , combined with the large sample of supernovae , will enable stringent control of systematic uncertainties .
the resulting data set will be used to determine the energy density of dark energy and parameters that describe its dynamical behavior .
the data also provide a direct test of theoretical models for the dark energy , including discrimination of vacuum energy due to the cosmological constant and various classes of dynamical scalar fields .
if we assume we live in a cosmological - constant - dominated universe , the matter density , dark energy density , and flatness of space can all be measured with snap supernova and weak - lensing measurements to a systematics - limited accuracy of 1% . for a flat universe
, the density - to - pressure ratio of dark energy or equation of state @xmath5 can be similarly measured to 5% for the present value @xmath6 and @xmath7 for the time variation @xmath8 . for a fiducial sugra - inspired universe , @xmath6 and @xmath9
can be measured to an even tighter uncertainty of 0.03 and 0.06 respectively .
note that no external priors are needed . as more accurate theoretical predictions for the small - scale weak - lensing shear develop , the conservative estimates adopted here for space - based systematics should improve , allowing even tighter constraints . while the survey strategy is tailored for supernova and weak gravitational lensing observations , the large survey area , depth , spatial resolution , time - sampling , and nine - band optical to nir photometry will support additional independent and/or complementary dark - energy measurement approaches as well as a broad range of auxiliary science programs . |
avalanche behavior @xcite has been found in many first order phase transitions at low temperatures . when the external parameters are slowly driven , the transition , instead of occurring as a sharp change of the system properties at a certain point of the phase diagram , splits into a series of discontinuous jumps that link metastable states .
this gives rise to hysteresis loops , whose branches consist of a sequence of random steps . in many cases the sizes of such steps range from microscopic to macroscopic scales , distributed according to a power - law .
the phenomenon has been found to be associated with magnetic transitions @xcite , capillary condensation @xcite , martensitic transformations @xcite , and others @xcite . a crucial ingredient in order to observe such avalanches is that thermal fluctuations are very small compared with the energy barriers that separate transformed and untransformed domains .
for this reason , such first - order phase transitions have been called `` athermal '' @xcite or `` fluctuationless '' @xcite . within this context
the 3d - grfim with metastable dynamics is a prototype model : the complexity of what we call `` disorder '' in a real system is simplified into a series of quenched random fields , gaussian distributed with zero mean and standard deviation @xmath1 , that act on every spin of a 3d ising model .
in addition , one assumes that temperature is zero ( @xmath0 ) and provides the model with a particular metastable dynamics in order to study the evolution of the magnetization @xmath2 when the external field @xmath3 is swept .
the details of the dynamics were introduced by sethna and coworkers a decade ago @xcite .
the basic assumption is that the driving field rate is slow enough so that system relaxation can be considered instantaneous ( adiabatic driving ) .
such relaxations are the so - called magnetization avalanches .
after the introduction of the model several works @xcite described the basic associated phenomenology : ( a ) the existence of a disorder - induced critical point at @xmath4 associated with the change from a continuous to a discontinuous hysteresis loop and ( b ) the fact that within a large region around the critical point the distribution of avalanche sizes @xmath5 exhibits almost power - law behavior .
nevertheless , several questions remain unsolved , mostly related to the properties of the spanning avalanches which are responsible for the observed macroscopic discontinuities in the @xmath3-@xmath2 hysteresis loop and in the @xmath1-@xmath3 phase diagram .
more recently , a fss analysis of the number of avalanches and avalanche size distribution @xcite has revealed that the scenario is quite complex .
the study was restricted to the statistical analysis of the full set of avalanches recorded in a half loop , irrespective of the field values @xmath3 where such avalanches occur .
( the obtained statistical distributions are often called integrated distributions ) . a detailed study as a function of @xmath3
has not been done before .
let us summarize here the main results presented in ref .
@xcite , in order to introduce the notation .
according to the dependence on the system size @xmath6 of the average number @xmath7 and size distribution @xmath8 , avalanches can be classified into several categories , ( the subindex @xmath9 stands for the different categories ) as presented in table [ table1 ] .
the scaling behavior is written as a function of the variable @xmath10 which measures the distance to the critical value of the disorder @xmath11 .
its precise definition is given as a second - order expansion : @xmath12 with @xmath13 .
this expression was found to be the best choice for the collapse of the scaling plots .
the values of the different critical exponents are summarized in table [ table2 ] , together with the new exponents that will be computed in the present work . [ cols="<,^,<,<",options="header " , ] nevertheless
, we can provide an argument based on renormalization group theory that indicates that the critical points in the two models ( 3d - rfim in equilibrium and the 3d - rfim with metastable dynamics ) correspond to the same fixed point in a more general parameter space . within the framework of rg theory the critical surface ( or critical line )
is defined as the set of all points in the parameter space that flow to a certain critical fixed point when the renormalization group transformation is applied .
the variation of the tunable parameters of a model describes a `` physical '' trajectory in the parameter space .
according to these definitions , the critical point corresponds to the point where the `` physical '' trajectory intersects the critical surface .
the two models discussed here can be considered as particular cases of a more general model with the same 3d - grfim hamiltonian and the following @xmath0 adiabatic dynamics : when @xmath14 is varied , blocks of neighboring spins of size @xmath15 flip when such a flip represents an energy decrease .
the metastable dynamics introduced by sethna corresponds to @xmath16 ( only single spin flips are considered ) and the equilibrium model at @xmath0 ( exact ground - state ) corresponds to @xmath17 . the parameter @xmath18 is a new parameter that must be considered in the rg equations . since a critical point is found both with @xmath16 and @xmath17 , it is natural to assume that this is an irrelevant parameter .
thus , we propose the scenario presented in fig .
[ fig17 ] .
changing @xmath18 alters the position of the critical point , but not the critical exponents which correspond to the same critical fixed point .
numerical simulations of the 3d - grfim with @xmath19 dynamics will help in clarifying this picture . at present
we guess that the rg flow follows the arrows schematically indicated in fig .
[ fig17 ] . both the equilibrium critical point ( ecp ) and the metastable critical point ( mcp )
lie on the same critical surface ( cs ) .
in general a first - order phase transition occurs at the points in the parameter space that go towards a discontinuity fixed point when the rg transformation is applied .
@xcite we assume the existence of two discontinuity fixed points : the equilibrium discontinuity fixed point ( edfp ) and the metastable discontinuity fixed point ( mdfp ) .
the edfp controls the first - order phase transition in the equilibrium case ( @xmath20 and @xmath21 ) and the mdfp controls the first - order phase transition when @xmath22 .
all the points that flow towards any of the discontinuity fixed points define the discontinuity surface ( ds ) where the first - order phase transition occurs .
( -5,-3)(7,5 ) ( 0,0)(3,-1.5)(3.6,-1)(4.2,-0.4)(4.4,-0.15)(4.6,0.2)(4.8,1)(4.9,1.4)(5,2)(0,3 ) ( 4,1.7 ) ( -4,-2 ) ( -1.2,-0.6 ) ( -3.2,-1.6 ) ( -4.6,-2.25 ) ( 7,0 ) ( 6,-0.3)1 ( 6.8,0.3 ) ( 0,3 ) ( 0,1.8)(0,1.8 ) ( 0,0.8)(0,0.8 ) ( 0,3)(0,5 ) ( 0,4.2)(0,3 ) ( -0.2,5 ) ( 3,-1.5 ) ( 0.75,-0.375 ) ( 2.25,-1.125 ) ( 5,-0.5)(5,2 ) ( 1,-2.5)(6,0 ) ( 0,3)(5,2 ) ( 0,3)(1.25,2.75 ) ( 0,3)(3.75,2.25 ) ( 4.2,2.4 ) ( 5,2)(4.4,-0.15)(3,-1.5 ) ( 4.9,1.4)(4.8,1 ) ( 4.2,-0.4)(3.6,-1 ) ( 2.5,2.5)(2.25,0.75)(1.5,-0.75 ) ( -1,2)(-1.6,-0.15)(-3,-1.5 ) ( -1,2)2pt ( 0,3)(-1,2 ) ( 5,2)(-1,2 ) ( 0,3)(6,2.5 ) ( 6,2.5)2pt ( 6,0)(6,3 ) ( 0,3)(6,3 ) ( 5,2)(6,2.5 ) ( -0.1,3.2)ecp ( 5.2,2)mcp ( 3.2,-1.6)mdfp ( -0.1,0.2)edfp ( 0,3)(5,2 ) ( 0,0)(3,-1.5 ) ( 1,2.4)(2.4,1.8)(3.4,1.2)(4,0.6)(4.1,0)(3.05,-1.4 ) ( 1.4,0.2)(2,-0.3)(2.6,-0.8)(3,-1.5 ) ( 0.8,-0.6)(1.6,-1.2)(1.4,-2)(1,-2.4 ) another interesting question to be discussed is the determination of the correlation length @xmath23 in the 3d - grfim with metastable dynamics .
avalanches can be understood as the zero temperature fluctuations in the driven system .
is their average linear size related to @xmath23 ? .
the first thing to note is that we have found that avalanches display two different fractal dimensions ( and thus different associated @xmath24 exponents ) .
the @xmath25 and @xmath26 exponents , nevertheless , are the same for all the scaling collapses . for instance , this is illustrated by fig .
[ fig7 ] in this work and by figs . 8 , 9 , and 10 in ref .
thus the behavior of the correlation length is unique : @xmath27 with @xmath28 .
the fluctuations then can `` choose '' between two different mechanisms for propagation either with fractal dimension @xmath29 or with fractal dimension @xmath30 .
the second point to be considered is that @xmath23 can not be related to the size of the subcritical 3d - spanning avalanche since we have found that @xmath23 is finite below @xmath31 .
keeping these two observations in mind we propose that the correlation length is related to the average radius of the largest non - spanning avalanche .
below @xmath32 the existence of a compact subcritical 3d - spanning avalanche does not allow for the non - spanning avalanches to overcome a certain finite length and thus @xmath23 is finite . only at the critical point
does the subcritical 3d - spanning avalanche become fractal and allows for other spanning avalanches to exist and @xmath23 becomes infinite .
this behavior is much similar to what has been recently found in percolation .
@xcite we conjecture that some of the theorems that have been rigorously proven concerning the uniqueness of the infinite percolating cluster should be applicable to our case concerning the compact subcritical 3d - spanning avalanche .
the present results should be considered as an interesting stimulus to proceed with the analysis of percolation theory .
for instance , we propose checking whether the fractal dimension of the spanning clusters is the same as that of the infinite cluster at distances lower than the correlation length at the percolation threshold .
the results presented in this paper are mainly related to two topics in the 3d - grfim : firstly , the field dependence of the spanning avalanches , and secondly , the geometrical properties of the avalanches .
we have extended the fss hypothesis proposed in @xcite to properly take into account the field dependence of the number densities @xmath33 and of the bivariate distributions @xmath34 . when carrying out such an extension , it is necessary to introduce a new scaling variable @xmath35 and a new exponent @xmath26 related to the divergence of the correlation length when @xmath3 approaches @xmath36 ( @xmath37 ) .
we have also introduced a scaling hypothesis for the field @xmath38 at which the different avalanches concentrate and their standard deviation @xmath39 . from the scaling collapses corresponding to the 1d- and 2d - spanning avalanches
we have found @xmath40 .
the study of the 3d - spanning avalanches is more intricate as already shown in @xcite , where we propose the existence of two different kinds of 3d - spanning avalanches . in this paper
we have proposed two approximate separation methods for classifying these avalanches as subcritical or critical . using these methods
we have found that @xmath41 for both cases .
scaling enables the following behavior to be scketched in the thermodynamic limit : the 1d- , 2d- , and critical 3d - spanning avalanches only exist at the critical point @xmath42 , where their number densities are infinite .
in contrast , one subcritical 3d - spanning avalanche exists below @xmath43 and it occurs on the line @xmath44 ( eq .
( [ eq:18.1 ] ) ) . from the average mass @xmath45
we have obtained the fractal dimensions corresponding to each of the types of spanning avalanches .
this has allowed us to confirm independently the results in @xcite : @xmath46 for the 1d- , 2d- , and subcritical 3d - spanning avalanches and @xmath47 .
furthermore , the behavior for @xmath48 of the mass corresponding to the subcritical 3d - spanning avalanches indicates that the correlation length @xmath23 is finite below @xmath31 . as a consequence ,
we conclude that the line @xmath49 , where the discontinuity in the order parameter occurs , corresponds to a standard first - order phase transition line and @xmath23 only diverges at the critical point .
we acknowledge fruitful discussions with x.illa , b.tadi , ll.maosa , and a.planes . this work has received financial support from cicyt ( spain ) , project mat2001 - 3251 and cirit ( catalonia ) , project 2000sgr00025 .
this research has been partially done using cesca resources ( project csiccf ) .
p. also acknowledges financial support from dgicyt . | spanning avalanches in the 3d gaussian random field ising model ( 3d - grfim ) with metastable dynamics at @xmath0 have been studied .
statistical analysis of the field values for which avalanches occur has enabled a finite - size scaling ( fss ) study of the avalanche density to be performed . furthermore , direct measurement of the geometrical properties of the avalanches has confirmed an earlier hypothesis that several kinds of spanning avalanches with two different fractal dimensions coexist at the critical point .
we finally compare the phase diagram of the 3d - grfim with metastable dynamics with the same model in equilibrium at @xmath0 . |
it is well known that a trapped atomic bose - einstein condensate ( bec ) with attractive interactions is stable if the number of atoms in it is below a critical value , above which collapse occurs @xcite . beneath the collapse threshold ,
the bec can form stable wave packets in a one - dimensional ( 1d ) cigar - shaped " trap , which is tightly confined in two ( transverse ) directions , and is unbound along the longitudinal axis . in that case , the stability of bright solitons is provided by balance between the quantum pressure , alias matter - wave ( mw ) dispersion , and mean - field attraction . a single mw soliton khaykovich02 and trains of interacting solitons @xcite have been created in the cigar - shaped optical traps . however , while the trapping geometry was nearly one - dimensional , the solitons themselves were far from being 1d objects .
in particular , in ref .
@xcite , a stable soliton was only possible if its longitudinal size exceeded the transverse size by no more than @xmath2 ( note that the situation was affected by an expulsive axial potential , unavoidable in the specific experimental setup ) .
recently , it was shown that the proximity of the soliton to being a 3d object strongly affects its properties , such as the character of its motion @xcite and interactions @xcite .
in particular , it was demonstrated that a moving soliton immersed in a cloud of thermal atoms is subjected to a temperature - dependent friction force @xcite .
a collision between two solitons , which are by themselves stable , in a confined geometry may readily lead to collapse , if the total number of atoms in the soliton pair exceeds the above - mentioned critical value , and the phase difference between the solitons is ( close to ) zero @xcite .
the significance of the effective dimensionality of mw solitary pulses is further emphasized by the recent observation of formation of a set of nearly-3d mutually repulsive mw solitons ( with a phase shift of @xmath3 between them ) as a result of incomplete collapse in an attractive bec with the number of atoms _ several times larger _ than the critical value cornish06 , adams06 .
in addition to being a profoundly important object of fundamental studies , mw solitary waves are also natural candidates for applications , such as high - precision atom interferometry and quantum - information processing .
thus , a thorough understanding of deviations of their behavior from that of ideal 1d solitons is important in this respect too .
in this paper we report results of theoretical investigation of the shape of stationary mw solitons and binary collisions between them in the quasi-1d regime , with the aim to identify manifestations of nonsolitonic behavior due to the residual multi - dimensionality .
the effect of the tightly confined transverse dimensions is taken into account through a perturbative _ self - focusing _ quintic term added to the corresponding one - dimensional gross - pitaevskii equation ( gpe ) , as per refs . @xcite and brand06 . in section
ii , we introduce this extended gpe , find a family of its exact one - soliton solutions , and demonstrate stability of the _ entire family _ , _ _
_ _ despite the fact that collapse occurs in the 1d equation with the quintic self - focusing term . in section iii , we investigate soliton collisions within the framework of this equation . on the contrary to completely elastic collisions between solitons in the cubic gpe ( alias cubic nonlinear schrdinger equation , nlse ) , in the presence of the quintic term colliding solitons with zero phase difference , @xmath4 , merge into a single pulse if their relative velocity is smaller than a critical value , @xmath5 .
we find the dependence of @xmath0 on the strength of the transverse confinement and number of atoms in the solitons .
for moderate quintic nonlinearity , good agreement with an analytic prediction derived from the perturbation theory is found . with a stronger quintic term ,
the numerical results deviate from the perturbation theory , although not dramatically .
finally , we demonstrate dynamical symmetry breaking between identical solitons colliding with @xmath6 ( in that case , the merger does not occur ) , as a function of the relative velocity .
an explanation to the latter effect is proposed .
it is based on estimation of a symmetry - breaking parameter , which is a mismatch between the _ amplitude center _ and _ phase center _ of the soliton pair with @xmath7 ( exact definitions are given below ) .
reasonably good agreement between numerical results and the analytical approximation is observed .
the paper is concluded by section iv .
we start with the standard gpe for a condensate tightly confined in the transverse plane , with the radial coordinate @xmath8 , and unconfined in the axial direction , @xmath9 : @xmath10 where operator @xmath11 acts in the transverse plane , @xmath12 is the frequency of the trapping potential in this plane , @xmath13 is the atomic mass , and @xmath14 is the scattering length .
transition to the quasi-1d description is possible if the change of the chemical potential due to the mean - field interaction is much smaller than the level spacing in the transverse trapping potential .
we briefly recapitulate the corresponding derivation , following , chiefly , ref . @xcite . in the quasi-1d limit
, the factorized ansatz , @xmath15 perezgarcia98 , is used to adiabatically separate fast transverse and slow longitudinal dynamics , by neglecting derivatives of @xmath16 with respect to the slow variables , @xmath9 and @xmath17 . by substituting the ansatz into eq .
( gpe ) , two decoupled equations are obtained , within the framework of the tight - transverse - confinement approximation : @xmath18 @xmath19 where the transverse chemical potential , @xmath20 , has to be found from the ground - state solution of eq .
( [ gpetransv ] ) as a function of the 1d density , @xmath21 .
physical solutions of eq .
( gpetransv ) exist only if @xmath22 @xcite , otherwise transverse collapse occurs @xcite . in the quasi-1d limit , corresponding to @xmath23 , the transverse wave function , @xmath16 , is close to the ground state of the 2d harmonic potential , and can be expanded over the set of transverse eigenmodes , @xmath24 : @xmath25 .
coefficients @xmath26 are small and can be calculated perturbatively .
accordingly , the transverse chemical potential @xmath27 can be expanded over powers of the density by means of the perturbative theory , @xmath28 , where @xmath29as shown in ref .
@xcite ( the subscript
1d " implies that the corresponding coefficient appertains to the standard 1d model ) . substituting the expansion for @xmath20 in eq .
( [ gpelong ] ) , one arrives at an effective equation describing the soliton dynamics in the quasi-1d limit : @xmath30which is nlse with the cubic - quintic ( cq ) nonlinearity .
other approaches to the derivation of the effective 1d gpe were also proposed @xcite . in particular , a more complex equation with nonpolynomial ( algebraic ) nonlinearity was derived , by means of the variational approach to the separation of the axial and transverse wave functions , in ref .
@xcite . expanding the nonlinearity up to the quintic term ,
one arrives at an equation similar to eq .
( [ gpequintic ] ) , but with a different numerical coefficient .
nlses with the cq nonlinearity are well known as model equations in nonlinear optics , starting with pioneer works @xcite .
gpes with the cq nonlinearity were also used in order to take into account three - body collisions in the bec @xcite . however , in the previously considered settings , these equations were always considered with a combination of self - focusing cubic and _ self - defocusing _ quintic terms . a drastic difference in the present case is that the quintic term is _ self - focusing _ [ as seen from eq .
( [ gg ] ) , this conclusion does not depend on the sign of scattering length @xmath31 , i.e. , on the self - focusing or defocusing character of the cubic term ; the same conclusion follows from the expansion of the above - mentioned nonpolynomial nlse derived in ref .
salasnich ] . the use of the gpe with the
double - self - focusing " cq nonlinearity , which is the case here , was tacitly assumed impossible , as in this case the equation gives rise to collapse .
nevertheless , we will show below that this equation generates meaningful stable solutions .
in fact , if the cubic nonlinearity is self - focusing , i.e. , the scattering length is negative ( the case considered throughout the present work ) , the presence of the collapse is a relevant qualitative feature of the effective gpe , as collapse takes place too in the full 3d equation , from which eq .
( [ gpequintic ] ) was derived ( even if the _ strong collapse _ in the full 3d gpe and _ weak collapse _ in the 1d cq equation bear essential differences ) . as shown in refs .
perezgarcia98,carr02,gammal02 , the collapse in the 3d equation may be avoided under the constraint of @xmath32 , where @xmath33 is the number of atoms in the condensate , and @xmath34 is the harmonic - oscillator length corresponding to the transverse confinement .
below , we use eq .
( [ gpequintic ] ) in the normalized form , @xmath35where @xmath36 and @xmath37 are dimensionless interaction constants . in fact , the absolute values of both of them may be additionally scaled to be @xmath38 , but we find it more convenient to keep these coefficients as free parameters . a family of exact soliton solutions to eq .
( [ gpedless ] ) can be found as an analytical continuation of the well - known solution of the equation with the self - defocusing quintic term @xcite .
the result is @xmath39where @xmath40 is the soliton s chemical potential that may take any value from @xmath41 .
the squared amplitude of this soliton , i.e. , the maximum atomic density at its center , is @xmath42and the norm of the soliton , which measures the total number of atoms , is @xmath43the soliton s norm and squared amplitude are shown , as functions of the chemical potential , in fig .
[ onesolitonsolution ] .
it is worth to note a drastic difference of this soliton family from its counterpart in the model with the self - defocusing quintic term , i.e. , @xmath44 : in that case , the norm takes all values , @xmath45 , while the chemical potential is limited to a finite interval , @xmath46 . on the contrary , for the present solution family , eq .
( [ normcq ] ) demonstrates that the norm is limited to a finite interval , @xmath47while @xmath48 is not limited from above ( as said above ) .
in fact , @xmath49 in eq .
( [ max ] ) is a collapse border of the soliton family .
further , in the usual cq model , with @xmath50 , the amplitude is limited by a finite value , @xmath51 , while the width of the soliton diverges @xmath52 at @xmath53 .
in contrast to this , eqs .
( [ cqsolution ] ) and ( [ amplcq ] ) show that the amplitude of the present soliton family diverges , @xmath54 , at @xmath55 , and the width of the soliton shrinks in the same limit , as @xmath56 .
this asymptotic behavior of the soliton solution clearly suggests a transition to a collapsing solution at @xmath57 , see eq .
( [ max ] ) .
equation ( [ normcq ] ) shows that condition @xmath58 holds for the entire soliton family ( see also the upper panel in fig .
onesolitonsolution ) , hence the solitons satisfy the known _ vakhitov - kolokolov _ ( vk ) _ _ _ _ stability criterion @xcite . as
this criterion is only a necessary one , but not sufficient , the stability of the solitons was tested in systematic direct simulations of eq .
( [ gpedless ] ) .
results clearly suggest that _ all _ the solitons are indeed stable against small perturbations ( of course , a large perturbation may provoke onset of the collapse ) .
it is relevant to mention that the 1d gpe with the full algebraic nonlinearity introduced in ref .
@xcite gives rise to _ two branches _ of ( implicit ) soliton solutions , one stable and one unstable ; the branches meet and disappear at the point of transition to collapsing solutions .
equation ( [ gpedless ] ) does not give rise to the second branch , as the combination of the cubic and quintic terms may be regarded as a truncated expansion of the full algebraic nonlinearity from the above - mentioned equation , and this truncation does not pick up the unstable branch .
it may also be relevant to note that , starting the derivation of the effective 1d equation from the 3d gpe with the positive scattering length ( corresponding to self - repulsive bec ) , one will arrive at eq .
( [ gpedless ] ) with @xmath59 ( and again with @xmath37 ) .
the corresponding equation , featuring competition between the cubic self - focusing and quintic self - defocusing terms , has a family of exact soliton solutions given by the same expressions ( [ cqsolution])-([normcq ] ) , in which @xmath60 is negative , as per eq .
( [ g ] ) . despite the formal similarity
, the latter soliton family is completely different from the one presented above . in particular , in the limit of @xmath61 the solution is not a usual broad small - amplitude soliton , but rather an algebraic one,@xmath62the most drastic difference of the soliton family with @xmath59 and @xmath37 from the above one is that it features @xmath63 , hence this _
entire family _ is unstable , according to the vk criterion [ algebraic solitons , such as one in eq .
( [ mu=0 ] ) , are known to be unstable for a different reason @xcite ] .
besides the fact that all the solitons in the model with the positive scattering length are unstable , their physical meaning is doubtful also because the quintic term , which appears as a perturbative correction to the cubic one @xcite , actually dominates over it in these solutions .
it is commonly known that collisions between solitons in the one - dimensional nlse , which is an integrable equation , are completely elastic .
the force of interaction between the solitons depends on the relative phase between them : with @xmath65 and @xmath66 , they are attract and repel each other , respectively @xcite .
the quintic term breaks the integrability of the equation , and is expected to make collisions inelastic . for @xmath65 , simulations reveal a critical collision velocity , below which two identical solitons merge into a single one .
the merger may be explained by the fact that radiation loss due to the inelastic collision becomes greater than the initial kinetic energy of the soliton pair @xcite .
this explanation can be implemented in an explicit form if the quintic term is treated as a small perturbation . to this end , defining @xmath67 , we rewrite eq .
( [ gpedless ] ) in the following form : @xmath68 where @xmath69 . in the zero - order approximation ( @xmath70 ) ,
the traveling - soliton solution to eq .
( gpeperturbed ) is @xmath71 where @xmath72 and @xmath73 are its amplitude and velocity , and the frequency @xmath74 is a sum of the kinetic energy and binding ( potential ) energy , @xmath75 , per particle .
the use of the perturbation theory makes it possible to obtain the following analytical result for the collision between solitons with equal amplitudes @xmath76 , velocities @xmath77 , and a phase shift @xmath78 between them @xcite : if the solitons are fast , @xmath79 , the energy loss generated by the emission of radiation during the collision is @xmath80\right\ } , \label{energyloss}\ ] ] where @xmath81,@xmath82 and @xmath83 .
note that the phase - dependent terms are exponentially small .
in the same approximation , the collision - induced loss of the number of atoms is @xmath84 to estimate a merger condition ( threshold ) , we assume that the velocities @xmath85 , which determine the collision - induced losses as per eqs .
( energyloss ) and ( [ numberloss ] ) , are actually acquired by originally quiescent ( or slowly moving ) solitons due to their mutual attraction ( if @xmath86 is close to zero ) . to this purpose
, we note that the effective potential of the interaction between far separated identical solitons is , in the case of @xmath70 , @xmath87@xcite , and the effective mass of the soliton is @xmath88 . in this approximation
, the attraction accelerates the two in - phase solitons to _ self - acquired velocities _ , @xmath89 , that can be found from the energy - balance equation , @xmath90 , hence @xmath91 .
substituting this velocity in eq .
( [ energyloss ] ) shows that the phase - dependent part is less than @xmath92 of the phase - independent one , and therefore we neglect it .
thus , the collision - induced loss of the energy and number of atoms ( for both solitons ) are predicted by the perturbation theory to be @xmath93where @xmath94 is the same numerical coefficient as in eq .
( [ energyloss ] ) .
the energy of a free soliton and its norm ( number of atoms ) , in the @xmath95 limit , are @xmath96 ( the negative term in @xmath97 is the binding energy ) .
first , the norm loss , @xmath98 , taken from eq .
( [ collisionloss ] ) , gives rise to the collision - induced change of the soliton s amplitude : @xmath99 .
the corresponding change in the binding ( potential ) energy of both solitons is positive , @xmath100 finally , the energy balance predicts a change in the total kinetic energy : @xmath101 the merger condition states that the loss of the kinetic energy is equal to or exceeds the initial kinetic energy @xcite . with regard to the expression for the total kinetic energy of both solitons which follows from eq .
( [ solitonenergy ] ) , @xmath102 , this condition means that the merger is expected if the initial velocity of each soliton falls below a critical value : @xmath103the derivation of the merger threshold implies that the critical velocity is much smaller than the above - mentioned self - acquired velocity , @xmath104 ( then , the initial velocities of the solitons may be disregarded in the above energy - balance analysis , in comparison with @xmath105 , as it was actually done ) .
expression ( [ criticalvelocity ] ) indeed satisfies condition @xmath106 , as @xmath107 is a small parameter . for simulations of soliton collisions in eq .
( [ gpedless ] ) , we chose parameter values close to those in the real experiment @xcite , where @xmath108li atoms were used : a very small scattering length , @xmath109 nm , transverse oscillation frequency @xmath110 hz , and the number of atoms @xmath111 .
however , we did not include any external longitudinal potential , in contrast to the expulsive potential that was present in the experiment .
recall that the expulsive potential made the soliton stability region very small @xcite , and actually caused the soliton to be very close to the 3d limit .
the present simulations do not include the external potential because we are interested not in effects produced by such a potential , but rather in small deviations from the one - dimensionality .
in fact , a modification of the above - mentioned experimental setup , with the aim to make the central segment of the cigar - shaped trap free of any tangible axial potential , is quite possible . to compare the analytical prediction for the critical velocity , given by eq .
( [ criticalvelocity ] ) , to numerical results , it is necessary to express perturbative parameter @xmath107 in terms of the transverse trapping frequency @xmath12 . undoing the above renormalizations ,
one arrives at a conclusion that eq .
( [ criticalvelocity ] ) implies a quadratic dependence , @xmath112 , within the framework of the perturbation theory .
this dependence is indeed observed in simulations at relatively small @xmath113 , as seen in fig .
( [ vcepsilon ] ) .
however , at larger @xmath12 , i.e. , for stronger transverse confinement , the numerical results feature a greater power in the @xmath114 dependence .
in particular , the best fit to the last four numerical points in fig .
( [ vcepsilon ] ) yields @xmath115 , which demonstrates a small but tangible deviation from the power law corresponding to the perturbation limit .
we now turn to the dependence of the critical velocity on the number of atoms , @xmath33 . the analytical prediction , eq .
( [ criticalvelocity ] ) , clearly implies @xmath116 . in fig .
( [ vcnumber ] ) , this dependence is indeed observed at smaller values of @xmath33 ( i.e. , for weaker nonlinearity ) , where the perturbation limit should naturally be valid .
perusal of numerical data shows that , in this range , the actual collision - induced radiation loss is very small , and , as a result , the merger does not lead to complete fusion of the colliding solitons into a single pulse , but rather to formation of a bound state of two solitons ( weak merger " ) , as can be seen in fig .
[ weakmergerprofiles ] .
namely , after the first collision , the solitons re - emerge as two distinct wave packets which then collide again many times .
a similar nearly radiationless inelastic collision , leading to the formation of a two - soliton loosely bound state , was recently observed in simulations of a weakly discrete cubic nlse @xcite .
a definite deviation from the @xmath117 dependence is observed in fig .
[ vcnumber ] for @xmath118 , which shows a limitation of the perturbative predictions . in this regime of strong nonlinearity ,
a smooth transition in the collision process occurs , from the formation of the above - mentioned long - lived bound state to direct ( strong " ) merger of two solitons into a single pulse , which is accompanied by a burst of radiation .
the conspicuous loss of matter with the radiation prevents the emerging single pulse from having the number of atoms above the collapse threshold , therefore the pulse _ does not _ blow up .
the transition is expressed in reduction of the life time of the loose bound state before the complete merger . in fig .
strongmergerprofiles , which represents the strongest nonlinearity included in the present framework , the bound state features only two oscillations . for even stronger nonlinearities ( which were also considered ) , the power - law dependence of @xmath0 on the number of atoms and transverse trapping frequency is observed in the form of @xmath119 and @xmath120 , respectively . however , in such an extreme regime , the relevance of the quasi-1d model is questionable . in any case , these results convey a clear caveat to soliton experiments , in which relative variations in the atom number may be as large as @xmath121 : the strong power - law dependence of the critical velocity on the number of atoms should be taken into account , to avoid occasional merger of solitons .
we proceed to inelastic collisions of identical solitons with the phase difference of @xmath123 .
numerical simulations of eq .
( [ gpedless ] ) show a salient effect of symmetry breaking in this case : while the solitons separate after the collision , they emerge as two pulses with _ different amplitudes _ ( then , the velocities are also different , to comply with the momentum conservation ) , as shown in fig . asymmetryprofiles .
it should be mentioned that a similar effect was observed in simulations of collisions between identical solitons in some other nonintegrable 1d models , chiefly in those describing transmission of nonlinear optical pulses , within the framework of the coupled - mode theory , in waveguides equipped with bragg gratings . in that context , the collision - induced symmetry breaking was reported in basic single - core models @xcite , and in more sophisticated dual - core ones ssb - dual - core .
a similar effect was also observed in collisions between moving solitons in the discrete nlse @xcite . in order to achieve qualitative understanding of the symmetry breaking ,
we resort to consideration of an _ ansatz _ based on a formal linear superposition of two completely overlapping identical solitons ( cqsolution ) , at some moment of time @xmath124 , with velocities @xmath77 and phases @xmath125 .
the ansatz _ _ _ _ yields the following expression : @xmath126an essential peculiarity of this expression is that the central points of the two last multipliers do not coincide : one is found at @xmath127 , while the other one at @xmath128 .
this simple observation suggests a concept of the mismatch between the _ amplitude center _ and _ phase center _ of the pair of colliding identical solitons .
the mismatch was considered as a cause of breaking the symmetry between colliding solitons in the above - mentioned model based on the discrete nlse @xcite . to characterize the asymmetry of ansatz ( [ cq2ssolution ] ) qualitatively
, we introduce its center - of - mass coordinate , @xmath129 } , \label{com}\]]with @xmath130 and @xmath131 , which corresponds to a maximum of @xmath132 for given @xmath133 .
for the weak quintic nonlinearity ( @xmath134 ) , one has @xmath135 } , \label{centerofmassweaknl}\ ] ] asymmetry parameter @xmath136is shown , as a function of @xmath73 , in fig .
( amplituderatiopi10 ) by the solid line for @xmath137 .
it characterizes the degree of the collision - induced symmetry breaking , and predicts a maximum at some _ nonzero _ velocity .
quite a similar dependence is indeed produced by numerical simulations of eq .
( [ gpedless ] ) for the same value of @xmath133 , as shown by dots in fig .
( amplituderatiopi10 ) .
the dots display values of the amplitude ratio of the output soliton pair , as found from the simulations . actually , the pulses emerging from the inelastic collisions are breathers with time - dependent amplitudes .
therefore , we averaged the amplitudes over long propagation distances after the collision . generally , the numerical data in fig .
( [ amplituderatiopi10 ] ) follow the predicted symmetry - breaking parameter .
however , two notable deviations are observed : a nonsmooth shape of the numerical curve ( local modulations near the maximum , which tend to give rise to two extra local maxima , definitely exceed an error of the numerical simulations ) , and a weakly decaying tail , which implies that the asymmetry generated by collisions between fast solitons is essentially larger than predicted by the above model .
note that the analytical model does not include radiation loss .
the latter may enhance the asymmetry , as the loss itself is , plausibly , asymmetric too .
we note that ref .
@xcite , in which collisions of nearly 3d solitary waves were considered through simulations of the full 3d gpe , showed very little symmetry breaking ( population transfer " ) between colliding solitons with the initial phase difference of @xmath138 , less than @xmath139 .
however , our results predict that the matter transfer ( symmetry breaking ) would be conspicuous at specific values of the collision velocities , which might not be included in the analysis reported in ref .
@xcite for very small @xmath140 , we observed chaotic behavior in the output of the collision , similar to what was reported in a weakly discrete nlse @xcite ( see also ref .
@xcite ) . very recently ,
chaotic behavior was predicted for collisions of more than two mw solitons , in the presence of a longitudinal parabolic trapping potential martin06 . in our model ,
the collision between two solitons is sufficient to observe chaotic behavior , which will be reported elsewhere .
this work aims to understand how the tight confinement in transverse directions affects the longitudinal dynamics of matter - wave solitons in the quasi-1d setting . within the framework of the known model , which reduces the multidimensional character of the full gross - pitaevskii equation to the appendage of an additional self - focusing quintic term to the effective 1d equation
, we have investigated deviations from the ideal soliton behavior
. a family of exact stationary solutions for the solitons has been constructed , and it was demonstrated that the entire family is stable , despite the possibility of collapse in the modified 1d equation ( with the negative scattering length ) . we have found inelastic effects in soliton collisions , which are impossible in ideal solitons .
two identical in - phase solitons merge into a single pulse , if the collision velocity is smaller than a critical value .
in fact , two different types of the merger were observed , strong " and weak " ones , the former leading to the formation of a loose bound state of two solitons that feature repeated collisions , with very weak radiation loss , while the latter means direct fusion into a single pulse , which is accompanied by a burst of radiation ( in that case , the radiation loss helps the emerging pulse to drop the number of atoms below the collapse threshold , and thus avoid the blowup ) .
both the analytical approximation , based on the perturbation theory , and numerical results highlight the strong dependence of the critical velocity on the strength of the transverse confinement and the number of atoms in the solitons .
symmetry breaking in collision between identical solitons with nonzero phase difference was also found , and partially explain by means of the calculation of a phenomenologically defined symmetry - breaking parameter , which measures the mismatch between amplitude and phase centers of the colliding solitons .
99 j. l. roberts , n. r. claussen , s. l. cornish , e. a. donley , e. a. cornell , and c. e. wieman , phys .
* 86 * , 4211 ( 2001 ) ; j. m. gerton , d. strekalov , i. prodan , and r. g. hulet , nature * 408 * , 692 ( 2000 ) ; e. a. donley , n. r. claussen , s. l. cornish , j. l. roberts , e. a. cornell , and c. e. wieman , _ ibid_. * 412 * , 295 ( 2001 ) .
b. b. baizakov , b. a. malomed and m. salerno , phys .
a * 70 * , 053613 ( 2004 ) , and an article in _ nonlinear waves : classical and quantum aspects _
, ed . by f. kh .
abdullaev and v. v. konotop , pp .
61 - 80 ( kluwer academic publishers : dordrecht , 2004 ; also available at http://rsphy2.anu.edu.au/asd124/baizakov_2004_61_nonlinearwaves.pdf ) . | by means of analytical and numerical methods , we study how the residual three - dimensionality affects dynamics of solitons in an attractive bose - einstein condensate loaded into a cigar - shaped trap . based on an effective 1d gross - pitaevskii equation that includes an additional quintic self - focusing term , generated by the tight transverse confinement
, we find a family of exact one - soliton solutions and demonstrate stability of the entire family , despite the possibility of collapse in the 1d equation with the quintic self - focusing nonlinearity . simulating collisions between two solitons in the same setting , we find a critical velocity , @xmath0 , below which merger of identical in - phase solitons is observed .
dependence of @xmath1 on the strength of the transverse confinement and number of atoms in the solitons is predicted by means of the perturbation theory and investigated in direct simulations .
the simulations also demonstrate symmetry breaking in collisions of identical solitons with a nonzero phase difference . this effect is qualitatively explained by means of an analytical approximation . |
the reliable prediction of the equilibrium structure and energetics of hybrid inorganic / organic systems from first principles represents a great challenge for theoretical methods due to the interplay of covalent interactions , electron transfer processes , pauli repulsion , and van der waals ( vdw ) interactions . during recent years
, huge efforts have been made to incorporate vdw interactions into density functional theory ( dft ) calculations in order to determine the structure and stability of @xmath0-conjugated organic molecules on solid surfaces @xcite .
understanding these interface properties is relevant , _ inter alia _ , for electron transfer processes in organic devices . until now and despite the obvious benefit , there are only few a studies of metal - organic interfaces combining theory and experiment . here ,
x - ray standing wave ( xsw ) measurements can provide an important test for dft calculations @xcite .
this is particularly important for systems with strong vdw contributions to the overall bonding , for which no simple substrate dependence is expected . as model system we chose diindenoperylene ( dip , @xmath3 ) , a @xmath0-conjugated organic semiconductor with excellent optoelectronic device performance , which has been studied over the last decade both in thin - films @xcite and in monolayers on noble metal surfaces @xcite .
with respect to its chemical structure , dip is a relatively simple , planar hydrocarbon without heteroatoms .
in contrast to the intensely studied perylene derivative 3,4,9,10-perylene tetracarboxylic dianhydride ( ptcda , @xmath4 ) @xcite with its four keto groups , the dip substrate interaction is not complicated by polar side groups , and the influence of intermolecular interactions is expected to be smaller than for ptcda @xcite . here , we present a systematic study with high - precision experimental data and state - of - the - art calculations of dip adsorbed on cu(111 ) , ag(111 ) , and au(111 ) .
this allows us to assess the role and relative contribution of the vdw interactions , which , contrary to simplistic pictures , we find here to be lowest for the most polarizable substrate .
dft calculations were performed using a method that extends standard pairwise vdw approaches @xcite to model adsorbates on surfaces @xcite .
this was achieved by combining the dft+vdw scheme @xcite with the lifshitz - zaremba - kohn ( lzk ) theory for vdw interaction between an atom and the surface @xcite of a solid . in our approach ( dft+vdw@xmath1 ) , the vdw energy is given by a sum of @xmath5 terms , where @xmath6 are the distances between atoms @xmath7 and @xmath8 , in analogy to standard pairwise dispersion corrected dft methods . however , by employing the lzk theory we include the many - body collective response ( _ screening _ ) of the substrate electrons in the determination of the @xmath9 coefficients and vdw radii , going effectively beyond the pairwise description .
interface polarization effects are accounted for via the inclusion of semi - local hybridization due to the dependence of the @xmath10 interatomic coefficients on the electron density in the dft+vdw method .
the dft+vdw@xmath1 method has been shown to yield remarkably accurate results for the structure and adsorption energies of xenon , benzene , and ptcda on a variety of ( 111 ) metal surfaces @xcite .
the fhi - aims code @xcite was employed for our dft calculations .
the repeated - slab method was used to model all systems with the vacuum gap set to 20 . in all calculations , convergence criteria of 10@xmath11 electrons for
the electron density and 10@xmath12 ev for the total energy of the system were used .
a convergence criterion of 0.01 ev / for the maximum final force was used for all structure relaxations . the scaled zeroth - order regular approximation ( zora )
was applied for inclusion of scalar relativistic effects @xcite . the dft+vdw@xmath13 method employed the perdew - burke - ernzerhof ( pbe ) functional @xcite .
the sampling of the brillouin zone was done using a @xmath14 @xmath15-point grid .
we used a @xmath16 unit cell composed of a metal surface of three layers and one single dip molecule . in the absence of experimental data for the in - plane registry we placed the central ring of the molecule aligned with a topmost metal layer atom and the major axis of the molecule aligned along the cell diagonal .
this structure was adopted for the cu(111 ) , ag(111 ) , and au(111 ) surfaces . in each simulation
we obtained the adsorption energy curve using a rigid dip molecule and tuning the surface
molecule distance @xmath17 .
the adsorption energy per molecule @xmath18 was calculated from @xmath19 , where @xmath20 is the energy per unit cell of the isolated metal surface , @xmath21 is the energy per unit cell of the isolated dip molecule , and @xmath22 is the energy per unit cell of the combined system .
we also obtained the relaxed geometries for all three systems starting from the static equilibrium geometry . during geometry relaxation
, we allowed only the topmost metal layer and the molecule to relax while the other two metal layers were fixed . from the final relaxed configurations we obtained the bonding distance @xmath17 by taking the average position of all dip carbon atoms with respect to the unrelaxed topmost surface layer .
this definition is consistent with the analysis of the xsw data .
to measure the bonding distance of dip we used the xsw technique @xcite , which yields precise and element - specific structural data .
the experiments were performed at beamline id32 of the esrf @xcite .
dip films were prepared and studied _ in situ _ under ultrahigh - vacuum conditions . a separate preparation chamber contained a knudsen cell , a quartz crystal microbalance , installations for ar@xmath23 sputtering , and a temperature - controlled sample stage .
the main chamber , in which the xsw measurements were performed , was equipped with a sample manipulator and a hemispherical specs phoibos 225 hv photoelectron analyzer .
the xsw experiments were carried out at room temperature in back - reflection geometry using the ( 111 ) bragg reflection of the crystals for at least two films per substrate to check for reproducibility of the results .
the detection angle of the analyzer was @xmath2490@xmath25 relative to the surface normal with an acceptance angle of @xmath26 .
we note that in this configuration non - dipolar contributions to the photoelecton yield can be effectively avoided @xcite .
the cu(111 ) , ag(111 ) , and au(111 ) single crystals were mounted on different sample holders for individual treatment .
the surfaces were prepared by repeated cycles of ar@xmath23 bombardment and annealing at 700k .
surface cleanliness was confirmed with x - ray photoelectron spectroscopy ( xps ) as well as low - energy electron diffraction ( leed ) .
sublimation grade dip was evaporated from a home - built knudsen cell .
the intensity ratio of the c 1__s _ _ signal relative to a substrate core - level , normalized with the corresponding photoemission cross sections , was used to determine the number of dip molecules on the surface . with the unit cell size of dip on cu(111 ) @xcite , ag(111 ) @xcite , and au(111 ) @xcite , the coverages were calculated to be between @xmath27 and @xmath28 ml . with the xsw setup at id32 . from each signal
a shirley background was subtracted and then fitted with a voigt function for the main peak and a gaussian function for possible shake - up peaks . ]
the c 1__s _ _ core - level signals of dip on cu(111 ) , ag(111 ) , and au(111 ) , which were used for the xsw measurements , are shown in fig .
[ fig : xps_fit ] .
the main peaks are expected to consist of two principal components ( c c vs c h bound atoms ) which , however , could not be resolved with the energy resolution of the xsw setup . and coherent fraction @xmath29 .
bragg energies are @xmath30 kev [ cu(111 ) ] and @xmath31 kev [ ag(111 ) and au(111 ) ] .
( b ) all xsw results for dip on cu(111 ) , ag(111 ) , and au(111 ) displayed in an argand diagram . here
, each datapoint represents one single xsw measurement yielding @xmath29 ( length of a vector ) and @xmath32 ( angle of a vector ) . the three vectors point to the average values of @xmath29 and @xmath32 for dip on cu(111 ) , ag(111 ) , and au(111 ) .
film 1 of dip on ag(111 ) was measured with a different crystal compared to films 2 and 3 . ]
in addition to each main peak , a second weak feature at @xmath241ev higher binding energy possibly related to a shake - up process can be observed .
obviously , the binding energy of the c 1__s _ _ main line of dip follows @xmath33 , being 284.5 ev on ag(111 ) , 284.2 ev on cu(111 ) , and 283.7 ev on au(111 ) .
furthermore , the c 1__s _ _ peak of dip on ag(111 ) exhibits a stronger asymmetry than on cu(111 ) and au(111 ) .
a detailed discussion of the spectroscopic features is beyond the scope of this paper in which we focus on the xsw results .
representative results of the xsw experiments are shown in fig .
[ fig : dip_yield](a ) . in each panel
the measured reflectivity of the substrate and the corresponding c 1__s _ _ photoelectron yield is displayed .
least - squares fits of the data give the coherent position @xmath32 and hence the average bonding distance @xmath34 @xcite , where @xmath35 is the substrate lattice plane spacing .
based on results of all xsw experiments we calculate the average bonding distance @xmath36 and the standard deviation ; see fig . [
fig : dip_yield](b ) . for cu(111 ) we thus find @xmath37 , and for ag(111 ) @xmath38 . due to the reconstruction of the au(111 ) surface , which results in a 3% larger spacing between the first and second au layers @xcite , the bonding distance decreases from the measured apparent value @xmath39
to @xmath40 .
all experimental results are summarized in table [ tab : xsw_results ] .
note that although the coverage of the two ( three ) dip films prepared on each substrate was not identical , we did not observe a significantly coverage - dependent bonding distance @xmath36 .
.results of xsw experiments : coherent fraction @xmath29 , coherent position @xmath32 , and bonding distance @xmath36 of dip on the three noble metals .
the parameters refer to an average of several xsw measurements with the corresponding standard deviation as error bars .
[ cols="<,^,^,^",options="header " , ] [ tab : dft_results ] having established precise experimental data , we now turn to the results of our dft calculations .
the average bonding distances of dip obtained from fully relaxed structures are @xmath41 on cu(111 ) , @xmath42 on ag(111 ) , and @xmath43 on au(111 ) , see table [ tab : dft_results ] and fig .
[ fig : adsorption ] .
we hence find that the pbe+vdw@xmath44 method applied to dip on cu(111 ) , ag(111 ) , and au(111 ) yields an agreement better than 0.1 between theoretical calculations and experiments . in accordance with the bonding distances ,
the calculated adsorption energies listed in table [ tab : dft_results ] follow the trend @xmath45 for small distances ] compared to ag(111 ) and au(111 ) , which is due to significant interaction between dip and cu(111 ) .
one may speculate that the interaction mechanism includes hybridization between dip and cu states .
in addition to the adsorption energies and average bonding distances , table [ tab : dft_results ] holds the minimal and maximal bonding distances @xmath46 of individual carbon atoms in dip . these values indicate that the molecule adsorbs in a slightly tilted or distorted geometry . for cu(111 ) , where the effect is most pronounced , the calculated bonding distances @xmath47 and @xmath48 differ by @xmath240.4 , which is equivalent to a molecular tilt angle of 1.5@xmath25 .
the corresponding spread of vertical positions of the carbon atoms leads to a reduced @xmath29 in the xsw scans .
model simulations similar to those presented in ref .
show that the dft - derived adsorption geometry on cu(111 ) results in a relatively small decrease of the coherent fraction ( @xmath49 ) , which lies within the standard deviation of our xsw measurements . to obtain a better understanding of the influence of lateral intermolecular interactions on the dip adsorption geometry
, we also computed the relaxed dip geometry for different cu(111 ) , ag(111 ) , and au(111 ) unit cells . for dip on cu(111 ) , we increased the unit cell from @xmath16 to @xmath50 in order to reduce the molecule molecule interactions .
we studied various configurations , finding a flat relaxed geometry for each case considered .
the bonding distance is slightly larger ( 2.64 ) than for the calculation with the smaller unit cell . for dip on ag(111 ) , we also considered a unit cell which was determined from a closed packed monolayer on ag(111 ) @xcite .
the relaxed geometry of the molecules in the monolayer is flat and the bonding distance @xmath51 in almost perfect agreement with the experimental one , i.e. , even better than the result for ag(111 ) shown in table [ tab : dft_results ] . for a @xmath52 unit cell of au(111 ) , the relaxed dip geometry yields an equilibrium distance of 3.15 , also in slightly better agreement with experiment than the result shown in table [ tab : dft_results ] .
overall , these calculations agree with the experimental observation that the vertical dip position depends only weakly on surface coverage .
for the unrelaxed dip molecule as a function of its averaged distance @xmath17 from the cu(111 ) , ag(111 ) , and au(111 ) surfaces .
the curves are shown for the pbe functional with and without the inclusion of long - range vdw interactions using the vdw@xmath1 method .
the reported contribution of the vdw energy is shown at the equilibrium distance corresponding to the fully relaxed dip surface geometry ( see text ) . ] with the experimental and theoretical values at hand , and in view of their excellent agreement , we are in a good position to discuss the vdw interactions and the bonding distances in more detail . as described above , the ( atom - atom ) vdw energy is computed as @xmath53 , where the @xmath10 coefficient determines the strength of the interaction between atoms @xmath7 and @xmath8 , while @xmath6 is the distance between adsorbate and substrate atoms ( fig . [
fig : e_vdw ] ) .
integration of the vdw energy for a single atom adsorbed on a semi - infinite surface yields the atom
surface vdw energy as @xcite @xmath54 , where now @xmath55 determines the interaction strength between atom and surface , @xmath56 corresponds to the distance of the atom to the uppermost surface layer , and @xmath57 indicates the position of the surface image plane . in a rather naive picture ,
the @xmath55 coefficient can be determined simply from the @xmath58 and the @xmath59 coefficients that correspond to the adsorbed atom and the metal atom , respectively . however , the situation for real surfaces is more complex because both localized and bulk metal electrons contribute to the @xmath55 coefficient in a non - trivial way , meaning that this coefficient depends on the dielectric function of the underlying solid .
we computed the @xmath55 coefficients corresponding to the interaction between a carbon atom and the cu(111 ) , ag(111 ) , and au(111 ) surfaces . when describing the metal surface as a simple collection of non - interacting atoms we obtain @xmath60 , @xmath61 , and @xmath62 hartree@xmath63bohr@xmath64 .
in contrast , when using the more appropriate lifshitz - zaremba - kohn expression @xcite for @xmath55 , we obtain @xmath65 , @xmath66 , and @xmath67 hartree@xmath63bohr@xmath64 .
this clearly illustrates that the vdw interaction between an atom and a solid surface is significantly modified by the collective electronic response within the substrate surface @xcite . between adsorbate and substrate atoms .
( b ) vdw energy of dip adsorbed on cu(111 ) , ag(111 ) , and au(111 ) as a function of the distance between adsorbate and substrate atoms . ] the very similar lzk @xmath68 coefficients for cu , ag , and au lead to essentially the same adsorption energy at large distances for dip on cu(111 ) , ag(111 ) , and au(111 ) ( fig .
[ fig : adsorption ] ) .
however , at shorter molecule
surface distances , which include the equilibrium distance , the adsorption energy is determined by an interplay between the vdw attraction and the pauli repulsion with a possible covalent component .
the pauli repulsion follows roughly the trend of decreasing vdw radii , with a faster onset in terms of the molecule
surface distance for au ( with the largest vdw radius ) , and then decreasing for ag and cu .
therefore , for au the balance between the vdw attraction and the pauli repulsion is obtained _ further away _ from the substrate ( i.e. , at larger adsorption distances ) than for cu , which in turn makes the adsorption energies _ lower _ for au than for cu , in contrast to the possible naive expectation of au with its higher polarizability and @xmath9 coefficient exhibiting a stronger vdw interaction than cu .
the difference in the vdw energy distribution for dip on cu(111 ) , ag(111 ) , and au(111 ) is visualized in fig .
[ fig : e_vdw](b ) , where the vdw energy between dip and substrate atoms is plotted as a function of their distance @xmath6 .
in contrast to ag(111 ) and au(111 ) , the small bonding distance of dip on cu(111 ) results in a second peak in the histogram at @xmath243.6 , which originates from the higher atomic density of the cu substrate .
in conclusion , the bonding distances calculated with the pbe+vdw@xmath1 method are in excellent agreement with the xsw data for dip on cu(111 ) , ag(111 ) , and au(111 ) ( 2.51 , 3.01 , and 3.10 , respectively ) .
our combined study demonstrates that the vdw energy is larger for dip on cu(111 ) than for dip on ag(111 ) and au(111 ) .
future investigations on the electronic properties of these systems , which can draw on the findings presented here , will contribute to an even better understanding of the adsorption process .
this work was financially supported by the dfg ( schr700/14 - 1 and sfb951 ) and mext . n.f . and a.t .
are grateful for support from the fp7 marie curie actions of the eu , via the initial training network small ( grant no .
mcitn-238804 ) .
the authors gratefully acknowledge the esrf for providing access to beamline id32 .
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 * * , ( ) @noop ph.d .
thesis , @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop * * , ( ) @noop _ _ ( , ) @noop ph.d .
thesis , @noop * * , ( ) | we present a benchmark study for the adsorption of a large @xmath0-conjugated organic molecule on different noble metal surfaces , which is based on x - ray standing wave ( xsw ) measurements and density functional theory calculations with van der waals ( vdw ) interactions .
the bonding distances of diindenoperylene on cu(111 ) , ag(111 ) , and au(111 ) surfaces ( 2.51 , 3.01 , and 3.10 , respectively ) determined with the normal incidence xsw technique are compared with calculations . excellent agreement with the experimental data , i.e. , deviations less than 0.1 , is achieved using the perdew - burke - ernzerhof ( pbe ) functional with vdw interactions that include the collective response of substrate electrons ( the pbe+vdw@xmath1 method ) .
it is noteworthy that the calculations show that the vdw contribution to the adsorption energy increases in the order au(111 ) @xmath2 ag(111 ) @xmath2 cu(111 ) . |
the ability of one star to gravitationally amplify the light from another was considered previously by einstein ( 1936 ) , and developed more generally by , for example , refsdal ( 1964 ) .
realistic rate estimates for gravitational microlensing by massive parallel photometry were made by paczynski ( 1986 ) for the lmc and again by paczynski ( 1991 ) for the galactic bulge .
the first measurements of microlensing were reported by udalski et al .
( 1993 ) and alcock et al .
several microlensing searches are currently underway , including the ogle collaboration ( udalski et al .
1993 ) , the macho collaboration ( alcock et al . 1993 ) , the eros collaboration ( aubourg et al . 1993 ) , the agape collaboration ( alard et al . 1995 ) , and a collaboration monitoring m31 ( crotts et al . 1992 ) .
groups that track microlensing events , once announced , include the planet collaboration ( albrow et al .
1995 ) , and the gman collaboration ( becker et al .
1997 ) .
combined , microlensing search collaborations currently monitor tens of millions of stars down to a visual magnitude of about 19.5 over a small fraction of the sky , currently reporting over 100 microlensing events ( for a good review , see paczynski 1996 ) . of the first 45 gravitational lens candidate events toward the galactic bulge , the brightest star recorded to undergo a microlensing event by the macho collaboration was of unlensed visual magnitude 16.4 ( alcock et al .
1997 ) .
a primary motivation for this study was curiosity about how bright a star was undergoing a detectable microlensing event at any given time .
previously , colley & gott ( 1995 ) estimated that a microlensing events becomes bright enough for naked eye observation nearly every 5000 years . perhaps , however , a star that is significantly brighter than those being monitored is _ currently _ undergoing detectable microlensing .
microlensing of this star might be relatively easy to track , as it would be relatively bright and in a less crowded field .
this paper is an attempt to estimate the magnitudes of the brightest microlensing events on the entire sky , and to discuss the possible advantages of finding these events .
2 will detail how this estimate will be made . in 3 ,
results will be presented , while some discussion and conclusions will be given in 4 .
nemiroff ( 1986 ) gave a first estimate of the all - sky microlensing rate .
there it was shown that to amplify source light at least @xmath1 , a lens with schwarzschild radius @xmath2 must fall into a detection volume " @xmath3 , where distance @xmath4 has subscripts @xmath5 , @xmath6 , and @xmath7 which stand for observer , lens , and source , respectively .
@xmath8 , and here carries the subscript abs " which refers to absolute amount of source magnification .
assuming all stars are identical and uniformly distributed , the likely number of lens events @xmath9 above absolute amplitude @xmath1 out to distance @xmath10 is found by summing the space internal to all the lens detection volumes , so that @xmath11 , where @xmath12 is the number density of stars .
this early estimate did not include the brightness of the lensing star itself .
estimates of the effects of the brightness of the lens or an unlensed component have been done previously with regard to existing microlensing surveys by kamionkowski ( 1995 ) , nemiroff ( 1997 ) and han ( 1998 ) .
nemiroff ( 1997 ) has shown that for a field of lenses with identical brightness , the apparent optical depth is a factor of about six less than the absolute optical depth , which includes the assumption that the lens is dark . including this factor in the previous estimate , the number of lenses expected above the _ apparent _ amplification factor of @xmath13 would be @xmath14 given @xmath15 0.1 stars pc@xmath16 and @xmath17 km , @xmath18 , the single closest lensing event is expected at @xmath19 pc .
the total number of stars in this volume is @xmath20 = 7.0 x 10@xmath0 .
given the canonical star is like the sun and has absolute visual magnitude @xmath21 , its apparent magnitude at 1200 pc is an estimate of the brightest microlensing event occurring on the sky at any one time : @xmath22 . although instructive in calibrating intuition ,
this estimate did not include the anisotropic and inhomogeneous stellar distributions that surround our earth and our galaxy .
a more accurate estimate might be significantly different . toward this end ,
an all - sky microlensing rate estimate is made here by employing the galactic model published originally by bahcall & soneira ( 1980 ) and subsequently updated by bahcall ( 1986 ) .
the bahcall - soneira galactic model obtains a reasonable fit to the observed star counts with a relatively simple disk and spheroid model .
bahcall ( 1986 ) demonstrates accuracy of the model in several directions down past 20th magnitude , although all of these directions are at least 20 degrees away from the galactic plane .
the model also included large - scale affects of reddening and obscuration by galactic dust .
a fortran - program version of the bahcall - soneira model was downloaded from j. bahcall s web site in december 1997 and adapted to address microlensing . the bahcall - soneira code tracks both main - sequence and giant stars for both a galactic disk and spheroid . for main - sequence disk stars ,
the code was run with absolute visual magnitude @xmath23 cut - offs of @xmath24 at the bright end at @xmath25 at the dim end . for disk giants ,
@xmath23 was taken to be @xmath26 at the bright end .
for spheroid stars , @xmath23 was taken to be @xmath27 at the bright end . in general ,
the total number of lensing events visible is given by @xmath28 where @xmath29 is a solid angle on the sky , @xmath30 is the maximum source distance , @xmath31 is the cutoff for absolute visual magnitude at the dim end , @xmath32 is the cutoff for @xmath23 at the bright end , @xmath33 is the absolute visual magnitude of a source star , and @xmath34 is the absolute visual magnitude for the lens .
the impact parameter @xmath35 is the furthest distance from the observer - source axis a lens could be placed , at distance @xmath36 , and still amplify source light by the amount @xmath13 needed for detection . from nemiroff ( 1986 , 1989 , 1997 ) this impact parameter was found to be @xmath37 the brightness of the stellar lens itself works to dampen the magnitude of observed microlensing events .
more precisely , to achieve an apparent lensing amplification of @xmath13 , the lens must create an absolute amplification @xmath38 to be seen over its own brightness ( nemiroff 1997 ) : @xmath39 , where @xmath40 indicates apparent brightness . microlensing rates from several candidate scenarios were made . in the first
, it was assumed that all stars brighter than a given unlensed magnitude limit would be monitored for microlensing , a case referred to as without magnification bias . " in the second scenario , it was assumed that a star more faint than a given magnitude limit could be amplified above the limit and be detected .
the second scenario therefore includes magnification bias ( nemiroff 1994 ) . in the third scenario , only stars in the galactic disk of the bahcall - soneira model were included , without magnification bias , to give a indication of the angular distribution of the microlensing events . in all of the computations ,
giant stars were not allowed to be lenses , since one or both amplified images might impact the lens surface and be absorbed ( ftaclas 1998 ) . the mass of main sequence stellar lenses was taken from mihalas & binney ( 1981 ) to be @xmath41 .
equation 2 above was used to compute @xmath42 , the number of expected lens events anywhere on the sky . for any magnitude limit @xmath43 ,
however , the maximum expected amplification @xmath44 on the sky can be computed directly from @xmath42 since @xmath42 scales as @xmath45 .
in fact , when @xmath18 , @xmath46 , the optical depth ( vietri & ostriker 1983 ; nemiroff 1989 ) .
when @xmath9 events are expected at the level of @xmath18 ( where @xmath47 1.34 ) , then fewer events are expected at a higher @xmath13 ( nemiroff 1997 ) . specifically , the level of @xmath45 where one expects a single event is at @xmath48 .
the corresponding amplitude is @xmath49
figure 1 shows a plot of the expected optical depth to stellar microlensing as a function of apparent visual magnitude @xmath50 .
when the inverse of the optical depth becomes on the order of the number of stars being observed , then there is a 63 % chance that at least one of the stars is amplified by a factor greater than 1.34 , and detection of gravitational lensing becomes likely . including stars in both the galactic disk and the central spheroid , stellar microlensing becomes likely between 16th and 17th magnitude .
the bahcall - soneira model used predicts almost exactly 100 million stars exist over the entire sky that are brighter than 17th magnitude .
perhaps unexpected is that figure 1 shows that optical depth can actually be a decreasing function of the survey apparent magnitude limit .
this is caused by the shape of the stellar luminosity function , which may cause a disproportionate number of stars in a given magnitude range to be relatively close to the observer , hence contributing relatively little to the optical depth .
this decrease in optical depth means only that the probability of detectable lensing _ per star _ has decreased . since star counts
increase greatly as magnitude dims , the total probability of finding microlensing somewhere on the sky at fainter magnitudes still sharply increases .
figure 2 shows plots of the expected maximum amplitude of the brightest lens event on the sky as a function of the faintest apparent visual magnitude monitored .
inspection of this plot again indicates at about 17th magnitude , an @xmath51 event is expected , even without magnification bias .
a search sensitive to magnification bias could discover that a star temporarily brighter than 16th magnitude is actually being detectably microlensed , somewhere on the sky .
the more stable the relative photometry , the smaller the microlensing enhancements it would be sensitive to , the brighter the star that would show detectable microlensing .
alternatively , were all stars brighter than 20th magnitude monitored , at least one would be expected to be undergoing an @xmath52 event at any given time .
figure 3 shows the number of @xmath53 lensing events expected over the sky as a function of visual magnitude .
this plot once again indicates a 63 % chance that one star with unlensed magnitude brighter than 17 is being amplified by @xmath53 , and that it is likely a star appearing to be brighter than 16th magnitude is undergoing an @xmath53 microlensing event .
additionally , figure 3 indicates that on average , two stars with unlensed magnitude brighter than 18 are undergoing detectable microlensing , while potentially 9 stars brighter than 20th magnitude are expected to undergo simultaneously detectable gravitational lens magnifications with @xmath53 .
figure 4 shows a contour plot of the number of stars per square degree brighter than 17th magnitude expected to show detectable microlensing .
figure 4 is plotted in galactic coordinates and is centered on the galactic center .
the plot shows that the most probable place to find gravitational lensing in our galaxy is first toward the galactic center , and next in our galactic plane . in the direction of the galactic center
, the success rate of a lensing search monitoring all stars brighter than 17th magnitude would be on order 3 x 10@xmath54 lenses degree@xmath55 .
these results estimate the _
static _ probability of gravitational lensing
assuming stars do nt move .
but stars do move and lensing events have finite durations .
the methodology for making _ dynamic _ lensing probability estimate , allowing for relative lens , observer , and source motion , was discussed previously by nemiroff ( 1991 ) .
taking the detection volume defined above and allowing it to move will allow it to catch " increasingly more lenses as a function of time .
were the source to move with transverse speed @xmath56 relative to the observer for a time @xmath57 , it would sweep out a volume @xmath58 were @xmath59 is the semi - major axis , and @xmath60 is the semi - minor axis of the ellipsoidal detection volume . given canonical values of @xmath17 km ,
@xmath61 kpc , @xmath18 , @xmath62 km sec@xmath63 , and @xmath64 3.1 x 10@xmath65 sec , the ratio of dynamic to static volumes is about 5 .
this means that the probability of lensing is about 5 times as great over the course of a year than during a single observation . in general
, the duration of a microlensing events would be on order @xmath66 . given the canonical values listed above , a typical duration for a microlensing event would be @xmath67 days . the brightness of the lens and the maximum amplification of the event is automatically included in @xmath45 .
in general , higher amplification events , closer sources , and brighter lenses would cause shorter duration events . from figure 3 ,
the magnitude at which @xmath68 is about 15th magnitude without magnification bias , and about 14th magnitude with magnification bias .
therefore , were all stars brighter than 15th magnitude monitored for a year with observations spaced shorter then the event duration , at least one is likely to show an @xmath53 event .
directly applying stellar densities derived from the bahcall - soneira model to microlensing leads to several inaccuracies .
for example , this model is not designed to directly indicate the space density of binary stars , dark matter lenses , or the galactic bar , each of which could alter the microlensing rate . no corrections for these were made in the present calculation , although it is believed that these factors would only increase the actual observed lensing rate .
one conclusion that can be drawn is that microlensing effects are present in _ any _ set of 100 million source stars that can possibly be monitored .
in fact , the brightest 100 million stars are among the _ least _ likely set to show microlensing , since their proximity allows relatively little room to place the lenses . additionally , these stars are so close that wide - angle measurements are needed just to monitor them . searching for all sky microlensing , however ,
should yield the closest lenses and the brightest sources .
the relative brightness of these microlensing events might make them more than a curiosity . at 15th magnitude
, most observatories would be capable of follow - up coverage of an announced , on - going microlensing event .
this could lead to microlensing events with extraordinarily good time coverage .
good time coverage , in turn , creates an increased chance of recovering brief bumps in microlensing light curves caused by planets orbiting the stellar lens ( see , for example , mao & paczynski 1991 ; gould & loeb 1992 ) .
the relative proximity of these lenses and/or sources might be expected to create an increased chance that parallax could be measured for them , and hence foster the ability to better understand lens and source distances , masses , and even transverse velocities . for lenses crossing the disk of the source
, the lens will act like a huge magnifying glass being swept across the source star face .
potentially recoverable information includes the radius , shape , and brightness distribution of the source star ( nemiroff & wickramasinghe 1994 ; gould 1994 ; witt & mao 1994 ; gaudi & gould 1998 ) . the relatively high angular separation of these microlensing events might be expected to allow optical depth estimates toward widely separated directions in our galaxy .
therefore , event rates would give information on the galactic stellar and dark matter distributions . the optical depth to lensing has yet to be accurately predicted in any direction
the result is always a surprise .
possibly optical depth estimates along new directions would also yield new surprises . some might consider the above results discouraging , indicating that the number of stars is too high , the event rate too low , the data rate too high , and the monitoring instruments needed are too unusual to search for microlensing over the entire sky .
others might consider that the technology need to carry out a large area search capable of detecting microlensing is not only possible , but surveys that could be considered prototypical are already being implemented .
examples of such monitoring systems include lotis ( park et al .
1997 ) and rotse ( marshall et al .
1997 ) , originally created to search for optical counterparts to gamma - ray bursts .
lotis is already generating daily images of approximately @xmath69 steradians down to 13th magnitude while searching for the optical counterparts of gamma - ray bursts . a planned improvement creating super - lotis " will scan the sky every three weeks down to a limiting visual magnitude of 19 , at which time bright star microlensing events
would be expected to exist in its database .
lastly , it should be pointed out that the secondary " information on the intrinsic variability of bright star systems gathered in a wide - angle microlensing search should be of significant value .
interesting causes of variability includes stellar pulsations , binary star eclipses , and recurrent novae . arguably , since transient events of even unknown cause could be followed by small telescopes common both to professionals and amateurs , triggers " created by a wide - angle monitor could be of interest regardless of whether they were related to microlensing .
i thank c. ftaclas and j. b. rafert for helpful comments , g. marani for a careful reading of the manuscript , and b. paczynski for pointing out parallax possibilities .
this research was supported by grants from nasa and the nsf .
alard , c. , mao , s. , guibert , j. 1995 , astron .
, 300 , l17 albrow , m. et al .
1995 , , 187 , # 117.04 alcock , c. et al .
1993 , nature , 365 , 621 alcock , c. et al .
1997 , , 479 , 119 aubourg , e. et al .
1993 , nature , 365 , 623 bahcall , j. 1986 , , 24 , 577 bahcall , j. n. & soneira , 1980 , , 44 , 73 becker , a. et al .
1997 , , 191 , # 83.05 colley , w. n. & gott , j. r. iii , 1995 , , 452 , 82 crotts , a. 1992 , , 399 , l43 einstein , a. 1936 , science , 84 , 506 ftaclas , c. 1998 , personal communication gaudi , b. s. & gould , a. 1998 , , submitted : astro - ph/9802205 gould , a. 1994 , , 421 , l71 gould , a. & loeb , a. 1992 , , 396 , 104 han , c. 1998 , , in press kamionkowski , m. 1995 , , 442 , l9 marshall , s. et al .
1997 , , 191 , 48.15 mihalas , d. & binney , j. 1981 , galactic astronomy , 2nd edition , ( freeman : san francisco ) , p. 113
mao , s. & paczynski , b. 1991 , , 374 , 37 nemiroff , r. j. 1986 , , 123 , 381 nemiroff , r. j. 1989 , , 341,579 nemiroff , r. j. 1991 , astron . ap . , 247 , 73 nemiroff , r. j. 1994 , , 435 , 682 nemiroff , r. j. 1997 , , 486 , 693 nemiroff , r. j. & wickramasinghe , w.a.d.t .
1994 , , 424 , l21 paczynski , b. 1986 , , 304 , 1 paczynski , b. 1991 , , 371 , l63 paczynski , b. 1996 , , 34 , 419 park , h. s. et al , 1997 , 490 , l21 refsdal , s. 1964 , , 128 , 295 udalski a. et al , 1993 , acta astron . , 43 , 69 vietri , m. & ostriker , j. p. 1983
, , 267 , 488 witt , h. , & mao , s. 1994 , , 430 , 505 | it is estimated that a star brighter than visual magnitude 17 is undergoing a detectable gravitational microlensing event , somewhere on the sky , at any given time .
it is assumed that both lenses and sources are normal stars drawn from a standard bahcall - soneira model of our galaxy .
furthermore , over the time scale of a year , a star 15th magnitude or brighter should undergo a detectable gravitational lens amplification .
detecting and studying the microlensing event rate among the brightest 10@xmath0 stars could yield a better understanding of galactic stellar and dark matter distributions .
diligent tracking of bright microlensing events with even small telescopes might detect planets orbiting these stellar lenses . |
was discovered as part of an exosat high galactic latitude survey ( giommi et al .
the source was associated with the @xmath3 = 15.9 mag galaxy mcg-01 - 12 - 005 , a radio source ( with 6 cm flux density 44 mjy ) with no optical emission lines .
it was classified as a candidate bl lacertae object on the basis of its x - ray luminosity ( above @xmath4 erg s@xmath5 ) and its position in the @xmath6 diagram , and is included as a bl lac object in the vron - cetty & vron ( 2003 ) catalogue .
observations with cangaroo give an upper limit for the @xmath7-ray flux above 2 tev of less than @xmath8 photons @xmath9 s@xmath5 ( roberts et al .
1998 ) but the results are not conclusive on the nature of the source .
kirhakos & steiner ( 1990 ) determined a redshift of 0.0392 for the source they designate iras 04235 - 0840b and identify with the heao source 1h 0422 - 086 , which they classify as a type 2 seyfert , and associate with .
mcg-01 - 12 - 005 is an early type galaxy close to the centre of galaxy cluster clg0422 - 09 , at @xmath10 .
white , jones & forman ( 1997 ) described it as a cooling - flow cluster and give a temperature of 2.9 kev ( from edge & stewart 1991 ) .
asca data analysed by ikebe et al .
( 2002 ) implied a temperature of k@xmath11 kev from the @xmath12 relation , and a 0.1 - 2.4 kev luminosity of ( 5.3@xmath13 erg s@xmath5 .
the galactic absorption for is 6.22@xmath14 @xmath9 ( dickey & lockman 1990 ) , and the @xmath15 reported in x - ray analysis of the galaxy cluster is 6.4@xmath14 @xmath9 ( e.g ikebe et al .
2002 ) .
in this paper we present observations of which give us new insights into the nature of this source . throughout this paper ,
we adopt a redshift for of @xmath10 and a cosmology with @xmath16 = 70 km s@xmath5 mpc@xmath5 , @xmath17 and @xmath18 = 0.7 . the angular diameter distance to this source
is 159 mpc , and its luminosity distance is 172 mpc . at the redshift of the source , 1 arcsec corresponds to 773 pc .
[ fig : imaflux ] [ fig : dss ] was observed with in december 2002 ( observation i d 03970 ) as part of a small sample of bl lac objects with relatively weak cores , which included pks 0521@xmath2356 ( birkinshaw , worrall , hardcastle 2002 ) , with the primary objective of studying the interaction between the jet and the group or cluster x - ray medium .
since the bl lac core of was expected to produce a high x - ray count rate , the observation was performed in faint mode and with a small window with the back - illuminated chip s3 on the advanced ccd imaging spectrometer ( acis ) .
this window was 128 pixels ( of [email protected] arcsec@xmath20 each ) wide , and 1024 pixels long .
we processed the level 1 event data using v3.0.2 and calibration database ( caldb ) 2.26 .
we removed bad pixels using the supplied bad pixel file , and retained events of grade 0,2,3,4 , and 6 .
we verified that there are no periods of statistically higher background by generating light curves in different energy bands , and the useful exposure time is 14280 s. the online aspect calculator shows that no aspect correction was necessary for these data , therefore the x - ray , radio and optical frames should align to the aspect uncertainty @xmath21 arcsec .
the analysis of the diffuse emission started with our use of wavdetect to identify possible point sources in the field .
the algorithm was used with a source detection threshold set to 10@xmath22 ( the minimum value for the algorithm to give an accurate enough detection ) , and only found a single point source at ra = @xmath23 , dec = @xmath24 ( j2000 ) .
the characterisation of the diffuse emission is affected by the half - chip observation , which induces border effects to the north - east and south - west .
we corrected these effects via the exposure map , generated using the aspect histogram and the instrument map .
when appropriate , the counts image was converted into a flux image by dividing it by the exposure map . for spectral analysis
, we accounted for the position - dependent redistribution matrix file ( rmf ) and the auxiliary response file ( arf ) for the source , which were computed from calibration database 2.26 .
this release takes into account a correction for the degradation of the low - energy quantum efficiency .
we extracted spectra in successive circular annuli , excluding the point source and small scale structures ( see sec . [ sec : centre ] ) .
the background for all annuli was chosen to be the same , and was extracted in a circle of radius 28.5 arcsec ( see fig .
[ fig : imaflux ] ) centred at ra = @xmath25 , dec = @xmath26 ( j2000 ) .
for the spectral analysis we excluded the regions close to the chip boundary . if not otherwise specified , spectra were fitted with absorbed plasma models in xspec v11.3 .
we tried fits using the apec ( smith et al .
2001 ) and ( mewe , gronenschild , van den oord 1985 ; liedahl , osterheld , goldstein 1995 ) codes ; the latter gave better fits .
we scaled abundances to the standard solar values from grevesse & sauval ( 1998 ) .
a - array vla data for exo 0423.4@xmath20840 at 1.4 , 4.9 , and 8.6 ghz were extracted from the national radio astronomy observatory ( nrao ) archive , and processed into maps through the normal processes of calibration and cleaning . at 1.4 and 4.9
ghz it was possible to self - calibrate the data to improve the final maps , but the high noise and short observation at 8.6 ghz made this impossible .
all three maps easily detect a bright extended radio source near the centre of the x - ray emission . at 1.4
ghz the radio map has a synthesized beam of @xmath27 arcsec , and does not well define the structure of the radio source . however , at this lowest frequency we also find a diffuse source located near ra = @xmath28 , dec = @xmath29 which may ( or may not ) be associated with . unfortunately
, this structure lies off the field of the _ chandra _ x - ray observation .
the 4.9-ghz image of ( fig . 2 ) shows the radio emission to be from a small double source . at this resolution ( @xmath30 arcsec ) the two components
are well separated , but it is unclear based on the radio data whether one of the components is a core and the other is a one - sided jet or both components are lobes symmetrically disposed about the core .
the 8.4-ghz image has @xmath31 arcsec full - width - half - maximum ( fwhm ) but a high noise level , so that we can not currently distinguish which component has the flatter spectrum , or whether there is a core present between the two components .
further high - resolution , higher - sensitivity radio data are needed to resolve this issue .
shows clear diffuse x - ray emission ( see fig . [ fig : imaflux ] ) .
the x - ray peak is at ra = @xmath32 , dec = @xmath33 ( j2000 ) which is close ( less than 5 arcsec ) to the nasa / ipac extragalactic database ( ned ) position of galaxy mcg-01 - 12 - 005 , and coincident with the point source detected by wavdetect , dec = @xmath34 ( j2000 ) .
the data show this to be an error , since no x - ray emission is associated with mcg-01 - 12 - 006 .
simbad correctly associates with galaxy mcg-01 - 12 - 005 . ] .
the centre of the cluster of galaxies clg 0422 - 09 is at j2000 coordinates ra = @xmath35 , dec = @xmath36 , less than 2 arcsec from the x - ray peak .
the image suggests that the bulk of the x - ray emission from is not due to the expected bl lac object , but is diffuse in nature and originates from the intergalactic medium ( igm ) of galaxy mcg-01 - 12 - 005 , and intracluster medium ( icm ) of galaxy cluster clg0422 - 09 .
the optical emission of mcg-01 - 12 - 005 from the digital sky survey ( dss ; fig .
[ fig : dss ] ) is elongated in the n - s direction , while the x - ray emission in the central regions displays a slight elongation in the e - w direction .
the x - ray contours become more circular at larger distance from the centre .
a more extended structure lies east of the x - ray peak and causes the elongation of the central region of the x - ray contours .
we applied a gaussian filter of fwhm 0.98 arcsec to the ( 0.4 - 7.0 kev ) image , to enhance the extended structure ( see fig .
[ fig : imagauss ] ) .
two x - ray enhancements to the east of the x - ray peak , labelled with @xmath37 , seem to be connected to the compact source ( labelled with @xmath38 ) .
however this structure is barely significant .
the other extended structure ( marked as @xmath39 in the insert of fig . [ fig : imagauss ] ) is significant at 11@xmath40 above the surrounding area . in this figure
we also show the radio contours from the 4.9 ghz map .
we observe that the radio emission is located in a region of lower x - ray surface brightness than its immediate surroundings .
we note that the optical centre of the galaxy and the southern peak of the radio emission at 4.9 ghz coincide , suggesting that this is a likely location for the x - ray centre of the object .
unfortunately , the radio data are not sufficient to define the location of the core emission .
to avoid difficulties arising from the uncertain location of the x - ray centre , we excluded structure @xmath39 and the compact source ( @xmath38 ) for our analysis of the large scale structure .
we will discuss the structure at the centre in a separate section .
we replaced these two structures with their local surrounding mean counts and calculated the centroid of the resulting x - ray emission , which was found to be at ra = @xmath41 , dec = @xmath33 ( j2000 ) .
for any extended source analysis in the following , this will be fixed as the centre of the data .
we extracted a surface - brightness profile by integrating photons in concentric 1-arcsec wide annuli ( masked by the rectangular regions in [ fig : imaflux ] ) out to 118 arcsec and in the 0.4 - 2.0 kev energy range .
we subtracted a local background represented by the ( masked ) annular region between 120 and 130 arcsec .
the inner 3 arcsec of the profile display some fine - scale structure ( see fig . [
fig : sbprof ] ) , presumably from errors in the interpolation over structures @xmath38 and @xmath39 as discussed above .
the excess emission represented by bins 2 and 3 has little effect on the parameters of fits to the overall structure of the cluster , but it increases the total @xmath42 .
we thus excluded these two bins when fitting .
we binned the surface - brightness profile to have a signal - to - noise ( s / n ) of at least 3@xmath40 above the background in each bin .
we then initially fitted the profile with a 1d : @xmath43^{-3\beta+1/2 } \label{eq : beta}\ ] ] where @xmath44 and @xmath45 are the core radius and the index , and @xmath46 is the central surface brightness .
the model in equation [ eq : beta ] is not a good representation of the data , providing best - fitting parameters of @xmath47 counts s@xmath5 arcmin@xmath48 , @xmath49 , @xmath50 arcsec ( [ 10.7@xmath51@xmath52 kpc ) for a reduced @xmath42 , @xmath53 = 1.64 = 147.9/90 d.o.f .
( the @xmath53 rises to 1.95 if bins 2 and 3 are taken into account , but @xmath45 and @xmath44 do not change within the errors ) . and the 2-@xmath45 models . bins 2 and 3 were excluded when fitting .
] the fit improved when we fitted the profile with a sum of two @xmath45 models , @xmath54 where @xmath55 and @xmath56 are defined as in eq .
[ eq : beta ] .
we required the @xmath45 parameters of both models to be the same in order to have a smooth power law shape for the density profile parameters , but the @xmath57-test indicates no significant improvement in the fits when the second @xmath45 is allowed to vary independently from the first .
] at large radii .
this model gives an excellent fit of @xmath53 = 1.03 = 90.2/88 d.o.f .
, with @xmath58 counts s@xmath5 arcmin @xmath48 , @xmath59 arcsec ( [ 8.3@xmath60 @xmath52 kpc ) , @xmath61 counts s@xmath5 arcmin @xmath48 , @xmath62 arcsec ( [ 29.7@xmath63@xmath52 kpc ) , and @xmath64 . for comparison ,
the analysis of _ rosat _ data by reiprich & bhringer ( 2002 ) found a value of @xmath65 and @xmath66 70 @xmath5 kpc .
the @xmath45 values are in good agreement if the two-@xmath45 model is considered . in fig .
[ fig : sbprof ] we show the data points and both the single @xmath45 and double @xmath45 best - fit models .
the lower two windows show the residuals .
we derive the gas density distribution in combination with the spectral analysis .
we extracted the global spectrum of the source by selecting a circular region of radius 118 arcsec ( see fig .
[ fig : imaflux ] ) . to limit spurious effects produced by the boundary of the exposed part of the chip , we excluded two rectangles of size @xmath67 arcsec@xmath20 centred on ra = @xmath68 , dec = @xmath69 , and ra = @xmath70 , dec = @xmath71 ( j2000 ) , respectively with the same orientation as the ccd ( position angle = 30@xmath72 ) .
the compact source was also excluded .
we took background from the usual circle ( sec .
2.1.2 ; fig .
[ fig : imaflux ] ) .
the source has a 0.4 - 7.0 kev count rate of 1.27 cts s@xmath5 in this region .
we binned the spectrum to 60 counts per bin prior to background subtraction . the global spectrum
( see fig .
[ fig : spglob ] ) shows strong emission at the energy corresponding to the fe l complex .
other spectral features are also detected , although some of them are instrumental ( e.g. the region 1.9 - 2.1 kev near the ir edge ) and are excluded when fitting .
we fitted the global spectrum between 0.4 and 7.0 kev with thermal models in xspec ( version 11.3.0 ) , leaving the fractional abundance relative to solar free to vary , and initially adopting a galactic absorption of @xmath73 @xmath9 ( dickey & lockman 1990 ) .
a single model gives a bad fit with @xmath74 d.o.f .
, a temperature of about 2.8 kev and a metallicity of 1.4 solar , with large residuals at low energy . a much improved , though still unacceptable fit , with @xmath75 d.o.f .
, is obtained if the absorption is allowed to vary . the resulting = ( 1.19@xmath76 cm @xmath48 is roughly double the expected galactic absorption . an alternative explanation for the poor fit at low energy
might be thought to be systematic errors in the used ( version 2.26 ) calibration of the quantum efficiency degradation of the acis - s3 chip . to test this we repeated the fit but used only data above 1 kev .
the resulting fit ( table 1 , column 3 ) is statistically indistinguishable from a fit using the full energy range , and so we concluded that the high is not an artifact of the data processing . fitting a two- model with as a free parameter in the 0.4 - 7.0 kev range gives @xmath77 d.o.f .
the @xmath57-test indicates that the two - temperature model is a better representation of the data at 99.9% significance ( see fig .
[ fig : spglob ] ) .
the best - fitting results are listed in table [ tab : spglobfit ] .
the 0.2 - 10.0 kev total unabsorbed luminosity is 3.7@xmath78 erg s@xmath5 .
a model composed of a plus a power law gives a better fit ( @xmath42/d.o.f.=222.2/145 ) than the single temperature representation , but does not improve on the two - temperature model ( see table [ tab : spglobfit ] ) and the spectral index is not well determined .
we thus do not discuss this model any further .
[ cols="<,^,^,^,^,^,^ " , ] abundances are from grevesse and sauval ( 1998 ) . in the two - temperature model
the abundances of the two thermal components were tied .
@xmath79 is the normalisation given by xspec , as @xmath80 @xmath81 , where @xmath82 is the angular size distance to the source ( cm ) , @xmath83 is the electron density ( @xmath84 ) , and @xmath85 is the hydrogen density ( @xmath84 ) .
errors are quoted at 90% confidence for one interesting parameter .
the energy range used for fitting is listed for each fit .
the results on the global spectrum suggest the following : 1 .
the gas spectrum is consistent with a two - temperature model .
spectral fitting to asca data ( see e.g. ikebe et al .
1999 , makishima et al .
2001 ) suggested that a two - temperature model can mimic a multi - temperature gas as well as a cooling flow ( cf ) , and the contribution of the cd galaxy has to be taken into account .
we thus fitted the global spectrum with a model composed of a plus mkcflow ( mushotzky & szymkowiak 1988 ) .
the high temperature of the cf model was fixed to be the same as the model temperature .
the metallicities were forced to be the same and the absorbing column density left free .
the best - fit has @xmath86 d.o.f .
the high temperature is k@xmath87 kev , the low temperature is poorly constrained , with a best - fit value of k@xmath88 kev ( and upper limit of 0.45 kev ) , the metallicity @xmath89 , and the = ( 1.27@xmath90 @xmath9 , all in agreement , at 90% confidence , with the two - temperature model described above , though a worse fit .
the mass deposition rate is 10@xmath91 m@xmath92 yr@xmath5 .
2 . the absorption is higher than galactic by more than a factor of 2 . as shown later
4.3.1 ) the excess absorption encompasses the whole field of view and not only the cool central part of the cluster , suggesting that this excess absorption is of galactic rather than intrinsic origin .
no co map in the direction of our source is good enough to establish if molecular gas contributes the needed column .
however , there is a good correlation between galactic hi and the infra - red ( ir ) emitting dust ( see boulanger et al .
pointecouteau et al .
( 2004 ) found a remarkable agreement between their x - ray fitted profile , and the total @xmath15 profile derived from this correlation . to test if this interpretation is applicable to , we looked at the infra - red astronomical satellite ( iras ) map at 100 @xmath93 m , in the direction of the source ( schlegel , finkbeiner , & davis 1998 ) .
the map shows bright clumpy emission , with strong gradients .
the iras map is at far larger spatial scales than the field of view , but the 100 @xmath93 m flux in the vicinity of the field is @xmath94 mjy sr@xmath5 .
the ir - relation of boulanger et al . ( 1996 ) then predicts @xmath95 hi @xmath9 .
this is in good agreement with our best - fit results from the global spectrum , suggesting the absorption excess is , indeed , from molecular gas . in order to perform spatially
resolved spectroscopy , we selected concentric annular regions out to 118 arcsec , masked by the rectangles seen in fig .
[ fig : imaflux ] .
the first bin is of radius 5 arcsec , just outside the bright structures in the centre .
possible variations of the spectral modelling within this region are discussed in section [ sec : centre ] , but they do not affect our conclusions here . to the deprojected data points ( see sec .
[ sec : ps ] ) ; * bottom * : projected ( grey ) and deprojected ( black ) metallicity profiles.,title="fig : " ] to the deprojected data points ( see sec .
[ sec : ps ] ) ; * bottom * : projected ( grey ) and deprojected ( black ) metallicity profiles.,title="fig : " ] we obtained a projected temperature profile by fitting each annulus with a model .
we used the same background as for the global spectrum .
all spectra were binned to 15 counts per bin prior to background subtraction , and fitted between 0.4 and 7.0 kev . in order to test if is constant within the area of the x - ray detection , we fitted each annulus with a model with , temperature and abundances as free parameters .
we found that all the column densities are consistent with the value found in the global analysis ( within 1@xmath40 errors ) . to strengthen this result , we fitted the spectrum in each annulus by excluding data below 1 kev .
( this also tests for possible spatial variation of the quantum efficiency degradation at low energy . )
these fits are consistent with those in the full energy range .
we thus confirm that is a factor of two higher than the value predicted by the hi measurements at all radii , and that it is constant over a scale @xmath96 arcsec ( 90 kpc ) , with a global value of @xmath97 @xmath9 . in the following the column density
is fixed to this value .
we next fitted the spectra again in each annulus , between 0.4 and 7.0 kev , with an absorbed model with fixed .
a single - temperature model is an acceptable representation of the data in each annulus .
however , to test possible contamination from a non - thermal source in the inner bin , we added a second component , represented by a power law , when fitting the first bin .
the additional second component is not required ( we also added a second thermal model , and the conclusion is similar ) .
the projected radial temperature and metallicity profiles are shown in fig .
[ fig : tzprof ] by the grey data points ( crosses ) .
displayed errors are 1@xmath40 for one interesting parameter .
there is a clear temperature drop towards the centre , with a minimum projected k@xmath98 kev .
the temperature rises almost monotonically with radius to reach a maximum of 2.8 kev in the outermost bin , which is in agreement with the global temperature found with rosat ( white et al .
this suggests that the data are giving us a fine - scale measurement of the cool gas in the core region .
the projected metallicity profile drops continuously from a value of 0.8 @xmath0 in the centre to @xmath99=0.35 in the outermost region .
we accounted for projection effects in the profiles of fig .
[ fig : tzprof ] by using the projct model in xspec , and assuming spherical symmetry .
the volumes of each shell are calculated by the model once the inner and outer boundaries of each annulus are added to the fits headers of each spectrum .
since the cluster emission extends well beyond our outermost annulus , the emitting volume associated with the outermost annulus will be too small .
the effect is to underestimate the brightness of the next annulus in .
we fitted an absorbed , single - temperature mekal model in which the temperature and metallicity were left as free parameters at each radius .
the column density was fixed to the global value as before .
the best - fitting model gives @xmath42 = 802.6/791 d.o.f .
the asterisks in fig .
[ fig : tzprof ] show the deprojected temperature and abundance profiles of the cluster .
when projection effects are taken into account , the temperature in the central bin is significantly cooler than its respective projected value , since the hot gas from the outer regions seen in projection is accounted for . for the outer annuli the projected and deprojected temperatures agree within the errors .
the metallicity declines outwards and within the errors is also consistent with a constant value of 0.8 @xmath0 within 40 arcsec and then drops to @xmath100 in the outermost bin . from the deprojected emission measure it is possible to calculate the ion and electron density directly from xspec . under the assumption that the density is constant within each annulus ,
the ion density is simply : @xmath101 where @xmath79 is the normalisation from the deprojected model fitting , @xmath82 is the angular diameter distance and @xmath102 is the volume of the shell .
the electron density is then obtained from the relation ( see e.g. worrall & birkinshaw 2004 ) @xmath103 where @xmath104 is the mass fraction in hydrogen , @xmath105 for normal cosmic ( solar ) abundances .
the derivation of the electron density directly from xspec is a suitable alternative option to the classical method of inverting the surface brightness profile into a density profile .
the latter method requires us to estimate the emissivity function at each radius , and this is a strong function of the assumed temperature for a cluster at @xmath106 kev , even if energies below 2.0 kev are used when generating the surface - brightness profile .
we will see that the two methods are , in fact , in agreement . in the top panel of fig .
[ fig : nesp ] we show the electron density derived from the best - fit normalisation , assuming that the density is constant within each annulus .
since the cluster extends well beyond our outermost annulus , we overestimate the electron density in the last bin . to correct for this effect
we note that : 1 .
the density derived from the isothermal described in sec .
[ sec : sbprof ] is in agreement with the data points in fig .
[ fig : nesp ] , for all annuli but the last ( continuous line in fig .
[ fig : nesp ] ) , despite the clear temperature variations ( fig .
[ fig : tzprof ] ) ; 2 .
the rosat data provide the best measures of the cluster at larger scale , and the best - fitting @xmath45 found by reiprich & bhringer ( 2002 ) is in excellent agreement with our @xmath45 value .
this suggests that the extrapolation of our best - fitting is a reasonable representation of the outer shells of the gas emission . for these reasons
, we corrected the density in the last annulus to equal the -derived density at its distance from the centre ( 83 arcsec ) .
we added a 20% systematic error to this bin .
the corrected density is 50% lower than the original value , as shown in fig .
[ fig : nesp ] by the dashed point .
a similar argument applies to the proton density profile .
we used the corrected values to calculate the entropy and pressure profiles below .
we also note that the first data point is higher than ( but consistent with ) the profile ( continuous line ) .
this arises from the exclusion of the excess emission in the central 5 arcsec when fitting the @xmath45 model ( second and third bin in fig .
[ fig : sbprof ] ) , while this excess was not excluded when we extracted the central spectral annulus ( but note that the compact source and the extended structure @xmath39 described in sec .
[ sec : morphology ] were masked ) .
limits of this profile . *
middle : * entropy profile .
the continuous line is the best - fitting power law ( plus constant ) model ( see text ) .
the dashed line represents the entropy profile derived by using the best - fitting temperature profile model and the -derived density profile .
the dot - dashed line represents the entropy profile when the exponent is fixed to be 1.1 as expected from shock heating in spherical collapse ( e.g. tozzi & norman 2001 ; voit et al .
2002 ) ; * bottom * : pressure profile .
the continuous line is the pressure profile derived from the deprojected temperature profile and the density profile implied by the fit.,title="fig : " ] limits of this profile . *
middle : * entropy profile .
the continuous line is the best - fitting power law ( plus constant ) model ( see text ) .
the dashed line represents the entropy profile derived by using the best - fitting temperature profile model and the -derived density profile .
the dot - dashed line represents the entropy profile when the exponent is fixed to be 1.1 as expected from shock heating in spherical collapse ( e.g. tozzi & norman 2001 ; voit et al .
2002 ) ; * bottom * : pressure profile .
the continuous line is the pressure profile derived from the deprojected temperature profile and the density profile implied by the fit.,title="fig : " ] limits of this profile . *
middle : * entropy profile .
the continuous line is the best - fitting power law ( plus constant ) model ( see text ) .
the dashed line represents the entropy profile derived by using the best - fitting temperature profile model and the -derived density profile .
the dot - dashed line represents the entropy profile when the exponent is fixed to be 1.1 as expected from shock heating in spherical collapse ( e.g. tozzi & norman 2001 ; voit et al .
2002 ) ; * bottom * : pressure profile .
the continuous line is the pressure profile derived from the deprojected temperature profile and the density profile implied by the fit.,title="fig : " ] the middle panel of fig .
[ fig : nesp ] shows the profile of @xmath107 ( lloyd - davies et al .
2000 ) , which is related to the entropy per unit mass in the gas , derived from the temperature and density profiles obtained from the spectral fits .
it is useful to fit the temperature profile with an analytical function .
here we use ( allen et al .
2001 ) @xmath108 where our best - fit has @xmath109 kev , @xmath110 kev , and @xmath111 arcsec , close to the value of @xmath112 from the radial profile fit .
the exponent , @xmath113=2 , was fixed to the best - fitting value found by allen et al .
we then calculated the entropy profile using the best - fitting temperature model and the isothermal -derived electron density profile .
this is shown in fig .
[ fig : nesp ] by the dashed line .
the dot - dashed line shows the s @xmath114 behaviour found in analytical and semi - analytical models ( tozzi & norman , 2001 ; voit et al .
2002 ) , and which is also found in numerical simulations without complex gas physics ( e.g borgani et al .
the normalisation is arbitrary .
a simple power - law fit of the form @xmath115 to this entropy profile yields a slope @xmath116 ( @xmath117 is in arcsec ) .
the effect of adding a constant ( @xmath118 ) is to increase the slope .
the full details of the power law plus constant fit are @xmath119 , @xmath120 and @xmath121 .
an @xmath57-test indicates that there is less than 1 per cent chance that the simple power law is a better fit .
it is now common to find the entropy profile scaled by @xmath122 , the radius within which the mean cluster density is 200 times the mean density of the universe .
if we use the equation in evrard , metzler & navarro ( 1996 ) , and the external temperature of the cluster ( 2.9 kev ) , we obtain @xmath123 mpc for .
this implies that our outermost annulus corresponds to @xmath124 @xmath122 , which is too small to allow useful comparisons with previous work on clusters with similar temperatures .
pratt & arnaud ( 2004 ) found @xmath125 for a small sample of clusters with @xmath126 kev .
although this is calculated in the range ( 0.05 - 0.5 ) @xmath122 , it looks consistent with the result we find here for . the bottom of fig .
[ fig : nesp ] shows the pressure profile , where we compare the results when we use the xspec proton density profile ( data points ) and the isothermal density profile ( continuous line ) .
in previous sections we have described the small - scale structure in the central 4 arcsec of ( see fig .
[ fig : imagauss ] ) . in the surface - brightness profile ( sec .
[ sec : sbprof ] ) we found excess emission above a , despite the fact that the compact source @xmath38 detected with wavdetect , and the more elongated emission labelled as @xmath39 in fig . [
fig : imagauss ] were masked before extracting the radial profile .
we also noted that the centre of the galaxy ( and the cluster ) displays a lower x - ray surface brightness than the immediate surroundings . in this section
we explore the nature of the compact source and the extended structure , and investigate the relation between the central gas and the radio source .
we extracted a spectrum of source @xmath38 from a circular region of radius 1.35 arcsec centred on ra= @xmath35 , dec = @xmath127 ( j2000 ) .
the local background was estimated in the surrounding annular region within radii 1.35 and 2.95 arcsec .
there are only 38 net counts in this region . adopting c - statistics we found that a power law of spectral index 2.5@xmath128 is a reasonable description of the data ( the column density was fixed to the global value ) . with this model
the unabsorbed 0.2 - 10 kev flux of the source is 3.9@xmath129 ergs s@xmath5 @xmath9 .
a thermal model also describes the data well , but the temperature is not constrained . with the data available , we can not state if the source is a clump of thermal gas within the cluster or a background ( foreground ) quasar , although the non - thermal interpretation seems more plausible from the spectral analysis . given the positional offset between this source and the radio emission , it is unlikely that this is the bl lac object we were looking for .
we mask this source for the following analysis .
we also extracted a spectrum of structure @xmath39 .
we have even fewer photons but the spectrum shows a clear thermal signature , and a temperature consistent with the temperature that we found for nearby gas when this structure was excluded ( sec .
[ sec : tprof ] ) . since the surface - brightness profile displays excess emission between 1 and 4 arcsec from the centre , even after excluding structure @xmath39 , we suggest that @xmath39 is of the same nature as the excess surface brightness , and its brightness may be due to projection effects or the particular geometry of the source . to investigate the spectral signature of the excess surface brightness , we extracted the surface brightness profile of between 0 and 15 arcsec in smaller radial annuli than in sec .
[ sec : sbprof ] , and in two energy bands : a soft band ( s ) at 0.4 - 2.0 kev , and a hard band ( h ) at 2.5 - 6.5 kev . at larger scale
the profile was binned in 5 arcsec wide annuli .
we then obtained an hardness ratio ( hr ) profile by calculating @xmath130 ( fig .
[ fig : hr ] ) .
the average value of the hr is -0.77 , reflecting the soft spectrum of the gas .
points at radius @xmath131 arcsec are consistent ( within the statistical errors ) with this average value .
we generated s and h images and smoothed them with a gaussian filter of @xmath40=2 pixels ( 0.98 arcsec ) . in the bottom of fig .
[ fig : hr ] we show the soft image in grey scale .
the black contours are obtained from the gauss filtered hard image and the white contours are from the radio map at 4.9 ghz .
the hard emission is detected above the background at @xmath132 .
we observe that the hard emission is centred in the region of low surface brightness soft x - ray emission coincident with the optical centre of the galaxy , and that it is oriented in the same direction as the radio source but is more extended .
the peaks of the hard x - ray and radio emission are not coincident , but this is likely to be a consequence of the smoothing , and is not significant .
the peak to the north - west ( black contours ) is probably an unrelated source .
black contours are obtained from the gaussian filtered hard image with the same @xmath40 .
white contours are from the 4.9 ghz map .
the lower contour is traced at 0.5 mjy and they are logarithmically spaced up to 10 mjy .
, title="fig : " ] arcsec .
black contours are obtained from the gaussian filtered hard image with the same @xmath40 .
white contours are from the 4.9 ghz map .
the lower contour is traced at 0.5 mjy and they are logarithmically spaced up to 10 mjy .
, title="fig : " ] what is the physical relationship between the radio source and the surrounding icm ?
first of all , there are two peaks in the 4.9 ghz radio map .
we label as @xmath133 the southern component at ra = @xmath134 , dec = @xmath135 , and as @xmath79 the northern component at ra = @xmath136 , dec = @xmath137 .
the 8.46 ghz map shows a peak of flux density 0.8 mjy between component @xmath133 and @xmath79 , while components @xmath133 and @xmath79 are resolved and show steeper spectra .
more sensitive radio observations would be needed to decide whether the core of the source lies in @xmath133 or @xmath79 , or in the weak component that appears to lie between them .
if we make the assumption that the source is in equipartition between radiating particles and magnetic field , we find an equipartition magnetic field of 7.4 nt and a pressure of 1.5@xmath138 pa for both components ( we assumed a spectral slope of 2.2 , minimum electron lorentz factor of 100 , maximum electron lorentz factor of 10@xmath139 , and filling factor equal to 1 ) . from the pressure profile discussed in sec .
[ sec : ps ] the pressure of the external medium at the distance of each of the radio components from the optical centre is @xmath140 pa , which is only a factor of @xmath141 from the internal minimum pressure of the source , suggesting that the source is close to pressure balance .
we summarise below the results from our study of using the high resolution of . 1 .
the x - ray emission from is extended , without the strong point - like component expected for a bl lac object .
is in fact associated with a cluster of galaxies , and an early type galaxy which dominates the central optical light ; 2 .
we found absorption in excess of galactic .
this is not intrinsic to the source but of galactic origin , as shown by the agreement between our x - ray measure and an independent measure which uses the -ir emission relation ; 3 .
we find that the gas is locally isothermal , with a temperature decreasing towards the centre , as found for cooling - flow clusters , and with a temperature of @xmath142 kev at @xmath143 kpc ; however , a cf model gives a mass deposition rate of only 10 @xmath144 yr@xmath5 ; 4 .
the metallicity profile declines outwards and it is consistent with an almost constant @xmath145 at @xmath146 kpc , and then drops to 0.37@xmath0 ; 5 .
the gas entropy profile behaves as @xmath147 , close to the relation predicted by analytical models of shock heating in spherical collapse , but shows a small core reflecting possible entropy modification in the centre ; 6 .
the central radio source has a minimum pressure that is a factor of 2 below the external medium .
hard x - ray emission from the cluster centre has the same orientation as the radio source , but is more spatially extended .
the current observation gives us a precise look into the core of an intermediate temperature galaxy cluster .
the radial temperature profile is reminiscent of cooling core clusters , harbouring denser gas in the centre .
are we observing a classical cooling flow ?
arguments against the classical cooling - flow models have been extensively discussed since the advent of ( e.g. tamura et al .
2001 ; peterson et al .
2003 ; bhringer et al .
2002 ) . in our case , fitting a cooling - flow model to the global spectrum does not give a better fit than a two - phase model , but the temperature profile does suggest a cooling flow .
another interesting possibility is that we are observing the interstellar medium ( ism ) of the early type galaxy .
indeed our observation maps x - rays only within the optical envelope of the galaxy . a clear indication in favour of this interpretation is given by the metallicity profile .
we are not able to map individual elements with the current data , but the fitted metallicity should be dominated by the iron abundance .
the metallicity profile clearly shows a decline from the centre outwards , which can also be seen as a break at @xmath148 kpc , with an average global relative abundance of @xmath149 within this radius and @xmath99=0.37 outside .
work based on asca data found that clusters harbouring cd galaxies display a significant increase in metal abundance towards the centre ( e.g. fukuzawa et al . 2000 ; matsushita et al .
similar results were found with bepposax data ( e.g. de grandi & molendi 2001 ) , and suggest that this is due to metal enrichment by sn type ia in the cds ( e.g. finoguenov & ponman 1999 , matsushita et al . 2002 , 2003 ) . the optical galaxy that we associate with , mcg-01 - 12 - 005 , is not well studied and it is not catalogued as a cd galaxy . however , this interpretation is still plausible , since prior to the advent of it was almost impossible to map in such detail the inner regions of galaxy clusters , and objects without the striking optical properties of cd galaxies are just starting to be discussed .
an acceptable estimate of the galaxy x - ray temperature is given by the first bin of the temperature profile ( k@xmath150@xmath1 1.3 kev ) .
we find a corresponding x - ray luminosity of @xmath151 erg s@xmath5 , once projection is taken into account , and after extrapolating to 40 kpc ( when the cluster emission dominates ) .
although the temperature is rather high , the values we find are in agreement with the @xmath12 relation of osullivan et al .
( 2003 ) for early type galaxies , adding another piece of support to the ism interpretation for the x - ray emission within 40 kpc of this object .
while the temperature increases rapidly after the first bin , and at 40 kpc it is already @xmath152 kev which is too high for the atmosphere of a galaxy , we note that the metallicity and temperature profiles could change independently with different radial functions because of the different physics associated with the transport of heavy ions outwards and thermal energy inwards .
we also note that the gas is consistent with a two - phase metallicity , with a value close to solar within 35 arcsec and close to a common value for galaxy clusters beyond 35 arcsec .
this suggests that little mixing between the ism and icm has occurred .
we can use the comparison between the radio source and the ambient medium properties to map the formation history of the core of .
we find that the gas entropy declines continuously towards the centre , in agreement with recent results on galaxy groups and clusters ( e.g. ponman , sanderson & finoguenov 2003 ; pratt & arnaud 2003 ; khosroshahi , jones & ponman 2004 , pratt & arnaud 2004 ) , typically of the same temperature as .
moreover , the large - scale radial dependence of the entropy we measure is in agreement with both analytical models of spherical collapse , and numerical simulations based solely on gravitational collapse ( tozzi & norman , 2001 , borgani et al .
2001 , voit et al .
it does seem that there is or there has been some modification of the core entropy of this cluster ( galaxy ) .
this is seen in the radial dependence of the entropy , where a simple power - law fit gives a shallower slope than expected from pure shock heating , and where a power law plus constant model is clearly a better fit .
however there is no flat isentropic core ( entropy plateau ) .
the deviation of other classical scaling laws , such as the @xmath153 relation ( e.g. arnaud & evrard 1999 ) , from the prediction of purely gravitational models has suggested that some extra heating has occurred .
it has been proposed that sources at the centres of clusters ( such as agn or radio galaxies ) could cause this heating ( e.g. valageas & silk 1999 ; wu , fabian & nulsen 2000 ; churazov et al .
2002 ; ruszkowski & begelman 2002 ) .
the small core in the entropy distribution of this object would suggest that entropy has been modified in some way , and the central radio source , although weak , may be a plausible candidate for this modification ( e.g. roychowdhury et al .
2004 ) .
an alternative observational indication is given by the comparison between the radio and x - ray luminosity in the cooling region , which should be similar , once a reasonable factor for the conversion between power and radio emission is assumed .
the radio source at the centre of is particularly small .
the radio luminosity is of order @xmath154 erg s@xmath5 , while the x - ray luminosity of the soft thermal component within 35 arcsec is @xmath151 erg s@xmath5 .
if only 0.1% of the total power output goes into the radio emission , then the rest of the power can heat the gas .
however , this is a rather low efficiency for conversion of the source power into radio emission , and also a high efficiency for heating . from this evidence
, it is unlikely that the radio source is supplying sufficient heat to stop any cooling flow .
since we do not have an independent measure of the magnetic field , we assumed equipartition , yielding the result that the minimum internal pressure of the source is about a factor of 2 lower that the external pressure . the minimum pressure is a lower limit ,
thus we are facing two possibilities for the small size of the source : the radio source is old and is dying , or the radio source is young . in the first case ,
the source was bigger and more powerful in the past .
it may have heated the external medium which is now hotter than at the origin , as can be inferred from the relative distribution of hard and soft emission in the core of ( fig .
[ fig : hr ] ) .
it is also possible that the source inflated bubbles of rising gas , carrying energy and pressure , as observed for example in m87 ( e.g ghizzardi et al .
however , radio sources are short - lived ( 10@xmath155 yr ) and when they are starved of fuel they lose power
. if this is the scenario we are observing , the radio source should be relatively old , and is in its fading phase .
hints in favour of this picture are the small size of the radio source , the distribution of the hard emission in the central 5 arcsec ( fig .
[ fig : hr ] ) , the small entropy core , and the relatively large external pressure at the location of the radio galaxy .
we can not constrain the spatial distribution of the hard emission , which is strongly limited by the statistics .
it is unlikely that the hard emission is non - thermal , since high energy electrons can not live longer than their parent population .
thus it is possible that what we are observing is the signature of gas heating .
the second possibility is that the source is young .
the engine has just turned on and the source is expanding .
this would explain i ) the low surface brightness of the gas at the location of the radio source , and ii ) the high surface brightness just outside it .
this can be interpreted as an expanding source which is pushing the ambient gas away .
if the source is young then it may not have had time to affect the entropy of the gas , which could explain the relatively small heating signatures in the entropy profile .
the observation of shows that a bl lac object is absent and we observe an extended , cluster - scale source , which was previously discussed by edge & stewart ( 1991 ) .
the central x - ray emission is likely to be due to the ism of the early - type galaxy at the centre of the cluster , and only at @xmath143 kpc does the icm dominate over the ism .
the central x - ray emission was little affected by heating in the way observed for other low - temperature clusters .
if the source is relatively young , our results represent indirect support for models where agn activity in the central galaxy may be cause of the breakdown of self - similarity .
we are very grateful to g.w .
pratt for many helpful discussions .
we thank m.j .
hardcastle for helping in the initial phase of the work .
we thank an anonymous referee for a thorough referee s report which enabled us to improve the manuscript .
e.b . and d.m.w .
thank pparc for support .
this research has made use of the simbad database , operated at cds , strasbourg , france , and the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration .
the national radio astronomy observatory is a facility of the national science foundation operated under cooperative agreement by associated universities , inc .
allen , s.w . ,
schmidt , r. , fabian , a.c .
, 2001 , mnras , 328 , l37 arnaud , m. , evrard , a.e . , 1999 ,
mnras , 305 , 631 birkinshaw , m. , worrall , d. m. , hardcastle , m. j. , 2002 , mnras , 335 , 142 bhringer , h. , matsushita , k. , churazov , e. , ikebe , y. , chen , y. , 2002 , a&a , 382 , 804 borgani s. , governato , f. , wadsley , j. , menci , n. , tozzi , p. , lake , g. , quinn , t. , stadel , j. , 2001 , apj , 559 , l71 boulanger , f. , abergel , a. , bernard , j .- p . , burton , w. b. , desert , f .- x .
, hartmann , d. , lagache , g. , puget , j .- l . , 1996 , a&a , 312 , 256 churazov , e. , sunyaev , r. , forman , w. , bhringer , h. , 2002 , mnras , 332 , 729 de grandi , s. , molendi , s. , 2001 , apj , 551 , 153 dickey , j. m. , lockman , f. j. , 1990 , ara&a . 28 , 215 edge , a. c. , stewart , g. c. , 1991 , mnras , 252 , 414 evrard , a.e . ,
metzler , c.a . ,
navarro , j.f .
, 1996 , apj , 469 , 494 finoguenov , a. , ponman , t. j. , 1999 , mnras , 305 , 325 fukuzawa , y. , makishima , k. , tamura , t. , et al . ,
2000 , mnras , 313 , 21 ghizzardi , s. , molendi , s. , pizzolato , f. , de grandi , s. , 2004 , apj , 609 , 638 giommi , p. , tagliaferri , g. , beuermann , k. et al .
, 1991 , apj , 378 , 77 grevesse , n. , sauval , a.j . , 1998 ,
space science reviews , 85 , 116 ikebe , y.,makishima , k. , fukuzawa , y.,tamura , t. , xu , h. , ohashi , t. , matsushita , k. , 1999 , apj , 525 , 58 ikebe , y. , reiprich , t. h. , bhringer , h. , tanaka , y. , kitayama , t. , 2002 , a&a , 383 , 773 khosroshahi , h.g . , jones , l.r . , ponman , t.j .
, 2004 , mnras , 349 , 1240 kirhakos , s.d . ,
steiner , j.e .
, 1990 , aj , 99 , 1722 liedahl , d. a. , osterheld , a. l. , goldstein , w. h. , 1995 , apj , 438 , 115 lloyd - davies , e.j . , ponman , t.j . , cannon , d.b . , 2000 , mnras , 315 , 689 makishima , k. , ezawa , h. , fukuzawa , y. , et al . , 2001 , pasj , 53 , 401 matsushita , k. , makishima , k. , ikebe , y. , et al .
, 1998 , apj , 499 , l13 matsushita , k. , belsole , e. , finoguenov , a. , bhringer , h. , 2002 , a&a , 386 , 77 mewe , r. , gronenschild , e.h.b.m . ,
van den oord , g.h.j .
, 1985 , a&as , 62 , 197 mushotzky , r. f. , szymkowiak , a. e. , 1988 , in `` cooling flows in clusters and galaxies '' ; proceedings of the nato advanced research workshop , cambridge , england , june 22 - 26 , 1987 ( a89 - 27526 10 - 90 ) .
dordrecht , netherlands , kluwer academic publishers , 1988 , p. 53 - 62 .
osullivan , e. , ponman , t. j. , collins , r. s. , 2003 , mnras , 340 , 1375 peterson , j.r . ,
kahn , s.m .
, paerels , f.b.s . , et al . , 2003 , apj , 590 , 207 ponman , t.j . , sanderson , a.j . , & finoguenov , a. , 2003 , mnras , 343 , 331 pratt , g.w . , arnaud , m. , 2003 , a&a , 408 , 1 pratt , g.w . , arnaud , m. , 2004 , a&a , in press , preprint astro - ph/0406366 pointecouteau , e. , arnaud , m. , kaastra , j. , de plaa , j. , 2004 , a&a , in press , astro - ph/0403596 reiprich , t.h . , bhringer , h. , 2002 , apj , 567 , 719 roberts , m.d . , et al . , 1998 , a&a , 337 , 25 roychowdhury , s. , ruszkowski , m. , nath , m. m. , begelman , m.c . , 2004 , astro - ph/0401161 ruszkowski , m. , & begelman , m. c. , 2002 , apj , 581 , 223 schlegel , d.j . , finkbeiner , d.p . , davis , m. , 1998 , apj , 500 , 525 smith , r. k. , brickhouse , n. s. , liedahl , d. a. , raymond , j. c. , 2001 , apj , 556 , 91 tamura , t. , kaastra , j.s . ,
peterson , j.r .
, et al . , 2001 , a&a , 365 , l87 tozzi , p. , norman , c. , 2001 , apj , 546 , 63 valageas , p. , silk , j. , 1999 , a&a , 350 , 725 vron - cetty , m .-
, vron , p. , 2003 ,
a&a , 412 , 399 voit , g.m . ,
bryan , g.l . ,
balogh , m.l .
, bower , r.g . , 2002 ,
apj , 576 , 601 white , d.a . ,
jones , c. , forman , w. , 1997 , mnras 292 , 419 worrall , d. m. , birkinshaw , m. , 2004 , in `` physics of active galactic nuclei at all scales '' .
d. alloin , r. johnson , p. lira , in press wu k.k . ,
fabian , a.c .
, nulsen , p.e.j . , 2000 ,
mnras , 318 , 889 | we present a observation of the candidate bl lac object . the x - ray emission from
is clearly extended , and is associated with an optical early - type galaxy , mcg-01 - 12 - 005 , at the centre of cluster clg 0422 - 09 .
we do not detect a point source which can be associated with a bl lac , but we found a small radio source in the centre of mcg-01 - 12 - 005 . the cluster gas temperature mapped by the observation drops continuously from 80 kpc towards the centre , and
is locally single phase .
we measure a metallicity profile which declines outwards with a value 0.8@xmath0 in the centre , dropping to 0.35@xmath0 at larger radius , which we interpret as a superposition of cluster gas and a dense interstellar medium ( ism ) in the central galaxy .
although the temperature profile suggests that conduction is not efficient , the ism and intra - cluster medium seem not to have mixed .
the entropy profile declines continuously towards the centre , in agreement with recent results on groups and clusters .
the radio source appears to have had some effect in terms of gas heating , as seen in the small scale ( @xmath110 kpc ) entropy core , and the asymmetric hard emission on the same scale .
galaxies : individual : mcg-01 - 12 - 005 and exo 0423.4@xmath20840 x - ray : galaxies : interstellar medium radio : galaxies galaxies : active : bl lacertae galaxies : clusters : intergalactic medium : individual : exo 0422 , clg0422 - 09 |
since the discovery of fractional quantum hall effect(fqhe ) and high @xmath1 cuprates , it has been realized that a large class of phases and phase transitions can not be described by landau symmetry breaking theory .
enormous efforts have been made to understand the underlying physics for these new systems during the last two decays .
it has been realized that the strongly correlated nature plays an essential role for these new classes of quantum matter .
the most successful and powerful approach to study these new systems is based on new classes of variational wave functions .
for example , the famous laughlin wave function@xcite successfully explains the quantized nature of hall conductance at rational filling .
it turns out such a new state is very different from a symmetry breaking state and it can describe new orders of quantum matter topological order@xcite . although the essential physics of high-@xmath1 is still under controversial , it is believed the relevant low energy physics is controlled by a class of metastable states the resonating - valence - bond(rvb ) states@xcite .
a quantitative description for the rvb states is based on the projective wave function approach , which was first proposed to satisfy the no - double - occupancy constraint for ( repulsive)hubbard model in the strong coupling limit@xcite .
it turns out this new class of states can also describe new phases of matter with topological order or quantum order@xcite .
later , those projective functions are widely used to describe the novel phenomena in strongly correlated systems , including frustrated magnets@xcite as well as the lattice fractional quantum hall states@xcite . despite the success of projective states ,
they are specially designed to describe states with particular topological order or quantum order and it is very difficult to study the competing effect between different orders .
thus , it is very important to establish a unified frame work to encode different orders of quantum matter . in ref.@xcite , a natural generalization of the projective states , the grassmann tensor product states(gtps )
have been proposed as generic variational wave functions to study interacting boson / fermion systems .
however , only local gtps(gtps with short range bonds)can be efficiently simulated approximately@xcite . as a result , it is also very important to understand what kind of states can be faithfully represented in a local way . for spin systems , refs . @xcite
have shown that the ground states of non - chiral topological phases , the so called string - net condensates@xcite , admit a local tensor product states(tps ) representation .
later , the fermionic version of string - net states was proposed in ref.@xcite and it has been shown they can describe non - chiral topological orders(e.g . ,
fractional topological insulators ) in fermion systems .
similar as the bosonic string - net states , their ground states can also be faithfully represented as gtps since the parent hamiltonians for these new classes of phases can be described as summations of ( fermionic)commuting projectors . in a recent work
, it has been shown even for systems with chiral topological orders , all the physical quantities still admit a local gtps representation@xcite despite that their ground states wave functions contain long range bonds @xcite .
thus , best to our knowledge , the gtps variational approach can in principle describe all kinds of gapped local boson / fermion systems in @xmath2d .
clearly , the advantage of this new variational approach is that it provides a unified description for many different orders of quantum matter and allows us to study the competing effect between these different orders . on the other hand , from the quantum information and computation perspective
, it has been shown that ground states of gapped local hamiltonians obey so - called area laws . for local boson / spin systems with translational invariant , states satisfy such a property can be efficiently represented by the class of so - called matrix product states(mps ) in one dimension and tensor product states(tps ) or projected entangled pair states ( peps ) in higher dimensions@xcite .
recently , a fermionic generalization of those states the fermionic projected entangled pair states ( fpeps ) were proposed and have been bench marked for many interesting free / interacting fermion systems@xcite . in ref.@xcite , it has been shown all fpeps can be represented as ( local)gtps .
although gtps are conceptually useful , the implementations in generic strongly correlated boson / fermion systems are still not easy since the tensor contraction for generic gtps is an exponential hard problem .
similar difficulties occur for peps(fpeps)@xcite and many efforts have been made based on the mps algorithm@xcite
. however , the mps algorithm still has a very big cost to handle large system with periodical boundary condition(pbc)@xcite .
alternatively , based on the concept of renormalization@xcite , the so - called tensor - entanglement renormalization group(gterg)@xcite method and its recent development@xcite turn out to be very successful for systems with pbc .
similar as terg , based on the renormalization principle for grassmann variables , the grassmann - tensor - entanglement renormalization group(gterg ) was proposed in ref.@xcite to simulate physical measurements for gtps approximately . however , a naive minimization procedure for generic gtps variational approach will still be very hard due to the large number of variational parameters when inner dimension @xmath0 increases(scale as @xmath3 on honeycomb lattice and @xmath4 on square lattice ) . for tps
, it is well known the imaginary time evolution algorithm is the best method to solve this problem .
thus , it is natural to generalize this algorithm into gtps . in this paper , we will show based on the fermion coherent state representation , all the algorithms developed for usual tps can be easily generalized into gtps .
the paper is organized as following : in section ii , we represent a standard form of gtps , which only contains one specie of grassmann variable for each inner index and significantly simplifies the representation for numerical calculations . in section iii , we first give a brief review about the concept of imaginary time evolution algorithm for tps and then present the detail implementation of the imaginary time evolution algorithm for gtps .
finally , we demonstrate the algorithm for a simple spinless fermion system on honeycomb lattice , including both off - critical and critical cases . in section iv
, we describe the environment effect of the gterg algorithm and present a simple improved algorithm .
we implement the algorithm for a critical free fermion system on honeycomb lattice and find a significant improvement .
finally , we briefly summarize our results and discuss possible future developments along this direction .
in this section , we will introduce a standard form to represent gtps . in the standard form ,
each link only associates with one grasmann variable , thus , the representation in the numerical calculations will be simplified significantly . a graphic representation for the gtps . the open circle which connects to links @xmath5 and @xmath6 represents the grasmann metric @xmath7 . the filled circle on the physical site @xmath8 represents the grassmann tensors @xmath9 .
the grassmann numbers @xmath10 associate with the grasmann metric @xmath7 and the dual grassmann numbers associate with the grassmann tensor @xmath9 .
there are a pair of indices @xmath11 live on link @xmath5 .
@xmath12 is called the bosonic index while @xmath13 are called the fermionic indices . ]
let us recall the generic gtps wavefunctions(defined in the usual fork basis ) : @xmath14 where @xmath15 here @xmath16 label different physical sites , @xmath17 label different links and @xmath18 means the link @xmath5 connects to the site @xmath8(see in fig .
[ grassmanntg ] , any link @xmath5 uniquely belongs to one physical site @xmath8 ) . on each link @xmath5 ,
@xmath19 label the bosonic inner indices , @xmath20 label the fermionic inner indices and @xmath21 label different species of grassmann variables .
@xmath22 is the physical index . @xmath23 and @xmath24 are grassmann numbers and dual grassmann numbers that satisfy the standard grassmann algebra : @xmath25 note that @xmath26 and @xmath27 have opposite orders : @xmath28 the symbol @xmath29 represents a projection of the result of the integral to the term containing no grassmann variables @xmath23 . for fermion(electron )
systems , the physical indices @xmath22 on a local hilbert space are always associated with a definite fermion parity @xmath30 .
hence , we can impose the following constraints to issue that eq .
( [ gtps ] ) do represent fermion wavefunctions .
although the original form eq.([gtps ] ) provides us a good physical insight of the state , especially for strongly correlated systems from projective constructions , it is not an efficient representation for numerical simulations . in the following
we will derive the standard form of gtps to simplify the representation . by using the grassmann version of singular - value - decomposition(gsvd )
method proposed in ref.@xcite , under the constraint @xmath32 , we can decompose @xmath33 into : @xmath34 with @xmath35\th_i^{n_i}\nonumber\\ s_{j;a_jq_j}&=\sum_{\{l_j^{\al_j}\}n_j}{\rm{s}}_{j;a_jq_j}^{\{l_j^{\al_j}\}n_j } \left[\prod_{\al_j } ( \th^{\al_j}_{j})^{l^{\al_j}_j}\right]\th_j^{n_j}\end{aligned}\ ] ] and @xmath36 where @xmath37 and @xmath38 are determined by the singular - value - decomposition(svd ) for the matrix @xmath39 with @xmath40 .
( notice the the constraint @xmath32 implies a @xmath41 symmetry for the matrix @xmath42 and it can be block diagonalized , with each sector labeled as @xmath43 or @xmath44 . ) again , the symbol @xmath45 represents a projection of the result of the integral to the term containing no grassmann number @xmath46 .
we call @xmath47 the standard metric for gtps , which is the grassmann generalization of the canonical delta function @xmath48 . a graphic representation of the decomposition eq .
( [ decompose ] ) .
we use double lines(red and blue ) to represent the standard metric @xmath49 , which is the @xmath41 graded version of the canonical delta function .
the blue line represents the channel with no inner fermion(@xmath43 ) and the red line represents the channel with one inner fermion(@xmath50 ) for the standard metric .
the arrow of the red line represents the ordering for the dual grassmann variables @xmath51 and @xmath52 .
we notice the standard metric @xmath49 only has one specie of grassmann variable despite the original grassmann metric @xmath53 contains many species of grassmann variables on its link @xmath5 and @xmath6(labeled by @xmath54 and @xmath55 ) . here
@xmath56 and @xmath57 are bosonic indices . ] put eq.([decompose ] ) into eq.([gtps ] ) , we have : @xmath58 the grassmann matrices @xmath59 defined on all links contain even number of grassmann numbers and they commute with each other .
such a property allows us to regroup them as : @xmath60 we use the fact that each link @xmath5 uniquely belongs to a site @xmath8 to derive the above expression .
we note @xmath61 defined here have opposite orders according to @xmath62 defined in eq .
( [ tg ] ) thus , we can integral out all the grassmann numbers @xmath63 and sum over all the bosonic indices @xmath64 to derive a simplified wave function : @xmath65 where the new grassmann tensor @xmath66 associate with physical site @xmath8 can be expressed as : @xmath67 with @xmath68 a graphic representation for the standard grassmann tensor @xmath69 , which is a combination of the grassmann tensor @xmath70 and those @xmath71(green solid circles ) surrounding it . ]
we call eq.([simplegtps ] ) the standard form of gtps which only contains one specie of grassmann variable on each link
. we can further simplify the expression by group the bosonic index @xmath72 and fermionic index @xmath73 into one super index @xmath74 .
@xmath75 with @xmath76 where @xmath77 .
notice the super index @xmath78 has a definite fermion parity @xmath79 and the corresponding fermion number is determined as @xmath80 . a graphic representation for the standard form of gtps . the filled circle on the physical site @xmath8 represents the standard grassmann tensors @xmath81 .
the link index @xmath78 has a definite fermion parity @xmath82 and we use blue(red ) line to represent the fermion parity even(odd ) @xmath83 indices . if @xmath84 , we associate a grassmann number @xmath46 with the standard grassmann tensor @xmath81 while associate its dual @xmath85 with the standard metric @xmath86 . since the standard metric @xmath86 is actually just a grassmann generalization of the canonical delta function @xmath87 ,
we only need to use an arrow to specify the ordering of the two grassmann variables @xmath88 and @xmath85 . ] the new form eq.([ngtps ] ) is extremely useful in general purpose of numerical calculations .
we will use this new form to explain all the details of our algorithms .
in this section , we will give a brief review of imaginary time evolution algorithm of usual tps first and then generalize all those algorithms into gtps . finally , we apply the algorithms to a simple free fermion example on honeycomb lattice .
let us consider the imaginary time evolution for generic tps @xmath89 .
@xmath90 if we do nt make any approximation , the true ground state can be achieved in the @xmath91 limit .
@xmath92 where @xmath93 is the number of evolution steps and @xmath94 is a sufficiently thin imaginary - time slice .
however , without any approximation , the inner dimension of tps will increase exponentially with the number of evolution steps .
hence , we need to find out the best tps approximation with fixed inner dimension .
wlog , we use the honeycomb lattice geometry to explain the details here and for the rest parts of the paper . to illustrate the key idea of the algorithm ,
let us consider a simple case that the model hamiltonian @xmath95 only contains a summation of nearest neighbor two - body terms : @xmath96 let us divide the hamiltonian into three parts : @xmath97 where @xmath98 label the sublattices @xmath98 and @xmath99 label three different nearest neighbor directions . by applying the trotter expansion
, we have : @xmath100 notice each @xmath101 only contains summation of commuting terms , hence we can decompose them without errors : @xmath102 let us expand @xmath89 under the physical basis : @xmath103 here @xmath104 denote two different sublattice in a unit cell and @xmath105 denote the physical indices on site @xmath106 , e.g. , @xmath107 for a spin @xmath108 system .
the canonical delta function @xmath109 defined on link @xmath110 can be regarded as the metric associated with tensor contraction , which can be generalized to its grassmann variable version for fermion systems . after acting one evolution operator @xmath111 onto the corresponding link @xmath110 , we can expand the new state @xmath112 as : @xmath113 is the matrix element of the evolution operator on link @xmath110 . use the svd decomposition
, we can decompose the rank @xmath114 tensor @xmath115 as : @xmath116 here the indices @xmath117 have dimension @xmath118 , where @xmath0 is the inner dimension and @xmath119 is the physical dimension of the tensor @xmath120 . after applying @xmath121 on state @xmath89 ,
the tensor @xmath70 will be replaced by @xmath122 : @xmath123 notice all @xmath124 have enlarged inner dimension @xmath118 instead of @xmath0 for their inner indices @xmath125(the bond along @xmath126 direction ) .
similarly , the dimension of indices @xmath127(the bonds along @xmath128 directions ) will also be enlarged to @xmath118 after applying @xmath129 and @xmath130 on @xmath89 .
thus , it is easy to see that the inner dimension will increase exponentially with increasing evolution steps if we do nt make any truncation . to solve the above difficulty
, we need to find a new set of @xmath131 with fixed inner dimension @xmath0 that minimizes the distance with @xmath132 .
if we start from @xmath89 and evolve it by @xmath133 in a sufficient thin time slice , the cost function @xmath134 takes the form : @xmath135 where @xmath136 @xmath134 is a multi - variable quadratic function of @xmath137 , hence we can use the sweep method to minimize it .
the advantage of the above algorithm is that the trotter error will not accumulate after long time evolution .
however , calculating the cost function @xmath134 explicitly is an exponentially hard problem and we need further approximations at this stage .
some possible methods have been proposed based on the mps algorithm@xcite , but the calculational cost can still be very large and the method has only been implemented with open boundary condition(obc ) so far . nevertheless , for translational invariant tps ansatz , it is possible to develop efficient method to simulate the cost function by using the terg method .
we assume @xmath138 if @xmath139 and @xmath140 if @xmath141 .
the cost function can be expressed as : @xmath142 where @xmath143 and @xmath144(fig .
[ environmenttensor ] ) are the so called environment tensors . here
we use the convention that all the repeated indices will be summed over and use @xmath145 to represent complex conjugate of @xmath146 . strictly speaking , the environment tensors for @xmath143 and @xmath144
are also dependent on @xmath147 and @xmath148 , thus @xmath134 is no longer a quadratic multi - variable function .
however , for sufficiently thin time slice , up to @xmath149 error(same order as trotter error ) , we can replace @xmath150 by @xmath151 when calculating the environment tensor @xmath143 .
@xmath144 can be derived from @xmath143 : @xmath152 again , repeated indices need to be summed over here .
we notice @xmath143 can be expressed as a tensor trace of double tensors @xmath153 with an impurity tensor @xmath154 sites on the link @xmath110 : @xmath155,\end{aligned}\ ] ] where the impurity double tensor @xmath154 is just a projector : @xmath156 now it is easy to see we can first decompose the impurity tensor on the link @xmath157 to two rank @xmath158 impurity tensors on site @xmath157 and then implement the usual terg algorithm .
we can also use a more efficient but complicated way to compute @xmath143 by applying the coarse grain procedures for all sites except sites @xmath157 , as introduced in ref.@xcite .
( ( a ) ) and @xmath144((b ) ) on honeycomb lattice . [ environmenttensor ] ] on honeycomb lattice . [ environment ] ] to the @xmath110 link for tps(with inner dimension @xmath0 )
, we get new tps with enlarged inner dimension(@xmath118 , @xmath119 is the physical dimension ) for the corresponding link .
( b ) for large classes of translational invariant tps , we can reduce the inner dimension from @xmath118 back to @xmath0 with respect to some environment wight @xmath159 .
[ evolution ] ] the above algorithm can be further simplified if we assume the environment tensor @xmath143 has specific forms for certain physical systems .
one interesting attempting is proposed in ref.@xcite by assuming @xmath143 can be factorized as : @xmath160 where @xmath161 is a positive weight vector defined on links along @xmath162 direction .
the above form can always to be true for 1d systems@xcite due to the existence of canonical form of a mps .
although the above form is not generic enough in 2d , it still works well in many cases , especially for those systems with symmetry breaking order . in this case , the cost function eq.([costf ] ) can be solved by svd decomposition and keep the leading @xmath0 singular values for the following matrix : @xmath163 the new tensors @xmath164 and @xmath165 can be determined as : @xmath166 similar as in 1d , the environment weight associated with bonds along @xmath126 direction is updated as @xmath167 .
it is not hard to understand why the above simplified algorithm works very well for systems with symmetry breaking orders , but not in the critical region , since in those cases the ground states are close to product states and the entanglements between @xmath99 directions in the environment tensors become pretty weak .
however , for topological ordered states , the environment tensors can be more complicated .
for example , the @xmath41 topological ordered state in the toric code model will have an emergent @xmath41 symmetry and could not be factorized as a product . hence , for topological ordered states , it is important to calculate the full environment in the imaginary time evolution . in this subsection , we will show how to generalize the above algorithms to gtps .
the key step is to introduce fermion coherent state representation and treat the physical indices also as a grassmann variable .
wlog , we use a spinless fermion system as a simple example here and for the rest parts of the paper : @xmath168 is a grassmann number . as already having been discussed in ref.@xcite , in this new basis , the gtps wavefunction eq.([gtps ] ) and its standard form eq.([ngtps ] )
can be represented as : @xmath169 where @xmath170 is the complex conjugate of the grassmann number @xmath171 and @xmath172 . on honeycomb lattice with translational invariant , we can simplify the above expression as : @xmath173 with @xmath174 comparing with usual tps , here the link indices @xmath175 always have definite fermion parity @xmath176 .
@xmath177 when @xmath178 while @xmath179 when @xmath180 .
let us start from the simplest case with the assumption the environment can also be approximated by some weight . in this case , the imaginary time evolution for gtps can be reduced to a svd problem of grassmann variables . the grassmann version of the matrix @xmath42 in eq.([m ] )
can be constructed as following : \(a ) let us first sum over the bond indices @xmath117 and integral out the grassmann variable @xmath181 : @xmath182n_f(a)}(-)^{m_j\left[m_i+n_f(b)+n_f(c)\right ] } { \rm{\t
t}}^{m_{i}}_{a;abc}{\rm{\t t}}^{m_{j}}_{b;a b^\prime c^\prime } \bar \eta_j^{m_j}\bar \eta_i^{m_i } \theta_\beta^{n_f(b ) } \theta_\gamma^{n_f(c ) } \theta_{\beta^\prime}^{n_f(b^\prime ) } \theta_{\gamma^\prime}^{n_f(c^\prime)}.\label{gintegral}\end{aligned}\ ] ] here the sign factor @xmath183n_f(a)}(-)^{m_j\left[m_i+n_f(b)+n_f(c)\right]}$ ] comes from the anti - commutating relations of grassmann variables .
\(b ) next we derive the matrix element of the evolution operator under fermion coherent state representation .
let us first calculate them under the usual fock basis @xmath184 with @xmath185 .
let us define : @xmath186 thus , we can expand @xmath187 as : @xmath188 in the fermion coherent state basis , we have : @xmath189 \(c ) then we can evolve the state @xmath190 to a new state @xmath191 : @xmath192 put eq.([gintegral ] ) and eq.([corepresentation ] ) into the above equation , it is easy to derive the grassmann version of the rank @xmath114 tensor @xmath193 defined on the link @xmath110 .
we have : @xmath194n_f(a)}(-)^{m_j^\prime\left[m_i^\prime+n_f(b)+n_f(c)\right ] } e_{m_i^\prime m_j^\prime}^{m_i m_j } { \rm{\t t}}^{m_{i}^\prime}_{a;abc}{\rm{\t t}}^{m_{j}^\prime}_{b;a b^\prime c^\prime } { ( \bar\eta_j^\prime)}^{m_j } { ( \bar \eta_i^\prime ) } ^{m_i } \theta_\beta^{n_f(b ) } \theta_\gamma^{n_f(c ) } \theta_{\beta^\prime}^{n_f(b^\prime ) } \theta_{\gamma^\prime}^{n_f(c^\prime)}\nonumber\\ & = \sum_{m_i^\prime m_j^\prime a } ( -)^{\left[m_i^\prime+m_j^\prime\right]n_f(a ) } ( -)^{m_j^\prime\left[m_i^\prime+n_f(b)+n_f(c)\right]}(-)^{m_j\left[m_i+n_f(b)+n_f(c)\right ] } \nonumber\\ & \times e_{m_i^\prime m_j^\prime}^{m_i m_j } { \rm{\t t}}^{m_{i}^\prime}_{a;abc}{\rm{\t t}}^{m_{j}^\prime}_{b;a
b^\prime c^\prime } { ( \bar \eta_i^\prime ) } ^{m_i } \theta_\beta^{n_f(b ) } \theta_\gamma^{n_f(c ) } { ( \bar\eta_j^\prime)}^{m_j } \theta_{\beta^\prime}^{n_f(b^\prime ) } \theta_{\gamma^\prime}^{n_f(c^\prime)}.\end{aligned}\ ] ] we notice there is an extra sign factor @xmath195}$ ] after we reorder these grassmann variables .
\(d ) finally , we can define the grassmann generalization of the @xmath42 matrix after putting the environment weight for all the inner indices .
@xmath196 where the coefficient matrix @xmath197 reads : @xmath198n_f(a ) } ( -)^{m_j^\prime\left[m_i^\prime+n_f(b)+n_f(c)\right ] } \nonumber\\ & \times ( -)^{m_j\left[m_i+n_f(b)+n_f(c)\right ] } e_{m_i^\prime m_j^\prime}^{m_i m_j } { \rm{\t t}}^{m_{i}^\prime}_{a;abc}{\rm{\t
t}}^{m_{j}^\prime}_{b;a b^\prime c^\prime}\end{aligned}\ ] ] since @xmath199 is a local fermionic operator , it will always contain even number of fermion operators . as a result
, the nonzero elements of @xmath200 matrix will always contain even number of grassmann variables and we can apply gsvd as discussed before .
we keep the largest @xmath0 eigenvalues : @xmath201 the coefficient matrix @xmath42 will have a block diagonalize structure , hence the new index @xmath125 will have a definite fermion parity @xmath202 . and @xmath203 . in step ( b ) , we express the two sites evolution operator @xmath204 under fermion coherent basis . in step ( c ) and ( d ) , we apply the evolution operator @xmath204 to the two sites grassmann tensor @xmath205 and derive a new two sites grassmann tensor @xmath206 .
then we approximate it by two new adjoint grassmann tensors @xmath207 and @xmath208 with respect to the environment weights @xmath159 .
we also use double lines to label different fermion parities for the physical indices . for the simple spinless fermion example
, red line represents one fermion state and blue line represents non fermion state .
the metric @xmath209 is the standard metric for the physical indices .
[ evolutionf ] ] similar as in usual tps case , the new grassmann tensors @xmath210 and @xmath211 have the form @xmath212 however , due to the reordering of grassmann variables , extra signs appear in the new tensors @xmath213 and @xmath214 .
@xmath215 again @xmath216 is used as the new environment weight for all links along @xmath126 direction .
the full environment tensors can be very similar as in usual tps case .
the grassmann version of the environment tensor @xmath143 can be efficiently simulated by gtps .
the environment tensor @xmath144 can be calculated from @xmath143 and @xmath217 .
the cost function can be derived from eq.([costf ] ) after replacing @xmath218 and @xmath146 to their grassmann version .
after we contract the tensor - net and integral out all the grassmann variables , it can be reduced to a usual multi - variable quadratic minimization problem for the coefficient tensors @xmath219 and @xmath220 , we can solve it by using the sweep method .
although the algorithm with full environment is general , it is still very time consuming and a much more efficient method is very desired .
nevertheless , it turns out that the simple updated method works very well in many cases and we will focus on the application of this method in this paper . in this subsection
, we demonstrate the above algorithm by studying a free fermion model on honeycomb lattice .
we consider the following ( spinless)fermion hamiltonian : @xmath221 we first test our algorithm in the parameter region where the system opens a gap . for example , we can fix @xmath222 and take different values for @xmath223 from @xmath224 to @xmath225 . see in fig .
( [ freegapless ] ) , even with the minimal inner dimension @xmath226 , the gtps variational approach already gives out very good ground state energy .
here and through out the whole paper , we fix the total system size to be @xmath227 sites and with pbc.(the gteg algorithm will allow us to reach huge size in principle , however , for better convergency , we choose relatively large but not huge size . )
we note the agreement for the ground state energy is better for small @xmath223 , which is expected since the system becomes a trivial vacuum state @xmath228 in the limit @xmath229 . in the insert of fig .
[ freegapless ] , we also plot the ground energy as a function of @xmath230 and it is shown the energy converges for very small @xmath230(around @xmath231 , where @xmath230 is the number of singular value we keep in the gterg algorithm ) .
we find the relative error can be very small , e.g. , @xmath232 for @xmath233 . for fixed @xmath222 .
as a bench mark , we also plot the exact energy of the free fermion hamiltonian .
insert : relative error of ground state energy as a function of @xmath230 for @xmath233 .
@xmath230 is the number of singular values kept in the grassmann svd decomposition@xcite.[freegap ] ] although we get good results for the above simple example , it is not quite surprising since a trivial gapped fermion system only evolves local physics .
a much more challenging and interesting example is at @xmath234 , in which case the low energy physics of the system is described by two dirac cones in the first brillouin zone(bz ) .
actually , up to a particle - hole transformation on one sublattice , the above fermion paring hamiltonian is equivalent to the fermion hopping hamiltonian that describes the novel physics in graphene(the spin less version ) where low energy electrons deserve a dirac like dispersion . in fig .
[ freegapless ] , we plot the ground state energy as a function of @xmath230 for inner dimensions @xmath226 and @xmath235 . in this case
, the @xmath226 approach only gives very poor results comparing to the gapped case .
however , when we increase the inner dimension to @xmath235 , we find that the ground state energy agrees pretty well with the exact one , with a relative error of @xmath236 .
another interesting feature is that such a critical system would require a much larger @xmath230(around @xmath237 for @xmath235 . ) for the convergence of the gterg algorithm .
later , we would discuss a possible improvement for the gterg algorithm , which allows us to access much larger inner dimension @xmath0 .
finally , we would like to make some comments and discussions about the above results : \(a ) although a gapless fermion system with dirac cones is critical , it does not violate the area law because it only contains zero dimensional fermi surfaces .
\(b ) the good agreement in ground state energy does not imply the good agreement in long range correlations .
indeed , our conjecture is that for any finite @xmath0 , the variational wave function we derive may be always associated with a finite correlation length which scales polynomially in @xmath0 .
we note that an interesting critical fpeps state with finite inner dimension @xmath226 was proposed in ref.@xcite , however , that model is different from our case since the dirac cones in that model contain a quadratic dispersion along some directions in the first bz .
\(c ) although the gtps variational approach can not describe the above system with finite inner dimension @xmath0 , the variational results will still be very useful . from the numerical side
, we can apply both finite @xmath0 and finite size scalings to estimate the physical quantities in the infinite @xmath0 and infinite size limit . from the analytic side
, the finite correlation length corresponds to a finite gap in the system , which is known as dirac mass term in the effective theory@xcite . in quantum field theory ,
a controlled calculation(without singularities in calculating correlation functions ) can be performed by first taking the dirac mass term to be finite and then pushing it to the zero limit . for fixed @xmath234 . as a bench mark
, we also plot the exact energy of the corresponding free fermion hamiltonian .
[ freegapless ] ]
in this section , we will discuss possible improvements for the terg algorithm and its grassmann generalization .
again , let us start from usual tps case
. its grassmann generalization would be straightforward by replacing the complex number valued tensors with those grassmann valued tensors .
we notice the simple svd method used in terg algorithm actually implies the following cost function : @xmath238 where the rank four double tensor @xmath239 is defined as @xmath240 and @xmath241 means summing over indices for the connected link , as see in fig.[honeycombrg ] ( a).(actually it is the inner product of two vectors if we interpolate @xmath242 as @xmath243 dimensional vectors , where @xmath0 is the inner dimension of the tps . )
@xmath244 is the usual @xmath245-norm of a vector .
( the rank @xmath246 tensor @xmath239 and rank @xmath158 tensors @xmath247 can be viewed as @xmath248 and @xmath243 dimensional vectors . )
such a cost function only minimizes the @xmath245-norm of local error @xmath249 for a given cut - off dimension @xmath230(the dimensional of the link shared by @xmath250 and @xmath251 ) and could not be the optimal one for minimizing global error . to figure out the optimal cost function ,
let us divide the system into large patches , as seen in fig.[tensortrace ] .
if we trace out all the internal indices inside the patch , we can derive a rank @xmath252(@xmath252 is the number of sites on the boundary of the patch ) tensor @xmath253 for the patch and the norm of tps can be represented as the tensor trace of the new double tensors @xmath253 . when we perform the terg algorithm and replace @xmath239 by @xmath254 , we aim at making a small error for the double tensor @xmath253 . and @xmath255 .
the symbol @xmath241 here means sum over the indices for the shared link for two connected tensors .
( b)a schematic plot for the terg scheme on honeycomb lattice .
the first step is approximate while the second step is exact.[honeycombrg ] ] now it is clear why the simple svd method that minimizes local errors could not approximate the patch double tensor @xmath253 in an optimal way .
let us rewrite @xmath253 as @xmath256.(notice here we regard the rank @xmath252 double tensor @xmath253 as a @xmath257 dimensional vector , the rank @xmath246 double tensor @xmath239 as a @xmath248 dimensional vector and the rank @xmath258 double tensor @xmath259 as a @xmath257 by @xmath248 matrix . )
the double tensor @xmath259 is called as environment tensor .
the cost function which provides the best approximation for @xmath253 can be represented as : @xmath260 again , @xmath241 means the vector inner product and @xmath244 means the usual @xmath245-norm . comparing with the simple svd method , the environment tensor gives a complex weight for each component of the local error @xmath261 .
notice the cost function proposed here is very different from the one in ref.@xcite , which aims at minimizing @xmath262 . .
the global cost function aims at finding the best way to minimize the error for the above patch double tensor.[tensortrace ] ] ] although the environment tensor @xmath259 is conceptually useful , it is impossible to compute this rank @xmath258 tensor when the patch size becomes very large .
nevertheless , we can derive a simplified environment through the simplified time evolution algorithm .
the key step is also based on the conjecture that the rank @xmath263 tensor @xmath264 can be factorized into a product form : @xmath265 where @xmath266,@xmath267 are the double indices . for any converged @xmath268 and @xmath269 from the simplified time evolution algorithm with converged weight vectors @xmath270 , the weight vectors @xmath271 for double tensors @xmath239 can be determined as : @xmath272 the above conjecture for the environment is reasonable when the environment of @xmath273 can be approximated as a product form since @xmath274 similar as the simplified time evolution algorithm , solving the optimal cost function in this case can be implemented by the svd decomposition for matrix @xmath275 : @xmath276 then the rank @xmath158 double tensors @xmath250 and @xmath251 can be determined as : @xmath277 finally , the weight vector @xmath278 for @xmath279 can be updated as @xmath280 . in general cases , the assumption eq.([environment ] ) could not be true , however , as long as @xmath281 has a much more uniform distribution(up to proper normalization ) : @xmath282 the above weighted terg(wterg ) algorithm can still improve the accuracy for fixed @xmath230 with the same cost .
this is because the svd method is the best truncation method if the environment has a random but uniform distribution . by replacing the
complex valued double tensors with grassmann variable valued double tensors , all the above discussions will be valid for gtps .
however , the definition of the inner product @xmath241 and the corresponding 2-norm @xmath244 should also be generalized into their grassmann version , which evolves the integration over grassmann variables for the connected links with respect to the standard grasamann metric .
for example , the cost function of the gsvd method discussed in ref.@xcite can be formally written as : @xmath283 only contains one species of grassmann variables .
we can express @xmath284 as : @xmath285 where @xmath286 is the complex coefficient of the grassmann number valued double tensor @xmath287 and @xmath288 is determined by the fermion parity of the inner index @xmath289 .
we notice @xmath290 if we express the grassmann number valued double tensors @xmath291 on sublattice @xmath292 as : @xmath293 similarly , we can express @xmath294 and @xmath295 as : @xmath296 again , @xmath297 and @xmath298 are the complex valued coefficients .
recall the definition of the standard grassmann metric @xmath299 : @xmath300 the inner product @xmath301 explicitly means : @xmath302 thus , we have : @xmath303 with @xmath304 now it is clear that up to a sign twist , the gsvd cost function is equivalent to the cost function of its complex coefficient tensors : @xmath305 explicitly the same as in the terg case , the best approximation for a given @xmath230 is nothing but the svd decomposition for the coefficient tensor @xmath253.(if we view @xmath306 as a matrix @xmath307 . ) on the other hand , as already having been discussed in ref.@xcite , the constraint eq .
( [ constraint ] ) of gtps implies their double tensors containing even number of grassmann number . as a result , @xmath275 is block diagonalized and the index @xmath308 will have a definite parity @xmath309 .
we notice the novel sign factor @xmath310 here arises from the anti - commuting nature of the grassmann variables and will encode the fermion statistics .
the second rg step remains the same as in gterg and a similar novel sign factor will also emerge there
. all the above discussion will still be correct if the inner index of the double tensor @xmath284 contain multiple species of grassmann variables .
indeed , starting from the standard gtps , the double tensor @xmath284 in the first rg step will contain two species of grassmann variables .
this is because @xmath311 and @xmath312 . in the second and later rg steps
, @xmath313 will only contain one specie of grassman variables .
the discussion for the environment effect will be in a similar fashion .
especially , if we use some weighting factor to approximate the environment effect , the coefficient tensors @xmath247 will take the same form as in usual tps case .
however , an important sign factor should be included when we define the matrix @xmath275 .
again , the environment weight for the first step is determined by the time evolution algorithm of gtps .
same as in the bosonic case , the singular value obtained from the gsvd would perform as the environment weight for next rg step . in fig .
[ gaplessweight ] , we implement the above algorithm to the free fermion hamiltonian eq.([freefermion ] ) at critical point(@xmath234 ) .
we see an important improvement that the ground state energy decreases when increasing @xmath230 and is strictly above the exact energy , unlike the simple gterg approach , which can overestimate the ground state energy for small @xmath230 .
however , for larger enough @xmath230 , the two approaches converge to the same values , as expected .
we further use the new algorithm to study the critical model with larger inner dimension @xmath0 . up to @xmath314 , we find that the ground state energy from the gtps approach is almost the same as the exact one(relative error @xmath315 ) . at @xmath234 . by introducing the environment weight in the gterg method ,
an important improvement is that the ground state energy is always above the exact value , which is very important for variational approach ] .[gaplessweight ]
in this paper , we first derive a standard form of gtps that only contains one specie of grassmann variable for each inner index and significantly simplifies the representations in our numerical calculations .
based on the fermion coherent state representation , we further generalize the imaginary time evolution algorithms into fermion systems .
we study a simple free fermion example on honeycomb lattice , including both off - critical and critical cases to test our new algorithms .
finally , we discuss the importance of the environment effect of the terg / gterg method and present a simple improvement by introducing proper environment weights .
although the simple time evolution algorithm discussed here is not generic enough , it has already allowed us to study many interesting and important models , such as hubbard/@xmath316 model , whose ground state is believed to be a superconductor .
the evidence for the existence of superconductivity in these models based on the gtps algorithm will be discussed and bench marked with other methods elsewhere@xcite .
of course , the generic algorithm is also very important and desired , especially for those systems with topological order . actually , the general discussion in section ii have already made some progress along this direction , but not efficient and stable enough at this stage . on the other hand , further improving the efficiency of contacting ( grassmanna )
tensor net is also very important .
although the gterg / terg algorithm provides us promising results in many cases , it is still not efficient enough since the algorithm is not easy to be parallelized .
recently , a novel idea of combination the concept of renormalization and monte carlo(mc)@xcite has made great success for boson / spin systems , it would be very natural to generalize it into fermion / electron systems based on the grassmann variable representations . | recently , the grassmann - tensor - entanglement renormalization group(gterg ) approach was proposed as a generic variational approach to study strongly correlated boson / fermion systems@xcite .
however , the weakness of such a simple variational approach is that generic grassmann tensor product states(gtps ) with large inner dimension @xmath0 will contain a large number of variational parameters and be hard to be determined through usual minimization procedures . in this paper
, we first introduce a standard form of gtps which significantly simplifies the representations .
then we describe a simple imaginary - time - evolution algorithm to efficiently update the gtps based on the fermion coherent state representation and show all the algorithm developed for usual tensor product states(tps ) can be implemented for gtps in a similar way . finally , we study the environment effect for the gterg approach and propose a simple method to further improve its accuracy .
we demonstrate our algorithms by studying a simple 2d free fermion system on honeycomb lattice , including both off - critical and critical cases . |
immunotherapy seeks to arm the patient with an increased number of effective immune cells that are specifically trained to seek out and destroy harmful cells .
one target for immunotherapy that has earned a place in the clinical spotlight recently is melanoma , since specific peptides have been isolated from melanoma cells that can be used in cancer vaccines .
a subset of immune cells , called dendritic cells , are able to stimulate the production of immune effector cells that are capable of recognizing and killing specific tumor cells .
researchers have been able to extract these cells from patients and culture them _ ex vivo _ to create a vaccine that can boost a patient s response against their own cancerous cells , @xcite .
the success of clinical trials of dendritic cell ( dc ) vaccines has resulted in the recent fda approval of the first cancer vaccine for prostate cancer , @xcite despite promising clinical responses in vaccine trials , it remains difficult to predict which patients will actually respond to these vaccines and why , .
this may be due to the complicated kinetics of the immune response to the presentation of antigen by the dendritic cells . in this paper
, we analyze the effect of the delay between the time the dcs come into contact with the t - cell population and the initiation of the expansion , or proliferating phase .
we focus on the dynamics describing the active effector t - cell population in the spleen , using a phenomenological description that captures the effect of the delay .
our goal here is to isolate the role of this delay term in order to analyze its effect on the sustainability of t - cell production .
the model we present is very simple , and is not intended to realistically capture the entire cascade of immune events in the response to antigen presentation by dendritic cells . nor does this model describe the trafficking of immune cells between compartments of the body , in particular between lymphoid organs , such as the spleen , and the tumor site .
these effects are crucial to the understanding of the immune response and must be considered in the design of vaccine therapies .
one group of experiments that looks at the trafficking of dcs between compartments is described in @xcite , and a mathematical model of dc trafficking and t - cell activation is presented in @xcite .
elsewhere we combine these ideas , along with previous work on cell - lysis rates , @xcite , in a more realistic model that includes several types of immune cells , and trafficking between compartments . in section [ modeldescription ]
we describe the model itself , and then we show stability switching as a function of the delay in section [ stability ] .
we conclude with some discussion of the model simulations and results in section [ conclusion ] .
after encountering antigen , such as tumor cells or tumor peptides , dendritic cells migrate to lymph organs such as the spleen . as they migrate , they mature and , upon arrival in the spleen , they are able to activate the proliferation of t - cells that have the ability to seek out and destroy ( _ lyse _ ) the specific target cells that provided the antigen .
this activation requires some contact time between the dendritic cells and the naive t - cells ; this connection time is called the
synaptic connection time . " in our model , we represent this time as a delay , @xmath0 .
once proliferation is initiated , new antigen - specific t - cells are created .
a somewhat simplified description of the fate of these new cells is as follows : the cells must go through a check - point during which the new cells are tested for efficacy against the antigen .
if they pass this test , they either move into the blood stream to migrate to the target cells , they become memory cells in order to protect against future challenges by the same antigen , or they return to the proliferating compartment to produce more specific t - cells .
proliferation continues only while certain cytokines are present .
these cytokines are produced by previously activated t - cells that have been in contact with dendritic cells for the synaptic connection time , @xmath0 .
when the dendritic cells have done their job educating " the t - cells , it is believed that they die , @xcite , and are then cleared " from the spleen compartment .
when the dcs are no longer present , cell proliferation ceases , and the activated t - cells either move into the blood stream or become memory cells . this process is represented graphically in figure [ fig : modelprocess ] . in this model ,
the concentration of the dendritic cells at a given time , @xmath1 , is given by @xmath2 , while the number of activated t - cells is given by @xmath3 .
the activation / proliferation rate is then given by @xmath4 where @xmath5 is represents the feedback " function .
feedback " function depends on the number of t - cells present in the proliferating compartment time @xmath0 earlier in the presence of the dcs ; therefore they are producing the necessary cytokines .
the current number of proliferating t - cells is represented by @xmath3 .
we assume that the feedback " function , @xmath5 , is a positive function that increases to some maximum level , @xmath6 , and then decreases .
the justification for this functional form is that activation by dcs is maximized at a certain concentration of dcs in the spleen .
the addition of more dcs has the effect of keeping the activated cells in the area .
if they are present for too long , the effectiveness of the t - cells begins to diminish .
this is sometimes called the decreasing potential hypothesis " in the literature , @xcite . one simple function that captures these dynamics
is given by : .3 in -.5 in .2 in where @xmath7 represents the population of active t - cells from time @xmath0 ago .
we assume that t - cells leave the proliferating compartment through death , moving into the blood stream or the memory compartment at an average per - cell rate , @xmath8 .
putting these pieces together results in the following single delay equation : @xmath9 where for simplicity : @xmath10 by rescaling the time variable , we can assume that @xmath11 , ( the decay parameter , @xmath12 , must also be appropriately scaled ) .
simulated solutions are shown in figures [ fig : sampletrajectories_tauzero ] and [ fig : switchingtimes ] .
these solutions show that changing the value of @xmath0 , as well as the initial history ( the values of @xmath3 for @xmath13 ) , can have a strong effect on the long - term behavior of the system .
we verify this in the next section by performing a stability analysis .
when dendritic cells are present in the spleen , @xmath14 , so that the differential equation becomes : @xmath15 equation [ eqn : scaleddde ] potentially has three equilibria .
one is the zero fixed point : @xmath17 , and the other two are solutions to @xmath18 these two solutions are depicted graphically in figure [ fig : twoequil ] .
note that the function @xmath19 has a maximum which occurs at @xmath20 .
when the maximum of this function is greater than @xmath12 the two additional equilibria will exist .
this occurs for @xmath21 .
we denote the greater of these two equilibria by @xmath22 , and the smaller by @xmath23 , if they exist . if @xmath22 is stable , then proliferation is sustained , and active effector cells continue to be produced and sent into the blood stream and into the memory compartment .
however , if @xmath22 is unstable , then production may die off and no new effector cells will be produced , even when dcs are still present in the spleen .
our goal in this section is to show that changing the value of the delay , @xmath0 , can change the stability of @xmath22 .
linearization around an equilibrium , @xmath24 gives : @xmath25 where @xmath26 , @xmath27 , evaluated at the equilibrium @xmath24 , and @xmath28 .
the characteristic equation is : @xmath29 evaluating @xmath30 and @xmath31 at @xmath32 gives : @xmath33 @xmath34 when we consider the characteristic equation in the case of the equilibrium @xmath17 , we always obtain @xmath35 , _ independent of @xmath0 . _
thus we see that the equilibrium at @xmath36 is * always stable*. on the other hand , linearizing around the upper equilibrium at @xmath22 in the case when @xmath16 gives : @xmath37 using the fact that @xmath38 , we see that stability of the upper equilibrium is ensured if @xmath39 we saw earlier that the maximum of @xmath40 occurs at @xmath41 , and we know that @xmath42 . because we assume @xmath43
, we see that equation [ eqn : xstarcondition ] is satisfied .
therefore , when @xmath16 , @xmath22 is * stable*. since @xmath16 corresponds to the ode case , and @xmath23 lies between the two stable equilibria @xmath17 and @xmath44 , we know that @xmath23 is * unstable*. sample trajectories are shown in figure [ fig : sampletrajectories_tauzero ] .
changes in stability can occur when the real part of @xmath45 switches sign , i.e. when @xmath45 is purely imaginary . in this case
, the characteristic equation becomes @xmath46 at the upper equilibrium , @xmath22 , @xmath47 .
equating the real and imaginary parts in equation [ eqn : om ] in this case gives : @xmath48 the first equation implies that @xmath49 .
substituting this into the second equation gives : @xmath50 for a fixed value of @xmath12 , this gives a sequence of possible switching times , at which the equilibrium at @xmath22 can change stability .
for example , taking @xmath51 , we see that @xmath52 , and possible switching times occur when @xmath53 figure [ fig : switchingtimes ] shows numerical solutions for increasing values of @xmath0 .
we see that the equilibrium point at @xmath52 loses stability when @xmath54 , when a limit cycle appears .
increasing @xmath0 destroys this limit cycle and , for larger values of @xmath0 , solutions that once tended to @xmath22 converge to the zero equilibrium .
we have presented a simplified model of the interaction between dendritic cells and t - cells at the initiation of a specific immune response .
the model contains one delay that represents the synaptic connection time between dendritic cells and t - cells which is necessary for the development of cytokines that stimulate proliferation .
there is experimental evidence that prolonged contact between dendritic cells and t - cells results in a decrease in viable t - cell production .
the model presented here shows a change in stability of the high t - cell equilibrium as the delay is increased .
this equilibrium goes from being stable when the delay , @xmath0 , is small to being unstable for larger values of @xmath0 .
these results support the experimental evidence for the decreasing potential hypothesis " .
the model also suggest a role for dendritic cell therapies in boosting the immune defense against certain diseases such as cancer .
however , the model indicates that levels of dendritic cells in the spleen should not be maintained at high levels for too long in order to optimize the expansion of active effector t - cells .
we would like to thank the organizers of the 8th aims conference on dynamical systems , differential equations and applications that took place in dresden , germany , may 25 - 28 , 2010 , at which this work was presented .
n. koike , s. pilon - thomas , and j.j .
mule , _ nonmyeloblative chemotherapy followed by t - cell adoptive transfer and dendritic cell - based vaccination results in rejection of established melanoma
_ j. immunotherapy * 31 * 4 , may ( 2008 ) , 402412 . | the activation of a specific immune response takes place in the lymphoid organs such as the spleen .
we present here a simplified model of the proliferation of specific immune cells in the form of a single delay equation .
we show that the system can undergo switches in stability as the delay is increased , and we interpret these results in the context of sustaining an effective immune response to a dendritic cell vaccine .
angela gallegos ami radunskaya |
quantum magnetism is one of the oldest central problems in condensed matter and many body physics .
few exact results are known except at one dimension or when long range spin order is established at low temperature . in the later case , landau s symmetry breaking paradigm
is applicable , and the long range spin order determines the low energy physics of the system .
the low energy spin excitations are spin waves or magnons.@xcite systematic corrections to spin wave theory can be obtained through semiclassical approach ( @xmath10 expansion@xcite ) and/or nonlinear @xmath11 model.@xcite however , it was suggested that for some anti - ferromagnetic materials , strong quantum fluctuations due to small spin magnitude and low dimensionality combined with geometric frustration may lead to quantum coherent , spin disordered ground states.@xcite in such cases , a quantum fluid approach@xcite is probably a better starting point to describe the low energy physics instead of usual spin wave or semiclassical approach .
there exist at present two common quantum fluid approaches based on different particle " representations of spins . in the bosonic "
approach@xcite spins are represented by spin up and down schwinger bosons with the constraint that total boson number at each site is @xmath12 .
a spin - disordered state is obtained in a mean field theory as long as bose condensation does not occur .
the low lying excitations coming from this theory are bosonic .
the other possibility is to represent spins by fermions . up to now ,
the fermionic approach is mainly restricted to @xmath0 spin systems@xcite . because of the pauli exclusion principle , we can not put more than two spin up or down fermions on a site to form @xmath2 objects as in bosonic approach .
one way out is to introduce more species of fermions .
it was proposed in ref . , that @xmath13-species of boson / fermions can be used to construct the spin-@xmath5 swapping operators .
later , it is shown that the fermionic representation can be used to describe a @xmath3 valence bond solid state.@xcite in this paper , we systematically study the fermionic representation for arbitrary spins . to construct a spin operator with spin magnitude @xmath5 , we follow ref . and introduce different fermion operators for different @xmath14 states .
we find that for half - odd - integer spins , the spin operator thus constructed is invariant under a @xmath15 transformations,@xcite i.e. the symmetry group is @xmath15 group whereas for integer spins the symmetry group is @xmath16 .
fermionic path integral formulations for spin systems with arbitrary spin-@xmath5 can then be formulated and the corresponding mean field theories can be studied .
we shall see that one of the mean field theories applied to 1d antiferromagnetic heisenberg model produces mean field excitation spectrums which agrees with haldane s conjecture.@xcite moreover , there exist zero energy majorana edge modes for open integer spin chains .
we also apply our approach to two - dimensional ( 2d ) antiferromagnetic heisenberg models on triangular lattices where we obtain two translational and rotational invariant spin liquid solutions , one ( with @xmath6-wave paring symmetry ) is gapped and the other one ( with @xmath7-wave pairing symmetry ) has nodes at the spinon fermi level .
comparing with experiments on @xmath17 , we argue that the @xmath7-wave pairing spin liquid is a plausible ground state .
it should be noted that rather different fermionic / bosonic spinon approaches have also been applied to @xmath18 spin models where the dynamic symmetry is generalized from @xmath15 to @xmath18 , and the hamiltonian is composed of @xmath18 generators .. in these approaches , @xmath19-species of particles are introduced to construct the @xmath18-spin operator and hamiltonians .
the major difference between our approach and the @xmath18 approaches is that @xmath13 species of fermions are introduced to form a @xmath15-spin ( or irreducible representation of @xmath15 group ) in our approach . unlike the @xmath18 models ,
our approach is not limited to bipartite lattices .
the paper is organized as follows . in section
ii , we introduce the general fermionic spinon representation for arbitrary spin @xmath5 and discuss the symmetry properties associated with the representation . in section iii , we introduce the path integral formalism where different ways of imposing particle number constraints are presented .
a particle - hole symmetric way of imposing constraint is discussed . in section iv.a and iv.b
, we focus on 1d heisenberg models for spin-1 and spin-3/2 , respectively , and study two versions of the mean field theories .
the cases for general spin @xmath5 are discussed in section iv.c . in section v we study 2d afm models on triangular lattice . our paper is summarized and concluded in section vi .
we begin with the fermionic representation for spins . for spin @xmath0 ,
two species of fermions @xmath20 and @xmath21 are introduced to construct the three spin operators @xmath22 . to generalize this fermionic representation to arbitrary spin-@xmath5
, we introduce @xmath23 species of fermionic operators @xmath24 satisfying anti - commutation relations , @xmath25 where @xmath26 .
the spin operator can be expressed in terms of @xmath24 and @xmath27 s , @xmath28 where @xmath29 and @xmath30 @xmath31 is a @xmath32 matrix whose matrix elements are given by @xmath33 it is easy to see that the operators @xmath34 satisfy the @xmath15 angular momentum algebra , @xmath35=i\epsilon^{abc}\hat s^c$ ] . under a rotational operation
, @xmath36 is a spin-@xmath5 spinor " transforming as @xmath37 and @xmath38 is a vector transforming as @xmath39 , here @xmath40 is the @xmath23-dimensional irreducible representation of @xmath15 group generated by @xmath41 and @xmath42 is the adjoint representation . as in the @xmath0 case , a constraint that there is only one fermion per site is needed to project the fermionic system into the proper hilbert space representing spins , i.e. @xmath43 where @xmath44 is the site index and @xmath45 ( the particle representation , one fermion per site ) .
alternatively , it is straightforward to show that the constraint @xmath46 ( the hole representation , a single hole per site ) represents a spin equally . the @xmath45 representation can be mapped to the @xmath46 representation _ via _ a _ particle - hole transformation_. for @xmath0 , the particle picture and the hole picture are identical , reflecting an intrinsic particle - hole symmetry of the underlying hilbert space which is absent for @xmath47 .
following affleck _
et al . _ , @xcite we introduce another spinor " @xmath48 , whose components can be written as @xmath49 , where the index @xmath50 runs from @xmath5 to @xmath51 as in @xmath36 . to examine whether @xmath52 is really a spinor "
, we construct a spin singlet state for two spins at site @xmath44 and @xmath53 , @xmath54 .
therefore @xmath55 is a scalar operator as @xmath56 , meaning that @xmath52 and @xmath36 must behave identically under spin rotation .
consequently , the spin operators can also be written in terms of @xmath52 , @xmath57 combining @xmath36 and @xmath52 into a @xmath58 matrix @xmath59 , @xcite we can reexpress the spin operators as @xmath60 and the constraint can be expressed as @xmath61 where the @xmath62 sign implies @xmath45 and @xmath63 sign implies @xmath46 .
we are interested in two kinds of operations ( or groups ) acting on @xmath64 .
the first kind belongs to a @xmath13-dimensional irreducible representation of @xmath15 group , acting on the left of @xmath64 .
suppose @xmath65 is an element of this irreducible representation , then @xmath66 ( @xmath42 belongs to the adjoint irreducible representation of @xmath15 group , which is a @xmath67 matrix ) . under transformation @xmath68 ,
the spin operator @xmath69 becomes @xmath70 , which means _ a rotation of the spin_. it is obvious that the particle number constraint eq .
( [ contr2 ] ) remains unchanged under the action of @xmath65 .
the other kind belongs to a @xmath71 unitary group acting on the right of @xmath64 which keeps the spin operator eq .
( [ spin ] ) and the fermionic statistics eq .
( [ fermi ] ) invariant .
this group reflects the symmetry properties of the underlying hilbert space structure in the fermionic representation .
we call it an _ internal symmetry group_. the internal symmetry group is different for integer and half - integer spins . it is @xmath72 for the former and @xmath15 for the latter .
we leave the rigorous proof in appendix a , and shall explain qualitatively the reason behind here .
notice that @xmath36 and @xmath52 are not independent .
the operators in the internal symmetry group will mix " the two fermion operators in the same row of @xmath36 and @xmath73 , i.e. @xmath74 and @xmath75 , @xmath76 and @xmath77 , etc .
for integer spins , @xmath78 and @xmath79 will be mixed " . to keep the relation @xmath80 invariant ,
there are only two methods of mixing " : one is an @xmath81 transformation , the other is interchanging the two operators .
these operations form the @xmath16 group . for half - odd - integer spins ,
the pair @xmath82 do not exist , and the symmetry group is the maximum group @xmath15 . the difference between integer and half - integer spins is a fundamental property of the fermionic representation as we shall see more later .
now let us see how the constraints eq .
( [ newcontr1 ] ) transform under the symmetry group . for @xmath0 , the constraint eq .
( [ newcontr1 ] ) is invariant under the transformation @xmath83 because the right hand side vanishes ( due to the particle - hole symmetry of the hilbert space ) . for integer spins , if @xmath84 , then @xmath85 , and eq .
( [ newcontr1 ] ) is invariant .
if @xmath86 , then @xmath87 , meaning that the particle " picture(@xmath62 sign in eq .
( [ contr2 ] ) ) and the hole " picture ( @xmath63 sign in eq .
( [ contr2 ] ) ) are transformed to each other .
for a half - odd - integer spin with @xmath88 , @xmath89 is a rotation and we need to extend the constraint into a vector form similar to @xmath0 case,@xcite so that eq .
( [ contr2 ] ) becomes @xmath90 under the group transformation @xmath83 , @xmath91 where @xmath92 , @xmath93 , i.e. @xmath42 is a 3 by 3 matrix representing a 3d rotation in the internal hilbert space .
the transformed constraint represents a new hilbert subspace which is still a @xmath94-dimensional irreducible representation of the spin @xmath15 algebra . any measurable physical quantity such as
the spin @xmath95 remains unchanged in the new hilbert space .
therefore , for half - odd - integer spins ( @xmath88 ) , there exists infinitely many ways of imposing the constraint that gives rise to a hilbert subspace representing a spin .
however , for integer spins , there exists only two possible constraint representations .
the different representations of constraints for @xmath2 systems result in different path integral representations and different mean field theories .
these mean field theories are equivalent in the sense that they can be transformed to each other by the internal symmetry group . in the next section
, we shall study the heisenberg model in different representations and shall construct a
mixed " path integral representation which restored particle - hole symmetry , and the internal symmetry group becomes almost " a gauge symmetry.@xcite the mixed " representation is studied in section iv where a new mean field theory is proposed which recovers haldane conjecture at 1d for the heisenberg model .
we shall focus on the antiferromagnetic heisenberg model @xmath96 with @xmath97 in the rest of the paper .
we start by presenting some useful algebraic manipulations of the hamiltonian , and then discuss the general path integral formalism .
the different ways of handling constraints and how they affect the symmetry of the lagrangian will be discussed in the process .
it is known that for spin-1/2 the heisenberg interaction can be written as @xcite @xmath98 where @xmath99 are spin - singlet operators . @xmath100 and
@xmath101 will be extended to arbitrary spin magnitudes with the above definition and will be used in the following discussions .
interestingly , an expression almost the same as eq .
( [ s1/2 ] ) holds for @xmath3 , but this is no longer true for larger spins .
we shall present precise formula for @xmath4 and @xmath102 , and provide a general discussion for higher spins .
for @xmath3 , the three spin matrices are @xmath103 where @xmath104 .
the matrix operator @xmath64 is given by @xmath105 the spin operator can be written in form of eq .
( [ spin ] ) , and it can be shown after some straightforward algebra that the hamiltonian can be written as @xmath106 for @xmath4 , the three spin matrices are @xmath107 and the matrix operator @xmath64 is given by @xmath108 where @xmath109 and @xmath110 . in this case
the singlet operators @xmath100 and @xmath101 alone are not enough to represent the heisenberg hamiltonian , and triplet hoping and pairing terms are necessary .
after some straightforward but tedious algebra , we find that the hamiltonian for spin @xmath4 heisenberg model can be written as @xmath111:\nonumber\\\end{aligned}\ ] ] it is interesting to note that the above form holds also for @xmath3 @xcite . however , for @xmath3 , the two terms in the square bracket can be transformed from one to another , which does not hold for @xmath4 .
this suggests that @xmath112 heisenberg models have a higher hidden symmetry " compared with @xmath4 . actually , there is another hidden symmetry for the spin @xmath0 systems which is absent for spin-1 , @xmath113 where @xmath114 is the particle number on site @xmath44 .
our analysis suggests that smaller spins have higher symmetry when the spin interaction is expressed in the fermionic representation . for @xmath5=2
, the heisenberg hamiltonian can be written as @xmath115 as in the case of lower spins , the terms @xmath116 and @xmath117 have finite overlaps and are not completely independent of each other . for @xmath118 , it is not possible to represent the hamiltonian in the above two terms alone .
quintet and higher multipolar hoping and pairing operators are needed to represent the heisenberg hamiltonian and we are not able to obtain any general expression .
summarizing , the heisenberg model can be written in the fermionic representation as @xmath119\end{aligned}\ ] ] where @xmath120 are parameters dependent on the spin magnitude @xmath5 and @xmath121 represents quintet and higher multiplet terms . in the following
, we shall study in detail a hamiltonian which keeps the first two terms only .
the hamiltonian can be considered as an effective hamiltonian for constructing trial ground state wavefunctions in a variational calculation .
notice that the effective hamiltonian is in fact exact " up to @xmath102 if @xmath122 and @xmath123 are chosen properly . in imaginary time path
integral formalism , the partition function is given by @xmath124+h\nonumber\\\end{aligned}\ ] ] where @xmath125 is the lagrange multiplier field introduced to impose the constraint eq .
( [ contr2 ] )
. the @xmath126 sign corresponds to the particle ( @xmath127 ) and hole ( @xmath128 ) representations . the grassmann number @xmath64 satisfy the antiperiodic boundary condition @xmath129 .
notice that the lagrangian eq .
( [ u ] ) is not invariant under the action of the internal symmetry group because of the non - invariant particle number constraint eq .
( [ contr2 ] ) , i.e. the internal symmetry group of the spin operator is not a symmetry group of the lagrangian . in the following
we will employ a trick to restore the symmetry of the lagrangian .
the integer- and half - odd - integer- spin models will be discussed separately because of the intrinsic difference in their internal symmetry groups .
our trick is to consider an average of all the partition functions with different particle - number constraint representations . in the integer spin case ,
the average is given by @xmath130+h)d\tau}\times\nonumber\\ & & \prod_{\{i,\tau\}}\frac{1}{2}\left[\delta(\hat n_i-1)+\delta(\hat n_i-2s)\right ] \nonumber\\ & = & \int d\psi d\psi^\dagger d\lambda e^{-\int_0^\beta ( \sum_i\frac{1}{2}\mathrm{tr}[\psi_i\partial_\tau\psi^\dagger_i]+h)d\tau}\times\nonumber\\ & & \prod_{\{i,\tau\}}\frac{1}{2}\left(e^{i\frac{\lambda_i}{2}[\mathrm{tr}(\psi_i\sigma_z \psi_i^\dagger)-2s+1]}\right.\nonumber\\&&\left .
+ e^{i\frac{\lambda_i}{2}[\mathrm{tr}(\psi_i\sigma_z\psi_i^\dagger)+2s-1]}\right).\end{aligned}\ ] ] notice that we do not make any approximations in deriving eq .
( [ path1 ] ) from eq .
( [ u ] ) and the new " partition function is a faithful representation of heisenberg model except that it averages over all possible particle " and hole " representations of constraints locally .
similar ideas have been applied to generate the supersymmetric representation and the @xmath15 representation of the @xmath131-@xmath132 model @xcite .
the fermion particle - hole symmetry is restored in this representation .
the lagrangian corresponding to eq .
( [ path1 ] ) is @xmath133 -2\ln\cos\frac{(2s-1)\lambda_i}{2}\right]\nonumber\\ & & + h.\nonumber\end{aligned}\ ] ] this form of lagrangian is invariant under the internal symmetry group @xmath16 and the symmetry group becomes almost a gauge symmetry " of the new lagrangian as we shall see in the following .
we note that the lagrangian is complex because of the multiplier @xmath134 .
this problem can be solved by lifting the contour of integration over @xmath134 into complex plane via analytic continuation to find a saddle point in the imaginary axis @xcite . for convenience ,
we write @xmath135 , and integrate @xmath136 from @xmath137 to @xmath138 ( the saddle point of @xmath136 is real )
. then the lagrangian becomes @xmath139 -2\ln\cosh\frac{(2s-1)\tilde\lambda_i}{2}\right]\nonumber\\ & & + h,\end{aligned}\ ] ] we shall now consider the path integral representation in terms of the effective hamiltonian eq .
( [ htr ] ) keeping only the first two terms .
the effective hamiltonian can be decoupled by a hubbard - stratonovich matrix field ( see appendix b ) @xmath140 where @xmath141 , and @xmath142 . note that @xmath143 and @xmath144 form two sets of spin triplet operators , respectively .
the hamiltonian becomes @xmath145\right.\nonumber\\ & & \left.+b(s)\mathrm{tr}[\hat { \mathbf v}_{ij}^\dagger\cdot\hat { \mathbf v}_{ij } -(\hat{\mathbf v}_{ij}\cdot \psi_j^\dagger\mathbf i\psi_i+h.c.)]\right\}.\end{aligned}\ ] ] the @xmath146 and @xmath136 fields define @xmath16 lattice gauge fields coupling to the fermionic particles which we shall call spinons in the following . under the local gauge transformation @xmath147 ,
the temporal component and the spacial component of the gauge fields transform as @xmath148 notice that @xmath125 changes sign under an uniform @xmath149 gauge transformation , and @xmath100 , @xmath101 exchange their roles under a staggered " @xmath149 transformation where we have @xmath150 on one sublattice and @xmath151 on the other sublattice .
notice also that the lagrangian is not invariant under a time - dependent gauge transformation , because the term @xmath152 is not invariant .
a similar situation occurs in the supersymmetric representation of heisenberg model . because of
this restriction the internal symmetry group does not generate a complete gauge symmetry " in the path integral formalism . integrating out the fermion fields , we get an effective action in terms of the @xmath153 , @xmath154 and @xmath125 fields . the mean field values of the these fields
are given by the saddle point of the effective action determined by the self - consistent equations @xmath155 , @xmath156 with the mean field hamiltonian @xmath157+b(s)[-(\mathbf v^*\cdot c_i^\dagger\mathbf
i c_j+\mathbf u^*\cdot\bar c_i^\dagger\mathbf
i c_j+h.c.)\right.\nonumber\\ & & \left.+|\mathbf u|^2+|\mathbf v|^2]\right ] + \sum_i[\tilde\lambda(\hat n_i-\frac{2s+1}{2})\nonumber\\ & & -\ln\cosh\frac{(2s-1)\tilde\lambda}{2}],\end{aligned}\ ] ] where @xmath158 , @xmath159 , @xmath160 , @xmath161 , and @xmath162 is determined by the condition @xmath163 notice that the averaged particle number satisfies @xmath164 because we are mixing the particle and hole pictures in the constraint .
if the ground state does nt break the particle - hole symmetry , then @xmath165 and we get the half - filling condition @xmath166 . we first examine how the two particle number constrains @xmath167 are transformed into each other by the @xmath15 internal symmetry group .
we consider the vector constraint eq .
( [ veccontr ] ) for @xmath168 @xmath169 where @xmath170 the @xmath171 and @xmath172 components indicate that the one - site pairings @xmath173 @xmath174 are equal to 0 , and the @xmath175 component imposes the constraint that there is exactly one fermion per site .
notice that the first two components of the constraint is automatically satisfied if the third component is satisfied rigorously . under the @xmath15 transformation @xmath176 ,
the constraint becomes @xmath177 , where @xmath178 , @xmath42 is an @xmath67 rotation .
since the constraint is invariant under an @xmath81 gauge transformation generated by @xmath179 , the continuum space formed by different constrains is the surface of a sphere @xmath180 . in particular , for @xmath181 , the north pole of the sphere which corresponds to the positive sign in eq .
( [ contr2 ] ) is transformed to the south pole , @xmath182 , corresponding to the negative sign in eq .
( [ contr2 ] ) .
now we construct a lagrangian which is invariant under the @xmath15 internal symmetry group . in the path integral formalism , the particle number constraint @xmath183 can be realized by introducing a vector langrange multiplier field @xmath184 @xmath185\}.\end{aligned}\ ] ] the vector @xmath186 changes into @xmath187 under @xmath15 gauge transformation , where @xmath188 is an unit vector on the surface of the sphere @xmath189 . as in the integer spin case , a @xmath15 invariant path integral formalism can be obtained by averaging over all possible particle number constraints : @xmath190 where @xmath191 .
introducing @xmath192 , we obtain a lagrangian with almost " @xmath15 gauge symmetry : @xmath193 -(2s-1)\ln\frac{\sinh\tilde\lambda_i}{\tilde\lambda_i}]\nonumber\\&&+h.\end{aligned}\ ] ] as in the case of integer spins , we replace @xmath194 by the effective hamiltonian eq .
( [ htr ] ) with only the first two terms . introducing the hubbard stratonovich field as in the integer spin case
( see appendix b for details ) , @xmath195 then @xmath196 and @xmath197 form an @xmath15 lattice gauge field coupling to the spinons .
( notice the sign difference in the matrix elements of @xmath198 and @xmath199 between integer and half - odd - integer spins . ) under gauge transformation @xmath147 ( @xmath200 ) the gauge fields transform as , @xmath201 and the lagrangian eq .
( [ lsu2 ] ) is invariant only under time - independent gauge transformations as in the integer spin case because of the non - invariant term @xmath202 .
this is different from the @xmath0 case , where the lagrangian is invariant under time - dependent gauge transformations .
the mean field effective hamiltonian corresponding to eq .
( [ lsu2 ] ) is @xmath203+b(s)[-(\mathbf v^*\cdot c_i^\dagger\mathbf
i c_j+\mathbf u^*\cdot\bar c_i^\dagger\mathbf
i c_j+h.c.)\right.\nonumber\\ & & \left.+|\mathbf u|^2+|\mathbf v|^2]\right ] -\sum_i{2s-1\over2}(\pmb{\tilde\lambda}\cdot\hat { \mathbf m}_i + \ln\frac{\sinh\tilde\lambda}{\tilde\lambda})\nonumber\\\end{aligned}\ ] ] where
the parameters @xmath204 are solved self - consistently by the mean field equations @xmath205 , @xmath206 and @xmath207 is determined by @xmath208 the averaged particle number per site is again given by @xmath166 if the ground state respects particle - hole symmetry @xmath209 .
it can be proved ( see appendix b ) that for integer spins , @xmath210 , @xmath211 , @xmath212 and @xmath213 , whereas for half - odd - integer spins , @xmath210 , @xmath214 , @xmath215 and @xmath213 . notice that the parity of the paring terms @xmath101 and @xmath143 are different for integer and half - odd - integer spins .
these operators are central to the mean field theory , because the ground states are completely determined by their expectation values .
in particular , for a mean field theory with @xmath216 and @xmath217 , the mean field ground state is a bcs spin - singlet pairing state where the order parameter @xmath218 has even parity for half - odd - integer spins , but has odd parity for integer spins .
we shall see how this important difference leads to different excitation spectrums between integer and half - odd - integer spin systems in the fermionic mean field theory .
similar result exists for states with @xmath219 and @xmath220 . in this case
the mean field ground state is a bcs spin - triplet pairing state and the parity of the order parameters are reversed .
we note that as in the @xmath0 case , the mean field hamiltonian should be viewed as a trial hamiltonian for the ground state wavefunction of the spin systems after gutzwiller projection . for the mean field theory with particle - hole symmetry
the gutzwiller projection is rather non - trivial .
the state is a coherent superposition of states which allows sites with both one fermion and with @xmath12 fermions .
the numerical analysis of such a state is complicated and we shall not go into details in this paper .
in this section we apply our mean field theory to the antiferromagnetic heisenberg model in one dimension .
we first consider the cases of spin @xmath3 and @xmath4 where two versions of mean field theories based on different methods of implementing the constraints will be discussed .
the mean field results are summarized in section iv.c where the case of general spin @xmath5 and haldane conjecture will be discussed . to simplify our analysis
we make use of the wagner - mermin theorem which asserts that a continuous symmetry can not spontaneously break in one dimension@xcite . therefore we assume in our mean - field theory that the ground state is a spin singlet ( spin liquid state ) and the expectation values of @xmath221 and @xmath222 are zero , since
the rotational symmetry will be broken otherwise . as a result , we keep only the @xmath122 term in the trial hamiltonian . for simplicity we shall also restrict ourselves to translational invariant solutions of the mean - field theory in this paper .
we choose @xmath223 to be the same as eq .
( [ h1 ] ) , i.e. , @xmath224 . for convenience of discussion
, we introduce two new fermions @xmath225 which are the symmetric and antisymmetric combinations of @xmath226 and @xmath227 , @xmath228 the mean field hamiltonian eq .
( [ heff1 ] ) becomes completely decoupled in terms of @xmath225 and @xmath78 . in fourier space , it can be written as @xmath229\nonumber\\ & & + jn(|\chi|^2+|\delta|^2)-n(\frac{3}{2}\tilde\lambda+\ln\cosh\frac{\tilde\lambda}{2}),\end{aligned}\ ] ] where @xmath230 , @xmath231 ( since the phases of @xmath232 and @xmath233 are unimportant in 1d , we choose @xmath232 and @xmath233 to be real numbers ) and @xmath19 is the length of the chain .
the mean field hamiltonian can be diagonalized by a bogoliubov transformation : @xmath234 the coefficients @xmath235 satisfy the bogoliubov - de gennes ( bdg ) equations , @xmath236 solving the equations we obtain @xmath237 where @xmath238 is given by @xmath239 and the diagonalized hamiltonian is @xmath240 where @xmath241 is the ground state energy .
minimizing @xmath242 , we obtain the self - consistent mean field equations at zero temperature , [ mfeq ] @xmath243 with @xmath3 .
the above equations are solved numerically where we obtain @xmath244 and @xmath165 as shown in the first row of table [ tab : spin1 ] .
the averaged particle number of the spinons per site is @xmath245 , indicating that the ground state respects particle - hole symmetry .
the ground state energy is @xmath246 , and the excitation spectrum is flat ( i.e. @xmath247-independent ) with a finite energy gap @xmath248 .
the spin - spin correlation function is @xmath249\\\end{aligned}\ ] ] and is nonzero only for @xmath250 and @xmath251 , with @xmath252 and @xmath253 .
the correlation function is zero for @xmath254 , indicating that the mean field theory describes a short - ranged valence - bond - solid ( vbs ) state .
.solutions of two versions of mean fields for spin-1 af heisenberg chain .
@xmath255 is the ground state energy per site and @xmath256 is the excitation gap . [
cols="^,^,^,^,^,^,^",options="header " , ] comparing the mean field energy dispersion eq .
( [ ex1 ] ) and eq .
( [ ex3/2 ] ) , we find that the excitation spectrum of spin-1 heisenberg model is gapped whereas the excitation spectrum for the spin-3/2 heisenberg model is gapless in the mean field formulation with particle - hole symmetry .
this difference between integer spin and half - odd - integer spin persists for any spin @xmath5 , if we consider the trial hamiltonian eq .
( [ heff1 ] ) or eq .
( [ heff2 ] ) with @xmath257 , i.e. if we consider bcs spin - singlet ground state wavefunctions .
the mean field equations for arbitrary @xmath5 are the same as eq .
( [ mfeq ] ) or eq .
( [ mf3/2 ] ) for arbitrary @xmath5 , and the dispersion is qualitatively the same as eq .
( [ ex1 ] ) or eq .
( [ ex3/2 ] ) , as long as we consider particle - hole symmetric mean field solutions which do not break translational invariance .
the main effect of changing @xmath5 in mean field theory is to change the number of fermionic spinon species .
this result is consistent with the haldane conjecture , @xcite which asserts that the integer spin heisenberg chains have singlet ground state with finite excitation gaps and exponentially decaying spin - spin correlation functions , while half - odd - integer spin chains have singlet ground states with gapless excitation spectrums and power law decaying correlations .
our particle - hole symmetric mean field results agree well with haldane conjecture .
haldane noticed that the time reversal operator satisfies @xmath258 for integer spin and @xmath259 for half - odd - integer spin , and this results in different berry phase contributions from the topological excitations ( skymion or instanton ) in the path integral formulation given by @xmath260 , where @xmath261 is the skyrmion number .
for @xmath262 odd integer , the berry phase @xmath263 is @xmath264 for integer spins and @xmath265 for half - odd - integer spins .
the difference in the @xmath266 factor between integer spin and half - odd - integer spin is the origin of haldane s conjecture .
the situation is quite similar in our mean field theory . noticing that @xmath267 is a singlet formed by two spin-@xmath5 particles .
the clebsch - gordan coefficients implies that @xmath211 for integer @xmath5 and @xmath214 for half - odd - integer @xmath5 ( see appendix b ) .
this sign or parity difference results in appearance of @xmath268 in eq .
( [ ex1 ] ) for @xmath3 and @xmath269 in eq .
( [ ex3/2 ] ) for @xmath4 , which leads to different symmetries of ground state wavefunctions and different excitation spectrums between integer and half - odd - integer spin chains . and
@xmath270 are two reciprocal vectors .
the brillouin zone is divided into three regular pieces by the three new bases @xmath271 , @xmath272 and @xmath273 . ,
width=316 ] the topological term eq .
( [ tplg ] ) leads to exponentially localized edge states for open integer spin chains which is absent for half - odd - integer spin chains .
@xcite this important difference between integer spin and half - odd - integer spin chains is also reflected in our mean field theory where zero energy majorana edge fermions exist at the open boundaries for integer spin chains which are absent for half - odd - integer spin chains .
the existence of majorana edge fermions for integer spin chains is a direct consequence of the odd - pairing symmetry for integer spin chains .
@xcite the details of the majorana edge fermions is discussed in appendix c.
we show in the previous section that our mean field theory is able to capture the essential physics of spin - liquid states in 1d antiferromagnetic quantum spin chains . in this section
we shall apply our mean field theory to the 2d @xmath274-@xmath275 heisenberg model on triangular lattice .
we shall show that our mean field theory admits new spin liquid solutions not explored before .
some results of this model have been reported in a previous paper@xcite , where particle number constraint is treated in the particle representation " ( @xmath276 ) . in this paper , we shall revisit this model with the particle - hole symmetric constraint .
the hamiltonian of the @xmath274-@xmath275 model is @xmath277}\mathbf s_i\cdot \mathbf s_{j},\end{aligned}\ ] ] where @xmath278 denotes nearest neighbor ( nn ) and @xmath279 $ ] the next next nearest neighbors ( nnnn ) . for @xmath3 , there are two channels of decoupling the spin interaction , namely , @xmath280 or @xmath281 .
since there is no mermin - wigner theorem to protect us at zero temperature at 2d , we have to keep both terms in the effective hamiltonian eq .
( [ heff1 ] ) .
we shall choose @xmath282 , with the weight @xmath283 determined by minimizing the ground state energy . .
we indicate the relative phase of the nearest neighbor pairing @xmath284 .
( a ) , @xmath7-wave pairing .
the dished lines are the zero lines of the pairing function ( in lattice space or momentum space ) .
( b ) , @xmath6-wave pairing .
there are no zero lines , the pairing function vanishes only at the @xmath285 point.,width=288 ] next we introduce the mean field parameters @xmath286 , @xmath287 , @xmath288 , @xmath289(for nn ) and @xmath290 , @xmath291 , @xmath292 , @xmath293 ( for nnnn ) to decouple the hamiltonian .
we assume that @xmath294 and @xmath295 are parallel , so we only keep the @xmath296 component of these vectors , and note them simply as @xmath297 and @xmath298 ( which are set as real numbers ) . as in 1-d case
we consider here only spin liquid solutions that respect translation and rotational symmetries .
since the pairings have odd parity @xmath299 , only @xmath300-wave and @xmath7-wave like pairings are allowed . solving the mean field equations we find two kinds of solutions ( see fig .
[ fig : pairing ] ) , which have @xmath7-wave and @xmath6-wave pairing symmetries , respectively . we shall concentrate on these two kinds of solutions in the following . in momentum space ( fig .
[ fig : bz ] ) , the mean field hamiltonian eq .
( [ heff1 ] ) can be written as @xmath301\nonumber\\&&+\sum_{\mathbf{k } } v_k ( c^\dagger_{1k}c_{1k}-c^\dagger_{-1k}c_{-1k } ) \nonumber\\&&-\sum_{\mathbf{k } } u_{k}^*(c_{-1-k}c_{1k}+h.c.),\ ] ] with @xmath302 where @xmath162 is the lagrange multiplier and @xmath303,\\ & \psi_k^{f}={1\over3}[\sin k_x+\sin(-{k_x\over2}+{\sqrt3 k_y\over2 } ) + \sin(-{k_x\over2}-{\sqrt3k_y\over2})],\\ & \psi_k^{p+ip}={1\over3}[\sin k_x+e^{i{\pi\over3}}\sin({k_x\over2}+ { \sqrt3 k_y\over2})\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \\ & \ \ \ \ + e^{i{2\pi\over3 } } \sin(-{k_x\over2}+{\sqrt3 k_y\over2})].\ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \ \\end{aligned}\ ] ] the pairing symmetries can be verified easily by expanding the pairing terms at small @xmath247 : @xmath304 , and @xmath305 .
there are three lines of zeros for the @xmath7-wave pairing function as shown in fig .
[ fig : pairing ] . introducing a vector @xmath306 , the hamiltonian eq .
( [ mf_hb ] ) can be written in a matrix form : @xmath307 where @xmath308 @xmath309 can be diagonalized by the bogoliubov transformation , @xmath310 where @xmath311 and @xmath312 .
the corresponding eigenvalues are given by @xmath313 and @xmath314 , where @xmath315 and the self - consistent mean field equations are @xmath316 , \nonumber\\ & & \bar \delta_{1,3}=\frac{1}{n}\sum_{\mathbf{k}}\psi_{k,2k}[\frac { |\delta_k|}{e_{1k}}+ \frac{|\delta_k|}{2e_{0k } } ] , \nonumber\\ & & \bar u_{1,3}=\frac{1}{n}\sum_{\mathbf{k}}\gamma_{k,2k}\frac { u_k}{e_{1k } } , \nonumber\\ & & \bar v_{1,3}=0 , \nonumber\\ & & \langle \hat n_i-\frac{3}{2}\rangle=\frac{1}{2}\tanh \frac{\tilde\lambda}{2 } , \ ] ] where we have adopted the particle - hole symmetric constraint in writing down the mean field equations .
we have also checked the mean field solutions with constraint @xmath317@xcite and found that the excitation spectrums are qualitatively the same as those in the particle - hole symmetric theory .
the ground state energy is lowest for vanishing @xmath318 , i.e. @xmath319 , @xmath320 and the ground state is a spin - singlet .
the excitations are characterized by three branches of fermionic spinons with @xmath321 and identical dispersion @xmath322 . -@xmath275 heisenberg model with @xmath323 .
we use the constraint @xmath324 .
the red dot line indicate energy per site for the @xmath6 state and the black square line is for the f - wave state .
dirac nodes exist for the f - wave state .
the number of nodes is 6 when @xmath274 is dominating and 24 when @xmath275 is dominating.,width=268 ] our mean field theory for the @xmath274-@xmath275 model contains two regimes of spin liquid states for both @xmath7 and @xmath6 pairing symmetries as a function of @xmath325 .
a first order phase transition occurs between the two regimes at @xmath326 ( see fig . [
fig : hsb ] ) .
when @xmath274 dominates , the spin liquid state is characterized by @xmath327 and @xmath328 ( consequently @xmath329 and @xmath330 ) ; while when @xmath275 dominates , @xmath331 and @xmath332 , @xmath333 ( consequently @xmath334 , @xmath335 ) .
the @xmath6 states remain lower in energy in both regimes .
the @xmath7-wave pairing solution respects particle - hole symmetry and has @xmath336 .
the excitation is gapless with several dirac cones in the first brillouin zone .
a cut of the spinon dispersion is shown in fig .
[ fig : f_band ] where the particle - hole symmetry is obvious .
the mean field solution with @xmath337 has similar properties , except that the particle - hole symmetry is lost and the position of the dirac nodes are shifted.@xcite for the @xmath6-wave pairing , we find that the ground state breaks particle - hole symmetry and @xmath338 .
the excitation spectrum is fully gapped .
similar to integer spin chains , we find that the @xmath6-wave state is topologically nontrivial ( see also appendix c ) .
the solution has @xmath339 , so that @xmath340 can be either positive or negative , depending on @xmath341 .
therefore the bogoliubov spinor space described by the vector @xmath342 ( which is @xmath189 ) covers the brillouin zone ( @xmath341 space , which is also @xmath189 ) at least once . in other words ,
the topological ( skyrmion ) number of mapping from the @xmath341 space to the spinor space is nonzero @xmath343 .
however in the vacuum where the spinon density is zero , @xmath344 is very big and @xmath340 can only take positive values , which gives a zero topological number .
@xcite since the bulk and vacuum belong to different topological sectors , the boundary defines a domain wall between the two phases , and there should exists gapless ( chiral ) majorana edge states in the @xmath6 state following the analysis of read and green .
@xcite in this sense , the @xmath6-state describes a time - reversal symmetry breaking topological spin liquid . for @xmath345 , @xmath346 .
the figure on the right shows the dispersion along the line linking two @xmath347 points of the brillouin zone ( see the blue lines on the left ) .
the negative energy part is filled and the excitation is gapless with 6 dirac nodes in the first brillouin zone .
the dispersion for the mean field solution with @xmath337 is similar to the above one except that the locations of the nodes are shifted a little bit.,width=316 ] next we consider the specific heat and spin susceptibility for the @xmath7- and @xmath6 states . for the @xmath6-wave pairing states ,
since the excitations are fully gapped as in bcs superconductors , the specific heat and spin susceptibility will show exponential behavior @xcite at low temperature . for the @xmath7-wave pairing states , both the specific heat and the spin susceptibility show power law behavior due to the dirac cone structure ( see fig . [
fig : f_band ] ) .
the energy for the system is @xmath348 in mean field theory . at low temperature ,
the specific heat is dominate by the low energy excitations near the nodes of the fermi surface , which is given approximately by @xmath349 , where @xmath350 is the fermi velocity ( the anisotropy of the dispersion can be removed by a re - scaling of momentum and will not affect the result qualitatively ) .
the magnetic specific heat is given by @xmath351 where @xmath352 is the number of dirac cones .
the magnetic susceptibility can be calculated from the linear response theory with @xmath353\nonumber\\\end{aligned}\ ] ] the static susceptibility is given by @xmath354 .
the detailed calculation is given in appendix d. recently , a spin-1 material @xmath355 was discovered , where the @xmath3 @xmath356 ions form a triangular lattice with antiferromagnetic ( afm ) interaction .
the system has a curie - weiss temperature @xmath357k , and is found to exhibit no conventional magnetic order down to @xmath358k .
several plausible ground states have been proposed for this system , @xcite including an antiferro - nematic ( afn ) state @xcite , a ferro - nematic state , @xcite and a kt phase driven by vortices .
@xcite spin liquid is also proposed to be a plausible ground state . in our mean field theory , the @xmath7-wave
pairing spin liquid has coherent gapless excitation with specific heat scaling as @xmath359 at low temperature which is consistent with the experimental result .
@xcite furthermore , the knight shift data in ref .
shows that the intrinsic susceptibility decreases with decreasing temperature ( but the very low temperature data are missing .
@xcite ) this matches qualitatively with our mean field result eq .
( [ kubo ] ) .
we therefore proposed that the @xmath7-wave pairing spin liquid state is a plausible ground state for the material .
summarizing , we show in this paper how the fermionic representation for spin @xmath0 systems can be generalized to spin systems with @xmath2 . the symmetry group of the spin operator is @xmath15 group for integer spin and @xmath16 group for half - odd - integer spin .
different path integral formulations and mean field theories are developed corresponding to different ways of handling the constraints . in 1d , we show that the particle - hole symmetric mean field theory for @xmath3 and @xmath4 spin chains are consistent with haldane s conjecture , and we argue that the difference reflects a fundamental difference between integer and half - odd - integer spin chains .
we also study 2d spin-1 afm on triangular lattice where we find two spin liquid states , one is a gapless @xmath7-wave spin liquid and the other is a topological @xmath6 spin liquid state .
we propose that the gapless @xmath7-wave spin liquid is a plausible ground state for the material niga@xmath8s@xmath9 .
our approach can be applied to any other spin models and provides a new approach to spin liquid states for @xmath2 spin systems .
we note that because of limitation in scope we have addressed only a very limited number of issues in the study of spin - liquid states in this paper . within the mean - field theory
we have restricted ourselves to spin liquids solutions of hamiltonian ( [ hamiltonian ] ) with spin rotational symmetry ( 1d ) and lattice translational symmetry ( 1d and 2d ) .
we note that solutions which break spin rotational symmetry exist in our theory which may serve as ground states of hamiltonian ( [ hamiltonian])@xcite .
for example , we show in ref .
that state with long - ranged magnetic order exists as lowest energy state of the mean - field theory of the @xmath274-@xmath275 model and anti - ferro nematic order may exist when the bi - quadratic ( @xmath360 ) spin - spin interaction exists in the hamiltonian .
for simplicity we only considered the mean field states without breaking the translational symmetry .
we note that states that break translational symmetry ( such as dimerized states ) are believed to be ground states of some 1d or 2d spin models.@xcite another very important issue is whether the spin liquid states we find are stable against gauge fluctuations . in 1-d
, the gauge fluctuations can be removed by a time - dependent read - newns gauge transformation and thus have no effect to the low energy properties@xcite .
the situation is very different in two or higher dimensions where the stability of the mean - field state depends on dimensionality and the ( gauge ) structure of the gauge field fluctuations .
for instance , @xmath149 spin liquid is believed to be stable at 2d@xcite and gapless @xmath81 dirac fermionic spin liquids is stable at 2d in the large-@xmath19 limit@xcite . in our case of spin liquid solutions for the @xmath3 heisenberg model , the @xmath81 gauge fluctuation is gapped by the spinon - pairing term via the anderson - higgs mechanism .
the ground state is stable against the low energy gauge fluctuations which is described by an effective @xmath149 gauge theory .
we note that the emergence of gauge - field structure is a direct consequences of particle number constraints and a reliable answer to the question of stability of spin liquid states can be obtained only if we can handle the particle number constraint reliably .
gauge field theory can only handle long distance , low energy gauge fluctuations and a more satisfactory answer to the question of stability of the spin liquid states can be obtained only after the gutzwiller projection wavefunctions are studied carefully .
we thank prof .
michael ma for the discussion about 1d models .
we also thank prof .
a. lee and naoto nagaosa for suggesting the @xmath6-ansatz for afm on triangular lattice , and thank mr .
cheung chan and dr .
xiao - yong feng for helpful discussions .
zxl and tkn are supported by rgc grant of hksar and hkust3/crf/09 .
yz is supported by national basic research program of china ( 973 program , no.2011cb605903 ) , the national natural science foundation of china(grant no.11074218 ) and the fundamental research funds for the central universities in china .
we shall look for allowed transformations @xmath361 , which keep the spin operator eq .
( [ spin ] ) invariant . at first glance
, this condition is satisfied as long as @xmath362 .
however , it is _ not true _ because @xmath24 s and @xmath363 s are not independent . by checking each row of @xmath64 directly ,
it is easy to see that @xmath364 transforms to @xmath365 , and @xmath366 . noticing that @xmath367
, we obtain @xmath368 or @xmath369 on the other hand , @xmath370 transforms to @xmath371 . noticing that @xmath372 and @xmath373
, we obtain @xmath374 or @xmath375 eq .
( [ w1 ] ) and eq .
( [ w2 ] ) impose the condition @xmath376 for integer spin , eq . ( [ w3 ] ) becomes @xmath377 , which results in @xmath378 or @xmath379 , i.e. @xmath380 , where @xmath381 is the usual gauge symmetry group and @xmath382 is the particle - hole symmetry group .
notice that the two groups do nt commute , thus the total symmetry group is a semidirect product of the @xmath81 group and the @xmath149 group @xmath383 .
for half - odd - integer spin , eq .
( [ w3 ] ) implies @xmath384 , which is equivalent to @xmath385 since @xmath386 .
this indicate that @xmath89 .
thus the symmetry group is @xmath15 group .
first we introduce a @xmath32 matrix @xmath387 , which is essential in our proof : @xmath388 the matrix is the cg coefficients combining two spin - s to a spin - singlet state , i.e. @xmath389 . notice that for integer spin @xmath390 while for half - odd - integer spin @xmath391 .
the following are some common properties of the @xmath387 matrix .
suppose @xmath392 is an rotation operation , and @xmath393 is its @xmath32 dimensional matrix representation , then we have @xmath394 , and @xmath395 on the other hand , @xmath396 , so @xmath397 comparing these two , we have @xmath398 , or @xmath399 _ this means that @xmath387 is the transformation matrix which transfers the representation matrix of a rotational operator to its complex conjugate . _ in general , @xmath393 can be written as @xmath400 , where @xmath401 are the @xmath15 group parameters ( real numbers ) for the operator @xmath392 , and consequently @xmath402 . from eq .
( [ bdb ] ) we have @xmath403 so we have @xmath404 noticing @xmath405 , above equation is equivalent to @xmath406 now let us consider the spin algebra .
the spin interaction is decomposed by the following two operators @xmath407 we define @xmath408 , @xmath409 and @xmath410 , @xmath411 .
the following discussions distinguish integer- and half - odd - integer- spins and prove the results in section iii.c . * integer spins * ( @xmath412 ) the second term on the first row of eq .
( [ cd ] ) can be written as @xmath413 in second step , we have used the fact that the transverse of a scalar operator is equivalent to reversing the order of the constituting components .
commuting two fermionic operators we obtain a minus sign . on the other hand , @xmath414 , so @xmath211 . the second term on the second row of eq .
( [ cd ] ) is @xmath415 it is obvious that @xmath210 . to simplify eq .
( [ uv ] ) , we will use the following relations : @xmath416 , @xmath417 and @xmath418 .
the second term on the first row of eq . ( [ uv ] ) is @xmath419 on the other hand , above operator can be written into another form , @xmath420 . comparing with eq .
( [ uij ] ) we get @xmath421 . the second term on the second row of eq . ( [ uv ] ) is @xmath422 and the relation @xmath423 manifests itself . * half - odd - integer spins * ( @xmath424 ) repeating the above procedures , it is straightforward to show that @xmath425 with @xmath426 .
it is known that integer spin antiferromagnetic heisenberg model has free edge states with spin magnitude - s/2 .
@xcite majorana fermion edge states also exist in our mean field theory under open boundary condition .
our discussion will follow the argument of ref . for 2d pairing fermions .
firstly , we point out the existence of two topologically distinct phases under periodic boundary condition and then we discuss the zero energy edge state solution at an open boundary . in our mean field theory , all the properties of the ground state are completely determined by @xmath427 and @xmath428 ( or @xmath340 and @xmath429 ) . noticing that @xmath427 and @xmath428 obey @xmath430 , so they can be viewed as a spinor .
this spionor is equivalent to a vector @xmath431 where we have used eq .
( [ ukvk ] ) . obviously , the spinor @xmath432 spans a @xmath433 space .
recalling that the first brillouin zone is also @xmath433 , the functions @xmath434 describe a mapping from @xmath433(k space ) to @xmath433(spinor space ) .
the mapping degree is characterized by the first homotopy group @xmath435 .
when @xmath436 ( which is the case for our mean field theory ) , the map is nontrivial because every point of the spinor @xmath433 is covered once , in other words , the mapping degree is @xmath437 .
this topological number defines a phase , we call it a phase . when @xmath438 , the mapping is topologically trivial since @xmath340 is always positive(or negative ) and the lower half circle(or upper circle ) of the spinor @xmath433 is never covered , so the mapping degree is @xmath439 .
we call this region b phase
. since the mapping can not be smoothly deformed from @xmath439 to @xmath437 , a topological phase transition occurs at @xmath440 .
we will see that the existence of zero energy edge states is tied with a phase boundary between the bulk ( a phase ) and the vacuum ( b phase ) .
next we consider an infinite chain with a single edge .
we assume that the edge is located at @xmath441 , and @xmath442 is the vacuum where the spinon density is zero .
the boundary can be described by a potential @xmath443 that is large and positive at @xmath442 so that the wavefunction of the spinon vanishes exponentially at @xmath442 .
equivalently , we can assume a position dependent lagrange multiplier term @xmath444 which becomes very large and positive(so that the condition @xmath438 is satisfied ) at @xmath442 .
this implies that the vacuum belongs to the b phase and the boundary is a domain wall between a phase and b phase .
we now study the edge states in the continuum approximation .
it is convenient to introduce @xmath445 such that @xmath446 is positive in the bulk , becomes zero near the edge and turns negative outside the edge .
expanding @xmath340 and @xmath429 to first order in @xmath247 near @xmath447 , we obtain @xmath448 and @xmath449 . to see the effects of the domain wall , we consider the bdg equation eq .
( [ bdg - k ] ) , which can be written in position space as @xmath450 where we have replaced @xmath255 and @xmath247 by @xmath451 and @xmath452 , respectively .
the bdg equation is compatible with @xmath453 , the spinor @xmath432 satisfying this relation describes majorana fermions .
there is a normalizeable zero energy solution for the above equations .
putting @xmath454 , then @xmath455 and the solution is @xmath456 notice that the bdg equation eq .
( [ bdg ] ) @xmath457 at @xmath458 admits _ no _ normalizable bound solutions .
this shows that the zero mode is the only plausible bound state solution . in the case of a pair of boundaries at @xmath441 and @xmath459
, we can replace the two first order equations by a second order equation , @xmath460(u\pm v),\ ] ] and the energy @xmath255 is not zero for finite @xmath461 .
however two bound state solutions with energy going to zero exponentially with increasing @xmath461 exist .
the solution @xmath462 is centered at @xmath441 and @xmath463 is centered at @xmath459 .
notice that there is only one solution for the above bdg equations , and the two modes @xmath464 are majorana fermions .
the above discussion is applicable to our mean field theory for @xmath3 open antiferromagnetic chain .
the discussion can be generalized to larger integer spins @xmath465 if we adopt the effective hamiltonian eq .
( [ hm1 ] ) . unlike the @xmath3 case ,
our @xmath4 mean field theory produces a mapping from @xmath247-space @xmath433 to the @xmath432 space @xmath433 which is topologically trivial ( essentially because of the even parity of the pairing term ) , and
the majorana edge states do nt exist anymore .
this is in agreement with the result that no exponentially localized edge state exists for open half - odd - integer spin chains .
the existence of power - law localized edge states in half - odd - integer spin chains @xcite is more subtle and is not produced by our mean field theory .
substituting the bogoliubov transformation eq .
( [ boglbv ] ) into the kubo formula eq .
( [ kubo ] ) , we obtain @xmath466 [ -\mathcal c_{q+k}^2 g_a(q+k,\tau)+\mathcal s_{q+k}^2 g_b(-q - k,-\tau)]\right.\nonumber\\&&\left .
-\mathcal c_q\mathcal s_q [ g_a(q,-\tau ) + g_b(-q,\tau)]\times\mathcal c_{q+k}\mathcal s_{q+k } [ g_a(q+k,\tau)+ g_b(-q - k,-\tau)]\right]\nonumber\\ & = & \frac{2}{\beta v}\sum_{q , i\omega } \left[[\mathcal c_q^2 g_a(q , i\omega)-\mathcal s_q^2 g_b(-q ,- i\omega ) ] [ -\mathcal c_{q+k}^2 g_a(q+k , i\omega+i\omega)+\mathcal s_{q+k}^2 g_b(-q - k ,- i\omega - i\omega)]\right.\nonumber\\ & & \left.-\mathcal c_q\mathcal s_q\mathcal c_{q+k}\mathcal s_{q+k } [ g_a(q , i\omega ) + g_b(-q ,-
i\omega ) ] [ g_a(q+k , i\omega+i\omega)+ g_b(-q - k ,- i\omega - i\omega)]\right]\nonumber\end{aligned}\ ] ] with the notations @xmath467 and @xmath468 . using the results @xmath469 and @xmath470
, we get the static susceptibility @xmath471\nonumber\\ & = & \lim_{k\to0}\frac{2}{v}\sum_{q}\left[-(\mathcal c_q\mathcal c_{q+k}+\mathcal s_q\mathcal s_{q+k})^2 \frac{n_f(\varepsilon_{q+k})-n_f(\varepsilon_q)}{\varepsilon_{q+k } -\varepsilon_q}\right]\nonumber\\ & = & -2\int d^2q\frac{\partial n_f(\varepsilon_q)}{\partial \varepsilon_q}\end{aligned}\ ] ] which is the pauli susceptibility for free fermions . at low temperature
, the excitation spectrum can be approximated by dirac cones , so we have @xmath472 actually , a similar expression like eq .
( [ h3/2 ] ) also holds for @xmath0 , @xmath473:$ ] .
notice that the first term in the square bracket has different sign comparing to the expression for @xmath3 and 3/2 .
the two terms in the square bracket are also identical . we note that another method to deal with disordered spin states is modified spin - wave theory where zero magnetization is enforced by a magnon number constraint .
minoru takahashi [ prog .
87 , 233 ( 1986 ) ; phys . rev .
b 40 , 2494 ( 1989 ) ] applied this method to study 1d ferromagnets and 2d antiferromagnets on square lattice .
the results agree well with that of bethe ansatz and schwinger boson mean field theory , respectively .
this is because that the elementary excitations in both situations are spin-1 magnons which can be viewed as bound states of two spin-1/2 spinons .
however , for 1d half - odd - integer antiferromagnetic spin chains , elementary exciations are spin-1/2 spinons .
so far we do not know how this modified spin wave - theory can be generalized to deal with this situation .
ian affleck , nucl .
b 265 , 409 ( 1986 ) ; s. k. yip , phys . rev .
90 , 250402(2003 ) ; j. j. garca - ripoll , m. a. martin - delgado , and j. i. cirac , phys .
93 , 250405 ( 2004 ) ; rajiv r. p. singh , zheng weihong , c. j. hamer , and j. oitmaa , phys .
b 60 , 7278 ( 1999 ) . | the fermion representation for @xmath0 spins is generalized to spins with arbitrary magnitudes .
the symmetry properties of the representation is analyzed where we find that the particle - hole symmetry in the spinon hilbert space of @xmath11/2 fermion representation is absent for @xmath2 . as a result
, different path integral representations and mean field theories can be formulated for spin models . in particular , we construct a lagrangian with restored particle - hole symmetry , and apply the corresponding mean field theory to one dimensional ( 1d ) @xmath3 and @xmath4 antiferromagnetic heisenberg models , with results that agree with haldane s conjecture . for a @xmath3 open chain , we show that majorana fermion edge states exist in our mean field theory . the generalization to spins with arbitrary magnitude @xmath5 is discussed .
our approach can be applied to higher dimensional spin systems . as an example
, we study the geometrically frustrated @xmath3 afm on triangular lattice .
two spin liquids with different pairing symmetries are discussed : the gapped @xmath6-wave spin liquid and the gapless @xmath7-wave spin liquid .
we compare our mean field result with the experiment on niga@xmath8s@xmath9 , which remains disordered at low temperature and was proposed to be in a spin liquid state .
our fermionic mean field theory provide a framework to study @xmath2 spin liquids with fermionic spinon excitations . |
fractoluminescence is the emission of light during fracture of solids .
this phenomenon has been reported in brittle materials such as sucrose @xcite , ice @xcite , mgo @xcite , mica @xcite , and in soft solids such as pressure - sensitive tape @xcite .
it fits into the more general framework of fracto - emission : the emission of particles , including electrons @xcite , positive ions @xcite or x - rays @xcite during fracture .
even though fracto - emission has been known by the scientific community since francis bacon s experiments in 1605 @xcite , it has been very little used as a tool to study the failure of materials .
the common experimental techniques to study material failure are based on acoustic emissions or on direct observations .
acoustic emissions allow to measure the elastic waves produced by events such as microcracking or dislocation avalanches . in a previous article @xcite
, we demonstrated that in certain scintillating materials used for particle detection , fracture was accompanied by emission of light with the same spectrum as the visible scintillating light , and that fractures could therefore contribute to backgrounds in rare - event searches .
we also argued that light emission could give new insight on the rupture dynamics and provide quantitative information about the energy fracture budget . to fully benefit from the excellent resolution of the photomultiplier ( single photons with nanosecond timing ) , to be able to follow the actual crack propagation , and to control the atmosphere around the sample
, we have developed an enhanced multi - channel setup .
it allows us to measure accurately light emissions , acoustic emissions , crack velocity , compression and loading force during stable and fast crack propagation in controlled atmosphere .
our multi - channel setup has been specially designed to study inorganic crystal scintillators such as bgo ( ) , and , which are all very brittle materials . in order to break the specimens in a controlled and reproducible manner , the double cleavage drilled compression ( dcdc ) geometry defined by janssen in 1974
@xcite has been adopted .
each sample is cut into a @xmath0 rectangular parallelepiped .
all faces are polished to optical quality and a 2 mm diameter circular hole is made through the center of the @xmath1 faces .
when such a sample is compressed along its long axis ( see figure [ figdcdc ] ) , a crack can be initiated on each side of the hole in the mirror plane parallel to the @xmath2 faces .
two cracks can then propagate in opposite directions in this plane .
the applied compression stress allows to control the velocity of both cracks up to a critical length , at which point the specimen abruptly breaks into two parts ( fast fracture ) @xcite .
the slow crack propagation below the critical length is not only determined by the applied stress but also by subcritical rupture processes i.e. a thermally activated energy path at the crack tip which ease bond breaking .
the surrounding gas is a well - known agent that may reduce the critical energy barrier required for thermally activated fracture @xcite . in order to take advantage of the sample geometry and cover the different regimes of fracture
, we developed a setup which precisely controls the applied stress on the specimen and the atmosphere surrounding it .
the setup is represented in figure [ figsetupoverview ] .
the sample to be studied is squeezed between a backstop and a pushrod . the backstop is fixed and the pushrod is actuated by a linear stepper motor .
a load cell monitors the force applied on the sample by the motor .
the deformation of the sample is measured by a custom - made displacement sensor .
two acoustic sensors record the elastic vibrations produced by the crack propagation , and a photomultiplier above the sample measures the visible light emitted . simultaneously , the propagation of cracks in the sample is observed using a near - infrared camera and a near - infrared light source ( filtered and not visible by the photomultiplier ) .
signals produced by the sensors are digitized and recorded by a computer ; the video is separately saved on another computer . in order to allow accurate light measurement and to control the atmosphere , the sample , and most of the instruments ,
are located in a sealed chamber .
this enclosure is light - tight , with the exception of a quartz viewport for the photomultiplier tube , and has been tested at vacuum pressures down to @xmath3 .
the vacuum is achieved using a turbo vacuum pump : hicube 80 eco manufactured by pfeiffer vacuum .
the pressure is measured by a pkr 251 compact fullrange gauge .
all sensors but the photomultiplier are inside the enclosure , and were tested for vacuum operations .
the photomultiplier is outside mostly for size and cabling reasons .
the linear actuator has also been left outside because its heat could not be properly dissipated in vacuum and the grease trapped inside may not have been suitable for vacuum operations . during fracture tests
, the sample is squeezed between the pushrod and the backstop .
the pushrod and the backstop are actually the parts which transfer the force to the specimen .
they are designed to compensate for parallel defects of crystal faces with a ball joint adapter at the tip of the pushrod .
otherwise , the load will not properly spread on the whole faces and the edges of the specimen will potentially break .
since the crystals we use have a relatively high young s modulus , the pushrod and the backstop are made out of stainless steel , and the faces in contact with the specimen are mirror - polished for a good mechanical contact . the linear stepper motor which drives the applied stress on the sample is the na23c60 model manufactured by zaber .
its best step resolution ( its smallest displacement ) is below @xmath4 and it allows a maximum continuous thrust of 950 n. the motor is position - controlled but its position is not relevant to determine the sample deformation .
indeed , it would be biased by the deformation of the materials surrounding the sample . moreover
, tests have shown some skipping of the motor gears and they do not seem to be taken into account by the motor controller .
the applied force on the sample is regulated in real - time via an action on the stepper motor position .
the load cell is the key component which gives the feedback on the applied stress .
it converts the applied force ( correlated to the stress in the double - cleavage drilled compression geometry ) into an electrical signal .
the load cell is the sml-300 model with a sga signal conditioner , both produced by interface .
its gain is set to 7.7 mv / n for our application .
according to the manufacturer , the sml-300 load cell had never been tested at low pressure .
however , its design does not allow any air to be trapped inside the gauge , so it is safe for tests in vacuum . in vacuum , the strain gauges glued on the load cell produce of the order of 1 w of heat , which can only be dissipated by conduction through the metal parts holding the gauge , and not by conduction and convection in the air .
the temperature of the cell is then expected to be slightly higher than at room pressure , but it should nt introduce a large error as the temperature is expected to remain within the compensated range ( the thermal error is @xmath5 for temperatures from @xmath6 to @xmath7 ) . a displacement sensor allows to measure the deformation of the sample , and then to get an evaluation of the work supplied by the motor .
we chose a capacitive sensor for its good resolution , and for its ability to work in vacuum and at low temperatures .
laser - based systems were disregarded because of potential interference with the optical readout .
a capacitive sensor is made of two electrodes separated by a variable gap .
the target electrode is fixed , while the probe electrode is attached to the point whose position is to be measured . neglecting fringe effects , two parallel plates
form a capacitance inversely proportional to their separation , @xmath8 the gap : @xmath9 where @xmath10 is the dielectric constant of the medium in the gap , @xmath11 the vacuum permittivity , and @xmath12 the surface area of each electrode .
the dielectric constant of air is equal to @xmath13 at normal pressure and temperature @xcite , meaning that measurements in air will be consistent with those in vacuum to @xmath14 .
our implementation is based on the toth - meijer design @xcite .
as shown in figure [ figcapasensor_electrodeprincipe ] , the use of a large , concentric , guard electrode around the small , circular , probe electrode allows to mitigate fringe effects with the larger target electrode @xcite .
the guard and probe electrode are isolated electrically but at the same virtual voltage .
the electric field lines are then linear between the probe and the target electrode making the capacitance relatively close to the parallel plate model of equation [ eq : capacitance_parallel_plate ] .
the electrodes ( shown in figure [ figcapasensor_electrodes ] ) have been made out of brass and are rotationally symmetrical .
the outer diameter of the target and the guard ring is 10 mm and the diameter of the probe is 6 mm .
the space between the probe and the guard ring is about @xmath15 and is filled with stycast 2850 ft epoxy ( catalyst 9 ) .
coaxial cables connect the electrodes to the capacitance - to - period converter .
readout is accomplished by a martin voltage - controlled oscillator @xcite that converts the capacitance into a period .
the components used for the capacitance - to - period converter are listed in table [ dispsensor_companents ] .
[ 1]>b#1 @xmath16 + @xmath17 & oa1 : ltc6244 + @xmath18 & oa2 : opa132u + this board runs a xilinx spartan-3e fpga which provides a basic package for fast prototyping .
a @xmath19 oscillator generates the main clock for the fpga .
a daughter board has also been designed to connect the capacitance - to - period converter inputs and outputs to the fpga and to carry the digital - to - analog converter ( dac ) .
the system output is thus a voltage proportional to the gap @xmath8 between the electrodes .
our capacitive sensor was calibrated using the linear stepper ( sec .
[ sec : compressionsystem ] ) to move the probe electrode .
results are shown in figure [ figcapasensor_calibration_maintext ] .
the sensor does not provide the absolute distance between the electrodes , but rather the change in distance between them as the stepper moves , which is the relevant quantity as the sample is compressed . for the calibrations ,
the linear stepper is directly attached to the probe electrode ( in standard operations , deformation of various intermediate objects like the force gauge and the pushrod mean that the displacement seen by the probe is different from that generated by the motor ) .
the linear stepper therefore provides a calibrated displacement .
comsol simulations have been used to determine the actual capacitance , taking into account the full geometry of the electrodes .
matlab was used to simulate the capacitance - to - period electronic conversion .
together , these simulations yield the offset to convert the motor steps into the absolute distance depicted in fig .
[ figcapasensor_calibration_maintext ] .
the sensor can measure displacements , rather than the absolute gap , with a relative accuracy of 1% , as determined by comsol and matlab simulations accounting for factors including nonlinearities and misalignment of the electrodes .
the sensor functions over gaps ranging from 0.2 mm to 0.8 mm .
the lower end of the range corresponds to the electrodes nearly touching ; the upper end corresponds to the dac saturating in amplitude .
use of updated components ( tab .
[ dispsensor_companents ] ) allows to refresh the measurement at a rate of 126 hz , compared to the 10 hz of the original design @xcite .
operation has been demonstrated from ambient pressure down to @xmath3 . during material failure , crack growth and avalanches of dislocations release the stored strain energy .
a part of this energy is converted into elastic waves , which propagates through the material .
these elastic waves are commonly called acoustic emission and are generally in the ultrasonic range . in order to monitor the acoustic activity , two piezoelectric transducers , converting the transient waves into an electrical signal ,
are placed on the surface of the specimen .
the pico model , manufactured by mistras group and shown in figure [ figae_pico ] , is employed .
it is a miniature and lightweight sensor ( outside diameter 5 mm , height 4 mm , weight @xmath20 ) .
it is well - suited for small specimens and it allows to observe a large range of frequencies , with a 20 db bandwidth between 200 khz and 1 mhz .
the current generated by the acoustic sensors is converted into a voltage and amplified using a custom circuit .
an external board cuts off all the frequencies outside the range 10 khz1 mhz before recording the signals . in order to be sensitive to ultrasonic waves ,
the sensors require a coupling to the specimen which minimizes the density of air bubbles trapped between the specimen and the sensors @xcite .
we therefore glue the sensors to the sample using dow corning 732 silicone adhesive , which offers good adhesion on multiple substrates .
a small force is applied to keep the sensors in place for at least 12 hours , allowing the glue to dry . during fracture ,
no added mechanical device is needed to couple the sensors to the specimen , leaving a clear field of vision for the light sensor and for the camera .
once a fracture test is done , the acoustic sensors can be simply removed by twisting .
the pico sensors , adhered with the dow corning 732 adhesive , have been operated successfully at pressures down to @xmath3 .
the light emissions during our fracture experiments are measured by a photomultiplier viewing the sample inside the experimental chamber through a quartz window .
photomultipliers use the photoelectric effect to convert an incident photon into an electron that is then multiplied into a measurable signal @xcite . during scintillation and fracture of the bgo crystals we often use ,
the light emissions are within the range 350650 nm with a peak at 480 nm @xcite ( see also sec . [
sec : spectro ] ) . the r7056 model from hamamatsu is employed because it is sensitive over the range 185650 nm and is well - suited for photon counting .
its maximum quantum efficiency is @xmath21 at 420 nm .
one contribution to background comes from dark counts caused by the emission of electrons from the photocathode by thermal fluctuations ( called thermionic emission ) .
since our setup does not allow to control the temperature of the photomultiplier , fracture tests are usually performed in less than two hours to avoid large fluctuations of temperature and ensuing changes in backround . the amount of light emitted during fracture tests turned out to be quite large @xcite . in order to deal with large amounts of light ,
two spacers are placed in front of the photomultiplier .
filters such as neutral density filters or hot filters ( see section [ sec : camera_tracking ] ) can then reduce the light reaching the light sensor . between calibrations and fracture measurements
, filters can be switched without changing the distance between the sensors and the specimen to be studied , in other words , without changing the solid angle of the specimen for the light sensor .
one of the novelties of our setup is the ability to observe the propagation of cracks within the sample , while measuring the light emission .
based on the fact that the photomultiplier is relatively insensitive to near - infrared light , the specimen is illuminated using a near - infrared light source to allow a modified webcam to record the material failure . a 980 nm laser - diode ( vcsel-980 from thorlabs )
is used as near - infrared light source because it can provide enough power ( 2 mw ) to illuminate the fracture scene and its narrow spectral emission range ( full width at half maximum 0.75 nm ) is outside the photomultiplier spectral range .
the narrow beam of the laser diode is spread thanks to some light - diffusing plastic sheets scavenged from a computer screen . a modified c920 webcam from logitech is employed to record high - definition videos 1080 pixels ) with h.264 hardware encryption . ] of samples during fracture tests .
its hot filter has been removed to broaden the camera range up to the actual limit of the cmos sensor ( above 1000 nm ) and to be then sensitive to the light at 980 nm .
the bulky microphones on the camera board and the blue leds have been unsoldered for practical reasons . the webcam is placed inside the vacuum chamber , as close as possible to the sample .
the cmos sensor quickly gets warm at ambient pressure , so a brass arm is used to make a thermal junction between the sensor and the metal baseplate of the setup . in vacuum , it helps heat to dissipate by conduction .
operation confirmed that the stock webcam lens was not damaged when the air was pumped out of the setup , allaying fears the lens would fail because of air trapped inside .
first tests showed that the photomultiplier was definitely sensitive to the light emitted by the laser - diode . an hypothesis to explain
this unexpected result is that multiple photon interactions with an electron might give to the electron enough energy to escape from the photocathode .
the relative high power of the laser - diode make this hypothesis plausible .
indeed , a power of 2 mw means about @xmath22 is @xmath23 . for a 980 nm photon , @xmath24 .
hence , a @xmath25 light source generates @xmath22 . ] which is far from being negligible .
short - pass filters have thus been added in front of the photomultiplier to attenuate the unwanted signal caused by the laser - diode . during fracture tests ,
the sample is illuminated in such a way that the cracks are dark on a bright background ( see figure [ figcamera_detection ] ) .
the good contrast allows to simply convert the images to grayscale and then to black - and - white using a well - chosen threshold which highlights the crack surfaces .
the crack tips are found as the horizontal extrema of the crack surfaces .
the absolute crack lengths are then computed using the pinhole camera model which describes the relationship between the 3d scene and the 2d projection on the cmos sensor .
this tracking algorithm has been fully implemented in matlab .
figure [ figcamera_detection ] shows the successive steps to get a measure of the crack length .
though pixel size corresponds to @xmath26 on the sample , the position of the crack front in the long dimension of the sample is known to about @xmath27 , and is dominated by the nonlinear shape of the crack front .
in addition , our tracking algorithm assumes that the fracture strictly propagates in the longitudinal plane of the sample .
however , certain cracks tend to slightly twist during their propagation , meaning the true crack length may be underestimated by up to 10% . during fracture tests ,
the frames generated by the camera have to be synchronized with the other signals ( acoustic , light , force and displacement signals ) .
flashes of light are generated using the 980 nm laser - diode to mark the time on the camera .
this is done manually by closing the circuit feeding the diode by tapping it against the vacuum chamber which is the ground of the circuit .
these taps also propagate acoustic waves through the setup to the acoustic sensors .
the propagation delay is quite negligible compared to the exposure time of the camera ( 0.2 s ) because the acoustic waves go through approximately @xmath28 of aluminum at a speed of @xmath29 which gives a delay of only @xmath30 .
we have verified that any drift between the camera and the data acquisition system remains shorter than the duration of a frame on tests lasting up to 2 hours . in order to obtain spectroscopic information on the emitted light , at the expense of any time resolution ,
the setup can be optically coupled to a spectrometer .
this is done by replacing the photomultiplier by an optical fiber leading to a horiba micro hr monochromator coupled to a liquid - nitrogen cooled horiba symphony ii ccd .
the small solid angle under which the scintillator sees the optical fiber means that light collection efficiency is much lower than with the photomultiplier .
the experimental protocol is to use an exposure of a few seconds , during which a sufficiently large force is applied to break the sample completely .
an example of a spectrum is shown in fig .
[ figspectro ] .
the spectrometer was previously calibrated in terms of wavelengths using a mercury lamp to within 2 nm , but the spectrum amplitudes are not corrected for the various efficiencies .
as described before , our study of fracture involves the use of multiple sensors : a photomultiplier , two acoustic sensors , a force gauge and a displacement sensor .
a data acquisition system has been developed to simultaneously digitize the signals generated by these sensors at drastically different sampling rates , and to control the force applied on the sample in real - time . since fracture is a one - shot process , fracture tests imply to continuously digitize all the channels , without any dead time , for hours .
two high - speed digitizers manufactured by cronologic are used .
the ndigo5 g is a 5 giga - sample - per - second ( gsps ) analog - to - digital converter ( adc ) with a 10 bit resolution and the ndigo250 m is a 250 mega - sample - per - second ( msps ) adc with a 14 bit resolution .
these digitizers are designed to be plugged on pcie slots for a high throughput .
the ndigo5 g is used to record the light signal at full speed since single photoelectron pulses are a few ns long .
the ndigo250 m is used to continuously record the slower acoustic , force and displacement signals at a reduced speed of 2 msps .
two important features of these digitizers are the synchronization , done internally , and the onboard zero suppression . on the ndigo5 g
, the onboard field programmable gate array suppresses noise between events ( see sec . [
sec : lightchannel ] ) , which significantly reduces the load on the downstream software and hardware . in a typical fracture experiment ,
less than @xmath31 of the light data are actually relevant , and then , automatically recorded by the data acquisition system . without this zero suppression , continuously sampling the light channel at 5 samples per nanosecond over an hour , would by itself lead to over 20 tbytes of data , straining data storage and slowing offline data reduction .
the architecture of the data acquisition system is shown in figure [ figdigitizers ] .
different software tools have been developed in c++ using boost and qt .
they allow to record the data generated by the digitizers , display some signals in real - time , and regulate the force applied on the specimen .
two computers are needed to perform a fracture test .
the ndigo5 g and the ndigo250 m are plugged into computer 1 .
the digitizing server on computer 1 collects all the data generated by both digitizers and saves them in binary files on two solid - state drives ( samsung ssd pro series ) .
the ndigo250 m is configured to create only 4 kib packets of data .
this size of packets allows a fast treatment by the drives .
the size of data packets for the light channel strongly depends on the light signal dynamic , but light data are saved only when a sufficient amount of data is collected to minimize the number of calls to the drives .
large buffers in the digitizer server are set in order to absorb punctual data floods caused by strong light emissions , and then spread over time the load on the solid - state drives . the data from the ndigo250 m are also sent to computer 2 over ethernet using the user datagram protocol ( udp ) .
it is a communication protocol designed to be minimalist , it does not guarantee that data will be delivered because packets are sent without prior communications , but it reduces the latency compared to connection - oriented communication protocols .
another advantage is that it makes computer 2 fully independent of computer 1 .
computer 2 collects the data from ndigo250 m received over ethernet .
the force signal is extracted to regulate the force applied on the specimen using a proportional - integral - derivative controller which commands the motor setpoint .
acoustic , force and displacement signals are sent to matlab where they are displayed in order for the user to monitor fracture tests in real - time .
the principle of the light acquisition is shown in figure [ figtriggering ] .
each photon detected by the photomultiplier produces a negative pulse on its output .
an offset is added to the signal to use the full range of the digitizer .
onboard zero suppression allows to only record the photons .
when the signal goes below a threshold , a trigger occurs on the digitizer . for each trigger
, a predefined number of signal samples before and after the trigger ( pre - trigger and post - trigger , respectively ) are grouped in a data packet and saved on the drives . if the signal is still below the threshold at the end of the post - trigger , the post - trigger is lengthened and the extra data samples are added to the same packet . packets also contains information such as timestamp , number of data samples and flags ( errors or warnings during the acquisition ) .
on the drives , only the signal samples containing useful information are then saved thanks to this onboard zero suppression .
when ionizing particles interact in a scintillator such as bgo , they deposit a certain amount of energy .
some of this energy is reemitted as visible photons .
the light yield of scintillators tells how many photons are actually emitted per unit of deposited energy . for @xmath32-ray sources ,
the light yield of bgo is 8 photons / kev @xcite . for a single @xmath32 interaction ,
the time distribution of emitted photons generally follows one or more exponential decays . in the case of bgo ,
the main time constant is @xmath33 ns @xcite . in @xmath34 , @xmath35 of the photons are emitted . in other scintillators , or at other temperatures ,
the time distribution of photons can be more complex @xcite .
the group of photons emitted by an interaction is called a scintillation event . to identify the scintillation events in the data , a sliding integral
is computed . for each saved signal sample ,
all the preceding signal samples within @xmath34 are integrated .
thus , an integral is calculated for each saved signal sample , called signal integral .
the analysis is described in figure [ figintegrationlightevent ] . to obtain energy spectrum . ] to identify scintillation events on the signal integral , a condition on the distance between two successive photons is used .
if two successive photons are separated by more than @xmath36 ( called waiting time in figure [ figintegrationlightevent ] ) , two different scintillation events are considered . in the case where two scintillation events are very close to each other ( successive photons separated by less than @xmath36 ) , a single scintillation event is considered .
if this method is compared to the detection of local maxima on the integral signal ( using a schmitt trigger for example ) , it is not required here to set threshold(s ) on the integral signal , which might depend on the photomultiplier and its voltage . for each scintillation event , the maximum integral is measured .
it represents the maximum integral on a sliding window of @xmath34 .
one of the first steps that needs to be carried out is determining the response of the system to single photons .
to do so , a data acquisition as described previously can be performed with the photomultiplier in a dark chamber .
it is then not excited by any light source but room - temperature thermal fluctuations eject some electrons from the photocathode .
these electrons generate pulses similar to pulses generated by single incident photons .
these pulses can be easily detected as separated events by the previous analysis and their integral can be calculated .
figure [ spe ] shows the distribution of the integrals of the events from which the typical value is taken as the central value ( here , 22,480 a.u . ) of a gaussian fit over part of the spectrum .
the typical rate of dark counts is a few hundred hertz .
next , once the scintillator is placed in the setup , and before it is fractured , the light collection efficiency of the configuration can be determined , and the light channel calibrated .
a calibration using two @xmath32-ray sources , ( mainly 122 kev photons ) and ( mainly 662 kev photons ) , is performed to determine a relation between the number of photons measured by the photomultiplier and the number of photons actually emitted by the sample .
the radioactive sources deposit a known energy in the sample ; therefore using the light yield , one can determine the number of photons emitted .
the light collection efficiency is determined by comparing this number to the one obtained by building the spectrum of the detected integrals , and scaling it to the response to single photelectrons .
figure [ figcs ] shows typical calibration spectra that we get during an experiment .
thanks to the average integral of the single photoelectron obtained with figure [ spe ] , we can deduce the number of detected photoelectrons for a given event integral .
thus , for the @xmath37cs full collection peak of 662 kev , a gaussian fit over part of the peak gives an average event integral of @xmath38 photoelectrons that are detected .
based on the light yield of 8 photons / kev for bgo @xcite , the light collection efficiency can be determined as @xmath39 .
since the light emission spectra during scintillation and fracture are identical @xcite , the relation can be extended to the energy , so the calibration gives the ratio between the measured energy and the actual energy radiated by the specimen . during the fracture measurement itself
, the photon rate is computed .
this is achieved by integrating the light signal over a sliding window .
the calculated integrals are converted into a number of emitted photons , using the light calibrations performed before the fracture test . by dividing the number of photons by the width of the sliding window ,
the photon rate is thus determined .
the typical width for the sliding window is about one second .
this duration is a compromise , as shorter windows generally suffer from statistical fluctuations , whereas longer ones do not allow to precisely observe sharp changes in the photon rate . the two acoustic sensors ( sec .
[ sec : acousttrans ] ) produce analog signals which are run through a 10 khz1 mhz bandpass filter before being continuously recorded at 2 msps , as shown in figure [ aes]a .
since the acoustic sensors are not very efficient below 100 khz , a 6th - order butterworth high - pass filter with 100 khz cut - off frequency is used offline .
it suppresses environmental noises ( figure [ aes]b ) . after this filtering
, well - separated events are determined when the power of the signal overcomes a threshold above the baseline ( figure [ aes]c ) .
the power of the data at each instant is computed as the average of the squares of the signal samples in the previous 0.5 ms .
next , at each point in time , the average ( @xmath40 ) of the power , and its standard deviation ( @xmath41 ) are both computed over the previous second .
at each point in time , an adaptive threshold is then defined as @xmath42 .
this adaptive threshold is designed to rise when the data become noisier , avoiding spurious events .
when the data surpass the threshold , the value of the threshold is frozen until the data drop again beneath it , and computation of the threshold resumes .
individual events are defined over the interval during which the power is above the threshold . at the end of this process
, all the individual events are saved to another file .
operation of the setup described above requires several steps to achieve controlled propagation of a fracture in a sample . the sample , cut into the geometry described in sec .
[ dcdcgeo ] , requires extra preparation .
first , the edges in contact with the backstop and the ball joint are sanded down , which avoids unwanted microcracks at the sharp edges during compression . then , using a surgical blade , small notches are made on either side of the hole and on each face of the sample , in the direction in which the fracture will propagate .
these artificial flaws will help to initiate the cracks .
next , the sample is cleaned , and the acoustics sensors are glued as described in sec .
[ sec : acousttrans ] .
once the crystal is prepared , it is carefully positioned in the setup between the ball joint and the backstop .
it must be precisely centered on the loading axis in order to avoid deviations of the crack from the nominal propagation plane .
the specimen is also tilted so that the infrared camera can monitor the whole crack propagation .
next , the enclosure is closed which allows to control the gases present during the fracture test and their pressures .
the sample is illuminated with infrared light and observed by the camera .
the data acquisition system is then started , which records all the sensors : load cell , displacement sensor , acoustic sensors , photomultiplier and camera .
the sample is then ready for a controlled crack growth experiment .
a typical fracture test at ambient pressure with a bgo sample is presented in figure [ bullegaryplot ] .
an initial force of @xmath43 n is applied .
the force is slowly increased at @xmath44 n / s up to @xmath45 n ( around @xmath46 s ) , and is then kept constant by the regulation system ( fig .
[ bullegaryplot]a , pink curve ) .
the opposite of the displacement , measured by the capacitive sensor ( fig .
[ bullegaryplot]a , purple curve ) , gives the variation of sample compression .
the compression increases by about @xmath47 @xmath48 during the force ramp .
afterwards , the displacement decreases slowly except for a jump that occurs at @xmath49 s and that is associated to fast and full rupture of the sample into two halves .
the propagation of the two opposite cracks observed by the infrared camera is shown in figure [ bullegaryplot]b .
the center of the sample is chosen as origin , hence the crack length only starts at a distance of @xmath44 mm which is the edge of the central hole . at @xmath50
s , a jump in crack growth is observed , corresponding to the reopening of the crack which had been initiated previously and allowed to close .
then , while the force is kept constant , a slow growth of the crack is observed up to a critical length ( near @xmath51 s ) when it accelerates and cleaves the material in two .
light and acoustic emissions are shown in figure [ bullegaryplot]c .
acoustic emissions are correlated with the whole crack propagation . in this setup ,
light emissions are prominently observed at the moment of fast fracture .
these correlations confirm our previous observations @xcite .
full analysis of these results , and a host of experiments , are underway .
in order to study fractures in scintillating materials , we have developed a novel multi - channel setup under controlled atmosphere . using the dcdc geometry for the sample ,
our setup allows the simultaneous measurement of the force applied to the sample , the compression , the acoustic and visible light emissions , all the while filming the propagation of the fracture under infrared lighting .
the data acquisition system enables continuous recording of all channels , allowing analysis of rupture dynamics over extremely large time and energy ranges .
for instance , nanosecond timing is achieved for the light channel , with onboard zero supression to limit the amount of data stored .
various loading profiles can be applied , allowing the study of different fracture regimes .
thanks to this setup , we will be able to precisely explore light emision during the subcritical regime in particular .
we also plan to study effects such as that of the atmosphere on the fracture and light emission .
lastly , our setup is not restricted to scintillators , but could be used for other materials such as pmma and glass @xcite .
we thank gary contant and chuck hearns ( queen s university ) for contributing to the mechanical development , and jean - michel combes and franois gay ( universit de lyon ) who developed the acoustic amplifiers and participated to the electronic design .
this work has been funded in canada by nserc ( grant no .
sapin 386432 ) , cfi - lof and orf - sif ( project no .
24536 ) , and by the france - canada research fund ( project `` listening to scintillating fractures '' ) .
alexis tantot has been supported by an exploradoc grant from rhnes - alpes ( france ) .
10 m. v sivers , m. clark , p. c. f. di stefano , a. erb , a. gtlein , j .- c .
lanfranchi , a. mnster , p. nadeau , m. piquemal , w. potzel , s. roth , k. schreiner , r. strauss , s. wawoczny , m. willers , and a. zller .
low - temperature scintillation properties of cawo4 crystals for rare - event searches . , 118(16):164505 , october 2015 . | to investigate fractoluminescence in scintillating crystals used for particle detection , we have developed a multi - channel setup built around samples of double - cleavage drilled compression ( dcdc ) geometry in a controllable atmosphere .
the setup allows the continuous digitization over hours of various parameters , including the applied load , and the compressive strain of the sample , as well as the acoustic emission .
emitted visible light is recorded with nanosecond resolution , and crack propagation is monitored using infrared lighting and camera .
an example of application to ( bgo ) is provided .
_ keywords _ : brittle fracture , fractoluminescence , scintillators + |
kj would like to thank a. n. korotkov for helpful discussions .
this work was supported by the hearne institute for theoretical physics , the army research office and the disruptive technologies office .
eyal buks , eran arbel - segev , stav zaitsev , baleegh abdo , and m. p. blencowe .
quantum nondemolition measurement of discrete fock states of a nanomechanical resonator .
eprint : arxiv : quant - ph/0610158 , 2006 . | we describe a nano - electromechnical system that exhibits the `` retroactive '' quantum jumps discovered by mabuchi and wiseman [ phys .
rev .
lett .
81 , 4620 ( 1998 ) ] .
this system consists of a cooper - pair box coupled to a nano - mechanical resonator , in which the latter is continuously monitored by a single - electron transistor or quantum point - contact .
further , we show that these kinds of jumps , and the jumps that emerge in a continuous quantum non - demolition measurement are one and the same phenomena .
we also consider manipulating the jumps by applying feedback control to the cooper - pair box . in 1998
mabuchi and wiseman discovered the existence of emergent jumps in purely diffusive , measured cavity qed system , and the termed the phenomena `` retroactive '' quantum jumps @xcite . since that time it has also been discovered that quantum jumps will emerge in a noisy system subjected to a continuous quantum non - demolition ( qnd ) measurement .
examples are the continuous measurement of the energy of a two - level system , such as a cooper - pair box ( cpb ) @xcite , and the measurement of the energy of a nano - mechanical resonator @xcite . our main purpose here is to show that the jumps discovered by mabuchi and wiseman , and the jumps in a qnd measurement are , in fact , the same phenomena having the same underlying origin . while there was no explicit qnd measurement in the cavity qed scenario of mabuchi and wiseman , the `` retroactive '' jumps are nevertheless due to a qnd measurement that arises as a result of the interaction between two subsystems .
this is a qnd measurement on one subsystem mediated through the other subsystem .
these jumps may therefore be described as arising from an _ indirect _ qnd measurement , rather than a direct one .
thus , if one interprets the cqed jumps of mabuchi and wiseman as retroactive , then the jumps induced by all qnd measurements , whether on quantum or classical systems , are retroactive in the same sense . in the process of our analysis
will also do a number of other things .
we first show how to design a nano - electro - mechanical system so that jumps of the `` retroactive '' type emerge in the diffusive motion .
this system consists of a nano - mechical resonator linearly coupled to a cpb @xcite , where the position of the resonator is continuously measured .
this measurement can be implemented using a single - electron transistor ( set ) @xcite or a quantum point contact ( qpc ) @xcite .
the dynamics of this system are not identical to the cavity - qed system analyzed in @xcite , and this difference makes it easier to see the connection with the usual direct qnd measurement . we also show that it is possible to effect a degree of control over the jumps by applying feedback control to the cpb .
if one places a cpb adjacent to a nano - resonator , and places a voltage on the nano resonator , then the resonator feels a linear force from the cpb , and this depends upon which state the cpb is in .
( that is , whether or not the cpb contains an excess cooper - pair ) .
the hamiltonian describing the dynamics of the coupled systems is @xcite @xmath0 .
\label{ham}\ ] ] here @xmath1 is the annihilation operator for the resonator , which has angular frequency @xmath2 .
the frequency corresponding to the charging energy of the cpb is @xmath3 , and the josephson tunneling frequency is @xmath4 .
the tunneling term allows us to perform @xmath5 rotations using voltage pulses , but otherwise can be ignored since we set @xmath6 .
the interaction between the resonator and the cpb is the linear force that the resonator feels from the charge on the cbp , and the strength of this force determines @xmath7 .
the expression for the above rate parameters in terms of the physical configuration of the resonator and cpb may be found , for example , in @xcite . as indicated above , the position of the resonator can be monitored continuously using an set or qpc .
we model this measurement process as an inefficient , but otherwise ideal , continuous position measurement @xcite .
the continuous stream of measurement results ( often referred to as the measurement record ) is @xmath8 where @xmath9 and @xmath10 is an increment of gaussian white noise @xcite . here
@xmath11 is a measure of the rate at which information is extracted from the system , and which we will refer to as the measurement strength .
the parameter @xmath12 is called the efficiency of the measurement .
the interaction of the system with its environment ( including the set ) continually carries information away from the system ( at a rate proportional to @xmath11 ) , and @xmath12 gives the fraction of this information which is actually collected by the observer . the resulting dynamics of the system density matrix , @xmath13 , is given by the stochastic master equation ( sme ) @xcite @xmath14 dt - k [ x,[x,\rho ] ] dt \nonumber + \sqrt{2 \eta k } ( x \rho + \rho x - 2\langle x\rangle \rho ) dw \label{msme}\ ] ] where @xmath15 is given by eq.([ham ] ) above and @xmath16 is the position operator for the resonator .
as such , @xmath11 has units of @xmath17 , and so we define the corresponding dimensionless rate @xmath18 .
we will consider two modifications of the above basic dynamics .
the first is that we will modulate the interaction strength @xmath7 between the resonator and the cpb at the resonant frequency of the resonator , so that @xmath19 .
this can be done by varying the voltage on the resonator .
the result is to allow the cpb to drive the resonator at its resonant frequency , generating the maximum steady - state displacement of the resonator .
the second modification is the application of a real - time feedback loop @xcite to damp the motion of the resonator .
this means that the observer continually applies a force @xmath20 to the resonator , where @xmath21 $ ] is the observers maximum likelihood estimate of the momentum of the oscillator at each time @xmath22 .
the result of this is to apply a ( somewhat noisy ) frictional damping force to the resonator .
we must also include in the dynamics the effects of temperature on the resonator .
the resonator is in contact with a thermal bath that induces damping and injects noise into the resonator .
since the quality factor of the resonator is above @xmath23 , the thermal damping is much smaller than the damping that will be induced via the feedback loop , and as a result we simply subsume this damping into the feedback .
the thermal noise can be taken into account by choosing an appropriate value for the efficiency @xmath12
@xcite .
the noise introduced by the ( low temperature ) thermal bath is just as if a position detector was carrying information away at the rate @xmath24 @xcite , where @xmath25 is the thermal damping rate and @xmath26 is the temperature . to include the thermal noise one therefore replaces @xmath11 in eq.([msme ] ) with @xmath27 , and @xmath12 with @xmath28 where @xmath12 is the set measurement efficiency .
the final thing to include is the environmental noise on the cpb .
the kind of noise that is of interest to us is noise which causes diffusion between the two energy eigenstates .
it is this noise which , coupled with the system s dynamics , induces the quantum jumps .
thermal noise is of this type , and is usually modeled with the master equation @xmath29 + \xi { \cal d}[\sigma_+ ] \ } \rho $ ] , where @xmath30 , and @xmath31\rho \equiv [ a^\dagger a,\rho]_+ - 2 a \rho a^\dagger$ ] for any operator @xmath1 .
however , the charging energy of a cpb is usually chosen so that @xmath32 . in this case
the thermal noise will only cause the upper state to decay , rather than generate diffusion between the two .
nevertheless there are other ways to induce the desired diffusion .
one is to apply a stochastic sequence of pulses to a voltage gate , where the pulses bring the cpb to the degeneracy point .
if the random pulse lengths are short compared to @xmath33 , then the josephson tunnelling term generates an evolution described by a hamiltonian @xmath34 , where @xmath35 is white noise with autocorrelation @xmath36 .
the result is the master equation @xmath37 \rho$ ] .
this master equation closely emulates the thermal master equation in the limit in which @xmath38 , the thermal noise term becomes @xmath39 .
since the system hamiltonian is approximately symmetric in @xmath16 and @xmath40 , the effect on the system is equivalent to @xmath41 . ] .
thirdly , one could instead increase the josephson term .
due to the measurement dynamics to be discussed later , this should have the same effect as thermal noise . in our numerical simulations we choose to explicitly model the second noise source discussed above , that of stochastic driving proportional to @xmath5 as a result , the full dynamics of the resonator - cpb system is @xmath42 dt -k_{\mbox{\scriptsize tot } } [ x,[x,\rho ] ] dt \nonumber \nonumber \\ & + & \kappa { \cal d}[\sigma_x ] \rho dt + \sqrt{2 \eta_{\mbox{\scriptsize tot } } k_{\mbox{\scriptsize tot } } } ( x \rho + \rho x - 2\langle x\rangle \rho ) dw \end{aligned}\ ] ] following the analysis of the cavity - qed system by mabuchi and wiseman
, we now examine the steady - state of the hamiltonian @xmath43 including the feedback damping , when the cpb is in either of it s energy eigenstates .
we will denote these eigenstates by @xmath44 they correspond to the presence or absence of a cooper - pair in the box .
in these two cases the resonator has the effective hamiltonian @xmath45 where @xmath46 is the maximum value of the driving force . since the hamiltonian is linear ,
the dynamics of the expectation values of @xmath16 and @xmath47 are simply those for the equivalent classical system , namely that of a driven , damped harmonic oscillator . as a result , the steady - state solution for @xmath48 is @xmath49 where @xmath50 is the mass of the resonator .
the state of the resonator in each case is a ( somewhat squeezed ) gaussian state .
thus , like the cavity - qed system in @xcite , our system has two stable orthogonal states , each of which is a product of a distinct state of the two - level system with a distinct gaussian state of the resonator . in our case
it is the steady - state phase of the resonator , @xmath51 , that is tightly correlated with the orthogonal states of the cpb . the measurement of the position of the oscillator continually provides the observer with information regarding the location of the oscillator in phase space , and thus about the phase of the oscillations .
since this phase is correlated with the eigenstates of the oscillator , the measurement will tend to continually collapse the state of the cpb to one of its eigenstates . because of this it is only the two eigenstates that are stable against the measurement process , and this is the reason that the two steady - states given by eq.([ssx ] ) are important
if the environmental noise is to induce jumps in the system it will be between these two stable states .
we now simulate the full dynamics of the observed nano - electro - mechanical system , including the environmental noise . in specifying values for the system parameters
, we will quote all rate constants in terms of the frequency of the resonator @xmath52 .
we set the interaction strength @xmath53 and the feedback damping rate @xmath54 , both of which are easily achievable @xcite .
we set the set measurement strength at @xmath55 , which is also not difficult to achieve , certainly with @xmath56 as high as @xmath57 @xcite .
we find from our simulations that relatively high efficiency ( @xmath58 ) is required for the observer to effectively track the quantum jumps .
this requires that the set have high efficiency , and that @xmath59 .
the question is still open as to whether such an efficiency can be reached with an set @xcite , but it is estimated that a qpc should in theory be able to achieve efficiencies above @xmath60 . setting @xmath61 , and using the parameters in @xcite , gives @xmath62 .
thus a factor of 20 increase in @xmath11 from that configuration would be required . while we would expect this to be possible ,
the overall efficiency requirement is the most challenging in the scenario . finally , we choose the cpb noise strength to be @xmath63 . in figure [ fig1](a )
we show the evolution of the phase of the resonator , @xmath64 , which we define by the relation @xmath65 . here
@xmath66 is as defined above , and @xmath67 .
we start the resonator in a coherent state with @xmath68 and @xmath69 , so that the initial phase is zero .
we see that the phase quickly drifts to one of the two values @xmath70 , and from then on exhibits jumps in the motion between these values . in figure [ fig1](b )
we show the position of the resonator as a function of time .
the amplitude of the position oscillations tends to reduce during phase flips , as expected . , and
for ( c ) @xmath71.,width=321 ] one of the most interesting aspects of these quantum jumps is that they are an emergent phenomenon ; the underlying dynamics does not contain jumps but consists purely of continuous diffusion .
the authors of @xcite explain this emergence by providing a detailed analysis of the interplay of the correlations produced by the hamiltonian dynamics , the measurement induced localization , and the diffusive noise . their analysis is certainly not wrong .
our point here is that the mechanism that produces jumps in this case is the same mechanism that produces jumps in the case of a continuous qnd measurement .
this provides us with a somewhat simpler picture of the cause of the `` retroactive '' jumps .
qnd measurements are measurements in which the observable being measured is not changed by the dynamics of the system ( that is , the observable commutes with the hamiltonian ) @xcite . as a result , once the measurement has projected the system onto a given eigenstate of the observable , it remains there throughout the remainder of the observation period .
now consider what happens when the observable is additionally subject to diffusion from environmental noise . in the absence of the measurement
the noise will cause the observable to diffuse from one eigenstate to another , but in the presence of a sufficiently strong continuous qnd measurement the dynamics is quite different . consider what happens during a small time interval when the system begins in one eigenstate . during the interval
the diffusion will generate a small probability that the system is in an adjacent eigenstate .
however , during the same interval the measurement will collapse the system to one of the eigenstates , and with very high probability this will be the initial eigenstate , since the diffusion has only managed to generate a small probability for the other eigenstates during the short time interval . as a result , it is only unlikely events , in which the noise has a particularly large fluctuation , and the measurement conspires by returning a low probability result , that will cause the system to transition from one eigenstate to the next .
the result is periods in which the system remains in a given eigenstate of the qnd observable , interspersed by quantum jumps between the eigenstates .
the jumps are _ quantum _ jumps since they only appear because the observable has a discrete spectrum .
this qualitative picture has been confirmed by numerical simulations of a measurement of the energy of a quantum harmonic oscillator @xcite , and the states of an electron in a coupled pair of quantum dots ( often referred to as a `` charge qubit '' ) @xcite .
the jumps discovered in @xcite , and those in the dynamics here , can now be seen to be caused by the same effect , except in that in these two cases the qnd measurement is mediated through a second system . in our case
the qnd observable is the energy of the cpb .
the interaction between the resonator and the cpb causes the phase of the resonator to be tightly correlated with the energy eigenstates of the cpb , as shown in eq.([ssx ] ) . in continually providing information about the phase of the resonator
, the position measurement necessarily provides information about the energy of the cpb , generating a qnd measurement , and resulting in quantum jumps .
interestingly it is not necessary to perform a qnd measurement on the mediating system ( the resonator ) to generate the jumps ; position is not a qnd observable for the resonator .
we now consider the use of feedback control to stabilize the system against the jumps . in doing so we reveal a fundamental limitation of feedback control in this context - hamiltonian feedback can not stop the system from jumping .
the reason for this is quite simple as discussed above , the jumps from @xmath72 to @xmath73 , for example , are due to the small probability for @xmath73 continually generated by the noise process . to generate this probability
the noise process merely reduces the length of the bloch vector . since the cause of the jumps is purely this reduction ( along with the qnd measurement ) , and
since hamiltonian evolution , feedback or otherwise , can rotate the bloch vector but can not change its length , feedback is powerless to prevent the jumping . however , feedback can be used to _ increase _ the rate of jumps from either state , thereby altering the relative time that the system spends in the two states .
ultimately this could be used to ensure that the system spends almost all its time in one of the states , creating an effective stability .
we implement this procedure by applying a feedback hamiltonian of the form @xmath74 to the cpb , where @xmath75 , so that @xmath76 determines the feedback strength . in each time step @xmath77
, we choose this hamiltonian so as to rotate the system towards the state @xmath72 .
we set @xmath78 , and find that @xmath71 is required to obtain a significant effect .
we plot the resulting evolution in fig [ fig1](c ) , for the same noise realization as before .
this shows that the phase is now more stable at @xmath79 than @xmath80 . |
the game of septoku is a sudoku variant played on a grid of hexagons .
the puzzle was invented in 2006 by bruce oberg @xcite .
septoku is played on a hexagonal board of side 4 with 37 cells , as shown in a sample puzzle in figure [ fig1]a @xcite .
we label the cells on the board by numbering them sequentially ( figure [ fig1]b )
. we will refer to a * row * of the board to include not only the horizontal rows , but also the
diagonal rows " parallel to the other sides of the board .
for example the cells numbered @xmath0,@xmath1,@xmath2,@xmath3 form one row , as do the cells @xmath0,@xmath4,@xmath5,@xmath6 and @xmath7,@xmath8,@xmath9,@xmath10,@xmath11 .
we will abbreviate these three rows as row 1@xmath124 " , row 1@xmath1216 " and row 3@xmath1228 " .
the puzzle is to fill the 37 cells with seven symbols ( the numbers 1 to 7 ) such that each row contains different symbols .
there are seven rows in each of three directions , so 21 rows in all .
note that most rows contain _ less _ than 7 cells .
the seven * * circular regions * * ( with grey centers ) must each contain all seven symbols .
for example the cells @xmath13 must contain all of the numbers 1 to 7 .
we will refer to a circular region by its center , i.e. circle 6 " refers to the cells @xmath0 , @xmath1 , @xmath4 , @xmath14 , @xmath15 , @xmath16 , @xmath17 . as a visual aid in all large diagrams , each circular region
is marked by a grey circle .
in general , any subset of the board that must contain distinct symbols will be called a * region*. a puzzle is created using a blank board with a number of seeds filled in ( figure [ fig1]a ) .
the puzzle is to fill out the rest of the board so the above rules are satisfied , such a completed board will be called a * valid * board .
a well - crafted puzzle has only one solution .
bruce oberg has created several pages of enjoyable septoku puzzles of varying difficulty starting from 7 to 13 seeds @xcite . in sudoku , each cell ( usually called a square in this case ) is in exactly three regions , a ( normal ) row , column , and @xmath18 block .
there are nine rows , nine columns , and nine @xmath18 blocks , so the total number of regions is 27 . in septoku , the total number of regions is 28 ( seven rows in each of three directions , plus seven circles ) , even though the board is less than half the size of a sudoku board .
each cell is in at least four regions and sometimes even five .
not all regions contain all seven symbols , but we shall see that a septoku solution is more constrained and there are fewer solutions .
a valid septoku board must also satisfy certain additional constraints that are not at all obvious .
in fact , we will see that the above rules above are so stringent that there are only six valid boards , up to reflections and rotations of the board , and permutations of the seven symbols .
* theorem 1 * : in a valid septoku board , each circle center @xmath19 must have a unique symbol .
: suppose otherwise
. then there must be some symbol that occurs twice among the circle centers .
the center symbol is clearly unique by the rules of the puzzle , so without loss of generality , assume a 1 appears in cells 6 and 21
. there must be a 1 in the center circle 19 , and this can only be in cell 13 or 25 .
but if it is in 25 , then circle 8 can contain no 1 at all .
thus there must be a 1 in cell 13 .
there must be a 1 in circle 17 , and this can now only be in cell 24 , and a 1 in circle 32 , which can only be in cell 31 .
but now this is not a valid septoku board , because circle 30 contains two 1 s .
* theorem 2 * : in a valid septoku board , each corner @xmath20 plus the center @xmath21 must have a unique symbol .
: suppose otherwise .
then there must be some symbol that occurs twice among the corners ( the center symbol is clearly unique by the rules of the puzzle ) . without loss of generality , assume a 1 appears in cells 1 and 22 . by theorem 1
there must be a 1 in some circle center , and this can only be in cell 8 or 30 . in either case , there can no longer be a 1 in the center circle 19 , so this is not a valid septoku board .
* theorem 3 * : in a valid septoku board , each symbol must appear at least five times .
: suppose the symbol 1 appears four or fewer times . by theorems 1 and 2
each symbol must appear in a circle center , and in a corner ( or once in the center , 19 ) .
if 1 occurs in a corner and one circle center , then this covers only two of the circles out of seven .
there is no way the last two 1 s can cover the remaining five circles , since one cell is in common with at most two circles .
it remains a possibility that the 1 is in the center , cell 19 .
however it is not possible for the three other 1 s to cover the six remaining circles without two being in the same row . if each symbol occurs 5 times this covers only 35 of the 37 board locations , so at least one symbol must occur more than 5 times .
in fact there are only two possibilities : six of the symbols appear 5 times , one symbol appears 7 times .
five of the symbols appear 5 times , two symbols appear 6 times .
we shall now see that both of these possibilities can occur .
* theorem 4 * : not counting reflections , rotations , and symbol permutations , there are only * six * valid septoku boards . : by applying a symbol permutation , any valid board has a version where the center looks like figure [ fig2 ] .
the reason why this configuration was chosen is that the symbol permutation @xmath22 rotates the pattern in the center circle counter - clockwise by 60 degrees , and the permutation @xmath23 does a clockwise rotation describes the permutation which sends @xmath24 , @xmath25 , @xmath26 , @xmath27 , @xmath28 , @xmath29 and @xmath30 . ] . from theorem 1 we know that the symbol 1 must appear in some circle center .
it suffices to consider only the cases where this 1 is in cell 8 or 21 . for
if the 1 is at cell 17 or 30 we reflect the board across row 1@xmath1237 , and then apply the permutation @xmath31 to bring it into standard form with a 1 at cell 8 or 21 . from theorem 2
we know that the symbol 1 must appear in some corner , and thus all possible septoku boards have a standard form that looks like one of the four in figure [ fig3 ] .
therefore , if we find all solutions to the four puzzles above , we will find all possible septoku boards .
this may seem difficult , but actually there are not many solutions .
* proposition 1 * : puzzle 4 is unsolvable . :
circle 17 must contain a 1 , but there is no cell where a 1 can be placed without producing two 1 s in some row .
* proposition 2 * : puzzles 1 - 3 each have 4 solutions . :
it is probably possible to prove this analytically , but we use an integer programming model to make sure we have found all solutions . we define an unknown binary @xmath32 matrix @xmath33 as : @xmath34 .constraints for the integer programming ( ip ) model . [ cols="^,^",options="header " , ] @xmath33 must satisfy the constraints in table [ table1 ] .
there are several programming languages for solving such problems , we use gams ( general algebraic modeling system ) .
the details of the model are in the gams input file in appendix a. a specialized solver is needed to solve integer programming problems .
fortunately , a free solver is available on the web , the neos server @xcite .
you can submit the gams model in appendix a using the web form .
the model will be solved ( in under 1 second ) and the results will be emailed to you . the solver will return a single solution , if one exists . to get additional solutions ,
we add another constraint that eliminates the solution that was just found . after doing this four times , the solver reports that no more solutions exist . for the four puzzles ,
the solutions returned by the gams solves are shown in figure [ fig4 ] .
since there are 12 solutions to puzzles # 1#4 , one might think there are exactly 12 septoku boards .
however , by applying rotations , reflections , and permutations , we can show that some boards can be converted into others .
we define : @xmath35 as the transformation that rotates the board clockwise @xmath36 .
therefore @xmath37 is a rotation counter - clockwise by @xmath36 .
@xmath38 as the reflection of the board across the @xmath39-axis ( row 16@xmath1222 )
. then we have the following identities between boards : @xmath40 thus , there are only six boards ( figure [ fig5 ] ) .
three of them contain five of each number except for 3 and 6 , these are boards @xmath41 , @xmath42 and @xmath43 in figure [ fig5 ] . boards @xmath44 , @xmath45 and @xmath46 in figure [ fig5 ] have five of each number , except for the central number 7 , of which there are seven
. the number 7 appears in each of the 28 regions .
finally , how do we know that none of the six boards in figure [ fig5 ] can be transformed into one another by applying reflections , rotations , and permutations ( which would mean there are fewer than 6 solutions ) ?
for example , how do we know that we ca nt apply some symmetry transformation @xmath47 followed by a permutation @xmath48 to convert board @xmath41 into @xmath42 ?
@xmath49 we note that no symmetry transformation changes the value of the center , so for equation ( [ eq : atob ] ) to hold , @xmath48 must leave 7 s alone .
therefore , @xmath47 must map 7 s on one board to 7 s on the other .
it is impossible for any board in @xmath50 to be transformed into one of @xmath51 , as they contain different numbers of 7 s .
as far as transforming @xmath41 to @xmath42 , there are only two rotations of @xmath41 that produce the pattern of 7 s in @xmath42
. the center circle must be the same in both boards , this fixes the permutation @xmath48 , and by direct calculation we find that neither of these transformations of @xmath41 yield @xmath42 .
similarly , it is time consuming but not difficult to verify that none of the six boards can be transformed into any of the others .
we also verified this using a computer , applying each of the 12 symmetry transformations , and then applying each of the @xmath52 permutations in turn .
a program that did this found that there was no way to transform a board of one of the six fundamental types @xmath41-@xmath46 into any of the others ( this program also identified @xmath53 , @xmath54 , @xmath55 , @xmath56 , @xmath57 , @xmath58 ) .
* theorem 5 * : counting reflections , rotations , and symbol permutations as separate boards , there are 120,960 valid septoku boards . :
each solution board has @xmath59 different versions using all possible permutations of symbols .
these boards are all distinct , since each permutation is distinct .
we can then apply the 12 symmetry transformations to generate @xmath60 board positions . however , not all of these will be distinct .
if @xmath47 is a symmetry transformation and @xmath48 is a permutation , we need to figure out under what conditions a is unchanged after application of @xmath47 followed by @xmath48 , or @xmath61 notice that none of the symmetry transformations change the value of the central cell . therefore , for equation
( [ eq : aunch ] ) to hold the permutation @xmath48 can not change the value of the central cell ( which is always 7 in standard form ) , so the symmetry transformation must leave all 7 s unchanged .
we note that for the boards @xmath41 , @xmath42 , and @xmath43 , the only symmetry transformation that leaves all 7 s unchanged is a rotation of the board by @xmath62 .
we can then calculate that @xmath63 and similarly for @xmath42 and @xmath43 .
this effectively halves the number of distinct boards generated from these three , giving us @xmath64 different solutions .
the 7 s in the boards @xmath44 , @xmath45 and @xmath46 are unchanged by a rotation of @xmath65 , and @xmath66 and similarly for @xmath45 and @xmath46 .
this reduces the number of distinct boards generated in these cases by a factor of 6 , giving us @xmath67 .
the total number of possible boards is therefore @xmath68 .
* theorem 6 * : the smallest number of seeds for a septoku puzzle to have a unique solution is six .
: clearly the answer can not be less than six .
if this were possible , there must be two of the seven numbers not even among the seeds .
if there is one solution , a different one can be obtained by swapping these two numbers . to show the answer is six ,
we need only find a puzzle with six seeds and a unique solution . by the same argument as the previous paragraph , the six seeds must be for six of the seven numbers ( all different values ) .
we wrote a program that takes one of the six valid boards , and considers all combinations of 6 different seeds .
the program then checks each such puzzle for a unique solution by matching against all possible symmetry transformations and permutations of the six possible boards . in this manner hundreds of puzzles
were generated with six seeds and a unique solution .
four of them are shown in figure [ fig6 ] .
working by hand , finding solutions to these puzzles is difficult ( although the first is not so bad as we will show below ) .
the construction technique assures that the solution is unique .
we have also verified uniqueness using an integer programming model .
theorems 1 - 3 are useful for solving septoku puzzles by hand .
there is one more theorem that is even more powerful .
* theorem 7 * : in a valid septoku board , the seven symbols can be partitioned into three pairs and one unpaired symbol which is the value of the central cell 19 .
let @xmath39 and @xmath69 be any two cells that map to each other under @xmath70 rotation of the board ( for example cells 1 and 37 , or 2 and 36 , or 7 and 31 , etc . ) .
then either @xmath39 and @xmath69 have ( different ) values taken from one of the three pairs , or @xmath39 and @xmath69 both have the value of the central cell ( clearly this is only possible if @xmath39 and @xmath69 are not in the same row ) .
_ proof _ : for a board in standard form , theorem 7 is just a restatement of equation ( [ eq:180deg ] ) ( which is true of boards @xmath44 , @xmath45 and @xmath46 as well as @xmath41 , @xmath42 and @xmath43 ) . since any solution is a symmetry transformation plus permutation of one of @xmath41@xmath46 , all solutions satisfy theorem 7 . in the solution boards @xmath41-@xmath46 in standard form ,
the pairs are always @xmath71 , @xmath72 and @xmath73 , while the unpaired central cell is 7 .
theorem 7 says that if we find a 1 in some cell , there must be a 4 in the cell diametrically opposite , and similarly for the other pairs .
if there is a 7 in some cell , there must be another 7 in the cell diametrically opposite . in solving septoku puzzles , once we figure out what the pairings are , theorem 7 is very powerful in filling out the board .
once we figure out one board location we can also fill in the cell diametrically opposite .
this actually makes many puzzles easy , and the only ones that are not easy are those where it is hard to figure out the pairs .
for example , we now show how theorem 7 can be used to solve the first 6-seed puzzle in figure [ fig6 ] .
first , it is clear that the center can only be the missing symbol , 4 , and the only possibility for cell 20 is a 2 .
the board now looks like figure [ fig7]a .
now consider the pairs of theorem 7 .
clearly @xmath74 from the center is one such pair , the other two can only be @xmath75 and @xmath76 , or @xmath77 and @xmath78 .
cells 4 and 34 must come from some pair , and they ca nt come from @xmath74 or @xmath75 because row 4@xmath1234 already contains a 1 and 2 , and cells 4 and 34 ca nt come from @xmath76 because by theorem 2 the corners must be distinct .
so we now know that the three pairs are @xmath74 , @xmath77 and @xmath78 and the board looks like figure [ fig8]a .
cell 6 must be either a 1 or a 7 ( by theorem 1 ) .
if it is a 7 then circle 32 can not contain a 7 , so cell 6 must be a 1 and cell 32 is a 7 .
we can now determine that cells 28 and 31 must be 1 s , and cells 10 and 7 must be 7 s ( figure [ fig8]b ) .
cell 34 must be a 3 or a 5 , but if it is a 5 then circle 8 can contain no 5 . thus cell 34
must be a 3 , and cell 4 must be a 5 .
circle 21 must contain a 3 , and it can only be at 27 .
cell 1 must be a 2 or a 6 , but if it is a 6 then circle 32 can contain no 6 .
thus , cell 1 must be a 2 , and cell 37 a 6 ( figure [ fig8]c ) .
we can now readily fill in the rest of the solution ( figure [ fig8]d ) .
in fact , this solution board @xmath33 is a version of board @xmath45 , as can be seen by verifying the identity @xmath79 where @xmath80 is just the reflection of the board across the row 1@xmath1237 .
we now consider how we could modify the rules to change the puzzle
. we may be able to increase the number of possible solutions by modifying the rules in several ways .
we could increase the number of symbols ( to eight , for example ) while keeping the requirement that the symbols must be distinct within each of the 28 regions .
another possibility is to remove some of the regions . to keep the symmetry of the regions ,
if we remove one region it should be the center circle 19 .
the following theorem shows , however , that dropping the center circle region does not change the puzzle at all ! * theorem 8 * : consider septoku , but drop the requirement that the symbols in the center circle 19 be distinct .
there are * no solutions * to the puzzle where the cells in the center circle are not distinct . :
suppose a solution exists with two symbols the same in circle 19 .
because the rows must be distinct , the central number can not be the duplicate , so without loss of generality assume that the symbol 1 appears in cells 13 and 26 . but circles 17 and 21 must both contain 1 s , and these can only be on row 16@xmath1222 .
there can not be two 1 s along this row , so no solution is possible . by theorem 8
, we can relax the original rules of septoku to remove the center circle , and the puzzle will be unchanged . however , it is possible to find new solutions if we remove one or more of the other circle regions .
we have considered how to relax the rules of the puzzle to _ increase _ the number of solutions .
now we wish to go in a different direction , namely keeping the rules the same , but making the board larger .
since a larger board must still satisfy all of the original rules of septoku , this has the effect of _ reducing _ the number of solutions .
some septoku solutions have the property that the solution can be extended off the board , while holding to a natural extension of the rules .
we can extend two opposite edges of the board to create a 49-cell * rhombus board * in figure [ fig10 ] .
the rhombus ( or diamond if rotated @xmath36 counter - clockwise ) board has two sets of rows which all have 7 cells , and another set of rows which have 2 to 7 cells .
there are also now 9 circular regions .
only two of the six septoku solutions can be extended in this way ( @xmath45 and @xmath46 ) , and therefore up to symmetry transformations and permutations there are only * two * valid rhombus septoku boards , a fact verified by an ip model .
the two possible valid rhombus septoku boards are shown in figure [ fig10 ] .
in addition , the rhombus pattern can be repeated periodically to infinity .
another variation which is possible is the 73-cell * star septoku * board ( figure [ fig11 ] ) . on this board , some of the rows have more than seven cells .
we therefore modify the rules to specify that the symbols in _ any connected subset of a row with seven or fewer cells must contain distinct symbols_. there are only two valid star septoku boards . not surprisingly , it is easy to create many 6-seed puzzles in rhombus or star septoku that have unique solutions .
an example of one such puzzle is shown in figure [ fig11]b .
in all the board types discussed so far , the circular regions overlap . it is possible to create a board on a hexagonal grid where the circular regions do not overlap , more reminiscent of ordinary sudoku .
figure [ fig12 ] shows a * flower septoku * board , which has 49 cells and 34 regions . since the circular regions in a flower septoku board no longer overlap , one might suspect that this would allow for more solutions
however , note that there are now 15 rows of length seven which must contain all seven numbers . an ip model has found that there are only * three * different flower septoku boards , up to a symmetry transformation and permutation .
these three solutions also satisfy theorem 7so despite the change in geometry , solutions to flower septoku have the same symmetry properties as solutions in regular septoku .
septoku is a puzzle where the set of solutions is far more constrained than it may first appear .
we have found that there are only six fundamentally different solution boards , and this knowledge has allowed us to create 6-seed puzzles that can be quite difficult to solve by hand , at least not without detailed knowledge of the puzzle .
theorems 1 , 2 , 3 and 7 are useful for solving septoku puzzles .
there are several possible ways to reduce the constraints in septoku , so that more solutions are possible .
the first possibility is to reduce the number of regions .
the most obvious to remove are some of the circle regions , we have shown that removing only the center circle does not change the puzzle .
a second possibility is to consider a board where the circular regions do not overlap , one example is the 49-cell flower septoku board . a third possibility is to allow more than 7 symbols , but keep them unique among rows or inside circles . even 8 symbols will give a much larger number of different solutions . the main difference between sudoku and septoku
are the extra constraints imposed by an additional set of parallel rows in a hexagonal grid . in any variation of septoku
we have considered , these additional constraints have resulted in a relatively small number of solutions .
unique solutions to the 6-seed puzzles of figure [ fig6 ] are given in figure [ fig9 ] .
for the star septoku problem in figure [ fig11]b , rotate the board in figure figure [ fig11]a @xmath36 clockwise , and then swap @xmath81 and @xmath1 ( apply the permutation @xmath82 ) . for
the flower septoku problem in figure [ fig12]b , simply label each circular region with the same pattern , the standard pattern shown in figure [ fig2 ] .
it is hard to believe that all these solutions are unique , but this has been confirmed by ip models .
mybib http://www.oberg.org 2006 google us puzzle championship , http://wpc.puzzles.com/g6/ b. oberg , septoku , g4g7 exchange book , volume 2 , 153160,http://www.oberg.org / septoku_g4g7.pdf j. czyzyk , m. mesnier and j. mor , the neos server , _ ieee j. comp .
science and eng .
_ , * 5 * ( 1998 ) , 6875 , http://neos.mcs.anl.gov/neos/
here is a sample gams program for solving septoku puzzles .
email the author for a copy of this file , or for model files in rhombus , star , and flower septoku . * the septoku board ( 37 cells ) : * 1 2 3 4 * 5 < 6 > 7 < 8 > 9 * 10 11 12 13 14 15 * 16 < 17 > 18 < 19 > 20 < 21 > 22 * 23 24 25 26 27 28 * 29 < 30 > 31 < 32 > 33 * 34 35 36 37 * regions that must contain distinct values * regions 1 - 7 are the ( horizontal ) rows * regions 8 - 14 are the ( up - right slanting ) rows * regions 15 - 21 are the ( down - right slanting ) rows * regions 22 - 28 are the circular regions ( with centers < * > in diagram ) * note that all regions have between 4 and 7 cells set regions ( r , m ) / 1.1 , 1.2 , 1.3 , 1.4 , 2.5 , 2.6 , 2.7 , 2.8 , 2.9 , 3.10 , 3.11 , 3.12 , 3.13 , 3.14 , 3.15 , 4.16 , 4.17 , 4.18 , 4.19 , 4.20 , 4.21 , 4.22 , 5.23 , 5.24 , 5.25 , 5.26 , 5.27 , 5.28 6.29 , 6.30 , 6.31 , 6.32 , 6.33 , 7.34 , 7.35 , 7.36 , 7.37 , 8.1 , 8.5 , 8.10 , 8.16 , 9.2 , 9.6 , 9.11 , 9.17 , 9.23 , 10.3 , 10.7 , 10.12,10.18,10.24,10.29 , 11.4 , 11.8 , 11.13,11.19,11.25,11.30,11.34 , 12.9 , 12.14,12.20,12.26,12.31,12.35 , 13.15,13.21,13.27,13.32,13.36 , 14.22,14.28,14.33,14.37 , 15.4 , 15.9 , 15.15,15.22 , 16.3 , 16.8 , 16.14,16.21,16.28 , 17.2 , 17.7 , 17.13,17.20,17.27,17.33 , 18.1 , 18.6 , 18.12,18.19,18.26,18.32,18.37 , 19.5 , 19.11,19.18,19.25,19.31,19.36 , 20.10,20.17,20.24,20.30,20.35 , 21.16,21.23,21.29,21.34 , 22.1 , 22.2 , 22.5 , 22.6 , 22.7 , 22.11,22.12 , 23.3 , 23.4 , 23.7 , 23.8 , 23.9 , 23.13,23.14 , 24.10,24.11,24.16,24.17,24.18,24.23,24.24 , 25.12,25.13,25.18,25.19,25.20,25.25,25.26 , 26.14,26.15,26.20,26.21,26.22,26.27,26.28 , 27.24,27.25,27.29,27.30,27.31,27.34,27.35 , 28.26,28.27,28.31,28.32,28.33,28.36,28.37/ ; * solutions that may or may not be allowed set sol1 ( m , n ) / 1.2 , 2.5 , 3.4 , 4.3 , 5.3 , 6.6 , 7.7 , 8.1 , 9.5 , 10.5 , 11.4 , 12.1 , 13.2 , 14.6 , 15.7 , 16.1 , 17.2 , 18.6 , 19.7 , 20.3 , 21.5 , 22.4 , 23.7 , 24.3 , 25.5 , 26.4 , 27.1 , 28.2 , 29.2 , 30.4 , 31.7 , 32.3 , 33.6 , 34.6 , 35.1 , 36.2 , 37.5/ ; set sol2 ( m , n ) / 1.2 , 2.6 , 3.4 , 4.3 , 5.7 , 6.3 , 7.5 , 8.1 , 9.6 , 10.3 , 11.4 , 12.1 , 13.2 , 14.7 , 15.5 , 16.1 , 17.5 , 18.6 , 19.7 , 20.3 , 21.2 , 22.4 , 23.2 , 24.7 , 25.5 , 26.4 , 27.1 , 28.6 , 29.3 , 30.4 , 31.2 , 32.6 , 33.7 , 34.6 , 35.1 , 36.3 , 37.5/ ; set sol3 ( m , n ) / 1.5 , 2.6 , 3.4 , 4.3 , 5.2 , 6.3 , 7.7 , 8.1 , 9.6 , 10.3 , 11.4 , 12.1 , 13.2 , 14.5 , 15.7 , 16.1 , 17.5 , 18.6 , 19.7 , 20.3 , 21.2 , 22.4 , 23.7 , 24.2 , 25.5 , 26.4 , 27.1 , 28.6 , 29.3 , 30.4 , 31.7 , 32.6 , 33.5 , 34.6 , 35.1 , 36.3 , 37.2/ ; set sol4 ( m , n ) / 1.2 , 2.7 , 3.4 , 4.3 , 5.3 , 6.6 , 7.5 , 8.1 , 9.7 , 10.7 , 11.4 , 12.1 , 13.2 , 14.6 , 15.5 , 16.1 , 17.5 , 18.6 , 19.7 , 20.3 , 21.2 , 22.4 , 23.2 , 24.3 , 25.5 , 26.4 , 27.1 , 28.7 , 29.7 , 30.4 , 31.2 , 32.3 , 33.6 , 34.6 , 35.1 , 36.7 , 37.5/ ; scalars allow1 set to 37 to allow sol1 and 36 to not allow it /36/
allow2 set to 37 to allow sol2 and 36 to not allow it /36/
allow3 set to 37 to allow sol3 and 36 to not allow it /36/
allow4 set to 37 to allow sol4 and 36 to not allow it /36/ ; obj the sum of all labels ( unimportant ) allfilled(m ) all cells must have exactly one entry regdist(r , n ) each region must have distinct entries puzzle the seeds defining the puzzle uniq1 determines if sol1 is allowed or not uniq2 determines if sol2 is allowed or not uniq3 determines if sol3 is allowed or not uniq4 determines if sol4 is allowed or not ; obj .. sum((m , n),x(m , n ) ) = e= object ; allfilled(m ) .. sum(n , x(m , n ) ) = e= 1 ; regdist(r , n ) ..
sum(m$regions(r , m),x(m , n ) ) = l= 1 ; puzzle ..
sum(seeds , x(seeds ) ) = g= card(seeds ) ; uniq1 .. sum(sol1,x(sol1 ) ) = l= allow1 ; uniq2 .. sum(sol2,x(sol2 ) ) = l= allow2 ; uniq3 .. sum(sol3,x(sol3 ) ) = l= allow3 ; uniq4 .. sum(sol4,x(sol4 ) ) = l= allow4 ; | septoku is a sudoku variant invented by bruce oberg , played on a hexagonal grid of 37 cells .
we show that up to rotations , reflections , and symbol permutations , there are only six valid septoku boards .
we show that the minimum number of starting seeds for a septoku puzzle to have a unique solution is six .
we generalize the puzzle to other board shapes , and devise a 6-seed puzzle on a star - shaped board with 73 cells that has a unique solution . |
in recent years , high resolution observations have revealed the existence of many young ob stars in the galactic center ( gc ) .
accurate measurement of the orbital parameters of these stars gives strong evidence for the existence of a massive black hole ( mbh ) which dominates the dynamics in the gc @xcite .
most of the young stars are observed in the central 0.5 pc around the mbh .
the young star population in the inner 0.05 pc ( the `` s - stars '' ) consists exclusively of b - stars , in an apparently isotropic distribution around the mbh , with relatively high eccentricities ( @xmath7 ; @xcite ) .
the young stars outside this region comprise o - stars in one or two disks , and present markedly different orbital properties @xcite . since regular star formation in the region near the mbh
is inhibited by tidal forces , many suggestions have been made regarding the origin of the s - stars .
many of these are probably ruled out by observations and/or by theoretical arguments ( see * ? ? ?
* ; * ? ? ?
* for a review ) .
the various scenarios for the origin of the s - stars predict very different distributions for their orbits , which in principle could be constrained by observations .
however , it is not clear to what extent relaxation processes can produce changes in the distribution of orbital parameters after the stars have been deposited near their current locations .
here we try to resolve this question .
we use @xmath0-body simulations to follow the evolution of stellar orbits around the gc mbh for @xmath3 myr , starting from various initial conditions that were motivated by different models for the origin of the s - stars .
however , we do not discuss here the possibility that an intermediate mass black hole was involved in the production and/or evolution of the s - stars , which is discussed in details elsewhere ( see * ? ? ? * and references therein ) .
we find our n - body results to be consistent with our analytical predictions , and compare them with current observations .
we then discuss the implications for the validity of the models for the production of the s - stars .
in addition we study the possible ejection of the s - stars outside the inner @xmath8 pc and the contribution of such ejected stars to the population of hypervelocity stars ( as suggested by @xcite ) and to the isotropic population of b - stars observed at distances of up to @xmath9 pc from the mbh ( i.e. at distances similar to those of the young o / wr stellar disks , but outside of these disks at high inclinations ) . in 2 we summarize the different models for the origin of the s - stars and the predictions that they make for the stellar orbits at the time when the stars are first deposited near the mbh .
3 describes the @xmath0-body simulations we carried out to follow the s - star orbital evolution starting from these initial conditions .
4 and 5 present the results of these simulations and discusses the implications for the origin of the stellar populations of b - stars in the gc and for the population of hypervelocity stars observed in the galactic halo . 6 sums up .
many solutions have been suggested for the origin of the s - stars , but many of these have been effectively excluded ( see @xcite for a review ) .
here we focus on two basic models which differ substantially in their predictions for the initial orbital distribution of the s - stars and/or the time passed since their arrival / formation at their current location . these are ( 1 ) formation of the s - stars in a stellar disk close to the mbh , followed by transport through a planetary - migration - like scenario to their current positions @xcite ; ( 2 ) formation of the s - stars in binaries far from the mbh followed by scattering onto the mbh by massive perturbers ( e.g. giant molecular clouds ) and tidal disruption of the binaries @xcite , leaving a captured star in a tight orbit around the mbh . binary disruption scenarios similar to ( 2 )
have been proposed , in which the s - stars formed in a stellar disk ( either the currently observed 6 myr old disk , or an older , currently not observed disk ) and later changed their orbits due to coherent torques through an instability of eccentric disks @xcite ; or through the kozai mechanism resulting from the presence of two disks @xcite .
the latter alternative is unlikely since the kozai mechanism is quenched in the presence of a massive enough cusp of stars such as exists in the gc @xcite ) . in any case , all the binary - disruption scenarios imply very similar initial distributions for the captured s - stars , and may differ only in their relevant timescales . in the following ,
we briefly discuss the initial distribution of the eccentricities and inclinations of the s - stars expected from the different scenarios for their production .
the models are summarized in table [ t : models ] .
llccccl # & origin & initial & time & model & survival & refs.@xmath10 + & & eccentricity & ( myr ) & probability@xmath11 & fraction@xmath12 & + 1 & capture following binary disruption due & high & @xmath13 & @xmath14 & @xmath15 & 1 + & to disk instability in the currently observed disk & ( @xmath16 ) & & & & + 2 & capture following binary disruption due & high & @xmath3 & @xmath17 & @xmath18 & 2 + & to massive perturbers or to disk instability & ( @xmath16 ) & & & & + & in an old non - observed disk & & & & & + 3 & disk formation + planetary like & low & @xmath13 & @xmath19 & @xmath20 & 3 + & migration ( currently observed disk ) & ( @xmath21 ) & & & & + 4 & disk formation + planetary like & low & @xmath3 & @xmath22 & @xmath15 & 3 + & migration ( possible old disk ) & ( @xmath21 ) & & & & + + + + + a close pass of a binary star near a massive black hole results in an exchange interaction , in which one star is ejected at high velocity , while its companion is captured by the mbh and left on a bound orbit .
such interaction occurs because of the tidal forces exerted by the mbh on the binary components .
typically , a binary ( with mass , @xmath23 , and semi - major axis , @xmath24 ) , is disrupted when it crosses the tidal radius of the mbh ( of mass @xmath25 ) , given by @xmath26 .
one of the binary components is captured by the mbh @xcite on a wide and eccentric orbit while the companion is ejected with high velocity @xcite . the capture probability and the semi - major axis distribution of the captured stars were estimated by means of numerical simulations , showing that most binaries approaching the mbh within the tidal radius @xmath27 are disrupted @xcite .
the harmonic mean semi - major axis for 3-body exchanges with equal mass binaries was found to be @xcite @xmath28 where @xmath24 is the semi - major axis of the infalling binary and @xmath29 that of the captured star ( the mbh - star `` binary '' ) .
most values of @xmath29 fall within a factor @xmath30 of the mean .
this relation maps the semi - major axis distribution of the infalling binaries to that of the captured stars : the harder the binaries , the more tightly bound the captured stars .
the periapse of the captured star is at @xmath31 , and therefore its eccentricity is very high @xcite , @xmath32 for values typical of b - type main sequence binaries and the mbh in the gc ( @xmath33 ; @xmath34 ; @xmath35 ) .
therefore , in order to study the evolution of s - stars from the binary disruption scenarios we assume that the initial eccentricities of s - stars are in the range @xmath36 ( where most are close to the mean value of @xmath37 ) . in principle
the binary disruption scenario has specific predictions for the semi - major axis distribution of the captured stars , which could also be used for constraining the model .
however such distribution is highly sensitive to differences in the ( unknown ) binary distribution in the gc region .
the prediction of high eccentricities for the captured s - stars , instead , is robust and has only a weak dependence on the mass of the binary .
the inclinations of the captured s - stars in the massive perturbers scenario @xcite are likely to be distributed isotropically since the stars originate in an isotropic cusp .
although in the disk instability scenario @xcite the progenitors of the captured stars form in the stellar disk , their original inclinations could be excited to higher inclinations than the typical inclinations observed in the stellar disk ( y. levin , private communication ) , and may resemble a more isotropic distribution .
@xcite suggested that the s - stars could have formed in the currently observed stellar disk in the gc @xcite , and then migrated inward in a way similar to planetary migration .
the migration time scale expected from such a scenario could be as short as @xmath38 yr ( for type i migration ) , which could be comparable to ( although possibly larger than ; @xcite ) the lifetime of the gaseous disk .
recent analytic work ( @xcite ; and references therein ) has shown that eccentricity is likely to be damped during migration , unless eccentricity excitation occurs , which requires the opening of a clean gap in the disk . in the latter case
the migration timescale might be larger @xcite , possibly inconsistent with the lifetime of the gaseous disk @xcite .
it is therefore more likely that the eccentricities of the stars are damped during the migration .
even eccentricity excitation , if such took place , is unlikely to excite very high eccentricities .
the mean eccentricity of the observed stars in the stellar disk is @xmath39 .
therefore , in order to study the evolution of the s - stars following their formation in a stellar disk , we assume them to have low eccentricities , or , being conservative , moderately high eccentricities ( @xmath40 ; where we use a thermal distribution of eccentricities , cut off at @xmath41 ) .
these simulations include the less likely ( @xcite ) possibility that the stellar disk extended inward to the current region of the s - stars , in which case the s - stars were formed in - situ and did not migrate .
to test these competing models , we carried out @xmath0-body simulations of the inner milky way bulge using models containing a realistic number of stars .
all integrations were carried out on the 32-node grape cluster ` gravitysimulator ` at the rochester institute of technology which adopts a parallel setup of grape accelerator boards to efficiently compute gravitational forces .
the direct - summation code @xmath42grape was used @xcite .
the simulations used a softening radius of @xmath43 comparable to the radius of the s - stars @xmath44 , so as to be able to follow even the closest encounters between stars .
our initial conditions were based on a collisionally evolved model of a cusp of stars and stellar remnants around the gc mbh ( * ? ? ?
* hereafter ha06 ; see also @xcite ) .
ha06 evolved the multi - mass isotropic fokker - planck equation representing the stellar distribution in the region extending to @xmath45 pc from the mbh ; in their models the contribution to the gravitational potential from the distributed mass is ignored .
ha06 fixed the relative numbers of objects in each of four mass bins by assuming a mass function consistent with continuous star formation . the @xmath2 stellar - mass black holes ( sbhs )
were found to follow a steep , @xmath46 density profile near the mbh while the lower - mass populations ( main - sequence stars , white dwarfs , neutron stars ) had @xmath47 , @xmath48 .
the sbhs were found to dominate the mass density inside @xmath49 pc .
based on this model , we constructed an @xmath0-body realization containing a total of 1200 objects within 0.3 pc of the mbh : 200 `` stars , '' with masses of @xmath50 , and 1000 `` black holes '' with masses @xmath51 , around a mbh of @xmath52 .
we set @xmath53 for the sbhs and @xmath54 for the lower - mass stars , and each density component was tapered smoothly to zero beyond @xmath55 pc when computing the corresponding @xmath56 . since the s - stars may have masses as high as @xmath57 , the higher mass stars in our simulations could also be treated as s - stars .
we did not see any major differences in the evolution of the more massive and the less massive stars , and we discuss the evolution of both together .
the number of sbhs contained within a radius @xmath58 in our @xmath0-body models was @xmath59 implying a distributed mass within @xmath55 pc of @xmath60 .
this is somewhat ( @xmath61 ) lower than the mass in sbhs in the ha06 or similar @xcite models at the same radius , and a factor @xmath62 lower than the _ total _ mass ( mostly in main - sequence stars ) in the fokker - planck models . in this sense , the rates of evolution that we infer below can be considered to be conservative . on the other hand
, we note that the late - type ( old ) stars that dominate the number counts in this region have a much flatter density profile than predicted by the ha06 models , possibly even exhibiting a central `` hole '' @xcite . only the b - type stars in the nuclear star cluster show a steeply - rising number density , @xmath63 @xcite but they presumably constitute a negligible fraction of the total mass in this region , and in any case are far too young to have reached a collisional steady state around the mbh .
while the origin of this discrepancy between models and observations is currently unresolved , it may imply that the other contributors to the distributed mass around the mbh , including the sbhs , also have a lower density than in the fokker - planck models .
for instance , relaxation times at the gc may be too long for collisionally relaxed steady states to have been established in the last 10 gyr @xcite .
modeling of the stellar proper motion data @xcite implies a distributed mass within 1 pc of @xmath64 , but these data are consistent with both rising and falling mass densities within this region and the distributed mass in the inner @xmath55 pc is essentially unconstrained @xcite . because of these uncertainties
, we discuss below how our results would vary if different numbers of sbhs were assumed .
as discussed in the previous section , we studied two basic sets of initial conditions for the s - stars . in the first model we assumed that the s - stars were captured by the mbh as in the binary disruption scenario @xcite , which leaves the captured stars in highly eccentric orbits ( @xmath65 ; cf .
section [ sec : binary - disruption ] ) . under these assumptions ,
the stars evolved for @xmath13 myr ( if formed in the stellar disk ) or longer ( if the s - stars formed outside the central pc , i.e. not in the young stellar disk ) .
we evolved the models for up to @xmath3 myr , which is comparable to the lifetime of the observed s - stars ( although some may have longer lifespans ) . in the second scenario we assumed that the s - stars formed in a gaseous disk and then migrated inwards ( or formed in situ in a disk extending close to the mbh ) .
for this case we assumed the s - stars to have low eccentricities ( @xmath66 ) , as typical of disk formation models , and to evolve for @xmath67myr ( the lifetime of the observed stellar disk ) . in order to check both scenarios we selected the stars with initially high eccentricity orbits ( @xmath68 ) and low eccentricity orbits ( @xmath69 ) and followed their evolution for a time appropriate to their presumed origin .
in addition to these evolutionary scenarios we also studied the possibility of ejection of s - stars as hypervelocity stars , following a close encounter with a sbh in the vicinity of the mbh , as suggested by @xcite .
our high resolution simulations can accurately follow close encounters between stars , and therefore track any resulting high velocity ejections of stars .
motivated by recent observations of b - type main sequence stars outside the inner @xmath8 pc of the galaxy , which show evidence of random orientations similar to those of the s - stars , we also looked for stars scattered to larger distances by gravitational encounters .
in other words , we considered whether captured s - stars could dynamically evolve to become the extended b - type stars population observed outside the central @xmath8 pc .
although some binaries could exist in the close regions near the mbh @xcite , these are likely to be rapidly disrupted and not play an important role in the dynamical evolution of the s - stars @xcite . we therefore do not include any binaries in our initial conditions for the s - stars and sbhs evolution .
we applied the correlation curve method @xcite to our simulations to identify the relaxation process responsible for the dynamical evolution of the stars ( fig . [
f : rr ] ) .
the method is able to detect and measure relaxation in nearly keplerian @xmath0-body systems . in the isotropic system
considered here , the angular momentum of the stars , @xmath70 , evolves both due to the slow stochastic two - body relaxation ( e.g. @xcite ) and to the rapid resonant relaxation ( rr ) @xcite .
two - body relaxation changes @xmath70 in a random walk fashion , @xmath71 over the long two - body relaxation time - scale @xmath72 , where @xmath73 is the maximal ( circular orbit ) angular momentum for a given energy , @xmath74 , @xmath0 is the number of enclosed stars on the distance scale of interest , and time @xmath75 is measured in terms of the orbital period on that scale .
resonant relaxation occurs when the symmetries of the potential act to constrain the stellar orbits ( e.g. closed ellipses in a kepler potential , or planar rosettes in a spherical one ) .
as long as the symmetry is approximately maintained on the coherence timescale @xmath76 , the stars experience coherent torques , and @xmath77 . in a nearly keplerian potential , as is the case in the inner parsec of the gc , rr can change both the magnitude of @xmath70 ( `` scalar rr '' ) and its direction ( `` vector rr '' ) . in the newtonian context
, the coherence of scalar rr is limited by the precession of the apoapse due to the enclosed stellar mass , on a time - scale @xmath78 .
on timescales @xmath79 , the coherent change @xmath80 becomes the mean free path in @xmath70-space for a rapid random walk , @xmath81 , where @xmath82 .
the coherence of vector rr is self - limited by the change in the orbital orientation due to rr , and is even faster , @xmath83 , where @xmath84 .
figure ( [ f : rr ] ) shows the rms change in the scalar and vector angular momentum of the stars in the simulation , as a function of the time lag @xmath85 ( correlation curves ) , up to the maximal time lag for which the simulation can still be analyzed with high statistical confidence , @xmath86 . in our simulation and @xmath87 ,
where @xmath88 and @xmath89 .
@xmath90 for our simulation . ] , the scalar rr coherence time expected from theory is @xmath91 and the scalar rr timescale is @xmath92 .
the behavior of the scalar correlation curve clearly indicates that rr dominates relaxation in our gc model .
the turn from the coherent phase to the random - walk phase of rr is seen at @xmath93 , close to the predicted value of @xmath94 .
the random - walk growth continues up to full randomization and saturation at @xmath95 , as expected @xcite .
the scalar rr timescale can be estimated directly from the simulated data by @xmath96 ; substituting @xmath97 at @xmath98 yields @xmath99 , in good agreement with the predicted value .
the correlation curves @xmath100 and @xmath101 as a function of the time lag @xmath85 for all the stars in the simulation . the simulation data ( points )
are compared to the best fit theoretical curves ( lines ) .
four regimes are seen ( * ? ? ?
* see text and also ) . at very short timescales ,
non - resonant two - body relaxation ( @xmath102 ) dominates over rr ( @xmath103 ) ; at @xmath104 the curves display a coherent , linear rise ; at @xmath105 , the curves display a random - walk growth and at @xmath106 they begin to saturate . ] furthermore , since @xmath107 , the rr - driven eccentricity evolution relates the final eccentricity , @xmath108 , after time lag @xmath85 to the initial eccentricity , @xmath109 .
we therefore expect the eccentricities at some given time lag to have a typical spread , @xmath110 , where @xmath111^{1/2}.\ ] ] the magnitude of the predicted change in eccentricity agrees well with that observed in the simulations .
for example , an s - star initially captured by a tidal exchange event on a @xmath112 yr ( @xmath113 pc ) , @xmath114 orbit can evolve by rr to an @xmath115 orbit in 20 myr .
the short vector rr timescale @xmath116 ( @xmath117 yr at 0.04 pc ) implies full randomization of the orbital planes after 6 myr , throughout the s - cluster volume , as is observed in the simulation .
we therefore conclude that rr is the dominant mechanism responsible for the dynamical evolution of the s - stars and other stars close to the mbh in the gc . in figure
[ f : eccentricities - cum ] we show the final cumulative eccentricity distribution of the s - stars for the different origin models ( table [ t : models ] ) .
these are compared to the the orbits of the observed s - stars ( taken from @xcite ) .
the probabilities for the samples of the observed and simulated s - stars to be randomly chosen from the same distribution ( calculated using a two - sample @xmath118 test ) are given in table [ t : models ] .
we find that the binary disruption model taking place at least @xmath3 myr ago is much favored over all other models tested here .
we find @xmath119 chance for the observed s - stars and the simulated stars in such model to originate from the same distribution ( with the shorter timescale of @xmath13 myr consistent at the @xmath120 level ) .
in contrast , the disk migration scenarios seem to be excluded ( for the given assumptions ) , since they have major difficulties in explaining the large fraction of eccentric orbits observed for the s - stars in the gc .
we find that the inclination distribution of the captured stars , initialized with the same inclination , is rapidly isotropized , to resemble a random distribution of inclinations ( consistent with random at the @xmath121 and @xmath122 level , after evolution of @xmath3 and @xmath13 myr , respectively ) .
this is expected from the rr process , as discussed in the previous section ( see also fig . [
f : rr ] ) .
we conclude that the observed isotropic distribution of the s - stars angular momentum direction is consistent with all the s - stars production models studied here , and can not be used to discriminate between them , although it constrains the lifetime of the s - stars system to be at least @xmath123 myr at the @xmath124 level , assuming all s - stars were initially put on the same plane .
as already discussed above , the s - stars can change their orbits due to their dynamical evolution .
a star could therefore be scattered very close to the mbh and be disrupted by it , if its pericenter distance from the mbh becomes smaller than the tidal radius of the star @xmath125 .
many of the s - stars could therefore not survive for long close to the mbh .
we followed the orbits of stars in our simulations and calculated the fraction of stars that have been disrupted .
for the tidal disruption calculations all stars were assumed to have the typical main sequence radius according to their mass .
we consider a star as being disrupted if its pericenter became smaller than twice the tidal radius during the simulation ( i.e. , when it is strongly affected by the mbh tidal forces or even totally disrupted in one pericenter passage ) .
we find that most of the s - stars survived to current times in all the models ( see survival fractions in [ t : models ] ) .
the s - stars population in the gc therefore gives a good representation of all the s - stars formed / captured in this region .
the production rate of the s - stars required to explain current observations is therefore only slightly higher ( @xmath126 times higher ) than that deduced from current number of s - stars observed . in principle
the s - stars could be ejected by strong encounters to orbits with larger semi - major axes , putting them outside the 0.05 pc region near the mbh @xcite , or even ejecting them as unbound hypervelocity stars @xcite .
the softening radius used in our simulation was @xmath127 , comparable to the radius of observed s - stars , allowing us to follow even very close encounters .
nevertheless , we find that only @xmath128 of the stars were ejected outside of the central @xmath8 pc , and even those had maximal semi - major axis in the end of the simulation not extending beyond 0.1 pc .
we therefore conclude that such ejected s - stars can not explain recent observations of many b - type stars outside the central @xmath8 pc ( h bartko , private communication ) . moreover ,
none of the @xmath50 stars in our simulations have been ejected as a hypervelocity star , suggesting that ejection of hypervelocity stars through encounters with sbhs is not an efficient mechanism ( see also @xcite for the related constraints on this mechanism ) .
we note that since our simulations can be rescaled , we can probe much higher stellar densities and smaller softening radius ( see next section ) .
when such rescaling is used , in which all the 1200 stars are distributed between @xmath129 pc to @xmath55 pc ( rescaling the radii by half ) , we find @xmath130 stars ejected beyond @xmath55 pc ( to have final semi - major axis of @xmath131 and @xmath132 pc ) , but none ejected as hypervelocity stars or even as slow unbound stars .
as discussed above , the density of the sbhs that are responsible for the evolution in our @xmath0-body models is not well determined . scaling the @xmath0-body results to different assumed values of @xmath0
is complicated by the fact that scalar resonant relaxation has two regimes , coherent and random walk . in our models ,
the transition occurs at @xmath133 where the @xmath134 scaling is for scattering by sbhs of a given mass . in the coherent regime , @xmath135 ,
orbital angular momenta grow as @xmath136 in the diffusive regime , @xmath137 , @xmath138 independent of @xmath139 .
we are interested in the orbital evolution of the s - stars over timescales of @xmath140 yr . in our simulations and those with larger @xmath0 , @xmath141 . in this large-@xmath0 regime ,
changes in eccentricity are dominated by the diffusive relation and are therefore expected to be nearly independent of @xmath0 for time scales of interest , at least up to @xmath0-values of @xmath142 where the distributed mass begins to approach the mass of the mbh and resonant relaxation is no longer effective . only for @xmath143 does @xmath144 approach @xmath145 yr and our results start depending significantly on @xmath139 . however , such a small number of sbhs in the volume of interest is highly unlikely , and therefore our results are robust to the details of the sbh cusp model .
the @xmath0-body simulations can also trivially be rescaled by @xmath146 at fixed mass .
this corresponds to placing the same number of sbhs into a smaller ( larger ) region and integrating for a shorter ( longer ) time .
for instance , if we rescale our simulations to three times smaller distances ( in which case the stars are distributed between @xmath129 and @xmath8 pc ) , the integration time becomes @xmath62 myr .
the approximately thermal eccentricity distribution of the s - stars near the massive black hole ( mbh ) in the galactic center ( gc ) , @xmath147 , is not naturally predicted by either of the two leading models for their production : migration of stars formed in a gaseous disk ; or capture of stars following binary disruption by the mbh . the former model predicts eccentricities that are too low , the latter too high . in this paper , we followed the dynamical evolution of orbits of various eccentricities near the gc mbh , including for the first time the cluster of stellar - mass black holes ( sbhs ) that is expected to form around the mbh via mass segregation .
we found that perturbations from the sbhs can reduce the eccentricities of initially very eccentric orbits ( @xmath148 ) into a distribution that is consistent with the observed one , in a time of approximately @xmath3 myr , comparable to s - star lifespans ; some of the stars change their eccentricities by more than @xmath9 to values as low as @xmath149 .
we confirmed these @xmath0-body results via a theoretical analysis of the relaxation process , and used that analysis to argue that our results are not strongly dependent on the ( unknown ) normalization of the sbh density near the mbh .
the same mechanism is unable to convert initially low - eccentricity orbits into very eccentric ones on the same time scale , arguing against the validity of the disk migration model for the origin of the s - stars .
we also found that most s - stars are not disrupted by the mbh during their lifetime , and very few are ejected outside the central @xmath8 pc near the mbh , and none having a semi - major axis beyond @xmath55 pc .
we did not find any hypervelocity star ejected in our simulation .
evolution toward a thermal eccentricity distribution is a natural consequence of random gravitational encounters with a population of massive perturbers . in this paper
, we considered the effect of a background of sbhs , which are expected on the basis of very general arguments to contribute a total mass of @xmath60 in the inner 0.1 pc around the mbh .
we showed that they were effective at moderating the eccentricities of initially highly eccentric orbits .
these results strengthen models in which the s - stars form from disrupted binaries , and disfavor models in which the s - stars are formed with low eccentricities . | we use @xmath0-body simulations to study the evolution of the orbital eccentricities of stars deposited near ( @xmath1 pc ) the milky way massive black hole ( mbh ) , starting from initial conditions motivated by two competing models for their origin : formation in a disk followed by inward migration ; and exchange interactions involving a binary star .
the first model predicts modest eccentricities , lower than those observed in the s - star cluster , while the second model predicts higher eccentricities than observed .
the @xmath0-body simulations include a dense cluster of @xmath2 stellar black holes ( sbhs ) , expected to accumulate near the mbh by mass segregation .
perturbations from the sbhs tend to randomize the stellar orbits , partially erasing the dynamical signatures of their origin .
the eccentricities of the initially highly eccentric stars evolve , in @xmath3 myr ( the s - star lifespan ) , to a distribution that is consistent at the @xmath4 level with the observed eccentricity distribution .
in contrast , the eccentricities of the initially more circular orbits fail to evolve to the observed values in 20 myr , arguing against the disk migration scenario .
we find that @xmath5@xmath6 of the s - stars are tidally disrupted by the mbh over their lifetimes , and that the s - stars are not likely to be ejected as hypervelocity stars outside the central 0.05 pc by close encounters with stellar black holes . |
stars like the sun , which have sub - surface convection zones , display a rich spectrum of acoustic ( p - mode ) oscillations .
the oscillations are stochastically excited and damped by the convection , and this gives rise to an extremely rich spectrum of modes .
measurement of the amplitudes and damping rates of the p modes therefore gives important information for constraining theories of convection in stellar interiors .
chaplin et al . ( 2007 ) used model computations of the excitation and damping of sun - like oscillations to make predictions of the p - mode spectra of a selection of stars on the lower main sequence .
the computations revealed an extremely interesting feature in the predicted appearance of the spectra of those model stars that had effective temperatures cooler than about 5400k : the modelled power spectral density of the modes showed two maxima , at different frequencies .
chaplin et al . found a pronounced dip in mode power between the maxima when the computations were made for young stars ; since the maxima are well separated in frequency , the predicted spectra took on a `` double humped '' appearance . in older main - sequence stars
the dip was found to be much less pronounced , and instead the spectra showed a broad plateau of power .
the k1 v main - sequence star @xmath0cen b ( hr5460 ) is a suitable candidate to test the predictions , since its effective temperature lies on the cool side of the 5400-k boundary given by the model computations , and data are available on its p - mode spectrum from observations made by kjeldsen et al .
as we shall demonstrate , the model computations predict a broad plateau of power for @xmath0cen b ; however , this broad plateau is not seen in the p - mode data of kjeldsen et al .
this paper reports on attempts to try and understand this disagreement between theory and observation .
the observational data for our study are the aforementioned doppler velocity observations of @xmath0cen b , made by kjeldsen et al .
the theoretical predictions we use are pulsation computations of the stochastic excitation rates and the damping rates of the radial p modes of the star . by making judicious comparisons of the theoretical computations and the observational data we show it is actually possible to make inference on the p - mode linewidths of @xmath0cen b , without directly measuring those linewidths , even though the kjeldsen et al .
observations span only a few nights .
the conclusions drawn do rest on one important assumption : that most of the disagreement that is observed between theoretical predictions and observations of the radial - mode amplitudes of @xmath0cen b resides in errors in the model damping rates , and not errors in the model excitation rates . we provide evidence in support of this rationalization from a similar comparison of theoretical and observational data of the p - mode spectrum of the `` sun as a star '' .
the sun - as - a - star data come from observations made by the ground - based birmingham solar - oscillations network ( bison ) ( chaplin et al .
1996 ) in doppler velocity .
the layout of our paper is as follows .
we begin in section [ sec : data ] with a description of the excitation and damping - rate computations .
we then proceed in section [ sec : res ] to compare theoretical predictions and observations of the mode amplitudes , for both @xmath0cen b and the sun .
then , in section [ sec : wid ] , we show how discrepancies between the theoretical and observed amplitudes may be explained largely in terms of errors in the theoretically computed damping rates .
we also demonstrate how inference may be made on the mode linewidths from use of the observed amplitudes and the theoretical excitation and damping computations .
we finish in section [ sec : disc ] with a brief discussion of the main points of the paper .
the stellar equilibrium and pulsation computations that we performed are as described by balmforth ( 1992a ) and houdek et .
al ( 1999 ) .
these computations gave estimates of the powers and damping rates of the radial p modes of @xmath0cen b. the model computations required four general input parameters to specify the stellar model : the mass , @xmath2 ; radius , @xmath3 ; effective temperature , @xmath4 ; and chemical composition .
we took values for @xmath0cen b of : @xmath5 ; @xmath6 ; and @xmath7 ( see yildiz 2007 , and references therein ) .
the composition was fixed at @xmath8 and @xmath9 , as for the model computations performed in chaplin et al .
( 2005 , 2007 ) . while this composition differs from estimates given in the literature for the @xmath0cen b system ( of @xmath10
; again , see yildiz 2007 ) , changes in composition at this level have only a second - order impact on the model calculated p - mode excitation and damping rates , and any changes that might be relevant here are smaller than the observational uncertainties ( e.g. , see fig .
17 of houdek et al . 1999 ) .
computation of the excitation rates and damping rates of the p modes demands a description of how the pulsations interact with the convection .
this requires computation of the turbulent fluxes associated with the convective heat and momentum transport .
these turbulent fluxes are obtained from a nonlocal , time - dependent generalization of the mixing - length formulation of gough ( 1977a , b ) , with a mixing length calibrated to the sun . in this generalization
there is a parameter , @xmath11 , which specifies the shape of the convective eddies .
then there are two parameters , @xmath12 and @xmath13 , which control respectively the spatial coherence of the ensemble of eddies contributing to the turbulent fluxes of heat and momentum and the degree to which the turbulent fluxes are coupled to the local stratification .
these two parameters control the degree of ` non - locality ' of convection ; low values imply highly nonlocal solutions , and in the limit @xmath14 the system of equations formally reduces to the local formulation ( except near the boundaries of the convection zone , where the local equations are singular ) .
gough ( 1977a ) has suggested theoretical estimates for their values , but it is likely that the standard mixing - length assumption of assigning a unique scale to turbulent eddies at any given location causes too much smoothing ; accordingly , somewhat larger values probably yield more realistic results . in this paper
we therefore adopt two sets of values for the non - local parameters which provide reasonable results for the sun , and other sun - like stars : @xmath15 ; and @xmath16 .
we comment further on the impact of this choice in section [ sec : cal ] below .
both the envelope and pulsation computations assumed the three - dimensional eddington approximation to radiative transfer ( unno & spiegel 1966 ) . the integration was carried out inwards , starting at an optical depth of @xmath17=@xmath18 and ending at a radius fraction @xmath19 .
the opacities were obtained from the opal tables ( iglesias & rogers 1996 ) , supplemented at low temperature by tables from kurucz ( 1991 ) .
the equation of state included a detailed treatment of the ionization of c , n , and o , and a treatment of the first ionization of the next seven most abundant elements ( christensen - dalsgaard 1982 ) , as well as ` pressure ionization ' by the method of eggleton , faulkner & flannery ( 1973 ) ; electrons were treated with relativistic fermi - dirac statistics .
perfectly reflective mechanical and thermal outer boundary conditions in the pulsation computation were applied at the temperature minimum in the manner of baker & kippenhahn ( 1965 ) . at the base of the model envelope
the conditions of adiabaticity and vanishing displacement were imposed .
only radial p modes were considered .
the amplitudes of stochastically excited oscillations are obtained in the manner of chaplin et al .
the procedure is based on the formulation by balmforth ( 1992b ) and goldreich & keeley ( 1977 ) but includes a consistent treatment of the anisotropy parameter @xmath11 of the turbulent velocity field ( for details see the discussion in chaplin et al .
2005 ) . for the anisotropy parameter
we adopt the value @xmath20 , a value that was also considered by chaplin et al .
the excitation model assumes a description in which the largest , energy - bearing eddies are described by the mixing - length approach .
the small - scale convection is modelled by a turbulence spectrum , for which we adopt the kolmogorov spectrum ( kolmogorov 1941 ) describing the spatial properties of the small - scale turbulence .
the temporal behaviour of the small - scale turbulent dynamics has a frequency spectrum that is approximated by a gaussian centred at zero frequency and with a width corresponding to the inverse of the correlation timescale of an eddy whose spatial extent ( or wavenumber ) is characterized by the mixing length .
the correlation timescale is not a well - defined quantity , and therefore we scale a ( well - defined ) characteristic eddy turnover time with a correlation parameter @xmath21 ( cf .
balmforth 1992b ) , which we set to @xmath22 . in this section
we describe how the results of the stellar model computations were used to make predictions of the parameters of the p - mode peaks observed in the frequency power spectrum . for given values of @xmath23 and
@xmath24 the model computations provided predictions of the linear damping rates , @xmath25 , and the acoustic energy supply rates , @xmath26 , of the radial p modes .
the observed parameters of the mode peaks are formed from these quantities .
the peak fwhm linewidths are given by @xmath27 the mode velocity powers , @xmath28 , are calculated from ( e.g. , houdek et al .
1999 ) : @xmath29 where @xmath30 is the mode inertia .
the maximum power spectral density , or height @xmath31 , of a mode peak in the frequency power spectrum depends on the effective length of the dataset , @xmath32 , and the linewidth , @xmath33 , via the formula ( fletcher et al .
2006 ) : @xmath34 when @xmath35 ( i.e. , when @xmath36 ) the mode is not resolved and power is confined largely in one bin of the frequency power spectrum , so that @xmath37 .
when , on the other hand , @xmath38 ( i.e. , @xmath39 ) , the mode is well resolved and the lorentzian shape of the peak may be inferred from the power in the bins occupied by the peak . for cases in the intermediate regime where @xmath40 is neither much greater , or much smaller , than two there is a gradual transition between @xmath31 for the unresolved and fully resolved regimes .
in other recent papers ( e.g. , chaplin et al . 2005 ) we have gone on to use @xmath31 to make comparisons of theory with observation . in this paper
we instead use data on heavily smoothed frequency power spectra , which give direct inference on @xmath28 ( and therefore the amplitudes , @xmath41 ) as opposed to the @xmath31 .
however , we do still make use of the heights to calibrate the model computations , as we now go on to discuss . we make three important points concerning the calibration .
firstly , the computations have been calibrated so that , for a model of the sun , the average maximum power spectral density , @xmath31 , of the five most prominent modes is the same as that observed in the real bison sun - as - a - star data . by using an average over several modes , as opposed to taking the @xmath31 of just the strongest mode , we seek to stabilize the calibration against small - scale fluctuations in the computations ( chaplin et al .
2007 ) . furthermore , by using data on @xmath31 , as opposed to @xmath28 ( or @xmath41 ) , we maintain consistency with our previous work ( e.g. , see chaplin et al .
2008 ) . to summarize our first point : data on the sun serve as a reference calibration for the model computations .
secondly , it is important to recognise that the reference calibration had to be performed independently for each of the non - local parameter choices @xmath15 and @xmath16 , respectively .
this is because the changes to the values of the non - local parameters can have a significant impact on the computed damping rates , which in turn affects the absolute magnitudes of the predicted velocity powers , @xmath28 ( equation [ eq : v2 ] ) and therefore the heights , @xmath31 ( equation [ eq : height ] ) .
thirdly , when it comes to comparing the results of the pulsation computations with the observational data on @xmath0cen b , we must remember that there will be instrument - dependent differences in the observed velocity amplitudes , due to the use of different spectral lines in the doppler velocity observations .
we have folded into the calibration of the model data the fact that the amplitudes of the solar p modes measured by the stellar techniques are a factor 1.07-times smaller than the amplitudes measured by bison ( kjeldsen et al . , 2008 ) .
the left - hand panel of fig .
[ fig : rawspecs ] shows the observed frequency power spectrum of @xmath0cen b , plotted on a logarithmic scale .
the spectrum was computed from a few nights of doppler velocity data collected by kjeldsen et al .
the dark solid line is a smoothed spectrum given by applying to the raw spectrum a gaussian filter of width @xmath42 , where @xmath43 is the large frequency spacing between consecutive overtones ( here @xmath44 ) .
the dashed line is a smooth estimate of the background power spectral density .
it was obtained by fitting , in regions outside the range occupied by the p modes , a second - order polynomial to the logarithm of power versus the logarithm of frequency .
we used the gaussian smoothed spectrum and the background fit to estimate the mode amplitudes , @xmath41 .
this was done by following the recipe outlined in kjeldsen et al .
( 2008)cen b and the sun that are a few percent lower than in kjeldsen et al .
these differences arise from differences in the fitting - function that was used to estimate the background .
the estimation of the background is the largest source of uncertainty for the method . ] . in summary , we began by subtracting the background fit from the gaussian - smoothed spectrum .
the residuals thereby obtained were converted to units of power per hertz , multiplied by the large frequency spacing @xmath43 , and finally divided by a constant factor ( see factors in table 1 of kjeldsen et al .
2008 ) to allow for the effective number of modes in each slice @xmath43 of the spectrum .
this gave observational estimates of the radial mode amplitudes , @xmath45 , which are plotted as a thick grey line in each panel of fig .
[ fig : cenamp ] .
the thin grey lines are an estimate of the uncertainty envelope on the observed amplitudes .
these uncertainties were estimated by using as a guide the results of analyzing many independent , short segments of bison sun - as - a - star data , as will be explained below .
we have used results on a parallel analysis of the bison sun - as - a - star data as a belt - and - braces check on the procedures and results .
the complete bison timeseries that we used is 4752days long .
this was split into independent segments of length 5days ( a length similar to the @xmath0cen b timeseries ) , and the analysis procedures that were applied to the @xmath0cen b frequency power spectrum were applied to the frequency power spectrum of each segment .
the right - hand panel of fig .
[ fig : rawspecs ] shows the frequency power spectrum of one of the 5-day bison segments ( the layout and linestyles are the same as in the left - hand panel , and the spectrum was again smoothed over @xmath42 , but with @xmath46 for the sun ) . the thick grey lines in both panels of fig .
[ fig : sunamp ] show the _
mean _ of the amplitudes that were obtained from the @xmath47 independent 5-day bison segments , using the method of kjeldsen et al .
outlined above .
the thin grey lines mark plus and minus the rms of the amplitudes .
since the total epoch covered by the data spans more than one 11-yr cycle of solar activity , these uncertainties reflect variation of the estimated amplitudes from short - term stochastic variability _ and _ long - term solar cycle variability .
there will also be a contribution from the finite signal - to - noise ratio of the observations .
the fractional uncertainties in the estimated 5-day bison amplitudes are about 15% in the middle of the spectrum ; and about 25% at the extreme frequencies @xmath48 and @xmath49 , respectively . in the light of these values
, we have assumed the estimated @xmath0cen b amplitudes are all determined to a fractional precision of 20% .
this value was used to make the uncertainty envelope for the estimated @xmath0cen b amplitudes plotted in both panels of fig .
[ fig : cenamp ] . with estimates of the uncertainties on @xmath45
now in hand , next we ask the question : how robust are the @xmath45 likely to be for @xmath0cen b ?
we can obtain some insight by testing the robustness of the 5-day bison sun - as - a - star amplitudes .
this test may be accomplished by comparing the 5-day estimates with estimates from a `` peak - bagging '' analysis of the entire 4752-day bison timeseries .
the bison mode amplitudes are usually estimated using the peak - bagging fitting techniques ( e.g. , see chaplin et al .
peak - bagging involves maximum - likelihood fitting of mode peaks in the frequency power spectrum to multi - parameter fitting models , where individual mode peaks are represented by lorentzian - like functions .
the points with error bars in fig .
[ fig : sunamp ] show estimated amplitudes @xmath50 from a full peak - bagging analysis of the 4752-day bison timeseries , in which the mode peaks were fitted in the high - resolution frequency power spectrum of the complete timeseries .
since these peak - bagging estimates are extremely precise , and are also expected to be fairly accurate , they serve to provide a robust cross - check of the 5-day estimates . as we can see , the 5-day bison estimates are in good agreement with the peak - bagging bison estimates .
this comparison suggests we may also expect to have reasonable confidence in the observed @xmath0cen b amplitudes .
predictions from the pulsation computations of the p - mode velocity amplitudes , @xmath41 , are shown as dotted lines in fig .
[ fig : cenamp ] ( @xmath0cen b ) and fig .
[ fig : sunamp ] ( sun ) .
the predicted amplitudes plotted in the left - hand panels of the figures are for @xmath15 , while those in the right - hand panels are for @xmath16 .
first , let us compare the modelled and observed amplitudes for @xmath0cen b ( fig .
[ fig : cenamp ] ) .
when @xmath15 ( left - hand panel ) the model amplitudes form a very broad plateau at frequencies @xmath51 , and the level of this plateau rises slowly with increasing frequency . while the match between the predicted amplitudes and the observed amplitudes is reasonable at frequencies @xmath52 , this is demonstrably not so at higher frequencies , where the predicted amplitudes are significantly higher than the observed amplitudes .
when @xmath16 the predicted amplitudes give a better match , on average , to the observed amplitudes , but some overestimation at higher frequencies remains .
high - frequency disagreement between theory and observation is also seen in fig .
[ fig : sunamp ] for the sun .
when @xmath15 , we again see a pronounced disagreement at high frequencies between the predicted amplitudes and the observed amplitudes . when @xmath16 the level of disagreement is less pronounced , but is nevertheless still present .
let us summarize the main points from figs .
[ fig : cenamp ] and [ fig : sunamp ] : when @xmath15 , theoretically computed high - frequency velocity amplitudes for @xmath0cen b _ and _ the sun significantly overestimate the observed velocity amplitudes .
although not as severe , this overestimation persists at @xmath16 .
what might be the cause of this disagreement between theory and observation ? that is the question we turn to next .
provided we trust the mode amplitudes @xmath45 estimated from the @xmath0cen b frequency power spectrum and the 5-day bison analysis above suggests the estimates should be reasonable there are two possible ways out of the problem .
equation [ eq : v2 ] implies that : @xmath53 so , either the model damping rates @xmath25 and therefore the model linewidths @xmath33 are too weak at high frequencies to explain the observed @xmath45 , or the modelled acoustic supply rates , @xmath26 , are too strong ( or there is a combination of the two effects ) .
there is good evidence from analysis of sun - as - a - star data ( e.g. , chaplin et al . 2005 ; houdek 2006 ) that a significant part of the disagreement may come from problems computing @xmath25 ( and therefore @xmath33 ) .
we can double - check the validity of this assumption here for the sun , because we have precise bison data on the linewidths and powers of the solar p modes from the peak - bagging analysis .
these peak - bagging data may be used as a precise and accurate reference against which to check the quality of the results obtained on the short 5-day bison segments .
we base our solar check on equation [ eq : v2 ] .
again , it tells us that @xmath54 . if we take the ratio of the observed to the model - computed velocity powers , we will therefore have : @xmath55 in the above , all observed quantities for the star , which we assume come from analysis of the short 5-day segments , carry the suffix `` obs '' ; and the model - predicted quantities are suffix - free .
rearrangement of the above gives : @xmath56 let us suppose that all the problems in the modelling lie in computation of the linewidths , @xmath33 .
the implication is then , of course , that the model - predicted values of the acoustic supply rates , @xmath26 , are accurate , i.e. , @xmath57 .
equation [ eq : ratio1 ] would then simplify to @xmath58 how well do the solar model computations and 5-day observations match equation [ eq : ratio2 ] , i.e. , how well is the relation @xmath59 satisfied ?
[ fig : sunwid ] shows plots of the inferred linewidths @xmath60 of the sun ( thick grey line ) .
the @xmath61 were computed from the observed 5-day @xmath45 in section [ sec : obs ] above and the model predictions of @xmath28 and @xmath33 .
the left - hand panel shows results for when the model computations use @xmath15 ; while the right - hand panel shows results with @xmath16 .
the thin grey lines show the uncertainty envelopes on the @xmath60 , which come from the uncertainties on @xmath45 shown in fig .
[ fig : sunamp ] . also plotted as points with associated best - fitting uncertainties in fig .
[ fig : sunwid ] are the fitted linewidths @xmath62 from a full 4752-day peak - bagging analysis of the bison sun - as - a - star data . for our purposes here , we may regard the peak - bagging linewidths as being good measures of the true linewidths of the sun
. our check on whether @xmath57 therefore amounts to seeing if the 5-day @xmath61 are a good match to the @xmath62 .
this does assume that there are no significant biases in the estimated @xmath63 , as was suggested by the good agreement of the @xmath45 and @xmath50 in fig .
[ fig : sunamp ] . at a first glance
, the most striking aspect of the solar plots is indeed the encouraging level of agreement between the @xmath61 and the @xmath62 .
[ fig : widrat ] shows the comparison in more detail . here
, we have plotted the fractional differences between the 5-day linewidths and the peak - bagging linewidths . when @xmath15 , the @xmath60 over the main part of the solar p - mode spectrum are seen to be about 10% higher on average than the @xmath62 , while they are about 35% lower when @xmath16 .
the implication is that in both cases there is actually an offset between @xmath26 and @xmath64 .
however , what we can say is that changes in the differences as a function of frequency are not significant over the main part of the p - mode spectrum , given the observational uncertainties ( we can disregard the differences at the lowest frequencies , where the error bars are very large , and the 5-day spectra have insufficient resolution to give robust estimates of @xmath60 in this part of the spectrum ) .
we are therefore in a position to conclude the following : the _ shape _ in frequency of the @xmath60 is a reasonable match to the shape of the @xmath65 at the level of precision of 5-day timeseries .
the above suggests we may use equation [ eq : ratio2 ] to infer the variation of the linewidths as a function of frequency , without the need to measure those linewidths directly .
there will remain some uncertainty over the absolute calibration of the linewidths .
let us now apply equation [ eq : ratio2 ] to the data on @xmath0cen b. we have used the observed estimates of @xmath45 from the @xmath0cen b spectrum , together with the theoretical computations of @xmath41 and @xmath33 for the star , to give the inferred linewidths @xmath61 plotted in fig .
[ fig : cenwid ] .
we do not require an explicit estimate of @xmath64 , because we rely on the assumption ( verified above for the sun ) that the theoretical @xmath26 has the same shape in frequency as @xmath64 .
there is no obvious reason why we should expect the acoustic supply - rate computations , @xmath26 , to be any less valid for @xmath0cen b than they are for the sun ( this might not have been so had @xmath0cen b been somewhat hotter than the sun , or much cooler than it actually is ; see houdek 2006 ) . at the very least
we may therefore have reasonable confidence in the shapes of the inferred linewidth curves for the star , assuming any deviations from a constant scaling offset are no more severe than they are for the sun . both curves in fig .
[ fig : cenwid ] appear to show a plateau in the linewidths at @xmath66 , like that seen in the solar linewidth curve .
other obvious features of the curves a decrease of the linewidths at lower frequencies , and an increase at higher frequencies are also sun - like in nature .
it is worth stressing that the linewidth curves have very similar shapes at @xmath15 ( left - hand panel ) and @xmath16 ( right - hand panel ) .
we have also shown on fig .
[ fig : cenwid ] observational linewidth estimates , which were obtained by kjeldsen et al .
( 2005 ) in two parts of the p - mode spectrum ( points with associated uncertainties ) .
these estimates were obtained from the same data but in a different way , namely by measuring the scatter of p - mode frequencies around smooth ridges in the echelle diagram .
comparison of the inferred @xmath61 with the kjeldsen et al .
estimates allows us to place constraints on the absolute values of the linewidths for @xmath0cen b. such a comparison suggests there is actually little to choose between the @xmath60 and the kjeldsen et al .
estimates when the energy supply rates are computed using @xmath15 ( giving the inferred linewidths in the left - hand panel of fig .
[ fig : cenwid ] ) or @xmath16 ( giving the inferred linewidths in the right - hand panel ) .
however , the @xmath16 result does give a slightly better match .
inspection of the @xmath61 linewidth curve then implies that the @xmath0cen b linewidths take values of about @xmath67 at a frequency of @xmath68 , about @xmath69 at a frequency of @xmath66 , and about @xmath70 at a frequency of @xmath71 .
these linewidths turn out to agree reasonably well with the solar linewidths , if the mode frequencies are multiplied by the ratio of the acoustic cut - off frequencies of the two stars ( which in effect scales the @xmath0cen b frequencies down to those shown by the sun ) . in short , the @xmath0cen b linewidth curve is similar to a frequency - scaled version of its solar cousin .
in this paper we presented an analysis of the amplitudes and linewidths of the low - degree , sun - like p modes displayed by the star @xmath0cen
b. these data were extracted from only a few nights of doppler velocity observations collected on the star by kjeldsen et al .
( 2005 ) , and were compared with theoretical predictions of the stochastic excitation and damping rates of the p modes .
we also performed a parallel analysis of sun - as - a - star p - mode data collected by the ground - based bison .
the very long bison timeseries allowed us to validate the analysis techniques . for the sun
, we found that model predictions of the mode amplitudes were significantly larger than the observed amplitudes in the high - frequency part of the p - mode spectrum .
we were able to confirm that most of the disagreements for the sun , which set in at frequencies @xmath72 , are due to problems computing the damping rates , not the excitation rates , of the modes .
the computed damping rates must be increased to explain the observed amplitudes .
similar disagreements are seen for @xmath0cen b. here , we do not have the luxury of cross - checking the analysis with results from very long datasets , since the latter at present do not exist [ although we may hope to obtain such datasets in the future from the likes of song ( grundahl et al .
2007 ) and siamois ( mosser et al . 2007 ) ] .
however , by assuming that the disagreements for @xmath0cen b are also due largely to problems with the model - predicted damping rates , as was shown to be the case for the sun , we were able to demonstrate that the model linewidths must also be increased significantly in order to explain the observed amplitudes .
the problems for @xmath0cen b set in at frequencies @xmath1 .
the conclusions above bear on a prediction of the pulsation computations mentioned in the introduction ( section [ sec : intro ] ) : that stars cooler than about 5400k will show a broad plateau or double - hump of power in their p - mode spectra .
we showed that @xmath0cen b does not have the predicted broad plateau in its _ observed _ p - mode spectrum .
if the high - frequency damping rates are increased as they must be to resolve the disagreement between theory and observation the broad plateau or second high - frequency hump disappears in the cooler models .
finally , we also showed how , by making suitable comparisons of theory and observation , it is possible to make inference on the variation with frequency of the p - mode linewidths without directly measuring those linewidths , even if the observations come from only a few days of data .
use of independent measures of the @xmath0cen b linewidths , made in two parts of its spectrum by kjeldsen et al .
( 2005 ) , allowed us to calibrate our inferred linewidth curve for @xmath0cen b. we found that the resulting , calibrated linewidth curve is similar to a frequency - scaled version of its solar cousin , with the scaling factor equal to the ratio of the respective acoustic cut - off frequencies of the two stars .
the ratio of the frequencies at which the onset of high - frequency problems is seen in both stars is also given approximately by the same scaling factor .
wjc thanks members of the school of physics at the university of sydney for their hospitality and support during a visit when some of this work was conducted .
wjc also acknowledges the support of the school of physics and astronomy at the university of birmingham .
gh acknowledges the support of the uk science and technology facilities council , and trb acknowledges the support of the australian research council .
mosser b. , and the siamois team , 2007 , in : 1st arena conference on `` large astronomical infrastructures at concordia , prospects and constraints for antarctic optical / ir astronomy '' , eds .
n. epchtein , m. candidi , eas publication series 25 , p. 239 | this paper presents an analysis of observational data on the p - mode spectrum of the star @xmath0cen b , and a comparison with theoretical computations of the stochastic excitation and damping of the modes .
we find that at frequencies @xmath1 , the model damping rates appear to be too weak to explain the observed shape of the power spectral density of @xmath0cen b. the conclusion rests on the assumption that most of the disagreement is due to problems modelling the damping rates , not the excitation rates , of the modes .
this assumption is supported by a parallel analysis of bison sun - as - a - star data , for which it is possible to use analysis of very long timeseries to place tight constraints on the assumption .
the bison analysis shows that there is a similar high - frequency disagreement between theory and observation in the sun .
we demonstrate that by using suitable comparisons of theory and observation it is possible to make inference on the dependence of the p - mode linewidths on frequency , without directly measuring those linewidths , even though the @xmath0cen b dataset is only a few nights long .
use of independent measures from a previous study of the @xmath0cen b linewidths in two parts of its spectrum also allows us to calibrate our linewidth estimates for the star .
the resulting calibrated linewidth curve looks similar to a frequency - scaled version of its solar cousin , with the scaling factor equal to the ratio of the respective acoustic cut - off frequencies of the two stars .
the ratio of the frequencies at which the onset of high - frequency problems is seen in both stars is also given approximately by the same scaling factor . |
particle acceleration at collisionless shocks is believed to be a common phenomenon in space plasmas belonging to a diversity of environments , ranging from the inner heliosphere to the central regions of distant galaxies ( e.g. blandford and eichler 1987 ) .
evidence to support the belief that such a mechanism can efficiently produce non - thermal particles includes direct measurements of accelerated populations in various energy ranges at the earth s bow shock ( e.g. scholer et al .
1980 ; ipavich , scholer and gloeckler 1981 ; mbius et al . 1987 ; gosling et al . 1989 ) and interplanetary shocks ( for the pre - ulysses era see , for example , sarris and van allen 1974 ; gosling et al . 1981
; decker , pesses and krimigis , 1981 ; kennel et al .
1984 ; tan et al . 1988 ) , and indirect evidence provided by the observation of non - thermal electromagnetic radiation in distant cosmic sources such as supernova remnants and exotic galaxies and the very existence of cosmic rays .
the motivations for developing theories of shock acceleration are therefore obvious , and several popular approaches have emerged over the last twenty years .
one of these is the monte carlo technique of ellison and jones ( e.g. see jones and ellison , 1991 , and references therein ) which nicely describes the injection and acceleration of particles from thermal energies , at the same time minimizing the complexities of the underlying microphysics while retaining the essential ingredients of the fermi acceleration mechanism .
while theoretical work can provide considerable understanding of the diffusive shock acceleration process , the test of its appropriateness to physical environments can only be made by comparison of model predictions with observational data .
this has been done very successfully ( ellison , mbius and paschmann 1990 ; see also ellison and mbius 1987 ) in the case of the quasi - parallel portion of the earth s bow shock , comparing spectral predictions of the aforementioned monte carlo simulation with proton , he@xmath2 and c , n and o ion particle distributions obtained by the ampte experiment . using the solar wind temperatures of the different ionic species as input for the simulation upstream of the bow shock , downstream particle distributions were obtained from the simulation and compared with the ampte data : the agreement was impressive .
an important conclusion of this comparison was that the data implied that the dynamic effects of the accelerated particles are crucial to the determination of the shock structure .
this feature is a consequence of the quasi - parallel bow shock being a fairly strong shock ( i.e. , mach number about 7 ) and a comparatively efficient accelerator ( @xmath3 of the solar wind energy flux goes into superthermal particles ; see ellison , mbius and paschmann 1990 ) . since this pioneering work
, successful comparisons of other theoretical techniques with data from the earth s bow shock have been performed .
these include the hybrid plasma simulations of trattner and scholer ( 1991 , 1993 ) and of giacalone and coworkers ( described in giacalone , burgess , schwartz and ellison 1992 , 1993 ) , which yielded quite reasonable agreement with both the bow shock data and the monte carlo technique ( ellison et al . 1993 ) .
very recently , kang and jones ( 1995 ) have obtained solutions to the convection - diffusion equation for particle acceleration at parallel shocks that generate particle distributions similar to the bow shock observations and both the monte carlo and hybrid simulations .
it is natural to extend such comparisons to other shock environments .
the ulysses spacecraft has provided the opportunity for an application of this work to interplanetary ( ip ) shocks , and the high quality and comparative abundance of the spectral data ( e.g. see ogilvie , et al .
1993 ; gloeckler , et al . 1995 ; baring et al .
1995 ) obtained by the swics instrument aboard ulysses enhances the significance of such tests of theoretical models .
this paper presents results from our program of fitting the ulysses observations at oblique interplanetary shocks with monte carlo simulation output .
this has become possible following the generalization of the monte carlo technique to treat oblique shocks ( baring , ellison and jones 1993 , 1994 ) and to examine cases of strong scattering turbulence ( baring , ellison and jones 1995 ; ellison , baring and jones 1995 ) .
shorter presentations of these comparisons between theory and experiment are enunciated in baring et al .
( 1995 ) and baring , ogilvie and ellison ( 1995 ) . in this paper , our simulations
are confined to test particle implementations for simplicity , which is quite appropriate for the weak ip shocks considered here ( discussed in section 4.2 ) .
note that this is not the first attempt to fit interplanetary shock data with monte carlo simulation results : ellison and eichler ( 1984 ) fit international sun - earth explorer ( isee ) 3 proton data ( taken from gosling et al .
1981 ) of the ip shock detected on 27th august 1978 with considerable success ( see also ellison 1983 for simultaneous data fits ) .
however , their simulations were restricted to the case of truly parallel shocks , while the observed shock was only quasi - parallel .
therefore , having simulations of oblique shocks now at our disposal , we are in a position to model ip shocks more realistically than ellison and eichler ( 1984 ) and assess how the strength of particle scattering influences the acceleration process .
we note also that jones and kang ( 1995 ) and kang and jones ( 1996 ) have very recently applied their convection - diffusion equation technique to the ip shock data presented in baring et al .
( 1995 ) , again with considerable success in their theory / observation comparison . in this paper
we compare the swics measurements of proton and he@xmath2 distributions downstream of interplanetary shocks observed at 2.73.2au from the sun with the simulation output , using upstream measurements of the solar wind quantities to determine the shock parameters for the model . using a single free parameter ,
the ratio of the mean - free path for scattering to the ionic gyroradius , to describe the strength of the scattering turbulence , excellent fits of the theory to the observations are obtained for two of our chosen shocks when strong scattering is present , i.e. when .
we regard this as significant confirmatory evidence for the existence of strong turbulence in ip shocks , a result that is borne out in magnetic field data ( e.g. tsurutani , smith and jones 1983 for isee 3 observations ; vias , goldstein and acua 1984 for voyager spacecraft measurements ; balogh et al . 1993 and
section 5.2 here for ulysses magnetometer results ) .
furthermore , the simultaneous observations of and spectra are remarkably similar ( except for normalization ) when plotted on a velocity scale , indicating that the acceleration process treats particles of a given speed almost identically despite differences in mass or charge .
such a situation is expected if all particles interact elastically with a massive background ; the monte carlo technique naturally reproduces this definitive property as a consequence of its assumption of elastic scattering of particles in the local plasma frame .
our attention is confined here to shocks less than 3.5au distant from the sun , encountered by ulysses in the first half of 1991 .
few pick - up ions , in particular protons , are evident at heliospheric distances less than around 3.2au .
interstellar pick - up ions , which can participate in the acceleration process , are not incorporated in the simulation results presented here .
the reader is referred to gloeckler et al .
( 1994 , 1995 ) for interesting ulysses observations of ip shocks with strong signals from pick - up hydrogen , and . before detailing our fitting procedure and results in section 4
, we describe briefly the swics instrument in section 2 and the monte carlo simulation ( see section 3 ) .
the paper concludes with a discussion of the differences between our ip shocks and the quasi - perpendicular portions of the earth s bow shock , possible signatures of pick - up protons in one of our candidate shocks , and concurrent ulysses magnetometer data .
the solar wind ion composition spectrometer ( swics ) on the ulysses spacecraft measures the intensity and distribution functions of solar wind and suprathermal ions over a range of energy per charge from to kev / e in logarithmically spaced steps with energy resolution .
the energy - per - charge analysis is combined with post - acceleration through a potential difference of kv , and followed by time - of - flight and energy measurement to identify ionic mass and charge states . because of the double and triple coincidences used , the background levels are very low , typically less than one count per hundred hours .
this permits measurements of very low flux levels .
the instrument uses a curved multi - slit collimator , which covers in angle , fixed to the spacecraft with the spacecraft spin axis passing through one end .
thus the instrument sweeps out a cone of half - angle as it rotates .
the cone is centered approximately in the solar wind direction , and is sampled once per spacecraft revolution , which is 12 seconds in duration .
the entire velocity range of observed ionic distribution functions is covered sequentially in 64 of these samples , i.e. once every 13 minutes ; hence any variability in the ion populations is undetectable on shorter timescales . in order to restrict the counting rate of the
`` start '' channel plate in the time - of - flight system , the swics data processing unit has a mode where the voltage sweep applied to the electrostatic analyser is reversed if the counting rate exceeds a preset number .
since the voltage sweep tracks particles from high to low energy , this safety feature ( gliem et al . 1988 ) prohibits measurement of the ion distributions much below the channel with peak count rate .
this mode was in use early in the ulysses mission , indeed for the data presented here , but its use was subsequently abandoned , having been deemed unnecessary for instrumental security .
data used in this manner implicitly assumes that the detected ions are samples of a distribution that is isotropic within this cone , an expedient approximation that is extremely difficult to relax .
careful consideration of particle anisotropies is therefore essential to more refined analysis of the experimental data .
for example , the solar wind beam is always completely sampled by a small portion of the collimator , whereas ions accelerated to well above 1000 km / sec are quasi - isotropic and therefore occupy a larger solid angle than does the rotating collimator . between these cases
are suprathermal ions , whose inferred distribution depends somewhat on the precise degree of anisotropy in the plasma , an unmeasured and therefore assumption - dependent quantity .
while such intricacies of data analysis do not affect the qualitative nature of the scientific conclusions presented ( this matter will be briefly addressed in section 4.2 ) , it must be noted that it is the suprathermal ( or injected ) ions , which have the largest uncertainties due to unknown anisotropies , that are the most influential in the shock acceleration process . for more complete details of the swics instrument ,
see gloeckler et al .
this study uses measurements of protons and doubly - charged helium only ; data on pick - up ions observed at greater heliospheric distances is presented in gloeckler et al .
also , note that data at energies higher than kev have been obtained by the hi - scale experiment on ulysses ( described in lanzarotti et al .
use of such data was not possible at this stage , and is deferred to future work .
the simulation technique ( described in detail in ellison , jones and eichler 1981 ; jones and ellison 1991 ; baring , ellison and jones 1993 ; ellison , baring , and jones 1996 ) is a kinematic model , closely following bell s ( 1978 ) approach to diffusive acceleration ; it essentially solves a boltzmann equation for particle transport involving a collision operator .
particles are injected upstream and allowed to convect into the shock ( i.e. mimicking the solar wind ) , colliding with postulated scattering centers ( presumably magnetic irregularities ) along the way .
as they diffuse between the upstream and downstream regions , they continually gain energy ( for a simulation example , see fig . 3 of baring , ellison and jones 1994 ) , sampling the velocity differential across the shock ; this is the principle of the fermi acceleration mechanism .
an important property of the model is that it treats thermal particles like accelerated ones , making no distinction between them .
hence , as the accelerated particles start off as thermal ones , this technique automatically injects particles from the thermal population into the acceleration process , thus circumventing the injection problem that plagues some models of shock acceleration .
one valuable consequence of this unified treatment for thermal and non - thermal particles is that modification of the shock hydrodynamics by the accelerated population can easily be incorporated to forge a fully self - consistent model of acceleration and shock structure .
such non - linear hydrodynamics were included in the modelling of the quasi - parallel portion of the earth s bow shock ( ellison , mbius and paschmann 1990 ) , but is not important for the interplanetary shocks examined in this paper : this is evident from the spectral data presented in section 4.2 below .
shock modification is therefore not included in the present modelling .
however , we note that it may prove necessary to include such non - linear effects when performing fits over a much broader range of energies than is the case here , for example when including hi - scale data .
the scattering is assumed to be elastic and to isotropize particle momenta in the local fluid frame , i.e. to mimic large - angle collisions with field turbulence that is anchored in this frame .
such collisions might well be expected in view of the quite turbulent fields observed in heliospheric shock environments ( e.g. hoppe et al .
1981 ; tsurutani , smith and jones 1983 ; balogh et al . 1993 ) . in the simulation ,
particles are allowed to travel between scatterings for a time that is exponentially distributed about the time , where is the particle speed in the fluid frame and is the scattering mean free path in the fluid frame , which is taken to be a power - law function of momentum : @xmath4 here is the gyroradius of a particle with momentum , charge , in a magnetic field of strength ( is the speed of light and is the electronic charge ) , and is the gyroradius of a proton with the far upstream speed ( i.e. the component of the upstream flow velocity normal to the shock plane in the rest frame of the shock ) in the upstream magnetic field , .
hereafter , denotes distance measured along the shock normal , and subscripts 1(2 ) will indicate upstream ( downstream ) . here
is set proportional to the gyroradius of ( low energy ) protons of speed : the ratio is an input parameter that is of crucial importance to this study . for simplicity , is taken to be the same both upstream and downstream of the shock , a subjective choice that is largely immaterial to the nature of the results presented below .
this simple prescription is perhaps surprisingly appropriate to shocked plasma environments .
bow shock observations ( ellison , mbius , and paschmann 1990 ) indicate that , while can be deduced from ions accelerated in solar particle events ( mason , gloeckler and hovestadt 1983 ) , and is obtained from turbulence in the interplanetary magnetic field ( moussas et al .
plasma simulations ( giacalone , burgess , and schwartz 1992 ) suggest a mean free path obeying eq .
( [ eq : mfp ] ) with .
clearly a variety of microphysics can be incorporated in specifying and both pitch - angle diffusion ( small - angle scattering ) and large - angle scattering can be modelled by the simulation
. therefore the attractiveness of this approach is that the overall nature of the diffusive acceleration process can be modelled without the burden of cpu intensive computations of the underlying physical plasma mechanisms .
all results presented here are for , a choice that nicely guarantees that the ratio is independent of the particle momentum .
the assumption of elasticity of scattering in the local fluid frame is expected to be accurate except perhaps at shocks with very low alfvnic mach numbers , where the flow speed does not far exceed the alfvn velocity and the scattering centers ( i.e. alfvn waves ) move in the fluid frame with significant speeds . in the shocks
studied here , as is typically the case for interplanetary shock systems , the magnetic field is dynamically important , with alfvnic mach numbers in the range of , so that to some degree the elastic scattering assumption is violated .
nevertheless , for the shock parameters relevant to this study , the contribution to particle acceleration resulting from motion of the scattering centers in the local fluid frame is expected to be overshadowed by the first - order acceleration process that is modelled in this paper , and which principally determines the overall shape of model particle distributions .
we expect that the major properties of the acceleration mechanism are therefore well - described by elasticity of scattering , and that the deviations from this in ip shock environments will incur only effects of secondary importance .
hence , in the spirit of this incipient comparison of theory and experiment , we adopt the elastic scattering approximation , and defer examination of the higher order effects that are associated with its violation to future work . in applications of the monte carlo simulation to oblique shocks ,
particle transport is followed in the de hoffmann - teller ( ht ) frame ( de hoffmann and teller 1950 ) , where the drift electric field is transformed to zero , and the fluid flow is everywhere parallel to the magnetic field .
this approach guarantees that shock drift acceleration is automatically included .
also , since our technique does not require detailed computation of the turbulent field structure , it is much faster than hybrid or full plasma codes , and thereby achieves much larger dynamic ranges of momenta for particle distributions than is possible for the plasma simulations .
this renders the monte carlo approach ideal for spectral comparisons with observational data .
there are two variations of the simulation that are employed here : ( i ) one in which a guiding - center approximation is made ( baring , ellison and jones 1994 ) , where the details of the particles gyromotions are ignored and they conserve their adiabatic moment ( the so - called _ adiabatic approximation ) when interacting with the shock ; and ( ii ) where the exact gyrohelix particle trajectories ( e.g. see ellison , baring and jones 1995 , 1996 ) are computed without imposition of the adiabatic approximation in particle - shock interactions . results from both variations are employed in this paper .
another simplification made in the simulation is the use of a single parameter to describe the strength of the scattering turbulence ( i.e. cross - field diffusion ; see baring , ellison and jones , 1995 for details ) .
this parameter plays a crucial role in determining the efficiency of injection of particles into the non - thermal tail , as is discussed in section 4 . _ our model
is implemented in this paper as a test particle simulation , where the accelerated particles make only a small contribution to the dynamics of the fluid flow .
this is appropriate for modelling interplanetary shocks , which are of low sonic mach number and therefore comparatively inefficient at accelerating particles ( as discussed in section 4.2 ) . for a full discussion of the non - linear effects associated with high mach number shocks , see ellison , baring , and jones ( 1996 ) .
the usual hydrodynamic quantities of the downstream fluid are therefore simply determined from those in the upstream region by the full mhd rankine - hugoniot relations ( e.g. decker 1988 ) . in the ip shock applications here
the magnetic field is high enough to be dynamically important .
hence , all characteristics of the shocks depend only on the shock compression ratio , the sonic mach number , where is the total upstream temperature , is the adiabatic index , and the alfvnic mach number .
simulation output principally includes particle distributions in any desired frame of reference at any position upstream or downstream of the shock .
before comparing in detail the monte carlo particle distributions with the swics measurements downstream of our three case - study ip shocks , it is appropriate first to indicate what input parameters for the simulations were derived from various pieces of observational data .
the three shocks chosen were observed early in the ulysses mission , specifically on day 97 of 1991 ( april 7 ; denoted 91097 ) , day 118 of 1991 ( april 28 ; denoted 91118 ) , and day 147 of 1991 ( may 27 ; denoted 91147 ) , when ulysses was in the ecliptic before its encounter with jupiter .
motivation for our choices included availability of high quality spectra ( i.e. with good statistics ) in these shocks , and _ relatively simple shock structure and spectra during this phase of the mission .
these shocks appear to be representative of the shocks encountered by ulysses in the first half of 1991 .
the prominence of accelerated particles in the environs of many shocks other than those studied here suggests that the basic physical conclusions drawn in this paper can be extended to a host of other ip shocks detected during this early phase of the ulysses mission . _ of the three shocks in our case study , two ( 91097 and 91118 ) are freely propagating ( travelling ) interplanetary shocks and the other ( 91147 )
is a member of a co - rotating shock pair .
_ co - rotating interaction regions ( cirs e.g. see hundhausen 1973 for a description ) , frequently observed by ulysses ( e.g. gosling et al . 1993 , 1995 ) , result from the evolution of regions where solar wind regions of higher speed impinge upon lower speed wind that emanates from the equatorial region of the sun . at the heliocentric distances considered here ,
the collision results in a forward and reverse shock pair on either side of an interface ; the shocks propagate away from the interface , which itself moves away from the sun at a speed greater than that of the solar wind .
we have selected the forward shock on day 91147 only for our study . _ following the approach that ellison , mbius and paschmann ( 1990 ) used when comparing monte carlo spectra with observations at the quasi - parallel portion of the earth s bow shock , we use here measurements of various solar wind quantities upstream of three interplanetary shocks to set the input parameters for the monte carlo simulation runs .
the simulation then makes predictions of downstream spectra , which are then compared with swics data downstream of these ip shocks to ascertain whether consistency between experiment and theory can be achieved .
the various input quantities used in the monte carlo runs are tabulated in table 1 for the three selected ip shocks , together with their corresponding heliocentric distances .
all the quantities in the upper zone except for the shock velocities were directly determined from measurements by various instruments on ulysses ; , the ion temperatures and the shock mach numbers are inferred from the measured quantities .
specifically , for shocks 91097 and 91147 are taken from the published listing of burton et .
1992 ) , which cites a shock speed of 2 km / sec for 91118 .
this erroneous result , probably typographical , can be corrected by inferring the value of based on the other measurable quantities and the total mach number of 1.97 cited by burton et .
al . ( 1992 ) and balogh et al . (
the result is an estimate of 165 km / sec , which implicitly has at least a 10% uncertainty since the mach number can be no more accurate than the least certain measurable , which is the field obliquity . unpublished data on plasma speeds combined with the measured obliquity were used to verify the consistency of this estimate .
the values listed in table 1 for the thermal velocities were taken from gaussian fits to spectral data less than an hour upstream of the respective shocks .
the peak of the phase space distribution was fit with a form proportional to .
upstream densities and are quoted for similar timescales .
the upstream magnetic field values are averages on timescales of about a minute , and their accuracy is limited by temporal / spatial fluctuations , which are discussed in section 5 . as a consistency check on the tabulated values , we used the monte carlo code to generate solutions of the rankine - hugoniot relations for the field compression ratio from the input solar wind quantities , and these agreed to within about 8% of the ratios published by experimental teams ( balogh et al .
1995 , hoang et al .
the solar wind electron temperatures for the 91097 and 91147 shocks are published in hoang et al .
( 1995 ) , where they indicate large uncertainties ( ) .
no such listing for the 91118 event was available , so we used the best fit radial dependence temperature at 2.85 au obtained from phillips et al .
( 1995 ) , representing a length scale of the order of 0.25au .
this choice is subject to large uncertainty since the electron temperature varies on short timescales . with the exception of the shock obliquity , uncertainties for the measured
solar wind quantities largely reflect limits set by fluctuations in these quantities and the choice of temporal binning .
determination of is intrinsically problematic because it is difficult to establish the plane of the shock in a turbulent environment with measurements made only along the spacecraft trajectory .
clearly the uncertainties in most shock parameters are of the order of 10% , which must be taken into account in the following theory - data comparison .
uncertainties are not quoted in table 1 for the derived quantities , i.e. ion temperatures and mach numbers . as indicated , the values of listed in table 1 for 91097 and 91147 are derived using the measured values in the upper zone , whereas the value of 2.07 for the 91118 shock was used to estimate the shock velocity .
simple formulae for the sonic and alfvnic mach numbers that are scaled to suit typical ip shock parameters can be quickly written down : @xmath5 where is the shock speed in units of 100 km / sec , and are the proton and electron temperatures in units of k , is the upstream field strength in units of gauss , and is the proton density in , which is approximately equal to the electron density ( an 8% difference arises due to the alpha particle abundance in the solar wind ) .
the comparable values of these two mach numbers for the ip shocks studied here indicate that the fields are dynamically important ; their proximity to unity suggests that the shocks are weak and therefore are relatively inefficient particle accelerators ( e.g. compared to the earth s bow shock ) , a feature that is borne out in their spectral data .
note that we neglected to explicitly include the effects of on the dynamics of the 91097 event .
while alpha particles comprise about 4% if the solar wind population , and therefore are about 16% abundant by mass , they also comprise roughly the same percentage of the thermal energy ( i.e. 16% ; a coincidence specific to 91097 ) . hence , dynamically they contribute about the same fraction of the ram pressure ( i.e. the numerator of ) and the thermal pressure ( denominator ) and hence have marginal influence on the value of .
helium data are considered only for the april 7 , 1991 shock since they were of almost comparable quality to the proton data ; the non - thermal tails of the he spectra for the 91118 and 91147 events had insufficient statistical quality for the purposes of this analysis .
the comparisons of the monte carlo model results with the ip shock data are performed in the spacecraft frame , using boost speeds between the shock and spacecraft frames in the range of around 360430 km / sec , depending on the shock .
such speeds were deduced from the speeds associated with the thermal peaks upstream and downstream of each of the shocks , and were consistent with estimates ( subject to large uncertainties ) determined from the shock parameters listed in burton et al .
a multiplicative factor is applied to the monte carlo upstream proton distributions to render them commensurate with the solar wind density .
the fitting of the observed spectral data with the simulation output is then essentially achieved by adjusting a _ single free model parameter by visual inspection .
this parameter , whose physical significance is discussed shortly , is a scaling parameter for the simulation , being the ratio of the mean free path to the gyroradius of particles with speed ( see section 3 ) .
the scientific conclusions of this paper are amply defined via this simple visual procedure .
_ figure [ fig:91118gc ] depicts the data / theory comparison for the ip shock on april 28 , 1991 ( day 91118 ) using the guiding center ( gc ) version of the simulation .
the experimental data are taken over a period of a few hours on the downstream side of the shock where the distribution was relatively steady in time ; throughout this paper the uncertainties on the data points denote one standard deviation .
a similar comparison is made in fig .
[ fig:91118orb ] using the gyrohelix ( gh ) version of the code .
the two fits are quite similar with the gyrohelix version requiring a somewhat lower value of to approximate the data .
note that this period of data collection is chosen to be quite long in order to achieve good statistics .
the base instrumental spectral time resolution of 13 minutes corresponds to the order of 400 gyroperiods for protons behind the shock ( where gauss ) , where sec .
here is the gyrofrequency and is the field in units of gauss .
hence the much longer integration times used in the spectral comparisons of this section generally represent several thousand gyroperiods .
the properties of the shock itself are deduced from the much more frequent measurements of the magnetic field .
the swics particle distribution data used here and throughout this paper are phase space densities , for which represents a particle density ; these are essentially the count rate ( i.e. detected flux ) divided by velocity to the fourth power .
as the velocities are in units of km / sec , the distribution functions depicted here integrate over velocity to yield densities in units of particles per cubic kilometre .
since data are taken in the spacecraft frame , the thermal peak exhibits a large positive offset of a few hundred km / sec relative to the frame of the shock , reflecting the solar wind speed in the ecliptic .
the counts are dead - time corrected , but do not include complete corrections for instrumental geometry and response .
such corrections are not yet fully implemented , but are expected to produce at most about a factor of a 23 difference to the distribution .
thus these corrections are small over the dynamic range of seven orders of magnitude for the distributions in the figures of this section , and it is evident that their inclusion would not significantly alter the physical conclusions deduced from the comparisons presented .
figures [ fig:91118gc ] and [ fig:91118orb ] can be used to estimate the contribution of the accelerated population to the dynamics of the 91118 shock .
the phase space densities used here can be multiplied by a factor of to yield energy density distributions .
this weighting should be considered in the shock rest frame , where treatment of the dynamics is simplest . in this frame , for the ip shocks considered here , the downstream thermal particles typically have speeds around km / sec ( i.e. the shock speeds listed in table 1 ) . hence in transforming data from the spacecraft to the shock rest frames , clearly a ratio of around 10 in speeds is encompassed in the shock frame by the distributions in figs .
[ fig:91118gc ] and [ fig:91118orb ] .
it follows that the high speed particles have at most a few percent of the energy density contained in the thermal population , a result that extends to the other two shocks .
the accelerated populations detected by the swics instrument are therefore dynamically unimportant , thereby justifying the test particle implementation in this paper .
the agreement between experiment and theory in figs .
[ fig:91118gc ] and [ fig:91118orb ] is impressive , with only small differences in the downstream temperature ( for the guiding center implementation of the code ) and shape of the suprathermal portion of the distributions being apparent .
the fits are of comparable quality for the two types of particle convection employed , namely ( i ) exact gyrohelix determination and ( ii ) use of the guiding center approximation with the added assumption of adiabatic moment conservation .
furthermore , the data strongly constrains the possible values of ( typically ) . while these two versions of the simulation produce somewhat different estimates of the parameter due to their differing prescriptions for the way particles cross the shock , the physical conclusions drawn from these values are similar .
it appears that exact gyrohelix computations are slightly less efficient at injecting particles from the thermal population in highly oblique shocks .
the physical importance of this parameter follows from its crucial role in determining the extent of cross - field diffusion .
if and are the diffusion coefficients respectively parallel and perpendicular to the average magnetic field , then the two can be related using the standard kinetic theory result ( e.g. axford 1965 ; forman , jokipii and owens 1974 ) for a mean free path and particle gyroradius in the local fluid frame : @xmath6 non - zero values of ( i.e. non - infinite ) amount to the presence of cross - field diffusion .
this identity is tantamount to shifting a particle by approximately one gyroradius in _ any direction _ during scatterings with the turbulent magnetic field .
jokipii ( 1987 ) observed that there is no general consensus as to the form of this relationship , so that the choice of the ratio is somewhat subjective , being restricted only by the requirement that a particle move of the order of a gyroradius in a scattering .
clearly then , large values of ( in this application ) represent weak cross - field diffusion , where the field lines are almost laminar , while small values ( ) define the regime of bohm ( isotropic ) diffusion , which may be attained in strongly turbulent fields . using an analysis of particle acceleration times , ellison , baring and jones ( 1995 )
demonstrated that the monte carlo technique accurately simulates the kinetic theory result in equation [ eq : kappas ] .
besides confirming jokipii s ( 1987 ) observation that shocks of higher obliquity are _ faster accelerators , the analysis of ellison , baring and jones ( 1995 ) indicated that the efficiency of injecting particles into the fermi acceleration mechanism is a strongly _ decreasing function of shock obliquity and the strength of the scattering parameter ( see also baring , ellison and jones 1993 ) . this sensitivity is a crucial feature of the acceleration process that compels the range of in the fit to be narrow .
the origin of this dependence can be simply understood . in the de - hoffmann teller frame
, the diffusion length scale of low energy ( i.e. thermal ) particles along the field is roughly given by , while their diffusion length normal to the field is represented by the gyroradius . in oblique shocks , specifically those with ,
convective return of thermal ions to the shock from the downstream side is inhibited so that the injection into the fermi acceleration mechanism is suppressed ( baring , ellison and jones 1993 , 1994 ) . from simple considerations of geometry ,
this is clearly circumvented by cross - field diffusion when ( for being the downstream field obliquity ) and particle return to the shock is enabled ( baring , ellison and jones 1993 ; ellison , baring and jones 1995 ) .
hence it is expected that the efficiency of injection and acceleration will scale inversely with the parameter . _ _
clearly then , when and cross - field diffusion is small , the accelerated population is strongly suppressed in highly oblique shocks , contrary to the observational data .
the light solid histogram in fig .
[ fig:91118gc ] is the predicted simulation in the limit of , starkly illustrating the lack of acceleration from thermal energies .
therefore in the fits in figs .
[ fig:91118gc ] and [ fig:91118orb ] , where the high velocity portions of the histograms shift up and down rapidly relative to the thermal peaks with only modest variations in , only small values of are allowed .
it can be simply deduced that , within the context of this monte carlo model , the significant acceleration efficiency implies that strong scattering is operating in this traveling interplanetary shock .
a similar conclusion is reached in the subsequent presentation for other ip shocks , which are also efficient accelerators , and is suggestive of turbulent magnetic fields in interplanetary shock environments .
since the efficiency is extremely sensitive to the shock obliquity , the most uncertain of the observables , improved measurements will refine estimates of , but will not alter the implication of strong scattering acting in interplanetary shocks . note that hereafter , and will be used interchangeably due to their identity , which is the result of assuming in the scattering formula in equation ( [ eq : mfp ] ) .
figure [ fig:91097 ] displays the spectral comparison for the ip shock on april 7 , 1991 ( day 91097 ) using the guiding center ( gc ) version of the simulation .
the swics data are again taken over a period of a few hours ( i.e. thousands of gyroperiods ) on the downstream side of the shock , but this time , the data permitted presentation of the phase space distribution also .
the relative normalizations of the upstream distributions for protons and were deduced from the abundance ratio implied by table 1 , and are not free parameters in the model .
the fit is clearly impressive for the proton spectrum , and is equally encouraging for the alpha particles considering that just _
one parameter ( ) is used to reproduce both the spectral shape _ and the relative populations of both accelerated ion species . again , the relatively low value of implies strong scattering . _
_ at first sight , it seems remarkable that just one choice of can fit the the two species so nicely .
however , this is consistent with the observed spectra being essentially independent of mass and charge , i.e. depending only on velocity kinematics . since the inferred upstream temperature is about four times that of the protons , the input for the simulation is roughly independent of mass and charge .
if , in addition , all particles interact nearly elastically with a massive background , namely magnetic field turbulence `` anchored '' in the fluid frame , the acceleration process , and therefore the spectral shape and injection efficiency , will depend only on particle velocities . since the monte carlo simulation makes just this elastic assumption ( i.e. in eq .
[ [ eq : mfp ] ] ) , it is not surprising that we can accurately model the observations in this test - particle regime .
given such dependence on the kinematics , the relative normalizations of the and data downstream are largely determined by their relative abundances upstream .
the results in fig .
[ fig:91097 ] indicate the high accuracy of such a scattering assumption in traveling interplanetary shocks when interstellar pick - up ions are unimportant . another noteworthy feature of this comparison is that this is the first time that the upstream abundance ratio _ of thermal ions _ has been used to set relative population levels in a theory / data comparison of acceleration at shocks somewhere in the heliosphere . due to the inability of the 3d - plasma instrument on the ampte / irm spacecraft to measure elements heavier than hydrogen , the study of the parallel portion of the earth s bow shock by ellison , mbius and paschmann ( 1990 ) did not have upstream measurements at thermal energies of elements heavier than hydrogen ( downstream observations of non - thermal alpha particles and heavier elements were made above 10 kev / q by the suleica instrument ) , and so could not determine relative abundances in the local solar wind . it should be noted that the values of obtained in the fits presented here differ somewhat from previous expositions of this data / theory comparison , notably in baring et al .
( 1995 ) and baring , ogilvie and ellison ( 1995 ) .
this difference arises because more refined estimates of the magnetic field and solar wind densities have been used here than in the earlier works , with finite electron temperatures being included , thus revising the input parameters for the shock simulation .
the shock strength is somewhat weaker than before , so that estimates for are now lower .
we note that recently with the aid of shock parameters from baring et al .
( 1995 ) , kang and jones ( 1995 ) and kang and jones ( 1996 ) have obtained , comparable fits to the proton data for the 91097 and 91118 shocks using their convection - diffusion equation model of the fermi acceleration mechanism . as a contrast to the above data comparisons , it is instructive to study a shock that was encountered at a later stage of ulysses passage away from the sun . figure [ fig:91147 ] displays the comparison for the ip shock on may 27 , 1991 ( day 91147 ) , again using the guiding center ( gc ) version of the simulation , and proton data accumulated over several hours .
the choice of is solely influenced by the fit to the high velocity portion ( 10001800 km / sec ) of the data , and again suggests strong scattering .
however , the value of is somewhat higher in this shock than in the previous two , due to its lower obliquity .
the inaccuracy of the comparison in the suprathermal region contrasts strongly with the previous fits , and is probably indicative of the presence of another proton population ( i.e. interstellar pick - up ions ) , as is discussed below , rather than representing the failure of the fermi acceleration mechanism .
the presence of such pick - up ions , with speeds ranging between zero and twice the solar wind speed , may account for the discrepancy , since the simulation deficiency in fig .
[ fig:91147 ] is in precisely this velocity range .
the present implementation of the monte carlo simulation models injection and acceleration of particles drawn only directly from the thermal population , however , it can be easily adapted to describe additional pre - accelerated populations . to summarize , in two of our three case studies excellent fits
have been obtained , though some small discrepancies exist at thermal and suprathermal energies .
the slight deficiencies of the simulation predictions relative to the data in figs .
[ fig:91118gc][fig:91097 ] ( particularly for the guiding center implementations of the simulation ) could be due to subtleties in the microphysics that are omitted from the simulation , such as the presence of a cross - shock potential that would modify the acceleration efficiency , or deviations from the scattering law in equation ( [ eq : mfp ] ) .
furthermore , there may be some suprathermal population present at a level smaller than in fig .
[ fig:91147 ] that may distort the distributions from the simple simulation predictions .
in addition , the omission from the simulation of non - linear modifications of the shock hydrodynamics by the accelerated population ( generally small for ip shocks ) might lead to a slightly incorrect estimation of the downstream temperature .
such non - linear effects smooth the shock velocity profile upstream ( decelerating the flow ) and increase the overall compression ratio , thereby modifying the width of the downstream thermal distribution ( e.g. ellison , baring and jones 1996 ) .
it must be remembered that the fits attempted here are confined to relatively low particle speeds , and may be degraded or modified somewhat by a future consideration of data at higher energies , such as that obtained by hi - scale , where the test - particle fermi acceleration distributions assume well - known ( e.g. see blandford and eichler 1987 ; jones and ellison 1991 ) power - laws .
perhaps the most remarkable aspect of the observations presented here is simply that low mach number , highly oblique , interplanetary shocks accelerate significant fractions of thermal solar wind protons and alpha particles to non - thermal energies , and do so with comparable efficiencies .
the smooth transition from the thermal portion of the distribution to a non - thermal tail seen in figs .
[ fig:91118gc][fig:91097 ] , and , to a lesser extent , [ fig:91147 ] , is a distinctive signature of the direct injection and acceleration of thermal particles , given that no such transition is seen less than an hour upstream of these shocks .
this close association of thermal and non - thermal ions in the neighbourhood of these quasi - perpendicular ip shocks is decidedly different behavior from that observed at highly oblique portions of the earth s bow shock , and also from plasma simulation results , which show no significant acceleration of thermal ions at all in shocks of high obliquity . at quasi - perpendicular parts of the bow shock ,
reflected field - aligned beams are typically seen propagating back from the shock , and these beams have a much lower /proton ratio than the ambient solar wind ( e.g. ipavich et al .
1988 ; fuselier and thomsen 1992 ) .
these beams are believed to originate when particles specularly reflect off the shock ( see thomsen 1985 for a review ) , with protons reflecting more easily than .
downstream from the bow shock , the helium remains at much lower relative density compared to the protons ( fuselier and schmidt 1994 ) , in contrast to what we see here downstream from ip shocks .
the radically different behavior of highly oblique regions of the earth s bow shock and the ip shocks considered here is probably the result of the very different relative length scales involved .
both travelling ip shocks and those associated with cirs have radii of curvature that are much larger than that of the earth s bow shock , at least for large distances from the sun .
the diffusion length of suprathermal particles at the bow shock is comparable to the bow shock radius , and reflected or accelerated particles that are produced near the tangent point of the solar wind magnetic field with the shock , where the field has very little turbulence , are quickly swept away and do not have the chance to generate local turbulence .
an ip shock , on the other hand , is much larger compared to diffusion length scales so that particles generated at quasi - perpendicular shock environments can remain in the system long enough to maintain much larger levels of turbulence .
an idea of the scale lengths associated with these shock systems can be gained from the gyroradii of the detected particles . at non - relativistic velocities ,
a particle of mass number and charge state has a gyroradius of , which can be translated into units of relevance to the plasma environment considered here : @xmath7 where is the field strength expressed in units of gauss , and is the particle energy in kev .
clearly , since the ip shock velocities are of the order of 100300 km / sec , thermal protons have gyroradii around km , and even the most energetic ions detected by swics have gyroradii ( and therefore mean free paths ) much less than the typically huge ip shock scale ( km ) .
this behaviour differs strongly from that at the earth s bow shock , which has particles accelerated to comparable energies but a much smaller radius curvature , of the order of km .
an important byproduct of these scale length estimates is that the use of a planar shock model in this study is both expedient and justifiable for interplanetary shock applications .
the angle between the shock normal and the magnetic field , , is constant for much longer distances along the faces of such shocks than typical particle mean free paths . as the efficiency of shock acceleration is a strong function of ( e.g. baring , ellison and jones 1993 ; ellison , baring and jones 1995 , 1996 ) , interplanetary shocks are very suitable for comparison with theory , since the observations at a given point with given are not contaminated by particles accelerated at locales of the shock where is different . in regards to full plasma simulations , to our knowledge , all published 1-and 2-dimensional plasma simulation results for shocks with @xmath8 fail to show significant acceleration of thermal ions ( e.g. liewer , goldstein , and omidi 1993 ; liewer , rath , and goldstein 1995 ; kucharek and scholer 1995 ) . we believe this to be a consequence of an unphysical suppression of cross - field diffusion in simulations with field geometries of less than three dimensions ( jokipii , kta , and giacalone 1993 ; jokipii and jones 1996 ) .
since cross - field diffusion is essential for allowing downstream particles to recross the shock , as noted above , its suppression will eliminate particle injection and possibly also subsequent self - generation of field turbulence .
the kinematic monte carlo technique simply incorporates a prescription of such particle diffusion via the parameter , and hence is inherently three - dimensional in its description of particle motions .
the excellence of the data fits to the 91097 and 91118 shocks , and the comparatively poor fit to the 91147 data , could be considered consistent results if an extra population of seed particles for the acceleration mechanism were present in the 91147 event , thereby circumventing the deficiency seen in fig . [ fig:91147 ] .
interstellar pick - up ions , which are neutral when they enter the heliosphere , are ionized by the solar uv flux or by charge exchange with the solar wind ( e.g. gloeckler and geiss 1996 ) .
they naturally present themselves as candidates for such a population , especially since they have a velocity distribution ranging between zero and roughly twice the solar wind speed ( e.g. vasyliunas and siscoe , 1976 ) , i.e. exactly where the monte carlo model underpredicts the accelerated population for the 91147 event .
this population is particularly interesting since the acceleration efficiency of suprathermal ions exceeds that of the thermal particles .
the density of pick - up ions in the heliosphere is not uniform due to the gravitational focusing of the ion streams in the solar neighbourhood ; in fact , the degree of focusing differs for protons and ( vasyliunas and siscoe , 1976 ) so that the radial variations of the densities of these pick - up species are not the same .
such density gradients are also influenced by variations in the ionization fractions .
helium atoms , whose ionization potential is high , penetrate as close as 0.5au , whereas hydrogen gas is highly ionized by the time it is less than a few au from the sun .
hence the density gradient for is much greater in the 2au5au range than for singly - ionized helium .
note that while pick - up ions are much more prevalent than pick - up ions , is the dominant charge state of helium in the solar wind plasma .
fortunately , due to measurements by swics , as presented by gloeckler and geiss ( 1996 ) , we have a strong indication of the variations in pick - up ion backgrounds at times when the spectra presented in this study were obtained .
they indicate that for , a slow decline in count rate was observed as ulysses made its way from 1au to beyond jupiter ( from 2 counts to 0.8 counts per 12 second instrumental sweep ) , whereas the flux ratio of increased by an order of magnitude between 3au and 5au .
consequently , it can be inferred that the proton pick - up density increased dramatically during this period of data collection , and specifically , that an increase of a factor of occurred between the 91097 and 91147 events .
the pick - up ion count rate in the 91147 shock was actually a significant fraction of the proton count rate at comparable suprathermal energies .
this implies that the data / theory comparison presented here can comfortably accommodate the hypothesis that there is a very small contribution of pick - up protons for the 91097 shock observed at au , but that this contribution has risen dramatically by day 147 in 1991 , corresponding to au .
we therefore contend that pick - up ions play an important role in the acceleration process in the 91147 event , and that our omission of them from the monte carlo model is primarily responsible for the poorness of the fit .
the question of plausibility of the values for obtained by our fits of the spectral data naturally arises .
precedence for such low exists in the literature .
forman , jokipii and owens ( 1974 ) determined the ratio of components of the diffusion tensor along and orthogonal to the field using quasi - linear diffusion theory applied to measurements of the interplanetary magnetic field power spectrum made by the mariner 4 spacecraft at locations near earth and also near mars .
they computed a ratio of , which corresponds to using the kinetic theory formula in equation ( [ eq : kappas ] ) .
similar estimates are obtained at much larger solar distances ( 20 au ) using somewhat different analyses of magnetic field data obtained by voyager 2 , to compute ( see valdes - galicia , quenby and moussas 1992 ) , and also field data acquired by pioneer 11 to determine ( see moussas et al .
while these results were obtained using measurements made near the ecliptic plane , it is anticipated that similar estimates will result near the heliospheric poles from information recently acquired by the ulysses mission .
the values for determined from the spectral comparisons here are therefore not extraordinary by heliospheric standards . yet the diffusion coefficients computed by the analyses of forman , jokipii and owens ( 1974 ) , valdes - galicia , quenby and moussas ( 1992 ) , and moussas et al .
( 1992 ) are all in the range of about @xmath10 sec@xmath11@xmath10 sec@xmath11 , implying a mean free path of around 1100 au for particles of speeds km / sec ( ) . these far exceed the estimate of a few times km ( au ) obtainable from equation ( [ eq : rgforms ] ) for similar for the ip shocks of studied here .
this dramatic difference is very probably not a conflict , but rather reflects a huge disparity in spatial scale of diffusion in the general interplanetary medium as opposed to diffusion in the environs of interplanetary shocks .
it must be emphasized that the spectral estimates of determined here are very sensitive to the shock obliquity , a quantity that is difficult to measure accurately in space plasma shocks .
the diffusion length scale associated with interplanetary shocks can be examined by consideration of the magnetic field turbulence in their environs . fig .
[ fig : bcomp ] displays the magnetic field data measured by the ulysses magnetometer in the temporal vicinity of the 91097 ip shock on 1 sec ( up to the 5.5 hour mark ) and 2 sec ( more than 5.5 hours into april 7 , 1991 ) timescales .
the orthogonal rtn components , , , of magnetic field and the total field strength comprise the data streams . here
the -direction points radially away from the sun , the -axis is perpendicular to the -axis and lying in the plane formed by the radial vector to the spacecraft and the solar rotation axis , and the -direction completes the orthogonal coordinates such that r , t , and n form a right - handed system . in fig .
[ fig : bcomp]a the field quantities are presented for the period of an hour on either side of the 91097 shock to illustrate properties coupled to the shock on smaller scale lengths .
the field turbulence appears much larger on the downstream side of the shock ( the right hand side of the plot , at later times ) , but this is somewhat deceptive since there the field has been enhanced due to the shock compression ( see the bottom panel ) .
the level of turbulence relative to the mean ambient field is physically a more meaningful quantity , and will be addressed shortly .
also in fig .
[ fig : bcomp]a there is evidence of sharp discontinuities in field components downstream .
these are not shocks since they are not accompanied by steep density gradients , yet they are representative of spatial / temporal structural features generally seen in the turbulent environs of ip shocks .
field data streams for the other candidate shocks look very similar to fig .
[ fig : bcomp ] . in fig .
[ fig : bcomp]b the field data are displayed for the 4.5 hour period downstream of the shock during which the spectral data in fig . [ fig:91097 ] were collected , revealing no significantly different behaviour from the downstream section in fig .
[ fig : bcomp]a . a more illuminating presentation of the turbulent magnetic field is given in fig . [ fig : turbulence ] , which exhibits the field _ turbulence measure derived from the measured data in fig .
[ fig : bcomp]a . the turbulence measure , which can formally be related to the magnetic reynolds number ( e.g. see boyd and sanderson 1969 , p. 126
) , was obtained by comparing vector components of the field in successive time bins ( labelled and ) according to @xmath12 on 1s , 10s and 100s binning timescales .
while the choice of turbulence measure is subjective , this definition is informative , and virtually coincides with definition of the normalized variance that is frequently used in the space plasma literature : see forsyth et al .
( 1996 ) for variances during out - of - the - ecliptic passages of ulysses , which are generally taken as averages over longer terms rather than successive data bins .
the remarkable feature of fig . [ fig : turbulence ] is the similarity of the turbulence measure on both sides of the shock : the shock enhances the field and the turbulence by similar factors so that the relative level of turbulence remains more or less the same .
it is unclear whether or not such a compressive enhancement is caused by shock - generated turbulence .
it should be noted that any differences in turbulence measure between the upstream and downstream regions would argue for different values for either side of the shock .
however , for the monte carlo technique , the key acceleration properties of spectral shape and injection efficiency depend mostly on the value of in the downstream region , so that only this value need be specified .
the similarity of the turbulence measures throughout the shock transition further indicates that the added complication of treating such non - uniformities in is unneccesary . _ the turbulence measure averages around 5% on second timescales , which is probably around the border of the quasi - linear regime , and increases to an average of 16% on 100 second timescales , reflecting a non - uniform power spectrum .
this latter level of turbulence is quite substantial and is suggestive that quasi - linear theory may not be applicable on these larger timescales .
note that at the shock on the 1 and 10 second timescales .
the binning is truncated at the onset of data gaps in fig .
[ fig : bcomp]a . without a more sophisticated analysis ( e.g. see forman , jokipii and owens 1974 ) , it is difficult to conclude whether or not these turbulence measures are closely consistent with the spectrally - determined estimates of . however , it is possible to rule out extremities .
the bohm diffusion limit is expected to arise when , which is not reached in this data . yet
the levels of turbulence are high enough that the plasma is probably much closer to the bohm diffusion limit than to a state . since the proton gyroperiod is sec , then protons behind the shock ( where gauss ) sample fluctuation timescales of the order of a few seconds .
this provides a rough guide to the level of turbulence expected to dominate the particle diffusion .
another consistency check concerns the acceleration timescale : are the ip shocks old enough to permit the diffusive fermi mechanism to accelerate ions to the energies observed ?
numerous theoretical treatments of diffusive acceleration timescales exist in the literature , and we refer the reader to jones and ellison ( 1991 ) for a review . for the monte carlo technique used here , where the mean free path is proportional to the ion rigidity , an explicit form for the acceleration time to a given energy is presented in equation ( 4 ) of ellison , baring and jones ( 1995 ) , and order of magnitude estimates can be gleaned from fig . 3 of that work . specifically , the acceleration time is approximately proportional to , and yields values of the order of minutes for energization to 10 kev , the typical upper range of swics observations , and timescales of the order of a few hours for the mev energies detected by hi - scale .
both these timescales are considerably shorter than the multi - day duration of passage of our candidate ip shocks through the heliosphere , thereby causing no temporal consistency problems for the fermi mechanism .
it is instructive to remark upon the upstream ip spectral structure and the simulation predictions .
the swics data upstream of the three candidate shocks are dominated by the beam - like distribution of the solar wind that is thermally - broadened , containing minimal evidence of suprathermal ions .
hence , the measured upstream densities of and ( for event 91097 ) are representative of the solar wind properties , containing virtually no pollution from non - thermal ions . yet
, there is marginal evidence in the swics data for particles seen upstream at low flux levels but with speeds considerably in excess of the solar wind speed .
no continuity of the distribution function is apparent , and it could even be described as consisting of two components of disparate velocities .
the upstream non - thermal `` component '' appears only near the shock .
evidence for such a non - thermal population that wanes with distance from the shock has been seen upstream of interplanetary shocks by previous instruments . gosling et al .
( 1984 ) display distributions that diminish in intensity upstream of ip shocks encountered by isee 1 and 2 , and that exhibit upstream non - thermal distributions that are flatter than their downstream counterparts .
tan et al .
( 1989 ) present isee 3 proton data that impressively exhibits a non - thermal distribution that flattens at greater distances upstream of an ip shock passage .
such properties are consistent with the theory of fermi acceleration .
lee s ( 1982 ) analytic modelling of ion acceleration and wave excitation at the earth s bow shock predicts that low energy suprathermal ions disappear at short distances upstream of the shock , with a depletion between the solar wind and high speed particles that grows both in size and range of energy as the distance upstream increases .
our monte carlo model displays virtually identical properties ( baring , ellison and jones 1994 ; ellison , baring and jones 1996 ) , which are a direct consequence of low energy suprathermal particles having shorter mean paths than ions of higher energies , and therefore having greater difficulty diffusing against the convective power of the solar wind to a given distance upstream of the shock .
the isee data add weight to the hypothesis that fermi acceleration is acting at interplanetary shocks .
the spectral comparison between swics measurements of ion distributions downstream of three interplanetary shocks observed at 2.73.2au from the sun , and our kinematic monte carlo model for diffusive shock acceleration , provides impressive fits of the data for two events ( detected on days 91097 and 91118 ) with theory by effectively varying just a single parameter , .
the low values ( ) of this parameter indicate strong cross - field diffusion ( not far from the bohm diffusion limit ) is operating in these systems , a property that is commensurate with magnetic field turbulence in the general interplanetary medium , and is consistent with self - generated turbulence predicted in diverse shock models .
furthermore , the turbulent magnetic field data presented for the 91097 shock appears , without sophisticated analysis , to be concordant with low values of deduced from the spectral considerations .
it is anticipated that the inference of strong diffusion normal to the mean ambient magnetic field direction applies to many of the ip shocks encountered by ulysses during the first half of 1991 , since they exhibit similar evidence of prominent downstream non - thermal ion populations .
the monte carlo model simulates these complex shock systems in an elegantly simple but elucidating manner , using the measured upstream solar wind quantities as input for the theoretical simulation .
it successfully models proton and he@xmath2 distributions in the 91097 event , both in spectral shape and normalization , using the _ same value of .
this is because the acceleration process at ip shocks seems dependent only on the velocity of the species , not its mass or charge .
such a kinematic property is not expected if reflection at the shock layer is an important part of the injection process ( e.g. at the quasi - perpendicular portion of the earth s bow shock ) .
however , it is an inherent assumption of the monte carlo technique in the test - particle acceleration regime that is appropriate for the generally weak ip shocks .
the theory makes clear , concise and eminently testable predictions about the fall - off of the accelerated population upstream of shocks ; evidence for this phenomenon occurring at interplanetary shocks already exists in the literature . in the case of the may 27 , 1991 shock , the model underpredicts the proton data in the 5001000km / sec range .
we propose that this discrepancy might be rectified if interstellar pick - up protons , which were observed at significant levels during late may 1991 and at subsequent stages of the ulysses mission , are included in the simulation as an extra population seeding the acceleration process .
anticipated developments of the present work include extension of the comparison to other shocks , and analysis of the field data for the three shocks considered here , using quasi - linear theory , to obtain alternative estimates for the ratio . _ we thank drs . tom jones and len burlaga for helpful comments following careful reading of the manuscript .
we are especially grateful dr .
george gloeckler , prof .
johannes geiss , and the many individuals at the university of maryland , the university of bern , the technische universitt braunschweig , the max - planck - institut fr aeronomie , and goddard space flight center who have contributed to the swics experiment .
we have benefited from discussions with dr .
daniel berdichevsky of hughes / stx , who has been responsible for production of the data .
we also thank dr .
frank jones , whose encouragement led to our theory / experiment collaboration and the instigation of this project .
this work was supported by nasa / jpl contract 955460 , the swiss nsf , the bundesminister fr forschung and technologie of germany , and the nasa space physics theory program .
dce thanks the service dastrophysique , cen - saclay , observatoire de paris meudon , and cnet / cetp ( issy - les - moulineaux ) for hospitality during much of the period in which work for this paper was completed .
axford , w. i. 1965 , planet .
space sci .
@xmath1313 , 115 .
balogh , a. , forsyth , r. j. , ahuju , a. , southwood , d. j. , smith , e. j. , and tsurutani , b. t. 1993 , adv .
space sci .
@xmath1313 , ( 6 ) 15 .
balogh , a. , gonzalez - esparza , j. a. , forsyth , r. j. , burton , m. e. , goldstein , b. e. , smith , e. j. , and bame , s. j. 1995 , @xmath1372 , 171 .
baring , m. g. , ellison , d. c. and jones , f. c. 1993 , @xmath13409 , 327 . baring , m. g. , ellison , d. c. and jones , f. c. 1994 , @xmath1390 , 547 . baring , m. g. , ellison , d. c. and jones , f. c. 1995 , adv .
space sci .
@xmath1315 ( 8/9 ) , 397 . baring , m. g. , ogilvie , k. and ellison , d. c. 1995 , proc .
24th international cosmic ray conference , ( rome ) @xmath133 , 229 . baring , m. g. , ogilvie , k. , ellison , d. c. , and forsyth , r. 1995 , adv .
space sci .
@xmath1315 ( 8/9 ) , 388 .
bell , a. r. 1978 , @xmath13182 , 147 .
blandford , r. d. and eichler , d. 1987 , phys . rep .
@xmath13154 , 1 .
boyd , t. j. m. and sanderson , j. j. 1969 , plasma dynamics , ( barnes and noble , new york ) .
burton , m. e. , smith , e. j. , goldstein , b. e. , balogh , a. , forsyth , r. j. , and bame , s. j. 1992 , @xmath1319 , 1287 .
decker , r. b. 1988 , @xmath1348 , 195 .
decker , r. b. , pesses , m. e. , and krimigis , s. m. 1981 , @xmath1386 , 8819 .
de hoffmann , f. and teller , e. 1950 , @xmath1380 , 692 .
eichler , d. : 1984 @xmath13277 , 429 .
ellison , d. c. 1983 , in proc .
18th international cosmic ray conference , ( bangalore ) @xmath1310 , 108 .
ellison , d. c. , baring , m. g. and jones , f. c. 1995 , @xmath13453 , 873 .
ellison , d. c. , baring , m. g. and jones , f. c. 1996 , , in press .
ellison , d. c. and eichler , d. 1984 , @xmath13286 , 691 .
ellison , d. c. , giacalone , j. , burgess , d. , and schwartz , s. j. 1993 , @xmath1398 , 21,085 .
ellison , d. c. , jones , f. c. , and eichler , d. 1981 , j. geophys .
@xmath1350 , 110 .
ellison , d. c. , and mbius , e. 1987 , @xmath13318 , 474 .
ellison , d. c. , mbius , e. , and paschmann , g. 1990 , @xmath13352 , 376 .
forman , m. a. , jokipii , j. r. and owens , a. j. 1974 , @xmath13192 , 535 .
forsyth , r. j. , horbury , t. s. , balogh , a. and smith , e. j. 1996 , @xmath1323 , 595 .
fuselier , s. a. , and schmidt , w. k. h. 1994 , @xmath1399 , 11,539 .
fuselier , s. a. , and thomsen , m. f. 1992 , @xmath1319 , 437 .
giacalone , j. , burgess , d. and schwartz , s. j. 1992 , in proc .
26th eslab symposium , study of the solar - terrestrial system ( esa , noordwijk ) p. 65
giacalone , j. , burgess , d. , schwartz , s. j. and ellison , d. c. 1992 , @xmath1319 , 433 .
giacalone , j. , burgess , d. , schwartz , s. j. and ellison , d. c. 1993 , @xmath13402 , 550 .
gliem , f. , et al .
1988 , solar wind ion composition spectrometer data processing unit , technical report of the technische universitt braunschweig , institut fr datenverarbeitungsanlagen .
gloeckler , g. , et al .
1992 , @xmath1392 , 267 .
gloeckler , g. , and geiss , j. 1996 , in cosmic winds and the heliosphere , eds .
j. r. jokipii , et al .
arizona press , tucson ) , in press .
gloeckler , g. , geiss , j. , roelof , e. c. , fisk , l. a. , ipavich , f. m. , ogilvie , k. w. , lanzerotti , l. j. , von steiger , r. , and wilken , b. 1994 , @xmath1399 , 17,637 .
gloeckler , g. , roelof , e. c. , ogilvie , k. w. , berdichevsky , d. b. 1995 , @xmath1372 , 321 .
gosling , j. t. , ashbridge , j. r. , bame , s. j. , feldman , w. c. , zwickl , r. d. , paschmann , g. , sckopke , n. , and hynds , r. j. 1981 , @xmath1386 , 547 .
gosling , j. t. , bame , s. j. , feldman , w. c. , paschmann , g. , sckopke , n. , and russell , c. t. 1984 , @xmath1389 , 5409 .
gosling , j. t. , thomsen , m. f. , bame , s. j. , and russell , c. t. 1989 , @xmath1394 , 3555 .
gosling , j. t. , bame , s. j. , mccomas , d. j. , phillips , j. l. , pizzo , v. j. , goldstein , b. e. , and neugebauer , m. 1993 , @xmath1320 , 2789 .
gosling , j. t. , bame , s. j. , mccomas , d. j. , phillips , j. l. , pizzo , v. j. , goldstein , b. e. , and neugebauer , m. 1995 , @xmath1372 , 99 .
hoang , s. , lacombe , c. , mangeney , a. , pantellini , f. , balogh , a. , bame , s. j. , forsyth , r. j. and phillips , j. l. 1995 , adv .
space sci .
@xmath1315 ( 8/9 ) , 371 .
hoppe , m. m. , russell , c. t. , frank , l. a. , eastman , t. e. , and greenstadt , e. w. 1981 , @xmath1386 , 4471 .
hundhausen , a. j. 1973 , @xmath1378 , 1528 .
ipavich , f. m. , scholer , m. and gloeckler , g. 1981 , @xmath1386 , 11,153 .
ipavich , f. m. , gloeckler , g. , hamilton , d. c. , kistler , l. m. , and gosling , j.t .
1988 , @xmath1315 , 1153 .
jokipii , j. r. 1987 , @xmath13313 , 842 .
jokipii , j. r. , and jones , f. c. 1996 , , submitted .
jokipii , j. r. , kta , j. , and giacalone , j. 1993 , @xmath1320 , 1759 .
jones , f. c. and ellison , d. c. 1991 , @xmath1358 , 259 .
jones , t. w. and kang , h. 1995 , proc .
24th international cosmic ray conference , ( rome ) @xmath133 , 245 .
kang , h. and jones , t. w. 1995 , @xmath13447 , 944 .
kang , h. and jones , t. w. 1996 , in press .
kennel , c. f. , et al .
1984 , @xmath1389 , 5436 .
kucharek , h. and scholer , m. 1995 , @xmath13100 , 1745 .
lanzerotti , l. j. , gold , r. e. , anderson , k. a. , armstrong , t. p. , lin , r. p. , krimigis , s. m. , pick , m. , roelof , e. c. , sarris , e. t. , simnett , g. m. , and frain , w. e. 1992 , @xmath1392 , 349 .
lee , m. a. 1982 , @xmath1387 , 5063 .
liewer , p. c. , goldstein , b. e. , and omidi , n. 1993 , @xmath1398 , 15,211 .
liewer , p. c. , rath , s. , and goldstein , b. e. 1995 , @xmath13100 , 19,809 .
mason , g. m. , gloeckler , g. , and hovestadt , d. 1983 , @xmath13267 , 844 .
mbius , e. , scholer , m. , sckopke , n. , lhr , h. , paschmann , g. , and hovestadt , d. 1987 , @xmath1314 , 681 .
moussas , x. , quenby , j. j. , theodossiou - ekaterinidi , z. , valdes - galicia , j. f. , drillia , a. g. , roulias , d. and smith , e. j. 1992 , @xmath13140 , 161 .
ogilvie , k. w. , geiss , j. , gloeckler , g. , berdichevsky , d. b. , and wilken , b. 1993 , @xmath1398 , 3605 .
phillips , j. l. , bame , s. j. , gary , s. p. , gosling , j. t. , scime , e. e. and forsyth , r. j. 1995 , @xmath1372 , 109 .
sarris , e. t. , and van allen , j. a. 1974 , @xmath1379 , 4157 .
scholer , m. , ipavich , f. m. , gloeckler , g. , and hovestadt , d. 1980 , @xmath1385 , 4602 .
tan , l. c. , mason , g. m. , gloeckler , g. , and ipavich , f. m. 1988 , @xmath1393 , 7225 .
tan , l. c. , mason , g. m. , gloeckler , g. , and ipavich , f. m. 1989 , @xmath1394 , 6552 .
thomsen , m. f. 1985 , in collisionless shocks in the heliosphere : reviews of current research , geophys .
ser . , 35 , eds .
b.t . tsurutani and r.g .
stone ( agu , washington , dc ) , p. 253 .
trattner , k. j. , and scholer , m. 1991 , @xmath1318 , 1817 .
trattner , k. j. , and scholer , m. 1993 , ann .
geophysicae @xmath1311 , 774 .
tsurutani , b. t. , smith , e. j. , and jones , d. e. 1983 , @xmath1388 , 5645 .
valdes - galicia , j. f. , quenby , j. j. , and moussas , x. 1992 , @xmath13139 , 189 .
vasyliunas , v. m. , and siscoe , g. l. 1976 , @xmath1381 , 1247 .
vias , a. f. , goldstein , m. l. , and acua , m. h. 1984 , @xmath1389 , 3762 .
-0.3truecm @xmath14 & & 153 & 165 & 132 & & ( 1 ) \cr \vsptwo \tbn & & 77^\circ\pm 7^\circ & 75^\circ\pm 7^\circ & 55^\circ\pm 16^\circ & & ( 2 ) \cr \vsptwo v_{th}(h^+ ) \ ; [ \hbox{km}\ , s^{-1 } ] & & 23\pm 2 & 29\pm 3 & 50\pm 2 & & ( 3 ) \cr \vsptwo v_{th}(he^{2 + } ) \ ; [ \hbox{km}\ , s^{-1 } ] & & 26\pm 3 & \dots & \dots & & ( 3,\ , 4 ) \cr \vsptwo n_{h^+ } \ ; [ \hbox{cm}^{-3 } ] & & 1.4\pm 0.1 & 0.37\pm 0.03 & 3.0\pm 0.3 & & \cr \vsptwo n_{he^{2 + } } \ ; [ \hbox{cm}^{-3 } ] & & 0.06\pm 0.005 & \dots & \dots & & ( 4)\cr \vsptwo b_1\ ; [ 10^{-5}\;\hbox{gauss } ] & & 2.05\pm 0.05 & 1.95\pm 0.05 & 4.30\pm 0.05 & & ( 5 ) \cr \vsptwo t_e\ ; [ 10 ^ 4\ ; k ] & & 6.4 & 6.5 & 12.4 & & ( 6 ) \cr \noalign{\horule } \vsptwo t_{h^+}\ ; [ k ] & & 3.2\times 10 ^ 4 & 4.9\times 10 ^ 4 & 1.5\times 10 ^ 5 & & ( 7 ) \cr \vsptwo t_{he^{2+}}\ ; [ k ] & & 1.6\times 10 ^ 5 & \dots & \dots & & ( 4,\ , 7 ) \cr \noalign{\horule } \vsptwo \machson & & 4.21 & 4.17 & 2.15 & & ( 8)\cr \vsptwo \machalf & & 4.05 & 2.39 & 2.44 & & \cr \vsptwo \machtot & & 2.92 & 2.07 & 1.61 & & ( 9)\cr } \htrule } \ ] ] notes : ( 1 ) shock velocities ( relative to solar wind ) for 91097 and 91147 were taken from burton et al .
( 1992 ) , but that for 91118 was estimated from the value of inferred from the given by burton et al .
( 2 ) shock obliquities were taken from burton et al .
( 1992 ) and their uncertainties are given in balogh et al .
( 3 ) the thermal velocities are measured gaussian widths .
( 4 ) omissions in data occur where such information was not used in the present comparison .
( 5 ) mean upstream fields are taken from magnetometer measurements ( hoang et al .
( 6 ) electron temperatures for 91097 and 91147 are from ( hoang et al .
1995 ) , while that for 91118 was estimated using the large scale averages of phillips et al .
( 7 ) ion temperatures are from .
( 8) the values assume that , and have included the dynamical effects of .
( 9 ) here .
uncertainties are quoted for measured quantities .
see the text for further elaboration on these parameters . 0.1truecm | various theoretical techniques have been devised to determine distribution functions of particles accelerated by the first - order fermi mechanism at collisionless astrophysical shocks .
the most stringent test of these models as descriptors of the phenomenon of diffusive acceleration is a comparison of the theoretical predictions with observational data on particle populations .
such comparisons have yielded good agreement between observations at the quasi - parallel portion of the earth s bow shock and three theoretical approaches , namely monte carlo kinetic simulations , hybrid plasma simulations , and numerical solution of the diffusion - convection equation .
testing of the monte carlo method is extended in this paper to the realm of oblique interplanetary shocks : here observations of proton and distributions made by the swics ion mass spectrometer on ulysses at nearby interplanetary shocks ( less than about 3au distant from the sun ) are compared with test particle monte carlo simulation predictions of accelerated populations .
the plasma parameters used in the simulation are obtained from measurements of solar wind particles and the magnetic field upstream of individual shocks ; pick - up ions are omitted from the simulations , since they appear , for the most part , at greater heliospheric distances .
good agreement between downstream spectral measurements and the simulation predictions are obtained for two shocks by allowing the parameter , the ratio of the mean - free scattering length to the ionic gyroradius , to vary in an optimization of the fit to the data ; generally , corresponding to the case of strong scattering .
simultaneous and data , presented only for the april 7 , 1991 shock event , indicate that the acceleration process is roughly independent of the mass or charge of the species .
this naturally arises if all particles interact elastically with a massive background , as occurs in collisionless `` scattering '' off a background magnetic field , and is a patent property of the monte carlo technique since it assumes elastic and quasi - isotropic scattering of particles in the local plasma frame .
= cmr5 = cmbx5 = cmr7 = cmbx7 = cmr8 = cmmi8 = cmsy8 = cmbx8 = cmtt8 = cmti8 = cmsl8 = cmr6 = cmmi6 = cmsy6 = cmbx6 # 1@xmath0 # 1#2 @xmath1 |
the large majority of the almost 500 extrasolar planets detected so far orbit main sequence ( ms ) stars , and these planets were detected mostly using radial velocities ( rvs ) or transits .
for various reasons , not only technical , current exoplanet research is strongly oriented towards ms stars , with the main goal of finding rocky analogs of our earth , and towards understanding planetary system formation , while the late - stage evolution after the ms remains poorly understood .
the first attempts to model the evolution of the planetary systems after the red giant phase are very recent ( villaver & livio 2009 and ref . therein ) , and in particular lack observational constraints .
the two main methods to detect exoplanets , rvs and transits , are not efficient for compact stars with high gravities and small radii , such as the extreme horizontal branch stars and the white dwarfs . an alternative method for these stars is given by the timing method , in analogy with the first pulsars planet discoveries @xcite .
this method is based on the reflex motion of the parent star due to the companion , which changes periodically the star - observer distance , causing a delay ( or advance ) on the arrival time of the photons . in order to use the timing method ,
what we need is a precise astrophysical clock . for the pulsars planets , this is given by the pulsar s radio signal modulated by the fast and stable rotation of the neutron star .
but other potentially precise clocks exist in astrophysics and are given , for example , by stellar pulsation timing or the eclipse timing in close binary systems .
an early example of application of the timing method is given by papar et al .
( 1988 ) , who calculated the orbital elements of an unseen stellar companion to the delta scuti star sz lyncis using the o
c technique ( sterken 2005 ) on the pulsation timing .
the method has also been discussed for pulsating white dwarfs by kepler et al .
( 1991 ) and provencal ( 1997 ) .
the next sections contain an introduction to the field of post - rgb planetary systems and a short review of the recent discoveries , respectively .
then we investigate the potential of the timing method to find new planetary systems around evolved stars using pulsation or eclipse timing , with particular attention to the ongoing space missions corot and kepler .
finally , in the last section , a few conclusions are given and future perspectives are discussed .
during the red giant ( rg ) expansion of a star , two main opposite phenomena influence the orbits of potential planetary companions : mass loss from the star , which lowers the binding energy and hence tends to cause an outward drift of the planet , and tidal effects , which tend to attract the planet close to the star and eventually inside the envelope .
if the planet enters the stellar envelope , it is not clear in which conditions it can survive and what can be the influence on the subsequent evolution of the star . on the other hand , we know that during the rg phase a small fraction of stars , of the order of 1% , loose almost all their envelope and are left with a very thin h envelope during their core he burning phase .
these are the so - called subdwarf b ( or sdb ) stars , that are located on the extreme horizontal branch ( ehb ) , at high effective temperatures ( 20,000k ) .
at least 50% of them are in binary systems and in this case the formation of an sdb star can be explained in terms of binary evolution ( han et al .
2002 , 2003 ) .
however , for the remaining fraction of apparently single sdb stars , it is more difficult to explain why the huge mass loss during the rgb occurred ( see review by heber 2009 ) . in this context planets
might play a role @xcite .
similar phenomena of planetary migration , planet engulfment and planet accretion from stellar ejected material can happen also during the second large expansion of a star , when it enters the asymptotic giant branch and the thermal pulses phase , and during the planetary nebula ( pn ) ejection . in order to achieve consistent modelling of all these complex phenomena , tests against observational constraints are essential . to detect planets around white dwarfs
is ( or more exactly will be ) a key step to study the late - stage evolution of planetary systems . equally important is to identify planetary systems in previous shorter and more rare evolutionary phases like the rgb or the horizontal branch . only in this way
can we try to disentangle the various phenomena that changed the configuration of the planets and their orbits , with the final goal of being able to trace back the entire history of a planetary system from the final white dwarf configuration to the original protoplanetary disk .
what we need first is to find a sufficient number of planetary systems orbiting post - rgb stars and white dwarfs , covering a wide range of orbital parameters and planetary masses .
concerning white dwarfs , what we know already is that some of these stars are surrounded by metal - rich debris disks : at least 17 white dwarfs show circumstellar dust at @xmath0 1,000 k ( reach et al .
2005 and ref
. therein , farihi et al . 2010 ) and about 20% of them show also caii emission from a hotter gaseous disk ( gnsicke et al . 2006 , 2010 ) .
these disks are generally very close to the white dwarf , within about 1 r@xmath1 .
recently , also much more distant and cooler debris disks have been detected ( su et al .
2007 , chu et al . 2010 and
bilkov et al . 2010 ) , which are probably the final evolution of the solar system s kuiper belt analogues . in other words
, we do nt know yet how many white dwarfs host planets and at which distances from the star , but we know that at least the material to build planets is there .
another interesting possibility is that second ( or even third ) generation planets can form from the stellar material ejected during stellar evolution ( perets et al . 2010 ) . but
again , to confirm these theoretical scenarios , we need observations and statistics .
from the observational side , the late - stage evolution of the planetary systems is a very recent field and only in the last three years a few first post - rgb planetary systems were detected .
they belong mainly to two classes : 1 .
extreme horizontal branch hot subdwarfs b : + v391 peg b @xcite , + hw vir b and c @xcite , + hs0705 + 67003 b @xcite , + hd149382 b ( geier et al . 2009 , but see also jacobs et al .
cataclysmic variables : + nn ser b and c ( qian et al .
2009b , beuermann et al
. 2010a , hessman et al . 2010 ) , + dp leo b ( qian et al . 2010a , beuermann et al .
2010b ) and + qs vir b ( qian et al .
2010b ) .
we note that the issue of nomenclature for circumbinary planets , in contrast to the nomenclature for planets around components of visually resolved binaries , is unsolved . on recommendation by the _ a&a _ editor , beuermann et al .
( 2010a , 2010b ) follow the rule by which any unseen companions , whether stellar or sub - stellar , are designated by lowercase letters .
they hence have introduced ( ab ) for the two components of the central binary and start numbering the third bodies with the letter c , resulting in the planet designations nn ser ( ab ) c , nn ser ( ab ) d , and dp leo ( ab ) c. another interesting new planet candidate was presented during this meeting by setiawan et al .
2010a ( a 1.3 m@xmath2 planet at 0.116 au from a red hb star of likely extragalactic origin ) .
in table 1 we summarize the list of known post - rgb planets , including the pulsars planets and a few brown dwarfs ( bds ) near the planet mass limit .
we include also the brown dwarf wd0137@xmath3349b because its very small orbital distance , only 0.375 r@xmath1 , implies that the bd was engulfed by the red giant progenitor of the white dwarf ( maxted et al .
2006 ) . as one can see ,
most of the detections reported in table 1 were obtained through timing measurements .
.post - rgb planets [ cols="<,<,>,^,<,>,^,<,>,^ , < , < , < , < , < " , ] a circumbinary planet with a minimum mass of about 5 m@xmath4 could be the reason of the periodic component in the o c plot of the eclipsing polar hu aquarii ( schwarz et al .
2009 , not included in table 1 ) .
otoole et al .
( 2010 ) also report early results from their on - going rv survey .
we refer to their proceeding papers for more details . for what concerns the white dwarfs ,
until now there are no clear detections of planets around single white dwarfs , with only two candidates : gd66 ( mullally et al .
2008 , 2009 ) and gd356 ( wickramasinghe et al .
2010 ) . we know at least three planet hosting ms stars having a wide binary wd companion : hd13445=gl86 , hd147513 and hd27442 ( desidera & barbieri 2007 and ref . therein ) .
in all these three systems the planet orbit the ms star at an orbital distance much smaller than the wd distance : 0.11 versus 19 au for hd13445 , 1.26 versus at least 100 au for hd147513 , and 1.18 versus 240 au for hd27442 .
therefore it is unlikely that the wd had an influence on the evolution of the planetary system and this is why these systems are not included in table 1 .
more interesting ( although quite exotic ) is the case of the circumbinary planet orbiting the pulsar+wd system psr 1620@xmath326 ( thorsett et al . 1999 ,
sigurdsson et al . 2003 ) in the globular cluster m4 . following sigurdsson et al . (
2003 ) , the most likely scenario is that the planet was initially in orbit around the ms star progenitor of what is now the wd companion .
the system underwent an exchange interaction with a neutron star binary , with the ms star replacing the original high - mass wd companion to the neutron star .
the end results were a new eccentric orbit for the ms star , and the displacement of the planet onto a wide circumbinary orbit .
similarly to the astrometric method ( and contrary to the rv and transit methods ) , the timing method is sensitive to large orbital separations . equation ( 1 ) shows that the ( maximum ) change in the arrival time of the photons ( or time shift ) @xmath5 is indeed proportional to the orbital distance @xmath6 ( @xmath7 , @xmath8 , @xmath9 and @xmath10 are , respectively , planet mass , stellar mass , light speed and inclination of the orbit to the line of sight ) :
@xmath11 for @xmath6=1au , @xmath7=1m@xmath2 , @xmath8=1m@xmath1(and @xmath10=90@xmath12 ) , the time shift is about 0.5 seconds ; while it is 1s for a stellar mass of 0.5m@xmath1 , as in the case of a typical sdb star or a slightly lower - than - average - mass white dwarf . as shown by the recent discoveries in the previous section , it is possible to detect planets with the timing method using ground - based data .
what is needed is a long - term photometric monitoring of the targets for many years with frequent runs , ideally each month or two .
a few ground - based programs to search for post - rgb planets exist , e.g.the exotime program ( schuh et al . 2009 , schuh 2010 , lutz et al .
2010 ) which is monitoring five sdb pulsators , a similar program at the mcdonald observatory on 15 zz ceti pulsators @xcite , and the long - term monitoring of one dbv pulsator by dalessio et al .
. a secondary goal of these programs is the measurement of the secular variation of the pulsation periods , which is related to the change of the stellar structure because of its evolution ( e.g. kepler et al .
2005 , costa & kepler 2008 , silvotti et al .
similar ground - based programs exist also for the eclipsing binaries .
however it is clear that , potentially , the maximum results can be achieved using continuous monitoring from space craft .
the exceptional photometric precision and duty cycle of the corot and kepler satellite data represent a unique opportunity to use the timing method to detect low mass substellar companions to pulsating stars and eclipsing binaries . in particular
, kepler has a longer coverage , up to 3.5 years , as the satellite will observe the same @xmath0105 square degree field for the whole mission duration ; while the corot `` long runs '' have a maximum duration of about 6 months . for more details concerning the kepler and corot missions we refer the reader to borucki et al .
( 2010 ) , gilliland et al .
( 2010 ) and auvergne et al .
the main limitation of kepler is the relatively long sampling time of 58 s ( short - cadence data ) .
although the photometer samples the field every 6.54 s , telemetry restrictions do not permit to save and transmit to the ground the original data .
this is a severe limitation only for the very rapid pulsators like the short - period sdb stars of the sdbv@xmath13 class and the white dwarfs , which can have pulsation periods as short as @xmath1412 minutes .
however , in the survey phase ( 1st year of observation ) that has just ended , no pulsating white dwarfs were found and only one short - period sdb pulsator was observed ( kawaler et al .
2010 ) plus an hybrid pulsator with a very rich amplitude spectrum in a eclipsing binary system with a m dwarf ( stensen et al .
2010a , see also end of this section on _ 5
. eclipse timing _ ) .
for all the other 13 long - period sdb pulsators ( sdbv@xmath15 ) discovered by kepler ( stensen et al .
2010b , 2010c ) , with typical periods between 30 min to few hours , the 58 s sampling is largely sufficient , even though longer pulsation periods mean higher minimum detectable mass of a substellar companion .
this is shown by the following equation ( 2 ) , in which @xmath16 is the timing error , @xmath17 is the number of data points , @xmath18 the pulsation period , @xmath19 the photometric ( relative intensity ) error , @xmath20 the pulsation amplitude and @xmath21 the amplitude error , which decreases with the square root of the length of the run and can be estimated directly by measuring the mean noise of the fourier amplitude spectrum .
equation ( 2 ) is valid under the assumption of uncorrelated errors and no cross terms between the different pulsation frequencies , and is obtained multiplying by @xmath18 equation ( a15 ) from breger et al .
1999 ( see also montgomery & odonoghue 1999 and silvotti et al .
2006 for more details ) .
@xmath22 a similar equation can be obtained also for the eclipse timing ( doyle & deeg 2002 ) , assuming a simple model of triangular eclipse ( which implies that the error in eclipse minimum timing @xmath16 is a lower limit to the real error ) . in the following equation ( 3 ) , @xmath23 is the duration of the eclipse from first to last contact , @xmath24 is the relative intensity of the eclipse ( or ( 1-@xmath24 ) is the eclipse s relative depth , with @xmath25=1 out of eclipse ) and @xmath26 is the number of data points taken during @xmath23 ; @xmath19 was defined above .
@xmath27 as we do not expect to find close - in planets with short orbital periods around post - rgb stars , we will need many months / years of kepler data to be able to set limits on the presence of planets around these stars . at the moment , with the available kepler data , we can only start making some tests to see what the typical phase errors are that we can obtain and therefore what we can expect in terms of minimum detectable masses and orbital periods .
the first goal of these preliminary tests is to verify the phase stability ( or `` phase coherence '' ) of different classes of pulsating stars and eclipsing binaries . regarding the pulsators , the most promising targets are those with stable pulsations ( in terms of period , amplitude and phase ) , short periods ( to reduce the minimum detectable mass ) , several independent periods and relatively large amplitudes . to have several independent periods
is important as it permits to check whether the same periodic pattern produced by a planet is present in all the periods ( see dalessio et al .
2010 for an example where this is not the case ! ) , helping to exclude false detections due to other periodic effects ( e.g. beating of close frequencies ) .
sdb pulsators fit well these constraints and , due to the relatively short periods , are the best candidates .
other potential targets are the delta scuti stars , @xmath28 cephei and rr lyrae . in the next pages
we show the results of a few preliminary tests on these pulsating stars using kepler and corot data , including also an eclipsing binary with a pulsating sdb component .
the kepler data were retrieved from the kasc data base ( kasc is the kepler asteroseismic consortium , gilliland et al .
the corot public data were downloaded from the corot archive at laeff ( laboratory for space astrophysics and theoretical physics ) in spain . *
subdwarf b stars * before the kepler observations , the oscillating sdb stars could be divided into two groups : the short - period ( p - mode ) pulsators ( or sdbv@xmath13=rapid pulsators ) of the v361 hya class from the prototype ( kilkenny et al .
1997 ) , with effective temperatures between about 27,000 and 35,000 k , and the longer period ( g - mode ) pulsators ( or sdbv@xmath15=slow pulsators ) of the v1093 her class ( green et al .
2003 ) , with effective temperatures between @xmath021,000 and @xmath028,000 k. four hybrid pulsators ( sdbv@xmath29 ) were known , showing both p- and g - modes at the same time ( schuh et al .
2006 , see also baran & fox - machado 2010 for the complete list of hybrid pulsators before kepler ) .
this classification ( compare also kilkenny et al .
2010 ) may need some adjustment because the kepler observations have shown that the temperature range of the two instability strips and the fraction of hybrid pulsators could be larger than previously believed . in order to have a first feeling of what can be obtained with the timing technique on kepler sdb data , we consider the star kic010139564 , the only rapid sdb pulsator observed by kepler @xcite , and one of the best candidates to search for planets .
figure 1 shows the o
c plot of six pulsation frequencies : these frequencies are considered also in figs.4 and 5 of kawaler et al . (
2010 ) . at 1 au from the star .
using longer subsets , this limit can be reduced to about 2 m@xmath2 and even further ( see text for more details ) . ] first of all , we see from fig.1 that the frequencies have in general stable phases .
this is true not only for these six frequencies but also for almost all the other frequencies of this star , although they are not represented here . by comparing fig.1 with fig.4 of kawaler et al .
( 2010 ) , we see that , for most of these frequencies , the error bars are smaller by a factor of ( 7.5/2)@xmath301.9 , which is exactly what we expect considering that each point in fig.1 represents @xmath07.5 days of data while in kawaler s plot each point corresponds to 2 days .
however , at least one frequency at 5760.2 @xmath31hz behaves differently : the error bar of fig.1 is larger than the one obtained by kawaler et al .
the reason is in their fig.5 , where we see that both phase and amplitude are changing periodically with a phase difference of about 90 degrees , suggesting that beating between closely spaced stable oscillation frequencies is occuring . from the smaller error bars of fig.1 we can estimate the minimum mass of a detectable planet
: we obtain about 7 m@xmath2 for an orbital distance of 1 au .
however , potentially , the minimum mass can be reduced to about 2 m@xmath2 using longer subsets of about 3 months each .
this limit could be further extended to about 0.7 m@xmath2 extrapolating from the small error bars in fig.4 of kawaler et al .
( 2010 ) , provided that the new kepler data that will be collected in the next months / years will actually allow to separate the close frequencies and resolve the amplitude spectrum of this star .
for the long - period sdb pulsators , the first observations have shown that in general the phases are stable allowing to use the timing method also for these stars . for them ,
the minimum detectable mass of a substellar companion should be @xmath141050 times higher than the numbers reported above considering that the periods are at least one order of magnitude longer .
however , some of these stars show also a few peaks at high frequency , that are probably due to a few excited p - modes ( reed et al .
2010 , charpinet et al .
if these short - period p - modes will be confirmed by the future kepler observations , they could bring down again the minimum detectable masses of the potential companions .
delta scuti stars * another interesting class of pulsators are the delta scuti stars . although the periods are relatively long and comparable with those of the long - period sdb pulsators , their amplitudes can be larger , allowing to reduce the timing errors to a few seconds ( see equation 2 ) .
an example of a delta scuti star with a relatively simple amplitude spectrum ( not too many frequencies ) and short periods is given in fig.2 , where we see that the phases are stable , at least during the 1-month kepler run in short cadence .
4 and @xmath3210 s respectively , suggest that delta scuti stars can be stable enough to use the timing technique .
the fractional amplitudes of the two frequencies are 0.25% and 0.17% respectively . ] * 3 .
@xmath28 cephei stars * the timing technique can also be used for @xmath28 cephei stars . here
we show an example on the @xmath28 cephei star hd 180642 ( v1449 aquilae ) , illustrating that these stars also may have stable phases .
hd 180642 was observed by the corot satellite during 156 days of the long run lrc01 , with a sampling cadence of 32s .
the corot data were analyzed in detail by degroote et al .
( 2009 ) , who found a complex amplitude spectrum with several combination frequencies .
we applied a least - square sinusoidal fit to the corot data using 15 frequencies obtained from pre - whitening , including a few combinations . in figure 3 ( upper panel ) a preliminary o
c plot shows an apparent phase variation of the order of 20s in 2 months , whose origin and nature is not clear .
if this effect were due to a secular variation of the pulsation period , we would obtain @xmath33 @xmath14 @xmath3410@xmath35 s / s .
this number is significantly larger than the value found by degroote et al .
( 2009 ) from direct measurement : about 1.3@xmath3610@xmath35 s / s .
this difference deserves further study .
cephei star hd 180642 ( v1449 aquilae ) .
each point represents about 17.4 days of the corot long run lrc01 ( 156 days ) .
the two panels suggest that phase and amplitude variations are not correlated and therefore these variations can not be attributed to beating of close frequencies .
see degroote et al .
( 2009 ) for more details on the pulsational behavior of this star . ] * 4 .
rr lyrae stars * from the kasc data base we selected two rr lyrae stars with the longest observational coverage ( about 6 months ) .
c diagram of these rr lyrae stars , shown in fig.4 , indicates that the phase errors are generally large and , even more important , they are highly inhomogeneous .
the reason of this behaviour is not clear .
although before the kepler observation , both stars were considered as non - blazhko stars , the non - uniform error distribution might be due to a low - level blazhko modulation of phases and/or frequencies .
noise due to convection might also play a role .
although a deeper testing on a larger number of rr lyrae stars is needed before drawing conclusions , this first example suggests that rr lyrae might not be stable enough for the timing technique . *
5 . eclipse timing * in this section we show an example of eclipse timing .
we note that this method is related to the search for transit timing variations .
the target , 2m1938 + 4603 ( or kic09472174 ) , is of particular interest because it is a sdb+dm eclipsing binary system and the sdb component shows a very rich pulsational spectrum with periods ranging from @xmath04 min to 5.5 hours .
the orbital period of the system is 0.12576530 d. for more details on 2m1938 + 4603 we refer the reader to the recent article by stensen et al .
( 2010a ) .
the eclipsing binary systems with a pulsating component , like this one , have the advantage of having two independent clocks given by the eclipse timing and by the pulsations ( provided that there is no coupling between them ) .
for this reason they are an ideal test case to verify the potential of the timing method in finding low - mass substellar objects .
detection limit would be reduced to about 4 m@xmath2 considering an orbital distance of 1 au from the binary .
potentially , the presence of a planet could be independently confirmed also from the pulsations of the sdb star . ]
in the last three years , the first discoveries of planets orbiting evolved stars after the red giant branch have proven that the timing method is an efficient way to detect substellar objects around compact stars , using either the stellar pulsation or the eclipse timing as a clock .
this method , which is sensitive to jupiter - mass planets in large orbits of @xmath141 au or more , requires long - time photometric monitoring with frequent , or , better still , continuous , runs . in order to validate the method
, it would be important to confirm these discoveries with other independent techniques .
this in general is not easy because the targets are generally faint and many have spectral lines broadened by high gravity .
the high quality data from the corot and kepler satellites represent a unique opportunity due to the exceptional duty cycle .
kepler has the additional advantage of longer runs , up to 3.5 years .
preliminary tests show that sdb stars are the best candidates , in particular the short - period pulsators .
other good candidates are the delta scuti stars , at least those with a relatively simple amplitude spectrum ( not too many frequencies ) .
an example on a @xmath28 cephei star shows that also these stars may have sufficiently stable phases , although the higher stellar mass implies a higher minimum mass of the potential substellar companions .
rr lyrae stars are more problematic and our preliminary tests suggest that phases and/or frequencies may have significant variations on time scales of weeks .
an example of an eclipsing binary is given and shows that also for these targets jovian planets can be detected .
moreover , if the binary system has a pulsating component like 2m1938 + 4603 , the pulsations can be used , potentially , as an independent clock to confirm the detections .
k. beuermann , f. v. hessman , s. dreizler et al . , _
astronomy & astrophysics _ , * 521 * , l60l64 ( 2010a ) .
k. beuermann , j. buhlmann , j. diese , et al . , _
astronomy & astrophysics _ , in press ( arxiv:1011.3905 ) ( 2010b ) .
j. bilkov , y .- h .
chu , k. y .- l .
su , r. a. gruendl , th .
rauch , spectral analysis of mid - ir excesses of wds , in _ planetary systems beyond the main sequence _ , edited by s. schuh , h. drechsel and u. heber , aip conference proceedings , american institute of physics , new york ( these proceedings ) .
y. h. chu , r. a. gruendl , j. bilkov , a. riddle , k. y .- l . su , spitzer surveys of ir excesses of white dwarfs , in _ planetary systems beyond the main sequence _ , edited by s. schuh , h. drechsel and u. heber , aip conference proceedings , american institute of physics , new york ( these proceedings ) .
j. dalessio , j. l. provencal , h. s. shipman , searching for planets with white dwarf pulsations : spurious detections in _ planetary systems beyond the main sequence _ , edited by s. schuh , h. drechsel and u. heber , aip conference proceedings , american institute of physics , new york ( these proceedings ) .
j. farihi , evidence for terrestrial planetary system remnants at white dwarfs , in _ planetary systems beyond the main sequence _ , edited by s. schuh , h. drechsel and u. heber , aip conference proceedings , american institute of physics , new york ( these proceedings ) .
b. t. gnsicke , gaseous debris disks around wds , in _ planetary systems beyond the main sequence _ , edited by s. schuh , h. drechsel and u. heber , aip conference proceedings , american institute of physics , new york ( these proceedings ) .
f. v. hessman , k. beuermann , s. dreizler , t. r. marsh , s. g. parsons , c. m. copperwheat , d. e. winget , g. f. miller , j. j. hermes , m. r. schreiber , w. kley , v. s. dhillon , s. p. littlefair , the planets around the post - common envelope binary nn serpentis in _ planetary systems beyond the main sequence _ , edited by s. schuh , h. drechsel and u. heber , aip conference proceedings , american institute of physics , new york ( these proceedings ) .
v. a. jacobs , r. h. stensen , h. van winckel , s. bloemen , p. i. ppics , g. raskin , j. debosscher , s. uttenthaler , e. van aarle , c. waelkens , e. bauwens , t. verhoelst , c. gielen , h. lehmann , r. oreiro , hermes high - resolution spectroscopy of hd 149382 : where did the planet go ? , in _ planetary systems beyond the main sequence _ , edited by s. schuh , h. drechsel and u. heber , aip conference proceedings , american institute of physics , new york ( these proceedings ) .
r. lutz , s. schuh , r. silvotti , the exotime targets hs0702 + 6043 and hs0444 + 0458 , in _ planetary systems beyond the main sequence _ , edited by s. schuh , h. drechsel and u. heber , aip conference proceedings , american institute of physics , new york ( these proceedings ) .
s. j. otoole , u. heber , s. geier , l. classen , o. de marco , radial velocity search for substellar companions to sdb stars in _ planetary systems beyond the main sequence _ , edited by s. schuh , h. drechsel and u. heber , aip conference proceedings , american institute of physics , new york ( these proceedings ) .
h. perets , planets in evolved binary systems
, in _ planetary systems beyond the main sequence _ , edited by s. schuh , h. drechsel and u. heber , aip conference proceedings , american institute of physics , new york ( these proceedings ) .
j. l. provencal ,
white dwarfs and planetary systems , in _ planets beyond the solar system and the next generation of space missions _ , edited by d. soderblom , asp conference series 119 , pp .
123126 ( 1997 ) .
j. setiawan , r. klement , th .
henning , h .- w .
rix , b. rochau , t. schulze - hartung , j. rodmann , a planetary companion around a metal - poor star with extragalactic origin
, in _ planetary systems beyond the main sequence _ , edited by s. schuh , h. drechsel and u. heber , aip conference proceedings , american institute of physics , new york ( these proceedings ) . | the timing method , using either stellar pulsations or eclipse timing of close binaries as a clock , is proving to be an efficient way to detect planets around stars that have evolved beyond the red giant branch . in this article
we present a short review of the recent discoveries and we investigate the potential of the timing method using data both from ground - based facilities as well as from the kepler and corot space missions . address = inaf - osservatorio astronomico di torino , strada dellosservatorio 20 , + 10025 pino torinese , italy address = konkoly observatory of the hungarian academy of sciences , konkoly thege mikls t 15 - 17 , 1121 budapest , hungary address = instituut voor sterrenkunde , k.u .
leuven , celestijnenlaan 200d , 3001 leuven , belgium address = instituut voor sterrenkunde , k.u .
leuven , celestijnenlaan 200d , 3001 leuven , belgium address = georg - august - universitt gttingen , institut fr astrophysik , friedrich - hund - platz 1 , 37077 gttingen , germany |
the first section of the paper reviews the concept of an euler characteristic for a first order structure in model theory .
the discussion is purely algebraic for the benefit of readers unfamiliar with model theory , and various examples are given .
once an euler characteristic is interpreted as an integral , it is natural to ask whether fubini s theorem holds ; that is , whether the order of integration can be interchanged in a repeated integral . in the second section we consider finite morphisms between smooth curves over any algebraically closed field , and
show that fubini s theorem is almost equivalent to the riemann - hurwitz formula .
more precisely , in characteristic zero the two are equivalent and so fubini s theorem is satisfied , whereas in finite characteristic the possible presence of wild ramification implies that , for any euler characteristic , interchanging the order of integration is not always permitted . the third section discusses a notion weaker than the full fubini property ( a so - called strong euler characteristic @xcite @xcite ) , but which is sufficent for our applications . we show that over an algebraically closed field of characteristic zero , there is exactly one strong euler characteristic ( over the complex numbers , this is the usual topological euler characteristic ) .
we return to finite morphisms between algebraic varieties , this time considering surfaces . again
, fubini s theorem is related to a riemann - hurwitz formula , originally due to iversen @xcite .
our methods provide a new proof of his result .
the paper finishes with a discussion of the geometric approach to ramification theory of local fields .
the conference motivic integration and its interactions with model theory and non - archimedean geometry at the icms , edinburgh , during may 2008 encouraged me to think about these ideas .
part of this text was written while visiting the ihes , and i am grateful for the excellent working environment which this provided .
this visit would not have been possible without the generosity of the cecil king foundation and the london mathematical society , in the form of the cecil king travel scholarship .
i thank my supervisor i. fesenko for his constant encouragement .
first we present some elementary objects from model theory from a perspective suitable for this work .
we must understand what sort of sets we can measure and what it means to measure them .
this material is well - known but hopefully this explicit exposition will appeal to those unfamiliar with the theory .
given a set @xmath0 , a _ ring _ of subsets of @xmath0 is defined to be a non - empty collection of subsets @xmath1 of @xmath0 such that @xmath2 it is enough to assume that @xmath1 is closed under differences and unions for this implies it is closed under intersections .
a ring of sets is said to be an _ algebra _ if and only if it contains @xmath0 .
following van den dries @xcite we define a _ structure _ @xmath3 on @xmath0 to be an algebra @xmath4 of sets on @xmath5 for each @xmath6 such that 1 . if @xmath7 then @xmath8 ; 2 .
@xmath9 ; 3
. if @xmath10 is the projection map to the first @xmath11 coordinates , then @xmath12 implies @xmath13 .
given a structure , one refers to the sets in @xmath4 as being the _ definable _ subsets of @xmath14 . if @xmath15 and @xmath16 then @xmath17 is said to be definable if and only if its graph belongs to @xmath18 .
let @xmath19 be a structure on a set @xmath0
. then 1 .
if @xmath20 then @xmath21 ; 2 . if @xmath22 , then @xmath23 is in @xmath4 ; 3 .
if @xmath24 is a permutation of @xmath25 , then the function @xmath26 given by permuting the indices of the coordinates by @xmath24 is definable .
moreover , if @xmath15 and @xmath27 is definable , then 1 .
@xmath28 is definable ; 2 . if @xmath29 is definable , then @xmath30 is definable , and the function given by restricting @xmath17 to @xmath31 is definable ; 3 . if @xmath32 , then @xmath33 ; 4 .
if @xmath17 is injective , then its inverse is definable ; 5 .
if @xmath34 and @xmath35 is definable , then @xmath36 is definable .
these are straightforward to check ; proofs may be found in @xcite .
if @xmath37 is a first order language of logic , and @xmath0 is an @xmath38-structure , then there is a structure on @xmath0 in which @xmath4 consists of precisely those sets of the form @xmath39 where @xmath40 is formula of @xmath38 in variables @xmath41 and @xmath42 ; that is , those sets which are definable with parameters in the sense of model theory .
realistically , any structure in which we will be interested will arise in this way as the parameter - definable sets of some language .
but for the reader less familiar with logic , the axiomatic approach above is more immediately appealing , though ultimately less satisfying .
[ example_structures ] we present some examples to explain what we can and can not study using structures .
all are well - known . 1 .
if @xmath0 is an arbitrary set , we may take @xmath4 to be collection of all subsets of @xmath14 ; that is , every set is definable . 2 .
if @xmath43 is an algebraically closed field , let @xmath44 be the ring of sets generated by the zariski closed subsets of @xmath45 ; such sets are called constructible .
it is known that @xmath46 forms a structure on @xmath43 .
the difficulty is establishing that such sets are closed under projection ; this may either be proved in a model theoretic setting , where it is equivalent to establishing that the theory of algebraically closed fields admits quantifier elimination , or it may be seen as a special case of a result of algebraic geometry concerning constructible subsets of noetherian schemes ( see e.g. @xcite exercises 3.17 - 3.19 ) .
3 . if @xmath43 is an arbitrary field , recall that an affine subset of @xmath45 is a set of the form @xmath47 where @xmath48 and @xmath49 is a @xmath43-subspace of @xmath45 . letting @xmath44 be the ring of sets generated by affine subsets of @xmath45 gives a structure on @xmath43 .
if @xmath50 is the real line , then let @xmath51 be the ring of sets generated by @xmath52 for @xmath53 $ ] ; the sets in @xmath51 are called semi - algebraic subsets of @xmath54 .
this gives a structure for @xmath50 .
again , the difficulty is verifying that such sets are closed under projection .
5 . none of the following give structures on the real line : the borel sets , the lebesgue measurable sets , the souslin sets .
so structures are typically quite coarse from the point of view of classical analysis and measure theory .
having introduced the sets of interest we now discuss what it means to take the measure of such a set .
let @xmath0 be a set with a structure @xmath19 .
euler characteristic _ is a map @xmath55 from the definable sets to some commutative ring @xmath56 , i.e. @xmath57 which satisfies 1 .
if @xmath58 are disjoint , then @xmath59 ; 2 . if @xmath7 , @xmath32 , then @xmath60 ; 3 .
if @xmath7 , @xmath32 and there is a definable bijection @xmath61 , then @xmath62 . from the additivity of @xmath55 , one might think that an euler characteristic is similar to a measure in the classical sense .
the vast difference between the two is the invariance of @xmath55 under definable bijections .
for example , if @xmath0 is a field @xmath43 , @xmath63 , and multiplication by @xmath64 is a definable map from @xmath43 to itself , then @xmath65 for definable @xmath66 ; in other words , scaling a set does not affect its size . or if @xmath0 is the real line and @xmath67 is definable , then for any definable @xmath28 of the positive reals , @xmath68 .
some authors prefer the term _ generalised euler characteristic _ or _ additive invariant _ , to avoid possible confusion with the _ topological euler characteristic _
@xmath69 for complex projective manifolds , defined as the alternating sum of the betti numbers .
the easiest example of an euler characteristic is counting measure : let @xmath0 be a finite set , @xmath4 the algebra of all subsets of @xmath14 , and set @xmath70 to define a @xmath71-valued euler characteristic . explicitly exhibiting more interesting euler characteristics requires some work , so we present here without proof some known examples using the structures of example [ example_structures ] . 1 .
let @xmath43 be a field , equipped with the structure generated by the affine subsets .
if @xmath43 is infinite then there is a unique @xmath72$]-valued euler characteristic @xmath55 which satisfies @xmath73 where @xmath48 and @xmath49 is a @xmath43-subspace of @xmath45 .
2 . give @xmath50 the structure of semi - algebraic sets .
then there is a unique @xmath71-valued euler characteristic @xmath55 which satisfies @xmath74 3 . give @xmath75 the structure of constructible sets
; then there is a unique euler characteristic @xmath69 which agrees with the topological euler characteristic for any projective manifold .
let @xmath0 be a set with structure @xmath19 . the associated _ grothendieck ring _
, denoted @xmath76 ( though it does of course depend on the structure , not just the set @xmath0 ) , is defined to be the free commutative unital ring generated by symbols @xmath77 $ ] for @xmath28 a definable subset of @xmath14 , any @xmath78 , modulo the following relations 1 .
if @xmath58 are disjoint , then @xmath79=[a]+[b]$ ] ; 2 . if @xmath7 , @xmath32 , then @xmath80=[a][b]$ ] ; 3 . if @xmath7 , @xmath32 and there is a definable bijection @xmath61 , then @xmath77=[b]$ ] .
the map @xmath81 $ ] defines a @xmath76-valued euler characteristic on @xmath0 , which is universal in the sense that if @xmath82 is an euler characteristic , then there is a unique ring homomorphism @xmath83 such that @xmath84)$ ] for any definable @xmath28 . thus @xmath85 $ ] is the most general euler characteristic of a structure .
note that if @xmath86 is a single point , and @xmath87 is definable , then projection induces a definable isomorphism @xmath88 .
so @xmath89[a]=[a]$ ] for all definable @xmath28 and therefore @xmath89$]=1 ; more generally , @xmath90={\vertb\vert}$ ] for any finite definable set @xmath31 .
[ remark_euler_characteristic_of_varieties ] _ extending the euler characteristic to varieties . _
assume that @xmath91 is an algebraically - closed field with the structure @xmath19 of constructible subsets .
let @xmath92 be a separated algebraic variety over @xmath43 - our varieties shall usually consist only of the closed points of the corresponding scheme - and let @xmath93 be the ring generated by the zariski closed subsets of @xmath92 , i.e. the algebra of constructible subsets of @xmath92 .
it is straightforward to prove that @xmath55 uniquely extends to @xmath93 in such a way that if @xmath94 is an affine open or closed subset , @xmath95 is constructible , and @xmath96 is an open or closed embedding for some @xmath97 , then @xmath98 .
[ remark_measure_to_integral ] _ extending the measure to an integral . _
if @xmath0 is a set equipped with a structure and euler characteristic @xmath55 , then there is a unique @xmath56-linear map @xmath99from the space of functions spanned by characteristic functions of definable sets to @xmath56 which satisfies @xmath100 for any definable @xmath28 .
we will allow ourselves to use typical notation for integrals , writing @xmath101 .
here we relate fubini s theorem for euler characteristics to the riemann - hurwitz formula for morphisms between curves ; then we produce a startling result implying that in finite characteristic it is always possible for fubini s theorem to fail . throughout this section @xmath43
is an algebraically closed field of arbitrary characteristic , @xmath19 is the structure of constructible sets , and @xmath55 is a fixed @xmath56-valued euler characteristic on @xmath19 . by a curve @xmath102 over @xmath43 , in this section , we mean a smooth , one - dimensional , irreducible algebraic variety over @xmath43 ; we only consider the closed points of @xmath102 .
following remarks [ remark_euler_characteristic_of_varieties ] and [ remark_measure_to_integral ] the space of integrable functions on @xmath102 is the @xmath56-module generate by characteristic functions of constructible sets ; the integral on this space will be denoted @xmath103 .
let @xmath104 be a non - constant morphism of curves .
we will study whether fubini s theorem holds for the morphism @xmath105 , which is to say that for each @xmath106 , the fibre @xmath107 is constructible , that @xmath108 is integrable , and finally that @xmath109 .
the problem immediately simplifies : [ lemma_fubini_formula_for_curves ] fubini s theorem holds for a separable morphism @xmath104 of projective curves if and only if the following formula relating the euler characteristics of @xmath110 and @xmath111 is satisfied : @xmath112 where @xmath113 is the ramification degree of @xmath105 at @xmath114 .
let @xmath115 be the finite set of points at which @xmath105 is ramified .
let @xmath116 be a point of @xmath111 .
the fibre @xmath107 is finite ; moreover , it contains exactly @xmath117 points when @xmath118 .
so each fibre is certainly constructible and @xmath119 .
thus @xmath108 is constant off the finite set @xmath120 and hence is integrable on @xmath111 ; integrating obtains @xmath121 the fundamental ramification equality @xmath122 transforms this into @xmath123 which completes the proof
. more generally , if @xmath124 and @xmath104 is a morphism of projective curves which is not necessarily separable , then we decompose @xmath105 as @xmath125 ; here @xmath126 is the frobenius morphism of @xmath110 , @xmath127 is a separable morphism , and @xmath128 is a non - negative integer .
the previous proof shows that fubini holds for @xmath105 if and only if @xmath129 so fubini holds for @xmath105 if and only if it holds for the separable part @xmath130 ; in particular , fubini holds for any purely inseparable morphism of projective curves for this reason we are justified in focusing our attention on separable morphisms . more usually fubini s theorem
is concerned with measuring subsets of product space via repeated integrals ; let us show that this is the same as our current activity considering fibres of morphisms between projective curves .
suppose @xmath104 is a separable morphism of projective curves over @xmath43 .
then @xmath105 is a finite morphism , so that if @xmath131 is a non - empty , affine , open subset then the same is true of @xmath132 .
choose closed embeddings @xmath133 , @xmath134 and let @xmath135 be the graph of @xmath136 .
it is immediate that the integral @xmath137 is well - defined and equal to @xmath138 .
conversely , if we fix @xmath139 then @xmath140 ; arguing as in the previous lemma now obtains @xmath141 so interchanging the order of integration preserves the value of the integral if and only if @xmath142 further , @xmath143 and @xmath144 are finite sets and it is straightforward to verify , similarly to the previous lemma , that @xmath145 taking the sum of the previous two formulae shows that fubini s theorem holds for @xmath104 if and only if the the repeated integrals of @xmath146 are equal .
recall that the riemann - hurwitz formula states that if @xmath104 is a non - constant morphism of projective curves , then there are integers @xmath147 for each @xmath148 ( which we shall call the _ riemann - hurwitz ramification degrees _ ) such that @xmath149 , with equality if and only if @xmath105 is tamely ramified at @xmath114 , and such that @xmath150 where @xmath151 is the genus of @xmath152 .
it is apparent that fubini s theorem and the riemann - hurwtiz formula are related .
[ remark_rh_ramification_degree_is_almost_discriminant ] the non - negative integer @xmath153 is equal to the different of the extension @xmath154 of discrete valuation rings , though we will not use this fact .
it is useful to have some explicit examples of morphisms between projective curves .
let @xmath155 be a polynomial over @xmath43 and let @xmath156 be the algebraic variety over @xmath43 which is the graph of @xmath17 , i.e. @xmath157 let @xmath158 be the morphism @xmath159 and let @xmath160 be the projection map @xmath161 .
note that @xmath126 is an isomorphism of algebraic varieties and that @xmath162 ; here we abuse notation and write @xmath17 for the morphism @xmath163 induced by the polynomial @xmath155 .
let @xmath164 denote the projective closure of @xmath156 , obtained by adding a single point at infinity .
the morphisms @xmath165 extend to morphisms @xmath166 , @xmath167 , @xmath168 .
the previous remark implies that the following are all equivalent : 1 .
fubini holds for @xmath168 ; 2 .
fubini holds for @xmath169 ; 3 .
the repeated integrals of @xmath170 are equal . to make use of the examples afforded by the previous remark we now calculate the ramification degrees : we retain the notation of the previous remark .
the ramification degrees of @xmath168 are @xmath171 and the riemann - hurwitz ramification degrees are @xmath172 here @xmath173 denotes the @xmath174-adic valuation on @xmath175 .
the ramification degrees are clear so we only consider the riemann - hurwitz degrees . write @xmath176 so that @xmath168 corresponds to the inclusion of function fields @xmath177 .
a local coordinate @xmath178 at @xmath179 is @xmath174 ; a local coordinate @xmath180 at @xmath181 is @xmath182 . by definition of the riemann - hurwitz ramification degree
, @xmath183 writing @xmath184 for some polynomial @xmath185 gives @xmath186 secondly , @xmath187 and local parameters are given by @xmath188 , @xmath189 ; therefore the riemann - hurwitz ramification degree at infinity is given by @xmath190 for any integer @xmath191 not divisible by @xmath192 , let @xmath193 in the previous remark
. then @xmath168 is unramified away from @xmath194 and infinity , with @xmath195 .
thus fubini s theorem holds for @xmath17 ( or , equivalently , for the set @xmath196 ) if and only if @xmath197 ; that is , if and only if @xmath198 . however ,
now assume @xmath124 and set @xmath199 .
then @xmath168 is unramified outside infinity , where it is wildly ramified of degree @xmath200 .
thus fubini s theorem holds for @xmath17 ( or , equivalently , for the set @xmath196 ) if and only if @xmath201 ; that is , if and only if @xmath202 .
taking @xmath203 in the previous two paragraphs shows that fubini fails for one of the sets @xmath204 , @xmath205 or that @xmath200 is an idempotent in @xmath56 .
the example shows that fubini s theorem can fail when in finite characteristic : [ theorem_fubini_fails_in_char_p ] assume @xmath206 and that @xmath207 in @xmath56 . then there exists a subset of @xmath208 for which fubini s theorem does not hold .
if fubini does hold for the sets @xmath209 and @xmath210 of the previous example then it follows that @xmath211 . but
then fubini does not hold for @xmath204 , unless @xmath212 in @xmath56 .
now we prove the next main result , namely that fubini s theorem forces @xmath55 , our arbitrary euler characteristic on the algebra of constructible sets , to be the usual euler characteristic of a curve : [ theorem_euler_characteristic_is_usual_one ] suppose that @xmath213 and that fubini s theorem is true for any non - constant , separable , tame morphism @xmath214 from a projective curve to the projective line .
then for any projective curve @xmath102 we have @xmath215 , where @xmath185 is the genus of @xmath102 . for any integer @xmath191
not divisible by @xmath192 , the morphism @xmath168 induced by @xmath193 is separable and tame ; therefore we may apply fubini s theorem to deduce @xmath198 .
therefore @xmath211 , which agrees with the desired genus formula .
now let @xmath102 be a projective curve over @xmath43 . by a classical result of algebraic geometry (
* prop 8.1 ) there is , for any @xmath11 sufficiently large ( depending on the genus @xmath185 of @xmath102 ) , a non - constant morphism @xmath214 of degree @xmath11 with the property that any fibre contains at least @xmath216 points . for @xmath11
not divisible by @xmath192 such a morphism is separable and tame ; therefore we are permitted to apply fubini s theorem , deducing @xmath217 but this is nothing other than the riemann - hurwitz formula for the morphism @xmath105 ; so we obtain @xmath215 as claimed .
this allows us to strengthen the observation that fubini fails in finite characteristic : suppose that @xmath213 and that fubini s theorem is true for any non - constant , separable , tame morphism between projective curves
. then fubini s theorem holds for a separable morphism between projective curves if and only if the morphism is tame .
the previous result implies that @xmath215 is the usual euler characteristic of any projective curve @xmath102 .
suppose that @xmath104 is a separable morphism of projective curves which is not everywhere tame .
then the riemann - hurwitz formula tells us that @xmath218 which is incompatible with fubini s theorem for @xmath105 as @xmath149 for all @xmath148 with at least one value of @xmath114 for which we do not have equality .
[ remark_wild_ramification ] more precisely , in the situation of the previous result , we have @xmath219 where @xmath220 is defined by @xmath221 ; here @xmath222 denotes the different of the extension @xmath154 of discrete valuation rings ( see also remark [ remark_rh_ramification_degree_is_almost_discriminant ] ) . @xmath220
measures the wild ramification at @xmath114 .
an euler characteristic is typically considered an object of tame mathematics @xcite , and so this formula is remarkable in that it expresses wild information purely in terms of tame .
a two - dimensional local field is a complete discrete valuation field @xmath126 whose residue field @xmath223 is a usual local field .
such a field is not locally compact but a theory of integration on such spaces has been developed @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite .
in @xcite the author proved that the characteristic function of @xmath224 fails to satisfy fubini s theorem ; in fact , @xmath225 and @xmath226 .
this is similar phenomenon to what we have just observed for the euler characteristic @xmath55 .
these results suggest interpreting the riemann - hurwtiz formula as a modified repeated integral , adjusted in a suitable way to ensure that fubini s theorem holds .
perhaps it is possible to modify the two - dimensional integration theory in a similar way .
in the previous section , we in fact only considered interchanging the order of integration in morphisms all of whose fibres were finite .
this brief section is a study of the possible euler characteristics which do satisfy this restricted version of fubini s theorem .
let @xmath0 be a set with structure @xmath19 .
an euler characteristic @xmath55 is said to be _ strong _ if and only if whenever @xmath61 is a definable function between two definable sets such that there exists a positive integer @xmath11 , with @xmath227 for all @xmath228 , then @xmath229 .
a strong euler characteristic satisfies fubini s theorem in a very weak sense . for suppose
@xmath55 is an euler characteristic , @xmath230 , @xmath231 are definable , and @xmath61 is an @xmath232 mapping as in the definition ; set @xmath233
. then fubini s theorem holds for @xmath146 if and only if @xmath229 .
it is straightforward to establish non - existence in certain cases and uniqueness in others : suppose @xmath43 is an algebraically closed field , of finite characteristic @xmath234 , with the structure of constructible sets ; then no strong euler characteristic exists .
this is just a restatement of theorem [ theorem_fubini_fails_in_char_p ] , where the counterexample did not require @xmath55 to satisfy the full fubini property , but merely be strong .
suppose @xmath43 is an algebraically closed field , of characteristic zero , with the structure of constructible sets ; then at most one strong euler characteristic exists , and it is @xmath71 valued .
let @xmath235 be strong euler characteristics , for @xmath236 .
the algebra of constructible subsets of @xmath45 is generated by the irreducible closed subsets , and therefore it is enough to establish @xmath237 for any irreducible closed @xmath238 ; this we do by induction on the dimension @xmath97 of @xmath92 .
let @xmath239 be the closure of @xmath92 in @xmath240 ; then @xmath241 has dimension strictly less than that of @xmath92 , and so , by the inductive hypothesis , it is enough to establish @xmath242 .
let @xmath243 be a finite projective morphism ; this always exists ( see e.g. ( * ? ? ?
6.4.27 ) ) .
let @xmath244 denote the points at which @xmath239 is non - singular , or at which @xmath17 is not tale ; this is closed in @xmath239 by ( * ? ? ?
4.2.24 , cor .
since morphisms of finite type are closed , @xmath245 is an open subset of @xmath246 , and it is non - empty because it contains the generic point ( here it is important to observe that @xmath247 is a _
separable _ extension of fields ) .
hence the restriction of @xmath17 to @xmath248 is a finite tale morphism to @xmath249 , i.e. an tale cover , of degree @xmath250 ; the assumption that each @xmath235 is strong implies @xmath251 for @xmath236 .
moreover , @xmath252 and @xmath253 are both @xmath254 , and therefore the inductive hypothesis lets us deduce @xmath255 it remains only to prove that our two euler characteristics agree on @xmath246 .
decomposing projective space into a disjoint union of constructible sets @xmath256 and using multiplicativity of each @xmath235 on products , we have finally reduced the problem to proving that @xmath257 .
but the argument of the first paragraph of theorem [ theorem_euler_characteristic_is_usual_one ] , which is valid for any strong euler characteristic , establishes that @xmath258 for @xmath236 .
if @xmath259 then a strong euler characteristic does exist on the structure of constructible sets , namely the topological euler characteristic .
this follows from the classical result that if @xmath260 is an @xmath11-sheeted covering of a cw - complex @xmath49 , then @xmath261 . the lefschetz principle ( i.e. that the first order theory of algebraically closed fields of characteristic zero is complete ; see @xcite for a classical discussion of this principle ) now implies that a strong euler characteristic exists for any algebraically closed field of characteristic zero . the inclusion of this material is inspired by @xcite and @xcite , where strong euler characteristics ( in fact , the definition of strong in these papers is slightly stronger than the definition given here ) are discussed from the perspective of model theory . in @xcite , it is proved that a universal strong euler characteristic @xmath262 exists , and so our previous theorem and remark prove that if @xmath43 is algebraically closed of characteristic zero , then @xmath263 .
now we generalise the results of the previous section from curves to surfaces .
@xmath43 continues to be an algebraically closed field , and @xmath55 is a fixed @xmath56-valued euler characteristic on the structure of constructible sets . in this section
, surface means a smooth , two - dimensional , irreducible algebraic variety over @xmath43 , whereas a curve is merely a one - dimensional , reduced , algebraic variety over @xmath43 if @xmath264 is a finite morphism between projective surfaces of degree @xmath11 , then let @xmath265 be the set of @xmath266 such that @xmath107 does not contain @xmath11 points .
zariski s purity theorem ( see e.g. ( * ? ? ?
8.2.15 ) or @xcite ) states that @xmath31 is pure of dimension one ; let @xmath267 be its irreducible components , and let @xmath268 be the degree of the morphism @xmath269 ( note that the degree is well - defined , as the base curve is irreducible , though the covering curve @xmath270 may be reducible ) . using this data
we may prove an analogue of lemma [ lemma_fubini_formula_for_curves ] : let @xmath264 be a finite morphism between projective surfaces , with notation as in the previous paragraph .
then fubini holds for @xmath105 ( in the same sense as the previous section ) if and only if the following formula relating @xmath271 and @xmath272 is satisfied : @xmath273 where @xmath274 denotes the number of local branches of @xmath275 at @xmath114 . if @xmath55 is a strong euler characteristic then this formula holds .
the normalisation of @xmath31 is by definition @xmath276 , where @xmath277 is the normalistion of the irreducible curve @xmath275 .
write @xmath278 , and let @xmath279 be its normalisation in the same way as @xmath31 ; the functoriality of normalising implies that there is an induced morphism @xmath280 such that @xmath281 .
let @xmath282 be a large enough finite set of points such that @xmath283 includes all singular points of the curve @xmath31 , @xmath284 includes all singular points of the curve @xmath285 , and @xmath286 includes all points of ramification of @xmath287 .
then @xmath288 and @xmath289 induce isomorphisms @xmath290 and @xmath291 ; therefore @xmath292 further , as we saw in the proof of lemma [ lemma_fubini_formula_for_curves ] , @xmath293 since @xmath294 , we have @xmath295 ; combining the last few identities gives @xmath296 to complete the proof , combine this identity with @xmath297 where @xmath298 denotes the number of irreducible components of @xmath31 which pass through @xmath116 ( note that @xmath299 ) .
when @xmath259 and @xmath300 is the topological euler characteristic , which we have remarked earlier is a strong euler characteristic , then the theorem proves that @xmath301 the lefschetz principle now implies that the formula remains true if we replace @xmath43 by any algebraically closed field of characteristic zero , and @xmath302 by the @xmath303-adic euler characteristic ( = alternating sum of betti numbers of @xmath303-adic tale cohomology of @xmath304 , = degree of the second chern class of @xmath304 ) .
this generalisation of the riemann - hurwitz formula to surfaces is due to b. iversen @xcite , who established it with purely algebraic techniques by studying pencils of curves on the surfaces .
iversen remarks in his paper that a more topological proof should be possible when @xmath259 , and our approach provides that . a natural question now to ask
is whether an analogue of the theorem holds in higher dimensions .
if @xmath305 is a finite morphism between @xmath97-dimensional smooth projective varieties over @xmath43 , then the branch locus will be pure of dimension @xmath306 , so one can hope to obtain results by induction on dimension . the difficulty which appears when the branch local has dimension @xmath307 is that there is no functorial way to desingularise .
it is unclear to the author at present how significant a problem this is .
another interesting question concerns the situation in characteristic @xmath200 .
we noted in remark [ remark_wild_ramification ] that , for curves , the difference between the euler characteristic and the integral over the fibres was a measure of the wild ramification . for surfaces ,
the situation is more complex , since the wild ramification of surfaces is not fully understood .
however , assuming that there is no ferocious ramification present ( this is when inseparable morphisms between curves appear ) , i. zhukov @xcite has successfully generalised iversen s formula by defining appropriate ramification invariants ; this provides an explicit formula for @xmath308 in terms of the wild ramification of the cover .
the most interesting related problem is the local situation ; that is , ramification of local fields .
fix a complete discrete valuation field @xmath126 with perfect residue field @xmath309 , and let @xmath310 denote its algebraic closure .
fix a finite galois extension @xmath311 with galois group @xmath312 , and define the usual ramification objects as follows : @xmath313 one proves that @xmath314 is a strictly increasing , piecewise linear , function @xmath315 , and defines the _ hasse - herbrand _
function @xmath316 to be its inverse .
upper ramification on the galois group is defined by @xmath317 .
we now explain a geometric interpretation of these formulae .
since @xmath309 is perfect , @xmath318 is monogeneic ; let @xmath319 be a chosen generator , with minimal polynomial @xmath320 $ ] .
extend @xmath321 to all of @xmath310 to give a @xmath322-valued valuation ; we will write @xmath323 for any real number @xmath324 to denote the closed ball of radius @xmath324 . by some rigid geometry , model theory , or explicit calculations ,
it is known that @xmath325 may be written in a unique way as a disjoint union of closed balls .
let @xmath326 denote this set of balls , and note that @xmath312 acts on it transitively since each ball contains at least one root of @xmath17 .
so , for any @xmath327 , the kernel of the action of @xmath312 on @xmath331 is @xmath332 .
further , the definition of the hasse - herbrand function implies that @xmath333 at least away from the ramification breaks , and therefore that @xmath334 since both sides are @xmath335 at @xmath336 .
but @xmath337 , and so @xmath338 for all @xmath327 .
if we think of `` the number of connected components '' as a measure , then ( @xmath339 ) is a repeated integral taken over certain fibres .
this geometric approach to ramification is used by a. abbes and t. saito @xcite @xcite to develop ramification theory for complete discrete valuation fields with imperfect residue field , using rigid geometry .
l. xiao has written a good overview @xcite of their theory and established integrality of various conductors @xcite @xcite .
i would not have been able to give the discussion above without the help of g. yamashita .
perhaps it is possible to replace the rigid geometry techniques by model theory , using the rich model theoretic structure of algebraically closed valued field , just as we have explored ramification of surfaces using the model theory of algebraically closed fields . , measure , integration and elements of harmonic analysis on generalized loop spaces , in proceedings of the st .
petersburg mathematical society .
xii , vol . 219 of amer .
2 , providence , ri , 2006 , amer . math .
soc . , pp . | we relate fubini s theorem for euler characteristics to riemann - hurwtiz formulae , and reprove a classical result of iversen .
the techniques used include algebraic geometry , complex geometry , and model theory .
possible applications to the study of wild ramification in finite characteristic are discussed . |
x - waves , originally introduced by lu and greenleaf in acoustics,@xcite can be simply described as a non - monochromatic superposition of plane - waves with the same wavevector component along a given direction of propagation .
the resulting pulsed beam can travel undistorted without neither diffraction or dispersion , even in the absence of a nonlinear self - action .
such x packets have been the subject of intense research in several fields , as recently reviewed in @xcite .
the first experimental evidence of x - wave formation during nonlinear optical processes was reported in @xcite .
subsequently , x - related results have appeared in @xcite , including a theoretical analysis and discussion on the generation and existence of this class of spatially and temporally localized waves in nonlinear media @xcite .
the above mentioned experimental and theoretical results have stimulated further investigations on the role of x - waves in nonlinear processes , as well as their description in terms of linear waves with a convenient envelope approach , @xcite and their extension to quantized fields .
@xcite within the active field of optical self - invariant pulsed beams ( see among others @xcite ) , it was recognized that x - waves may play a fundamental role in all the nonlinear processes encompassing spatio - temporal mechanics , well beyond the realm of quadratic frequency mixing where they had been originally observed .
x - waves have been also predicted in periodic media and bose condensed gases .
@xcite x- or _ progressive undistorted waves _ in focusing or defocusing kerr media can be associated to the instability of wide beams propagating in the presence of a refractive index linearly dependent on intensity .
@xcite this effect has also been investigated in quadratically nonlinear media @xcite .
very recently light filamentation with femtoseconds pulses in water , @xcite was explained by the `` nonlinear x - wave paradigm '' .
@xcite the authors pointed out that these spatio - temporal packets can be highly dynamic , i.e. they are continuously generated , undergoing splitting and replenishment with an `` average '' invariant propagation .
such a picture points out the robustness of x - waves with respect to the numerous non - trivial effects that may take place in 3 + 1d nonlinear processes . on the contrary , self - trapped light bullets , which do not exist in the absence of a nonlinear response , are extremely sensitive to the specific model .
x - wave dynamics seems strongly connected to pulse splitting during self - focusing in normally dispersive media , as originally predicted in 1992 @xcite , and later investigated by several authors .
@xcite experimental investigation of splitting / replenishment has been reported in @xcite .
the generation of x - waves may also be taken as the basis for the interpretation of the conical emission observed during the propagation of powerful ultrashort laser pulses in air , or water .
@xcite in this manuscript a preliminary settlement of x - wave nonlinear dynamics is attempted .
while the forthcoming theory is developed with reference to nonlinear optics , as outlined in @xcite it applies to bose einstein condensation ( i.e. to `` matter x - waves '' ) , as well .
the central idea is that , if these wave - packets can be considered somehow as `` modes of free space '' , with a given direction of propagation @xmath0 , then their nonlinear evolution is essentially two dimensional , involving @xmath0 and time .
thus , under certain conditions , the problem can be reformulated and strongly simplified by adopting the well known approach of guided wave nonlinear optics .
once an appropriate x - wave expansion is determined , it is not straightforward to write down `` coupled x - wave equations '' , in analogy with coupled mode equations , for the simple reason that these spatio - temporal beams are not normalizable .
the first step is then to build finite energy solutions , and then use the resulting superposition of x - waves to investigate nonlinear dynamics .
section i and ii cover the early stages of this approach . in section
iii nonlinear regimes are addressed via a perturbative expansion .
this is legitimated by the fact that progressive undistorted waves do exist in the absence of nonlinearity .
the analysis clarifies in which sense x - waves are robust , and may constitute a `` paradigm '' for nonlinear 3d+1 dynamics . in section iv
i consider the highly nonlinear regime .
this is possible while limiting to a specific x - wave .
if the wave packet is spontaneously generated during a nonlinear process ( e.g. from a bell shaped pulsed beam ) the resulting picture , for example the emergence of breathing solutions , provides a relevant insight for interpreting reported numerical and experimental findings .
the last sections deal with some corollaries : compression and chirp of x - waves , and the relation with the integrable 1d nonlinear schroedinger equation .
the appearance of an integrable model , enforcing the use of categories such as `` breathers '' , and its effectiveness in interpreting experimental and numerical investigations , points to an intriguing connection between nonlinear x - waves and solitons .
the wave equation ( at angular carrier frequency @xmath1 ) describing paraxial propagation in normally dispersive media , at the lowest order of approximation , can be cast as @xmath2 where @xmath3 and its derivatives provide the dispersion . in a reference system traveling at the group velocity of the medium , @xmath4 , it is @xmath5 for the sake of simplicity hereafter i will consider radially symmetric beams , with @xmath6 , . progressive undistorted waves , propagating with inverse differential velocity @xmath7 @xcite can be retrieved by looking for solutions of the form @xmath8 , thus ( @xmath9 ) @xmath10 if @xmath11 is written as a superposition of monochromatic bessel beams , @xmath12 , with @xmath13 in frequency units , the corresponding spatio - temporal dispersion relation is @xmath14 in order to have a continuous spectrum along @xmath15 , the left - hand side must be positive , thus ensuring the absence of evanescent waves .
this can be achieved , in the simplest way , by letting @xmath16 , which gives @xmath17 eq . ( [ dispersionx ] ) implies the existence of two types of x - waves : @xmath18 and @xmath19 with the corresponding `` spectra '' @xmath20 and @xmath21 .
they will be denoted as `` slash '' and `` backslash '' x - waves , because of the shape of their spatio - temporal frequency content , as discussed below ( see fig.[figurepsix ] ) .
a general linear x - wave solution,@xcite traveling with inverse differential velocity @xmath7 , is given by @xmath22\text { , } \end{array}\ ] ] which can be rewritten as @xmath23\text{,}\ ] ] with @xmath24 the `` x '' stands for either @xmath25 or @xmath26 , and @xmath27 corresponds to x - waves of the helmholtz equation.@xcite the spatio - temporal spectrum of @xmath28 is given by the fourier - bessel transform pair @xmath29(k_\perp,\omega , z)=\displaystyle \frac{1}{2\pi } \int_{-\infty}^{\infty}\int_0^{\infty } a(r , t , z ) j_0(k_\perp r ) \exp(i\omega t)r dr dt\\ \\ a(r , t , z)=\displaystyle \int_{-\infty}^{\infty}\int_0^{\infty } \mathcal{b}[a](k_\perp,\omega , z ) j_0(k_\perp r ) \exp(-i\omega t)k_\perp dk_\perp d\omega\text { , } \end{array}\ ] ] it is centered at the shifted central frequency @xmath30 , determined by the velocity : @xmath31=\displaystyle \frac{1}{k_\perp } \,f_\slash(\frac{k_\perp}{\sqrt{k''k } } ) \delta(\omega-\frac{k_\perp}{\sqrt{k''k}}-\frac{\beta}{k '' } ) e^{i(\frac{\beta^2}{2k''}+\frac{\beta\,k_\perp}{\sqrt{k''k}})z}+ \\
\displaystyle \frac{1}{k_\perp } \,f_\backslash(\frac{k_\perp}{\sqrt{k''k } } ) \delta(\omega+\frac{k_\perp}{\sqrt{k''k}}-\frac{\beta}{k '' } ) e^{i(\frac{\beta^2}{2k''}-\frac{\beta\,k_\perp}{\sqrt{k''k}})z}\text { , } \end{array}\ ] ] and it looks like an x in the angle - frequency plane . the two terms in ( [ axspectrum ] ) correspond to slash and backslash x - waves , as shown in figure [ figurepsix ] .
since the pulsed beam travels rigidly ( in modulus ) , an x - shaped spectrum determines the x - shape in the @xmath32 space ( roughly , the far field , i.e. the fourier - bessel transform , resembles the near field ) .
sketch of spatio - temporal spectra of `` slash '' and `` backslash '' x - waves.,width=302 ] the following relation holds useful for any @xmath33 : @xmath34 therefore an x - wave can also be defined as a solution of the equation ( [ nlsdimensional2 ] ) of the type @xmath35 with @xmath36 a formulation that will become handy below .
such a solution contains , in general , an infinite energy .
this is due to the idealized situation ( never available in experiments ) of a precisely defined velocity ( or inverse differential velocity @xmath7 ) .
any finiteness introduced by the experimental setup , such as the spatial extension of the sample , will in general fade the spectrum lineshape around the x , determining uncertainty in @xmath7 .
i will show below that this can be described by a packet of x - waves , with velocities around a given value and finite energy .
considering an x - wave with a specific velocity is thus as idealized as considering an elementary particle of given momentum .
@xcite before proceeding further i emphasize that entire more general classes of x - waves can be generated as discussed in @xcite or , for instance , by taking @xmath37 in ( [ x1 ] ) .
however , the case considered ( @xmath38 ) suffices to represent a wide class of beams , as it will be shown in the next section .
the general solution of eq .
( [ nlsdimensional2 ] ) can be expressed by the fourier - bessel spectrum of the field at @xmath39 , denoted by @xmath40(\omega , k_\perp,0)$ ] : @xmath41 with @xmath42 . by a simple variable change
, the field can be written as a superposition of slash - x - waves , traveling with different velocities .
letting @xmath43 gives @xmath44 while being @xmath45 and @xmath46(\alpha,\beta)\equiv k\,\alpha\,s(\alpha+\frac{\beta}{k''},\sqrt{k\,k''}\alpha)\text{.}\ ] ] an equivalent representation is obtained by backslash - x - waves , replacing the equation for @xmath15 in ( [ ansatzx ] ) with @xmath47 , and being @xmath48(\alpha,\beta)= k\,\alpha\,s(-\alpha+\frac{\beta}{k''},\sqrt{k\,k''}\alpha)\text{.}\ ] ] the variable change ( [ ansatzx ] ) corresponds to span the @xmath49 space by oblique ( slash ) parallel lines in ( [ generallinearsolution ] ) .
( [ xtrasform ] ) is a formulation of the `` x - wave transform '' , first introduced in @xcite and indicated @xmath50(\alpha,\beta)$ ] .
hence the spatio - temporal evolution , including diffraction and dispersion , can be represented by one dimensional propagation of packets with different velocities .
the energy of the pulsed beam is @xmath51 with @xmath52 showing that @xmath53 can be taken as the energy distribution function with respect to the inverse differential velocity @xmath7 . summarizing
, an arbitrary beam can be expanded in a superposition of x - waves traveling with different velocities .
conversely , such a superposition can be used to construct new - classes of physically realizable finite - energy x - waves .
to this extent , orthogonal x waves -first introduced in @xcite for the wave equation- are a fruitful approach ( see also @xcite ) .
with reference to two ( either slash or backslash ) x - wave solutions of ( [ nlsdimensional2 ] ) , denoted by @xmath54 and @xmath55 , with inverse differential velocities @xmath7 and @xmath56 and spectra @xmath57 and @xmath58 , respectively , the inner - product can be defined as the integral of @xmath59 with respect to @xmath60 , extended on the whole space : @xmath61
if @xmath62 and @xmath63 by defining ( @xmath64 and @xmath65 is the generalized laguerre polynomial ) @xmath66 with @xmath67 a parameter with the dimension of time,@xcite it is @xmath68 and the orthogonality condition ( @xmath69 and @xmath70 ) holds : @xmath71 in the following the _ fundamental slash - x - wave _ , with spectrum @xmath72 and spatial profile @xmath73^{3/2}(s - i\delta)^2 } \text{,}\ ] ] will be taken as a prototype of the simplest x - wave with finite power ( i.e. converging transverse integral of @xmath74 , see also @xcite ) .
its intensity profile can be found , e.g. , in @xcite .
when @xmath75 , it is @xmath76 , and @xmath77 the resulting beam is a finite energy x - wave that spreads according to a prescribed velocity distribution function @xmath78 ; this corresponds to the existence of solutions with an arbitrary `` depth of focus '' .
this kind of wave - packet can be written , with reference to slash x - waves , as @xmath79 ( with obvious notation ) while being , as above , @xmath80 by introducing the fourier transform pair of @xmath78 @xmath81 with @xmath82 in time units ( @xmath82 can be roughly kept in mind as the on - axis temporal variable @xmath83 ) , it is from ( [ largeceq ] ) @xmath84 hence 3d linear propagation of an x - wave packet in a normally dispersive medium can be reduced to that of a 1d pulse with anomalous dispersion .
the energy is @xmath85 as discussed in @xcite , by expanding @xmath86 into generalized laguerre polynomials it is possible to express the general solution of ( [ nlsdimensional2 ] ) by orthogonal x - waves ( either @xmath25 or @xmath26 ) .
this has been used in @xcite as an approach to field quantization , and applied to quantum optical parametric amplification .
since x - waves exist even in the absence of a nonlinear susceptibility , it is legitimate to resort to a perturbative approach .
a basic model for nonlinear optical interactions ( under standard approximations ) can be cast in the form @xmath87 where @xmath88 is a nonlinear source , weighed by @xmath89 . after a straightforward expansion of @xmath90 in powers of @xmath89 , the relevant equation is ( [ generalnonlinear ] ) at any order , where the right - hand side stems from the solutions at lower orders .
i will show below that if @xmath91 , the evolution according to ( [ generalnonlinear ] ) always provides a spatio - temporal spectrum corresponding to a progressive undistorted wave .
thus , if an x - wave is taken as the solution of the linear model ( @xmath92 ) , the nonlinearity has the role of `` dressing '' that solution , which still continues to exist .
since this result is valid at any order , it furnishes a general picture of the propagation of self - invariant beams in the nonlinear regime . for an x - wave propagating in a nonlinear medium
, @xmath93 can be interpreted as a function of its field and of its complex conjugate . in the case of harmonic generation
, @xmath93 is some power of the pump beam , traveling at the group velocity of the fundamental frequency .
@xcite what follows can be viewed as a generalization of what discussed in @xcite , with the inclusion of second order dispersion and for a wide class of nonlinear processes ( such as third and higher harmonics generation ) . in a kerr medium , with a refractive index @xmath94 , with @xmath95 the optical intensity , eq .
( [ generalnonlinear ] ) becomes @xmath96 for the solution at the lowest order ( @xmath97 ) , it is possible to take either an x - wave packet @xmath54 around @xmath98 or a wide pulsed beam with negligible diffraction ( with @xmath99 ) .
higher orders are obtained in the form ( [ generalnonlinear ] ) : at the first the rhs is proportional to @xmath100 . as a result of the following analysis ,
the correction to @xmath54 is still a progressive undistorted wave traveling at the same velocity .
since this can be applied at any order , linearly - self - invariant beams are very robust with respect to the nonlinearity . by writing @xmath90 , as a superposition of slash - x - waves , @xmath101 and inserting into ( [ generalnonlinear ] ) ,
one obtains @xmath102 taking the fourier - bessel transform of eq.([tempor1 ] ) , multiplying it by @xmath103 and integrating over @xmath104 and @xmath83 , one founds @xmath105(\alpha,\beta ) e^{i(\alpha+\beta / k'')(\bar\beta-\beta)z}\text{,}\ ] ] where eqs .
( [ ansatzx ] ) apply , and @xmath106(\alpha,\beta)= k\alpha\mathcal{b}[p](\alpha+\frac{\beta}{k''},\sqrt{k\,k''}\alpha)\text{.}\ ] ] eq .
( [ cequation ] ) can be readily integrated with the boundary condition @xmath107 at @xmath39 : @xmath108 \frac{sin(g z)}{g}e^{i[(\alpha+\frac{\beta}{k '' } ) ( \bar\beta-\beta)-\frac{\beta^2}{4k''}]z}\text{,}\ ] ] with @xmath109\text{.}\ ] ] eq .
( [ csolution ] ) can be interpreted in the @xmath49 , or @xmath110 , planes and shows that , for large propagation distances , @xmath111 tends to a dirac @xmath112 centered at @xmath113 : the propagation acts as a spatio - temporal filter , selecting specific combinations of frequencies and wavevectors .
the condition @xmath113 , in the @xmath49 plane , is @xmath114 which gives a hyperbola , as sketched in figure [ figureiperbola ] .
sketch of the generated spatio - temporal spectrum ( symmetrized for @xmath115 ) when @xmath116 .
the straight lines are slash and backslash spectra.,width=302 ] it is apparent that eq .
( [ iperbola ] ) is the dispersion relation corresponding to eq .
( [ x1 ] ) , therefore the resulting pulsed beam is a progressive undistorted wave traveling at inverse differential velocity @xmath98 . in reported experimental results
there is a clear evidence of this spectral hyperbola , see e.g. @xcite , as well as of a structured spectrum,@xcite as that corresponding to the splitting described in the following .
the asymptotes over which energy is concentrated correspond to slash and backslash x - waves .
thus , in any nonlinear process which can be reduced to equation ( [ generalnonlinear ] ) , x - wave packets are spontaneously generated .
i like to stress that one of the arms always starts at @xmath117 , while in the case @xmath99 eq .
( [ iperbola ] ) yields the exact x - shaped spectrum .
the specific features will in general depend on the source spectrum , as elucidated below with an example .
note also that the same analysis holds true , with simple changes , when @xmath118 , with @xmath119 depending on the specific nonlinear process ( e.g. @xmath119 is the wavevector mismatch for optical harmonics generation ) , detailed cases will be reported elsewhere .
the energy distribution function is , after ( [ energy1 ] ) , @xmath120|^2\,(\alpha,\beta)}{k\alpha}\frac{sin(gz)^2}{g^2}\,d\alpha\text{.}\ ] ] as @xmath121 @xmath122 with @xmath123 hence for large @xmath0 @xmath124(\bar\alpha+\frac{\beta}{k''},\sqrt{kk''}\bar\alpha)|^2 \theta_0(\bar\alpha)\text{,}\ ] ] with @xmath125 the unit step function .
the generated x - wave packet grows with a linear efficiency with respect to the propagation distance . in the case
@xmath126 , from the condition @xmath127 in ( [ asymptenergy ] ) it appears that energy is distributed in the intervals @xmath128 and @xmath129 when @xmath116 , and in @xmath130 and @xmath131 when @xmath132 .
this implies that @xmath53 does not have a continuous support ( see the example in figure [ figurebeta ] , discussed below ) for large @xmath133 , with sufficiently separated domains along @xmath7 , this is expected to provide pulse - splitting .
in fact , some components of the generated x - wave will travel at a velocity sensibly different from the pump wave , and after some propagation length satellite packets may appear ( such as , e.g. , those discussed in @xcite ) .
there are two relevant limits for eq .
( [ asymptenergy ] ) .
as @xmath134 , i.e. the pump travels at the linear group velocity of the medium , and @xmath135 , it is @xmath136(\frac{\beta}{2k''},-\sqrt{kk''}\frac{\beta}{2k''})|^2 \theta_0(-\beta)\text{.}\ ] ] eq .
( [ asymptenergy0 ] ) shows that only the portion of the pump spectrum laying on the x - spectrum , i.e. on the line @xmath137 , contributes to the energy of the generated beam .
the propagation filters out all of the spectral components that do not belong to an x - wave .
the other relevant limit corresponds to a pump velocity largely different from the linear group velocity of the medium : @xmath138 . from a physical point of view
, the nonlinear interaction is strongly hampered by the generated beam which rapidly gets separated from the pump , producing a typical `` cerenkov emission '' halo . from ( [ asymptenergy ] ) it is ( @xmath139 ) @xmath140(0,-\sqrt{kk''}\frac{\beta}{k''})|^2 \theta_0(-\beta)\text{.}\ ] ] for large @xmath133 the x - wave generation is strongly inhibited , and exclusively the on - axis spatio - temporal spectrum plays a role . only the branch @xmath128 is relevant in this limit .
furthermore , because of the factor @xmath141 , the energy distribution is expected to be peaked at some value @xmath128 , as shown below .
it is instructive to examine an example , by assuming a gaussian lineshape for the source : @xmath142(k_\perp,\omega)=\mathcal{b}_0 ^ 2\,exp(-\frac{k_\perp^2}{\delta k_p^2}-\frac{\omega^2}{\delta\omega_p^2})\text{,}\ ] ] such that @xmath143 measures the pump beam fluence , and @xmath144 and @xmath145 provide the on axis temporal and spatial spectra , respectively . in the case
@xmath99 ( source traveling at the linear group velocity at @xmath1 ) the energy distribution function , after eq . ( [ asymptenergy ] ) , is @xmath146 showing that the generated spectrum encompasses slash x - waves faster than the source ( @xmath128 ) , and spread in velocity with the characteristic value @xmath147^{-1}\text{,}\ ] ] such that the more the source is spectrally ( in space or time ) narrow the more `` ideal '' ( i.e. un - spreading , @xmath148 small ) is the generated x - wave - packet .
in the case @xmath126 the situation is more complicated . from ( [ asymptenergy ] ) and ( [ pumpexample ] ) , it is @xmath149 when @xmath150 ; while for other values of @xmath7 @xmath151\ ] ] with @xmath152 . in figure [ figurebeta ]
i show an example of the application of eq .
( [ energybeta ] ) , by reporting the normalized energy distribution @xmath153 .
the pump beam , with @xmath1 such that the wavelength is @xmath154 , is chosen with @xmath155 , with @xmath156 , and @xmath157 with @xmath158 .
for the medium parameters @xmath159 , @xmath160 , @xmath161 . as a result , @xmath162 , with a velocity bandwidth @xmath163 , and @xmath164 . in figure [ figurebeta ] , ( a ) and ( b )
, the two branches are displayed for @xmath165 , corresponding to a pump beam velocity of @xmath166 . in this case
, the energy distribution tends to that at @xmath99 , and no pulse splitting is expected ( the two branches are almost contiguous ) . in figure [ figurebeta ] , ( c ) and ( d ) , the two branches are shown for @xmath167 , corresponding to a pump beam velocity of @xmath168 .
two distinct lobes are present , with the most energetic peaked around @xmath169 . in this case
a satellite x - wave packet is expected to separate from the pump beam in propagation .
normalized energy distribution @xmath170 vs inverse differential velocity @xmath7 for two different values of pump velocity determined by @xmath98 . ( a ) branch @xmath128 for @xmath165 ;
( b ) branch @xmath171 for @xmath165 ; ( c ) as in ( a ) with @xmath167 ; ( d ) as in ( b ) with @xmath167 .
all quantities are dimension - less . , width=453 ] the x - wave will in general split for sufficiently long propagation , following the peaks of @xmath53 . at higher orders the resulting packets will become new sources for the perturbation equation , and additional splitting
will appear , consistently with the numerical results ( see e.g. @xcite ) .
this is expected in second - harmonic generation , initially considered in @xcite , where @xmath98 is the temporal walk - off between the two harmonics ( details will be given elsewhere ) ; as well as in dispersive propagation , as described in @xcite , and investigated experimentally in @xcite .
the spectral widening can indeed provide sources traveling at a group velocity different from that at @xmath1 ( @xmath172 ) . in figure [ figurebeta ]
note the difference in the vertical scale between panels ( a ) , ( b ) and ( c ) , ( d ) . in all cases the pump velocity is about @xmath172 , showing that the mismatch between the pump and the linear group velocities at @xmath1 drastically affects the generated x - wave spectrum . with regards to the efficiency of x - wave generation , in the case @xmath99 , by integrating ( [ asymptenergy0 ] ) it
is found @xmath173 the energy grows with @xmath148 , i.e. with an extension of the spatio - temporal spectrum of the pump .
after eq.([efficiency0 ] ) one founds that , for a wide spatial spectrum ( i.e. large @xmath145 , corresponding to a tightly focused pump beam ) , the efficiency is mainly determined by the on - axis temporal spectrum .
conversely , for large on - axis spectrum @xmath144 , the energy of the generated x - wave grows with @xmath145 .
the other relevant limit is @xmath174 , such that the resulting energy distribution is solely determined by the branch @xmath128 , and it is @xmath175 thus the efficiency is mainly affected by the spatial profile of the pump beam and it goes like @xmath176 , consistently with eq .
( [ asymptenergyinfty ] ) .
even when using some working hypotheses , it is possible to deepen the previous study and consider the highly nonlinear regime .
i will consider the multidimensional schroedinger equation : @xmath177 this may provide the correct trend observed in the experiments , including quadratic nonlinearity in certain regimes , ( see e.g.@xcite ) as long as envelope shocks ( see @xcite ) , or higher order phenomena such as plasma formation play a negligible role .
writing @xmath90 as a x - wave expansion ( @xmath178 and @xmath179 ) : @xmath180 and taking the fourier bessel transform of eq .
( [ nlsdimensional3 ] ) , evaluated at @xmath181 , and @xmath182 , the _ coupled x - wave equations _ are derived : @xmath183 the nonlinear polarization is @xmath184e^{-i(\alpha+\frac{\beta}{k''})\beta\,z } = \\ \\ k\alpha\int\,\mathcal{k}(\alpha,\vec\alpha ) x(\alpha_1,\beta_1,z)\,x(\alpha_2,\beta_2,z)\,x(\alpha_3,\beta_3,z)^ * \\ \\ \delta[\alpha+\alpha_3-\alpha_1-\alpha_2+(\beta+\beta_3-\beta_1-\beta_2)/k '' ] \theta(\alpha,\beta,\vec\alpha,\vec\beta , z)d\vec\alpha d\vec\beta\text { , } \end{array}\ ] ] with @xmath185 , @xmath186,@xcite @xmath187 and @xmath188z\}\text{.}\ ] ] a remarkable result is obtained if the solution of ( [ coupledxwaves ] ) is approximated by an x - wave - packet which , in order to avoid unnecessary complexities , is centered around the medium group velocity , i.e. @xmath99 . this is standard in usual coupled mode theory ; @xcite in the present case a `` mode '' is an x - shaped 3d wavepacket . as discussed below , if @xmath0 is sufficiently small , it is possible to take @xmath189 in ( [ nlsq ] ) ; in this regime i write @xmath190 in order to represent the nonlinear modulation ( with envelope @xmath111 ) of an x - wave . for definiteness ,
i consider @xmath62 , where @xmath191 is the spectrum of a basis x - wave .
if @xmath111 is peaked around @xmath99 , all components travel nearly at the same velocity and @xmath189 .
multiplying by @xmath192 and integrating over @xmath13 , from eq .
( [ coupledxwaves ] ) one obtains : @xmath193 with the interaction kernel : @xmath194\\ d\vec\alpha\text{.}\ ] ] eq .
( [ spectralxnls ] ) is a zakharov equation , and taking the fourier transform of @xmath111 ( see ( [ smallcdef ] ) ) one obtains the 1d nonlinear schroedinger equation ( nls ) : @xmath195 hence _ the evolution of an x - wave packet in a nonlinear kerr medium can be approximated by an effective 1 + 1d nonlinear schroedinger equation with a non - homogeneous nonlinearity . _
the latter , given by @xmath196 with dimensions of an inverse area , is expressed by the fourier transform of the interaction kernel @xmath89 : @xmath197 after some algebra it can be written as @xmath198 which is the spatial overlap of the component x - wave profile at @xmath199 , with @xmath200 taking for @xmath201 a solution of ( [ 1dnls ] ) , and @xmath111 its fourier transform according to ( [ smallcdef ] ) , @xmath90 reads @xmath202 with @xmath203 @xmath204 can be considered the basis for the wavelet transform of a with respect to the on - axis temporal variable.@xcite if @xmath189 the spreading due to the different velocities of the component x - waves ( given by the quadratic terms in @xmath7 in the exponential in ( [ waveletbasis ] ) ) is negligible , and @xmath205 can be approximated by its value at @xmath39 : @xmath206 and ( [ avsc ] ) becomes @xmath207 the 3d+1 nonlinear evolution problem ( [ nlsdimensional3 ] ) reduces , under suitable approximations , to a 1 + 1d model ( [ 1dnls ] ) . in the following
i will address the hypotheses leading to this result and to its straightforward consequences .
( [ 1dnls ] ) is valid as far as @xmath189 in ( [ nlsq ] ) .
writing @xmath208 with @xmath209 it must be @xmath210 .
@xmath211 is a quadratic function of @xmath7s ; if @xmath212 is the velocity bandwidth ( @xmath213 ) , and observing that the spectrum of the fundamental x - waves decays as @xmath214 ( thus , roughly , @xmath215 ) , the validity of this approximation leads to a constrained maximization of @xmath211 .
after a simple analysis , it must be @xmath216 while , if @xmath217 , _ the soliton regime _ : @xmath218 and , if @xmath219 , _ the chirp regime _ : @xmath220 as expected , the smaller @xmath221 , the larger @xmath222 .
the above can be recast in a much simpler and more insightful formulation by introducing the dispersion ( @xmath223 ) and the diffraction ( @xmath224 ) lengths . if @xmath225 is the s - width of @xmath201 , which in general does not correspond to the on - axis temporal duration of the pulsed beam ( this holds approximately only in the soliton regime , see ( [ avscsemplified ] ) ) , it is @xmath226 .
if @xmath227 is the beam waist at the center of the pulse , with @xmath228 the spatial spectrum , by the general properties of x - waves it is ( @xmath215 ) @xmath229 , and @xmath230 .
when @xmath67 is much smaller than @xmath231 , the x - wave dominates the whole spatio - temporal beam profile , this is _ the chirp regime_. conversely , when @xmath232 , _ the soliton regime _ is attained . in this case , the on - axis temporal duration is also affected by the envelope @xmath201 , while the x - wave mainly determines the spatial profile ( see ( [ avscsemplified ] ) ) .
for instance , assuming to be in the soliton regime ( as witnessed by non - trivial nonlinear dynamics ) , taking a beam waist of @xmath233 and an on - axis temporal duration @xmath234 it is @xmath235 , taking @xmath159 , @xmath236 and @xmath237 , a value far beyond the propagation distances reported in experiments .
@xcite @xmath222 is typically of the order of the smallest between @xmath223 and @xmath224 , because it measures the distance after which the x - wave packet starts to spread due to its finite energy .
therefore one expects the approximations leading to the 1d nls model eq.([1dnls ] ) to remain valid at least in the early stages of the dynamics , a few centimeters in experiments .
when @xmath0 increases , the nonlinear response gets reduced because of the sliding between the component x - waves , and the propagation becomes in essence the linear evolution of the generated x - wave pattern .
once the validity of ( [ 1dnls ] ) is ascertained , the corresponding nonlinear dynamics is determined by considering the the _ effective nonlinear length _ @xmath238 , with @xmath239 , which relates to the input peak intensity , omitting inessential numerical factors , by @xmath240 ( i.e. @xmath241 is approximately given by the peak power , see below ) . in the definition of @xmath242 ,
the value of @xmath196 is considered at @xmath243 , because prominent nonlinear effects are expected near the peak of the pulsed beam .
the depth of focus of the progressive undistorted wave is such that the on - axis intensity keeps constant in propagation , thus enhancing nonlinear effects .
this is quite similar to the nonlinear dynamics of guided modes , where the tightly confined light may favor nonlinear phenomena . when the temporal duration @xmath231 of the envelope @xmath201 is smaller than the duration of the modulated x - wave , approximately @xmath67 , as far as @xmath244 the nonlinearity will mainly play as a self phase modulation .
when @xmath245 the so called @xmath246 fundamental soliton solution ( with a sech profile for @xmath201 , and @xmath247 the number of eigenvalues in the inverse scattering problem of ( @xmath248 ) , with @xmath249 ) is attained .
however , it can be readily seen that @xmath250 ( in the previous example @xmath251 ) , hence the x - wave dynamics dominates the dispersion .
even if a @xmath246 soliton emerges , it essentially behaves as a non - dispersing non - diffracting x - wave . the situation can be completely different when the peak power increases and @xmath242 reduces . in this case
the parameter @xmath247 , given by @xmath252 , becomes greater than unity , and hence higher - order solitons and breathers are foreseen .
they will be discussed below .
under suitable conditions , satisfied in the early stages of propagation , the nonlinear dynamics of an x - wave in a kerr medium can be described by the 1d nonlinear schroedinger equation with a non - uniform nonlinearity profile @xmath196 , determined by the spatial ( self-)overlap of the progressive undistorted wave .
as i have shown earlier , the whole x - wave space is spanned by the fundamental x - waves .
the fundamental x - shaped profile , integrable with respect to @xmath104 and typically observed in simulations and experiments , is given by @xmath253 ( see ( [ fundamentalxwavespectrum ] ) ) .
i will consider the effective 1d nls with such spectrum .
it can be found that @xmath254 ^ 3}\text{,}\ ] ] which gives @xmath255 , width=302 ] @xmath256 , plotted in figure [ figurefsigma ] , and such that @xmath257 , can be evaluated either numerically or analytically . in the latter case it can be expressed as @xcite @xmath258 \}\text{. } \end{array}\ ] ] therefore , @xmath196 is a bell - shaped function . observing that its spatial extension is of the order of @xmath67 , when the on - axis temporal dynamics is dominated by the @xmath201 envelope ( i.e. the temporal width of @xmath201 is smaller that @xmath67 ) , it can be approximated by @xmath259 : @xmath260 the last relation stems from the fact that the spatial spectrum is @xmath261 .
the effective nls can thus be approximately rewritten as @xmath262 so that the effective nonlinear coefficient has an additional factor , roughly given by @xmath263 ( or equivalently @xmath264 ) , if compared to the plane wave value .
this shows the compensation of diffraction owing to the x - waves , which behave as `` modes of free space '' .
@xmath196 resembles the mode - overlap in a waveguide and the corresponding support to the nonlinear response along propagation .
this holds valid as long as @xmath265 , i.e. as long as the nonlinearity is not averaged out by the `` sliding '' of the x - waves .
the fundamental soliton ( when @xmath266 ) can be expressed in terms of the peak @xmath241 of the energy distribution function ; it reads @xmath267\xi_\slash^{(0)}(r , t , z , s)ds\text{,}\ ] ] which , using ( [ avscsemplified ] ) and ( [ sigma0simplified ] ) , can be approximated by @xmath268 as anticipated , from the expression of @xmath269 , the peak intensity @xmath270 relates to @xmath271 by @xmath272 ( omitting inessential factors )
( [ integratesech ] ) and ( [ sechx ] ) show the `` dressing '' mechanism associated with the nonlinearity : the latter acts only on the shape of the envelope which , even in the linear limit , would travel almost undistorted ( as far as @xmath265 ) . since this analysis can be repeated for any x - wave , nonlinear x - waves are not numerable families of solutions ( like multidimensional solitary waves ) , but have the same `` cardinality '' of the `` space '' of linear x - waves .
conversely , in spite of this remarkable difference with solitary waves , the ( approximate ) validity of an integrable model such as ( [ 1dnlssimplified ] ) seems to establish a strong link with solitons .
clearly this analysis is far from being a rigorous settlement of this conclusion , but could stimulate further research . in the following section
i will discuss some additional consequences of ( [ 1dnls ] ) and ( [ 1dnlssimplified ] ) , pointing out non - trivial nonlinear dynamics of x - waves .
the fundamental x - wave / fundamental soliton discussed above is of limited physical interest , because its invariant propagation is already `` embedded '' in its linear counterpart . if @xmath78 is a very narrow function , or equivalently the @xmath273width @xmath231 of @xmath274 is much wider than @xmath67 , the on - axis profile of the 3d - beam is mainly determined by the x - wave .
this has been indicates as _ the chirp regime_. in eq .
( [ 1dnls ] ) the second derivative can be neglected , and the solution is essentially self - phase modulation : @xmath275\text{.}\ ] ] proceeding as before ( [ xspm ] ) yields the approximate _ nonlinearly chirped basis x - wave _ @xmath276.\ ] ] there is a fundamental difference between this type of chirp and what is typically considered in fiber propagation , @xcite determined by the temporal power profile @xmath277 .
even an idealized x - wave ( corresponding to @xmath78 proportional to a dirac delta with respect to @xmath7 , and hence to a constant @xmath201 ) is chirped in a nonlinear medium because of its spatial profile ( reflected in @xmath278 ) .
since the earliest work on pulse splitting in normally dispersive kerr media , it has been known that , at the initial stages , the on - axis temporal spectrum exhibits a double peak .
@xcite figure [ figurefreqshift ] shows the instantaneous frequency @xmath279 for higher order basis x - waves , more complicated spectral modulations can be expected .
it is natural to identify the origin of this spectral splitting as a consequence of the self - phase modulation considered here .
.[figurefreqshift],width=302 ] it is also well stated that a chirped pulse may compress when propagating under anomalous dispersion .
with reference to the two frequencies at the peaks in figure [ figurefreqshift ] , they propagate according to eq .
( [ 1dnls ] ) with opposite velocities , such that the pulse gets compressed in the temporal domain . in the pulse
splitting considered in @xcite , on - axis temporal compression immediately follows the spectral splitting .
thus , taking the whole process of pulse - splitting and admitting that an x - wave is formed in the initial stage ( as in @xcite ) , the origin of the spectral splitting can be attributed to self - phase modulation of an x - wave .
nonlinear processes are therefore well suited for generating chirped x - waves , such as those recently addressed in @xcite .
while the fundamental soliton and the chirp provide the simplest corollaries of eq .
( [ 1dnls ] ) , the self - trapped behavior of the 3d beam is fundamentally due to the x - shape .
more interesting dynamics can be described referring to multi - soliton solutions ( obviously in _ the soliton regime _ ) .
the @xmath280 solitons @xcite are natural concepts in explaining splitting and replenishment , investigated numerically in @xcite and experimentally in @xcite .
noteworthy , breathing linear x - waves have been reported .
@xcite even if the replenishment dynamics has been examined with reference to water while including additional terms to eq .
( [ nlsdimensional3 ] ) , the most relevant features are somehow taken into account by this `` simple '' model , with plasma formation and higher order dispersion playing a perturbative role .
@xcite highly nonlinear phenomena , such as the shocks considered in @xcite , are expected at very high fluences and will be neglected hereby .
for example , the n=2 soliton @xcite provides the approximate _ breathing nonlinear x - wave _ @xmath281^{3/2}}\text{,}\ ] ] with @xmath282 figure [ fbx ] shows a typical spatio - temporal profile obtained after eq .
( [ breathingx ] ) .
the periodic depletion and replenishment of the x - shaped distribution is apparent . ) of a breathing x - wave .
two isosurfaces are displayed : the darkest ( red ) corresponds to higher intensity.[fbx],height=302 ] in propagation , the two - solitons , or breather- , solution pulsates and the beam evolves retaining most of its energy localized , but exhibiting the non - trivial nonlinear dynamics of the x - wave .
the generation of a breather can also be thought at the origin of splitting in the first stages of propagation .
it is also noticeable that higher order solitons exhibit the spectral splitting typically described in numerical simulations . before concluding , it is fruitful to summarize the picture of the splitting / replenishment phenomenon . at the beginning a wide bell - shaped pulsed beam evolves into an x - wave , owing to the spatio - temporal pattern formation of x - wave instability @xcite
then , an effective anomalous dispersion is experienced by the envelope of the finite - energy x - wave packet , and the _ chirp regime _ is entered with spectral splitting and on - axis compression .
once the envelope width is sufficiently reduced , the _ soliton regime _ starts , and the increased intensity through compression feeds the generation of a higher order soliton , or breather . after some spatio - temporal oscillations , several mechanisms may intervene to stop the periodic behavior , e.g. losses ( eventually of nonlinear origin , such as two - photon absorption ) or simply that , for large propagation distances , the nonlinear response average out due to the sliding between components of the finite energy x - wave packet .
progressive undistorted waves , a generalization of the non - monochromatic realm of self - invariant beams , seem to emerge as a valuable tool in numerous fields of applied research , from telecommunications to biophysics ( supported by the recent experiments in water ) .
their natural appearance in nonlinear processes , analyzed hereby in the optical domain but also available in acoustics as well as in bose einstein condensation , can be considered a fundamental result .
the formation of x - waves during frequency generation appears as natural as the use of their paradigm in interpreting 3d+1 nonlinear dynamics . while x - waves are known to carry infinite energy , and their superposition has been previously employed to build finite energy solutions , in this paper for the first time i have attempted to establish a general -albeit preliminary- theoretical framework for the extension of this approach to the nonlinear case .
the various steps of a basic nonlinear process , such as pulse - splitting in normally dispersive media , can be revisited in terms of x - wave nonlinear dynamics : from the initial pulse compression to splitting / replenishment .
the results open the way to the investigation of elastic collisions between nonlinear x - waves , as well as to establishing a link with parametric solitary waves in quadratic media . according to standard textbooks , a soliton ( or a bullet )
is a non - perturbative solution of a nonlinear wave equation . in this sense
, the non dispersive and non diffracting wave - packets analyzed here can not be considered solitons , because they exists even when a nonlinear susceptibility lacks .
nevertheless , an intriguing connection between progressive undistorted waves and solitons seems to be at the inner stem of numerically and experimentally investigated phenomena . in this respect
nonlinear x - waves can be considered a sort of _ chimera _ of modern nonlinear physics .
i am indebted with prof .
g. assanto ( nooel ) for the fruitful discussions and for a critical revision and with prof . s. trillo ( university of ferrara ) for his interest in this work . | the generation of finite energy packets of x - waves is analysed in normally dispersive cubic media by using an x - wave expansion .
the 3d nonlinear schroedinger model is reduced to a 1d equation with anomalous dispersion .
pulse splitting and beam replenishment as observed in experiments with water and kerr media are explained in terms of a higher order breathing soliton .
the results presented also hold in periodic media and bose - condensed gases . |
the tendency of photon bubbles " to form in magnetized , radiation pressure - supported atmospheres is well - established . in photon bubble instability ,
the atmosphere spontaneously forms a propagating pattern of low - density channels separated by regions of high density .
radiation tends to diffuse through the underdense regions , avoiding the regions of high density . both analytic calculations ( begelman 2001 ) and numerical simulations ( turner et al .
2005 ; hsu , arons & klein 1997 ) indicate that the instability can lead to very large and persistent density contrasts . in this case , the atmosphere becomes leaky " and the net radiation flux can greatly exceed the equilibrium value for a homogeneous atmosphere .
two families of linear photon bubble instabilities were discovered by arons ( 1992 ) and gammie ( 1998 ; see also blaes & socrates 2003 ) , respectively .
the gammie modes depend on the interplay of gas pressure and radiation pressure and are destabilized slow magnetosonic waves . in the nonlinear limit
they become trains of gas pressure - dominated shocks .
the arons modes are an outgrowth of internal entropy modes , and are related to previously - discovered instabilities driven by convection in a strong magnetic field , in the presence of heat conduction ( syrovatskii & zhugzhda 1968 ) .
they are driven by buoyancy , are governed by the rate of radiative diffusion , and do not depend on the presence of gas pressure .
the arons modes occur where the parameter @xmath0 , denoting the ratio of dynamical time to radiation diffusion time across the atmosphere , is smaller than one , indicating that radiation is relatively well - coupled to the gas .
in contrast , the gammie modes occur where radiation can diffuse across the atmosphere faster than the sound crossing time in the gas , a reflection of weak coupling the relevant condition is @xmath1 , where @xmath2 is the ratio of gas pressure to radiation pressure .
both modes depend for their existence on the presence of a magnetic field stiff enough to enforce approximately one - dimensional motion .
the purpose of this paper is to study the nonlinear development of the buoyancy - driven waves discovered by arons ( 1992 ) , and to assess their possible effects on radiation - dominated atmospheres . as in our earlier analysis of the nonlinear development of gammie modes
, we will use analytic techniques based on the assumption of periodic wave trains . in keeping with the essential physical characteristics of the arons modes
, we will ignore gas pressure in our analysis .
this eliminates the possibility of obtaining a series of shock fronts , which prove to be the dominant feature of nonlinear gammie waves .
instead , we will seek and find continuous periodic solutions . because the waves are driven by buoyancy ( and in contrast to our earlier analysis ) , we must carefully account for the vertical structure of the atmosphere .
we do this by expanding our solution in powers of the vertical coordinate @xmath3 , in addition to using a similarity variable to describe the waves . by focusing on wavelengths that are much shorter than the scale height of the atmosphere , we are able to perform a multi - scale perturbation analysis in the radiation pressure ( analogous to a wkb approximation , but nonlinear ) to obtain a model for the nonlinear arons waves .
the plan of the paper is as follows . in
2 we describe the basic equations and approximations relevant to steady - state , nonlinear waves driven by buoyancy in a radiation - dominated atmosphere .
the section concludes with a derivation of the basic equation describing the nonlinear waves .
we solve this equation in 3 , where we also discuss the main features of the waves . in
4 we interpret our results and discuss the waves global effects on atmospheres .
we emphasize that the existence of steady - state modes requires the presence of a distributed radiation source throughout the atmosphere .
absent this source , the atmosphere is expected to collapse due to the increasing leakage of radiation with height .
we attribute the collapse of the atmospheres in the hsu et al .
( 1997 ) simulations to this effect , and speculate that atmospheres subject to buoyancy - driven photon bubble instability will generally lose pressure support and collapse on a timescale short compared to the radiation diffusion time in the equivalent homogeneous atmosphere .
we summarize our results and present our conclusions in 5 .
we base our model on a simplified set of radiation - hydrodynamical equations , in which the magnetic field is stiff and vertical .
the latter was assumed in arons ( 1992 ) , although hsu et al .
( 1997 ) performed simulations for an inclined magnetic field .
all results below can be readily generalized to an arbitrary ( uniform ) magnetic field direction .
we assume that all vectors lie in the @xmath4 plane , and that there is a uniform gravitational field @xmath5 .
the five equations that must be satisfied describe continuity , radiative diffusion in both the @xmath6 and @xmath3 directions , momentum conservation parallel to @xmath7 , and conservation of radiation energy . eliminating the radiation flux through the diffusion equation @xmath8 , where @xmath9 is the opacity ( assumed constant ) and @xmath10 is the radiation pressure , we obtain ( for a vertical magnetic field ) @xmath11 @xmath12 @xmath13 - s_0 , \ ] ] where subscripts denote partial differentiation and dimensional coordinates are surmounted by a tilde .
the source term @xmath14 in the energy equation represents an assumed divergence @xmath15 of radiation flux that is intrinsic to the fluid . as we will show in 3 ,
this added energy flux is necessary in order to obtain steady - state wave trains otherwise the atmosphere collapses and no steady solution is possible . for simplicity , we will assume that @xmath16 is a constant , corresponding to a linear variation of injected radiation flux with height .
this is an arbitrary choice ; steady wave train solutions may exist for various dependences of @xmath16 on local state variables .
we nondimensionalize the equations by normalizing the pressure , density , and velocity to fiducial values : @xmath17 , @xmath18 , @xmath19 .
we then define a pressure scale height @xmath20 , where @xmath21 is the isothermal sound speed associated with the radiation pressure .
dimensionless coordinates are given by @xmath22 , @xmath23 , @xmath24 , and a characteristic mach number is defined by @xmath25 .
the diffusivity parameter from arons ( 1992 ) is given by @xmath26 finally , we define a dimensionless radiation source term by @xmath27 .
the dimensionless equations are then @xmath28 @xmath29 @xmath30 - 4\hat \sigma .\ ] ] we seek periodic plane - wave " solutions with wavelength @xmath31 , in which quantities associated with the wave depend on position and time through the combination @xmath32 we henceforth denote differentiation with respect to @xmath33 by a prime .
thus , the wavevector makes an angle @xmath34 with respect to the @xmath35axis , as in the analysis by arons ( 1992 ) .
the quantity @xmath36 is then the @xmath35component of the phase velocity .
the speed of the intersection of a wave front with the @xmath37axis is given by @xmath38 , while the overall phase speed of the waves is @xmath39 .
in addition to the periodic dependence on @xmath33 there must be a secular dependence on @xmath3 .
we will seek solutions in which the wavelength is short compared to the scale height , hence we can expand @xmath40 , @xmath41 and @xmath42 as a taylor series in @xmath3 , with @xmath43 : @xmath44 with similar relations for @xmath41 and @xmath42 .
since we demand that all functions of @xmath33 be periodic , we can define @xmath3-independent wave - averaged quantities by the integral @xmath45 we are free to choose the density normalization so that @xmath46 .
we seek steady - state solutions with no secular mass flow , implying that @xmath47 for all @xmath48 and @xmath49 , where @xmath50 this implies @xmath51 and @xmath52 .
similarly , all higher order terms in the taylor expansion for @xmath53 must vanish .
let us now consider the continuity equation .
to @xmath54 we have @xmath55 which is easily integrated to yield @xmath56 to @xmath57 we have @xmath58 implying @xmath59 const . , and
@xmath60 note that the density scale height is @xmath61 , with @xmath62 corresponding to the @xmath63 atmosphere considered by arons ( 1992 ) .
it is straightforward to carry the analysis of the continuity equation to higher orders in @xmath3 with , for example , @xmath64 const . and @xmath65 .
however , these higher - order terms are not required for our analysis .
we next consider the energy equation .
to @xmath54 we have @xmath66 - 4\hat \sigma,\ ] ] while to @xmath57 the equation is @xmath67 , \qquad\end{aligned}\ ] ] where the terms in the expansion @xmath68 are obtained from the momentum equation ( [ momnon ] ) : @xmath69\ ] ] @xmath70 .\ ] ] we argue below that @xmath71 is not needed at the desired level of approximation . multiplying eq .
( [ momnon ] ) by @xmath40 , we also obtain the useful results : @xmath72\ ] ] @xmath73 .\ ] ] our expansion in @xmath3 is valid only if the wavelength projected on the @xmath35axis is short compared to the scale height , @xmath74 .
( note that this approximation just breaks down for the fastest growing linear waves , according to arons [ 1992 ] . ) in this limit , the inertial terms of the momentum equation those on the left - hand side of eq .
( [ momnon]) should be small compared to the terms on the right - hand side .
this ordering goes along with the assumption that the phase speed of the waves projected along the @xmath35axis should be very subsonic , i.e. , @xmath75 . in this spirit , we introduce @xmath76 as a second expansion parameter for the dimensionless pressure , so that @xmath77 we are free to set @xmath78 , @xmath79 , in order to mimic an exponential atmosphere with @xmath80 , to lowest order . by renormalizing @xmath33 to @xmath81 ( where a prime henceforth denotes differentiation with respect to @xmath82 ) , and treating all orders of @xmath42 and @xmath40 and their derivatives with respect to @xmath83 as @xmath84 , we can expand equations ( [ enzero ] ) and ( [ enone ] ) in powers of @xmath85 . in these expansions
we also treat @xmath86 and @xmath87 as @xmath88 .
the leading terms in both equations are then @xmath89 ; in order to describe periodic wave trains we will need the @xmath90 and @xmath84 solutions to eq .
( [ enzero ] ) , and the @xmath90 solution to eq .
( [ enone ] ) .
carrying out this program , we first note that the terms proportional to @xmath91 in eq .
( [ enzero ] ) and @xmath92 in eq .
( [ enone ] ) are @xmath93 , so they can be ignored .
next , by expanding equations ( [ etazero ] ) and ( [ etaone ] ) we obtain : @xmath94 these results allow us to write down eq .
( [ enzero ] ) to @xmath90 : @xmath95 ( where we have used eq . [ [ uzero ] ] ) , implying @xmath96 now consider eq .
( [ enone ] ) to @xmath90 : @xmath97 substituting @xmath98 and using equations ( [ uzero ] ) , ( [ uone ] ) , and ( [ enzeromp1 ] ) , we obtain @xmath99 , where we have imposed the condition @xmath100 for a periodic solution .
we may then set @xmath101 . finally , we consider the @xmath84 terms in eq .
( [ enzero ] ) . using previous results , we find that @xmath102 .
therefore , the surviving terms are @xmath103 - 4\hat \sigma = { m_0\over m_p } \left [ \left ( { \eta_0'\over \eta_0 ^ 3}\right ) ' + \tan^2\theta \left ( { \nu_{01 } - u_0 ' \over \eta_0}\right ) ' \right ] - 4\hat \sigma , \ ] ] where we have used eq .
( [ nuexp2 ] ) and substituted @xmath101 . using the result that @xmath104 and expressing @xmath105 and @xmath106 in terms of @xmath107 using equations ( [ uzero ] ) and ( [ uone ] ) , we obtain @xmath108 which is the equation describing steady wave trains . to simplify our study of solutions , we renormalize the independent variable to @xmath109 and set @xmath110 , thus eliminating the factor @xmath111 from eq .
( [ etaeq1 ] ) .
the basic equation for our study of wave solutions is therefore @xmath112 where a prime now denotes differentiation with respect to @xmath83 and we have dropped the subscript 0 " from @xmath107 .
we seek continuous periodic solutions of eq .
( [ etaeq ] ) , subject to the constraint that @xmath113 . that such solutions might exist is evident from the fact that eq .
( [ etaeq ] ) can be written in the form @xmath114 where @xmath115 is a function of @xmath40 and @xmath116 .
@xmath117 is also an implicit function of @xmath40 through @xmath115 .
this is , of course , just a generalized version of the hamiltonian equation of energy conservation in a potential . in order for eq .
( [ genericode ] ) to have periodic solutions , @xmath118 must have two real roots , @xmath119 , with @xmath120 for @xmath121 , and the integral @xmath122 which equals half the wavelength , must be finite .
the requirement of an intrinsic energy source in the atmosphere , @xmath123 , is also apparent from the form of eq .
( [ etaeq ] ) .
if we choose @xmath124 , for example , then the first term on the right - hand side of eq .
( [ etaeq ] ) is @xmath125 , precluding a periodic solution unless @xmath126 . in general , for a given @xmath127 and wavelength , @xmath128 is an eigenvalue of the problem
. equation ( [ etaeq ] ) is most easily solved by choosing a maximum density , @xmath129 , then integrating to find the wavelength , minimum density @xmath130 , and @xmath128 , subject to the mean density constraint .
it is better to start integrating from the maximum density rather than the minimum density because , in solutions with large density contrasts , @xmath129 is exponentially sensitive to @xmath130 .
figure 1 shows the density profile of the waves for @xmath131 1.1 , 2 , and 11 , with @xmath124 .
parameters for a wider range of solutions are shown in table 1 . even for wave trains with very low amplitudes , @xmath132 ( fig .
1a ) , the density pattern is far from sinusoidal . at low amplitudes , @xmath133 , the depression in the low - density portion of
the wave is sharper and deeper than the enhancement in the dense region
. this behavior switches markedly for @xmath134 , and for high density contrasts ( fig .
1b ) the dense regions are extremely sharp and narrow , while the low density regions are broad and flat .
figure 2 and table 1 illustrate the variation of minimum density @xmath130 , wavelength @xmath31 ( in units of @xmath83 ) , and intrinsic flux divergence @xmath128 with @xmath135 .
the minimum density varies in the opposite sense to the maximum density , but at large density contrasts this variation is slower than logarithmic .
the wavelength decreases with increasing density contrast , also slower than logarithmically , while the intrinsic radiation source @xmath128 , required for a steady - state solution , increases slightly faster than logarithmically , although still very slowly .
interestingly , the inverse relationship between density contrast and wavelength is opposite to that found in the nonlinear shock train solutions ( begelman 2001 ) . numerical simulations ( turner et al .
2005 ) indicate that the shock train solutions evolve toward longer wavelengths , higher density contrasts , and larger luminosities .
if higher density contrasts and larger luminosities are also energetically favored in the case of buoyancy - driven photon bubbles , then we would expect evolution toward _ shorter _ wavelengths .
indeed , the fastest - growing arons modes have long wavelengths , close to the pressure scale height ( see 4.2 ) , whereas the fastest - growing gammie modes are short . [ cols="^,^,^,^,^,^,^,^,^,^,^,^,^",options="header " , ] the parameter @xmath127 represents the underlying density gradient in the atmosphere , with @xmath124 corresponding to the isothermal " atmosphere @xmath136 assumed by arons ( 1992 ) .
the radiation entropy gradient @xmath137 , so all atmospheres with @xmath138 are convectively stable according to the schwarzschild criterion .
note that , because @xmath139 for atmospheres subject to buoyancy - driven photon bubbles , it may be possible for large convective cells to transport radiation energy without excessive leakage .
( recall that @xmath0 is the ratio of the dynamical time to the diffusion time across a scale height . )
the models with @xmath124 and @xmath140 correspond to convectively stable atmospheres , while the model with @xmath141 is strongly unstable to convection . from fig .
2b , it is apparent that density contrasts are relatively insensitive to @xmath127 .
the one glaring exception to this statement is the fact that no periodic solutions exist for low density contrasts , in the case of @xmath141 .
this is easily understood by considering the right - hand side of eq .
( [ etaeq ] ) . in order to have a periodic solution
, the right - hand side must be negative at the upper turning point ( @xmath129 ) and positive at the lower turning point ( @xmath130 ) .
for sufficiently weak density contrasts and @xmath142 , it is possible to have @xmath143 . in this case , the first term on the right - hand side is negative , implying that @xmath128 must be negative in order to make the right - hand side positive .
but at @xmath129 the first term is positive implying that @xmath144 must be positive , leading to a contradiction .
the quantities @xmath128 and @xmath31 are more sensitive to the density gradient than is @xmath130 . at fixed @xmath129 , @xmath128 increases and @xmath31 decreases with @xmath127 . since larger @xmath127 means a steeper density gradient , these trends might reflect the increasing effectiveness of buoyancy in driving radiation through the atmosphere via the low - density channels .
larger @xmath128 means that more energy is being transported by diffusion through the photon bubbles , while a smaller wavelength could mean that a smaller fraction of a pressure scale is sufficient to get the required driving force .
note that this is opposite to the trend for ordinary convection , in which energy is assumed to be advected by fluid motions , and energy transport is expected to be larger for smaller @xmath127 . in buoyancy - driven photon bubble instability ,
diffusion rather than advection transports the energy .
the dimensional wavelength of solutions , @xmath145 , is related to the dimensionless wavelength @xmath31 ( in units of @xmath83 ) , by @xmath146 combining this equation with eq .
( [ enzeromp1 ] ) , we obtain @xmath147 from our numerical solutions ( table 1 ) we have determined that @xmath148 for the cases with high density contrast . in order for our analysis to be valid , however , it is not sufficient for the wavelength merely to be smaller than the pressure scale height . a stronger condition required by our expansion in powers of @xmath3 is that the wavelength projected onto the @xmath35axis , @xmath149 , must be smaller than @xmath150 .
thus , a necessary condition for the validity of our analysis is that @xmath151 from eq .
( [ enzeromp1 ] ) this condition implies that @xmath152 , thus ensuring the validity of our expansion in @xmath76 . in all cases
@xmath128 is positive , implying that the atmosphere itself must supply extra energy at all heights , in order to support a steady - state wave solution . in physical units ,
the intrinsic flux radiated by the atmosphere is @xmath153 since @xmath154 for the high contrast solutions , these atmospheres must radiate furiously . by definition ,
the energy loss rate due to diffusion in the equivalent homogeneous atmosphere is @xmath155 .
the inhomogeneous atmospheres in our model are losing energy @xmath156 times faster possibly leaking energy on less than a dynamical time .
it is important to remember that our adoption of an intrinsic flux divergence @xmath16 was arbitrary we took this step only because a steady - state wave solution could not be obtained without it . what would happen if , as is more likely , such a powerful intrinsic source term is absent ?
although this case is not covered explicitly by our calculations , the only sensible deduction would seem to be that the atmosphere loses radiation pressure support and collapses on the leakage time or the dynamical time , whichever is longer .
indeed , since the flux used in our radiation hydrodynamical equations is defined relative to the fluid frame , one way for the atmosphere to generate a flux divergence is for the gas to sink in the gravitational potential .
we precluded this possibility by demanding @xmath157 , for mathematical reasons , but this outcome can not be excluded physically . while we have not proven that our nonlinear wave trains represent the outgrowth of arons s ( 1992 ) instability , there are a number of aspects which make this identification plausible .
first , our wave trains are driven by buoyancy forces we guaranteed this by setting the gas pressure to zero ( as did arons in his linear analysis ) and treating inertial terms as higher order than buoyancy terms .
second , we exploited the same near - cancellation in the flux - divergence as arons did .
in his case , a slight imbalance drove the instability by diffusing more energy into the lower parts of bubbles than is removed at the top . in our case
, it supplied the crucial driving term @xmath158 in the wave equation ( [ etaeq ] ) .
third , our waves only appear in the @xmath159 terms of an expansion in height , further illustrating that they are driven by buoyancy effects .
fourth , the vertical component of the wave phase velocity is negative , in agreement with the linear instability ( arons 1992 , equations [ 19 ] and [ 42 ] ; in contrast to the group velocity of the linear waves , which is positive see arons eq . [ 22 ] ) . finally , the wave speeds in our model scale @xmath160 , indicating that the wave speed is set by the diffusion of radiation across the bubble , as in the linear instability . according to the linear analysis ,
the most rapidly growing modes have @xmath161 ( arons 1992 , eq .
[ 44 ] ) , where @xmath162 is the wavenumber . projected on the @xmath35axis , the wavelength is about 10 times longer than the scale height , putting it outside the limit of validity for our analysis .
the fastest growing linear mode that is ( marginally ) consistent with our analysis has @xmath163 .
if we assume that this mode determines the orientation of the nonlinear wave pattern , then the inequality in eq .
( [ kzh ] ) is replaced by an approximate equality . in the limit @xmath164
, we then have @xmath165 , and the dominant density contours will tend to align with the magnetic field .
thus , our nonlinear analysis reproduces another key feature of the linear instability , as well as the numerical simulations by hsu et al .
( 1997 ) .
finally , we note that the only existing numerical simulations of buoyancy - driven photon bubbles ( hsu et al .
1997 ) , while reproducing the linear growth rates adequately , are plagued by the rapid collapse of the simulated atmosphere as the waves saturate .
we conjecture that this outcome is real and reflects the lack of a compensating intrinsic source of radiation in the simulations .
the fastest growing modes grow on a dynamical timescale ( arons 1992 , eq .
[ 45 ] ) , which is similar to the collapse time .
the radiation pressure perturbations associated with the waves is @xmath166 .
these pressure fluctuations will be largely compensated by fluctuations in the magnetic field strength , in order to maintain overall pressure balance in the @xmath37direction .
the change in the magnetic pressure @xmath167 is related to sideways compressional or expansional motion by @xmath168 , where @xmath169 is the sideways displacement caused by the passage of the wave .
now , in order for the magnetic field to enforce approximately one - dimensional motion , we require that the sideways displacement be much smaller than the parallel displacement of a typical fluid element , @xmath170 .
this implies @xmath171 equation ( [ maglimit ] ) appears somewhat different from the analogous criterion presented by arons ( 1992 ) in connection with the linear instability .
it is beyond the scope of this paper to carry out a fully magnetohydrodynamical analysis .
the effects of gas pressure will be most important in the densest regions of the waves , where @xmath172 .
rapid energy exchange between matter and radiation will enforce approximately isothermal conditions across the wavefronts , implying that the gas pressure is proportional to the local density .
if the mean gas pressure is @xmath173 , where @xmath174 , then gas pressure will affect the wave dynamics when @xmath175 approaches @xmath176 , or equivalently , when @xmath177 in such cases , gas pressure may limit the maximum density but is unlikely to have much effect on the values of @xmath31 or @xmath128 .
we have produced an analytic model for periodic , nonlinear waves driven by buoyancy in a radiation pressure - dominated atmosphere with a stiff magnetic field . to isolate the effects of buoyancy
, we ignored gas pressure and performed a multi - scale expansion in powers of the vertical coordinate @xmath3 and wave mach number @xmath76 . all solutions for steady - state ,
periodic wave structures require the existence of a large radiation source term in the atmosphere , which we modeled as a uniform , positive flux divergence term in the energy equation .
physically , we believe that this source term compensates for the enhanced loss of radiation via the waves . without it
the atmosphere would lose pressure support and collapse , on a timescale much shorter than the diffusion time associated with the equivalent homogeneous atmosphere .
based on several similarities of generic properties and physical mechanisms , we identify our wave solutions with the nonlinear development of the photon bubble instability discovered by arons ( 1992 ) .
simulations of the linear growth and saturation of this instability ( hsu et al .
1997 ) were plagued by the rapid collapse of the simulated atmosphere shortly after the waves reached nonlinear amplitudes .
we suggest that this collapse resulted from the rapid release of energy via the waves , as indicated by our analytic model .
buoyancy - driven photon bubbles occur primarily when the radiation diffusion timescale across a pressure scale height is longer than the dynamical time , corresponding to the parameter @xmath0 being smaller than one .
( it may be possible for buoyancy - driven photon bubbles to develop in atmospheres with @xmath178 , provided that the wavefronts are sufficiently oblique to the magnetic field . )
this condition is satisfied in a number of astrophysical systems , including the equatorial regions of radiation - dominated accretion disks ( where @xmath179 , the viscosity parameter ) , accretion columns onto neutron stars in x - ray binaries , massive stellar envelopes well below the photosphere , and gamma - ray burst fireballs .
whether the linear instability develops and saturates in any of these systems is not clear , because the growth of the instability is rather slow .
for example , in accretion disks subject to the magnetorotational instability ( mri ) , the magnetic field structure is expected to change on a timescale comparable to the dynamical ( keplerian ) time .
this is shorter than the growth time of all but ( possibly ) the fastest growing buoyancy - driven photon bubbles .
hence , it would not be surprising if buoyancy - driven photon bubbles were suppressed in accretion disks .
( gas pressure - driven photon bubbles , on the other hand , can grow faster than mri , and are expected to be important in accretion disks [ begelman 2002 , 2005 ; turner et al . 2005 ] . ) in accretion columns onto a neutron star polar cap , however , the magnetic field is anchored into the neutron star crust and its structure is fixed .
we would expect buoyancy - driven photon bubbles to grow to nonlinear amplitudes under these circumstances , perhaps leading to the collapse of the accretion column . what would happen to an atmosphere that collapsed under the action of buoyancy - driven photon bubbles ?
if we treat the collapsing atmosphere as a sequence of hydrostatic models , then both the diffusion parameter , @xmath0 ( @xmath180 for a fixed column density ) , and the ratio of gas pressure to radiation pressure , @xmath181 , are expected to increase .
if @xmath0 became large , then the atmosphere would enter the strong - diffusion regime and gas pressure - driven shock trains would become the dominant form of photon bubble . alternatively , @xmath181 could increase until gas pressure halted the collapse .
a third possibility , which may be relevant in the case of neutron star accretion columns , is that the gravitational binding energy liberated by the accretion flow provides at least part of the compensating flux divergence .
it is not clear , however , whether this would be adequate to prevent the collapse of the column .
i am grateful to jon arons for pointing out the atmospheric collapse problem in his 1997 simulations with hsu and klein .
this research was supported in part by nasa astrophysical theory program grant nag5 - 12035 and by the national science foundation under grants ast-0307502 and phy-990794 .
part of this work was carried out at the kavli institute for theoretical physics at the university of california , santa barbara ; i thank the members of kitp for their hospitality . | we derive an analytic model for nonlinear photon bubble " wave trains driven by buoyancy forces in magnetized , radiation pressure - dominated atmospheres .
continuous , periodic wave solutions exist when radiative diffusion is slow compared to the dynamical timescale of the atmosphere .
we identify these waves with the saturation of a linear instability discovered by arons therefore , these wave trains should develop spontaneously .
the buoyancy - driven waves are physically distinct from photon bubbles in the presence of rapid diffusion , which evolve into trains of gas pressure - dominated shocks as they become nonlinear . like the gas
pressure - driven shock trains , buoyancy - driven photon bubbles can exhibit very large density contrasts , which greatly enhance the flow of radiation through the atmosphere .
however , steady - state solutions for buoyancy - driven photon bubbles exist only when an extra source of radiation is added to the energy equation , in the form of a flux divergence .
we argue that this term is required to compensate for the radiation flux lost via the bubbles , which increases with height .
we speculate that an atmosphere subject to buoyancy - driven photon bubbles , but lacking this compensating energy source , would lose pressure support and collapse on a timescale much shorter than the radiative diffusion time in the equivalent homogeneous atmosphere . |
the last few years have witnessed an impressive development of first - principles description of several hadronic observables by means of lattice qcd ( lqcd ) @xcite .
the studies of the lowest - lying baryon mass spectrum that have been undertaken by different collaborations using dynamical actions with light quark masses close to the physical point @xcite are a remarkable example of this progress .
the quark - mass dependence of the baryon masses provides information on their scalar structure , i.e. sigma terms .
the importance that a model - independent determination of these quantities have for dark matter searches has been recently pointed out .
chiral perturbation theory ( @xmath0pt ) , as the effective field theory of qcd at low - energies @xcite is a suitable tool to analyze the lqcd results . in principle , adjusting the proper low - energy constants ( lecs ) one shall optimally reproduce the quark mass dependence of the lqcd results and extrapolate them to the physical point . @xmath0pt
also provides a framework to ascertain systematic errors like those given by the boundaries of the finite box in which the simulation is performed .
however , a full understanding of current @xmath1-flavor lqcd results using su(3)-@xmath0pt is not yet available .
a neat example in the baryon sector ( b@xmath0pt ) is the difficulties found to describe the quark - mass dependence of the octet- and decuplet - baryon masses @xcite within the heavy - baryon ( hb)@xmath0pt approach @xcite .
different solutions , like the su(2)-flavor hyperon b@xmath0pt @xcite or the use of different variants of cut - off renormalization @xcite , have been proposed besides that , in the last decades there have been a sustained interest in the su(3)-b@xmath0pt description of the baryon - masses @xcite .
recently , it has been shown that a successful study of hyperon phenomenology can be achieved using the covariant approach of su(3)-b@xmath0pt @xcite .
examples include the description of the baryon - octet magnetic moments @xcite the analysis of the semileptonic hyperon decays @xcite and the prediction of the magnetic dipole moments and electromagnetic structure of the decuplet resonances @xcite .
the power counting is restored @xcite using the so - called extended - on - mass - shell ( eoms ) renormalization scheme @xcite and the contributions of decuplet resonances are systematically and explicitly considered @xcite .
a successful extension of this formalism to the analysis of the lqcd results may lead to important and far - reaching phenomenological applications .
the purpose of the present work is to explore the potential of the covariant formulation of su(3)-b@xmath0pt to conciliate the description of hyperon phenomenology and the description of the quark mass dependence of @xmath1-flavor lqcd results , studying those of the pacs - cs collaboration on the baryon - octet and -decuplet masses @xcite .
we have chosen the pacs - cs results because they contain simulations around the physical point and with different strange - quark masses .
this gives us an opportunity to study the extrapolation in both the pion and kaon masses .
moreover , the fine - grained lattices used by this collaboration , @xmath2 fm , allow to neglect the discretization errors in a first instance .
at @xmath3 the following terms in the chiral lagrangian contribute to the octet and decuplet masses @xmath4_\pm\rangle , \label{eq : octetct}\\ & & \mathcal{l}_{t}^{(2)}=\frac{t_0}{2}\bar{t}^{abc}_\mu g^{\mu\nu}t_\nu^{abc}\langle\chi_{+}\rangle + \frac{t_d}{2}\bar{t}^{abc}_\mu g^{\mu\nu}\left(\chi_+,t_\nu\right)^{abc } , \nonumber\end{aligned}\ ] ] where @xmath5 is the trace in flavor space and @xmath6 . in the eqs .
( [ eq : octetct ] ) , @xmath7 introduces the explicit chiral symmetry breaking , and the coefficients @xmath8 , @xmath9 , @xmath10 , and @xmath11 , @xmath12 are unknown lecs . for the calculation of the leading loop contributions to the masses with pseudoscalar mesons ( @xmath13 ) , octet- ( @xmath14 ) and decuplet - baryons ( @xmath15 ) we use the lowest - order @xmath16 lagrangian and the @xmath17 and @xmath18 ones of refs .
the lagrangians involving decuplet - baryons are @xmath19 in the sense that they constrain the unphysical degrees of freedom contained in the relativistic spin-3/2 fields @xcite .
we take the values @xmath20 and @xmath21 for the @xmath16 couplings , and @xmath22 @xcite and @xmath23 @xcite for the @xmath24 and @xmath18 ones .
notice that due to the factor 2 of difference between the hb and covariant conventions on @xmath25 and @xmath26 @xcite , our value for @xmath25 is larger than the one often used in the hb literature @xcite . for the meson decay
constant we take an average @xmath27 with @xmath28 mev .
for the masses of the pseudoscalar mesons we use @xmath29 mev , @xmath30 mev .
the mass of the @xmath31 meson is fixed with the gell - mann - okubo mass relation , @xmath32 . at @xmath3 , the lagrangians in eqs .
( [ eq : octetct ] ) give a tree - level contribution and provide the leading su(3)-breaking to the chiral limit masses @xmath33 and @xmath34 appearing in the free lagrangians .
the breaking is linear in the quark masses and leads to the gell - mann - okubo mass relation for the octet , @xmath35 , and the gell - mann decuplet equal spacing rules , @xmath36 @xcite . in the following ,
we refer to these results that we recover in b@xmath0pt at the leading order ( lo ) calculation of the self - energy as the gell - mann - okubo ( gmo ) results .
the next - to - leading order ( nlo ) contribution to the self - energy of the baryons is given at @xmath37 by the leading loop diagrams @xcite .
these only depend on phenomenologically determined couplings and masses and no new unknown lecs are introduced at this order .
we calculate the loops in the covariant approach and recover the power - counting using the eoms renormalization prescription introduced in ref .
@xcite and generalized for the decuplet loops in ref .
@xcite . from the covariant results we have obtained the hb ones by defining @xmath38 and expanding the loop - functions about the limit @xmath39 @xcite .
a full presentation of the nlo results used in this work can be found in the appendix . at @xmath37
the graphs @xmath40 and @xmath41 give the nlo su(3)-breaking corrections to the baryon masses @xmath42 the coefficients @xmath43 are again clebsch - gordan coefficients that can be found in table [ table : coefficients ] and @xmath44 are loop functions which are different for octet ( @xmath45 ) or decuplet ( @xmath46 ) external baryon lines .
the loop functions can be cast in a compact form in term of feynman - parameter integrals @xmath47 @xmath48 in the last equations we use the dimensionless quantities @xmath49 and @xmath50 for the baryon masses and @xmath51 or @xmath52 for the renormalization scale .
moreover , @xmath53 , where @xmath54 or @xmath55 and @xmath56 with @xmath57 the mass of the meson @xmath13 in the loop . the loop - integrals in eqs .
( [ eq : hbb])-([eq : htc ] ) contain the divergent piece @xmath58 ( @xmath59 ) and we cancel it renormalizing the lecs in the @xmath60-scheme .
the resulting loop - functions still contain analytical finite pieces that break the power counting .
we absorb them into the lecs within an extension of the @xmath60-scheme called eoms - renormalization scheme @xcite .
this is equivalent to redefining the @xmath60-renormalized lecs as @xmath61,\nonumber\end{aligned}\ ] ] @xmath62,\nonumber\end{aligned}\ ] ] @xmath63,\nonumber\end{aligned}\ ] ] @xmath64,\nonumber\end{aligned}\ ] ] @xmath65,\nonumber\end{aligned}\ ] ] @xmath66,\nonumber\end{aligned}\ ] ] @xmath67,\end{aligned}\ ] ] where the functions @xmath68 , @xmath69 , @xmath70 and @xmath71 @xmath72 @xmath73 depend only on the renormalization scale @xmath74 whenever the chiral - limit mass ratio @xmath75 is fixed .
explicit analytical forms of the eoms - regularized loop functions follow doing the feynman - parameter integrals in eqs .
( [ eq : hbb])-([eq : htc ] ) and applying the redefinitions of eqs .
( [ eq : eomslecs ] ) .
one can obtain the heavy - baryon results taking @xmath76 with @xmath77 and expanding about @xmath78 @xmath79 with the function @xmath80 defined as @xmath81 k. jansen , pos * lattice2008 * , 010 ( 2008 ) .
a. walker - loud _ et al .
_ , phys .
d * 79 * , 054502 ( 2009 ) .
s. aoki _ et al . _
[ pacs - cs coll .
] , phys .
d * 79 * , 034503 ( 2009 ) .
h. w. lin _ et al . _ [ hsc coll .
] , phys .
d * 79 * , 034502 ( 2009 ) .
s. durr _ et al .
_ , science * 322 * , 1224 ( 2008 ) .
c. alexandrou _ et al .
_ , phys .
d * 80 * , 114503 ( 2009 ) .
s. aoki _ et al .
_ [ pacs - cs coll .
] , arxiv:0911.2561 [ hep - lat ] . w. bietenholz _ et al .
_ , phys .
b * 690 * , 436 ( 2010 ) .
j. r. ellis , k. a. olive and c. savage , phys .
d * 77 * , 065026 ( 2008 ) ; j. giedt , a. w. thomas and r. d. young , phys .
* 103 * , 201802 ( 2009 ) .
s. weinberg , physica a * 96 * , 327 ( 1979 ) ; j. gasser and h. leutwyler , nucl .
b * 307 * , 779 ( 1988 ) ; h. leutwyler , annals phys .
* 235 * , 165 ( 1994 ) .
j. gasser , m. e. sainio and a. svarc , nucl .
b * 307 * , 779 ( 1988 )
. k. i. ishikawa _ et al .
_ [ pacs - cs collaboration ] , phys .
d * 80 * , 054502 ( 2009 ) .
m. procura , t. r. hemmert and w. weise , phys .
d * 69 * , 034505 ( 2004 ) ; m. procura _ et al .
_ , phys .
d * 73 * , 114510 ( 2006 ) ; v. bernard , t. r. hemmert and u. g. meissner , phys .
b * 622 * , 141 ( 2005 ) ; v. pascalutsa and m. vanderhaeghen , phys .
b * 636 * , 31 ( 2006 ) ; b. c. tiburzi and a. walker - loud , phys . lett .
b * 669 * , 246 ( 2008 ) .
j. f. donoghue , b. r. holstein and b. borasoy , phys .
d * 59 * , 036002 ( 1999 ) ; d. b. leinweber _ et al .
_ , phys .
d * 61 * , 074502 ( 2000 ) ; d. b. leinweber , a. w. thomas and r. d. young , phys .
lett . * 92 * , 242002 ( 2004 ) ; v. bernard , t. r. hemmert and u. g. meissner , nucl .
a * 732 * , 149 ( 2004 ) .
r. d. young and a. w. thomas , phys .
d * 81 * , 014503 ( 2010 ) .
p. langacker and h. pagels , phys .
d * 8 * , 4595 ( 1973 ) ; e. e. jenkins , nucl .
b * 368 * , 190 ( 1992 ) ; v. bernard , n. kaiser and u. g. meissner , z. phys .
c * 60 * , 111 ( 1993 ) ; m. k. banerjee and j. milana , phys .
d * 52 * , 6451 ( 1995 ) ; b. borasoy and u. g. meissner , annals phys .
* 254 * , 192 ( 1997 ) ; p. j. ellis and k. torikoshi , phys .
c * 61 * , 015205 ( 2000 ) ; m. frink and u. g. meissner , jhep * 0407 * , 028 ( 2004 ) ; a. walker - loud , nucl .
a * 747 * , 476 ( 2005 ) ; b. c. tiburzi and a. walker - loud , nucl .
a * 748 * , 513 ( 2005 ) ; b. c. lehnhart , j. gegelia and s. scherer , j. phys .
g * 31 * , 89 ( 2005 ) , a. semke and m. f. m. lutz , nucl .
a * 778 * , 153 ( 2006 ) , a. semke and m. f. m. lutz , nucl . phys .
a * 789 * , 251 ( 2007 ) .
l. s. geng , j. martin camalich , l. alvarez - ruso and m. j. v. vacas , phys . rev .
* 101 * , 222002 ( 2008 ) .
l. s. geng , j. martin camalich and m. j. vicente vacas , phys .
b * 676 * , 63 ( 2009 ) .
l. s. geng , j. martin camalich and m. j. vicente vacas , phys .
d * 79 * , 094022 ( 2009 ) .
l. s. geng , j. martin camalich and m. j. vicente vacas , phys .
d * 80 * , 034027 ( 2009 ) .
j. gegelia and g. japaridze , phys .
d * 60 * , 114038 ( 1999 ) ; t. fuchs , j. gegelia , g. japaridze and s. scherer , phys .
d * 68 * ( 2003 ) 056005 .
v. pascalutsa , phys .
b * 503 * , 85 ( 2001 ) ; v. pascalutsa , m. vanderhaeghen and s. n. yang , phys .
* 437 * ( 2007 ) 125 .
e. e. jenkins , phys .
d * 53 * , 2625 ( 1996 ) , e. e. jenkins _ et al .
_ , phys .
d * 81 * , 014502 ( 2010 ) .
s. r. beane , phys .
d * 70 * , 034507 ( 2004 ) [ arxiv : hep - lat/0403015 ] . | we present an analysis of the baryon - octet and -decuplet masses using covariant su(3)-flavor chiral perturbation theory up to next - to - leading order . besides the description of the physical masses we address the problem of the lattice qcd extrapolation . using the pacs - cs collaboration data we show that a good description of the lattice points can be achieved at next - to - leading order with the covariant loop amplitudes and phenomenologically determined values for the meson - baryon couplings .
moreover , the extrapolation to the physical point up to this order is found to be better than the linear one given at leading - order by the gell - mann - okubo approach .
the importance that a reliable combination of lattice qcd and chiral perturbation theory may have for hadron phenomenology is emphasized with the prediction of the pion - baryon and strange - baryon sigma terms . |
heavy ion collisions at high energy are a situation in which the colliding projectiles contain high densities of gluons , and are possibly subject to the phenomenon of gluon saturation @xcite : when the gluon occupation number approaches the inverse of the strong coupling constant @xmath0 , the interactions among the gluons become strong , and non - linear effects such as recombination become important .
moreover , in this regime a typical collision involves multiple gluon interactions , as depicted in the figure [ fig : cms ] ( left ) . unlike in the dilute regime , multiple gluons from each projectile can participate in the reaction .
this leads to two complications : we a way to organize and resum all the relevant graphs , and we need multi - gluon distributions to describe the projectiles .
both are provided by the color glass condensate effective theory @xcite , in which the incoming nuclei are described as collections of color charges moving at the speed of light along the light - cones . the object that encodes
this information is a probability distribution @xmath1 $ ] , where @xmath2 is the color charge per unit of transverse area of a given nucleus . in the saturated regime , this density is inversely proportional to the coupling , and observables at leading order are obtained by summing an infinite series of tree diagrams , which can be done more conveniently by solving the classical yang - mills equations , in which the charge density @xmath3 plays the role of a source term .
next - to - leading order corrections are made of one - loop diagrams in a classical background field .
when computing nlo corrections , the cgc should be viewed as an effective theory with an upper cutoff @xmath4 on the longitudinal momentum running in the loop ( see the figure [ fig : cgc ] , left ) , in order to prevent double counting the modes already taken into account via the source @xmath3 .
nlo corrections in general contain logarithms of this unphysical cutoff .
for the cgc framework to be consistent , observables should be cutoff independent in the end .
for a given observable , it is possible to make the distribution @xmath1 $ ] cutoff dependent in order to cancel the cutoff dependence coming from the loop correction . however , for this strategy to be useful in practice , two requirements should be satisfied : * the same @xmath1 $ ] should be able to cancel the cutoff dependence of a wide range of observables , * when colliding two such projectiles , each of them should have its own @xmath1 $ ] .
moreover , these distributions should be the same as in simpler collisions such as deep inelastic scattering ( dis ) , that involve only one saturated projectile .
under these conditions , the distribution @xmath1 $ ] can be viewed as an intrinsic property of a saturated projectile , and its cutoff dependence reflects changes in its apparent color content due to a change in the resolution scale .
let us start with a reminder of the situation in inclusive dis off a dense nucleus .
thanks to the optical theorem , the total dis cross - section is related to the forward elastic scattering amplitude of the virtual photon on the nucleus , and at lo it corresponds to the scattering of a quark - antiquark fluctuation of the photon off the classical color field @xmath5 of the nucleus ( see the figure [ fig : cgc ] , right ) : @xmath6 note that the @xmath7 pair couples only to the sources up to the longitudinal coordinate @xmath8 .
the other sources are too slow to be seen by the dipole . at nlo
, one needs corrections involving a gluon , such as the one in the right of the figure [ fig : dis - nlo ] .
the longitudinal momentum of this gluon should be integrated only up to the cutoff @xmath9 of the cgc effective theory . in practice , it is more convenient to include only longitudinal modes in a slice @xmath10 . at leading log accuracy ,
the contribution of the quantum modes in that strip is : @xmath11 where @xmath12 is an operator that contains functional derivatives with respect to the source @xmath3 , called the jimwlk hamiltonian .
these nlo corrections can be absorbed in the lo result , @xmath13\,w_\lambda[\rho]\,\cdots\ ] ] provided one defines a new effective theory with a lower cutoff @xmath14 and a modified distribution of sources @xmath15 $ ] : @xmath16\;w_{\lambda_0 ^ -}\ ; , \quad \mbox{i.e.}\quad \frac{\partial w_{\lambda}}{\partial\ln(\lambda ) } = -{\cal h}\,w_{\lambda}\ ; .\ ] ] by solving this equation ( known as the jimwlk equation ) until the cutoff has been lowered to the value @xmath17 , one can sequentially resum the leading log contributions from each of the @xmath18 slices below the original cutoff .
in high energy nucleus - nucleus collisions , there are now two saturated projectiles , and it is more difficult to extract the logarithms that arise in observables at nlo .
there is however a simple argument explaining qualitatively the universality of the distributions @xmath1 $ ] that describe these projectiles in the cgc framework , illustrated in the figure [ fig : cms ] ( right ) .
the soft gluon radiation responsible for the logarithms of the cutoff takes a long time , much shorter than the collision time that goes like the inverse of the collision energy .
therefore , these gluons must be emitted before the collision , at a time where the separation between the projectiles is space - like .
for this reason , there can not be any cross - talk between the distributions @xmath1 $ ] that describe the two nuclei , and they should be the same as in dis . in order to illustrate how this works in an actual calculation , let us consider the inclusive gluon spectrum at lo and nlo , illustrated in the two diagrams on the left of the figure [ fig : gluon ] . at lo ,
the gluon spectrum is quadratic in the retarded solution of the classical yang - mills equation , @xmath19 @xmath20={\colorb j_1^\nu}+{\colorb j_2^\nu}}\quad,\quad \lim_{t\to -\infty}{\colord{\cal a}^\mu(t,\vec\x)}=0\ ; .\ ] ] at nlo , it involves one - loop graphs in the presence of a background color field .
one can formally express the nlo result as@xcite @xmath21_\u \big[{\colorb a_\k^*}\,{\mathbbm t}\big]_\v + \int\limits_{_{\vec\u\in\sigma } } \big[{\colorb{\bs\alpha}}\,{\mathbbm t}\big]_\u \bigg]\ ; \left.{\colorb\frac{dn_g}{dyd^2\vec\p_\perp}}\right|_{_{\rm lo}}\ ; , \label{eq : lonlo}\ ] ] where @xmath22 and @xmath23 are small perturbations to the classical field @xmath24 , @xmath25 a surface used to set the initial condition for the fields ( e.g. the light - cone ) , and @xmath26 the functional derivative with respect to the initial classical field at the point @xmath27 .
in this formula , the integration over @xmath28 has a logarithmic dependence on the cutoff @xmath29_\u \big[{\colorb a_\k^*}\,{\mathbbm t}\big]_\v + \int\limits_{_{\vec\u\in\sigma } } \big[{\colorb{\bs\alpha}}\,{\mathbbm t}\big]_\u = \ln\left(\lambda^+\right)\,{\colorb{\cal h}_1 } + \ln\left(\lambda^-\right)\,{\colorb{\cal h}_2 } + \cdots\end{aligned}\ ] ] where @xmath30 are the jimwlk hamiltonians of the two nuclei .
this formula is the formal realization of the causality argument exposed at the beginning of this section , since the logarithms associated to the two cutoffs are accompanied by the jimwlk of the corresponding nucleus , without any mixing .
thanks to this formula , one can absorb the logarithms of the cutoff in jimwlk - evolved distributions of color sources , @xmath31 \ ; { \colora w_1\big[\rho_{_1}\big]}\ ; { \colorb w_2\big[\rho_{_2}\big ] } \ ; { \frac{dn_g[\rho_{1,2}]}{dy d^2\vec\p_\perp}}\ ; .\ ] ] the same factorization holds for multi - gluon inclusive spectra@xcite , and more generally for all the inclusive observables , for which eq .
( [ eq : lonlo ] ) is valid .
a natural extension of these results is to study it in the case of quark production . to that effect
, one needs expressions for the quark spectrum at lo and nlo ( see the two diagrams on the right of the figure [ fig : gluon ] ) .
these contributions have been worked out , and it turns out that they are related by a formula similar to eq .
( [ eq : lonlo])@xcite , @xmath32_\u \big[{\colorb a_\k^*}\,{\mathbbm t}+\underline{{\colorb b_\k^\dagger}\,{\mathbbm t}_{_\psi}}\big]_\v } \nonumber\\ & & \qquad\qquad\qquad\qquad\quad { + \int\limits_{_{\vec\u\in\sigma } } \big[{\colorb{\bs\alpha}}\,{\mathbbm t}+\underline{{\colorb{\bs\beta}}\,{\mathbbm t}_{_\psi}}\big]_\u \bigg]\ ; \left.{\colorb\frac{dn_{_q}}{dyd^2\vec\p_\perp}}\right|_{_{\rm lo}}}\ ; , \end{aligned}\ ] ] where the underlined terms are new and contain purely fermionic quantities : @xmath33 and @xmath34 are small fermionic fields , and where the operator @xmath35 is a functional derivative with respect to the initial value of fermion fields on @xmath25 .
this formula is only the first step ; the second part of this program consists in extracting the logarithms from the new fermionic terms in the operator that appears in the right hand side , and to prove that they are also proportional to the jimwlk hamiltonian .
this work is supported by the anr project # 11-bs04 - 015 - 01 . | in this talk , we discuss the factorization of the logarithms of energy in the color glass condensate framework .
# 1#2 p l u v # 1 |
properties of hadrons in nuclear medium is interest in studying the quantum chromodynamics ( qcd ) structure in dense matter , in particular related to the chiral symmetry restoration , phase - transition from quark - glue plasma to hadrons , dynamics of hypernucleus formation , nuclear equation of state ( eos ) etc @xcite .
high energy heavy - ion collisions in terrestrial laboratory provide a unique possibility to study the in - medium properties of hadrons in dense nuclear matter and to extract the high - density behavior of the nuclear symmetry energy ( isospin asymmetric part of eos ) .
it has obtained progress in extracting the in - medium properties of hadrons in dense nuclear matter , in particular for strange particles @xmath2 , @xmath3 , @xmath4 and @xmath5 @xcite .
dynamics of particles produced in heavy - ion collisions near threshold energies is a complicated process , in which the production and transportation in dense nuclear medium would be different in comparison to the in - vacuum cases .
density and momentum dependent potentials have to be implemented in correctly understanding the phase - space distributions of particles .
furthermore , dynamics of isospin particles could be modified by the mean - field potentials in constraining the high - density information of symmetry energy .
besides nucleonic observables @xcite , particles produced in heavy - ion collisions would be preferable probes for extracting the information of high - density phase diagram .
kaons as probing the high - density eos were proposed for the first time @xcite .
the available experimental data from kaos collaboration for @xmath6 production favored a soft eos at high baryon densities associated with transport model calculations @xcite .
similar structure for @xmath4 production on the eos was found in ref.@xcite .
the ratios of isospin particles produced in heavy - ion collisions such as @xmath7 , @xmath8 , @xmath9 etc @xcite , neutral particles such hard photons , @xmath1 etc @xcite , and the flow difference of isospin particles @xcite have been proposed as sensitive probes for extracting the high - density behavior of the nuclear symmetry energy ( isospin asymmetric part of eos ) .
although different interpretations on the @xmath7 ratios are concluded in constraining the high - density symmetry energy with transport models @xcite in combination with the experimental data from the fopi collaboration @xcite .
interplay of the mean - field potentials and corrections on threshold energies associated with production cross sections of particles impacts the constraining of stiffness of symmetry energy .
the in - medium effects on pions and @xmath10 dynamics in heavy - ion collisions have been studied in refs @xcite for threshold energy corrections and in refs @xcite for the pion optical potential . in this work ,
the dynamics of pseudoscalar mesons and hyperons with s=-1 in heavy - ion collisions and the in - medium properties of particles in dense nuclear matter are to be investigated with the lanzhou quantum molecular dynamics ( lqmd ) transport model .
the high - density behavior of nuclear symmetry energy from isospin particles and eta production will be explored .
the article is organized as follows . in section
ii we give a brief description of the recent version of the lqmd model .
the in - medium properties and isospin effects on particle dynamics are discussed in section iii .
summary and perspective on the mechanism of particle production near threshold energies are presented in section iv .
in the lqmd model , the dynamics of the resonances ( @xmath11(1232 ) , n*(1440 ) , n*(1535 ) , etc ) , hyperons ( @xmath4 , @xmath5 , @xmath12 , @xmath13 ) and mesons ( @xmath0 , @xmath1 , @xmath2 , @xmath3 , @xmath14 , @xmath15 ) is described via hadron - hadron collisions , decays of resonances , mean - field potentials , and corrections on threshold energies of elementary cross sections @xcite . besides the hadron - hadron collisions
, we have further included the annihilation channels , charge - exchange reaction , elastic and inelastic collisions in antinucleon - nucleon collisions for understanding antiproton induced reactions @xcite .
the temporal evolutions of the baryons ( nucleons and resonances ) and mesons in the reaction system under the self - consistently generated mean - field are governed by hamilton s equations of motion , which read as @xmath16 the hamiltonian of baryons consists of the relativistic energy , the effective interaction potential and the momentum dependent part as follows : @xmath17 here the @xmath18 and @xmath19 represent the momentum and the mass of the baryons .
the effective interaction potential is composed of the coulomb interaction and the local interaction potential @xmath20 the coulomb interaction potential is written as @xmath21 where the @xmath22 is the charged number including protons and charged resonances .
the @xmath23 is the relative distance of two charged particles , and the @xmath24 being the square of the gaussian wave - packet width .
the local interaction potential is derived from the skyrme energy - density functional as the form of @xmath25 .
the energy - density functional reads @xmath26^{2},\end{aligned}\ ] ] where the @xmath27 , @xmath28 and @xmath29 are the neutron , proton and total densities , respectively , and the @xmath30 being the isospin asymmetry .
the coefficients @xmath31 , @xmath32 , @xmath33 , @xmath34 , @xmath35 and @xmath36 are set to be the values of -215.7 mev , 142.4 mev , 1.322 , 23 mev @xmath37 , -2.7 mev @xmath37 and 0.16 @xmath38 , respectively . a skyrme - type momentum - dependent potential is used in the lqmd model @xcite @xmath39^{2 } f_{j}(\textbf{r},\textbf{p}',t).\end{aligned}\ ] ] here @xmath40 , @xmath41 ( @xmath42 ) and the isospin symbols @xmath43(@xmath44 ) represent proton or neutron .
the parameters @xmath45 and @xmath46 was determined by fitting the real part of optical potential as a function of incident energy from the proton - nucleus elastic scattering data . in the calculation
, we take the values of 1.76 mev , 500 c@xmath47/gev@xmath47 for the @xmath45 and @xmath46 , respectively , which result in the effective mass @xmath48=0.75 in nuclear medium at saturation density for symmetric nuclear matter .
the parameter @xmath49 as the strength of the isospin splitting with the value of -0.65 is taken in this work , which has the mass splitting of @xmath50 in nuclear medium .
a compression modulus of k=230 mev for isospin symmetric nuclear matter is concluded in the lqmd model .
the symmetry energy is composed of three parts , namely the kinetic energy of free fermi gas , the local density - dependent interaction and the momentum - dependent potential as @xmath51 the local part is adjusted to mimic predictions of the symmetry energy calculated by microscopical or phenomenological many - body theories and has two - type forms as follows : @xmath52 and @xmath53 the parameters @xmath54 , @xmath55 and @xmath56 are taken as the values of 52.5 mev , 43 mev , -16.75 mev .
the values of @xmath57=0.5 , 1 , 2 lead to the soft , linear and hard symmetry energy in the domain of high densities , respectively , and the eq .
( 9 ) gives a supersoft symmetry energy , which cover the largely uncertain of nuclear symmetry energy , particularly at supra - saturation densities .
all cases cross at saturation density with the value of 31.5 mev .
the values of slope parameters @xmath58 are 203.7 mev , 124.9 mev , 85.6 mev and 74.7 mev for the hard , linear , soft and supersoft symmetry energies , respectively . and the corresponding 448 mev , -24.5 mev , -83.5 mev and -326 mev for the curvature parameters @xmath59 are concluded .
it should be mentioned the short - range correlation of fermi gas reduces the kinetic energy part of symmetry energy @xcite .
the hyperon mean - field potential is constructed on the basis of the light - quark counting rule .
the self - energies of hyperons are assumed to be two thirds of that experienced by nucleons .
thus , the in - medium dispersion relation reads @xmath60 with @xmath61 and @xmath62 , which leads to the optical potential at the saturation density being the value of -32 mev .
the antinucleon - nucleon potential is similar to hyperons , a factor @xmath63 is introduced to mimic the antiproton - nucleus scattering @xcite and the real part of phenomenological antinucleon - nucleon optical potential @xcite as @xmath64 and @xmath65 with @xmath63=0.25 , which leads to the optical potential @xmath66=-164 mev for an antinucleon at the zero momentum at and normal nuclear density @xmath36=0.16 @xmath38 .
the hamiltonian of mesons ( here mainly concentrating on pseudoscalar mesons ) is constructed as follows @xmath67 here the coulomb interaction is given by @xmath68 where the @xmath69 and @xmath70 are the total numbers of mesons and baryons including charged resonances , respectively .
the energy of pion in the nuclear medium is composed of the isoscalar and isovector contributions as follows @xmath71 the coefficient @xmath72 mev , and the isospin quantity @xmath73 1 , 0 , and -1 for @xmath74 , @xmath75 and @xmath76 , respectively . the isospin asymmetry @xmath30 and the quantity @xmath77 adjusts the isospin splitting of pion optical potential .
usually , we take the @xmath78 in the model . here , we have two choices in evaluation of the isoscalar part , the phenomenological ansatz @xcite and the @xmath11-hole model @xcite . with the framework of the phenomenological ansatz ,
the dispersion relation reads @xmath79 with @xmath80 the phenomenological medium dependence on the baryon density is introduced via the coefficient @xmath81 with the parameter @xmath82 and the saturation density @xmath36 in nuclear matter .
impact of the phenomenological approach on pion dynamics in heavy - ion collisions has been investigated in refs @xcite . on the other hand ,
the pion self - energy in the nuclear medium have been studied via the @xmath11-hole model @xcite .
the in - medium pion dispersion relation consists of a pion branch ( smaller value ) and a @xmath11-hole ( larger value ) branch , which become softened and hardened with baryon density in nuclear matter , respectively .
thus , the dispersion relation reads @xmath83 the probabilities of the pion component satisfy the relation @xmath84 the value of the probability is determined from the pion self - energy as @xcite @xmath85 where the pion self - energy is given by @xmath86 with the migdal parameter @xmath870.6 and @xmath88 the @xmath89 , the @xmath90 , @xmath91 and @xmath92 are the masses of pion , nucleon and delta , respectively .
the @xmath93 coupling constant @xmath94 and the cutoff factor @xmath95 .
two eigenvalues of @xmath96 and @xmath97 are obtained from the pion dispersion relation as @xmath98 the energy balance in the decay of resonances is satisfied with the relation @xmath99 where the @xmath100 and @xmath101 are the momenta of resonances and pions , respectively .
the optical potential can be evaluated from the in - medium energy @xmath102 . shown in fig .
1 is the pion energy and optical potential as functions of the pion momentum and baryon density in units of pion mass and saturation density .
the attractive potential is obtained with increasing the pion momentum .
isospin splitting of the pion potential appears and the effect is pronounced in the domain of high baryon density , which impacts the charged pion ratios in heavy - ion collisions .
influence of the in - medium effects on the charged pion ratio is also investigated within a thermal model in ref .
@xcite .
the hamiltonian of @xmath1 is composed of @xmath103 the eta optical potential is evaluated from the dispersion relation based on the chiral perturbation theory @xcite as @xmath104 with @xmath105 and @xmath106 . the pion decay constant @xmath10792.4 mev , @xmath108280 mev and @xmath1090.4 fm .
the optical potential is given by @xmath110 with the eta mass @xmath111=547 mev .
the value of @xmath112 -94 mev is obtained with zero momentum and saturation density @xmath113 .
the attractive potential is used in this work as shown in fig .
2 . up to now , although it has not consistent conclusions on the depth of the @xmath1 potential in nucleus
. basically the attractive @xmath1-nucleon interaction got to be favored by different models @xcite .
even the existence of bound @xmath1-nucleus was pointed out @xcite .
the kaon and anti - kaon energies in the nuclear medium distinguish isospin effects based on the chiral lagrangian approach as @xcite @xmath114^{1/2 } \nonumber \\ & & + b_{k}\rho_{i}+\tau_{3}d_{k}\rho_{i3}\end{aligned}\ ] ] and @xmath115^{1/2 } \nonumber \\ & & -b_{k}\rho_{i}-\tau_{3}d_{k}\rho_{i3},\end{aligned}\ ] ] respectively . here
the @xmath1160.333 gev@xmath117 , the @xmath118 and @xmath119 are 0.18 gev@xmath47@xmath117 and 0.31 gev@xmath47@xmath117 , respectively , which result in the strengths of repulsive kaon - nucleon ( kn ) potential and of attractive antikaon - nucleon @xmath3n potential with the values of 27.8 mev and -100.3 mev at saturation baryon density for isospin symmetric matter , respectively .
the @xmath120=1 and -1 for the isospin pair k@xmath121(@xmath122 ) and k@xmath123(k@xmath124 ) , respectively .
the parameters @xmath125=0.0298 gev@xmath47@xmath117 and @xmath126=0.111 gev@xmath117 determine the isospin splitting of kaons in neutron - rich nuclear matter .
the optical potential of kaon is derived from the in - medium energy as @xmath127 .
the values of @xmath128=1.056 and @xmath129=0.797 at normal baryon density are concluded with the parameters in isospin symmetric nuclear matter .
the effective mass @xmath130 is used to evaluate the threshold energy for kaon and antikaon production , e.g. , the threshold energy in the pion - baryon collisions @xmath131 . shown in fig .
3 is a comparison of the kaon energy in units of its free mass and the optical potential at the momentum of @xmath132=0 as a function of baryon density . in the neutron - rich nuclear matter
, an isospin splitting for the pairs ( k@xmath123 , k@xmath121 ) and ( @xmath122 , k@xmath124 ) is pronounced with the baryon density .
the equations of motion include the spatial component of the vector potential , which would lead to an attractive lorentz force between kaons and nucleons as @xcite @xmath133 where @xmath134 is the velocity of kaons ( anti - kaons ) , and @xmath135 for k and @xmath3 , respectively .
the lorentz force increases with the kaon momentum .
however this contradicts the predictions from the theoretical studies which show an opposite momentum dependence @xcite .
the isospin effect was considered in the vector potential @xmath136 @xcite .
the kn potential and the lorentz force influence the kaon dynamics in heavy - ion collisions and in proton induced reactions , which lead to a flat structure of direct flows and well reproduce the experimental data @xcite .
the scattering in two - particle collisions is performed by using a monte carlo procedure , in which the probability to be a channel in a collision is calculated by its contribution of the channel cross section to the total cross section .
the primary products in nucleon - nucleon ( nn ) collisions are the resonances of @xmath11(1232 ) , @xmath137(1440 ) , and @xmath137(1535 ) .
we have included the reaction channels as follows : @xmath138 here hadron - hadron collisions take place as two - body process and three - body ( @xmath139state pion production ) reactions .
at the considered energies , there are mostly @xmath11 resonances which disintegrate into a @xmath0 and a nucleon in the evolutions .
the momentum - dependent decay widths are used for the resonances of @xmath11(1232 ) and @xmath137(1440 ) @xcite .
we have taken a constant width of @xmath140=150 mev for the @xmath137(1535 ) decay .
elastic scattering of nn , nucleon - resonance ( @xmath141 ) and resonance - resonance ( @xmath142 ) collisions and inelastic collisions of nucleon - resonance ( @xmath143 , @xmath144 ) and resonance - resonance ( @xmath145 , @xmath146 , @xmath147 and @xmath148 being different resonances ) , have been included in the model .
the elastic cross sections of the available experimental data are parameterized in the energy range of 1 mev - 2 tev for nn collisions @xcite , in which the @xmath149 ( proton - neutron ) cross sections are about 3 times larger than the @xmath150 ( proton - proton)/@xmath151 ( neutron - neutron ) cases at the incident energies from 1 mev to 400 mev .
the in - medium elastic cross section is scaled according to the effective mass through @xmath152 with the @xmath153 and @xmath154 being the reduced masses of colliding nucleon pairs in the medium and in the free space , respectively @xcite .
the parameterized cross sections calculated by the one - boson exchange model @xcite for the @xmath11(1232 ) and @xmath137(1440 ) production are used in the model .
the cross sections for @xmath137(1535 ) are estimated from the empirical @xmath1 production , such as @xmath155 ( a ) @xmath156 , ( b ) @xmath157 and ( c ) @xmath158 in mb and @xmath159 with @xmath160 being the invariant energy in gev and @xmath161 gev . the case ( b ) is used in this work .
the @xmath162 cross sections are about 3 times larger than that for @xmath151 .
half probabilities of the resonances @xmath163 decay into the @xmath1 production .
different stiffness of symmetry energy leads to difference of @xmath164 and @xmath162 collision probabilities in producing @xmath163 .
therefore , the production of @xmath1 could be probe of the high - density symmetry energy .
the strangeness and vector mesons ( @xmath14 , @xmath15 ) are created in inelastic hadron - hadron collisions without intermediate resonances .
we included the channels as follows : @xmath165 here the b stands for ( n , @xmath166 , n@xmath167 ) and y(@xmath4 , @xmath5 ) , k(k@xmath123 , k@xmath121 ) and @xmath3(@xmath122 , k@xmath124 ) .
the parameterized cross sections of each isospin channel @xmath168 @xcite are used in the calculation .
we take the parametrizations of the channels @xmath169 @xcite besides the @xmath170 reaction @xcite .
the results are close to the experimental data at near threshold energies .
the cross section of antikaon production in inelastic hadron - hadron collisions is taken as the same form of the parametrization used in the hadron string dynamics ( hsd ) calculations @xcite .
furthermore , the elastic scattering and strangeness - exchange reaction between strangeness and baryons have been considered through the channels of @xmath171 , @xmath172 and @xmath173 and we use the parametrizations in ref .
the charge - exchange reactions between the @xmath174 and @xmath175 channels are included by using the same cross sections with the elastic scattering , such as @xmath176 , @xmath177 etc @xcite .
the cross sections for @xmath14 and @xmath15 production are taken from the fitting of experimental data @xcite .
superdense hadronic matter can be formed in high - energy heavy - ion collisions and exists in the compact stars , such as neutron stars .
the hadron - hadron interaction in the superdense matter is complicated and varying with the baryon density . the hadron - nucleon potential impacts the ingredients of hadrons in the compact stars . in nuclear reactions ,
i.e. , heavy - ion collisions , antiproton ( proton ) induced reactions etc , the production and phase - space distribution of particles were modified in nuclear medium @xcite .
the isospin dependence of the potential and the corrections on threshold energies influence the ratios of isospin particles .
consequently , the extraction of high - density symmetry energy is to be moved from particle production . on the other hand ,
the yields of particles and bound fragments such as hypernuclides , kaonic nucleus , antiprotonic nucleus is related to the potential of particles in nuclear medium .
pion as the lightest meson could be easily produced in heavy - ion collisions and attracts much attention both in theories and in experiments .
besides as probing the high - density symmetry energy , the pion - nucleon potential in dense matter is not well understood up to now . in this work
, we thoroughly investigated the pion dynamics in heavy - ion collisions , in particular in the domain of near threshold energies ( e@xmath178=280 mev for @xmath75 ) .
impacts of the isospin and momentum dependent pion potential and the stiffness of symmetry energy on the pion production are stressed .
shown in fig .
4 is a comparison of total pion and ratio of charged pion produced in central @xmath179au+@xmath179au collisions with the experimental data from fopi collaboration @xcite .
it is obvious that the total number of pions is reduced with inclusion of the pion - nucleon potential .
however , the @xmath7 ratio is enhanced , in particular with the phenomenological approach for the isoscalar part of the pion energy in medium . the decrease of the pion yields is caused from the attractive interaction between pions and nucleons , which enhances the absorption collisions of pions and nucleons via the channels @xmath180 and @xmath181 .
the effect increases with the pion momentum and baryon density .
the isospin related pion - nucleon potential , i.e. , more attractive interaction for the @xmath76n potential , leads to the increase of the @xmath7 ratio .
precise structure of the charged pion ratio can be observed from the transverse momentum distribution as shown in fig .
5 . similar trends for free pion transportation ( only including coulomb interaction ) and the @xmath11-hole model are found .
a flat structure is concluded with the phenomenological approach .
however , the effect of symmetry energy from the transverse momentum spectra is negligible for the different @xmath182 potentials . to eliminate the coulomb interaction of charged particles ,
neutral particles produced in heavy - ion collisions could be nice probes in extracting the in - medium potential . usually , the neutral particles are reconstructed via the decays in experimentally , e.g. , @xmath183 . shown in fig .
6 is the @xmath75 production in @xmath179au+@xmath179au collisions at the incident energy of 300 mev / nucleon with different @xmath0n potentials , in which only the isoscalar part of the @xmath0n interaction contributes the pion dynamics .
it is obvious that the number of @xmath75 is reduced from the rapidity and transverse momentum distributions with the @xmath0n potentials .
the impact of the @xmath1n potential on @xmath1 dynamics is shown in fig .
one notices that the production cross sections are reduced in the domain of mid - rapidities and high momenta because of attractive interaction between etas and nucleons in nuclear medium .
the attractive optical potential enhances the absorption reactions of particles in nuclear medium , in particular at high baryon densities , i.e. , via the channels @xmath180 , @xmath184 . shown in fig .
8 is the phase - space distribution of @xmath1 production in central @xmath179au+@xmath179au collisions at incident energy of 600 mev / nucleon with different stiffness of symmetry energy .
the effect of the @xmath1n potential is similar to the @xmath185ca+@xmath185ca reaction shown in fig .
however , the difference of hard and soft symmetry energies at high - baryon densities is negligible in the spectra . it is concluded that constraining the symmetry energy from the transverse momentum or kinetic energy spectrum is not possible , but more sensitive to the @xmath1n potential . a systematic comparison of the in - medium corrections on neutral particle distributions in phase space is shown in fig .
9 . the mean - field potentials of particles in nuclear medium contribute the dynamical evolutions .
consequently , the kn potential reduces the kaon production at midrapidities and at high transverse momenta .
however , an opposite contribution of the @xmath1n potential is obtained because of the attractive interaction of eta and nucleon in nuclear medium .
the hyperon - nucleon interaction is negligible for @xmath4 dynamics . shown in fig .
10 is the inclusive spectra of @xmath75 , @xmath1 , @xmath186 , @xmath122 and neutral hyperons ( @xmath4+@xmath187 ) in @xmath179au+@xmath179au collisions .
similar to the transverse momentum spectra , the eta - nucleon , kaon - nucleon and hyperon - nucleon potentials weakly impact the particle emission .
however , the interaction of pions ( antikaons ) and nucleons contributes the pion ( antikaon ) dynamics , i.e. , increasing the antikaon production at low kinetic energies .
the in - medium effects of strange particles have been investigated from the k@xmath124/k@xmath121 spectrum in heavy - ion collisions and in proton induced reactions both in experiments @xcite and in theories @xcite .
a deeply attractive @xmath188 potential being the value of -110@xmath13515 mev was obtained at saturation density and weakly repulsive @xmath189 potential has been concluded to be 25@xmath13510 mev from heavy - ion collisions . shown in fig .
11 is a comparison of the k@xmath124/k@xmath121 ratio with and without the kaon(antikaon)-nucleon potentials in the @xmath190ni+@xmath190ni reaction as a function of transverse mass ( @xmath191 with @xmath192 being the transverse momentum and the mass of kaon ( antikaon ) @xmath193 ) . the reduction of the threshold energies increases the production cross sections of antikaons .
furthermore , the @xmath194 potential enhances the low - momentum ( kinetic energy ) @xmath195 production .
the @xmath196 ratio could be sensitive observable to extract the in - medium potentials .
more information of the in - medium effects is also investigated from the total multiplicity of particles produced in central @xmath185ca+@xmath185ca collisions as shown in fig .
it has been shown that the mean - field potentials of particles in nuclear medium have significant contributions on the total number of particle production .
specifically , the @xmath0n potential increases the collision probabilities of pions and nucleons , which leads to a reduction of pion numbers , in particular near threshold energies .
moreover , the @xmath1 and strange particles ( @xmath186 , @xmath122 , @xmath4 , @xmath187 ) are enhanced owing to the @xmath0 induced reactions , such as @xmath197 , @xmath198 and @xmath199 . the available data from fopi collaboration for pions @xcite and from taps collaboration for etas @xcite can be well reproduced .
it should be mentioned that the inclusion of the kn potential in the model leads to about 30@xmath200 reduction of the total kaon yields in the subthreshold domain @xcite .
the competition of the @xmath201n and kn potentials results in a bit of increase of kaon production because of the contribution of @xmath202 .
the studies of the transverse flows of nucleons , light fragments , pions and strange particles in heavy - ion collisions have been motivated a lot of issues , such as the symmetry energy , in - medium nn cross section , optical potentials of particles in nuclear matter etc @xcite . to investigate neutral particles production in the reaction plane and impacts of the optical potentials on particle dynamics
, we computed the rapidity distributions of transverse flows for neutral particles ( @xmath75 , @xmath1 , @xmath186 and @xmath4 ) produced in the peripheral @xmath190ni+@xmath190ni collisions ( b=7 fm ) at the incident energy of 1.93__a _ _ gev as shown in fig . 13 .
the attractive @xmath1n potential enlarges the transverse emission of @xmath1 in comparison to the in - vacuum case in heavy - ion collisions .
similar effect is also found for the @xmath4 production because of weakly attractive interaction between hyperons and nucleons below the baryon densities of 2.5@xmath36 .
almost isotropic emission and a clock rotation for @xmath186 production take place with the kn potential .
the @xmath0n potential enhances the absorption of @xmath75 by surrounding nucleons and even appears the evidence of antiflow in comparison with the proton flow .
the in - medium and isospin effects of pseudoscalar mesons and hyperons produced in heavy - ion collisions near threshold energies have been investigated within an isospin and momentum dependent hadron - transport model ( lqmd ) .
the in - medium potentials and corrections on threshold energies on particle production are of importance on particle transportation and distribution at freeze - out in phase space .
the isospin related pion - nucleon potential reduces the total pion production , but enhances the @xmath7 ratio , in particular in the domain of subthreshold energies .
the transverse flows , invariant spectra , rapidity and transverse momentum distributions in heavy - ion collisions are promising observables in extracting the in - medium properties of pions in dense nuclear matter .
experiments in the near future such as hirfl - csr ( lanzhou ) , riken - samurai in japan etc , are expected for extracting the high - density symmetry energy , pion - nucleon potential , in - medium properties of @xmath10 .
the dynamics of etas , kaons , antikaons and hyperons in heavy - ion collisions is influenced by the in - medium potentials , in particular at the high - momentum tails .
the transverse flow structure , invariant spectra , transverse mass spectra of @xmath196 , rapidity and transverse momentum distributions are related to the mean - field potentials , in particular in the domain of subthreshold energies .
the stiffness of symmetry energy weakly impacts the eta distribution in phase space .
we would like to thank lie - wen chen , maria colonna , massimo di toro , alexei larionov , bao - an li , che ming ko , ulrich mosel , hermann wolter for fruitful discussions .
this work was supported by the major state basic research development program in china ( no .
2014cb845405 and no .
2015cb856903 ) , the national natural science foundation of china projects ( nos 11175218 and u1332207 ) , the youth innovation promotion association of chinese academy of sciences , and yuncheng university research project ( no .
yq-2014014 ) . b. e. gibson and e. v. hungerford iii , phys . rep . *
257 * , 349 ( 1995 ) .
e. friedman and a. gal , phys .
rep . * 452 * , 89 ( 2007 ) .
m. di toro _ et al .
c * 83 * , 014911 ( 2011 ) .
g. q. li , c. m. ko , and b. a. li , phys .
rev . lett . * 74 * , 235 ( 1995 ) ; g. q. li and c. m. ko , nucl .
a * 594 * , 460 ( 1995 ) .
g. q. li , c. h. lee , and g. e. brown , nucl .
a * 625 * , 372 ( 1997 ) ; g. q. li and g. e. brown * 636 * , 487 ( 1998 ) . w. cassing and e. l. bratkovskaya , phys . rep .
* 308 * , 65 ( 1999 ) . c. fuchs , prog .
nucl . phys .
* 56 * , 1 ( 2006 ) . c. hartnack , h. oeschler , y. leifels , e. bratkovskaya , and j. aichelin , phys
. rep . * 510 * , 119 ( 2012 ) .
b. a. li , phys .
lett . * 85 * , 4221 ( 2000 ) ; z. q. feng , phys .
b * 707 * , 83 ( 2012 ) .
j. aichelin and c. m. ko , phys .
lett . * 55 * , 2661 ( 1985 ) . c. sturm _ et al . _
( kaos collaboration ) , phys .
lett . * 86 * , 39 ( 2001 ) .
g. q. li and c. m. ko , phys .
b * 349 * , 405 ( 1995 ) . c. fuchs , a. faessler , e. zabrodin , and y. m. zheng , phys .
* 86 * , 1974 ( 2001 ) . c. hartnack , h oeschler , and j. aichelin , phys .
lett . * 96 * , 012302 ( 2006 ) .
z. q. feng , phys .
c * 83 * , 067604 ( 2011 ) ; z. q. feng , nucl .
phys . rev . *
31 * , 326 ( 2014 ) .
b. a. li , phys .
lett . * 88 * , 192701 ( 2002 ) ; b. a. li , l. w. chen , and c. m. ko , phys . rep . * 464 * , 113 ( 2008 ) .
q. li , z. li , e. zhao , and r.k .
gupta , phys .
c * 71 * , 054907 ( 2005 ) .
g. ferini , t. gaitanos , m. colonna , m. ditoro , and h. h. wolter , phys . rev
. lett . * 97 * , 202301 ( 2006 ) . m. di toro , v baran , m colonna , and v greco , j. phys .
g : nucl . part .
* 37 * , 083101 ( 2010 ) .
v. prassa , t. gaitanos , g. ferini _
et al . _ ,
a * 832 * , 88 ( 2010 ) . z.
q. feng and g.m .
jin , phys .
c * 82 * , 044615 ( 2010 ) ; z. q. feng , phys .
c * 87 * , 064605 ( 2013 ) .
g. c. yong , b. a. li , and l. w. chen , phys .
b * 661 * , 82 ( 2008 ) ; g. c. yong and b. a. li , phys .
b * 723 * , 388 ( 2013 )
. v. giordano , m. colonna , m. di toro , v. greco , j. rizzo , phys .
c * 81 * , 044611 ( 2010 ) . z. q. feng , phys .
c * 84 * , 024610 ( 2011 ) ; nucl .
phys . a * 878 * , 3 ( 2012 ) ; nucl .
tech . , * 24 * , 050504 ( 2013 ) .
z. g. xiao , b. a. li , l. w. chen , g. c. yong , and m. zhang , phys .
* 102 * , 062502 ( 2009 ) . z.
q. feng and g. m. jin , phys .
b * 683 * , 140 ( 2010 ) .
w. j. xie , j. su , l. zhu , and f. s. zhang , phys .
b * 718 * ,
1510 ( 2013 ) . w. reisdorf _
( fopi collaboration ) , nucl .
a * 781 * , 459 ( 2007 ) .
g. ferini , m. colonna , t. gaitanos , and m. di toro , nucl .
a * 762 * , 147 ( 2005 ) .
t. song and c. m. ko , phys .
c * 91 * , 014901 ( 2015 ) .
l. xiong , c. m. ko , and v. koch , phys .
c * 47 * , 788 ( 1993 ) . c. fuchs , l. sehn , e. lehmann , j. zipprich , and a. faessler , phys .
c * 55 * , 411 ( 1997 ) .
j. hong and p. danielewicz , phys .
c * 90 * , 024605 ( 2014 ) .
z. q. feng and g. m. jin , chin .
* 26 * , 062501 ( 2009 ) ; phys .
c * 82 * , 057901 ( 2010 ) . z. q. feng , h. lenske , phys .
c * 89 * , 044617 ( 2014 ) ; z. q. feng , nucl .
sci . tech .
, * 26 * , s20512 ( 2015 ) . o. hen , b. a. li , w. j. guo , l. b. weinstein , and e. piasetzky , phys
c * 91 * , 025803 ( 2015 ) .
a. b. larionov , i. a. pshenichnov , i. n. mishustin , and w. greiner , phys .
c * 80 * , 021601 ( 2009 ) .
j. cnt , m. lacombe , b. loiseau , b. moussallam , and r. vinh mau , phys .
lett . * 48 * , 1319 ( 1982 ) . c. gale and j. kapusta , phys .
c * 35 * , 2107 ( 1987 ) .
g. e. brown and w. weise , phys
* 22 * , 279 ( 1975 ) ; b. friemann , v. p. pandharipande , and q. n. usmani , nucl .
a * 372 * , 483 ( 1981 ) .
j. xu , c. m. ko , and y. oh , phys .
c * 81 * , 024910 ( 2010 ) ; j. xu , l. w. chen , c. m. ko , b. a. li , and y. g. ma , phys .
c * 87 * , 067601 ( 2013 ) . p. z. ning , _ strangeness nuclear physics ( in chinese ) _
( science press , 2008 ) ; x. h. zhong , g. x. peng , lei li , and p. z. ning , phys .
c * 73 * , 015205 ( 2006 ) . h. c. chiang , e. oset , and l. c. liu , phys . rev .
c * 44 * , 738 ( 1991 ) ; t. waas and w. weise , nucl .
a * 625 * , 287 ( 1997 ) ; k. tsushima , d. h. lu , a. w. thomas , and k. saito , phys .
b * 443 * , 26 ( 1998 ) ; t. inoue and e. oset , nucl .
a * 710 * , 354 ( 2002 ) .
l. c. liu and q. haider , phys .
c * 34 * , 1845 ( 1986 ) .
d. b. kaplan and a. e. nelson , phys .
b * 175 * , 57 ( 1986 ) .
j. schaffner - bielich , i.n .
mishustin , and j. bondorf , nucl .
phys . a * 625 * , 325 ( 1997 ) . z. q. feng , nucl .
a * 919 * , 32 ( 2013 ) ; z. q. feng , w. j. xie , and g. m. jin , phys .
c * 90 * , 064604 ( 2014 ) .
y. m. zheng , c. fuchs , a. faessler _ et al .
c * 69 * , 034907 ( 2004 ) .
e. shuryak and v. thorsson , nucl .
a * 536 * , 739 ( 1992 ) ; a. sibirtsev and w. cassing , nucl .
a * 641 * , 476 ( 1998 ) .
s. huber and j. aichelin , nucl .
a * 573 * , 587 ( 1994 ) . z.
q. feng , phys .
c * 85 * , 014604 ( 2012 ) . k. tsushima , a. sibirtsev , a. w. thomas , and g. q. li , phys .
c * 59 * , 369 ( 1999 ) .
k. tsushima , s.w .
huang , and a. faessler , phys .
b * 337 * , 245 ( 1994 ) ; j. phys .
g * 21 * , 33 ( 1995 ) .
j. cugnon and r.m .
lombard , nucl .
a * 422 * , 635 ( 1984 ) .
w. cassing _
et al . _ ,
a * 614 * , 415 ( 1997 ) .
j. cugnon , p. deneye , and j. vandermeulen , phys .
c * 41 * , 1701 ( 1990 ) .
b. a. li , a. t. sustich , b. zhang , and c. m. ko , int . j. mod
e * 10 * , 1 ( 2001 )
. f. laue _
et al . _ ,
82 * , 1640 ( 1999 ) .
a. frster _
et al . _ ( kaos collaboration ) , phys .
lett . * 91 * , 152301 ( 2003 ) .
w. scheinast _ et al .
lett . * 96 * , 072301 ( 2006 ) .
r. averbeck , r. holzmann , v. metag , and r. s. simon , phys .
c * 67 * , 024903 ( 2003 ) . | dynamics of pseudoscalar mesons ( @xmath0 , @xmath1 , @xmath2 and @xmath3 ) and hyperons ( @xmath4 and @xmath5 ) produced in heavy - ion collisions near threshold energies has been investigated within the lanzhou quantum molecular dynamics ( lqmd ) transport model .
the in - medium modifications on particle production in dense nuclear matter are implemented in the model through corrections on the elementary cross sections and by inclusion of the meson ( hyperon)-nucleon potentials , in which the isospin effects are considered .
it is found that the transportation of particles are influenced with the in - medium corrections .
the total number of pions is reduced with an isospin , density and momentum dependent pion - nucleon potential . however , the ratios of charged pions is enhanced with inclusion of the potential .
the production of eta in the domain of mid - rapidities and high momenta is sensitive to the @xmath1-nucleon potential , but weakly depends on symmetry energy .
the attractive antikaon - nucleon potential enhances the subthreshold @xmath3 production and also influences the structure of phase - space distributions .
dynamics of etas , kaons , antikaons and hyperons is also influenced by the pion potential because of collisions between pions and nucleons ( resonances ) .
the impacts of mean - field potentials on particle dynamics are investigated , such as the phase - space distributions from rapidity and transverse momentum spectra , inclusive invariant spectra , collective flows etc .
pacs number(s ) : : 21.65.ef , 21.65.jk , 24.10.jv |
let @xmath2 and @xmath3 be two finite borel measures having compact and infinite supports @xmath6 in the complex plane @xmath7 , with @xmath8 then there exists a measure @xmath9 such that @xmath10 and we denote the support of @xmath9 by @xmath11 .
we shall regard @xmath2 as a perturbation of @xmath3 and investigate when such a perturbation is small " in the sense of definition [ psp ] below .
the three measures yield three lebesgue spaces @xmath12 with respective inner products @xmath13 and norms @xmath14 let @xmath1 denote the sequence of orthonormal polynomials associated with @xmath15 ; that is , the unique sequence of the form @xmath16 satisfying @xmath17 the corresponding monic polynomials @xmath18 , can be equivalently defined by the extremal property @xmath19 a related extremal problem leads to the sequence @xmath20 of the so - called _ christoffel functions _ associated with the measure @xmath15 .
these are defined , for any @xmath21 , by @xmath22 where @xmath23 stands for the space of complex polynomials of degree up to @xmath24 . using the cauchy - schwarz inequality
it is easy to verify ( see , e.g. , ( * ? ? ?
* section 3 ) ) that @xmath25 clearly , @xmath26 is the reciprocal of the diagonal of the kernel polynomial @xmath27 since @xmath28 , the following inequality is an immediate consequence of ( [ eq : chrfun - def ] ) @xmath29 [ psp ] with the @xmath15 s as in , we say that @xmath2 is a * polynomially small ( ps ) perturbation * of @xmath3 provided that @xmath9 is not the zero measure and @xmath30 the next result emphasizes the fact that @xmath2 being a ps perturbation of @xmath3 implies strong constraints on the relative position of the support of @xmath9 . its proof will be given in section 4 .
we denote by @xmath31 the unbounded component of @xmath32 and by @xmath33 the polynomial convex hull of @xmath34 , i.e. @xmath35 .
we use @xmath36 to denote the _ ( logarithmic ) capacity _ of a compact set @xmath37 .
[ prop : co ] if @xmath2 is a ps perturbation of @xmath3 , then @xmath38 ; hence , @xmath39 and @xmath40 .
below we provide a list of @xmath41 , where @xmath2 is a ps perturbation of @xmath3 . throughout the paper we use @xmath42 to denote the area measure on a bounded set @xmath37 and @xmath43 to denote the arclength measure on a rectifiable curve @xmath44 .
* let @xmath45 be a bounded jordan domain ( or the union of finitely many bounded jordan domains with pairwise disjoint closures ) and let @xmath46 be a compact subset of @xmath45 .
take @xmath47 , @xmath48 and @xmath41 , where @xmath49 is integrable on @xmath46
. then lemma 2.2 of @xcite implies the ps property .
* let @xmath44 be a closed piecewise analytic jordan curve without cusps and let @xmath46 be a compact subset in the interior of @xmath44 .
take @xmath50 , @xmath48 , where @xmath49 is integrable on @xmath46 .
then , @xmath41 is a ps perturbation of @xmath3 .
( see theorem 2.1 of @xcite . ) * here we assume @xmath3 is in the szeg class on the unit circle ; i.e. , the absolutely continuous part @xmath51 of @xmath3 with respect to arclength on the unit circle @xmath52 satisfies the condition @xmath53 , and we let @xmath9 be a finite measure supported on a compact set inside the unit circle . then @xmath41 is a ps perturbation of @xmath3 thanks to corollary 2.4.10 of @xcite .
( we remark that @xcite contains a related result for the case of purely absolutely continuous measures . ) * let @xmath44 be a piecewise analytic jordan curve without cusps , let @xmath45 denote its interior and let @xmath54
. if the exterior angle at @xmath55 is less than @xmath56 , then @xmath57 ; see ( * ? ? ?
* theorem 1.3 ) , and therefore @xmath58 , @xmath59 , where @xmath60 is the dirac measure at @xmath61 , is a ps perturbation of @xmath3 . on the other hand , if the exterior angle at @xmath61 is @xmath62 , then @xmath63 , for some positive constant @xmath64 and infinitely many @xmath24 ; see @xcite , and thus @xmath2 is _ not _ a ps perturbation of @xmath3 , in all the above examples that establish that @xmath2 is a ps perturbation of @xmath3 , the polynomials @xmath65 go to zero uniformly on the support of @xmath9 .
the following proposition provides a class of examples of ps perturbations where this uniform convergence ( and in fact pointwise convergence ) to zero does not hold uniformly in @xmath34 .
its proof will be given in section [ proofs ] .
we say that a jordan curve @xmath44 is @xmath66-smooth if @xmath44 has an arclength parametrization @xmath67 that is @xmath68-times differentiable and its @xmath68-th derivative belongs to the class lip @xmath69 , @xmath70 .
[ prop1 ] let @xmath45 be a bounded jordan domain with boundary @xmath44 in the class @xmath71 , @xmath72 .
if @xmath50 is the arclength measure on @xmath44 and @xmath73 is the area measure on @xmath45 , then @xmath41 is a ps perturbation of @xmath3 .
we remark that under the assumption of the proposition , the inequality @xmath74 , holds locally uniformly inside @xmath44 ; see ( * ? ? ?
* theorem 2.4 ) .
we also note that for the special case of analytic curves @xmath44 , proposition [ prop1 ] follows from ( * ? ? ?
* corollary 2.5 ) .
the paper is organised follows : in section 2 we state the main results along with some essential lemmas .
section 3 is devoted to the behaviour of the zeros of certain ps perturbations of bergman polynomials on the union of jordan regions . in section 4
we provide the proofs of our main results .
the main purpose of the paper is to show that the following two associated pairs of sequences @xmath75 have comparable asymptotics when the measure @xmath2 is a polynomially small perturbation of @xmath3 .
furthermore , under a somewhat stronger condition we show that the two christoffel sequences @xmath76 likewise have comparable asymptotics .
the proofs of our results are given in section 4 .
we use @xmath77 to denote the convex hull of a set @xmath37 and note that @xmath78 ; see proposition [ prop : co ] .
[ thm : main - gen ] , was presented by the second author at the cmft 2013 conference , held in shantou , china , in june 2013 . ]
if the measure @xmath2 is a ps perturbation of the measure @xmath3 , then the following hold : * @xmath79 * @xmath80 * uniformly on compact subsets of @xmath81 ( and for @xmath82 , provided @xmath83 does not vanish in @xmath84 for large @xmath85 ) , @xmath86 furthermore , if @xmath87 then @xmath88 uniformly on compact subsets of @xmath31 .
the appearance of @xmath89 in the denominator of assertion ( iii ) causes no difficulties , since by a classical result of fejr _ all the zeros of the orthonormal polynomials stay within the convex hull of the support of the measure of orthogonality _ see , e.g. , @xcite . theorem [ thm : main - gen ] was motivated by the following example .
[ ex1 ] consider the case @xmath90 , where @xmath91 denotes the canonical pentagon with corners at the five roots of unity and @xmath92 denotes the disk with center at @xmath93 and radius @xmath94 and let @xmath46 be the closed disk inside the pentagon with center at @xmath95 and radius @xmath96 .
then , take @xmath97 and @xmath98 , so that @xmath99 is a ps perturbation of @xmath3 , by an application of lemma 2.2 in @xcite .
in figure [ fig1 ] we plot the zeros of @xmath89 and of @xmath65 , for @xmath100 and @xmath101 .
the close resemblance of the zeros in the two plots suggested that the behaviour of the associated orthonormal polynomial sequence outside @xmath45 should be essentially the same , which is indeed what theorem [ thm : main - gen ] asserts . on the other hand , we wish to emphasize that , in general , the zeros of orthogonal polynomials for ps perturbations of a measure @xmath3 need not have the same limit behaviour as the zeros of the orthogonal polynomials generated by @xmath3 ; see example [ ex3 ] and figure [ fig4 ] below
. however , the balayage of their limit measures must be the same ; see theorem [ th : st4.7 ] .
the constructions of the orthonormal polynomials utilized for the plots in this paper were performed by applying the arnoldi variant of the gram - schmidt method ; see e.g. ( * ? ? ?
* section 7.4 ) .
all the computations were carried out on a macbook pro using maple .
essential to the proof of theorem [ thm : main - gen ] is the following simple lemma , which is of independent interest .
[ lem : main1 ] let @xmath2 and @xmath3 be two measures having compact and infinite support in @xmath7 , with @xmath28 .
then for all @xmath102 , @xmath103 where @xmath104 is non - negative and satisfies @xmath105 furthermore , @xmath106 and @xmath107 finally , for any @xmath108 , @xmath109
^ 2.\ ] ] if @xmath110 , then @xmath3 belongs to the important class of measures * reg * investigated in @xcite . furthermore
, if in a neighbourhood of infinity @xmath111 for some analytic function @xmath112 , then we say @xmath3 belongs to the class of measures * ratio*(@xmath113 ) .
the following result shows that both classes of measures are invariant under ps perturbations .
[ cor : inva ] let @xmath3 be in the class * reg * , respectively * ratio*(@xmath113 ) .
if @xmath2 is a ps perturbation of @xmath3 , then @xmath2 is in the class * reg * , respectively * ratio(@xmath113)*. we refer to @xcite , @xcite , @xcite , @xcite , @xcite and @xcite , for a list of recent applications of measures in the class * ratio*(@xmath113 ) , where proposition [ cor : inva ] is expected to have an impact ; namely results concerning measures in the class * ratio*(@xmath113 ) should be easily extended to hold for ps perturbations of these measures . as an illustration
, we consider the infinite upper hessenberg matrix @xmath114 associated with the orthonormal sequence @xmath115 ; that is , @xmath116_{k , j=1}^\infty$ ] , where @xmath117 ; see , e.g. , @xcite and @xcite .
then , the following result is an easy consequence of theorem [ thm : main - gen ] , proposition [ cor : inva ] and ( * ? ? ?
* corollary 1.4 ) .
[ cor : sima ] assume that @xmath118 * ratio(@xmath113 ) * satisfies @xmath119 if @xmath2 is a ps perturbation of @xmath3 , then the matrix @xmath114 is weakly asymptotically toeplitz ; i.e. for all @xmath120 , @xmath121 exists . in the following proposition
we give an example where sharp rates of convergence are obtained for the assertions ( i ) , ( ii ) and ( iii ) of theorem [ thm : main - gen ] .
[ thm : disk ] let @xmath122 be a compact subset of the unit disk @xmath123 and let @xmath124 . if @xmath125 and @xmath126 , then the following hold : * @xmath127 and the order @xmath128 is sharp ; * @xmath129 * @xmath130 * @xmath131 in the above , @xmath132 depends on @xmath133 only .
next we state a proposition that is needed in establishing the last assertion in theorem [ thm : main - gen ] .
it is a generalisation of assertion ( 2.10 ) in @xcite but we provide a shorter proof . [ lem : diagkernelout ] if @xmath122 is a compact subset of @xmath31 , then @xmath134 or , equivalently , @xmath135 uniformly for @xmath136 .
our goal in this section is to provide results concerning the asymptotic behaviour of zeros of ps perturbations of bergman polynomials .
( by the term _ bergman polynomials _ we mean polynomials orthonormal with respect to the area measure on bounded regions . )
to do this it will be convenient to use the notations introduced in @xcite regarding _ archipelaga with lakes_. more precisely , let @xmath137 be a finite union of bounded jordan domains @xmath138 , @xmath139 , in the complex plane @xmath7 , with pairwise disjoint closures , let @xmath122 be a compact subset of @xmath45 and consider the set @xmath140 obtained from @xmath45 by removing @xmath122 ; i.e. , @xmath141 . set @xmath142 for the respective boundaries and let @xmath143 denote the boundary of @xmath45 . let @xmath31 denote the unbounded component of @xmath144 , so that here @xmath145 ; see figure [ fig : archi - lakes ] . note that @xmath146 .
we refer to @xmath45 as an archipelago and to @xmath140 as an archipelago with lakes .
in this case , @xmath147 and @xmath148 , so that according to lemma 2.2 in @xcite , @xmath2 is a ps perturbation of @xmath3 .
further , from lemma 3.2 in @xcite , @xmath2 belongs to the class of measures @xmath149 and , in view of theorem [ thm : main - gen ] , so does @xmath3 . here
, we will use the notation @xmath150 and @xmath151 , in place of @xmath89 and @xmath65 , for the corresponding bergman polynomials .
also , we will use @xmath152 , @xmath153 , @xmath154 and @xmath155 , in the place of @xmath156 , @xmath157 , @xmath158 , and @xmath159 , respectively .
the main task of this section is to describe the asymptotic behaviour of the zeros of the polynomials @xmath151 under various assumptions on the boundary @xmath160 of each individual island @xmath138 .
the behaviour of the zeros of @xmath150 was investigated in @xcite .
our main tool is the _ normalized counting measure _
@xmath161 for the zeros of a the bergman polynomial @xmath151 ; that is , @xmath162 where @xmath60 is the unit point mass ( dirac delta ) at the point @xmath61 .
note that , since all the zeros of @xmath151 lie in the convex hull of @xmath45 , it follows from helly s selection theorem that the sequence of measures @xmath163 has convergent subsequences with limit measures supported on @xmath164 . with @xmath37 a compact set in the complex plane of positive capacity ,
we denote by @xmath165 the _ equilibrium measure _ ( energy minimizing borel probability measure on @xmath37 ) for the logarithmic potential on @xmath37 ; see e.g. , ( * ? ? ?
* chapter 3 ) and ( * ? ? ?
* section i.1 ) .
as is well - known , the support @xmath166 of @xmath165 lies on the boundary of the unbounded component of @xmath167 .
our first result is a consequence of the fact that @xmath168 is a measure in the class * reg*. its proof is given in section 4 .
[ th : st4.7 ] if @xmath169 is any weak - star limit measure of the sequence @xmath170 in , then @xmath169 is a borel probability measure supported on @xmath171 and @xmath172 , where @xmath173 is the balayage of @xmath169 out of @xmath171 onto @xmath174 .
similarly , the sequence of balayaged counting measures converges to @xmath165 : @xmath175 by the weak - star convergence of a sequence of measures @xmath176 to a measure @xmath177 we mean that , for any continuous @xmath113 with compact support in @xmath178 there holds @xmath179 for properties of balayage see ( * ? ? ? * sect .
ii.4 ) .
the next example illustrates that the zero behaviour of polynomials orthogonal with respect to a ps perturbation of a measure @xmath3 can have an asymptotic limit that is different from that of the zeros of orthogonal polynomials with respect to @xmath3 .
however , as theorem [ th : st4.7 ] asserts , the balayage onto the boundary of these two limiting distributions of zeros must be the same .
[ ex3 ] consider the case @xmath180 and @xmath181 , @xmath182 , @xmath183 .
as will be established in detail in the forthcoming paper @xcite , the distribution of zeros of @xmath184 satisfies @xmath185 where @xmath186 denotes the arclength measure on the circle @xmath187 , with @xmath188 being the unique point of @xmath122 having a common reflection in the circles @xmath52 and @xmath189 . in figure [ fig4 ]
we plot the zeros of @xmath151 , for @xmath190 and @xmath191 , for @xmath192 , @xmath193 and @xmath194 .
below we need the notion of an _ inward - corner _ ( ic ) point .
following @xcite , we say that a point @xmath195 on the boundary @xmath160 of @xmath138 is an ic point if there exists a circular sector of the form @xmath196 with @xmath197 whose closure is contained in @xmath138 except for @xmath195 .
the following theorem is an immediate consequence of corollary 2.2 in @xcite ; cf .
the proof of theorem [ th : st4.7 ] in section 4 .
[ th : ic ] with the notation above , assume that for each @xmath198 the boundary @xmath160 of @xmath138 contains an ic point s is irrelevant here . ] .
then , with @xmath161 as in , @xmath199 where @xmath200 is an open set containing @xmath201 , such that if @xmath202 the distance of @xmath203 from @xmath204 is positive
. the result of the theorem remains valid under the weaker condition that the boundary @xmath138 , for each @xmath198 , contains a _ non - convex type singularity _ ; see @xcite for details .
theorem [ th : ic ] is illustrated by the following example .
[ ex : sector - disk ] consider the case @xmath90 , where @xmath91 is a circular sector of opening angle @xmath205 and radius @xmath206 with center at the origin and @xmath92 is the disk with center at @xmath93 and radius @xmath94 and let @xmath122 be the closed disk inside @xmath91 with center at @xmath95 and radius @xmath96 . in figure [ fig3 ]
we plot the zeros of @xmath151 , with @xmath207 , for @xmath190 and @xmath191 . according to theorem [ th : ic ] ,
@xmath208 , as @xmath5 , through the whole sequence @xmath209 , where @xmath200 is some open neighbourhood of @xmath210 .
this , in particular , implies that every point of the boundary of @xmath91 will attract zeros of @xmath151 , a fact which is corroborated by the position of zeros in the plot .
the obvious reluctance of the zeros to approach the reentrant corner of the sector can be accounted for by the fact that the equilibrium measure at this corner has zero density .
we end this section by noting that there is a recent growing interest for asymptotics of bergman polynomials with weights supported on the unit disk ; see , e.g. @xcite , @xcite and the references therein .
it is evident that much of our results above hold true if the bergman polynomials @xmath151 are replaced by weighted bergman polynomials , defined by the inner product @xmath211 for some reasonable weight function @xmath212 .
see , for example , the book by suetin ( * ? ? ?
* chapter ii ) and the more recent results of totik in ( * ? ? ? * theorems 1.3 & 1.5 ) .
clearly , the polynomials @xmath213 correspond to the characteristic function @xmath214
this section contains the proofs of results in the preceding sections .
we shall show that there are no points of @xmath215 in @xmath216 . then , since @xmath217 we have @xmath39 .
furthermore , strict inequality can not hold for the capacities , since @xmath218 ; thus @xmath219 . to see that there are no points of @xmath215 in @xmath31 first observe that there can not be any isolated points of @xmath215 in @xmath31 , since at any such point , say @xmath195 , we would have from , @xmath220 but by a theorem of ambroladze ( * ? ? ? * theorem 1 ) , if @xmath221 denotes the _ green function _ with pole at infinity associated with @xmath31 , then @xmath222 which yields a contradiction .
( if the complement @xmath223 of @xmath31 has capacity zero , then @xmath224 ; see ( * ? ? ? * appendix v ) . )
next , suppose that @xmath225 and @xmath195 is an accumulation point of @xmath215 .
let @xmath226 be a closed disk centered at @xmath195 . by a result of widom @xcite
; see also ( * ? ? ? *
theorem 2.1.1 ) , there exists an integer @xmath227 such that each @xmath65 has at most @xmath227 zeros in @xmath228 .
let @xmath229 and consider a subsequence @xmath230 such that @xmath231 converges locally uniformly in @xmath7 to a monic polynomial @xmath232 of degree at most @xmath227 .
by corollary 1.1.5 in @xcite @xmath233 uniformly on @xmath234 , where @xmath234 is a closed disk centered at @xmath195 and contained in @xmath235 . then , there exists a constant @xmath236 such that @xmath237 hence , @xmath238 this yields a contradiction , since @xmath9 has infinitely many points of support in @xmath234 where @xmath232 must be zero
. let @xmath239 denote the associated sequence of orthonormal polynomials with respect to the arclength measure on @xmath44 and let @xmath240 denote the conformal map from the exterior of @xmath44 to the exterior of the unit circle , normalized by @xmath241 and @xmath242 .
then , from ( * ? ? ?
* theorem 2.3 ) we have @xmath243,\ ] ] uniformly for @xmath55 .
our assumption on @xmath44 implies that @xmath244 is finite and non - vanishing on @xmath44 and that @xmath245 is bounded on @xmath44 ; see e.g. ( * ? ? ?
* sections 3.33.4 ) .
let @xmath246 be fixed .
then , from integration along @xmath44 we have @xmath247d\zeta\\ & = \int_{z_0}^z \phi^n(\zeta)\frac{\phi^\prime(\zeta)}{\sqrt{\phi^\prime(\zeta)}}d\zeta+o\left(1/n\right ) , \quad z\in\gamma.\end{aligned}\ ] ] integrating by parts gives uniformly for @xmath55 , @xmath248^{3/2}}d\zeta+o\left(1/n\right)=o\left(1/n\right).\end{aligned}\ ] ] now by using greens formula ( see e.g. @xcite ) and the cauchy - schwarz inequality we have @xmath249^{1/2}\left[\int_\gamma |q_{n+1}(z)|^2 ds\right]^{1/2}\end{aligned}\ ] ] and the required result follows from ( [ eq : qp ] ) because of the normality of the polynomials @xmath250 .
we aim first at relating the leading coefficients @xmath156 and @xmath157 . to this end
, we note that @xmath251 which is a simple consequence of the minimal property ( [ eq : minimal1 ] ) .
the fact that @xmath252 is evident from the inequality ( [ eq : comp - pri - gamma ] ) . to prove ( [ eq : lem - main1b ] ) we note first
the two obvious relations @xmath253 and @xmath254 since @xmath3 has infinite support , shows that @xmath255 next we note the following two relations , which hold for all @xmath102 : @xmath259 and @xmath260 these can be readily seen by expanding @xmath261 and @xmath262 , together with the definition of @xmath104 in ( [ eq : lem - main1a ] ) and the four relations in ( [ eq : png=1- ] ) , ( [ eq : png*=1 + ] ) and ( [ eq:<pn*,pn > ] ) .
then , the inequality in is immediate from ( [ eq : lem - main2 - 2 ] ) , because @xmath252 .
let @xmath264 be fixed .
by runge s theorem there exists a sequence of polynomials @xmath265 such that @xmath231 converges uniformly to zero on the boundary of @xmath31 ( and hence on @xmath34 ) and also converges to one on a sufficiently small closed disk @xmath266 centered at @xmath188 .
but then it is immediate from the extremal property of christoffel functions that @xmath267 , uniformly on @xmath268 . since @xmath122 is compact , it can be covered by finitely many such disks , and the assertion of the proposition follows .
next we consider the sequence of numbers @xmath270 which by the assumption in tends to zero as @xmath271 .
in contrast , from proposition [ lem : diagkernelout ] it follows that for any given compact set @xmath272 and positive integers @xmath273 and @xmath274 , @xmath275 for all sufficiently large @xmath24 .
furthermore , from the orthogonal expansion of @xmath287 with respect to @xmath3 , we have @xmath288 by using the cauchy - schwarz inequality for the sum in ( [ eq : pndef ] ) we get that @xmath289 and , therefore , we obtain from ( [ eq : vare_m ] ) that @xmath290 now adding up ( [ eq : pnovergs ] ) and ( [ eq : pnoverb ] ) we conclude , in view of ( [ eq : lamgzle ] ) , that @xmath291 the upper estimate in ( [ eq : ggs-1 ] ) then follows from ( [ eq : chrfun - pro ] ) , ( [ eq : pnoverb ] ) and , because @xmath292.\ ] ] the lower estimate in is simply the comparison property ( [ eq : comp - pri - lambda ] ) .
when @xmath293 , the result is evident from theorem [ thm : main - gen](i ) , in view of proposition [ prop : co ] . for @xmath294
, the result follows at once from theorem [ thm : main - gen](iii ) , because outside @xmath263 , @xmath295 see also the remark following the statement of theorem [ thm : main - gen ] .
set @xmath297 , @xmath298 , @xmath299 and consider @xmath300 .
then it is easy to check that @xmath301 where @xmath302 since @xmath303 , it follows from the minimal property ( [ eq : minimal1 ] ) that @xmath304 hence , ( [ eq : pnd ] ) implies @xmath305 with @xmath306 , which yields ( i ) .
the sharpness claim is evident by letting @xmath307 .
finally , the uniform estimate in part ( iv ) follows from the @xmath309-estimate in ( ii ) by using the inequality @xmath310 which is valid for any polynomial @xmath287 of degree at most @xmath85 , where the constant @xmath64 depends on @xmath45 only ; see @xcite .
as was noted in section 3 , @xmath168 belongs to the class @xmath149 .
thus , from @xcite and ( * ? ? ?
* theorems 3.2.1 & 3.2.3 ) we have @xmath311 because @xmath312 and every point of @xmath44 is regular for the dirichlet problem in @xmath31 ; see , e.g. , @xcite . hence @xmath313 constitutes a sequence of asymptotically extremal monic polynomials on @xmath171 and the result follows from ( * ? ? ?
* theorem 2.3 ) ; see also ( * ? ? ?
* theorem iii.4.7 ) and ( * ? ? ?
* theorem 1.1 ) d. khavinson and n. stylianopoulos , _ recurrence relations for orthogonal polynomials and algebraicity of solutions of the dirichlet problem _ , around the research of vladimir mazya .
ii , int . math .
( n. y. ) , vol .
12 , springer , new york , 2010 , pp .
219228 .
to3em , _ remarks on relative asymptotics for general orthogonal polynomials _ , recent trends in orthogonal polynomials and approximation theory , contemp .
507 , amer .
soc . , providence , ri , 2010 , pp .
233239 .
j. l. walsh , _ interpolation and approximation by rational functions in the complex domain _ , fourth edition .
american mathematical society colloquium publications , vol .
xx , american mathematical society , providence , r.i . , 1965 . | we survey and present some new results that are related to the behavior of orthogonal polynomials in the plane under small perturbations of the measure of orthogonality .
more precisely , we introduce the notion of a polynomially small ( ps ) perturbation of a measure .
namely , if @xmath0 and @xmath1 are the associated orthonormal polynomial sequences , then @xmath2 a ps perturbation of @xmath3 if @xmath4 , as @xmath5 .
in such a case we establish relative asymptotic results for the two sequences of orthonormal polynomials .
we also provide results dealing with the behaviour of the zeros of ps perturbations of area orthogonal ( bergman ) polynomials . |
the search for primeval galaxies has been one of the long term programs of observational cosmology , along with the development of sensitive detectors , and is currently receiving an exciting boost with the new generation of 8metre class telescopes .
however , the search itself is somewhat hampered by the fuzziness of the concept of `` primeval galaxy '' .
this is generally understood as being `` a galaxy which is captured at the epoch of its formation '' . since there are several competing theories about what a forming galaxy should look like , it is readily possible that we are missing part of the process of galaxy formation because we do not know yet what to search for .
more specifically , two general paradigms have been proposed in the last twenty years or so . in the picture of monolithic collapse , galaxies form at a given epoch @xmath0 ,
when the physical conditions of the universe are favourable , and evolve at different rates which are fixed by the initial conditions . in the picture of hierarchical collapse , there is nothing like a given redshift of galaxy formation .
larger galaxies form from the merging of smaller ones , which on their turn have formed from the merging of still smaller lumps , and so on .
the beginning of the process took place at some early redshift @xmath1 ( when the first objects can cool ) and is still going on now .
as a result , the `` epoch of galaxy formation '' can be defined e.g. as the epoch when the first stars formed , when 50 % of the stars have formed , or when the morphology was fixed after the last major merging event .
this hierarchical galaxy formation is now modelled in the context of hierarchical clustering where dark matter completely rules gravitational collapse .
the models of galaxy formation have to reproduce the wide variety of objects which are now observed at high redshift , probably after strong observational biasses that are not fully understood .
these objects are generally selected according to photometric criteria .
we now have luminous blue compact galaxies ( lbcgs ; @xmath2 and @xmath3 ) , lyman break galaxies ( lbgs ; @xmath4 or @xmath5 , and @xmath6 ) , extremely red objects ( eros ; @xmath7 and @xmath8 ) , lyman alpha galaxies ( lags ; @xmath5 or @xmath9 , and @xmath10 ) , faint submm sources that are likely to be the high@xmath11 counterparts of the local `` ultraluminous infrared galaxies '' discovered by iras ( high@xmath11 ulirgs ; @xmath12 and @xmath13 @xmath14 ) , damped lyman @xmath15 absorbers ( dlas ; @xmath16 atom @xmath17 ) , etc . in order to link the predictions of the models of galaxy formation with these observations , it is necessary to model the spectral energy distributions ( seds ) on the widest wavelength range , from the rest frame uv to the rest frame submm .
as a matter of fact , several pieces of observational evidence are now converging to show that there is a significant amount of extinction in high redshift galaxies .
consequently , the absorption and emission processes due to dust can not be neglected in assessing the luminosity budget of forming galaxies . among these pieces of evidence
, we can quote : \(i ) the discovery of the cosmic ir background at a level 10 times higher than the no evolution predictions based on the iras luminosity functions , and twice as high as the cosmic optical background ( puget _ et al .
_ 1996 , guiderdoni _ et al .
_ 1997 , fixsen _ et al .
_ 1998 , hauser _ et al .
_ 1998 , lagache _ et al .
_ 1999 ) .
\(ii ) the ir and submm counts that have broken the cirb into its brightest contributors at 15 @xmath18 m ( isocam down to @xmath19 mjy , aussel _ et al .
_ 1999 , elbaz _ et al .
_ 1999 ) , 175 @xmath18 m ( isophot down to @xmath20 mjy , kawara _ et al .
_ 1998 , puget _
et al . _ 1999 ) , and 850 @xmath18 m ( scuba down to 2 mjy , smail _ et al .
_ 1997 , hughes _ et al .
_ 1998 , barger _
_ 1998 , 1999a , eales _
although the poor spatial resolution of the observations makes the identification of the optical counterparts somewhat difficult , the first results of the spectroscopic follow
ups seem to show that some of the sources are at @xmath12 ( smail _ et al .
_ 1998 , lilly _
_ 1999 , barger _
1999b ) .
\(iii ) at least some of the eros ( e.g. hr10 at @xmath21 , cimatti _ et al . _
1998a ) are high redshift dusty objects .
\(iv ) the extinction in lbcgs at @xmath2 corresponds to a factor 3 extinction on the rest frame flux at 2800 ( flores _ et al .
_ 1999 ) .
\(v ) the extinction in lbgs at @xmath22 and 4 is high ( @xmath23 ) , with a trend of a larger extinction for the brightest objects ( steidel _ et al .
_ 1999 , meurer _ et al .
_ 1999 ) . this corresponds to a factor 5 extinction on the rest frame flux at 1600 .
\(vi ) dust is present in high redshift radiogalaxies and qsos , up to @xmath24 ( see e.g. hughes _ et al .
_ 1997 , cimatti _ et al . _
. the optical spectrum of a gravitationally lensed galaxy at @xmath25 already shows a reddening factor amounting to @xmath26 ( soifer _ et al .
_ 1998 ) .
\(vii ) the correlation of dust with metallicity has still to be understood , since , for instance , the extremely metal poor galaxy sbs0335 - 052 ( 1/30 of solar ) has a significant ir emission ( thuan _ et al .
we hereafter present a model of synthetic seds , called stardust , which contains up to date spectrophotometric modelling coupled to chemical evolution and dust absorption .
the model allows one to make consistent predictions for the seds of galaxies in a very broad wavelength range .
details can be found in devriendt _
( 1999 ; hereafter dgs ) .
the model uses the so called `` isochrone scheme '' and includes a compilation of the geneva tracks ( see e.g. charbonnel _ et al .
_ 1996 , and references therein ) for various metallicities @xmath27 , and masses 0.8 @xmath28 120 . for stars less massive than @xmath29 ,
some of the old tracks stop at the giant branch tip .
more recent grids of models , based on geneva tracks and covering the evolution of low mass stars ( 0.8 to 1.7 @xmath30 ) from the zero age main sequence up to the end of the early agb , are included for @xmath31 0.001 and @xmath31 0.02 . for the late stages of other metallicities ( horizontal branch , early
agb ) , we either interpolate or extrapolate @xmath32 , @xmath33 and @xmath34 versus @xmath35 from the available tracks .
the final stages of the stellar evolution ( thermally pulsing agb and post agb ) are not included .
we use the grid of theoretical fluxes from kurucz ( 1992 ) which covers all metallicities from @xmath36 to @xmath37 , and 61 temperatures from 3500 k to 50 000 k. each spectrum spans a wavelength range between 90 @xmath38 and 160 @xmath18 m , with a mean resolution of 20 @xmath38 . for the coldest stars ( k and m - type stars ) with @xmath39 3750 k ,
kurucz s models fail to reproduce the observed spectra .
therefore , we prefer to use models coming from different sources ( see dgs ) . fig 1 gives the passive evolution of a simple stellar population with salpeter imf ( slope @xmath40 from @xmath41 to @xmath42 @xmath43 ) .
fig 2 compares our stardust model with other models available in the literature ( see dgs for details ) : gissel 1998 ( bruzual and charlot , 1993 ) , pgase ( fioc and rocca volmerange , 1997 ) , and our stardust .
though the stellar data are different , the agreement is generally very good in the uv / visible , except when post agb are dominant ( that is , in old stellar populations without star formation activity ) . in particular , the uv to sfr ratio that is used to derive the cosmic sfr from rest
frame 1600 or 2800 observations is remarkably similar in the three models .
our model computes chemical evolution and takes into account the effect of metallicity on the stellar tracks and stellar spectra in a consistent way .
in addition , the metallicity of the gas is followed . under simple assumptions on the variation of the extinction curve with metallicity ( based on the study of the milky way , the large magellanic cloud , and the small magellanic cloud ) , and on the geometry and relative distributions of dust and stars ( homogeneous mix in an oblate spheroid )
, transfer can be easily solved , and the amount of luminosity absorbed by dust is estimated .
the seds in the ir / submm are then computed to reproduce the correlation of iras colours with total ir luminosity @xmath44 , following guiderdoni _
( 1997 , 1998 ) .
the model reproduces the obscuration curve of calzetti _ et al . _
( 1994 ) observed in a sample of nearby starbursts .
it also reproduces the correlation of ir to 2200 flux with the uv slope @xmath45 of the seds ( around 2200 ) observed by meurer _
( 1995 ) . finally , the optical and ir / submm spectra are connected to give evolving synthetic seds from the far uv ( 90 ) to the submm ( @xmath46 mm ) . in the centimetre and metre range
, a radio component can be added under assumptions on the slope and on the correlation of radio fluxes with ir fluxes .
the seds depend on three parameters ( in addition to the imf ) : the sfr timescale @xmath47 , the age @xmath48 , and a concentration parameter @xmath49 that describes the size of the gaseous disk ( @xmath50 is for normal spirals , and a radial collapse by a factor 10 corresponds to @xmath51 ) .
fig 3 gives typical spectra for a spiral and a forming galaxy , without and with extinction ( respectively with @xmath50 and @xmath51 , this latter value being typical of an ulirg ) .
fig 4 gives the fits of a sample of local galaxies obtained with our theoretical seds and a @xmath52 procedure .
the sample gathers virgo cluster spirals ( boselli _ et al .
_ 1998 ) and ulirgs ( rigopoulou _ et al .
_ 1996 ) that have a sufficient number of photometric points at optical , nir , mir , fir , and submm wavelengths ( and , for some of them , in the radio ) .
nine spectra have been ordered in fig 4 according to their @xmath44 , in a sequence that parallels the compilation of sanders & mirabel ( 1996 ) .
there is some degeneracy between the parameters @xmath47 and @xmath48 of the fits , but their combination corresponds to similar sfrs .
this is illustrative of the ability of the model to capture the characteristic features of the objects .
this can be used to study the efficiency of the uv and ir fluxes to trace the underlying sfr .
as shown in fig 5 , the uv fluxes do scale with sfr for optically thin galaxies ( slope @xmath53 ) , but the proportionality breaks down for ulirgs . without any information on the optical thickness of the high redshift galaxies , the rest
frame uv fluxes could lead to erroneous estimates .
in contrast , the ir fluxes trace sfrs on several orders of magnitudes , with a slope 0.95 , and independently of the optical thickness .
the origin of this behaviour is not clear yet .
it is probably due to the fact that there are basically two regimes : either the galaxy is optically thin , and the ir roughly corresponds to a constant fraction of the uv , or it is optically thick , and all the uv flux is absorbed by dust and released in the ir .
these galaxies can also be used to predict the photometric properties at higher redshift .
fig 6 and 7 give the observer
frame optical magnitudes and ir / submm fluxes versus redshift .
these synthetic seds can subsequently be implemented into semi analytic models of galaxy formation .
the interest of such an approach is that the distribution of the sfr timescales @xmath47 can be computed from the distribution of dynamical timescales @xmath54 of the host haloes , under the assumption @xmath55 , where @xmath45 is an efficiency parameter which is fixed observationally ( see e.g. kennicutt 1998 ) .
the ages @xmath48 of the stellar populations are computed from the formation redshifts and the redshifts at which the galaxies are observed .
the distribution of the gas column densities is also computed from the size distribution of disks ( obtained from the size distribution of haloes after conservation of angular momentum ) . the ulirgs correspond to a further radial collapse by a factor 10 ( @xmath51 ) .
following guiderdoni _
( 1997 , 1998 ) , devriendt & guiderdoni ( 1999 ) proposed results of faint galaxy counts that are produced by a simple , semi
analytic prescription using the peaks formalism .
the usual recipes are implemented for gas cooling , dissipative collapse , and stellar feedback .
it is assumed that dust is heated by starbursts , and that the star formation rate is @xmath56 , with @xmath57 .
the spectra are taken from stardust .
this paper gives predictions of faint counts at optical , ir , and submm wavelengths , for a variety of cosmologies .
it turns out that the ionizing flux escaping from high redshift galaxies estimated with this model is unable to ionize the intergalactic medium ( devriendt _ et al .
these synthetic seds and the spectral templates of fig 4 are useful tools to analyse the panchromatic view on high redshift , young and `` primeval '' galaxies .
the spectral templates are available upon request to the authors .
the implementation of such spectra into semi analytic models is a promising way to predict the statistical properties of a host of objects that are now observed at high redshift , through various spectral windows , and with various selection criteria .
however , the modelling of the sfr history of galaxies in the context of hierarchical galaxy formation is still incomplete . in particular , the ulirgs appear to be key objects to understand the formation of bulges .
a more realistic treatment of such objects requires the monitoring of the merging rates , and involves keeping track of the spatial and dynamical information in models of galaxy formation .
this can be done , for instance , with the new generation of the so called `` hybrid '' models in which the merging history trees are directly built from the outputs of cosmological n body simulations .
the implementation of stardust into such models is in progress .
aussel , h. , cesarsky , c.j . ,
elbaz , d. , starck , j.l . , 1999 , _
a&a _ , 342 , 313 barger , a.j . ,
cowie , l.l . , sanders , d.b . ,
fulton , e. , taniguchi , y. , sato , y. , kawara , k. , okuda , h. , 1998 , _ nature _ , 394 , 248 barger , a.j . ,
cowie , l.l . ,
sanders , d.b .
, 1999a , _ apj _ , 518 , l5 barger , a.j . , cowie , l.l . , smail , i. , ivison , r.j . , blain , w. , kneib , j.p . , 1999b , _ aj _ , 117 , 2656 boselli , a. , lequeux , j. , sauvage , m. , boulade , o. , boulanger , f. , cesarsky , d. , dupraz , c. , madden , s. , viallefond , f. , vigroux , l. , 1998 ,
a&a _ , 335 , 53 bruzual , a.g . ,
charlot , s. , 1993 , _ apj _ , 405 , 538 calzetti , d. , kinney , a.l .
, storchi bergmann , t. , 1994 , _ apj _ , 429 , 582 charbonnel , c. , meynet , g. , maeder , a. , schaerer , d. , 1996 , _ a&as _ , 115 , 339 cimatti , a. , andreani , p. , rottgering , h. , tilanus , r. , 1998 , _ nature _ , 1998a , 392 cimatti , a. , freudling , w. , rottgering , h. , ivison , r.j . ,
mazzei , p. , 1998b , _ a&a _ , 329 , 399 devriendt , j.e.g . ,
sethi , s. , guiderdoni , b. , nath , b. , 1998 , _ mnras _ , 298 , 708 devriendt , j.e.g .
, guiderdoni , b. , sadat , r. , 1999 , _ a&a _ , 350 , 381 devriendt , j.e.g .
, guiderdoni , b. , 1999 , _ submitted _ eales , s. , lilly , s. , gear , w. , dunne , l. , bond , j.r . , hammer , f. , le fvre , o. , crampton , d. , 1999 , _ apj _ , 515 , 518 elbaz d. , aussel h. , cesarsky c.j .
, desert f.x . , fadda d. , franceschini a. , harwit , m. , puget j.l . , starck j.l . , 1999 , in _ the universe as seen by iso _ ,
p. cox & m.f .
kessler ( eds ) , 1998 , unesco , paris , esa special publications series ( sp-427 ) fioc , m. , rocca
volmerange , b. , 1997 , _ a&a _
, 326 , 950 fixsen , d.j . , dwek , e. , mather , j.c . ,
bennett , c.l . , shafer , r.a .
, 1998 , _ apj _ , 508 , 123 flores , h. , hammer , f. , thuan , t.x . ,
cesarsky , c. , dsert , f.x . ,
omont , a. , lilly , s.j . ,
eales , s. , crampton , d. , le fvre . ,
o. , 1999 , _ apj _ , 517 , 148 guiderdoni , b. , bouchet , f.r . , puget , j.l . ,
lagache , g. , hivon , e. , 1997 , _ nature _ , 390 , 257 guiderdoni , b. , hivon , e. , bouchet , f.r . ,
maffei , b. , 1998 , _ mnras _ , 295 , 877 hauser , m.g . , arendt , r. , kelsall , t. , dwek , e. , odegard , n. , welland , j. , freundenreich , h. , reach , w. , silverberg , r. , modeley , s. , pei , y. , lubin , p. , mather , j. , shafer , r. , smoot , g. , weiss , r. , wilkinson , d. , wright , e. , 1998 , _ apj _ , 508 , 25 hughes , d. , dunlop , j.s . ,
rawlings , s. , 1997 , _ mnras _ , 289 , 766 hughes , d. , serjeant , s. , dunlop , j. , rowan robinson , m. , blain , a. , mann , r.g . , ivison , r. , peacock , j. , efstathiou , a. , gear , w. , oliver , s. , lawrence , a. , longair , m. , goldschmidt , p. , jenness , t. , 1998 , _ nature _ , 394 , 241 kawara , k. , sato , y. , matsuhara , h. , taniguchi , y. , okuda , h. , sofue , y. , matsumoto , t. , wakamatsu , k. , karoji , h. , okamura , s. , chambers , k.c . , cowie , l.l . ,
joseph , r.d . , sanders , d.b . , 1998 , _
a&a _ , 336 , l9 kennicutt , r.g . , 1998 , in _ starbursts : triggers , nature and evolution _ ,
b. guiderdoni & a. kembhavi ( eds ) , edp sciences / springer verlag kurucz , r. , 1992 , iau symposium 149 , 225 lagache , g. , abergel , a. , boulanger , f. , dsert , f.x . ,
puget , j.l . , 1999 , _
, 344 , 322 lilly , s.j . ,
eales , s.a . ,
gear , w.k.p . ,
hammer , f. , le fvre , o. , crampton , d. , bond , j.r .
, dunne , l. , 1999 , _ apj _ , 518 , 641 meurer , g.r . , heckman , t.m . , leitherer , c. , kinney , a. robert , c. , garnett , d.r . , 1995 , _ aj _ , 110 , 2665 meurer , g.r . , heckman , t.m .
, calzetti , d. , 1999 , _ apj _ , 521 , 64 puget , j.l . , abergel , a. , bernard , j.p . , boulanger , f. , burton , w.b . , dsert , f.x . , hartmann , d. , 1996 , _ a&a _ , 308 , l5 puget , j.l . , lagache , g. , clements , d.l
reach , w.t .
, aussel , h. , bouchet , f.r . , cesarsky , c. , dsert , f.x . ,
dole , h. , elbaz , d. , franceschini , a. , guiderdoni , b. , moorwood , a.f.m .
_ a&a _ , 345 , 29 rigopoulou , d. , lawrence , a. , rowan robinson , m. , 1996 , _ mnras _ , 278 , 1049 smail , i. , ivison , r.j . ,
blain , a.w . , 1997 , _ apj _ , 490 , l5 smail , i. , ivison , r.j . ,
blain , a.w .
, kneib , j.p . , 1998 , _ apj _ , 507 , l21 soifer , b.t . , neugebauer , g. , franx , m. , matthews , k. , illingworth , g.d .
, 1998 , _ apj _ , 501 , l171 steidel , c.c . ,
adelberger , k.l . ,
giavalisco , m. , dickinson , m. , pettini , m. , 1999 , _ apj _ , 519 , 1 thuan , t.x . , sauvage , m. , madden , s. , 1999 , _ apj _ , 516 , 783 | the current and forthcoming observations of large samples of high redshift galaxies selected according to various photometric and spectroscopic criteria can be interpreted in the context of galaxy formation , by means of models of evolving spectral energy distributions ( seds ) .
we hereafter present stardust which gives synthetic seds from the far uv to the submm wavelength range .
these seds are designed to be implemented into semi analytic models of galaxy formation . |
the discovery of unusual optical transients is a goal of modern surveys .
focused supernova searches ( e.g. the texas supernova search ) or all - sky surveys , such as the panoramic survey telescope @xmath3 rapid response system ( pan - starrs ) , the catalina real - time transient survey ( crts ) , the palomar transient factory ( ptf ) and skymapper are expected to discover a large number of new types of stellar explosions in the next years .
the preliminary results are remarkable , and newly discovered transients are revolutionizing our knowledge of stellar explosions .
ultra - bright supernovae ( sne ) associated with faint and , presumably , metal - poor host galaxies are the most spectacular recent discoveries @xcite .
the field has moved quickly , prompted by the unusual transient , discovered in the hubble space telescope cluster supernova survey @xcite .
its lightcurve was symmetric , with a @xmath4100d rise time in the observed frame .
the spectrum showed broad absorption features and the transient was associated with no obvious host galaxy ( although a weak source , 1.5 from the transient , was marginally detected at magnitude @[email protected] ) . without robust constraints on the absolute magnitude for this transient ,
even the discrimination between galactic and extra - galactic origin was uncertain .
possible scenarios proposed by @xcite for were an outburst of a galactic c - rich white dwarf ( wd ) , a broad absorption lines quasar or a micro - lensing event , but none of them was fully convincing . assuming that the broad features in the spectra of were the c@xmath6 swan bands , @xcite tentatively fixed the redshift at z=0.14 , implying an absolute peak magnitude of about -18 , suggesting a sn - like explosion of a c - rich wolf - rayet ( wr ) star . a tidal disruption of a c - rich star by a black hole @xcite , a galactic wd - asteroid merger and a type ia sn in a dense , c - rich wind produced by a companion star @xcite were also proposed as alternative explanations .
a few events have recently been discovered sharing observed properties with .
data for a total sample of 6 objects have been presented by @xcite .
one of them was , an enigmatic object originally presented in @xcite and classified as a peculiar , over - luminous sn iil . through the detection of narrow interstellar mg ii lines ,
@xcite have definitely proved that these transients are not located in the galaxy or in the local group , but are relatively distant objects , with redshifts between 0.26 and 1.19 .
consequently , they are extremely luminous , with @xmath7-band absolute magnitudes spanning between -22 and -23 . on the basis of the lack of any evidence of a slope consistent with @xmath2co decay in the late - time lightcurve of both and , @xcite favored either a pulsational pair - instability outburst scenario , or core - collapse sne powered by rapidly rotating young magnetars .
unfortunately , follow - up observations collected so far for these transients and the information available for properly studying and modeling their data have been incomplete .
the discovery of a relatively nearby object of this class caught early and followed in detail , has provided us with a new opportunity to study the energy output and spectral evolution of one of nature s brightest explosions .
the crts team @xcite first announced the discovery of an optical transient at ra=11:25:46.71 and dec=-08:49:41.4 , on images obtained on march 13 , 2010 ( magnitude 18.5 , * ? ? ? * ) .
its optical spectrum showed a blue , featureless continuum , and the initial redshift determination ( z=0.17 ) was later corrected by the same authors to z=0.23 @xcite . on
the following day @xcite reported the independent discovery by the ptf survey @xcite of the same variable source ( labeled as ) at several epochs between march 5 and 16 , whilst no object was seen on march 4.27 ut to a limiting magnitude of 20.4 .
optical spectra on march 18.27 ut showed that was a luminous sn similar to the ultra - bright @xcite .
the spectrum showed broad features attributed to o ii @xcite .
the presence of narrow lines attributed to a host galaxy allowed them to estimate the redshift to z=0.23 .
in the course of the pan - starrs1 telescope ( ps1 ) 3@xmath8 survey , we recovered the transient ( , hereafter sn 2010gx ) between march 12 and 17 showing that its luminosity was still rising @xcite .
@xcite noted the presence of a faint host galaxy in archive sdss images ( ) with magnitudes @xmath9=22.7 , @xmath10=0.3 , and confirmed the @xcite redshift estimate of z=0.23 . at this redshift ,
the @xmath9-band absolute magnitude of the host galaxy is about -18 , similar to that of the lmc .
we carried out an extensive @xmath11 photometric follow - up campaign of using the telescopes listed in table [ tab1 ] .
the observed lightcurves , calibrated using 10 sdss stars in the field of the transient , are shown in figure [ fig1 ] .
the transient was discovered in the rising phase , and its lightcurves are asymmetric .
the pre - discovery limit of march 4 @xcite indicates that experienced a fast rise to maximum , followed by a slower magnitude decline .
a similar asymmetry was also observed in the lightcurve of @xcite .
the photometric evolution of is somewhat different from the bell - like shape observed in the slow - evolving lightcurve of @xcite .
assuming negligible host galaxy reddening ( galactic reddening of @xmath12=0.04 mag , * ? ? ?
* ) and accounting for redshift effects - band from observed sdss @xmath13-band . ] , an absolute rest - frame peak magnitude of @xmath14@xmath15 - 21.2 ( vega system ) is determined for . in figure [ fig2 ]
we compare the rest - frame , @xmath16-band absolute lightcurve of with those of a few ultra - bright events and classical type ib / c sne , including broad - lined energetic sne ic .
the epoch of the @xmath16-band maximum for was computed with a low - order polynomial fit to the lightcurve , and found to be at jd=2455283@xmath172 .
the absolute peak magnitude of is slightly fainter than that of , magnitudes @xcite . ] while no direct comparison is possible with the peculiar ( observed well past - maximum in the @xmath16-band , * ? ? ?
* ) and ( * ? ? ?
* for which a reliable rest - frame absolute lightcurve was computed only for the @xmath7-band )
. however , appears to be 2.5 - 5 mag brighter than sne ib / c reported in figure [ fig2 ] , and its overall evolution is much slower than that of normal ib / c events , although faster than that experienced by .
a major difference between the lightcurves of sne ib / c and is the _ apparent _ lack of a radioactive tail , in analogy to that observed in the case of other objects of the quimby et al . sample .
however , with the data collected so far , we can not exclude the possibility that the lightcurve flattens onto a radioactive tail at later epochs . in that case
, the expected amount of @xmath2ni ejected by would be comparable ( or only marginally higher , e.g. @xmath181m@xmath19 ) to that of type ib / c sne .
a sequence of spectra of was obtained with the 2.56-m nordic optical telescope and the 4.2-m william herschel telescope ( la palma , canary islands ) , and the 8.1-m gemini south telescope ( cerro pachn , chile ) .
pre - maximum spectra obtained on march 21 and 22 show a very blue continuum ( with a blackbody temperature @xmath20 k ) with broad absorption features below @xmath45700 ( figure [ fig3 ] ) .
weak , narrow emission lines ( h@xmath21 , h@xmath22 and the [ o ii ] doublet at 4959,5007 ) of the host galaxy are also visible , confirming the identification of the broad features as o ii , also identified by @xcite in the spectra of ( figure [ fig4 ] , top ) .
a spectrum obtained on april 1 ( + 4d ) is still blue ( t@xmath23 k ) but is almost featureless .
a significant evolution of the spectra of then occurred at 10 - 20d after peak . at these epochs
the spectra show very broad p - cygni absorptions of ca ii , fe ii and si ii , very similar to those observed in spectra of young sne ic @xcite .
the subsequent spectrum , obtained on may 2 ( + 30d ) , is markedly more similar to those of sne ic soon after maximum light .
finally , a further spectrum was obtained on jun 5 ( + 57d ) , and showed only a mild evolution in the spectral features .
the spectral evolution of from an -like event to a type ic sn provides an unexpected clue for understanding the evolutionary path of this class of transients . in order to produce o ii features ( together with the si iii ,
c ii and mg ii lines observed in spectra of other objects of this family , see * ? ? ?
* and figure [ fig4 ] , top ) , high photospheric temperatures are necessary .
interestingly , @xcite noted a short - life `` w '' feature in a very early spectrum of the type ib . that feature , visible in a spectrum taken 1.84d after the x - ray flash associated with the sn , disappeared 1d later .
@xcite noted striking similarity with the early - time spectrum of and ( following * ? ? ?
* ) tentatively identified such short - life features as a combination of o iii , n iii and c iii lines .
however , the `` w '' feature in is slightly blueshifted compared to the analogous feature visible in the early spectra of and .
therefore , fleeting lines due to ionized intermediate - mass elements could be common in very early spectra of some type ib / c sne
. however , these lines are visible for several weeks after the explosion in -like objects , which is likely due to higher densities and temperatures of the ejecta which persist for longer than in canonical sne ib / c .
as the sn expands , the ejecta become cooler and other broad lines appear ( ca ii , mg ii , fe ii and si ii ) .
these features are commonly visible in ib / c spectra around maximum @xcite . in figure [ fig4 ] ( middle ) , a later spectrum of ( + 21d ) is compared with a pre - maximum spectrum of the normal type ic sn 1994i @xcite and a slightly post - maximum spectrum of the broad - line ic sn 2003jd @xcite . the striking similarity among these
3 spectra is a confirmation that ( and possibly all -like objects ) should be considered spectroscopically as sne ic , although with rather extreme photometric properties ( section [ ph ] ) .
the similarity with normal stripped - envelope sne is even more evident in the comparison of the last spectrum of ( june 5 ) with spectra of the type ic @xcite and @xcite obtained about 1 week after their @xmath16-band peaks .
provides important clues to understand the nature of ultra - bright events .
its spectro - photometric similarities with this family is well established : high luminosity , slow - evolving lightcurves , similar spectral properties and faint host galaxies .
the spectral evolution of now links this family of transients to the more common type ib / c sne , and by implication the progenitor stars .
the overall spectral evolution is indeed similar to that of sne ib / c , although spectroscopically evolved on a much longer timescale .
the observed parameters of present several problems in interpreting the explosion .
its impressive luminosity at maximum and slower evolution could simply be interpreted as implying large photospheric radii ( @xmath24 ) and large ejecta masses ( @xmath25 ; for radiative diffusion from a sphere ) .
the energy source for sne ib / c is the decay of radioactive isotopes , but the 80d - long post - peak decline of ( fig .
2 ) is too steep to be due to @xmath2co decay .
it is plausible that a radioactive tail could be detected at later phases , if the lightcurve flattens to the luminosity of the type ic sn 1998bw . but
this would imply @xmath181m@xmath19 of @xmath2co decaying into @xmath2fe to power the tail luminosity , and such a moderate mass of @xmath2ni can not account for the high bolometric luminosity at peak , which is @xmath43 - 4@xmath2610@xmath27 erg s@xmath28 . assuming that the bolometric luminosity from the earliest ptf detection to our earliest multi - band observation was constant , the energy radiated by during the first 100d is @xmath46@xmath2610@xmath29 erg .
the peak luminosity of sn 2010gx is quite similar to that of sn 2007bi , but the lightcurve and spectral evolution are completely different @xcite . the slow decay time , and appearance of strong [ feii ] lines in sn 2007bi suggested a kinetic energy of few @xmath2610@xmath30 erg , very massive ejecta and 3 - 6m@xmath19 of @xmath2ni synthesized .
@xcite postulated this was the explosion of a 100m@xmath19 core in a pair - instability sn . while this possibility was also noted by @xcite , the gravitational collapse of the c+o core of a massive star ( m@xmath31 = 50 - 100m@xmath19 ) is a viable mechanism ( as recently calculated by * ? ? ?
whatever the explosion scenario , a large amount of @xmath2ni is necessary .
however , sn 2010gx is markedly different in its properties , particularly the more rapid decay in its lightcurve indicates that the pair - instability scenario and a large @xmath2ni production are unlikely to be the explanation . the apparent lack of any evidence of lightcurve flattening to a radioactive tail led @xcite to favor the _ pulsational _ pair - instability eruption scenario over a genuine sn explosion for sn 2005ap , scp-06f6 and other ptf sne . in the pulsational pair - instability model ,
the luminosity is generated by the collision of shells of material material ejected at different times by the pulsations .
little or no @xmath2ni powers the lightcurve for long periods .
the outbursts are expected to be energetic , reaching very high peak luminosities and creating hot ( t@xmath3225000k ) , optically thick photospheres @xcite .
all of this is consistent with the parameters observed in ultra - bright sne , and our well sampled lightcurve of sn 2010gx is not too different from those calculated by @xcite .
however , these models are for supergiant progenitors with large , extended h - rich envelopes .
the energy released in the pulsations is predicted to be 0.005 - 2@xmath2610@xmath29erg , in most cases below @xmath33erg .
this is enough to eject the loosely bound envelope of an extended supergiant , but weather or not this mechanism could eject a substantial part of a more compact wr star @xcite remains to be calculated in detail .
additionally , we do not see any sign of interaction between dense gas shells in the form of narrow circumstellar lines . while the pulsational pair - instability model is appealing as it can produce the high luminosity
, it needs further consideration to determine if it is physically viable for h - free progenitor stars .
another possibility , also discussed in @xcite , is that ultra - bright sne are powered by the spin down of newly born magnetars @xcite .
a magnetar with a moderate magnetic field ( b@xmath1510@xmath34 gauss ) and spinning periods of 2 - 20ms can produce peak luminosities similar to that observed in sn 2010gx ( @xmath43 - 4@xmath2610@xmath27 erg s@xmath28 ) .
in addition , magnetar - powered sn models do not need large @xmath2ni and total ejected masses @xcite , and can show ic sn features @xcite .
however our extensive lightcurve coverage of sn 2010gx shows a faster decline than the model lightcurves , and our estimated photospheric temperatures are a factor 2 - 4 higher than model predictions . the observational evidence presented in this paper links ( and probably the entire family of transients described by * ? ? ?
* ) with sne ic .
the very luminous and broad lightcurve implies much larger ejecta masses than inferred even for the broad - lined sne ic ( @xmath48 - 15m@xmath35 , * ? ? ?
the close similarity in the spectra implies the progenitor was a massive wr star , but the energy source powering the remarkable luminosity is uncertain .
in fact we have a sn - like transient which does not comfortably match any of the known sn scenarios , i.e. core - collapse and @xmath2ni - powered explosion , pair - instability , pulsational pair - instability nor magnetar - powered event .
the key diagnostics in the future will be late - time photometric monitoring after solar conjunction , and very early detection of new events .
the presence of a late - time lightcurve tail with a slope roughly consistent with the @xmath2co decay could support a real sn explosion , and the early rise - time can help to determine the progenitor radius and possibly signs of interaction between colliding shells .
the recent suggestion that the most massive stars in the lmc may be up to 320m@xmath35 @xcite could lead to more diverse sn progenitor populations than is currently appreciated @xcite .
the ps1 surveys have been made possible through contributions of the institute for astronomy at the university of hawaii in manoa , the pan - starrs project office , the max - planck society and its participating institutes , the max planck institute for astronomy , heidelberg and the max planck institute for extraterrestrial physics , garching , the johns hopkins university , the university of durham , the university of edinburgh , the queen s university belfast , the harvard- smithsonian center for astrophysics , and the los cumbres observatory global telescope network , incorporated .
this work was conducted as part of a euryi scheme award .
barbary , k. et al .
2009 , , 690 , 1358 baron , e. , hauschildt , p. h. , branch , d. , kirshner , r. p. , filippenko , a. v. 1996 , mnras , 279 ,
799 campana , s. et al .
2006 , , 442 , 1008 cobb , b. e. , bailyn , c. d. , van dokkum , p. g. , natarajan , p. 2006
, , 645 , l113 crowther , p. a , schnurr , o. , hirschi , r. , yusof , n. , parker , r. j , goodwin , s. p , & abu kassim , h. 2010 , arxiv:1007.3284 about dessart , l. , livne , e. , & waldman , r. 2010 , , 405 , 2113 drake , a. j. et al .
2009 , , 696 , 870 filippenko , a. v. 1997 , ara&a , 35 , 309 folatelli , g. et al .
2006 , , 641 , 1039 foley , r. et al .
2003 , , 115 , 1220 galama , t. j. et al .
1998 , , 395 , 670 gnsicke , b. t. , levan , a. j. , marsh , t. r. , wheatley , p. j. 2009 , , 697 , 129 gal - yam , a. et al .
2009 , , 462 , 624 gezari , s. et al .
2009 , , 690 , 1313 treffers , r. r. , richmond , m. w. 2001 , , 113 , 1349 hunter , d. j. et al .
2009 , , 508 , 371 kasen , d. , bildsten , l. 2010 , , 717 , 245 law , n. m. et al .
2009 , , 121 , 1395 mahabal , a. a. et al .
2010 , atel 2490 , 1 mahabal , a. a. , drake , a. j. 2010 , atel 2508 , 1 mazzali , p. a. et al .
2008 , science , 321 , 1185 mckenzie , e. h. , schaefer , b. e. 1999 , , 111 , 964 miller , a. a. et al .
2009 , , 690 , 1303 mirabal , n. et al .
2006 , , 643 , l99 modjaz , m. et al .
2009 , , 702 , 226 moriya , t. , tominaga , n. , tanaka , m. , maeda , k. , nomoto , k. 2010 , submitted , preprint arxiv:1004.2967 pandey , s. b. , anupama , g. c. , sagar , r. , bhattacharya , d. , sahu , d. k. , pandey , j. c. 2002 , , 340 , 375 pastorello et al .
2008 , , 389 , 955 pastorello et al .
2010 , atel , 2504 , 1 patat , f. et al .
2001 , , 555 , 900 pian , e. et al .
2006 , , 442 , 1011 quimby , r. m. , aldering , g. , wheeler , j. c. , hflich , p. , akerlof , c. w. , rykoff , e. s. 2007 , , 668 , 99 quimby , r. m. et al .
2010 , , submitted , preprint : arxiv0910.0059 quimby , r. m. , kulkarni , s. r. , ofek , e. , kasliwal , m. m. , levitan , d. , gal - yam , a. , cenko , s. b. 2010 , atel 2492 , 1 rau , a. et al .
2009 , , 121 , 1334 richmond , m. w. et al .
1996 , , 111 , 327 rosswog , s. , ramirez - ruiz , e. , hix , w. r. 2009 , , 695 , 404 schlegel , d. j. , finkbeiner , d. p. , davis , m. 1998 , , 500 , 525 smartt , s. j. 2009 , , 47 , 63 soderberg et al .
2008 , , 453 , 469 sollerman , j. , kozma , c. , fransson , c. , leibundgut , b. , lundqvist , p. , ryde , f. , woudt , p. 2000 , , 537l , 127 sollerman , j. et al
2006 , , 454 , 503 soker , n. , frankowski , a. , kashi , a. 2009 , new astr . , 15 , 189 taubenberger , s. et al .
2006 , , 371 , 1459 tomita , h. et al .
2006 , , 644 , 400 valenti , s. et al . 2008a , , 383 , 1485 valenti , s. et al .
2008b , , 673 , l155 woosley , s. e. , blinnikov , s. , heger , a. 2007 , , 450 , 390 woosley , s. e. 2009 , submitted ( arxiv:0911.0698 ) yoshii , y. et al .
2003 , , 592 , 467 young , d. r. et al .
2010 , , 512 , 70 ccccccccc date & jd & phase@xmath36 & u & g & r & i & z & telescope + 12/03/2010 & 2455267.95 & -12.2 & & & & 18.95 ( 0.05 ) & & ps1 + 13/03/2010 & 2455268.99 & -11.4 & & & 18.65 ( 0.03 ) & & & ps1 + 17/03/2010 & 2455272.96 & -8.2 & & 18.65 ( 0.09 ) & & & & ps1 + 20/03/2010 & 2455276.42 & -5.3 & 18.33 ( 0.05 ) & 18.45 ( 0.03 ) & 18.54 ( 0.05 ) & 18.76 ( 0.10 ) & 18.82 ( 0.18 ) & lt + 21/03/2010 & 2455276.61 & -5.2 & & & 18.53 ( 0.02 ) & 18.76 ( 0.02 ) & & not + 22/03/2010 & 2455277.77 & -4.3 & & & 18.50 ( 0.05 ) & & & gs + 22 - 23/03/2010 & 2455278.50 & -3.7 & 18.33 ( 0.02 ) & 18.44 ( 0.02 ) & 18.48 ( 0.02 ) & 18.71 ( 0.02 ) & 18.80 ( 0.04 ) & lt + 23/03/2010 & 2455279.31 & -3.0 & & 18.43 ( 0.02 ) & 18.48 ( 0.02 ) & 18.73 ( 0.05 ) & & lot + 24/03/2010 & 2455280.14 & -2.3 & & 18.40 ( 0.02 ) & 18.48 ( 0.02 ) & 18.72 ( 0.03 ) & 18.82 ( 0.07 ) & lot + 25/03/2010 & 2455281.07 & -1.6 & 18.38 ( 0.13 ) & & & & & uvot + 27/03/2010 & 2455282.52 & -0.4 & 18.48 ( 0.03 ) & 18.43 ( 0.02 ) & 18.50 ( 0.03 ) & 18.67 ( 0.03 ) & 18.84 ( 0.04 ) & lt + 31/03/2010 & 2455286.58 & 2.9 & 18.66 ( 0.07 ) & 18.48 ( 0.04 ) & 18.53 ( 0.05 ) & 18.72 ( 0.04 ) & 18.87 ( 0.07 ) & lt + 31/03/2010 & 2455286.70 & 3.0 & 18.67 ( 0.21 ) & & & & & uvot + 01/04/2010 & 2455287.99 & 4.1 & 18.72 ( 0.07 ) & 18.51 ( 0.13 ) & 18.56 ( 0.10 ) & 18.75 ( 0.13 ) & 18.85 ( 0.17 ) & ftn + 02/04/2010 & 2455288.85 & 4.8 & 18.78 ( 0.02 ) & 18.52 ( 0.04 ) & & & & ftn + 04/04/2010 & 2455291.45 & 6.9 & 18.95 ( 0.04 ) & 18.57 ( 0.01 ) & 18.60 ( 0.02 ) & 18.71 ( 0.02 ) & 18.84 ( 0.05 ) & lt + 06/04/2010 & 2455293.48 & 8.5 & 19.08 ( 0.25 ) & 18.64 ( 0.02 ) & 18.64 ( 0.04 ) & 18.78 ( 0.03 ) & 18.91 ( 0.03 ) & lt + 08/04/2010 & 2455294.59 & 9.4 & 19.26 ( 0.29 ) & & & & & uvot + 09/04/2010 & 2455295.53 & 10.2 & & 18.71 ( 0.06 ) & & & & gs + 09/04/2010 & 2455296.46 & 10.9 & 19.37 ( 0.04 ) & 18.72 ( 0.02 ) & 18.70 ( 0.08 ) & 18.87 ( 0.16 ) & 18.96 ( 0.13 ) & lt + 11/04/2010 & 2455297.84 & 12.1 & 19.44 ( 0.02 ) & 18.75 ( 0.01 ) & 18.69 ( 0.01 ) & 18.88 ( 0.03 ) & 18.96 ( 0.04 ) & ftn + 11/04/2010 & 2455298.48 & 12.6 & & 18.82 ( 0.01 ) & 18.72 ( 0.01 ) & 18.89 ( 0.07 ) & 18.99 ( 0.04 ) & lt + 15/04/2010 & 2455301.80 & 15.3 & 19.98 ( 0.04 ) & 18.97 ( 0.03 ) & 18.78 ( 0.02 ) & 18.95 ( 0.04 ) & 19.00 ( 0.05 ) & ftn + 16/04/2010 & 2455303.04 & 16.3 & 20.13 ( 0.31 ) & & & & & uvot + 19/04/2010 & 2455305.88 & 18.6 & & & & 19.04 ( 0.02 ) & 19.12 ( 0.06 ) & ftn + 21/04/2010 & 2455308.38 & 20.6 & & 19.38 ( 0.02 ) & 19.02 ( 0.03 ) & 19.10 ( 0.03 ) & 19.24 ( 0.07 ) & lt + 22/04/2010 & 2455308.58 & 20.8 & & & 19.03 ( 0.05 ) & & & gs + 23/04/2010 & 2455310.13 & 22.1 & 20.71 ( 0.23 ) & & & & & uvot + 23/04/2010 & 2455310.40 & 22.3 & & 19.48 ( 0.33 ) & 19.06 ( 0.28 ) & & 19.28 ( 0.15 ) & lt + 24/04/2010 & 2455310.84 & 22.6 & 20.75 ( 0.14 ) & 19.49 ( 0.05 ) & 19.08 ( 0.04 ) & & & ftn + 24/04/2010 & 2455310.88 & 22.7 & & & & 19.16 ( 0.08 ) & 19.30 ( 0.13 ) & ftn + 26/04/2010 & 2455313.42 & 24.7 & & 19.77 ( 0.07 ) & & & & lt + 26/04/2010 & 2455313.44 & 24.7 & & 19.77 ( 0.08 ) & 19.17 ( 0.05 ) & 19.30 ( 0.10 ) & 19.38 ( 0.12 ) & lt + 01/05/2010 & 2455318.39 & 28.8 & 21.74 ( 0.30 ) & 20.12 ( 0.03 ) & 19.43 ( 0.02 ) & 19.41 ( 0.03 ) & 19.48 ( 0.07 ) & lt + 02/05/2010 & 2455318.84 & 29.1 & 21.77 ( 0.31 ) & & & 19.43 ( 0.18 ) & 19.49 ( 0.29 ) & ftn + 02/05/2010 & 2455319.09 & 29.3 & 21.81 ( 0.40 ) & & & & & uvot + 03/05/2010 & 2455319.54 & 29.7 & 21.84 ( 0.29 ) & & & & & lt + 03/05/2010 & 2455320.43 & 30.4 & & 20.29 ( 0.04 ) & 19.55 ( 0.04 ) & 19.54 ( 0.05 ) & 19.54 ( 0.11 ) & lt + 05/05/2010 & 2455321.88 & 31.6 & 22.02 ( 0.16 ) & & & & & ftn + 05/05/2010 & 2455321.94 & 31.7 & & & & 19.61 ( 0.04 ) & 19.58 ( 0.08 ) & fts + 05/05/2010 & 2455322.03 & 31.7 & & 20.40 ( 0.27 ) & 19.62 ( 0.08 ) & & & fts + 06/05/2010 & 2455322.88 & 32.4 & & 20.48 ( 0.03 ) & 19.66 ( 0.04 ) & & & fts + 06/05/2010 & 2455322.94 & 32.5 & & & & & 19.61 ( 0.06 ) & fts + 06/05/2010 & 2455323.02 & 32.5 & & & & 19.63 ( 0.14 ) & & fts + 06/05/2010 & 2455323.48 & 32.9 & & 20.54 ( 0.03 ) & 19.68 ( 0.03 ) & 19.65 ( 0.04 ) & 19.63 ( 0.16 ) & lt + 07/05/2010 & 2455324.38 & 33.6 & & 20.57 ( 0.04 ) & 19.71 ( 0.03 ) & 19.68 ( 0.11 ) & 19.65 ( 0.15 ) & lt + 08/05/2010 & 2455325.11 & 34.2 & & 20.66 ( 0.08 ) & 19.76 ( 0.05 ) & & & fts + 09/05/2010 & 2455326.09 & 35.0 & & & & 19.74 ( 0.03 ) & 19.74 ( 0.07 ) & fts + 12/05/2010 & 2455328.95 & 37.4 & & 20.90 ( 0.04 ) & 19.93 ( 0.02 ) & & 19.91 ( 0.07 ) & fts + 14/05/2010 & 2455330.84 & 38.9 & @xmath3722.49 & & & & & ftn + 14/05/2010 & 2455330.90 & 38.9 & & & & 19.92 ( 0.03 ) & & fts + 14/05/2010 & 2455331.01 & 39.0 & & & & & 20.03 ( 0.05 ) & fts + 14 - 15/05/2010 & 2455331.50 & 39.4 & & 21.13 ( 0.09 ) & 20.09 ( 0.14 ) & 19.97 ( 0.15 ) & & lt + 15/05/2010 & 2455331.88 & 39.7 & & 21.14 ( 0.07 ) & 20.09 ( 0.03 ) & & & fts + 16/05/2010 & 2455332.82 & 40.5 & @xmath3722.53 & & & & & ftn + 17/05/2010 & 2455334.09 & 41.5 & & & & 20.07 ( 0.05 ) & & fts + 18/05/2010 & 2455335.02 & 42.3 & & 21.52 ( 0.08 ) & 20.40 ( 0.08 ) & & 20.20 ( 0.06 ) & fts + 19/05/2010 & 2455336.40 & 43.4 & & 21.54 ( 0.06 ) & 20.44 ( 0.06 ) & 20.14 ( 0.07 ) & 20.25 ( 0.18 ) & lt + 25/05/2010 & 2455342.39 & 48.3 & & 21.91 ( 0.30 ) & 20.70 ( 0.22 ) & 20.40 ( 0.12 ) & 20.49 ( 0.23 ) & lt + 30/05/2010 & 2455347.42 & 52.4 & & 22.51 ( 0.12 ) & 20.99 ( 0.06 ) & 20.87 ( 0.07 ) & 20.73 ( 0.29 ) & lt + 01/06/2010 & 2455348.95 & 53.6 & & & & 20.91 ( 0.16 ) & & fts + 04/06/2010 & 2455352.47 & 56.5 & & & & 21.13 ( 0.19 ) & & gs + 08/06/2010 & 2455355.43 & 58.9 & @xmath3723.88 & 23.55 ( 0.28 ) & 21.52 ( 0.09 ) & 21.31 ( 0.12 ) & 21.33 ( 0.41 ) & wht + 08/06/2010 & 2455355.98 & 59.3 & & & 21.58 ( 0.13 ) & 21.31 ( 0.16 ) & & fts + 13/06/2010 & 2455360.97 & 63.4 & & 23.91 ( 0.30 ) & 21.81 ( 0.20 ) & 21.56 ( 0.13 ) & & fts + 13/06/2010 & 2455361.42 & 63.8 & & 23.95 ( 0.49 ) & 21.83 ( 0.12 ) & 21.58 ( 0.29 ) & 21.69 ( 0.39 ) & lt + 16/06/2010 & 2455363.98 & 65.8 & & @xmath3723.73 & 21.91 ( 0.16 ) & & & fts + 29/06/2010 & 2455376.88 & 76.3 & & & 22.47 ( 0.21 ) & 22.38 ( 0.33 ) & & fts + | recent searches by unbiased , wide - field surveys have uncovered a group of extremely luminous optical transients .
the initial discoveries of sn 2005ap by the texas supernova search and scp-06f6 in a deep hubble pencil beam survey were followed by the palomar transient factory confirmation of host redshifts for other similar transients .
the transients share the common properties of high optical luminosities ( peak magnitudes @xmath0 to @xmath1 ) , blue colors , and a lack of h or he spectral features .
the physical mechanism that produces the luminosity is uncertain , with suggestions ranging from jet - driven explosion to pulsational pair - instability .
here we report the most detailed photometric and spectral coverage of an ultra - bright transient ( sn 2010gx ) detected in the pan - starrs1 sky survey . in common with other transients in this family ,
early - time spectra show a blue continuum , and prominent broad absorption lines of o ii .
however , about 25d after discovery , the spectra developed type ic supernova features , showing the characteristic broad fe ii and si ii absorption lines .
detailed , post - maximum follow - up may show that all sn 2005ap and scp-06f6 type transients are linked to supernovae ic .
this poses problems in understanding the physics of the explosions : there is no indication from late - time photometry that the luminosity is powered by @xmath2ni , the broad lightcurves suggest very large ejected masses , and the slow spectral evolution is quite different from typical ic timescales .
the nature of the progenitor stars and the origin of the luminosity are intriguing and open questions . |
the functional behavior of neurons in primary visual cortex can be partially modeled as a linear response to visual stimuli .
this linear response can be formally described as a generalized wavelet analysis of images , intended as functions from @xmath0 to @xmath1 ( we consider greyscale images ) , with respect to a family of functions @xmath2 .
such a family can be indexed by a parameter set @xmath3 that encodes some relevant geometric properties , which correspond to local features of the input image , and that is mapped on the cortex .
each neuron can be identified with a given collection of such properties , i.e. by an element of @xmath3 , and it is said to be _ selective _ with respect to the associated features . a neuron receives a visual input only from a bounded region in the visual plane . inside this region , which is called the _ receptive field _ , a neuron can be selective with respect to several _ local features _ , such as the local orientation , the local spatial frequency ( the rate of oscillations ) , the local apparent velocity ( the component of the speed of motion that is orthogonal to the locally detected orientation ) .
the neurophysiological and computational meaning of this selectivity will be discussed in sections [ sec : neuro ] and [ sec : mathmod ] .
the primary visual cortex ( v1 ) of many mammals , notably including humans , shows a remarkably regular spatial organization of the features its neurons are selective to@xcite .
v1 is arranged in such a way that each feature varies smoothly on the cortical layer : neurons that are close to one another are selective to similar features .
receptive fields are arranged in a topographic , or _ retinotopic _ way ,
i.e. neurons that are close to one another in the cortex are associated to similar regions in the visual plane .
they are also highly overlapping , in the sense that two adjacent neurons respond to two almost entirely overlapping regions .
orientation varies smoothly , in the sense that two adjacent neurons are selective for similar local orientations , and the same is true for the other listed features .
moreover , the variability of some of the features is mainly associated to the horizontal position of the neurons , according to a so - called hypercolumnar structure whose discovery dates back to the nobel prizes hubel and wiesel .
the cerebral cortex can be thought as a thick layered stack of ( folded ) planes and , roughly speaking , the hypercolumnar paradigm postulates that the function ( feature selectivity ) of neurons does not vary along the transversal direction but depends only on the position on the cortical plane ( see also@xcite ) . this structure has led to the representation of the _ functional architecture _ of primary visual cortex in terms of maps from the cortical plane to the space of a given feature , called _ cortical maps _ , which describe the displacement of the neurons in terms of their selectivity@xcite .
probably the best known cortical map is that of orientation preferences , depicted in figure [ fig : ohki ] .
it is a map @xmath4 that indicates , for each point on the cortex , what is the orientation the underlying neuron ( population ) is selective to .
this means that if a light stimulus is present in the receptive field of a neuron , and that stimulus is mainly oriented along the depicted direction ( e.g. it contains an edge with that orientation ) , then the neuron will be maximally stimulated .
conversely , if the stimulus contains an edge in the orthogonal direction , the neuron is minimally stimulated .
a satisfactory understanding of the properties of the family @xmath5 or of the cortical representation of the set @xmath3 from the point of view of image processing is still an open problem , even if many partial results are available .
this issue is relevant for artificial vision , because the capabilities of analyzing , compressing , and processing images possessed by the visual cortex still outperform any artificial system .
it is also relevant for neuropsychology and cognitive studies , because it is still not clear what is the role of the different constituents of the brain in order to build the structured and organized visual representation generated by the visual cortex . on the other hand
, this approach suffers from several limitations : just to mention a few , it does not include the various nonlinear mechanisms that contribute to the neural responses , and it does not take into account the dynamics of the network of connected neurons as well as the dynamic of activation of each single neuron .
the purpose of this paper is to review some key facts about mathematical models of neurophysiological behaviors , to show some new properties of these models , to present some open problems that can be of some interest to the community working on harmonic analysis and approximation theory , and to provide some relevant terminology and references commonly used in the neuroscience literature .
in this section we will introduce the main features of v1 that we will consider .
we will focus on linear behaviors and on cortical maps , disregarding the several issues on nonlinear behaviors , such as normalizations or surround effects@xcite , on dynamical effects , such as spatio - temporal behaviors of receptive profiles@xcite , neural oscillations@xcite , neurodynamics@xcite , or on properties of connectivities@xcite and their geometric models @xcite . _
primary visual cortex _
v1 is the first and largest cortical area dedicated to the processing of visual stimuli .
it is located in brodmann area 17 , involving both hemispheres of the occipital lobe of the brain , and it is also called striate cortex . it is followed by higher visual areas , from v2 to the so - called v5/mt , also called _ extrastriate cortices _ , that are located in brodmann areas 18 and 19 .
v1 receives direct sensory input from the _ lateral geniculate nucleus _
lgn , which is located in the thalamus and acts as a preprocessing unit of the visual stimuli collected by the retina .
the @xmath6 retinal receptors are connected by @xmath7 optic nerve fibers to the lgn and finally to the @xmath8 v1 cells@xcite . on the one hand
, this gives a strong indication that alredy in the retina there must occur a cospicuous processing of information . on the other hand
, the @xmath9 units of infomations reaching v1 are evidently highly reprocessed .
v1 transmits its information to extrastriate cortical areas by connections to v2 , which is then connected to higher visual areas . the electric inputs that v1 receives are due to three main classes of _ connectivities _ : _ feedforward _ inputs coming from the eye , through the lateral geniculate nucleus ; _ lateral _ inputs coming from other v1 neurons ; and _ feedback _ inputs coming from higher cortical areas .
v1 neurons are characterized by their feature selectivity , by their physical displacement , and by their connectivities . just like all neurons ,
the neurons that populate v1 accumulate an electric potential on their boundary membrane by means of chemical mechanisms involving ion concentrations . when this electric potential exceeds a given threshold , the neuron releases an impulsive discharge called _ action potential _ , that is transmitted by synapses as an electric current .
such current flows through connections until it reaches other synapses , where it is again converted into an electric potential that accumulates in the neuron that receives the signal . a neuron that emits an action potential
is said to be _ firing _ , the action potential itself being called a spike .
the activity of a neuron is its temporal sequence of spikes , and it is generally quantified by a nonnegative scalar quantity that measures its average _ firing rate _ , that is the frequency of spikes per second that it emits@xcite .
a neuron has always a positive firing rate : even when there is no external stimulus acting on the network , each neuron fires from time to time ( rest state ) .
the _ activity _ of a neuron is then measured as the difference between its firing rate and the firing rate at rest .
the firing rate depends in a nonlinear way on the membrane potential , which is generally modeled as a sigmoid function ; in particular , each neuron has a maximum firing rate , a linear range of membrane potentials where the firing rate is linear , and a minimum firing rate equal to zero .
a v1 neuron s visual receptive field is defined as the region of the visual field where the presence of light ( i.e. a visual stimulus ) causes a neural response .
the direct activity response of a neuron to a stimulus does not depend only on the stimulus location on the visual field , but also on the spatio - temporal distribution of the light in the receptive field region .
one of the basic needs of a neurophysiologically based formal theory of visual perception is to have a quantitative description of such dependence , that is sufficiently accurate to allow to predict the response of a neuron to any chosen visual stimulus .
the simplest ( though very effective ) model is the _ linear - nonlinear - poisson _ model .
even if it does not account for many nonlinear and/or collective phenomena@xcite , it is able to reproduce much of the single - neuron behavior .
it consists of three stages : 1 . the visual stimulus produces a feedforward input that feeds a neuron @xmath10 with a voltage : this voltage is considered as depending linearly on the visual stimulus . if @xmath11 is a visual stimulus , then the resulting membrane potential is a linear map @xmath12 generally realized as an @xmath13 linear functional @xmath14 = \langle f , \psi_{\xi}\rangle_{l^2({\mathbb{r}}^2)}\ ] ] where the vector @xmath15 depends on the specific neuron .
activity is computed by applying to @xmath16 $ ] a static nonlinearity , generally a sigmoid @xmath17 .
the firing process is considered as a poisson process whose rate is given by @xmath18 = { \mathscr{s}}(v_{\xi}[f])$ ] .
this is equivalent to assume that no information is encoded in the time process of spikes .
the fundamental objects of the lnp model are then the filters @xmath5 , where @xmath3 represents the family of visual neurons .
they are generally called _ receptive profiles _ , or also _ classical _ receptive fields or its support . ] .
other effects , which take place in outer regions and/or that contribute nonlinearly to the receptive profile behavior , are referred to as _ nonclassical _ or _ _ extra - classical__@xcite . in order to describe quantitatively the receptive profiles
, one needs to set up an experimental framework that allows to measure them .
the technique used is borrowed from a technique introduced by wiener , that is called _ white noise analysis _ , or _
spike - triggered average _ or , more commonly , _
reverse correlation_. it basically consists of showing random sequences of stimuli while performin electrophysiological recordings of the spike trains produced by a neuron , and performing averages over the stimuli that preceded a spike . in this way
one identifies the type of stimuli that generate the highest linear response of the neuron .
this in general is not sufficient to describe the complete behavior , which is more properly described in terms of so - called _ wiener series _ including nonlinearities of different types .
the procedure is however made consistent by imposing the lnp model to this stochastic analysis@xcite .
effective experimental techniques have been developed to visualize large portions of cortical tissue activity in vivo .
they are called _ optical imaging _ , and rely either on _ voltage - sensitive dyes _ or on _ _ intrinsic signals__@xcite . the first one , developed in the 70 s and 80 s makes use of dye molecules that are capable to transform changes in the membrane potentials into optical signals .
this technique offers a temporal resolution of about @xmath19 and a spatial resolution of about @xmath20 , while the produced signals do not correspond to single cell activity but are due to the averaged electric activity of a local population of neurons ( about 250 - 500 neurons ) .
this technique is considered most often for real - time measurements .
the second one , developed in the 90 s , is based on the intrinsic changes in the amount of light reflected by brain tissue , which is correlated with the presence of neuronal activity .
this mechanism is thought to arise at least in part from local changes in the blood composition that accompany such activity , which can be detected when the cortical surface is illuminated with red light .
active cortical regions absorb more light than less active ones .
this technique provides a more accurate spatial resolution of @xmath21 or better , but it provides a very low temporal resolution .
however , since the experimental set up is easier , it represents the most frequently used technique to obtain cortical maps .
more recently , noninvasive techniques with fmri have been used to provide evidence of orientation maps in humans @xcite . in figure
[ fig : bosking ] one can see the experimental data that produce the map of figure [ fig : ohki ] .
the anesthetized animal is presented with a plane wave ( called _ grating _ ) drifting back and forth along a given direction , and the cortical activity is measured in terms of intrinsic signals .
one obtains then a collection of maps indexed by a parameter @xmath22 , that we can denote with @xmath23 , where @xmath24 represent the local activity around the point @xmath25 on the cortex with respect to the orientation @xmath26 .
observe also that @xmath27 provides the _ orientation tuning _ of the neurons located at @xmath25 , i.e. their selectivity curve to different orientations@xcite .
the information about the local orientation preference given by the map @xmath28 depicted in figure [ fig : ohki ] is obtained by the vector sum @xmath29 since at each @xmath25 this depends only on one fourier coefficient of @xmath27 , of course it can not contain all the information given by the function @xmath30 . as a consequence
, different activity maps can produce the same orientation preference map .
more precisely , since @xmath31 then for all @xmath30 such that the integral at the right hand side is real one obtains the same map @xmath28 .
several experiments have thus focused on the joint displacement of orientation preferences and orientation tuning@xcite .
another well studied cortical map concerns the columnar organizations associated to _
_ ocular dominance__@xcite , that consists of columns where all cells respond preferentially to the same eye .
they show a regularly organized structure , which is strictly related to that of orientation preference
. however , for the present purposes of the analysis of purely two dimensional visual stimuli , binocularity does not seem to play a central role@xcite , so we will disregard completely issues regarding ocular dominance .
this section is devoted to the introduction of some formal properties of classical receptive fields , regarding locations , sizes and shapes , that can be deduced from experimental data .
v1 possesses a three dimensional layered structure .
we will consider each layer as flat and continuous , and distinguish coordinates on the two dimensional leaves from the transversal coordinate
. some relevant features that will be addressed are indeed approximately constant when measured transversally on v1 @xcite , while others are commonly considered in average over the layers : for this reason the two dimensional space corresponding to layers is sometimes called the _
cortical plane_. each v1 cell will then be identified by a point @xmath32 .\ ] ] we will keep the subscript @xmath33 in order to distinguish each such plane from the retinal plane @xmath34 , which we will identify with the visual field ( this is reasonable only locally : the optical properties of the eye and the retina require spherical coordinates , and for this reason the distances on the visual field are generally measured in angular degrees ) .
v1 cells can be classified in terms of several properties which concern their functional behavior , typically parameters of their receptive profiles .
let a given property be described by the elements of a set @xmath35 .
cortical map _ is then defined as a function @xmath36 that associates to a cell at @xmath10 a property @xmath37 .
the first properties one is interested to are related to the location and size of receptive fields : if @xmath38 is the receptive field of the cell at @xmath10 , denote its center with @xmath39 and its size with @xmath40 each v1 layer @xmath41 is organized in a
_ retinotopic _
way @xcite , i.e. for any fixed @xmath42 $ ] we have that @xmath43 is a smooth bijection .
moreover , for each fixed @xmath44 the variation of @xmath45 with @xmath46 , usually referred to as _
@xcite , is small and for the present purposes it can be considered negligible , so we will drop the subscript @xmath46 and simply denote the topographic map as @xmath47 .
this map is approximately given by the ( principal branch of the ) complex logarithm , also called _ log - polar map _ @xcite : for @xmath48 let @xmath49 , then @xmath50 the map @xmath47 is conformal , i.e. it preserves angles , because @xmath51 , and its geometry can be understood by noting that rotations and exponential dilations are mapped as translations : for @xmath52 , and @xmath53 @xmath54 a quantity commonly used to quantify properties of the distribution of receptive fields is the _ cortical magnification factor _
@xmath55 @xcite .
it measures the distance on the cortical surface between two neurons having rf positions whose distance @xmath56 on the visual field is small : @xmath57 making use of the log - polar mapping and denoting with @xmath58 , since @xmath59 we get @xmath60 a related notion is that of _ point image _ @xmath61 : it is defined as the square root of the cortical surface area activated by a point in the visual space .
it is then approximately proportional to the distance on the cortical surface at which two cells have separate receptive fields : @xmath62 the point image is measured@xcite as the product of the average receptive field size times the @xmath55 at @xmath63 degree @xmath64 which is consistent with the linear approximation ( [ eq : cmf ] ) . while @xmath47 and its derived quantity @xmath55 depend only on the surface coordinate @xmath25 , it is not so for @xmath65 , which actually depends both on @xmath25 and on @xmath46 : at any fixed @xmath25 there are several scales @xmath65 , which suggests that a _ multiscale analysis _ is performed . on the other hand ,
if one considers the average @xmath65 over @xmath46 and looks at the variability with respect to the horizontal position @xmath25 , one sees that the size of receptive fields becomes larger as approaching the periphery of the visual field , at almost the same rate due to the log - polar map ( see figure [ fig : gattass ] ) .
this makes the point image an almost constant quantity over the whole v1 surface@xcite .
the log - polar mapping and the enlargement of the receptive fields towards the periphery of the visual field imply that the visual system performs a more refined analysis close to the fovea , losing precision towards the periphery .
this is coherent also with the distribution of the density of retinal receptors , which is higher close to the fovea .
the constancy of the point image implements the complementary principle of dedicating approximately the same amount of cortical tissue to the processing of local information over the whole visual field . and of the local average of @xmath65 against distance from the v1 point where the fovea is mapped ( `` eccentricity '' ) .
right : fmri measurements in humans@xcite.,title="fig : " ] and of the local average of @xmath65 against distance from the v1 point where the fovea is mapped ( `` eccentricity '' ) .
right : fmri measurements in humans@xcite.,title="fig : " ] and of the local average of @xmath65 against distance from the v1 point where the fovea is mapped ( `` eccentricity '' ) .
right : fmri measurements in humans@xcite.,title="fig : " ] and of the local average of @xmath65 against distance from the v1 point where the fovea is mapped ( `` eccentricity '' ) .
right : fmri measurements in humans@xcite.,title="fig : " ] the majority of v1 neurons respond mostly to moving oriented stimuli , so their receptive profiles are more properly described by spatio - temporal filters@xcite . neurons with static behavior are called _ simple cells _ , in contrast to spatio - temporal behaved neurons that belong to a class called _ _ complex cells__@xcite . however , simple and complex cells have some common spatial characteristics @xcite , which consitute the only behaviors we will consider here
. measurements of lnp receptive profiles@xcite show very good fits with the so - called _ gabor model _
, that is given by a sinusoid oscillation under a gaussian bell .
they are defined on the retinal plane / visual field in terms of parameters @xmath66 , @xmath67 and @xmath68 , where @xmath69 as @xmath70 and are considered to be translated and rotated in all configurations on the visual field .
more precisely , for @xmath71 let us denote with @xmath72 and @xmath73 translations and counterclockwise rotations on the plane @xmath74 any neuron located at @xmath75 is associated to a receptive profile @xmath76 , where the parameters @xmath77 , @xmath78 , @xmath79 , @xmath80 , @xmath26 depend on the neuron @xmath10 .
the dependence of @xmath80 on @xmath10 was discussed in the previous section . for @xmath79
, one can approximately consider the introduced quantity @xmath65 as corresponding to @xmath81 , since most of the power is contained within two standard deviations .
the cortical map @xmath28 accounts then for the dependence of @xmath26 on @xmath10 . with respect to @xmath77
, a common simplification can be performed : since most v1 neurons@xcite have a value of @xmath77 equal to @xmath82 or @xmath83 , i.e. they are described either by a pure sine or a pure cosine , the cosine+phase term can be approximately substituted by a complex exponential without phase , meaning that both the real and the imaginary part provide a receptive profile .
this produces the simplified model @xmath84 this model allows also to include easily a characteristic of complex cells .
their response is indeed generally well described by a slight modification of the lnp model , called the _ _ energy model__@xcite , which consists of passing under the static nonlinearity not the linear filtering , but the square modulus of ( [ eq : rp ] ) , hence _ losing the phase _ : an input stimulus @xmath85 induces an activity on this type of neurons determined by a firing rate @xmath86={\mathscr{s}}\big(|\langle f , \psi_{\xi}\rangle_{{\mathbb{r}}^2_{\textnormal{retina}}}|^2\big ) .\vspace{-8pt}\ ] ] the model of v1 receptive profiles given by ( [ eq : rp ] ) is made of 6 parameters @xmath87 and @xmath88 , but v1 is only three dimensional , so only a subset of parameters is actually implemented .
the bijection due to the retinotopic mapping allows to consider the parameter @xmath80 to depend approximately ( up to the scatter ) only on the two horizontal cortical coordinates @xmath25 . on the other hand ,
the evidence for orientation hypercolumns shown in figure [ fig : bosking ] allows us to consider the orientation parameter @xmath26 , which depends essentially only on @xmath25 , to be a function of @xmath80 . with respect to the spatial frequency parameter @xmath78 ,
experimental evidence for a columnar structure is less clear , as weakly ordered maps are observed from intrinsic signals@xcite but their structure is debated@xcite . in any case
the system of gabor functions ( [ eq : rp ] ) can not make use of all frequencies at any point .
moreover , since the transversal dimension of v1 is finite , clearly also all scales can not be represented .
if one wants to consider the linear transformation performed by v1 receptive profiles , which is the main contribution to the feedforward voltage of cells , as a generalized wavelet transform of an image , some special characterization of the space of parameters actually present in v1 is needed in order to understand its functional behavior and its computational purposes . in order to reduce the problem , a useful simplification is that of disregarding the log - polar mapping and replace it with a linear mapping from the retina to the cortex .
this can be considered as a local approximation , which amounts to forget about the center - periphery difference in resolution and to consider the @xmath55 to be a constant . in order to keep a constant size for the cortical point image
, the dependency of scales should consequently be reduced to the vertical coordinate @xmath46 only . from now on we will then consider the topographic mapping to be simply the identity , so that a first set of interparametric dependencies is provided by @xmath89 where @xmath80 is then the independent coordinate on the cortical plane , which coincides with the image plane , while @xmath46 represents a transversal dimension defining the available scales over each position .
the dependence of @xmath78 is left for the moment free on @xmath80 and @xmath46 . moreover , by ( [ eq : pointimage ] ) , this implies that the point image can be approximately set to @xmath90 i.e. the receptive field size is considered to be given approximately by 4 standard deviations .
an additional interparametric dependence is provided by the observation that spatial frequency and scale appear to be highly correlated .
several experiments have been performed in this respect , here we will focus on data of orientation selective receptive profiles from cats fitted with a gaussian gabor model@xcite and on previously cited data on macaques@xcite , where the author has introduced the adimensional _ shape indices _ @xmath91 the index @xmath92 measures the number of oscillations under the spatial window , while the ratio between the two indicates the anisotropy of the two dimensional gaussian bell .
the relationship between spatial frequency @xmath78 and scale @xmath93 is shown in figure [ fig : josaa ] , left , as a scatter plot where each point represent a neuron . among the simplest model functions ,
one of the best fits is provided by @xmath94 ( red curve , @xmath95 root mean square error ) .
a pure inverse proportionality provides a slightly poorer fit ( green curve , 0.095 ) with @xmath96 which is about the mode of the distribution for the product @xmath97 ; this differs slightly from the mean value @xmath98 ( blue curve , 0.99 ) .
nevertheless , due to the shape indications provided by @xmath92 , it is relevant to observe in figure [ fig : josaa ] , center , how that shape index is distributed among the observed neurons .
these are cells that have been chosen for being selective to orientation .
this selection was not performed in the data displayed in figure [ fig : josaa ] , right , showing the relationship between the two shape indices .
one can see the actual presence of a significative amount of cells with poor orientation selectivity , corresponding to low values of @xmath92 . by computing the _ rotational uncertainty _ of receptive profiles ,
it is indeed possible to classify cells whose value of @xmath92 is smaller than @xmath99 as unselective with respect to orientations@xcite .
moreover , one can see that the distribution of @xmath92 in macaques is very similar to that found in cats , as was already pointed out@xcite .
we remark here that this correlation between receptive field size and spatial frequency , together with the variability of size with the transversal direction in the cortex , may be related to the apparently weak evidence of columnar organization of spatial frequencies@xcite .
( red ) , with @xmath100 ( green ) and with @xmath101 ( blue ) in cats@xcite .
center : histogram of @xmath92 ( cats@xcite ) .
right : shape indices distribution in macaques@xcite.,title="fig : " ] ( red ) , with @xmath100 ( green ) and with @xmath101 ( blue ) in cats@xcite .
center : histogram of @xmath92 ( cats@xcite ) .
right : shape indices distribution in macaques@xcite.,title="fig : " ] ( red ) , with @xmath100 ( green ) and with @xmath101 ( blue ) in cats@xcite .
center : histogram of @xmath92 ( cats@xcite ) .
right : shape indices distribution in macaques@xcite.,title="fig : " ] finally , it may be relevant to observe that the available data show evidence of a _
cutoff frequency_. moreover , while the actual sizes and frequencies may vary among different species , the adimensional shape indices show an apparent universal behavior which is common at least to cats and macaques ( and whose structure is not easily justified by means of standard optimization criteria @xcite ) , with an approximate upper bound @xmath102
the linear behaviors of receptive field can be studied with instruments of harmonic analysis , focusing on the properties of the wavelet decomposition they perform on images and on the group symmetries associated to their shapes .
this allows in particular to concretely formulate some open problems related to our visual system .
the set of receptive profiles can be fruitfully described , in an approximate way , in terms of the so - called _ _ affine weyl - heisenberg group__@xcite in two dimensions .
this is a semidirect product lie group possessing a combined structure of time - frequency , scales and rotations .
it reads @xmath103 , with composition law @xmath104 let us consider an approximate relation between the two scales as @xmath105 for a fixed value @xmath106 , ( figure [ fig : josaa ] , right , suggests a @xmath107 ; the case @xmath108 was treated in@xcite ) .
then , calling @xmath109 , denoting with @xmath110 and with @xmath111 the ordinary two dimesional dilation operator , we have that ( [ eq : rp ] ) reads @xmath112 now , @xmath113 is a projective unitary representation of @xmath114 , because @xmath115 the affine weyl - heisenberg group provides relevant symmetries for the analysis of multivariate data , as its structure contains as special cases the usual time - frequency , wavelet analysis and wave packets , as well as mixed behaviors @xcite .
it is well known that , in order to obtain resolutions of the identity , the whole group is too large and a proper subset needs to be considered . in the present case ,
a first reduction is provided by condition ( [ eq : section1 ] ) , which amounts to align the direction of modulations with the rotations parameter .
if the mother wavelet is isotropic ( i.e. @xmath108 ) , this is equivalent to the elimination of the rotations , whose effect is totally incorporated in the changing of orientations of the modulation parameter ( in ( [ eq : rfrepresentation ] ) simply commute @xmath73 with @xmath116 and use the rotational invariance of @xmath117 ) . for anisotropic mother wavelets ,
relation ( [ eq : section1 ] ) constrains the dilation axes to be aligned with the reference frame provided by the modulation vector @xmath118 .
a second reduction in the parameters space @xmath114 implemented by v1 can be ascribed to the approximate relation @xmath119 showed in figure [ fig : josaa ] , left , where the function @xmath120 represents the fitting curve while @xmath121 accounts for the discrepancies , which give rise to the variability in the distribution of @xmath92 .
the models for @xmath120 given by the inverse proportionality @xmath122 or by the rational dependence @xmath123 are well - known relationships in the affine weyl - heisenberg group@xcite .
this amounts to model the linear filtering as a usual wavelet analysis ( or a trivial modification of it ) with a two - dimensional morlet wavelet , whose role in image analysis for the detection of oriented features is well - known@xcite . since the data show that a relation @xmath124 is not implemented exactly , this leaves space for some additional fine adjustments : the proposed dependence on @xmath80 represented by the @xmath121 term is suggested by the mentioned spatial frequency maps@xcite . combining the constraints ( [ eq : section1 ] ) and
( [ eq : section2 ] ) with the ones given by ( [ eq : localfeatures ] ) , we can then approximately describe the parameter set @xmath3 representing classical receptive fields of v1 as the subset of @xmath114 given by @xmath125 \bigg\}\ ] ] where the function @xmath126 represents the orientation preference map and the function @xmath127 \to { \mathbb{r}}$ ] represents the variability of scales with the transversal cortical direction . with this approximation ,
the set of cortically implemented receptive profiles can then be written , for an appropriate parameter @xmath106 , as @xmath128 . while there are apparently no models concerning the function @xmath129 , which can probably be considered simply as a smooth bijection of @xmath130 $ ] onto some interval of scales @xmath131 $ ] , several models@xcite have been proposed for the map @xmath28 .
a common one@xcite is based on the observation that the power spectrum of orientation preference maps is approximately concentrated on a thin annulus , where it shows and apparently random behavior .
thus , a simple way to obtain quasi - periodic orientation preference - like structures is given by @xmath132 where @xmath133 \to \mathbb{c}$ ] is a highly irregular function ( eventually a random process ) and @xmath134 is the radius of the circle in the fourier plane , corresponding to the inverse of the typical correlation length , see figure [ fig : niebur ] . for what concerns spatial frequency maps and the term @xmath121 , despite some evidence@xcite of correlations with the map @xmath28 , they are still actively investigated@xcite . ) and an example of function @xmath135 on the circle , where the height represents the modulus and the color represents the phase.,title="fig : " ] ) and an example of function @xmath135 on the circle , where the height represents the modulus and the color represents the phase.,title="fig : " ] ) and an example of function @xmath135 on the circle , where the height represents the modulus and the color represents the phase.,title="fig : " ] this class of maps is well studied in optics@xcite , in the present context they have been related to the irreducible representations of the group of rotations and translations@xcite , which is the symmetry involved in the detection of orientations for the families ( [ eq : rp ] ) , and some of their geometric properties as random fields can be extended to other symmetries@xcite . by replacing the integral with a discrete sum ( [ eq : pinwheels ] ) , the associated quasi - periodic structure can also be obtained as results of nonlinear evolution equations@xcite for the optimization of the so - called _ coverage - continuity _ tradeoff , that is `` to preserve as far as possible neighborhood relations in parameter space''@xcite .
similar patterns were obtained by generative models via hebbian learning@xcite , and lateral connections seem to play a crucial role for their formation during animal development@xcite
. a currently , which has been scarcely addressed in its full complexity@xcite , concerns the properties of the system @xmath128 ( or of even more realistic models of the implemented family of receptive profiles ) considered as a wavelet analyzing family . in particular , it seems natural to ask what are its capabilities to represent images , or more in general what can be said about the function spaces associated to its linear decomposition .
this appears to be a basic step to understand also the computational role of the several nonlinearities that intervene in the neural code with different modulations of this first linear processing@xcite .
it would also provide some clearer indications on the primal information collected by sensory area v1 and later processed by the different types of connectivities . if one plans to perform a digital wavelet analysis in two dimensions with anisotropic wavelets
, a natural approach may seem that of using all orientations over each point , i.e. to consider a stack of different oriented local measurements and to repeat it over the whole signal .
however , visual cortex does not choose this strategy , since at each point @xmath80 of the image all the filters available ( by changing the transversal cortical coordinate ) have the same orientation , which changes as the ( horizontal ) position @xmath80 changes .
a notion that tries to quantify the completeness of the representation of orientations is that of _ _
coverage__@xcite , and an optimization with respect to it means requiring that all orientations be represented over each point , by means of the overlapping of adjacent receptive fields . in a quasi - periodic configuration like the one of orientation preference maps
, this condition implies that the length corresponding to the point image must be of the same order of the correlation length of these maps .
indeed , in that case all orientations would be represented in average within that distance , and all neurons lying within that distance would have receptive profiles that overlap at least partially .
this necessary condition for coverage can then be summarized , in terms of the discussed models , by the condition @xmath136 where @xmath134 is the parameter in ( [ eq : pinwheels ] ) and @xmath61 is given by ( [ eq : pi ] ) .
this condition was actually checked experimentally by observing the activity produced by a thin line@xcite , see figure [ fig : crowley ] .
such a stimulus can indeed activate neurons at a distance not larger than the point image , and only neurons selective to the same orientation will respond .
this experiment shows then that the point image is approximately of the same size as the distance between two cortical regions whose neurons have the same preferred orientation .
we observe also that condition ( [ eq : constraint ] ) is actually a constraint that links the maps @xmath137 \to [ \underline{s},\overline{s}]$ ] and @xmath4 .
selectivity of v1 neurons to local frequencies is often measured in terms of _ orientation tuning _ and of _ spatial frequency tuning _
curves , which quantify the decay of neural response for stimulus parameters that differ from the preferred one@xcite .
while this phenomenon can be partially explained in terms of classical receptive fields , it also offers a favorable experimental setting for checking for nonclassical effect , typically dynamical@xcite and nonlinear / nonlocal@xcite deformations .
here we simply observe that the qualitative behavior of these tuning curves is captured by the elements of the _ gram matrix _ of the system @xmath138 . since this is meant to be only a rough approximation of the experimental results , for simplicity we will consider @xmath139 ( i.e. @xmath108 ) , and denote with @xmath140 , @xmath141 , the receptive profiles .
the square modulus of the gram matrix for two elements at the same scale @xmath142 reads @xmath143 because ( calling @xmath144 ) @xmath145 the elements ( [ eq : gram ] ) can be considered as the energy of the response of a neuron to a stimulus shaped as a receptive profile .
since such a stimulus can provide the maximal linear response ( diagonal elements ) and is also parametrized by the features themselves , it offers a simple way to obtain tuning curves .
the typical experimental setup for obtaining tuning curves consists of showing stimuli that produce an electrophysiologically measured response that does not depend on the absolute position of the neuron and is @xmath146-periodic on the angular variable .
a simple way to obtain such a curve for a cell with preferred spatial frequency @xmath147 and preferred orientation @xmath148 is then via the following function @xmath149 it qualitatively resembles well known models@xcite and shows the role of spatial frequency in the orientation tuning width .
it is compared with actual measurements in cats@xcite in figure [ fig : webster ] .
the observed higher sharpness of the real tuning curves is thought to be related to nonlinear / nonlocal additional cortical mechanisms@xcite . ) for a typical cell @xmath150 , plotted with respect to @xmath78 and @xmath26 for different values of the other parameter ( normalized to have a maximum value 1 ) .
observe that @xmath78 is in logarithmic scale.,title="fig : " ] ) for a typical cell @xmath150 , plotted with respect to @xmath78 and @xmath26 for different values of the other parameter ( normalized to have a maximum value 1 ) .
observe that @xmath78 is in logarithmic scale.,title="fig : " ] ) for a typical cell @xmath150 , plotted with respect to @xmath78 and @xmath26 for different values of the other parameter ( normalized to have a maximum value 1 ) .
observe that @xmath78 is in logarithmic scale.,title="fig : " ] the elements ( [ eq : gram ] ) can also be used to obtain a _ relationship between the shape index cutoff and the characteristic length _ ( [ eq : constraint ] ) , by an argument introduced in@xcite concerning correlations . the function @xmath151 r.18 can indeed be considered as a measurement of _ energy autocorrelation _ in space and orientation for a receptive profile having spatial frequency @xmath78 and scale @xmath142 , denoting with @xmath152 its ( isotropic ) shape index . since
an orientation difference of @xmath153 provides the minimum correlation that can be attained due to the orientation selectivity mechanism only , it may be relevant to observe what is the spatial distance at which a receptive profile has the same correlation due to spatial decay only . by ( [ eq : correlations ] )
one obtains @xmath154 the dependency of the ratio @xmath155 on the shape index @xmath156 is shown in figure [ fig : shape ] .
it is possible to observe that , for a value of @xmath156 around @xmath157 , which coincides approximately with the cutoff bound ( [ eq : upperbound ] ) one obtains a distance that is approximately 4 times the scale , providing an effective size of a receptive profile corresponding to what is inside 2 standard deviations . by ( [ eq : pi ] ) , this was considered to correspond approximately to the point image , and by ( [ eq : constraint ] ) this length is apparently mapped on v1 as the characteristic length of the quasi - periodic map @xmath28 .
this can be interpreted as a principle of _ dimensionality reduction constrained to optimally independent representation of orientations _ , in the following sense .
as previously observed , v1 does not have a sufficiently high topological dimension to implement all scales / frequencies and orientations over each point , so it has adopted compromises such as the one described by the coverage strategy . on the other hand ,
if we measure the independence of receptive profiles in terms of correlations , then the maximal independence with respect to orientations can be obtained equivalently by translations at a given distance .
this distance can be estimated to grow with the shape index , as described by figure [ fig : shape ] , and reaches the characteristic length of the orientation preference maps at about the cutoff value for the shape index . at that distance , two receptive profiles
can then be considered as collecting a sufficiently independent information that justifies a repetition of a new full set of orientations .
say it from another point of view , orientation preference maps may be a way to map a compact variable on the highly redundant sampling space of positions .
however , spatial frequency variable is not compact , so in order to be projected down one needs a cutoff .
the one provided by ( [ eq : upperbound ] ) is consistent with the purpose of having maximal decorrelation at the distance corresponding to the quasi - periodicity of the map @xmath28 .
finally , we observe that frequency cutoff could be related to the finite resolution at the retinal level@xcite , but due to the intermediate optic nerve information compression and to the lgn preprocessing@xcite this dependence may be highly nontrivial .
even if much is known about the computations performed by primary visual cortex v1 , still many problems concerning the fine structure of these computations as well as their purposes in the processing of visual information are open .
this paper has focused on a restricted class of behaviors concerning classical receptive fields , which provide a primal linear response of v1 neurons to visual stimuli that is later processed by nonlinear mechanisms and by dynamical lateral and feedback connectivities .
many studies approach the computation implemented by v1 as being optimized for best information detection ( even if other purposes have been proposed , such as that of maximizing inference@xcite via bayesian@xcite schemes ) , and we have provided evidence that some of the design principles encountered in v1 are actually compatible with this view .
the presented approach focuses on the geometry of classical receptive fields , by considering their contribution to neural code as a linear wavelet analysis of visual stimuli . due to the prominent role of selectivity for local orientation ,
the substructure given by the group of translations and rotations has been often considered in relation with with the cortical two dimensional space - frequency analysis@xcite , as well as with the modeling of contour perception and of lateral connectivities@xcite , and it has provided a fruitful environment for cortical - inspired image processing on groups@xcite .
more in general , the geometry associated to v1 receptive profiles has provided a well - established set of models of the cortical architecture@xcite , and is thought to play a key role in the constitution of perceptual units@xcite . most of this geometry can be obtained by experimental setups that make use of parametric sets of simple stimuli , but the more recent approach to the study of receptive fields is actually based on the exploitation of nonlinearities and on the use of natural stimuli@xcite .
this provides indeed the framework for obtaining families of receptive field - like linear analyzers in terms of optimality criteria extracted from the statistics of natural images @xcite , and to reconcile some behaviors observed in v1 with recent approaches to high dimensional statistics , such as compressed sensing and sparsity@xcite . on the other hand ,
still much of the linear behavior continues to provide unsolved problems form the point of view of the strategies used by v1 for collecting informations .
we have mentioned that of characterizing the space of signals that can be represented by v1 classical receptive fields , but also the appearance of _ quasi - periodic structures _ in the space of coefficients of a gabor - like system seems to be suggestive in view of recent advances on similar approaches to sampling problems@xcite . moreover , the energy model for complex cells activity should be actually related to issues of _ _ phase retrieval__@xcite .
finally , the way nonlinearities intervene into the neural coding seems to be quite unusual with respect to the mainstream approaches to nonlinear approximation , as some of them seem to act mainly as _ deformations of the linear behavior _ , with frequent evidence of adaptivity mechanisms .
few theoretical studies in approximation theory deal with similar kinds of nonlinear behaviors@xcite , which however could provide effective alternative strategies from the ones presently considered in signal processing .
99 s. t. ali , j .-
antoine , j .-
gazeau , _ coherent states , wavelets and their generalizations_. springer , @xmath158 ed . 2014 .
a. afgoustidis , _ orientation maps in v1 and non - euclidean geometry_. j. math .
5:12 ( 2015 ) .
a. angelucci , j. b. levitt , e. j. s. watson , j .-
hup , j. bullier , j. s. lund , _
circuits for local and global signal integration in primary visual cortex_. j. neurosci .
22:8633 - 8646 ( 2002 ) .
a. angelucci , p. c. bressloff , _ contribution of feedforward , lateral and feedback connections to the classical receptive field center and extra - classical receptive field surround of primate v1 neurons_. prog .
brain res .
154:93 - 120 ( 2006 ) .
antoine , r. murenzi , p. vandergheynst , s. t. ali , _ two - dimensional wavelets and their relatives_. cambridge university press 2004 .
d. barbieri , g. citti , g. sanguinetti , a. sarti , _ an uncertainty principle underlying the functional architecture of v1_. j. physiol .-
paris 106:183 - 193 ( 2012 ) .
d. barbieri , g. citti , _ reproducing kernel hilbert spaces of cr functions for the euclidean motion group_. anal .
13:331 - 346 ( 2015 ) .
d. barbieri , g. citti , a. sarti , _ how uncertainty bounds the shape index of simple cells_. j. math .
4:5 ( 2014 ) .
d. barbieri , g. citti , g. cocci , a. sarti , _ a cortical - inspired geometry for contour perception and motion integration _ j. math .
imaging vision 49:511 - 529 ( 2014 ) .
m. v. berry , m. r. dennis , _ phase singularities in isotropic random waves_. proc .
lond . a 456:2059 - 2079 ( 2001 ) .
i. bojak , d. t. j. liley , _ axonal velocity distributions in neural field equations_. plos comput .
biol . 6:e1000653 ( 2010 ) .
w. h. bosking , y. zhang , b. schofield , d. fitzpatrick , _ orientation selectivity and the arrangement of horizontal connection in tree shrew striate cortex_. j. neurosci .
17:2112 - 2127 ( 1997 )
. w. h. bosking , j. c. crowley , d. fitzpatrick , _ spatial coding of position and orientation in primary visual cortex_. nature neurosci .
5:874 - 882 ( 2002 ) .
t. bonhoeffer , a. grinvald , _ optical imaging based on intrinsic signals : the methodology_. in `` brain mapping ; the methods '' .
a. w. toga and j. c. mazziotta ( eds . ) . academic press 1996 .
p. c. bressloff , _ waves in neural media_. springer 2014
. m. carandini et al . ,
_ do we know what the early visual system does ?
_ j. neurosci .
25:10577 - 10597 ( 2005 ) .
m. carandini , d. j. heeger , _ normalization as a canonical neural computation_. nature rev .
13:51 - 62 ( 2012 ) .
g. citti , a. sarti , _ a cortical based model of perceptual completion in the roto - translation space_. j. math .
imaging vision 24:307 - 326 ( 2006 ) .
g. citti , a. sarti , _ neuromathematics of vision_. springer 2014 .
g. citti , a. sarti , _ a gauge field model of modal completion_. j. math .
imaging vision 52:267 - 284 ( 2015 ) .
g. cocci , d. barbieri , a. sarti , _ spatiotemporal receptive fields of cells in v1 are optimally shaped for stimulus velocity estimation_. j. opt .
am . a 29:130 - 138 ( 2012 ) .
g. cocci , d. barbieri , g. citti , a. sarti , _ cortical spatiotemporal dimensionality reduction for visual grouping_. neural comput .
27:1252 - 1293 ( 2015 ) .
s. coombes , r. thul , k. c. a. wedgwood , _ nonsmooth dynamics in spiking neuron models_. physica d 241:2042 - 2057 ( 2012 ) .
m. c. crair , e. s. ruthazer , d. c. gillespie , m. p. stryker , _ ocular dominance peaks at pinwheel center singularities of the orientation map in cat visual cortex_. j. neurophysiol .
77:3381 - 3385 ( 1997 ) . p.
m. daniel , d. whitteridge , _ the representation of the visual field on the cerebral cortex in monkeys_. j. physiol .
159:203 - 221 ( 1961 ) .
p. dayan , l. f. abbott , _
theoretical neuroscience_. mit press 2001 .
g. c. deangelis , i. ohzawa , r. d. freeman , _ receptive - field dynamics in the central visual pathways_. trends neurosci .
18:451 - 458 ( 1995 ) .
r. l. de valois , k. k. de valois , _ spatial vision_. oxford university press 1990 .
r. duits , h. fhr , b. janssen , m. bruurmijn , l. florack , h. van assen , _ evolution equations on gabor transforms and their applications_. appl . comput . harmon .
35:483 - 526 ( 2013 ) .
r. durbin , g. mitchinson , _ a dimension reduction framework for understanding cortical maps_. nature 343:644 - 647 ( 1990 ) .
m. ehler , m. fornasier , j. sigl , _ quasi - linear compressed sensing_. multiscale model .
12:725 - 754 ( 2014 ) .
g. felsen , y. dan , _ a natural approach to studying vision_. nature neurosci .
8:1643 - 1646 ( 2005 ) .
d. fitzpatrick , _ seeing beyond the receptive field in primary visual cortex_. curr . opin .
10:438 - 443 ( 2000 ) .
a. k. fletcher , s. rangan , l. r. varshney , a. bhargava , _ neural reconstruction with approximate message passing_. adv .
neur . in .
24:2555 - 2563 ( 2011 ) .
p. fries , j .- h . schrder , p. r. roelfsema , w. singer , a. k. engel , _ oscillatory neuronal synchronization in primary visual cortex as a correlate of stimulus selection_. j. neurosci .
22:3739 - 3754 ( 2002 ) .
k. friston , _ a theory of cortical responses_. phil .
b 360:815 - 836 ( 2005 ) .
s. ganguli , h. sompolinsky , _ compressed sensing , sparsity , and dimensionality in neuronal information processing and data analysis_. annu .
35:485 - 508 ( 2012 ) .
j. l. gardner , a. a. anzai , i. ohzawa , r. d. freeman , _ linear and nonlinear contributions to orientation tuning of simple cells in the cat s striate cortex_. visual neurosci .
16:1115 - 1121 ( 1999 ) r. gattass , c. g. gross , j. h. sandell , _ visual topography of v2 in the macaque_. j. comp .
201:519 - 539 ( 1981 ) .
w. gerstner , w. m. kistler , _ spiking neuron models_. cambridge university press 2002 .
k. grill - spector , r. malach , _ the human visual cortex_. annu .
27:649 - 677 ( 2004 ) .
b. m. harvey , s. o. dumoulin , _ the relationship between cortical magnification factor and population receptive field size in human visual cortex : constancies in cortical architecture_. j. neurosci .
31:13604 - 13612 ( 2011 ) .
j. c. horton , d. l. adams , _ the cortical column : a structure without a function_. philos .
t. r. soc .
b 360:837 - 862 ( 2005 ) .
d. h. hubel , t. n. wiesel , _ uniformity of monkey striate cortex . a parallel relationship between field size , scatter and magnification factor_. j. comp . neurol .
158:295 - 306 ( 1974 ) .
d. hubel , eye brain and vision .
freely available at http://hubel.med.harvard.edu/ ( 1988 ) .
m. hbener , d. shoham , a. grinvald , t. bonheffer , _ spatial relationships among three columnar systems in cat area 17_. j. neurosci .
17:9270 - 9284 ( 1997 ) .
g. huguet , j. rinzel , j .-
hup , _ noise and adaptation in multistable perception : noise drives when to switch , adaptation determines percept choice_. j. vision 14:19 ( 2014 ) .
n. p. issa , c. trepel , m. p. stryker , _ spatial frequency maps in cat visual cortex_. j. neurosci .
20:8504 - 8514 ( 2000 ) .
s. jayasuriya , z. p. kilpatrick , _ effects of time - dependent stimuli in a competitive neural network model of perceptual rivalry_. b. math .
74:1396 - 1426 ( 2012 ) . c. kalisa , b. torresani , _
n - dimensional affine weyl - heisenberg wavelets_. ann .
i. h. poincar a 59:201 - 236 ( 1993 ) .
m. kaschube , m. schnabel , s. lwel , d. m. coppola , l. e. white , f. wolf , _ universality in the evolution of orientation columns in the visual cortex_. science 330:1113 - 1116 ( 2010 ) .
[ includes online material ] w. keil , f. wolf , _
coverage , continuity , and visual cortical architecture_. neural syst .
circuits 1:17 ( 2011 ) .
t. s. lee , _ image representation using 2-d gabor wavelets_. ieee t. pattern anal .
, 18:959 - 71 ( 1996 ) .
d. marr , t. poggio , e. hildreth , _ smallest channel in early human vision_. j. opt .
70:868 - 870 ( 1980 ) .
b. matei , y. meyer , _ quasicrystals are sets of stable sampling_. c. r. acad .
paris sr .
346:1235 - 1238 ( 2008 ) .
r. miikkulainen , j. a. bednar , y. choe , j. sirosh , _ computational maps in the visual cortex_. springer 2005 .
i. nauhaus , a. benucci , m. carandini , d. l. ringach , _ neuronal selectivity and local map structure in visual cortex_. neuron 57:673 - 679 ( 2008 ) .
e. niebur , f. wrgtter , _ design principles of columnar organization in visual cortex_. neural comput .
6:602 - 614 ( 1994 ) .
k. ohki , s. chung , p. kara , m. hbener , t. bonhoeffer , r. clay reid , _ highly ordered arrangement of single neurons in orientation pinwheels_. nature 442:925 - 928 ( 2005 ) .
b. a. olshausen , d. j. field , _ sparse coding with an overcomplete basis set : a strategy employed by v1 ? _ vision res .
37:3311 - 3325 ( 1997 ) .
j. ribot , y. aushana , e. bui - quoc , c. milleret , _ organization and origin of spatial frequency maps in cat visual cortex_. j. neurosci .
33:13326 - 13343 ( 2013 ) .
d. ringach , _ spatial structure and symmetry of simple cells receptive fields in macaque primary visual cortex_. j. neurophysiol .
88:455 - 463 ( 2002 ) .
d. l. ringach , m. j. hawken , r. shapley , _ dynamics of orientation tuning in macaque v1 : the role of global and tuned suppression_. j. neurophysiol .
90:342 - 352 ( 2002 ) .
a. romagnoni , j. ribot , d. bennequin , j. touboul , _ parsimony , exhaustivity and balanced detection in neocortex_. preprint , http://arxiv.org/abs/1409.4927 m. g. p. rosa , r. tweedale , _ maps of the visual field in the cerebral cortex of primates : functional organization and significance_. in `` the primate visual system '' , j. h. kaas , c. e. collins ( eds . ) .
crc press 2005 .
g. sanguinetti , g. citti , a. sarti , _ a model of natural image edge co - occurrence in the rototranslation group_. j. vision 10:37 ( 2010 ) .
a. sarti , g. citti , j. petitot , _ the symplectic structure of the primary visual cortex_. biol .
98:33 - 48 ( 2008 ) .
a. sarti , g. citti , _ the constitution of visual perceptual units in the functional architecture of v1_. j. comput .
38:285 - 300 ( 2015 ) .
m. p. sceniak , m. j. hawken , r. shapley , _ contrast - dependent changes in spatial frequency tuning of macaque v1 neurons : effects of a changing receptive field size_. j. neurophysiol .
88:1363 - 1373 ( 2002 ) .
u. sharma , r. duits , _ left - invariant evolutions of wavelet transforms on the similitude group_. appl .
39:110 - 137 ( 2015 ) .
y. shechtman , y. c. eldar , o. cohen , h. n. chapman , j. miao , m. segev , _ phase retrieval with application to optical imaging_. preprint , http://arxiv.org/abs/1402.7350 e. p. simoncelli , l. paninski , j. pillow , o. schwartz , _ characterization of neural responses with stochastic stimuli_. in `` the new cognitive neurosciences '' , @xmath159 ed .
, m. gazzaniga ( ed . ) .
mit press 2004 .
l. sirovich , r. uglesich , _ the organization of orientation and spatial frequency in primary visual cortex_. proc .
usa 101:16941 - 16946 ( 2004 ) .
n. v. swindale , _ orientation tuning curves : empirical description and estimation of parameters_. biol .
78:45 - 56 ( 1998 ) .
n. v. swindale , d. shoham , a. grinvald , t. bonhoeffer , m. hbener , _ visual cortex maps are optimized for uniform coverage_. nature neurosci .
3:822 - 826 ( 2000 ) .
d. teller , _ vision and the visual system_. notes , http://faculty.washington.edu/jpalmer/files/teller/ a. tonda , e. lutton , g. squillero , p .- h .
wuillemin , _ a memetic approach to bayesian network structure learning_. lect .
notes comput . sc .
7835:102 - 111 ( 2013 ) .
s. d. van hooser , j. a. f. heimel , s. chung , s. b. nelson , l. j. toth , _ orientation selectivity without orientation maps in visual cortex of a highly visual mammal_. j. neurosci .
25:19 - 28 ( 2005 ) .
w. e. vinje , j. l. gallant , _ sparse coding and decorrelation in primary visual cortex during natural vision_. science 287:1273 - 1276 ( 2000 )
. m. a. webster , r. l. de valois , _ relationship between spatial - frequency and orientation tuning of striate - cortex cells_. j. opt .
am . a 2:1124 - 1132 ( 1985 ) .
e. yacoub , n. harel , k. ugurbil , _ high - field fmri unveils orientation columns in humans_. proc .
usa 105:10607 - 10612 ( 2008 ) .
s. w. zucker , _ which computation runs in visual cortical columns ? _ in `` 23 problems in systems neuroscience '' , j. l. van hemmen , t. j. sejnowski ( eds . ) .
oxford university press 2006 . | some geometric properties of the wavelet analysis performed by visual neurons are discussed and compared with experimental data . in particular ,
several relationships between the cortical morphologies and the parametric dependencies of extracted features are formalized and considered from a harmonic analysis point of view . |
although microlensing was originally proposed by paczyski ( 1986 ) as a method to detect massive astrophysical compact halo objects ( machos ) , it can also be used in various other fields of astronomy .
one important application is searching for extra - solar planets .
photometrically , planets can be found via the distortions they create in the lensing light curve compared to the smooth and symmetric one of a single lens event ( mao & paczyski 1991 ; gould & loeb 1992 ; bolatto & falco 1994 ; wambsganss 1997 ) . due to the short duration ( @xmath3 hour
@xmath3 day ) of the planet - induced deviations , it is difficult to detect them from survey type experiments ( e.g. , ogle : udalski et al .
1993 ; macho : alcock et al . 1993
; eros : aubourg et al .
1993 ; duo : alard & guibert 1997 ) .
however , with sufficiently frequent and accurate observations from follow - up monitoring of ongoing events alerted by the survey experiments , it is possible to detect such perturbations and characterize the mass ratio and the projected separation of the planet .
currently , several groups are carrying out such observations ( mps : rhie et al .
2000 ; planet : albrow et al . 1998 ; moa : bond 2000 ) .
it was recently shown by safizadeh , dalal & griest ( 1999 ) that planets can also be detected from astrometric follow - up observations of lensing events by using the next generation of high - precision interferometers such as the _ space interferometry mission _ ( sim , unwin et al .
1997 ) and those to be mounted on the keck ( colavita et al .
1998 ) and the vlt ( mariotti et al . 1998 ) .
precise astrometry using these interferometers will permit measurements of the displacements in the source star image center of light with respect to its un - lensed position , @xmath4 ( miyamoto & yoshii 1995 ; h-1pt g , novikov , & polnarev 1995 ; walker 1995 ; boden , shao , & van buren 1998 ; miralda - escud 1996 ; paczyski 1998 ) .
the trajectory of the light centroid shifts ( astrometric curve ) caused by a single point mass is an ellipse ( walker 1995 ; jeong , han & park 1999 ) .
a planet can be detected because it can distort the elliptical astrometric curve of the single lens event , which is analogous to the deviation in the light curve . in this paper , we investigate the properties of the planet - induced deviations in the astrometric curves . for this
, we construct a vector field map of the excess centroid shift @xmath0 , which represents the difference of @xmath4 from that expected in the absence of the planet as a function of source positions . with the map ,
we examine in detail the correlations between the astrometric and photometric deviations , which were briefly discussed by safizadeh et al .
if a source located at @xmath5 is gravitationally lensed by a coplanar @xmath6 point - mass lens system , where the individual masses and locations are @xmath7 and @xmath8 , the positions of the resulting images @xmath9 are obtained by solving the lens equation of the form @xmath10 where @xmath11 is the total mass of the lens system and @xmath12 is the angular einstein ring radius . the einstein ring radius is related to the physical parameters of the lens system by @xmath13 where @xmath14 and @xmath15 represent the distances to the lens and source from the observer , respectively . for the case of a single point - mass lens ( @xmath16 )
, there are two solutions ( and thus the same number of images ) of the lens equation and the resulting total magnification and the centroid shift vector take the simple forms @xmath17 @xmath18 where @xmath19 is the dimensionless lens - source separation vector normalized by @xmath12 .
the separation vector is related to the single lensing parameters by @xmath20 where @xmath21 is the einstein ring radius crossing time ( einstein time scale ) , @xmath22 is the closest lens - source separation ( impact parameter ) , @xmath23 is the time at the moment of the closest approach , and the unit vectors @xmath24 and @xmath25 are parallel and normal to the direction of the lens - source transverse motion , respectively .
note that the sign ` @xmath26 ' is used because @xmath22 is positive definitive .
the lens system with a planet is described by the formalism of the binary lens system ( i.e. , @xmath27 ) with a very low mass companion . for this case ,
the lens equation becomes a fifth degree polynomial equation ( witt & mao 1995 ) and there are three or five solutions ( images ) depending on the source position with respect to the lenses .
the magnifications of the individual images are given by the jacobian of the transformation ( 1 ) evaluated at the image position , i.e.@xmath28 then , the total magnification is given by the sum of the magnifications of the individual images , i.e. @xmath29 , where @xmath30 is the total number of images . since the position of the image centroid , @xmath31 , equals the magnification weighted mean position of the individual images , the source image centroid shift with respect to the unlensed source position
is computed by @xmath32 the fundamental difference in the geometry of a binary lens system from that of a single point - mass lens is the formation of caustics .
the caustics are the set of source positions at which the magnification of a point source diverges in the geometric optics limit , i.e. @xmath33 .
hence , the most significant planet - induced deviations both in the light and astrometric curves occur when the source passes the region close to the caustics .
the set of caustics form closed curves .
the maximum size of the caustics occurs when the planet is located in the lensing zone of @xmath34 ( gould & loeb 1992 ) , where @xmath35 is the projected primary - planet separation in units of @xmath12 ( planetary separation ) .
we investigate the properties of planet - induced deviations in the astrometric curves by constructing maps of excess centroid shifts .
the excess centroid shift is defined by @xmath36 where @xmath37 and @xmath4 represent the centroid shifts with the presence and absence of the planet , respectively .
therefore , the excess centroid shift vector tells us how much the astrometric curve of an event caused by a lens system with a planet deviates from the elliptical one of a single point - mass event and in which direction the deviation is directed .
the excess centroid shift map is analogous to the excess magnification map that was often used to estimate the probability of photometric planet detections ( gould & loeb 1992 ; gaudi & gould 1997 ; gaudi & sackett 2000 ) . the excess magnification is defined by @xmath38 where @xmath39 and @xmath40 are the magnifications with and without the planet , respectively .
we note , however , that @xmath0 is a vector quantity , while @xmath41 is a scalar quantity .
hence the excess centroid shift map is a vector field map . with the map
, one can obtain an overview of the astrometric behaviors of planet - induced perturbations of all events caused by a lens system , without testing their individual astrometric curves ( han 2001 ) .
for the construction of the map , the excess centroid shift on each source position is determined by calculating @xmath4 and @xmath37 .
we calculate @xmath37 by numerically solving the binary lens equation , i.e. by solving the fifth degree polynomial equation ( witt & mao 1995 ) .
as will be shown in the following subsections , the properties of astrometric perturbations are greatly different between the systems with @xmath1 and @xmath2 .
hence , we separately investigate the properties of @xmath0 for these two types of systems . for this type of lens system ,
there exist two disconnected sets of caustics .
one is located near the center of mass ( central caustic ) and the other caustic ( planetary caustic ) is located between the primary and the planet ( griest & safizadeh 1998 ; bozza 1999 ) . in the upper panel of figure 1 ,
we present the excess centroid shift map of an example lens system composed of a planet with @xmath1 .
the grey - scales in the map are used to represent the regions of significant deviations with @xmath42 ( dark shaded region ) and moderate deviations with @xmath43 ( light shaded region ) .
the angular einstein ring radius of a galactic bulge event caused by a stellar mass lens with @xmath44 located midway between the observer and the source , i.e. @xmath45 , is @xmath46 @xmath47-arcsec . for this event , the deviations in these regions respectively are @xmath48 @xmath47-arcsec and @xmath49 @xmath47-arcsec . in figure 2 and 3 , we present the light and astrometric curves of the events resulting from the source trajectories marked in the map ( long straight lines with arrows ) .
the assumed planetary separation is @xmath50 and the planet / primary mass ratio is @xmath51 , which is roughly equivalent to the mass ratio between the jupiter and the sun . to show how the astrometric deviation is correlated to the photometric one
, we also present the excess magnification map ( the contour map in the second panel of fig . 1 ) and the curves of @xmath41 and @xmath0 around the time and near the region of significant deviations ( figure 4 ) .
the contours of the excess magnification map are drawn at the levels of @xmath52 , @xmath53 , 5% , and 10% and the regions of positive excesses are distinguished by grey - scales . from the analysis of the figures , we find the following properties of @xmath0 induced by planets with separations from the primary larger than @xmath12 . 1 .
the most significant astrometric deviations ( @xmath54 ) occur in the region near the caustics and also in the region close to the primary - planet axis between the two caustics .
since the planetary caustic for this system is located on the planet side with respect to the center of mass , the major deviation region is also located on the planet side .
the biggest deviations occur when the source crosses the caustics ( e.g. , the event caused by the source trajectory 3 ) . for this case , the resulting astrometric curve exhibits a sharp spike due to the sudden shift of the centroid ( see the corresponding @xmath4 and @xmath0 of the event in fig . 3 and 4 ) .
for caustic crossing events , see 3.3 of han ( 2000 ) .
] even for non - caustic crossing events , however , significant deviations can also occur if the source trajectory passes through the primary - planet axis region between the central and planetary caustics . for this case
, the deviation is smooth as shown in fig . 3 and the right panels of fig . 4 . 2
the deviation vectors in most of the major deviation region are directed towards the planet along the primary - planet axis . as a result ,
a majority of the events caused by a planetary lens system with @xmath1 will have major deviations oriented towards the planet .
the only exception occurs if the source passes through the small region on the opposite side of the primary lens ( e.g. , the event caused by the source trajectory 6 ) , within which the major deviation vector is directed away from the planet .
3 . as the distance from the primary - planet axis increases , the amount of deviation decreases .
in addition , the direction of @xmath0 is reversed compared to that of @xmath0 in the major deviation region . since the major deviation region
is surrounded by the region of moderate deviations , the astrometric curve of a typical event with significant deviations will have smaller deviations with an opposite orientation to that of the major deviation vectors before and after the major deviation .
we note that planet - induced photometric deviations also exhibit a similar property , i.e. having smaller deviations with an opposite sign of @xmath41 around the major deviation ( gaudi & gould 1992 ) .
the photometric and astrometric deviations are closely correlated , as discussed by safizadeh et al .
( 1999 ) . from the comparison of the sizes between the photometric and the corresponding astrometric deviations
, one finds that the amount of @xmath55 increases as @xmath56 increases . from the comparison of the sign of @xmath41 and the direction of @xmath0 ,
one also finds that @xmath0 is directed towards the planet when @xmath57 ( positive deviation ) and vice versa .
since the major astrometric deviation vectors of the events caused by this type of lens system are usually directed towards the planet , the light curves of these events are more likely to have positive major deviations ( see fig . 2 and the right panels of fig . 4 )
the location of the major astrometric deviation regions and the orientations of @xmath58 in this region can be understood from the correlations between @xmath0 and @xmath41 .
the most significant photometric deviations occur when the source is located near the primary - planet axis because at this moment the images are located near the axis and approaches most closely the planet ( gould & loeb 1992 ) . as a result
, the region of major astrometric deviations is located along the axis and @xmath0 in this region is directed also along the axis .
in addition , the similarity in the properties of moderate size deviations between the astrometric and photometric perturbations around the major deviations mentioned above can also be understood from the correlations between @xmath0 and @xmath41 .
the lens system composed of a planet with @xmath2 also has a single central caustic , but it has _ two _ planetary caustics , which are located on the _ opposite _ side of the planet with respect to the center of mass . unlike the planetary caustic of the planetary lens system with @xmath1 ,
the planetary caustics of this lens system are _ not _ located on the primary - planet axis , although they are symmetric with respect to the axis ( griest & safizadeh 1998 ) . in the upper panel of figure 5 , we present the map of @xmath0 for a planetary lens system with @xmath59 and @xmath51 and it is compared to the excess magnification map presented in the second panel .
also presented in figure 6 and 7 are the light and astrometric curves of several example events resulting from the source trajectories marked in the upper panel of fig . 5 . in figure 8 , we present the photometric and astrometric deviations around the time and near the region of significant deviations for these events . from the investigation of the properties of @xmath0 induced by the planet with @xmath2 and comparison to those of the deviation vectors induced by the planet with @xmath1 , we find the following similarities and differences . 1 . for this system , the most significant deviations also occur near the caustic regions and the region close to the primary - planet axis between the two types of caustics .
however , since the planetary caustics are located on the opposite side of the planet with respect to the center of mass , the significant deviation regions are also located on the opposite side of the planet .
2 . in addition , besides the small areas around the planetary caustics , @xmath60 in most of the major deviation region are directed away from the planet along the primary - planet axis ( see fig . 7 and right panels of fig .
3 . similar to the behaviors of @xmath0 induced by the planet with @xmath1 , the major deviation region
is surrounded by the region of moderate deviations , within which the direction of @xmath0 is reversed to that of @xmath0 in the major deviation region .
the relation between the astrometric and photometric deviations is also similar to the case of the planetary system with @xmath1 ; the amount of astrometric deviation increases as @xmath61 increases and its direction points towards the planet when the light curve has a positive deviation and vice versa .
because the major astrometric deviations induced by this type of planets are usually directed towards the opposite side of the planet , the light curves have negative deviations ( @xmath62 ) with the same probability .
another interesting finding is that most of the major planet - induced astrometric deviations point _
away _ from the opening of the astrometric ellipse regardless of the planetary separation ( see fig . 3 and 7 ) .
in this section , we investigate the variation of the excess centroid shift patterns with the planetary separation and the planet / primary mass ratio . for this
, we present excess centroid shift maps of lens systems with various values of @xmath35 and @xmath63 .
in addition , to examine the effect of finite source size on the pattern of @xmath0 , we present maps of an example lens system expected for different source sizes . in figure 9 , we present the locations of significant astrometric deviation regions for lens systems with various values of @xmath35 . in the following figures 10 and 11 , we present the vector field maps of @xmath0 of the corresponding lens systems in the regions around the significant deviations .
the tested lens systems have a common mass ratio of @xmath51 and a variety of planetary separations of @xmath64 , 1.7 , 1.5 , 1.3 ( whose maps are presented in fig .
10 ) , 1.0 , 0.9 , 0.7 , and 0.5 ( whose maps are presented in fig .
one finds that similar to photometric deviations ( gould & loeb ) , the size of significant deviation regions is maximized when @xmath65 and becomes smaller as the separation increases or decreases from this value .
we note , however , that the general patterns of @xmath0 described in 4.1 and 4.2 still apply . in figure 12 and 13 ,
we present the excess centroid shift maps of lens systems with various values of planet / primary mass ratios .
the tested systems have a common planetary separation of @xmath66 and various mass ratios of @xmath67 , @xmath68 , @xmath69 , @xmath70 ( whose maps are presented in fig .
12 ) , @xmath71 , @xmath70 , @xmath72 , and @xmath73 ( whose maps are presented in fig .
one finds that the location of the major astrometric deviation region does not vary with mass ratio , but the size of the region decreases rapidly as @xmath63 decreases .
this property is also similar to that of photometric deviations . in figure 14
, we presented the excess centroid shift maps of a common lens system ( with @xmath50 and @xmath51 ) but with different source sizes .
considering the finite source effect , the position of the image centroid is given by @xmath74 where @xmath75 is the surface intensity distribution of the source star and @xmath76 represents the displacement vector of an area element on the source star surface with respect to the lens .
for the construction of the maps , we assume the distribution of the source star surface intensity is uniform .
we note that @xmath0 in the maps are the deviations from the centroid shifts of the single primary lens which are also affected by the finite source effect . when a single lens event is affected by the finite source effect , the resulting centroid shift trajectory is no longer an ellipse ( mao & witt 1998 ) .
we test three different source sizes of @xmath77 , @xmath78 , and @xmath79 . from the figure
, one finds that the detailed structure of @xmath0 is smeared out with the increasing source size , as pointed out by safizadeh et al .
for example planetary microlensing astrometric curves affected by the finite source effect , see fig . 3 of safizadeh et al .
we investigate the astrometric properties of planet - induced deviations in the trajectory of the microlensed source star s centroid shift motion . for this purpose , we construct maps of excess centroid shifts from which one can obtain an overview about the behaviors of @xmath0 of all events without testing their individual centroid shift trajectories . from the analysis of the maps for various types of lens systems , we find that major astrometric deviations occur not only in the region around caustic but also in the region close to the primary - planet axis between caustics .
the region of major deviations is surrounded by the region of moderate deviations in which the orientation of @xmath0 is reversed compared to that in the major deviation region . due to the difference in the locations of the caustics between the planetary lens systems with @xmath1 and @xmath2 , the locations of the major deviations and the orientations of @xmath0 within these regions of the two types of lens systems are different from each other . from the investigation of the correlations between the astrometric and photometric deviations
, we find that both the size and the direction of the astrometric deviation are closely related to the magnitude and sign of the photometric deviation , as discussed by safizadeh et al .
( 1998 ) . by investigating the variation of the pattern of @xmath0 with the planetary separation , planet / primary mass ratio , and source size , we find the astrometric and photometric deviations both have similar dependence upon these parameters .
alard c. , guibert j. , 1997 , , 326 , 1 albrow m. d. , et al .
, 1998 , , 509 , 684 alcock c. , et al .
1993 , nat , 365 , 621 aubourg e. , et al . , 1993 , nat , 365 , 623 boden a. f. , shao m. , van buren d. , 1998 , , 502 , 538 bolatto a. d. , falco e. e. , 1994 , , 436 , 112 bond i. , 2000 , in asp conf .
microlensing 2000 : a new era of microlensing astrophysics , eds . j. w. menzies and p. d. sackett , in press bozza v. , 1999 , , 348 , 311 colavita m. m. , et al . , 1998 , proc .
spie , 3350 - 31 , 776 gaudi b. s. , gould a. , 1997 , , 486 , 85 gaudi b. s. , sackett p. d. , 2000 , , 528 , 56 gould a. , loeb a. , 1992 , , 396 , 104 griest a. , safizadeh n. , 1998 , , 500 , 37 han c. , 2001 , , 325 , 1281 h-1pt g e. , novikov i. d. , polnarev a. g. , 1995 , , 294 , 287 jeong y. , han c. , park s .- h . , 1999 , , 511 , 569 mao s. , paczyski b. , 1991 , , 374 , l37 mao s. , witt h. j. , 1998 , , 300 , 1041 mariotti j. m. , et al . , 1998 , proc .
spie , 3350 - 33 , 800 miralda - escud j. , 1996 , , 470 , l113 miyamoto m. , yoshii y. , 1995 , , 110 , 1427 paczyski b. , 1986 , , 304 , 1 paczyski b. , 1998 , , 494 , l23 rhie , s. h. , et al .
, 2000 , , 533 , 378 safizadeh n. , dalal n. , griest k. , 1999 , , 522 , 512 udalski a. , szymaski m. , kaluzny j. , kubiak m. , krzemiski w. , mateo m. , preston g. w. , paczyski b. , 1993 , acta astron . , 43 , 289 unwin s. , boden a. , shao m. , 1997 , in aip conf .
387 , space technology and applications international forum 1997 , ed .
m. s. el - genk ( new york : aip ) , 63 walker m. a. , 1995 , , 453 , 37 wambsganss j. , 1997 , , 384 , 172 witt h. j. , mao s. , 1995 , , 447 , l105 | an extra - solar planet can be detected by microlensing because it can distort the smooth lensing light curve created by the primary lens . as a new method to search for and characterize extra - solar planets , safizadeh , dalal & griest proposed to detect the planet - induced distortions in the trajectory of the microlensed source star s centroid motion ( astrometric curve ) , which is observable by using the next generation of high - precision interferometers . in this paper , we investigate the properties of the planet - induced deviations in the astrometric curves ( excess centroid shifts @xmath0 ) and the correlations of @xmath0 with the photometric deviations .
for this , we construct vector field maps of @xmath0 , which represent the difference of the centroid shift from that expected in the absence of the planet as a function of source positions . from this investigation
, we find that significant astrometric deviations occur not only in the region near the caustics but also in the region close to the planet - primary axis between the caustics .
however , due to the difference in the locations of the caustics between the two types of systems with the planet - primary separations ( normalized by the angular einstein ring radius ) @xmath1 and @xmath2 , the locations of the major deviation regions of the two systems are different each other . for systems with @xmath1 , the major deviation vectors have orientations , in most cases , pointing towards the planet , while they point away from the planet for systems with @xmath2 .
the major deviation region is surrounded by the region of moderate deviations , within which the orientation of @xmath0 is reversed compared to the orientation in the major deviation region .
we also find that the astrometric deviation is closely correlated with the photometric one as discussed by safizadeh et al .
the astrometric deviation increases as the photometric deviation increases and @xmath0 is directed towards the planet when the light curve has positive deviation and vice versa .
we also present excess centroid shift maps for lens systems with various values of the planetary separation , planet / primary mass ratio , and source size to show the changes in the pattern of @xmath0 with these parameters .
epsf [ firstpage ] gravitational lensing planets and satellites : general |
the triangulum - andromeda ( triand ) star cloud was identified as a low - latitude overdensity of m giant stars in the two micron all - sky survey ( 2mass ) by rocha - pinto et al .
( 2004 , hereafter rp04 ) and as a strong sequence of main sequence turn - off ( msto ) stars by majewski et al .
( 2004 , hereafter m04 ) .
rp04 and m04 found the structure to extend across at least @xmath0 of the halo at a heliocentric distance of 15 - 30 kpc .
the mean heliocentric radial velocity of the structure is @xmath1 km s@xmath2 , and its mean [ fe / h ] is @xmath3 dex based on calcium infrared triplet measurements by rp04 . though a subsequent megacam survey of msto stars in this sky region ( martin , ibata , & irwin 2007 ) confirmed a @xmath4 kpc distance for this stellar structure , to date little additional attention has been given to triand , including determining its properties at any level of detail useful for ascertaining the origin of the structure . in contrast , significantly more attention has been given to the galactic anticenter stellar structure ( gass ) , also called the `` monoceros stream '' or `` monoceros ring '' , which was discovered only two years before triand as several overdensities of presumed msto stars ( newberg et al .
2002 ) in the same general region of the galactic anticenter as triand but at a closer mean @xmath5 kpc heliocentric distance .
subsequent study of this ring - like structure with the isaac newton telescope wide field camera ( ibata et al .
2003 ; conn et al . 2005 ) , sloan digital sky survey spectroscopy and photometry ( yanny et al .
2003 , hereafter y03 ) , and as traced by 2mass m giants ( crane et al .
2003 , hereafter c03 ; majewski et al .
2003 , hereafter m03 ; rocha - pinto et al .
2003 , hereafter rp03 ) has shown that the low latitude gass ring spans at least the second and third galactic quadrants and has a wide metallicity spread , from [ fe / h]= @xmath6 ( y03 ) to @xmath7 ( c03 ) .
nevertheless , the origin of gass remains controversial .
principal scenarios include those where gass is a piece / warp of the galactic disk ( momany et al .
2006 ; kazantzidis et al . 2008 ; younger et al . 2008 ) versus those where it represents tidal debris from the disruption of a milky way ( mw ) satellite galaxy ( ibata at al . 2003 ; y03 ; c03 ; rp03 ; helmi et al . 2003 ; frinchaboy et al .
2004 ; martin et al . 2004 ; pearrubia et al . 2005 ; conn et al . 2005 ; rocha - pinto et al . 2006 ) .
a comparison of chemical patterns between the gass and other dwarf spheroidal ( dsph ) galaxies has recently been made in chou et al .
( 2010b , hereafter c10b ) .
their gass sample reveals similar abundance patterns for ti , y and la as in the sagittarius dwarf+stream system and other satellite galaxies , which suggests that stars in gass likely formed in a dwarf galaxy - like environment .
however , this chemical observation does not in itself conclusively resolve the debate over the _
dynamical _ origin of gass because theoretical models show that galactic disks grow outward by accretion of dwarf galaxies ( e.g. , abadi et al .
2003 ; yong , carney , & teixera de almeida 2005 ) , and whether gass represents an intact tidal stream versus some other galactic structure generated from a previously hierarchically - formed , outer disk is not clarified by this chemistry . the lack of a definitive origin model for gass notwithstanding , soon after the triand discovery it was proposed by pearrubia et al .
( 2005 ) that this structure may simply be an outlying piece of gass / mon , which these authors hypothesized to be a structure formed from the dissolution of a mw satellite galaxy .
both structures are at similar galactic latitudes , and pearrubia et al . were able to find an @xmath8-body simulation that could incorporate triand as a dynamically older , earlier - stripped piece from the putative mon progenitor galaxy , under the assumption that the position of this progenitor is in canis major ( the latter aspect motivated by the claim of a canis major overdensity interpreted as the core of a disrupting dsph by martin et al .
2004 itself a greatly debated issue ; see rocha - pinto et al .
2006 , mateu et al .
2009 , and references therein ) .
if triand and gass / mon are parts of the same dwarf galaxy system , then the stars in these two structures share a common chemical enrichment history and should in principle show similar enrichment patterns e.g. , those seen by c10b for gass / mon . in effect
, one could apply `` chemical fingerprinting '' to determine whether the stars in triand belong to the gass / mon system .
this technique was recently applied by chou et al .
( 2010 , hereafter c10a ) to show that a group of m giant stars in the north galactic hemisphere ( the `` ngc moving group '' ) are likely older pieces of the sagittarius ( sgr ) tidal stream .
a comparison to the sgr system is quite apropos in the study of mon and triand , since sgr exhibits many similar properties to both , including a relatively high metallicity with the presence of m giant stars in all three cases ( m03 ; rp03 ; rp04 ) that suggest the progenitor systems of all three substructures may have been somewhat similar .
therefore , following the above precedents from our previous chemical studies of halo substructure , in this paper we make the first high resolution spectroscopic assessment of the chemical abundance patterns in the triand system , and use these abundance measurements to investigate the hypothesized connection between the triand and gass / mon systems .
our analysis of triand stars is identical to that used for sgr and gass m giants described in c07 , c10a and c10b .
we use spectra from echelle spectrographs mounted on the apache point observatory 3.5-m and kitt peak national observatory 4-m telescopes , with resolutions @xmath9 and 35,000 , respectively .
we measure equivalent widths ( ews ) of eleven lines , two lines , and one line , and use spectral synthesis for one line in the spectral region 7440 - 7590 previously studied by smith & lambert ( 1985 ; 1986 ; 1990 ) .
we use the lte code moog ( sneden 1973 ) to derive abundances of these elements following the same procedures as in c07 , c10a and c10b , and adopt the same model atmospheres , generated by linear interpolation from the grids of atlas9 models in castelli & kurucz ( 2003 ) .
our analysis used their odfnew models without convective overshooting .
we observed six triand m giants selected to have the photometric parallax and radial velocity of that system from the sample described in rp04 .
observed gass stars range from @xmath10 , while the triand stars span a narrower , @xmath11 range .
nevertheless , both samples have the same @xmath12 blue limit and about the same intrinsic rgb luminosity , as evidenced by a comparison to the fitted isochrones ( see 3 ) .
the primary difference is that the gass sample includes more very red , presumably asymptotic giant branch ( agb ) stars so that the red limits are @xmath13 for triand and @xmath14 for gass ; this yields mean @xmath15 colors of 1.05 and 1.09 dex for the full triand and gass samples , respectively .
the apparent @xmath16 magnitudes and dereddened @xmath15 colors ( schlegel et al .
1998 ) for our target triand stars are listed in table 1 .
the same information for gass stars is found in c10b .
the derived abundances are also summarized in table 1 .
table 1 columns give the derived effective temperature using the houdashelt et al .
( 2000 ) color - temperature relation applied to the 2mass @xmath15 color , and the derived values of the surface gravity ( @xmath17 ) , microturbulence ( @xmath18 ) , abundance @xmath19(x ) , and abundance ratios [ fe / h ] or [ x / h ] for each element x as well as the standard deviation in the abundance determinations .
the latter represent the line - to - line scatter ( for fe , ti and y ) for the ew measures , or different continuum level adjustments ( for la ) , as discussed in c07 and c10a .
uncertainties in abundance ratios are derived as in c10b .
briefly , we account for both uncertainties in measured ews as well as propagated uncertainties from the adopted stellar parameters .
the latter mainly come from the color - calibrated effective temperatures and imprecise knowledge of the proper isochrones to use for each star .
the net uncertainties are estimated to be 0.14 , 0.20 , 0.19 , 0.16 for [ fe / h ] , [ ti / fe ] , [ y / fe ] and [ la / fe ] , respectively .
figure 1 shows color - magnitude diagrams ( cmds ) of the triand and gass m giant candidates . based on our derived [ fe / h ] for each star ( indicated by the color scale )
we roughly matched these distributions with a set of marigo et al .
( 2008 ) isochrones for 8 gyr , [ @xmath20/fe]=0 , rgb / agb stars . the choice of 8 gyr matches the approximate age of sgr stars of [ fe / h ] @xmath21 ( siegel et al .
2007 ) ; analogy with sgr is the only information available to inform such a decision for the systems of interest here .
the corresponding isochrone distances are @xmath2212 kpc ( gass ) and @xmath2222 kpc ( triand ) ; these are consistent with previous estimates of the distances to these structures .
however , due to the age - metallicity degeneracy of rgb isochrones , other combinations of age and metallicity also fit reasonably well e.g. , 5 gyr old isochrones at 25 kpc and 14 kpc also fit the triand and gass distributions , respectively . the wider cmd spread of gass stars having similar metallicities to each other ( e.g. , [ fe / h]@xmath22@xmath23 ) suggests a potentially larger relative distance spread ( @xmath2211 - 22 kpc ) for gass stars compared to triand if we assume tight , monotonic age - metallicity relations in the two systems .
figure 2 compares derived distributions of [ ti / fe ] , [ y / fe ] , and [ la / fe ] as a function of [ fe / h ] for our triand , gass / mon ( from c10b ) and sgr stars ( from c07 , c10a ) .
a strength of the following comparison of triand , gass / mon and sgr stars is that not only are our abundances for all three systems derived homogeneously using the same methodology ( same lines , same model atmospheres , same derivation of stellar parameters , etc . ) , but the stars selected for comparison span a very narrow range of stellar atmospheric parameters ( @xmath24k in temperature and @xmath25 dex in @xmath17 ) so that the lines under study are forming under similar atmospheric conditions .
linear fits to the mw distributions ( see c10a ) are also shown in figure 2 .
in contrast to the general chemical similarity of gass and sgr ( except for [ la / fe ] at the highest metallicities ; see c10b ) , we note some larger differences between triand and these other two systems ( fig .
2 ) . while the mean trend of [ ti / fe ] versus [ fe / h ] for the triand stars is somewhat similar to the mean trends for sgr and gass ( though the three of six sampled triand stars at [ fe / h ] @xmath26 fall well below stars of similar metallicity in sgr and gass ) ,
the s - process element patterns reveal more obvious differences .
the mean level of [ y / fe ] ( middle panel ) for triand stars is about 0.2 dex higher ( or about 2.5 times the error in the mean for triand ) than that for either sgr or gass stars at the same [ fe / h ] , and almost at the solar enrichment level and near the mean for mw stars . and
while the [ la / fe ] pattern of triand stars closely resembles that of sgr stars ( including one star at [ fe / h ] = @xmath27 with enhanced [ la / fe ] , hinting at a possible upturn at high metallicity ) , it is less similar to , and slightly enhanced compared to , the gass pattern , at least in the mean . to put this comparison on a more quantitative footing ,
we apply the statistical method of bootstrap sampling ( e.g. , efron 1979 ; efron & tibshirani 1993 ) to judge the differences between triand and gass . for a more conservative test , we combine the @xmath28 data points of triand+gass stars for [ ti / fe ] and [ y / fe ] , and @xmath29 for [ la / fe ] as if all stars are from one dsph system .
we then randomly pick six stars from the 27 triand+gass star sample , and compute and compare the mean values for [ ti / fe ] , [ y / fe ] and [ la / fe ] for the six selected and 21 non - selected stars ( or , in the case of [ la / fe ] , five from 24 stars ) .
this process is repeated with 10,000 random selections of six ( five ) stars .
the derived distributions of the bootstrap samplings for ti , y and la are approximately normally distributed , and we find that there are only @xmath30 with mean [ ti / fe ] less than the mean value of triand stars in the 10,000 bootstrap samples , and @xmath31 with mean [ y / fe ] and @xmath32 with mean [ la / fe ] more than those in triand stars .
if the bootstrap sampling is repeated with the same procedure as above but for only gass stars , we find only @xmath33 with mean [ ti / fe ] less than triand in the 10,000 bootstrap samples , and only @xmath34 with mean [ y / fe ] and 0 with mean [ la / fe ] more than triand .
if we restrict the gass stars to the same color range spanned by the triand stars ( @xmath35_o \le 1.07 $ ] ) to minimize metallicity selection biases , we find the mean [ fe / h ] for eight gass stars to be @xmath36 dex with a dispersion ( standard deviation ) of 0.26 dex , compared with a mean [ fe / h ] of @xmath37 dex with a dispersion of 0.19 dex for the six triand stars . applying the bootstrap method for the more conservative test we described above
, we find that there are only @xmath38 with mean [ ti / fe ] less than the mean value of triand stars in the 10,000 bootstrap samples , and @xmath39 with mean [ y / fe ] and @xmath40 with mean [ la / fe ] more than those in triand stars for the combination of @xmath41 data points of triand+gass stars for [ ti / fe ] and [ y / fe ] , and @xmath42 for [ la / fe ] .
if we repeat the bootstrap sampling for the eight gass stars only and compare results with the mean values in triand , we find only @xmath43 with mean [ ti / fe ] less than triand in the 10,000 bootstrap samples , and 0 with mean [ y / fe ] and [ la / fe ] more than triand , respectively .
given the marked degree to which the mean values of _ all three _ of the [ ti / fe ] , [ y / fe ] and [ la / fe ] ratios for triand stars are different from corresponding means for the various bootstrap samples , we conclude that there might be significant chemical differences between the two systems , sufficient to suggest that triand is not likely a part of the gass system .
as discussed in c10a , distinctive abundance patterns in stars reflect the unique enrichment history of their parent system , and therefore `` chemical fingerprinting '' can help identify tidally stripped and captured stars in the galactic field .
the chemical patterns of @xmath20-elements ( e.g. , ti ) and s - process elements ( e.g. , y and la ) may indicate differences in the ( early ) star formation rate ( sfr ) between the progenitors of the triand , sgr , and gass systems , as well as the mw . for example , @xmath20-elements are mainly produced from type ii supernovae ( sn ii ) while iron is synthesized largely by type ia ( sn ia ) .
so [ @xmath20/fe ] is high in the early enrichment of a stellar system until sn ia occur after the first @xmath44 gyr .
a shift to a lower [ @xmath20/fe ] trend compared to the mw at some particular [ fe / h ] indicates a relatively slower initial sfr in that system , a feature seen in many dsph galaxies as well as the lmc ( e.g. , smith et al . 2002 , shetrone et al . 2003 ; tolstoy et al .
2003 ; venn et al .
2004 ; geisler et al . 2005 ; pompia et al . 2008 ; c10a ) .
triand follows the same @xmath20 trends as dsph galaxies , consistent with an origin in a dwarf galaxy - like environment and , like other mw satellites , a low early sfr compared to the mw .
the s - process elements are primarily generated in low mass agb stars . in more metal - poor environments
agb stars produce heavier s - process elements like la more efficiently than lighter species such as y ( see review by busso , gallino , & wasserburg 1999 ) .
c10a found underabundant trends in [ y / fe ] for all sgr stars compared to the mw , and an upturn in [ la / fe ] for sgr at [ fe / h]@xmath45@xmath46 .
the latter trend also can be seen in mcwilliam & smecker - hane ( 2005 ) , and is an indication of a slow enrichment history , so that the yields from low - metallicity agb stars have enough time to contaminate the interstellar medium and leave their signature on the metal - rich populations ( venn et al .
2004 ; pompia et al.2008 ) .
the [ y / fe ] trends in triand stars are slightly higher than those in either sgr and gass and somewhat closer to mw trends , whereas triand shows a slightly lower [ la / fe ] pattern compared to sgr .
this suggests that agb stars in the triand progenitor produced heavy s - process elements less efficiently than in sgr i.e. , that low metallicity agb progenitors in triand contributed relatively less heavy s - process enrichment than those in sgr . this may also indicate that the triand progenitor was more metal - rich than the sgr progenitor . on the other hand ,
the one triand star with [ fe / h]@xmath45@xmath46 shows slightly supersolar [ la / fe ] ( like the sgr [ la / fe ] pattern , only less extreme ) , but above the gass [ la / fe ] pattern at these metallicities .
this may suggest that the triand system enriched more slowly than the progenitor(s ) that generated the gass stars , but more similar in rate to the sgr dsph , to allow expression of the high la yields from those relatively fewer low metallicity agb stars that it did contain ; but this conclusion must be regarded as quite tentative , given the errors on individual abundances and that this is based on only a single high metallicity triand star .
the @xmath8-body simulations from pearrubia et al .
( 2005 ) suggest that triand may be an older piece of the mon tidal stream . according to the sgr paradigm ,
older debris should be more metal - poor than younger debris , but share the same chemical pattern trends if from the same parent system .
indeed , pearrubia et al .
associate older model stream wraps with lower metallicity stars , but these data derive from lower resolution spectroscopic measures of ca ii line indices for gass and triand stars ( c03 , rp04 ) . though our study here has focused on better precision , high resolution abundance determinations , including [ fe / h ] derived from eleven individual fe line measures , unfortunately the sample of stars is still too small to make definitive conclusions about the relative metallicity of the two systems . on the other hand , the sample of triand and gass stars in hand does offer tantalizing evidence that , at least at the metallicities that were sampled , triand does not reveal the same chemical pattern trends as gass and that therefore the triand star cloud is not part of an extended gass system .
that said , larger samples of gass and triand stars having precisely determined metallicities and chemical abundance patterns would certainly provide a more conclusive chemical test of the possible connection between these two systems , as well as more insights into the origin and evolution of their respective progenitors .
m .- y.c . , s.r.m . and r.j.p .
acknowledge support from nsf grants ast-0307851 and ast-0807945 .
this project was also supported by the nasa _
sim lite _ key project _ taking measure of the milky way _ under nasa / jpl contract 1228235 .
k.c . and v.v.s .
also appreciate support from nsf grant ast-0646790 .
d. m .- d . acknowledges funding from the spanish ministry of education and science ( ramn y cajal program contract and research project aya 2007 - 65090 ) .
finally , we appreciate comments from two anonymous referees that helped improve this paper .
sun & & & & & & & 7.45 & & 4.90 & & & 2.21 & & & 1.13 & & + + triand & & & & & & & & & & & & & & & & + @xmath47 & 10.65 & 1.06 & 3750 & 0.4 & & 1.71 & 6.86 & @xmath48 & 3.83 & @xmath49 & & 1.47 & @xmath50 & & 0.30 & & @xmath51 + @xmath52 & 10.73 & 1.07 & 3750 & 0.3 & & 1.74 & 6.74 & @xmath53 & 4.33 & @xmath54 & & 1.36 & @xmath55 & & 0.32 & & @xmath56 + @xmath57 & 11.11 & 1.00 & 3850 & 0.4 & & 1.74 & 6.54 & @xmath58 & 4.02 & @xmath59 & & 1.27 & @xmath60 & & 0.21 & & @xmath61 + @xmath62 & 10.60 & 1.07 & 3750 & 0.4 & & 1.66 & 6.82 & @xmath63 & 3.94 & @xmath64 & & 1.33 & @xmath65 & & & & @xmath66 + @xmath67 & 11.11 & 1.02 & 3850 & 0.9 & & 1.53 & 7.12 & @xmath68 & 4.35 & @xmath69 & & 1.69 & @xmath70 & & 0.82 & & @xmath71 + @xmath72 & 11.02 & 1.07 & 3750 & 0.4 & & 1.65 & 6.81 & @xmath73 & 3.85 & @xmath74 & & 1.61 & @xmath75 & & 0.15 & & @xmath76 + | we undertake the first high resolution spectroscopic study of the triangulum - andromeda ( triand ) star cloud an extended , mid - latitude milky way halo substructure about 20 kpc away in the second galactic quadrant through six m giant star candidates selected to be both spatially and dynamically associated with this system .
the abundance patterns of [ ti / fe ] , [ y / fe ] and [ la / fe ] as a function of [ fe / h ] for these stars support triand as having an origin in a dwarf galaxy with a chemical enrichment history somewhat similar to that of the sagittarius dwarf spheroidal ( dsph ) galaxy .
we also investigate the previously proposed hypothesis that triand is an outlying , dynamically older piece of the monoceros stream ( also known as the galactic anticenter stellar structure , `` gass '' ) under the assumption that both features come from the tidal disruption of the same accreted milky way satellite and find that net differences in the above abundance patterns between the triand and gass stars studied suggest that these two systems are independent and unrelated . |
this work has been supported by the german research foundation ( dfg ) grant ni 1282/2 - 1 , by the helmholtz international center for fair within the framework of the loewe program ( landesoffensive zur entwicklung wissenschaftlich - konomischer exzellenz ) launched by the state of hesse , by a continuing faculty grant from loyola marymount university , by the lancaster - manchester - sheffield consortium for fundamental physics under stfc grant st / j000418/1 and partially by the european cost action mp0905 `` black holes in a violent universe '' .
m. cavaglia , _
int.j.mod.phys . _ * a18 * , 1843 ( 2003 ) ; p. kanti , _
int.j.mod.phys . _
* a19 * , 4899 ( 2004 ) ; a. casanova and e. spallucci , _ class.quant.grav . _ * 23 * , r45 ( 2006 ) ; m. bleicher and p. nicolini , _ j. phys
ser . _ * 237 * , 012008 ( 2010 ) ; m. bleicher , p. nicolini , m. sprenger and e. winstanley , _
int.j.mod.phys . _
* e20s2 * , 7 ( 2011 ) .
s. hossenfelder , s. hofmann , m. bleicher and h. stoecker , _ phys.rev . _ * d66 * , 101502 ( 2002 ) ; s. hossenfelder , m. bleicher , s. hofmann , h. stoecker and a. v. kotwal , _ phys.lett .
_ * b566 * , 233 ( 2003 ) ; b. koch , m. bleicher and s. hossenfelder , _ jhep _ * 10 * , 053 ( 2005 ) ; p.
nicolini , _ j. phys . _ * a38 * , l631 ( 2005 ) ; p. nicolini , a. smailagic and e. spallucci , _ phys .
* b632 * , 547 ( 2006 ) ; e. spallucci , a. smailagic and p. nicolini , _ phys . lett .
_ * b670 * , 449 ( 2009 ) ; p. nicolini , _ int .
phys . _ * a24 * , 1229 ( 2009 ) ; l. modesto , j. w. moffat and p. nicolini , _ phys .
* b695 * , 397 ( 2011 ) ; p. nicolini , arxiv:1202.2102 [ hep - th ] ( 2012 ) ; p. nicolini and e. spallucci , arxiv:1210.0015 [ hep - th ] ( 2012 ) . | we review the phenomenology of mini black holes at colliders in light of the latest data from the lhc . by improving the conventional production cross - section ,
we show that the current non - observation of black hole signals can be explained in terms of quantum gravity effects . in the most optimistic case
, black hole production could take place at a scale slightly above the lhc design energy .
we also analyse possible new signatures of quantum - corrected planck - scale black holes : in contrast to the semiclassical scenario the emission would take place in terms of soft particles mostly on the brane .
the main argument in support of mini black hole ( bh ) production@xcite is based on a viable solution of the hierarchy problem of particle physics : by allowing gravity to probe additional spatial dimensions , one can avoid the discrepancy between the electroweak scale and the planck scale , placing quantum gravity phenomenology in reach of current particle accelerators@xcite .
such logic , however , is being questioned by the current non - observation of quantum gravity signals at the lhc@xcite , giving the topic of mini bhs a more speculative character with respect to early expectations ( see ref . for recent reviews of the status of quantum gravity and innovative proposals to test it ) . against this background ,
we show that the mechanism of bh production and their associated signatures might be drastically different from what is known to date .
our general argument is based on the fact that semiclassical formulations can not efficiently describe the conjectured quantum gravity at the terascale .
even if we acknowledge the difficulties in predicting quantum gravity phenomenology from first principles ( _ e.g. _ from string theory , loop quantum gravity , _ etc ... _ ) , we recall that irrespective of the specific formulation , quantum gravity exhibits a unique character : the emergence of a minimal resolution length @xmath0 . as a result , one can follow the strategy of improving the existing scenarios by any of the minimal length mechanisms proposed in the literature ( for recent reviews see ref . ) .
customarily , mini - bh formation is modelled via thorne s `` hoop conjecture '' : a bh forms whenever the impact parameter @xmath1 becomes smaller than the effective schwarzschild radius , @xmath2 , of the two - body colliding system .
while for classical bhs such a conjecture consistently implies a black disk cross section , @xmath3 , in the case of particle collisions quantum effects are expected to be relevant . indeed , the black disk profile leads to the phenomenologically inconsistent result that about @xmath4 would have formed at the super proton synchrotron ( sps ) in 1985 , provided additional spatial dimensions are assumed . to overcome this oddity , we recall that even the impact parameter can not be smaller than the minimal length , @xmath5 .
one can thus analytically improve the bh cross section as@xcite @xmath6 where @xmath7 is the mandelstam `` invariant mass '' of the colliding two - body system .
the above formula smoothly interpolates the black disk result in the trans - planckian regime and the expected suppression of sub - planckian bhs , without introducing `` by hand '' any threshold function . by setting @xmath8 and considering the current limits on the lhc luminosity@xcite @xmath9
, we can estimate the bh production rate @xmath10 . in doing so , one has also to take into account further quantum corrections to @xmath11 , which departs from the classical linear function of the beam energy @xmath7 due to bh remnant formation@xcite . in conclusion ,
one has a variety of parameter combinations but @xmath10 turns out to be negligible except for a high number @xmath12 of extra dimensions . for m - theory
inspired @xmath13-dimensional spacetimes , the `` new physics '' might be just behind the corner , _ i.e. _ , @xmath14 for @xmath15 tev .
we notice that the resulting suppression _ is _ a genuine effect of quantum gravity while the semiclassical regime occurs at energies well above the terascale .
another key point that needs revision is the profile of bh emission spectra .
conventionally these are obtained by considering greybody factors resulting from scattering equations on classical higher dimensional bh geometries .
this analysis can not work for planck scale black holes ( qbhs ) : strong quantum effects at scales @xmath16 not only smear out the curvature singularity but modify the global structure of the geometry with consequent horizon extremisation and halted evaporation @xcite . at the level of primary emission , qbhs exhibit a decreased temperature with respect to classical bhs with the same mass
: this implies reduced fluxes of particles and energy both in the brane and in the bulk .
such a reduction , however , is not `` homogeneous '' : in marked contrast to the results for schwarzschild bhs@xcite , the ratio of total bulk / brane emission can be seen to decrease rapidly as @xmath12 increases@xcite ( see tab .
[ aba : tbl1 ] ) .
[ aba : tbl1 ] even if our conclusions are based on the primary emission of scalar particles only and a study of the dynamics of the associated chromosphere and photosphere is still missing , we argue that qbhs can be characterized by a distinctive signature , _
i.e. _ , a peculiar emission of detectable soft particles mostly on the brane . |
bipartite and multipartite entanglement is the main resource that gives rise to many of the applications of quantum information and computation , like for example quantum teleportation @xcite and quantum cryptography @xcite , among others ( see for instance refs .
a compound system is entangled when it is impossible to attribute a complete set of properties to any of its parts . in this case , and for pure states , it is impossible to factor the state in a product of independent factors belonging to its parts . in this paper
we will consider bipartite systems composed of two @xmath3 fermions .
our aim is to uncover some specific features that apply when both particles are identical .
they appear itemized in the next page .
states of two identical fermions have to obey the symmetrization postulate .
this implies that they decompose into linear combinations of slater determinants ( sls ) of individual states . naively , as these sls
can not be factorized further , indistinguishability seems to imply entanglement .
this is reinforced by the observation that the entropy of entanglement ( eoe ) is bounded from below by @xmath4 , well above the lower limit @xmath5 for a pair of non - entangled distinguishable particles .
so , it looks like there is an inescapable amount of uncertainty , and hence of entanglement , in any state of two identical fermions .
the above issue has been extensively examined in the literature @xcite with the following result : part of the uncertainty ( giving @xmath6 ) corresponds to the impossibility to individuate which one is the first or the second particle of the system .
this explains why the lower limit for the eoe is 1 .
consider for instance two identical @xmath3 fermions in a singlet state @xmath7.\ ] ] the antisymmetrization does not preclude the assignment of properties to the particles , but only assigning them precisely to particle 1 or particle 2 .
the reduced density matrix of any of the particles is @xmath8 with an eoe @xmath9 . the portion of @xmath10 above 1 ( if any ) is genuine entanglement as it corresponds to the impossibility of attributing precise properties to the particles of the system @xcite .
assume for instance that we endow the previous fermions with the capability of being outside ( @xmath11 ) or inside ( @xmath12 ) the laboratory ( @xmath13 , @xmath14 , @xmath15 , @xmath16 ) .
we now have two different possibilities : either the fermion outside has spin up ( @xmath17 ) or spin down ( @xmath18 ) .
hence , there are two different sls for a system built by a pair of particles with opposite spins , one outside , the other inside the laboratory @xmath19 , \nonumber \\ { \rm sl}(1,2)_2 & = & \frac{1}{\sqrt{2}}[\psi(1)^{\downarrow}\,\varphi(2)^{\uparrow}- \varphi(1)^{\uparrow}\ , \psi(2)^{\downarrow } ] .
\label{j1}\end{aligned}\ ] ] they form two different biorthogonal states , the combination @xmath20/\sqrt{2}$ ] corresponding to the singlet and @xmath21/\sqrt{2}$ ] to the triplet state ( with respect to the total spin @xmath22 ) .
an arbitrary state @xmath23 would then be a linear combination of these two sls : @xmath24 giving an eoe @xmath25 clearly , when @xmath26 or @xmath27 vanish , we come back to @xmath6 , as the only uncertainty left is the very identity of the particles . summarizing ,
while indistinguishability is an issue to be solved by antisymmetrization within each sl , entanglement is an issue pertaining to the superposition of different sls @xcite . at the end
, we could even decide to call 1 to the variables of the outside particle , and forget about symmetrization @xmath28 as both particles are far away from each other . in this case , the eoe @xmath29 is lesser than the one corresponding to antisymmetrized states by a quantity of 1 , which is just the uncertainty associated to antisymmetrization . from now on we will consider the latter definition of @xmath10 , which gives the genuine amount of entanglement between the two particles .
notice that for half - odd @xmath30 , the number @xmath31 of slater determinants is bounded by @xmath32 , where @xmath33 is the dimension of each hilbert space of the configuration or momentum degrees of freedom for each of the two fermions .
much in the same way as above , we could consider one of the particles as right moving ( @xmath34 ) the other as left moving ( @xmath35 ) , giving rise to two sls in parallel with the above discussion .
this is the first step towards the inclusion of the full set of commuting operators for the system .
in addition to the spin components ( @xmath36 ) or helicities , there are the total @xmath37 and relative @xmath38 momenta . in the center of mass ( com )
frame we could consider the system described by the continuum of sls @xmath39 , \nonumber\\ { \rm sl}(1,2;\mathbf{p})_{-s } & = & \frac{1}{\sqrt{2}}[\psi(1)_0^{-s}\,\psi(2)_\pi^{s}- \psi(1)_\pi^{s}\ , \psi(2)_0^{-s } ] , \label{j5}\end{aligned}\ ] ] where @xmath40 and @xmath41 .
the labels 0 and @xmath42 are the azimuthal angles when we laid the axes along @xmath38 .
finally , there is a pair of sls for each @xmath38 , so that a general state made with two opposite spin particles with relative momentum @xmath38 could be written in the form : @xmath43 with @xmath44 .
again , we run into the impossibility to tell which is 1 and which is 2 .
in addition there may be some uncertainty about the total spin state , whether a singlet or a triplet , or conversely , about the spin component of any of the particles , @xmath45 or @xmath46 .
after this discussion it should be clear to what extent entanglement and distinguishability belong to different realms @xcite .
the only requirement to include identical particles is to symmetrize the expressions used for unlike particles . until now
, we have only considered the free case .
we have to examine the case of two interacting particles , as interaction is expected to be the source of subsequent entanglement @xcite .
obviously , the answer may depend on a tricky way on the detailed form of the interaction , of its spin dependence in particular .
it also seems that the role of particles identity , if any , will be played through symmetrization . in the following
we will show that spin entanglement is generated for the case of two interacting spin-@xmath47 identical particles , with the following features : * spin - spin entanglement is generated even by spin independent interactions . * in this case , it is independent of any symmetrization procedure .
* this phenomenon does not appear for unlike particles .
we first tackle the scattering of two unequal @xmath3 particles @xmath48 and @xmath49 which run into each other with relative com momentum @xmath38 .
we set the frame axes by the initial momentum @xmath38 of particle @xmath48 , and let the spin components be @xmath50 and @xmath51 along an arbitrary but fixed axis .
we will consider a spin independent hamiltonian @xmath52 , so the evolution conserves @xmath53 and @xmath54 .
we denote by @xmath55 @xmath56 the state of particle @xmath48 ( @xmath49 ) that propagates along direction @xmath0 with spin @xmath30 . in these conditions the scattering proceeds as : @xmath57 where @xmath0 is the scattering angle and @xmath58 the scattering amplitude
. we will consider @xmath0 different from 0 or @xmath42 to avoid forward and backward directions .
while the increase of uncertainty due to the interaction is clear , because a continuous manifold of final directions with probabilities @xmath59 opened up from just one initial direction , spin remains untouched .
the information about @xmath60 is the same before and after the scattering ; as much as we knew the initial spin of @xmath48 , we know its final spin whatever the final direction is . in other words , spin was not entangled by the interaction .
we will now translate these well known facts to the case of identical particles , where they do not hold true .
let particle @xmath49 be identical to @xmath48 .
consider the same initial state as before : a particle @xmath48 with momentum @xmath38 and spin @xmath30 runs into another @xmath48 with momentum @xmath61 and spin @xmath62 .
notice there is maximal information on the state .
we could write @xmath63 , and eventually symmetrize .
we now focus on the final state .
it is no longer true that particle @xmath48 will come out with momentum @xmath64 and spin @xmath30 with amplitude @xmath58 while the amplitude for coming out with momentum @xmath64 and spin @xmath62 vanishes .
recalling that @xmath49 above did become @xmath48 , the two cases @xmath65 and @xmath66 fuse into a unique state @xmath67 as shown in figure [ figsse1bis ] . , @xmath68 indicate the corresponding third component of spin.,width=302 ] notice the uncertainty acquired by the spin : now particle @xmath48 comes out from the interaction along @xmath0 either with spin @xmath30 or with spin @xmath62 , with relative amplitudes @xmath58 and @xmath69 respectively .
in other words , spin was entangled during the spin independent evolution . here , it is not the spin dependence of the interaction , but the existence of additional degrees of freedom which generate spin - spin entanglement .
these , act as ancillas creating an effective spin - spin interaction that entangles the two fermions .
the ancilla and the degree of entanglement depend on the scattering angle @xmath0 .
notice that for @xmath1 both amplitudes @xmath58 and @xmath69 become equal , so that the degree of generated entanglement is maximal , 1 e - bit . on the other hand , for @xmath70 , it generally holds @xmath71 , so that in the forward and backward scattering almost no entanglement would be generated .
however , this depends on the specific interaction . in sec .
[ asecslo ] we will clarify this point with coulomb interaction .
symmetrization does not change this , it only expresses that we can not tell which one is 1 and which one is 2 .
the properly symmetrized initial state is @xmath72 . \label{j9}\end{aligned}\ ] ] the scattering process could be written in terms of sls as @xmath73 where @xmath74 is the final momentum and the slater determinants are given in ( [ j5 ] ) .
both , this expression and eq.([j8 ] ) , describe the same physical situation and lead to the same entanglement generation .
the bosonic case may be analyzed in an analogous way .
the modification for two - dimensional spin hilbert spaces ( i.e. photons ) would be a sign change in eqs .
( [ j5 ] ) , ( [ j9 ] ) and ( [ j10 ] ) , as bosonic statistics has associated symmetric states . the equivalent of eq .
( [ j10 ] ) for bosons is a genuine entangled state for @xmath75 , much as in the fermionic case .
we now consider coulomb interaction at lowest order to illustrate the reasonings presented above . in this case @xmath76 where @xmath77 is a numerical factor depending on the charge @xmath78 .
@xmath79 and @xmath80 are two of the mandelstam variables , associated to @xmath81 and @xmath82 channels respectively , and depending on the scattering angle @xmath0 . for initial @xmath83 and final @xmath84 relative 4-momenta of the scattering fermions ,
they are given by @xmath85 , @xmath86 .
in the com frame , @xmath87 where @xmath88 is the mass of each fermion and @xmath89 is the available energy . according to this , the spin part of the state ( [ j8 ] ) for this case ,
properly normalized , is @xmath90 being @xmath91 the two amplitudes @xmath92 and @xmath93 vary monotonously as @xmath0 grows , becoming equal for @xmath1 . the physical meaning for this
is that for @xmath94 , the knowledge about the system is maximal and the entanglement minimal ( zero ) , and for increasing @xmath0 the knowledge of the system decreases continuously until reaching its minimum value at @xmath1 .
accordingly , the entanglement grows with @xmath0 until reaching its maximum value for @xmath1 .
we plot in figure [ figsse2 ] the eoe @xcite @xmath95 of state ( [ eqsse13 ] ) as a function of @xmath0 , for @xmath96 .
the entanglement grows monotonically until @xmath1 , where it becomes maximal ( 1 e - bit ) . as a function of @xmath0 . ]
in order to analyze the role the @xmath0 scattering angle plays in the generation of these genuine quantum correlations , we consider now the degree of violation of bell s inequality as a function of @xmath0 . to this purpose
, we define @xcite the observable @xmath97,\nonumber\end{aligned}\ ] ] where @xmath98 is the ( normalized ) state ( [ j8 ] ) and @xmath99 , @xmath100 are arbitrary unit vectors . in eq .
( [ eqsse15 ] ) we consider the amplitudes @xmath58 and @xmath69 normalized for each @xmath0 , in the form @xmath101 .
we consider three coplanar unit vectors , @xmath99 , @xmath100 and @xmath102 .
@xmath103 , @xmath104 and @xmath105 .
we have @xmath106 the bell s inequality , given by @xcite @xmath107 will then be @xmath108 for the particular case of coulomb interaction at lowest order here considered , @xmath109 and thus the critical angle for which the inequality becomes violated is @xmath110 for @xmath111 . for @xmath112 the bell s inequality does not hold .
we show in figure [ figsse3 ] the @xmath0 dependence of @xmath113 together with the classical - quantum border , @xmath114 , at @xmath110 .
thus , for experiments with @xmath115 one could be able in principle to discriminate between local realism and quantum mechanics . as a function of @xmath0 .
the classical - quantum border corresponds to @xmath111 , with @xmath110 . ]
this is in contrast with recent analysis of bell s inequalities violation in elementary particle systems @xcite , where the emphasis was placed on flavor entanglement , @xmath116 , @xmath117 , and the like .
these analysis presented @xcite two kinds of drawbacks coming from the lack of experimenter s free will and from the unitary evolution with decaying states .
these issues reduce the significance of the experiments up to the point of preventing their use as tests of quantum mechanics versus local realistic theories .
the spin - spin entanglement analyzed in this paper does not have this kind of problems and could be used in principle for that purpose .
in summary , we analyzed the relation between entanglement and antisymmetrization for identical particles , in the context of spin - independent particle scattering .
we showed that , in order to create genuine spin - spin quantum correlations between two @xmath3 fermions , spin - dependent interactions are not compulsory .
the identity of the particles along with an interaction between degrees of freedom different from the spin , suffice for this purpose .
the entanglement generated this way is not a fictitious one due to antisymmetrization , but a real one , and violates a certain bell s inequality for @xmath118 .
this work was partially supported by the spanish mec project no .
fis2005 - 05304 . l.l . acknowledges support from the fpu grant no .
ap2003 - 0014 . | we consider the bipartite spin entanglement between two identical fermions generated in spin - independent scattering .
we show how the spatial degrees of freedom act as ancillas for the creation of entanglement to a degree that depends on the scattering angle , @xmath0 . the number of slater determinants generated in the process is greater than 1 , corresponding to genuine quantum correlations between the identical fermions .
the maximal entanglement attainable of 1 e - bit is reached at @xmath1 .
we also analyze a simple @xmath0 dependent bell s inequality , which is violated for @xmath2 .
this phenomenon is unrelated to the symmetrization postulate but does not appear for unequal particles . |
on 22 september 2011 at 10:29ut , an x - ray flare ( goes class x1.4 ) began in an active region located on the east limb of the sun ( noaa active region 11302 ; n13e78 ) .
approximately 11 minutes after the flare start time , a bright wave - like front ( cbf ) was observed propagating away from the southern edge of the active region in sdo / atmospheric imaging assembly ( aia ; * ? ? ?
* ) 21.1 nm passband images .
the cbf then propagated along the east limb from @xmath215@xmath6south to @xmath250@xmath6south of the equator ( fig .
[ fig : figure_aia_nrh_c2 ] ) . during the same period of the cbf propagation , a bright 150.9mhz source formed above the cbf , imaged using the nanay radioheliograph ( nrh ; * ? ? ?
* ) ( contours fig .
[ fig : figure_aia_nrh_c2 ] ) . in each image
the contours range from @xmath7 to @xmath8 , where @xmath7 is the peak brightness temperature at the time of the image ; the intensity of the contours is indicated by the colour bar on the right .
initially , both the erupting structure seen in the aia image and the radio source had the same spatial extent over latitude ( fig .
[ fig : figure_aia_nrh_c2]a ) , showing they belong to a common structure . after this
, the most southern part of the radio source reached an extremely high brightness temperature ( @xmath210@xmath9k ) and closely followed the propagation of the cbf southward until it eventually diminished into the thermal background at 10:56ut .
another emission source at 150.9mhz was also observed at @xmath2@xmath10 latitude on the east limb at a height of 1.11.3@xmath11 ( fig .
[ fig : figure_aia_nrh_c2]c , d , e , f ) ; while this source was clearly associated with the eruptive active region , any link between it and the cbf is secondary , as it shows no temporal relationship with the start and stop time of the bright front .
similar radio source motion is observed at 173 , 228 and 270mhz , however any co - propagation of these radio sources with the cbf is of much shorter duration .
a movie showing the co - propagation of the cbf and 150mhz radio source can be found in supplementary movie 1 . for a movie and discussion of the multi - thermal nature of the cbf see supplementary movie 2 . to compare the motion of the cbf and radio source , the position angle ( pa ) trajectories
were analyzed ( fig .
[ fig : angle_time ] ) . both the cbf and radio source
clearly show common kinematics , with the two features having a consistent progression southward around the east limb .
the solid lines show a fit of @xmath12 to the data , where @xmath13 is the starting pa , @xmath14 is the angular velocity , and @xmath15 is time .
the slope of each line gives @xmath14 , from which the velocity of the source may be obtained using @xmath16 , where @xmath17 is the distance of the source from sun center .
for the cbf , an angular velocity of @xmath18rads@xmath5 was obtained , resulting in a velocity of @xmath19kms@xmath5 at @xmath20 . at @xmath21 ,
the cbf was found to have an angular velocity of @xmath22rads@xmath5 , resulting in @xmath23kms@xmath5 .
the 150mhz source had an angular velocity of @xmath24 ; the value for @xmath17 of this source was estimated by directly converting frequency @xmath25 to electron density @xmath26 , using the methods described in supplementary material .
this gave a height of @xmath27 for the radio source , resulting in a velocity of @xmath28kms@xmath5 .
this velocity is larger than the cbf velocity at @xmath20 , but is comparable to the cbf velocity at @xmath21 ( fig .
[ fig : angle_time ] ) .
the similar speeds and trajectories of the cbf and radio source suggest that they belong to a common propagating structure in the corona .
finally , the radio source motion speed was used to estimate its alfvn mach number @xmath29 , where @xmath30 is the alfvn speed determined from magnetic field and density measurements at the radio source height ( see supplementary material ) . with @xmath31kms@xmath5 ,
the alfvn mach number of the radio source is @xmath0 , showing that the source travelled in excess of the local wave speed in the corona .
finally , we note that a potential field source surface extrapolation ( pfss ) reveals the presence of open and radial field in the south east quadrant of the corona ( see supplementary figure 2 ) , revealing that the propagation was transverse to the magnetic field .
this is an important aspect of quasi - perpendicular shock orientation that we discuss in the last section .
the 150mhz source observed by nrh had a brightness temperature of @xmath2@xmath32k , indicating coherent plasma emission .
such emission is generated via plasma oscillations that are due to instabilities in the presence of high velocity electron beams @xcite .
the presence of electron beams was independently verified and revealed in detail using radio dynamic spectra . at @xmath210:40ut
the fundamental and harmonic bands of a type ii burst shock signature was observed at 45 and 90mhz ( fig .
[ fig : dyn_spec]b ) , respectively , using the nanay decametric array ( nda ; * ? ? ?
type iii bursts begin at the same time as the type ii ( fig .
[ fig : dyn_spec]a , b ) , observed using nasa s stereo - b / waves instrument @xcite .
the frequency drift of these type iii radio bursts provide a measure of velocity of the electrons causing the radio emission . converting frequency - time measurements to height - time via a coronal density model ( see supplementary material ) , the beam speed was estimated to be @xmath330.4c ( 46kev ) , where c is the speed of light .
this is a clear signature of relativistic electron acceleration in association with the shock .
these particles were eventually detected in - situ by the stereo - b spacecraft ( see supplementary figure 3 ) .
more striking evidence for shock - accelerated electrons is in the form of ` herringbone ' emission , observed using the ecallisto spectrometers @xcite at the rosse solar terrestrial observatory ( rsto ; * ? ? ?
* ) ( fig .
[ fig : dyn_spec]c ) .
the herringbones result from individual beams of shock - accelerated electrons @xcite , traveling towards and away from the sun i.e. , to higher and lower frequencies .
similar features occur between 100200mhz ( fig .
[ fig : dyn_spec]b ) , showing the same characteristics as herringbones ( a bursty nature and decreasing intensity with respect to time ) . in a similar manner to the type iii bursts ,
the beam velocity of herringbone electrons was estimated to be 0.15c , again showing the presence of near - relativistic electron acceleration in association with the presence of a shock .
while their structure in frequency reveal how fast these beams travel , their behavior in time can reveal detailed temporal characteristics of the shock acceleration process .
a wavelet analysis using a derivative of a gaussian wavelet performed on a time series at 54mhz reveals periodicity at 211 seconds ( see supplementary figure 1 ) .
previous authors have attributed this bursty nature to rippling and inhomogeneity along the shock front , possibly revealing some level of instability or shock turbulence in the acceleration region @xcite ; we discuss this in the last section .
we note that the features at 100200mhz appear to be the extension of the herringbones into higher frequencies .
these features in particular show good temporal correspondence with the radio source i.e. , they have a start - stop time comparable to the radio source .
this is particularly apparent for the group of bursts at 10:5210:56ut ( supplementary movie 1 ) .
the radio emission in the dynamic spectra have all the hallmarks of shock generation with particle acceleration closely tied to the process .
the association of shock radio activity with the imaged 150mhz source suggests that the two observables have a common origin in a plasma shock .
overall , the position of the radio source at the southern flank of the cme , the transverse motion of the source ( propagation parallel to the surface ) and the zero frequency - drift of the herringbones is suggestive of a shock driven parallel to the surface by the flank expansion , similar to the assertion by @xcite and @xcite .
indeed the association of the cbf with this radio activity is corroborative evidence of the wave / shock system at the flank .
further evidence of a shock having occurred in the corona was obtained through white - light observations , allowing us to study the position of this shock relative to the cme .
a cme associated with this event was observed by the large angle spectroscopic coronagraph ( lasco ; * ? ? ?
* ) , first appearing at 10:46ut , with an apex heliocentric distance of @xmath22.6@xmath34 ( fig .
[ fig : figure_aia_nrh_c2]c ) .
the next available image shows the bright cme front with a fainter , secondary front at the southern flank ( fig .
[ fig : wl_shock]b ) .
this ` two - front ' morphology is a common occurrence in white - light cme structure and constitutes a reliable signature of a cme front associated with a stand - off shock @xcite . in order to distinguish between the cme front and shock front
, we performed a 3d reconstruction of the cme using the elliptical tie - pointing method described in @xcite ( fig .
[ fig : wl_shock]d ) .
this reconstruction reveals that the bright front outlined in the c2 coronagraph ( ellipse in fig .
[ fig : wl_shock]b ) corresponds to the faint front outlined as a halo in stereo - b cor1 ( ellipse in fig .
[ fig : wl_shock]c ) .
furthermore , the observations reveal that the secondary and extremely faint front at the southern edge of the cme ( as imaged in lasco / c2 , fig .
[ fig : wl_shock]b ) can not be considered as part of the cme structure , but is actually an associated shock front .
we note that white - light shocks have been reported in the past , occurring both in the low corona as well as out to @xmath20.5a.u .
@xcite . here
, we have employed a 3d reconstruction from multi - viewpoint observations to qualitatively confirm the presence of a shock at the southern flank of the cme , in the same region as the cbf and radio burst .
there has been much debate surrounding the assertion that cbfs are a wave phenomenon @xcite , with numerous authors suggesting a pseudo - wave theory @xcite . in the past
, the association of cbfs with type ii and type iii bursts has been used as evidence against this pseudo - wave interpretation and more in favor of the mhd wave paradigm @xcite .
this study reveals that the cbf in this event was indeed closely associated with shock radio activity positioned on the flanks of an expanding cme .
this kind of behavior has been suggested before , but never directly imaged @xcite .
it shows how a combination of radio and euv imaging can reveal the evolution of plasmoid driven shocks in the solar atmosphere @xcite . of further interest in this study
is the likelihood of a quasi - perpendicular orientation of the shock , as revealed by the pfss extrapolation .
quasi - perpendicularity is an essential aspect of the shock drift acceleration ( sda ) mechanism @xcite , a process believed to be responsible for particle acceleration in planetary magnetospheres @xcite and solar radio bursts @xcite .
this mechanism involves an adiabatic reflection of particles from the shock , with the energy gain sourced in the @xmath35 electric field , where @xmath36 and @xmath37 are the upstream flow speed and magnetic field , respectively . a single reflection from the shock has limited energy gain , however multiple reflections may produce relativistic energies , which is particularly important for low mach number shocks such as that reported here ( @xmath38 ) and in @xcite .
this multiple reflection process may be explained by inhomogeneity in the shock front , a characteristic usually known as rippling @xcite .
2d hybrid simulations show that rippling is brought about by an instability @xcite and resembles a standing - wave mode of the shock surface @xcite .
the presence of ripples can lead to a quasi - sinusoidal variation in shock - normal orientation with respect to the upstream magnetic field .
since the efficiency of sda requires quasi - perpendicularity , there will be sites on the shock front that provide efficient acceleration and sites that do not @xmath39 a structure that may lead to magnetic trapping and multiple reflections @xcite , hence producing higher energy particle acceleration .
the presence of ripples can produce quasi - periodic herringbones in three ways .
firstly , it makes sda more efficient and capable of producing the observed herringbone energies , especially when particle scattering is considered @xcite .
secondly , the periodic spatial variation in the acceleration efficiency of the shock could explain the bursty and quasi - periodic nature of the herringbones ( supplementary figure 1 ) .
@xcite suggest that shock front inhomogeneity brought about by mhd turbulence is a possible explanation of bursty herringbones .
@xcite produced a detailed model of sda from a rippled shock , specifically on the flanks of an expanding cme .
their results suggest that herringbones could be produced by accelerated electrons at spatially intermittent regions of quasi - perpendicularity on a rippled shock surface .
thirdly , @xcite also predicts that , if rippling is present , the upstream and downstream electron beams should have similar energies , which is not predicted for a uniform or smooth shock . a sample of the oppositely drifting herringbones in fig .
[ fig : dyn_spec]c shows that positive and negative frequency drifts are both @xmath25mhzs@xmath5 , revealing that the upstream and downstream populations have similar energies ( although we note the possibility that both positive and negative drifting herringbone features may be accelerated upstream , as suggested by the schematic of @xcite ) .
there is also the possibility that the herringbones may be associated with a termination shock of a reconnection outflow occurring behind the cme @xcite . in such a scenario , this shock would have a more indirect relationship with the cme propagation .
however , the imaged radio source shows a good temporal correspondence with the shock activity in the dynamic spectra , especially between 10:52@xmath3910:56ut , suggesting the particle acceleration indicated in the spectra shares a close relationship with the propagating source .
our observations reveal the need for a more detailed modeling of herringbone solar radio bursts .
the quasi - periodic behavior of herringbones provides a possible direct measure of shock inhomogeneity and the spatial scales over which the magnetic field varies in the shock and ambient corona ; it may also provide a measure of the turbulence in these plasma flows @xcite . in the future , high cadence euv imaging from sdo , combined with sensitive radio imaging - spectroscopy observations from instruments such as the low frequency array ( lofar ; * ? ? ?
* ) , will reveal unprecedented detail of plasma shocks and their role in particle acceleration .
this may reveal the fundamental nature of a plasma shock process that is universal , but currently impossible to directly observe in any other area of astrophysics .
we would like to thank nasa s sdo , stereo , and esa / nasa s soho teams , and the nanay radio astronomy observatory for providing open access to their data .
financial support of e.p.c . was provided by the irish research council embark initiative .
is funded by the european commission s seventh framework programme under the grant agreement no .
284461 ( eheroes project ) .
is supported by shine grant 0962716 and nasa grants nnx08aj07 g and nnx13ag11 g to the institute for astronomy .
p.z is currently funded under the trinity college dublin innovation academy bursary .
is funded under the esa prodex programme .
we would also like to extend thanks to the birr scientific and heritage foundation , supported by the earl of rosse .
special thanks is extended to christian monstein for his support in setting up the callisto spectrometers and jasmina magdaleni for very useful scientific discussions .
e.p.c . performed the data analysis of the radio source kinematics , the radio burst analysis , the alfvn mach number calculations , and the in - situ particle analysis . e.p.c . also wrote the article .
d.m.l . performed the data analysis of the coronal bright front and gave constructive advice on the writing of the article .
j.p.b . performed the 3d reconstruction of the cme and gave advice on the white - light shock analysis section .
provided the density maps , and d.s.b . provided the magnetic field maps that were used in the radio source and cbf mach number calculations .
j.m . installed the electronic systems at rsto .
conceived of the project and guided data analysis and writing of the article . | cosmic rays and solar energetic particles may be accelerated to relativistic energies by shock waves in astrophysical plasmas . on the sun , shocks and particle acceleration
are often associated with the eruption of magnetized plasmoids , called coronal mass ejections ( cmes ) . however , the physical relationship between cmes and shock particle acceleration is not well understood . here
, we use extreme ultraviolet , radio and white - light imaging of a solar eruptive event on 22 september 2011 to show that a cme - induced shock ( alfvn mach number @xmath0 ) was coincident with a coronal wave and an intense metric radio burst generated by intermittent acceleration of electrons to kinetic energies of 246kev ( 0.10.4c ) .
our observations show that plasmoid - driven quasi - perpendicular shocks are capable of producing quasi - periodic acceleration of electrons , an effect consistent with a turbulent or rippled plasma shock surface .
1 . astrophysics research group , school of physics , trinity college dublin , dublin 2 , ireland .
2 . mullard space science laboratory , university college london , holmbury st .
mary , dorking , surrey , rh5 6nt , uk .
3 . institute for astronomy , university of hawaii , 2680 woodlawn drive , honolulu , hi 96822 , usa .
+ @[email protected] _
_ coronal mass ejections ( cmes ) are spectacular eruptions of magnetized plasma from the low solar atmosphere into interplanetary space @xcite . with kinetic energies of @xmath2@xmath3j @xcite ,
they are the most energetic explosive events in the solar system and are often associated with plasma shocks and the acceleration of particles to relativistic speeds @xcite . however , the underlying mechanism relating cmes , shocks , and particle acceleration is still a subject of intense debate @xcite . by clarifying the inherent characteristics of these phenomena we learn not only about the nature of explosive plasma events but also about how they drive shocks and accelerate particles to high energies .
such processes are ubiquitous in the universe , playing a role in the acceleration of cosmic rays in supernovae and active galactic nuclei shocks @xcite .
cme - associated shocks are often observed over a variety of spectral bands . at radio frequencies , high intensity ( @xmath2@xmath4jy ) emissions , known as type ii and type iii bursts ,
are associated with coronal shocks and accelerated particles in the solar corona @xcite .
fine structure in these radio bursts can often reveal a bursty nature to the shock particle acceleration @xcite , which can reveal details of the internal shock structure @xcite . at extreme ultraviolet ( euv ) wavelengths
, the shock or pressure pulse response of the corona to an eruption may be imaged as a bright pulse propagating across the entire solar disk at typical velocities of 200400kms@xmath5 @xcite .
these coronal bright fronts ( cbfs ) are a regular feature of solar eruptive events and often display wave - like properties such as reflection @xcite , refraction @xcite and pulse broadening @xcite . like cmes ,
cbfs are often accompanied by type ii and type iii radio bursts , with euv and radio images revealing a spatial link between the phenomena that is suggestive of a common origin @xcite .
it has been proposed that the common origin for these myriad phenomena may be a cme - driven shock @xcite . in this scenario ,
the cme eruption drives a pressure pulse , observable in the low corona as a propagating wave - like cbf .
higher in the corona this same pulse forms a shock , accelerating particles and producing type ii and iii emission .
however , much debate surrounds the suggestion that ( i ) the cbf is a plasma pressure wave driven by a cme , and ( ii ) the radio bursts , generated by accelerated particles , result from this same wave / shock system .
the contention has arisen from attempts to explain non - wave kinematics of cbfs @xcite .
pseudo - wave theories are employed to describe this behavior , where the erupting cme produces a large - scale restructuring of the coronal magnetic field , which results in a propagating bright pulse ( via joule plasma heating ) that is not actually a driven wave @xcite . in this scenario ,
any relationship with shock observables is indirect .
further confusion is added by the possibility that high energy particles in association with the eruption may be a consequence of magnetic reconnection in the flaring active region , and not the result of a shock @xcite .
collectively , cmes , cbfs and radio bursts provide direct measures of both shock kinematics and the characteristics of the accompanying accelerated particles . however , a common theory explaining these phenomena has yet to be verified .
this lack of clarity can be ascribed to an euv imaging cadence that was unable to match the fast time sampling of radio imaging and spectroscopy .
now , using the high image cadence of the solar dynamics observatory ( sdo ; * ? ? ?
* ) , combined with fast time sampling radio images and spectra , we can reveal previously unseen characteristics of the relationship between these phenomena , proving that a cme - driven shock is the feature unifying these observations and that this shock is responsible for bursty electron acceleration .
this greatly advances our understanding of the close relationship between solar eruptions , plasma shocks and their resulting euv , radio and particle acceleration signatures . |
the preparation of ultracold ions as well as the production of nonclassical motional states of these ions plays a central role in ion trap experiments@xmath0}$ ] . as the trapped ion system is in nearly perfect isolation , immune from the violation of the environment , and meanwhile the electric - dipole forbidden transition is usually introduced for avoiding the dissipation of the ion , the decoherence in the preparation and preservation of nonclassical states can be effectively suppressed .
consequently , the ion trap system is also being considered to be a promising tool for the quantum computing@xmath1}$ ] .
the various theoretical schemes for generating the nonclassical states of motion of the trapped ions and achieving quantum computing with trapped ions are based on the jaynes - cummings ( jc ) model@xmath2}$ ] , in which the ion is simplified to be of two levels , the trap s potential is quantized as a harmonic oscillator , and the radiating lasers are supposed to be classical forms of standing or travelling waves . the general consideration is taken for the case of lamb - dicke limit ( ldl ) under the weak excitation regime ( wer ) , which corresponds to the actual case in present ion trap experiments@xmath3}$ ] .
the ldl means the ions moving within the region of the space much smaller than the effective wavelength of the laser wave , and the wer means the laser - ion interaction much smaller than the trap frequency . for this case
, some techniques developed in the framework of cavity qed based on jc model can be immediately transcribed to the ion trap system by taking advantage of the analogy between the cavity qed and ion trap problem . in the case of motion of ions exceeding the ldl , we can use the technique proposed in ref.[8 ] for a treatment . in a word , under the rotating - wave approximation ( rwa ) , the problem of trapped ions interacting with lasers is solvable .
however , the strength of the coupling between the ions and lasers can be conveniently adjusted simply by changing the intensity of lasers . if the coupling strength is much larger than the trap frequency , called the strong excitation regime ( ser ) , then the rwa is no longer applicable since the rapidly oscillating ( i.e. , counter - rotating ) terms also make significant contribution in the interaction , which is much different from the case in cavity qed@xmath4}$ ] .
it has been reported that the ser is useful for preparing nonclassical motional states of a trapped ion and implementing quantum computing with trapped ions@xmath5}$ ] .
for example in ref.[10 ] , schrdinger cat states could be prepared readily in simple way in the ser . based on that work
, a proposal was made for motional state engineering and endoscopy in the ser@xmath6}$ ] .
as the quantum state is sensitive to decoherence , the manipulation of trapped ions in the ser is of importance to the experimental implementation and measurement because of the great reduction of the operation time . due to the same reason
, a scheme for the hot - ion quantum computing also requires the condition of the ser@xmath7}$ ] .
in contrast to the theoretical studies , the ion - laser interaction in the ser has not yet been reported experimentally since the high - laser intensity produces increased off - resonant excitation of carrier transitions which limits the final ground state occupation of the cooling process@xmath8}$ ] , and the strong laser also modifies the original electronic levels of the ions@xmath9}$ ] , making the problem somewhat complicated . in a numerical study of the strong raman sideband excitation driving the vibrational ground state of the trapped ion@xmath10}$ ] , it is shown that the non - rwa interaction terms produce some limitations for two strong laser beams detuned by the vibrational frequency to be used as raman motional displacement beams , whereas the interaction is useful for the preparation of the schrdinger cat state . generally speaking ,
as the rwa is invalid , the ion trap problem in the ser can not be solved analytically with jc model .
so some approximations have to be introduced in refs.[10,11 ] and the studies in refs.[12,15 ] are resorted to numerical calculations .
in fact , as shown in ref.[16 ] , even if the ions are governed in the wer , in which the rwa is valid , the deletion of the rwa can also present some interesting results much different from those under the rwa . in that work ,
some particular analytical solutions have been obtained in the absence of the rwa for large detuning cases , and schrdinger cat states could be readily prepared .
however , that work is too simple .
we hope to investigate the raman configuration which has been extensively applied in actual ion trap experiments@xmath3}$ ] , instead of the simple case in ref.[16 ] .
we also hope to exclude the assumption related to both the wer and ldl .
therefore , in this contribution , we will try to undertake a more general and complete investigation on the trapped ion under the raman process .
moreover , a comparison of our results with those under the rwa will be made .
besides , we will propose a scheme for preparing the schrdinger cat state beyond the wer .
we study the single ultracold ion radiated by lasers in raman-@xmath11-type configuration , as shown in fig.1 .
the electronic structure is employed with two lower levels @xmath12 and @xmath13 coupled to a common upper state @xmath14 , and the two lasers with frequencies @xmath15 and @xmath16 respectively are assumed to propagate along opposite directions . for a sufficiently large detuning to the level @xmath14 , @xmath14 may be adiabatically eliminated , and what we have to treat is an effective two - level system , in which the lasers drive the electric - dipole forbidden transition @xmath13@xmath17 .
the dimensionless hamiltonian of such a system in the frame rotating with the effective laser frequency @xmath18@xmath19 can be written as@xmath20}$ ] @xmath21 where the detuning @xmath22 with @xmath23 being the optical resonance frequency , i.e. , the transition frequency of two levels of the ion , and @xmath24 the frequency of the trap .
@xmath25 is the dimensionless rabi frequency and @xmath26 the effective lamb - dicke parameter given by @xmath27 with subscripts denoting the counterpropagating laser fields .
@xmath28 ( @xmath29 ) are pauli operators , @xmath30 is the dimensionless position operator of the ion with @xmath31 and @xmath32 being operators of creation and annihilation of the phonon field , respectively .
the notations + and - in front of @xmath33 indicate the absorption of a photon from one beam followed by emission into the other beam and vice versa , respectively .
@xmath24 is generally supposed to be much greater than the atomic decay rate , called the strong confinement limit , for neglecting the effect of the atomic decay . in general
, one may expand @xmath34 in eq.(1 ) to the first - order terms of @xmath35 and neglect other higher - order terms by supposing the ion governed under the wer ( @xmath36 ) and within the ldl ( @xmath37 ) .
however , as we hope to treat the problem more exactly and generally , we first perform following unitary transformations on eq.(1 ) , that is @xmath38 where @xmath39 and @xmath40 with @xmath41 , @xmath42 and @xmath43 . actually , the unitary transformation @xmath44 made above is identical to that in [ 17 ] .
but for coinciding with the standard form of non - rwa model , @xmath45 is performed .
comparing with the standard non - rwa jc model@xmath4}$ ] , there exist an additional driven term and an additional constant term in eq.(2 ) , and the optical resonance frequency is replaced by the rabi frequency . in what follows , we will adopt the coherent state representation@xmath46}$ ] , by which the non - rwa jc models have been analytically treated@xmath47}$ ] . thus eq.(2 )
is rewritten as @xmath48 where @xmath49 is a complex number , and @xmath50 with the relation between the coherent state and fock state @xmath51 using the same idea as in refs.[16,19 ] , we assume the eigenfunction of the schrdinger equation of eq.(3 ) to be the series form @xmath52 where @xmath53 , @xmath54 and @xmath55 are constants determined later . taking @xmath56 into the schrdinger equation of eq.(3 ) will yield two recurrence relations @xmath57,\ ] ] @xmath58\ ] ] where @xmath59 is the trial solution of the eigenenergy of eq.(3 ) . as the procedure of solution is similar to that in ref.[16 ] , in what follows , we just list the results : \(1 ) @xmath60 for @xmath61 , we have two solutions .
one is @xmath62 with @xmath63 the eigenenergy of the system in this case , corresponding to @xmath64 and a restricted relation of @xmath65 .
the other is @xmath66 , with @xmath67 and a restricted relation of @xmath68 .
the eigenfunctions for above two cases are respectively @xmath69 and @xmath70 , where @xmath71 , @xmath72 $ ] , @xmath73 , and @xmath74 $ ] .
@xmath75 and @xmath76 can be set to be 1 for normalisation .
\(2 ) @xmath60 for @xmath77 , the two solutions are @xmath78 corresponding to @xmath79 and two independent restricted conditions , which can be solved from the following equation , respectively @xmath80 @xmath81}\ ] ] the eigenfunctions at this moment in the unit of @xmath82 are , respectively , @xmath83 and @xmath84 where @xmath85 , @xmath86 $ ] , @xmath87 $ ] , @xmath88c_{0 } \}$ ] , @xmath89 , @xmath90 $ ] , @xmath91 $ ] , and @xmath92c^{'}_{0 } \}$ ] ; \(3 ) when above technique is extended to the case of @xmath60 for @xmath93 , the series solution becomes very complicated .
however the eigenenergy of the system is still of the simple form , that is , @xmath94 , corresponding to @xmath79 and two independent sets of much complicated expressions of restricted conditions which we have to omit here , respectively .
as limitations for both the rabi frequency and lamb - dicke parameter as well as the rwa are excluded in the present treatment , the results we obtained above are more exact and general than those depending on the rwa or the approximate expansion of small values of @xmath26 and @xmath25 . however , as the problem involving the counter - rotating terms is non - integrable@xmath95}$ ] , the present solutions obtained by the termination of the series of the trial solution @xmath56 are some particular ones , along with some restricted conditions related to some parameters of the system . nevertheless , with the present solutions
, we can investigate some cases in a wide range of parameters , particularly for the case beyond the wer and ldl .
it is also of interest to make a comparison of our results with those under the rwa .
let us first consider a special case with both @xmath96 and @xmath97 .
in such a case , eq.(3 ) is reduced to @xmath98 repeating the above procedure of the series solution for the simplest case , i.e. , @xmath60 for @xmath61 , we can find @xmath99 and @xmath100 with @xmath101 , @xmath102 and @xmath103 .
we can easily find that the eigenenergy @xmath104 is in good agreement with the solution in a fock state representation for @xmath105 .
however , to make a comparison for eigenfunctions with former solutions , we have to return to the original representation before eq.(2 ) .
the eigenfunction in the original representation is @xmath106 @xmath107 where we have used the fact@xmath108}$ ] that @xmath109 in a coherent state representation corresponds to the displaced fock states @xmath110 by means of eq.(4 ) and baker - campbell - hausdorff formula .
therefore , in the case of @xmath111 but @xmath112 , we have from eq.(8 ) @xmath113 or @xmath114 , which means no transitions existing in the large detuning case .
if we assume @xmath115 and @xmath25 being arbitrary positive real number , we have @xmath116 , which corresponds to the carrier excitation .
both solutions above are also in good agreement with the solutions in a fock state representation@xmath117}$ ] . however , from eq.(8 ) we can know that , different from former solutions for the wer case , the probability amplitudes of up and down states of the ion are different and complicated in the case of the ser with large detunings .
this is a new result which has not been obtained before by general approaches made in a fock state or dressed state representation . for a more general comparison
, we should first present the solutions under the rwa .
as referred to in sec.i , eq.(1 ) can be treated to be a nonlinear jc model with the approach proposed in ref.[8 ] under the rwa .
however , to make the comparison more easy and clear , we start our rwa treatment from eq.(2 ) . by performing a unitary transformation
@xmath118 $ ] on eq.(2 ) with @xmath119 , we can obtain @xmath120 corresponding to the resonance conditions of @xmath121 under the rwa , and the eigenenergy in a fock state representation is of the form @xmath122 with @xmath123 .
meanwhile , performing a unitary transformation of @xmath124 $ ] on eq.(2 ) with @xmath125 , under the rwa resonance conditions of @xmath126 , we have @xmath127 whose eigenenergy in a fock state representation is of the form @xmath128 with @xmath123 . as shown in the last section ,
although our method can in principle present all series solutions , we merely present the specific forms of the non - rwa solution for two simplest cases . here
we will use those two cases for a comparison with eqs.(10 ) and ( 12 ) . for the case
( 1 ) , we find that the two solutions for @xmath129 , along with two different restricted relations respectively , actually correspond to the same expression @xmath130 however , for the case ( 2 ) , the situation is somewhat complicated .
the direct algebra shown in appendix presents us four solutions of e for this case , whereas the specific calculation shows that the solutions of eqs.(a3 ) and ( a4 ) are actually identical . in figs.2 and 3 ,
the solutions under the rwa are compared with those for the non - rwa treatment in the case of @xmath1310.5 and 3.0 .
we can find that , in the case of small values of @xmath25 and @xmath26 , some of the solutions under the rwa can approach those without the rwa .
it is physically reasonable because the rwa is only valid for the wer and ldl .
however , although the solutions without the rwa are exact , only a few of these solutions can be obtained for a certain termination of the series solution .
we can not obtain all the particular solutions of the system unless we make the termination of the series solution for almost infinite times up to the case with infinite series terms .
in contrast , the rwa treatment can present general solutions , like eqs.(10 ) and ( 12 ) .
nevertheless , as it merely retains some resonance terms in the hamiltonian , a specific solution under the rwa only corresponds to a specific value of @xmath25 .
moreover , the figures tell us that , only when the value of @xmath26 is not taken to be larger than 0.1 , can we consider that the solutions under the rwa and without the rwa are in good agreement in the case of the wer .
the rwa description is obviously inaccurate for the case beyond the ldl although @xmath25 is less than 1.0 in this case . when the value of @xmath25 is larger , the difference between the rwa case and non - rwa one becomes great , as shown in figs.3 .
it is difficult to find any correspondence for the two cases .
however , there are some crossing points between the rwa solutions and non - rwa ones .
it means at certain values of @xmath26 , the solutions for eigenenergies in these two cases can be the same . in physics
, it can be considered that the non - rwa solutions are some isolated ones , merely corresponding to those of up internal levels of the ion .
one may not expect to obtain the general eigenenergies with our non - rwa treatment .
however , the demonstration of above series solutions means the existence of the finite dimensional invariant subspace of some operators . particularly from the simple forms of the eigenenergies
, one may expect that the quantum kam theorem should be applicable for the ser problem@xmath132}$ ] . before ending our discussion of the usefulness of non - rwa solutions ,
it is also of interest to present a practical example of preparing the nonclassical motional states of the ion with our solutions . as referred to in sec.i
, there is a considerable interest in investigating the ser , particularly for rapidly preparing nonclassical motional states of the ion .
in fact , with the results in the present paper , one can also generate some particular nonclassical states through suitable adjustment of the rabi frequencies and proper measurements .
for example , consider the simplest case , i.e. , the case ( 1 ) in sec ii .
for the case of the resonance ( @xmath133 ) , we have @xmath134 with the same @xmath25 and double degenerate eigenfunctions @xmath135 and @xmath136 .
transforming our results into the schrdinger representation will yield the states @xmath137 and @xmath138 , respectively . in the case of the ldl(@xmath139 ) , the rabi frequency @xmath25 would be @xmath140 .
so we have @xmath141 , @xmath142 , @xmath143 and @xmath144 . by controlling the evolution time @xmath145 , and measuring the excited state of the ion
, we can obtain a displaced even coherent state , a kind of schrdinger - cat states @xmath146 where the relation between the coherent state and fock state has been used . as the rabi frequency approaches 2 ,
such a preparation is obviously outside the wer .
as @xmath109 corresponds to the displaced fock states @xmath110 , the eigenfunctions of the system we obtained are infinite superposition of displaced fock states .
therefore , it is understandable that we can not obtain similar analytical results by general fock state expansion .
the obvious advantage of our approach is the possibility to obtain some relations between parameters of the system corresponding to certain nonclassical states by truncating the series expansion step by step .
so as long as we can reach these conditions experimentally , the different nonclassical states predicted by our theory would be obtained . in summary
, the raman interaction between a trapped ultracold ion and two travelling wave lasers has been treated analytically in an interaction representation , without consideration of the rwa and the limitation for both the ldl and wer .
although the coherent state expansion technique was used previously for treating a similar problem , and the solutions we presented here are some isolated ones under certain conditions related to @xmath25 , @xmath26 and @xmath147 , the present investigation is more general and exact so that we can study the ion trap problem in a wide range of parameters and compare our results with other approximate works or numerical solutions in this respect as we did in the present paper and for the cavity - atom problem@xmath148}$ ] .
more importantly , the raman process has been widely used for the laser - cooling in current ion trap experiments , and the work outside the wer is attracting much interest of experimentalists .
one possible way@xmath20}$ ] to experimentally realizing the ser is that the ion is first cooled within the ldl and under the wer , and then the trap frequency is decreased by opening the trap adiabatically so that the ratio of the rabi frequency to the trap frequency is increased to a large number . therefore , we believe that the difficulties in realizing the ser will soon be overcome , and our work would be helpful for any possible future exploration of the quantum properties of the ion - trap system outside the wer .
the author sincerely thanks m.a.kornberg for his carefully reading the manuscript . the work is partly supported by national natural science foundation of china .
for obtaining a more specific expression of the eigenenergy for the case ( 2 ) in sec.ii , we can suppose @xmath149 for the case of @xmath150 and @xmath151 for @xmath152 .
direct algebra on eq.(7 ) can present us following two second - order differential equations @xmath153 and @xmath154 with @xmath155 , @xmath156 and @xmath157 $ ] .
so we can obtain the eigenenergies @xmath158 for the case of @xmath150 and @xmath159 for the case of @xmath152 .
r.blatt , j.i.cirac , a.s.parkins , and p.zoller , phys.scrip.t * 59 * , 294(1995 ) .
d.j.wineland , c.r.monroe , c.sackett , d.kielpinski , m.rowe , v.meyer and w.itano , ann.phys.*9 * , 851(2000 ) .
f.diedrich , j.c.bergquist , w.m.itano , and d.j.wineland , phys.rev.lett . * 62 * , 403 ( 1989 ) ; c.monroe , d.m.meekhof , b.e.king , s.r.jefferts , w.m.itano , and d.j.wineland , phys .
lett . * 75 * , 4011(1995 ) ; d.m.meekhof , c.monroe , b.e.king , w.m.itano and d.j.wineland , phys.rev.lett .
* 76 * , 1796(1996 ) ; c.monroe , d.m.meekhof , b.e.king and d.j.wineland , science , * 272 * , 1131(1996 ) .
j.i.cirac and p.zoller , phys.rev.lett .
* 74 * , 4091(1995 ) ; c.monroe , d.leibfreid , b.e.king , w.m.itano , and d.j.wineland , phys.rev.a * 55 * , r2489(1997 ) ; l.x.li and g.c.guo , phys.rev.a*60 * , 696(1999 ) ; j.i.cirac , and p.zoller , nature , * 404 * , 579(2000 ) ; k.m@xmath160lmer and a.s@xmath160rensen , phys.rev.lett . *
82 * , 1935(1999 ) ; a.s@xmath160rensen and k.m@xmath160lmer , phys.rev.lett . *
82 * , 1971(1999 ) ; phys.rev.a * 62 * , 022311(2000 ) .
c.monroe , d.m.meekhof , b.e.king , w.m.itano and d.j.wineland , phys.rev.lett . * 75 * , 4714(1995 ) ; b.e.king , c.s.wood , c.j.myatt , q.a.turchett , d.leibfried , w.m.itano , c.monroe and d.j.wineland , phys.rev.lett . *
81 * , 1525(1998 ) ; c.a.sackett , d.kielpinski , b.e.king , c.langer , v.meyer , c.j.myatt , m.rowe , q.a.turchette , w.m.itano , d.j.wineland and c.monroe , nature , * 404 * , 256(2000 ) .
j.f.poyatos , j.i.cirac , and p.zoller , phys.rev.lett . *
81 * , 1322(1998 ) .
c.a.blockley , d.f.walls , h.risken , europhys.lett.*17 * , 509(1992 ) ; j.i.cirac , a.s.parkins , r.blatt , and p.zoller , phys.rev.lett.*70 * , 556 ( 1993 ) ; opt.commun . *
97 * , 353(1993 ) ; j.i.cirac , r.blatt , a.s.parkins , and p.zoller , phys.rev.lett . * 70 * , 762(1993 ) .
w.vogel and r.l.de.m.filho , phys.rev.a * 52 * , 4214(1995 ) ; r.l.de.m.filho and w.vogel , phys.rev.lett . ,
* 76 * , 608(1996 ) ; j.steinbach , j.twamley and p.l.knight , phys.rev.a * 56 * , 4815(1997 ) .
k.zaheer and m.s.zubairy , phys.rev.a * 37 * , 1628(1988 ) ; s.j.d.phoenix , j.mod.opti .
* 36 * , 1163(1989 ) .
j.f.poyators , j.i.cirac , r.blatt and p.zoller p , phys.rev.a * 54 * , 1532(1996 ) .
s.zheng , x.zhu and m.feng , phys.rev.a * 62 * , 033807(2000 ) .
j.i.cirac , a.s.parkins , r.blatt and p.zoller , adv.atom.mole.opt.phys . * 37 * , 237(1996 ) .
g.morigi , j.eschner and c.h.keitel , phys.rev.lett . * 85 * , 4458(2000 ) .
d.jonathan , m.b.plenio and p.l.knight , phys.rev.a*62 * , 042307(2000 ) .
h.zeng , f.lin , y.wang and y.segawa , phys.rev.a * 59 * , 4589(1999 ) .
m.feng , x.zhu , x.fang , m.yan and l.shi , j.phys.b*32 * , 701(1999 ) .
h.moya-cessa , a.vidiella-barranco , j.a.roversi , d.s.freitas and s.m.dutra , phys.rev.a*59 * , 2518(1999 ) . j.w.negele and h.orland , quantum many - particle systems , ( addison - wesley publishing company , 1987)p.20 .
m.ku@xmath161 and m.lewenstein , j.phys.a * 19 * , 305(1986 ) ; m.feng , x.fang , x.zhu , l.shi and k.gao , comm.theor.phys.,*34 * , 1(2000 ) .
g.hose and h.s.taylor , phys.rev.lett . , * 51 * , 947(1983 ) ; r.graham and m.hhnerbach , phys.lett a * 101 * , 61(1984 ) ; z.phys.b*57 * , 233(1985 ) .
g.qin , k.wang , t.li , r.han and m.feng , phys.lett.a * 239 * , 272(1998 ) .
fig.2 variation of e with respect to @xmath26 in the case of @xmath163 , where solid curves from the bottom to top are rwa solutions of eq.(10 ) , corresponding to @xmath1640 , 1 , 2 , 3 , 4 , 5 and 6 , respectively , dot - dashed curve is the solution of eq(13 ) and dashed curves are from eq.(a3 ) or eq.(a4 ) ( because the solutions of eqs.(a3 ) and ( a4 ) are identical ) .
( a ) demonstrates @xmath165 , and ( b ) is for @xmath166 .
fig.3 variation of e with respect to @xmath26 in the case of @xmath167 , where solid curves from the bottom to top are rwa solutions of eq.(12 ) for the case of @xmath165 , corresponding to @xmath1640 , 1 , 2 , 3 , 4 , 5 and 6 , respectively , the dot - dashed curve is the solution of eq(13 ) and dashed curves are from eq.(a3 ) for cases of @xmath168 respectively . | the raman interaction of a trapped ultracold ion with two travelling wave lasers is studied analytically with series solutions , in the absence of the rotating wave approximation ( rwa ) and the restriction of both the lamb - dicke limit and the weak excitation regime .
the comparison is made between our solutions and those under the rwa to demonstrate the validity region of the rwa . as a practical example
, the preparation of schrdinger - cat states with our solutions is proposed beyond the weak excitation regime . |
the collider facilities such as the large hadron collider ( lhc ) and astrophysical experiments are two main tools for exploring new physics beyond the standard model ( sm ) . recently , the plank collaboration , designed to detect the cosmic microwave background ( cmb ) temperature anisotropy and polarization , has reported the latest value of tensor - to - scalar ratio @xmath6 at @xmath7 cl @xcite .
this data excludes a few well known inflation models such as quadratic inflation , but still allows some simple examples such as starobinsky - like inflation @xcite , @xmath8-attractor inflation @xcite and higgs inflation @xcite . among these survivors ,
higgs inflation is rather special due to two considerations .
firstly , there is only one new parameter @xmath0 in this model in the lagrangian in this model , which reads as in the jordan frame , @xmath9 here , @xmath0 measures the non - minimal coupling between higgs scalar @xmath10 and gravity .
it was found that the higgs scalar potential @xmath11 in the einstein frame rapidly approaches to a plateau potential in the large field region @xmath12 , and the model predicts that for e - foldings number @xmath13 , @xmath14 which is in perfect agreement with the plank 2015 data @xcite . in terms of the present cosmological data
@xmath15 in eq.([lagrangian ] ) approximates to the reduced plank mass @xmath16 gev for our present universe .
moreover , unlike starobinsky - like inflation or @xmath8-attractor inflation , it is obvious that higgs inflation is of special interest from the viewpoint of particle physics .
unfortunately , there are two subjects against the higgs inflation .
the first subject shows that the sm is not valid above scale @xmath2 @xcite .
it implies that the plateau potential @xmath17 at large field value @xmath12 may be significantly modified by the ultra - violet completion so that realistic inflation driven by the higgs field can not occur at all .
it is proposed in @xcite that a scale symmetry in the ultra - violet completion may be the prescription to this problem .
the second subject arises since the discovery of the higgs mass @xmath18 gev @xcite , which implies that the sm higgs quartic coupling @xmath19 at the weak scale is not large enough so that @xmath20 above the high - energy scale @xmath21 gev .
the uncertainty about this critical scale mainly arises from the the uncertainty of top quark mass , as the renormalization group equation ( rge ) for @xmath19 is very sensitive to the top yukawa coupling .
obviously , positive @xmath19 is required during inflationary epoch .
this problem can be solved through introducing some new fields into the sm , which give rise to either correct threshold correction to @xmath19 or slowing down the rg evaluation for @xmath19 along high energy scale , see , .e.g .
, @xcite . in this letter
, we consider saving the higgs inflation with high - scale supersymmetry ( susy ) .
high - scale susy can provide the observed higgs mass at the lhc and stable dm candidate @xcite .
however , the detection of such models at colliders is unpromising . in this sense
the astrophysical probe is an important direction .
fortunately , they can be probed directly or indirectly in the light of the cosmological observations on the early universe @xcite .
we identify the cut - off scale @xmath22 as the typical mass scale @xmath23 in the susy mass spectrum . above the scale @xmath24 supergravity
is the natural ultra - violet ( uv ) completion , and for the matter content same as the next - to - minimal susy model ( nmssm ) it may maintain the plateau potential @xcite . for earlier discussions , see @xcite . on the other hand , the susy dark matter ( dm ) , whose mass is around the weak scale , is the natural choice on the new fields added to the sm below the cut - off scale @xmath23 . in this paper
, we consider two possibilities @xcite , where wino @xmath25 and higgsino / singlino mixing state serves as the dm , respectively .
the paper is organized as follows . in the next section
, we discuss the rge for @xmath19 in two classes of high - scale susy below scale @xmath23 . in section iii , we discuss embedding the higgs inflation to supergravity , where we uncover the constraint on parameters @xmath19 and @xmath0 at the end of inflation @xmath26 arising from the present cosmological data . in section iv , we discuss the constraint on the susy mass spectrum in terms of the constraint on the threshold correction to @xmath19 .
finally , we discuss our results in section v.
below the scale @xmath23 , the effective theory is described by the sm together with either wino - like or higgsino / singlino - like dm . in this section
we use the 2-loop rges for relevant couplings for our analysis .
we refer the reader to appendix in our previous work @xcite , where the 2-loop rges for sm gauge and yukawa couplings , and higgs quartic coupling in models of sm@xmath27 and sm@xmath28 are explicitly shown .
we ignore other sm yukawa couplings in the following discussion . for the calculation of beta functions ,
see the references therein . in the sm ( black )
, sm@xmath27 ( blue ) and sm@xmath28 ( green ) , respectively .
the solid lines correspond to the central value @xmath29 gev , @xmath30 gev and @xmath31 .
the dotted lines show the uncertainty due to the top quark mass with 1@xmath32 deviation . in the model of sm@xmath28
, we have chosen small value @xmath33 .
see the text for the reason for this choice.,scaledwidth=45.0% ] fig.[lambda1 ] shows the rg running for @xmath19 in high - scale susy discussed in this letter , which is modified by the susy dm ( wino or higgisno - singlino mixing state ) with mass of weak scale in compared with the sm . in this figure ,
black , blue and green curves corresponds to the sm , sm@xmath27 , and sm@xmath28 , respectively .
the solid lines refer to the central value @xmath29 gev , @xmath30 gev and @xmath31 , while the dotted lines show the uncertainty due to the top quark mass with 1@xmath32 deviation . from fig.[lambda1 ] we observe that more positive @xmath19 is obtained in high - scale susy with wino dm for smaller top quark pole mass @xmath34 gev along rg running to scale @xmath2 .
this is mainly due to the correction to the sm beta function @xmath35 induced by the wino dm .
unlike the case with wino dm , the sign of the correction induced by higgsino / singlino dm is only positive when model parameter @xmath36 is small .
otherwise , for large @xmath37 the sign of the correction will reverse and @xmath19 will be negative more rapidly along rg running to scale @xmath2 .
this implies that the correction in this case is upper bounded .
but with @xmath29 gev and @xmath18 gev.,scaledwidth=45.0% ] fig.[lambda2 ] shows the sensitivity of the rg running for @xmath19 to the higgs mass .
one finds that for about @xmath38 gev deviation to the central value @xmath30 gev the correction in each case is tiny . among the three models
, @xmath19 approaches to zero mostly in the case with wino dm , similarly to what fig.[lambda1 ] indicates . in summary , for about @xmath39 deviation to the central value of top quark mass ,
@xmath19 is tuned to be more positive in high - scale susy with the low - energy theory below scale @xmath2 described by either sm@xmath27 or sm@xmath40higgsino / singlino dm in compared with the sm . as we will see in the next section , more positive @xmath19 at scale @xmath2
is favored by the higgs inflation . because the threshold correction at scale @xmath2 , may be unable to tune @xmath19 into a positive parameter above this scale as required by higgs inflation .
we proceed to discuss the constraint arising from the condition of plateau potential for higgs scalar in the context of supergravity , which is automatically the ultra - violet completion . for this purpose , we focus on the scalar - gravity part of supergravity lagrangian , which is defined by the frame function @xmath41 , kahler potential @xmath42 and superpotential @xmath43 .
we use @xmath44 to label the chiral superfields including the two higgs doublets @xmath45 .
we follow the notation and conventions in @xcite , where the kahler potential and frame function are given by in unit of plank mass , @xmath46 given the explicit form of these functions , the scalar potential in the einstein frame is directly derived from the well - known formula , . ]
@xmath47 where @xmath48 , @xmath49 represents the d - term contribution . the first attempt to embedding higgs inflation into supergravity
was shown in @xcite . in the light of @xcite
there are two important observations .
@xmath50 , the supergravity version of eq.([lagrangian ] ) requires a holomorphic function @xmath51 in order to reproduce the @xmath0-term .
@xmath52 is a dimensionless coupling constant .
@xmath53 , the matter content of mssm is not viable for the higgs inflation . by following this line ,
the authors in @xcite proposed that the matter context of nmssm is a realistic choice , in which the potential in the nmssm depends on three complex superfields , @xmath54 , as long as the scalar @xmath55 in the singlet superfiled @xmath56 can be stabilized at @xmath57 and @xmath58-flat condition @xmath59 is satisfied .
this idea was firstly considered in @xcite , but a new @xmath60-term should be added to the frame function @xcite for stabilizing the @xmath55 scalar . in summary ,
the frame function and superpotential for the purpose of higgs inflation are given by , respectively , @xmath61 substituting eq.([frame2 ] ) into eq.([potential ] ) gives rise to the potential , @xmath62 for @xmath57 and @xmath59 .
here we have used @xmath63 . for more details about the stabilization of @xmath55 and angles in eq.([z ] ) , see @xcite . in the region
@xmath64 , eq.([potential2 ] ) approaches to @xmath65 ( in unit of plank mass ) , which verifies our statements above .
moreover , in terms of eq.([potential2 ] ) one can also verify eq.([nr ] ) .
parameters @xmath0 and @xmath19 in eq.([lagrangian ] ) are related to parameters @xmath52 and @xmath66 in eq.([frame2 ] ) as , @xmath67 here @xmath68 denotes the rg scale corresponding to inflation .
they are constrained by the present cosmological data as follows .
recall that @xmath69 , where @xmath70 is the amplitude of the power spectrum of the curvature perturbation and @xmath71 in the context of single field inflation . for @xmath72
as reported by plank collaboration @xcite one obtains , @xmath73
until now , we have obtained the value of @xmath19 below the rg scale @xmath74 ( as shown in fig.[lambda1 ] and fig .
[ lambda2 ] ) and above the rg scale @xmath75 ( as shown in eq.([constraint ] ) ) given a choice on @xmath0 .
the effective theory at the intermediate scale between scale @xmath2 and @xmath75 is described by the sm together with gauginos and squarks .
other susy particles such as higgsinos and charged higgs have masses of order @xmath75 @xcite , so they should be integrated out at this intermediate energy scale @xcite , especially for the discussion about rge for @xmath19 .
the threshold correction @xmath76 at scale @xmath77 arising from integrating out gauginos and squarks can be determined in terms of the differences between the values of @xmath19 below and above scale @xmath77 , @xmath78 where @xmath79 , and @xmath80 , @xmath81 , @xmath82 refers to the gluino mass , bino mass and squark masses , respectively .
as we will see below , the threshold correction @xmath83 can be measured in high precision , so it is a new and useful factor to constrain the gut - scale susy mass spectrum .
the value of @xmath84 in eq.([threshold1 ] ) is determined by the rge for @xmath35 in the effective theory at the intermediate scale , with the boundary value at the end of inflation @xmath85 here , we have ignored the effects due to higgsino - induced operators with mass dimension higher than four , which are at least one order of magnitude smaller than the threshold correction in our case . with mass dimension six ,
as induced by heavy susy particles with mass @xmath15 , contribute to the leading corrections to the rges for sm ew gauge coupling @xcite . as a result
, they modify the rge for @xmath86 indirectly , the significance of which is determined by the ratio @xmath87 , as seen from the modified beta function to sm gauge coupling , @xmath88 . for the three stages we set in this paper ,
the modification to the rge for @xmath86 during inflation is the maximal , with heavy susy particle identified as @xmath89 .
in contrast , the modification after inflation is only mild . in our case , @xmath90 and @xmath91 , which implies that this modification is smaller than the threshold correction @xmath92 . ] the threshold corretion @xmath93 in eq.([bound ] ) arises from the heavy higgsinos and singlet @xmath55 , which are given by , respectively @xcite , @xmath94 where we have assumed that the a - term @xmath95 is smaller than @xmath96 . substituting the stabilized value @xmath59 into eq.([values ] )
we find that @xmath97 .
as long as the mass scale @xmath98 , which is of order @xmath75 , is far larger than the hubble parameter @xmath99 and the reheating temperature after inflation @xcite , @xmath100 rapidly approaches to the stabilized value during and after inflation .
the matter content of the effective theory at the intermediate scale is composed of sm fermions , squarks , gauginos and electrically neutral scalars in the higgs sector .
the beta function coefficients @xmath101 for the sm gauge coupling in this effective theory are given by , respectively , @xmath102 where @xmath103 is the number of sm fermion generations . in terms of eq.([rges ] ) and one - loop rges for @xmath19 and sm top yukawa coupling we show in fig.[threshold ] the constraint on the magnitude of @xmath83 for the range @xmath104 .
it is shown that the value for @xmath83 as required by higgs inflation is very sensitive to parameter @xmath0 . in the light of fig.[threshold ]
we conclude that for small non - minimal coupling @xmath1 , the threshold correction at scale @xmath2 induced by higgs inflation is about @xmath105 and @xmath106 in wino and higgsino / singlino dm , respectively .
the uncertainties mainly arise from the uncertainty of top quark pole mass . in the large @xmath0 region with @xmath107 , the required threshold correction is @xmath108 , which is too large to exclude such models . in the wino dm model .
we have chosen @xmath34 gev , @xmath30 gev and @xmath31 .
similar plot can be obtained for the case with higgsino / singlino dm.,scaledwidth=45.0% ]
so far we have only discussed the constraint on the threshold correction @xmath109 at scale @xmath2 in the case for wino dm . in the case for higgsino / singlino dm
the one - loop rges for relevant couplings at the intermediate scale are the same as in section iv .
so the conclusion there is also applied to the model of higgsino / singlino dm .
actually , given the same @xmath0 it is expected that the threshold correction for the case with higgsino / singlino dm is larger than what is shown in fig.[threshold ] . one may wonder whether adopting larger @xmath110 than @xmath111 gev can reduce the threshold correction in fig.[threshold ] @xmath112 fig.[lambda1 ] and fig.[lambda2 ] have shown that @xmath113 becomes more negative when one chooses larger @xmath110 , and the value of @xmath114 , which is positive , will be smaller simultaneously .
therefore the deviation to the threshold correction is not too obvious to violate our conclusions . in summary
, the answer to the question whether higgs inflation can be saved by high - scale susy critically depends on the magnitude of @xmath0 . for small non - minimal coupling @xmath1
, the threshold correction at scale @xmath2 required by higgs inflation is constrained in high precision , the magnitude of which is about @xmath105 and @xmath106 for the wino and higgsino / singlino dm , respectively .
the uncertainties mainly arise from the uncertainty of top quark pole mass .
this amount of threshold correction can be explained , for example , for universal squark masses @xmath115 with vanishing @xmath116 terms . for large @xmath117 ,
the required threshold correction is smaller than @xmath118 , which excludes such high - scale susy .
finally , we believe our analysis can be applied to split susy , although this model is more complicated in compared with what have been discussed here .
we would like to thank the referee for valuable suggestions . the work is supported in part by national natural science foundation of china under grant no . 11247031 and 11405015 .
v. khachatryan _ et al . _
[ cms collaboration ] , `` observation of the diphoton decay of the higgs boson and measurement of its properties , '' eur .
j. c * 74 * , no .
10 , 3076 ( 2014 ) , arxiv:1407.0558 [ hep - ex ] .
g. aad _ et al . _ [ atlas collaboration ] , `` measurement of the higgs boson mass from the @xmath119 and @xmath120 channels with the atlas detector using 25 fb@xmath121 of @xmath122 collision data , '' phys .
d * 90 * , no . 5 , 052004 ( 2014 ) , arxiv:1406.3827 [ hep - ex ] .
e. e. jenkins , a. v. manohar and m. trott , `` renormalization group evolution of the standard model dimension six operators i : formalism and lambda dependence , '' jhep * 1310 * , 087 ( 2013 ) , arxiv:1308.2627 [ hep - ph ] . | it is shown whether higgs inflation can be saved with high - scale supersymmetry critically depends on the magnitude of non - minimal coupling constant @xmath0 . for small @xmath1 , the threshold correction at scale
@xmath2 is constrained in high precision .
its magnitude is in the narrow range of @xmath3 and @xmath4 for the wino and higgsino / singlino dark matter , respectively . while in the large @xmath0-region with @xmath5 ,
such high - scale supersymmetry is excluded by too large threshold correction as required by higgs inflation . |
in the next generation of @xmath8 colliders , multiple vector - boson production will provide a crucial test of the gauge structure of the standard model ( sm ) , since it will allow the study of the triple and quartic vector - boson couplings .
these vertices are strictly constrained by the @xmath9 gauge invariance and any small deviation from the sm predictions spoils the precise cancellation of the high - energy behavior between the various diagrams , giving rise to an anomalous growth of the cross section with energy .
therefore , the careful study of multiple vector - boson production , and consequently of vector - boson self - interactions , can give important clues about the existence of new particles and/or interactions beyond the sm .
the reaction @xmath10 will be accessible at lep ii and some information about the @xmath11 and @xmath12 vertices will be available in the near future @xcite .
nevertheless , we will have to wait for colliders with higher center of mass energies in order to produce a final state with three or more gauge bosons and to test the quartic gauge boson interaction .
this will be accomplished at the next linear @xmath3 collider ( nlc ) @xcite , which will reach an energy between 500 and 2000 gev with an yearly integrated luminosity of at least @xmath13 fb@xmath14 .
an interesting feature of these new machines is the possibility of transforming an electron beam into a photon one through the laser backscattering mechanism @xcite .
this process will allow the nlc to operate in three different modes , @xmath3 , @xmath15 , and @xmath4 , opening up the opportunity for a wider search for new physics .
however , it is important to point out that the collider can operate in only one of its three modes at a given time , therefore , it is essential to study comparatively the different features of each of these setups . the triple and quartic gauge
boson vertices , in the framework of gauge theories , have a common origin , and consequently , a universal strength dictated by the gauge symmetry of the model .
notwithstanding , in a more general context , anomalous quartic couplings can arise as the low energy limit of heavy state exchange , whereas trilinear couplings are modified by integrating out heavy fields . in this sense , it is possible to conceive extensions of the sm where the trilinear couplings remain unchanged , while the quartic vertices receive new contributions .
for instance , the introduction of a new heavy scalar singlet , that interacts strongly with the higgs sector of the sm , enhances the four vector - boson interaction without affecting either the triple vector - boson couplings or the sm predictions for the @xmath16 parameter @xcite .
therefore , the measurement of the three vector
boson production cross section can provide further non trivial tests of the sm that are complementary to the analysis of the production of vector boson pairs .
the cross section for the production of multiple gauge bosons , in the context of the sm , has already been evaluated for @xmath3 colliders operating in the @xmath8 @xcite , @xmath15 @xcite , and @xmath4 @xcite modes .
there has also been some studies of anomalous quartic vertices through the reactions @xmath17 @xcite , @xmath18 @xcite , and @xmath19 @xcite , where @xmath20 @xmath21 , @xmath22 or @xmath7 and @xmath23 @xmath24 or @xmath25 . in this work ,
we analyze the effect of some genuinely anomalous quartic operators , _
i.e. _ operators which do not modify the trilinear vertices and can not be bounded by the lep ii measurements . furthermore , since we are interested in probing these anomalous couplings in a @xmath26 collider , we concentrate our analyses on operators that involve at least one photon .
we studied the production of three vector bosons in @xmath4 collisions through the reactions @xmath27 @xmath28 in order to impose bounds on the vertices @xmath29 , @xmath30 , and @xmath31 .
these processes involve only interactions between the gauge bosons , making more evident any deviation from the sm predictions .
moreover , there is no tree
level contribution involving the higgs boson which evades all the uncertainties coming from the scalar sector , like the higgs boson mass .
our results show that the constraint on the anomalous couplings @xmath29 , coming from the reaction [ z ] , is as restrictive as the one obtained in @xmath15 colliders @xcite and a factor of 5 better than the one coming from @xmath3 colliders @xcite . on the other hand , the bounds in the vertices @xmath30 and
@xmath32 , stemming from the processes [ g ] and [ z ] , are a factor from 2 to 5 better than the ones that can be obtained in an @xmath33 collider , however , they are worse than the ones coming from the direct reaction @xmath34 @xcite .
the next section contains the effective operators that we analyzed and the calculational method employed in this paper .
our results and discussions are presented in sec .
in order to construct effective operators associated to exclusively quartic anomalous couplings we employed the formalism of ref . @xcite .
we required the existence of a custodial @xmath35 symmetry , which avoids any contribution to the @xmath16 parameter , and we also demanded that the phenomenological lagrangians are invariant under local @xmath36 symmetry .
the lowest order operators that comply with the above requirements , involving at least one photon , are of dimension six @xcite : there are two independent @xmath37 and @xmath38 conserving operators involving two photons @xcite @xmath39 and one operator exhibiting just one photon @xcite @xmath40 where @xmath41 is the @xmath35 triplet and @xmath42 is the electromagnetic field strength . in terms of the physical fields @xmath43 , @xmath44 , and @xmath21 , the effective lagrangians ( [ lag:0 ] ) and ( [ lag : c ] ) give rise to anomalous @xmath1 and @xmath2 couplings , while ( [ lag : new ] ) generates a new @xmath45 vertex .
we should notice that the @xmath2 vertex is particularly interesting since it is completely absent in the sm .
moreover , we note that the effective operators ( [ lag:0 ] ) and ( [ lag : new ] ) can parametrize the exchange of a neutral scalar particle while the operator ( [ lag : c ] ) corresponds to the exchange of a charged particle .
therefore , we expect that the characteristic scale of the above interactions is set by the masses of the exchanged states . in the limit of very heavy scalars ,
the expected values of the coefficients @xmath46 ( @xmath47 , @xmath48 , @xmath49 ) should be in the range of @xmath50 to @xmath51 .
for the sake of definiteness , our results are presented assuming that @xmath52
. an interesting feature of the above effective lagrangians is that they will not be directly constrained by lep ii since they do not contribute to triple gauge boson vertices . in order to obtain some bounds on these couplings at low energies
, we must rely on their contribution to one - loop processes . from the analysis of oblique radiative corrections to the @xmath21 physics @xcite
, we estimated that @xmath53 and @xmath54 at @xmath55 level , while the coupling @xmath56 is not constrained at all . in order to evaluate the triple vector boson production at nlc
, we assumed that this machine will reach a center of mass energy of @xmath57 ( @xmath58 ) gev with a yearly integrated luminosity of 10 fb@xmath14 .
the most promising mechanism to generate hard photon beams in an @xmath8 linear collider is laser backscattering .
assuming unpolarized electron and laser beams , the backscattered photon distribution function @xcite is @xmath59 \ ; , \label{f : l}\ ] ] with @xmath60 where @xmath61 is the compton cross section , @xmath62 , @xmath63 and @xmath64 are the electron mass and energy respectively , and @xmath65 is the laser - photon energy .
the quantity @xmath66 represents the ratio between the scattered photon and initial electron energy and its maximum value is @xmath67 in what follows , we assumed that the laser frequency is such that @xmath68 , which leads to the hardest possible spectrum of photons with a large luminosity .
the cross section for the triple vector boson production via @xmath4 fusion can be obtained by folding the elementary cross section for the subprocesses @xmath69 ( @xmath70 ) with the photon - photon luminosity ( @xmath71 ) , _ i.e. _ , @xmath72 where @xmath73 ( @xmath74 ) is the @xmath3 ( @xmath4 ) center - of - mass energy , @xmath75 , and the photon - photon luminosity is @xmath76 the analytical calculation of the cross section for the subprocess @xmath77 ( @xmath78 ) is very lengthy and tedious despite being straightforward . in order to perform these calculations in an efficient and reliable way
, we evaluated numerically the helicity amplitudes using the techniques outlined in ref .
the phase space integrations were performed numerically using the monte carlo routine vegas @xcite . as a check of our results
, we explicitly verified that the amplitudes were lorentz and @xmath36 invariant and that the kinematic distributions for @xmath43 and @xmath44 coincided .
the total cross sections for the processes [ g ] and [ z ] are quadratic functions of the anomalous couplings @xmath46 , _ i.e. _ @xmath79 where @xmath80 stands for the sm cross section @xcite and @xmath81 ( @xmath82 ) is the interference ( pure anomalous ) contribution .
we evaluated the cross sections involved in eq.([base ] ) , imposing that the polar angles of the produced vector bosons with the beam pipe are larger than @xmath83 and assuming @xmath84 . for the process [ g ]
, we also introduced a cut on the photon transverse momentum , @xmath85 ( @xmath86 ) gev , not only to guarantee that our results are free of infrared divergences but also to mimic the performance of a typical electromagnetic calorimeter . in table
[ sigmas : z ] , we present our results assuming in each case that only one anomalous coupling is nonvanishing . in order to quantify the effect of the new couplings , we defined the statistical significance @xmath87 of the anomalous signal @xmath88 which can be easily evaluated using the parametrization ( [ base ] ) with the coefficients given in table [ sigmas : z ] .
we list in table [ const ] the values of the anomalous couplings that correspond to a @xmath89 effect in the total cross section for the different processes , assuming an integrated luminosity @xmath90 fb@xmath14 for the correspondent @xmath3 collider .
it is interesting to point out that the most restringent bounds come from the process [ g ] and that our results are quite insensitive to different choices of the angular and @xmath91 cuts , as we can learn from table [ const ] .
we can see from table [ const ] that our @xmath89 limits for the coupling @xmath56 are a factor of @xmath92 better than the limits obtained in @xmath3 collisions @xcite , while they are of the same order of the ones originating from an @xmath15 machine @xcite .
the @xmath4 mode of the nlc is more efficient for studying this anomalous quartic coupling since it leads to cross sections for the production of three vector bosons that is more than one order of magnitude larger than the ones for a conventional @xmath3 collider with @xmath93 @xcite .
the limits on the anomalous coupling @xmath94 that can be established through the triple gauge boson production are one order of magnitude better than the ones coming from the @xmath8 mode and are comparable to the constraints that can be obtained in the @xmath15 mode .
however , the bounds shown in table [ const ] are a factor of 2 weaker than the ones obtained from the direct reaction @xmath34 ( @xmath95 ) , since the triple gauge boson production is a higher order process in @xmath96 .
our limits on the coupling @xmath97 are slightly better than the ones arising from the @xmath8 mode , while they are almost one order of magnitude worse than the ones obtained in @xmath15 @xcite or direct @xmath98 ( @xmath95 ) @xcite collisions .
the kinematical distributions of the final state particles can be used , at least in principle , to increase the sensitivity of the @xmath4 reactions to the anomalous couplings , improving consequently the bounds on the new interactions .
more than that , they could furnish further information that would allow us to distinguish among the different anomalous interactions . in figs .
[ fig:1 ] to [ fig:4 ] , we exhibit some representative distributions for the processes [ g ] and [ z ] , adopting the values of the anomalous coupling constants that lead to a @xmath89 deviation in the total cross section .
we present in fig .
[ fig:1 ] the normalized @xmath99 distribution for the production of @xmath100 , where @xmath101 is polar angle of the vector boson with respect to the beam pipe . from this figure
, we can see that the anomalous @xmath102 distributions differ from the sm one .
for instance , we can identify the cases with negative values of @xmath97 and @xmath94 , but we can not discriminate between the positive values of these anomalous couplings .
moreover , it is interesting to notice that the anomalous couplings enhance the production of @xmath22 in the central region of the detector , where they can be more easily reconstructed .
figure [ fig:2 ] contains the normalized distribution for the invariant mass ( @xmath103 ) of vector boson pairs @xmath104 for the process [ g ] . here
we can observe that the presence of the anomalous interactions increases the invariant mass of the @xmath104 pairs since the new couplings are proportional to the photon momentum .
furthermore , this distribution allows us to separate the negative @xmath97 and @xmath94 couplings , while the positive values of the anomalous couplings lead to distributions similar to the sm ones .
we notice in our analyses that the distributions involving photons are less sensitive to the anomalous couplings than the @xmath22 ones .
although the process @xmath105 has a lower sensitivity , when compared with the @xmath100 production , it exhibits some interesting features , as can be seen from figs.[fig:3 ] and [ fig:4 ] .
we show in fig .
[ fig:3 ] the normalized @xmath21 transverse momentum distributions for the process [ z ] using the @xmath97 , @xmath94 , and @xmath56 anomalous couplings .
we can learn from this figure that the anomalous couplings favor the @xmath21 s to have a higher @xmath106 since the momentum dependence of the anomalous vertices enhances their contribution at higher energies .
once again , we can distinguish between the different anomalous interactions provided the couplings @xmath97 and @xmath94 are negative , since these couplings lead to larger purely anomalous contributions .
it is interesting to remark that our analyses show that the @xmath107 distribution does not allow us to separate clearly the different anomalous interactions . in fig .
[ fig:4 ] , we show the @xmath22 rapidity distributions for the process [ z ] .
the anomalous interactions lead to a more copious production of @xmath22 in the central region and it is also possible to have an indication of which effective interaction is responsible for the departure from the sm predictions .
we found out that , as happened in @xmath100 process , it is not possible to distinguish between the anomalous interactions through the rapidity distribution of the neutral vector bosons .
the bottom line of this work is that the study of the production of @xmath100 and @xmath108 in a @xmath4 collider will be able to increase the potential of @xmath3 machines to search for anomalous four gauge boson interactions .
furthermore , the kinematical distributions of the final state particle can also help to distinguish between some of these interactions .
we collect in this appendix the expressions for anomalous contributions to the the amplitudes of the processes @xmath5 , with @xmath6 or @xmath7 .
the standard model expressions can be found in ref .
the momenta and polarizations of the initial photons were denoted by ( @xmath109 , @xmath110 ) and [ @xmath111 , @xmath112 , while the momenta and polarizations of the final state @xmath43 , @xmath44 and @xmath113 are given by ( @xmath114 , @xmath115 , @xmath116 ) and [ @xmath117 , @xmath118 , @xmath119 respectively .
the amplitude of these processes can be written as @xmath120\;,\ ] ] where @xmath121 is the standard model invariant amplitude , and @xmath122 , represents the different anomalous contributions .
the factor @xmath123 depends upon the process under consideration : @xmath124 for the production of @xmath100 and @xmath125 for the production of @xmath108 .
we can write a compact expression for the anomalous amplitudes of the process ( [ g ] ) , in the form @xmath126^{\mu\nu\alpha\beta\gamma}\ ; , \nonumber \\ \left(m_{\text{ano}}^\gamma \right)^{\mu\nu\alpha\beta\gamma } ( a_c ) = \left [ m_1(a_c ) + m_2(a_c ) + m_3(a_c ) + m_4(a_c)\right]^{\mu\nu\alpha\beta\gamma}\ ; , \label{ano : g}\end{aligned}\ ] ] while for the process ( [ z ] ) we have @xmath127^{\mu\nu\alpha\beta\gamma}\ ; , \nonumber \\
\left(m_{\text{ano}}^z \right)^{\mu\nu\alpha\beta\gamma}(a_c ) = \left[m_3(a_c ) + m_4(a_c ) + m_5(a_c)\right]^{\mu\nu\alpha\beta\gamma}\ ; , \nonumber \\ \left(m_{\text{ano}}^z \right)^{\mu\nu\alpha\beta\gamma } ( a_n ) = \left[m_1(a_n ) + m_2(a_n ) + m_6(a_n)\right]^{\mu\nu\alpha\beta\gamma}\ ; , \label{ano : z}\end{aligned}\ ] ] with @xmath128 \ ; , \nonumber \\
m_2^{\mu\nu\alpha\beta\gamma } ( i ) & = & \delta^{\mu\alpha\xi}(k_1,-p_+ ) d^w_{\xi\lambda}(k_1-p_+ ) \gamma^{\lambda\beta\nu\gamma}_{(i ) } ( k_2,-k_3,-p_-,k_1-p_+ ) + \ ; [ k_{1\leftrightarrow 2}\ ; ; \;\mu\leftrightarrow\nu ] \ ; , \nonumber \\
m_3^{\mu\nu\alpha\beta\gamma } ( i ) & = & \delta^{\alpha\gamma\xi}(p_+,k_3 ) d^w_{\xi\lambda}(p_++k_3)\gamma^{\lambda\beta\nu\mu}_{(i)}(k_1,k_2,0,0 ) \ ; , \nonumber \\
m_4^{\mu\nu\alpha\beta\gamma } ( i ) & = & \delta^{\gamma\beta\xi}(k_3,p_- ) d^w_{\xi\lambda}(-p_--k_3)\gamma^{\lambda\alpha\nu\mu}_{(i)}(k_1,k_2,0,0 ) \ ; , \nonumber \\
m_5^{\mu\nu\alpha\beta\gamma } ( i ) & = & \frac{1}{\cos{\theta_w}^2}\delta^{\alpha\beta\xi}(p_-,p_+ ) d^z_{\xi\lambda}(p_++p_-)\gamma^{\lambda\gamma\nu\mu}_{(i)}(k_1,k_2,0,0 ) \ ; , \nonumber \\
m_6^{\mu\nu\alpha\beta\gamma } ( i ) & = & -\frac{a_n\sin\theta_w}{16 \cos{\theta_w}^2 \lambda^2 } \biggl ( 2 p_2^\mu g^{\alpha\beta } g^{\nu\gamma } - 2 p_2^\gamma g^{\alpha\beta } g^{\mu\nu } - p_2^\mu g^{\alpha\gamma } g^{\nu\beta } + p_2^\beta g^{\mu\nu } g^{\alpha\gamma } \nonumber \\ - p_2^\mu g^{\beta\gamma } g^{\alpha\nu } & + & p_2^\alpha g^{\mu\nu } g^{\beta\gamma } - p_2^\alpha g^{\mu\beta } g^{\nu\gamma } + p_2^\gamma g^{\mu\beta } g^{\nu\alpha } - p_2^\beta g^{\mu\alpha } g^{\nu\gamma } + p_2^\gamma g^{\mu\alpha } g^{\nu\beta } + 1 \leftrightarrow 2 \biggr ) \ ; . \label{m123}\end{aligned}\ ] ] in eq .
( [ m123 ] ) , @xmath129 , @xmath130 , represents the vector boson propagator in the unitary gauge , @xmath131 is the usual triple gauge boson vertex function @xmath132 \ ; , \ ] ] and , @xmath133 are the different four gauge boson couplings for the anomalous interactions @xmath134 \ ; , \label{rf : a0}\end{aligned}\ ] ] @xmath135 \ ; , \label{rf : ac}\end{aligned}\ ] ] @xmath136 \nonumber \\ & - & g^{\nu\beta } ~ \left [ g^{\mu\alpha } ~ q_1.(q_2 - q_4 ) - q_{1}^ { \mu } ( q_2 - q_4)^\alpha \right ] \nonumber \\ & + & g^{\mu\nu } ~ \left [ g^{\alpha\beta } ~ q_1.(q_3 - q_4 ) - ( q_3 - q_4 ) ^\alpha q_{1}^ { \beta } \right ] \nonumber \\ & - & q_{2}^ { \mu } \left(g^{\alpha\nu } ~ q_{1}^ { \beta } - g^{\alpha\beta } ~ q_{1}^ { \nu } \right ) + q_{2}^ { \nu } \left(g^{\alpha\mu } ~ q_{1}^ { \beta } - g^{\alpha\beta } ~ q_{1}^ { \mu } \right ) \nonumber \\ & - & q_4^ { \beta } \left(g^{\alpha\mu } ~ q_{1}^ { \nu } - g^{\alpha\nu } ~ q_{1}^ { \mu } \right ) + q_3^ { \beta } \left(g^{\alpha\nu } ~ q_{1}^ { \mu } - g^{\alpha\mu } ~ q_{1}^ { \nu } \right ) \nonumber \\ & - & q_3^ { \nu } \left(g^{\alpha\beta } ~ q_{1}^ { \mu } - g^{\alpha\mu } ~ q_{1}^ { \beta } \right ) + q_4^ { \mu } \left(g^{\alpha\beta } ~ q_{1}^ { \nu } - g^{\alpha\nu } ~ q_{1}^ { \beta } \right ) \biggr\ } \ ; .
\label{rf : an}\end{aligned}\ ] ] the amplitude @xmath137 comes from the new anomalous vertex @xmath138 , and @xmath139 gives rise to a five
vertex @xmath140 represented by @xmath141 , which is necessary to preserve the gauge invariance of the amplitude .
k. hagiwara , r. d. peccei , d. zeppenfeld , and k. hikasa , nucl .
phys . * b282 * , 253 ( 1987 ) ; proceedings of the ecfa workshop on lep200 , cern report 87 - 08 , ecfa report 87/108 , ed.a .
bhm and w. hoogland , aachen 1987 .
f. r. arutyunian , and v. a. tumanian , phys .
lett . * 4 * , 176 ( 1963 ) ; r. h. milburn , phys .
* 10 * , 75 ( 1963 ) ; see also c. akerlof , university of michigan report no .
umhe 81 - 59 ( 1981 ) , unpublished . | we study the constraints on the vertices @xmath0 , @xmath1 , and @xmath2 that can be obtained from triple - gauge - boson production at the next generation of linear @xmath3 colliders operating in the @xmath4 mode .
we analyze the processes @xmath5 ( @xmath6 , or @xmath7 ) and show that these reactions increase the potential of @xmath3 machines to search for anomalous four - gauge - boson interactions .
1.in _ submitted to physical review d1 _ |
this paper deals with the existence of periodic points for annulus maps .
a map is any continuous endomorphism of the open annulus @xmath3 .
the _ degree _ of an annulus map @xmath1 is the integer @xmath4 such that @xmath5 , where @xmath6 is the endomorphism of the fundamental group @xmath7 induced by @xmath1 .
we are particularly interested in the case where the degree of the map has modulus greater than one .
we will consider two different ways of saying that a map has abundance of periodic orbits .
the first one is completeness , and is related to nielsen equivalence .
the second one is the growth rate inequality which will be discussed shortly .
a map of the annulus is complete if for every @xmath8 the number of nielsen classes of fixed points of @xmath9 is equal to @xmath10 .
this is the maximum possible for a map of degree @xmath4 . roughly speaking , a continuous map @xmath11 of degree @xmath4 ,
@xmath12 is said to be complete if it has as many different nielsen classes of periodic points as the map @xmath13 acting on the circle .
nielsen equivalence and completeness are explained in the next section .
this notion of completeness is related to a great deal of problems on existence of periodic orbits in surface dynamics .
for example , a complete map always satisfies the growth rate inequality : @xmath14 it is an open problem if the rate inequality holds for differentiable maps of degree @xmath4 acting on the sphere @xmath15 ( problem 3 posed in @xcite ) .
other results related to the existence of periodic points for @xmath16 maps , ( not necessarily homeomorphisms ) were obtained by hagopian in @xcite , where some conditions are imposed on plane continua in order to obtain the existence of fixed points for any @xmath16 map defined on it .
existence of recurrence is mostly present in maps with @xmath2 because @xmath2 implies an expansion on the direction of a simple nontrivial closed curve .
this idea will be used frequently throughout the article . the covering map
@xmath17 defined in the annulus @xmath18 is a trivial example of a map without periodic points in the open annulus , and so the mere existence of periodic orbits becomes of interest . as covering maps
lift to homeomorphisms of the plane , the problem also relates to surface homeomorphisms dynamics , in particular with what is called brouwer theory . in @xcite completeness of annulus coverings of degree @xmath4 , @xmath2 , was proved under the assumption that an essential continuum is preserved . in this paper
we study the same problem for general continuous maps ( not necesarily coverings ) .
some definitions are needed before stating the results .
we say that a map @xmath19 is complete on @xmath20 if @xmath1 is complete and every nielsen class has a representative in @xmath21 . whenever @xmath21 is a compact subset of the ( open ) annulus , denote by @xmath22 the union of @xmath21 with the bounded ( relatively compact ) components of its complement .
[ t1 ] let @xmath19 be a surjective degree @xmath4 map of the annulus , and @xmath20 an essential continuum such that @xmath23 . if @xmath0 , then @xmath1 is complete on @xmath22 .
the ingredientes for the proof of this theorem are lefschetz theory ( to obtain fixed points enclosed by curves with nonzero index ) , caratheodory theory ( constructions of crosscuts with determined properties ) and nielsen theory ( existence of fixed points in every nielsen class is assured by proving that every lift of @xmath1 to its universal covering has a fixed point ) . the proof obtained fails for @xmath24 and it is easy to construct counterexamples of the proof .
it is an open problem if the assertion of the theorem is true also for negative values of @xmath4 .
the hypothesis @xmath25 is also important in the proof . assuming that @xmath21 is locally connected
, this assumption and the surjectivity of @xmath1 can be dropped , and a satisfactory conclusion is obtained .
[ t2 ] let @xmath19 be a degree @xmath4 map of the annulus , and @xmath20 a locally connected essential continuum such that @xmath26 .
if @xmath2 , then @xmath1 is complete on @xmath22 .
these statements imply the existence of periodic points in the set @xmath22 and there are examples where the periodic points are not contained in @xmath21 . posing some conditions on the behaviour at infinity of the map @xmath1 , completeness is obtained without any other assumption . to state
these conditions , note that the annulus can be compactified with two points , called ends of @xmath3 .
[ t3 ] let @xmath19 be a degree @xmath4 map of the annulus , where @xmath2 .
assume that @xmath1 extends continuously to the ends of @xmath3 .
each one of the following conditions imply that @xmath1 is complete . 1 .
both ends of @xmath3 are attracting
both ends of @xmath3 are repelling .
see sections 4,5 and 6 where completeness is obtained for maps of the annulus under different sets of hypothesis . in section 2
the notion of nielsen class is explained , in section 3 the index and lefschetz fixed point theorem are introduced without proofs , and some of the lemmata needed in the proofs of the theorems are included .
section 4 is devoted to the proof of theorem [ t1 ] . in section 5 the proofs for locally connected @xmath21
are given , and the last section contains two simple applications of these results to maps of the sphere .
two fixed points @xmath27 and @xmath28 of a map @xmath11 are said _ nielsen equivalent _ if there exists a curve @xmath29 joining @xmath27 to @xmath28 such that @xmath30 is homotopic to @xmath29 relative to its endpoints .
this is an equivalent relation whose classes are called nielsen classes .
two fixed points that are very close are equivalent .
of course the existence of a continuum of fixed points gives just one class of nielsen equivalent fixed points . if @xmath2 , then the map @xmath31 on the annulus @xmath32 has exactly @xmath33 fixed points , and also @xmath33 nielsen classes .
it is not difficult to see that a degree @xmath4 map of the annulus can not have more than @xmath33 classes of fixed points .
a map @xmath1 is _ complete _ if for every @xmath8 the number of nielsen classes of fixed points of @xmath9 is equal to that of the map @xmath34 .
so , asking for a map @xmath1 to be complete is to ask that is has the maximum possible ( non artificial ) periodic orbits .
the following proposition is a standard result in nielsen theory ( see jiang s book @xcite for further information on the subject ) and will be used to prove all the results throughout this paper .
alternatively , you can see corollary 1 in @xcite .
[ comp ] a degree @xmath4 map of the annulus has @xmath33 nielsen classes of fixed points if and only if any lift of @xmath1 to the universal covering of the annulus has a fixed point . note that in order to prove completeness of a map @xmath1 satisfying the hypothesis of theorem [ t1 ] , theorem [ t2 ] or theorem [ t3 ] , it suffices to show that any lift of @xmath1 has a fixed point . indeed
, note that in any case , every iterate @xmath35 of @xmath1 satisfies the same hypothesis as @xmath1 .
it follows that @xmath35 has @xmath36 classes of fixed points .
let @xmath37 , defined by @xmath38 denote the universal covering of the punctured plane .
the index of a plane closed curve @xmath29 defined in an interval @xmath39 $ ] such that @xmath40 for every @xmath41 , is defined as the first coordinate of @xmath42 , where @xmath43 is any lift of @xmath29 under @xmath44 : that is , @xmath45\to { { \mathbb{r}}}^2 $ ] is a curve such that @xmath46 for every @xmath41 .
this does not depend on the choice of the lift .
note that the index is an integer .
we abuse notation and make no difference between the curve and its trace in the plane . [ index ]
let @xmath29 be a closed curve and @xmath1 a continuous map defined on @xmath29 and without fixed points on @xmath29 .
define the lefschetz index of @xmath1 in @xmath29 as the index of the closed curve @xmath47 .
this index will be denoted by @xmath48 .
let @xmath29 be a simple closed curve in the plane , so that its complement has two connected components , one is bounded and denoted @xmath49 and the other unbounded , denoted @xmath50 .
the curve @xmath29 is positively oriented if the index of @xmath51 is @xmath52 whenever @xmath53 is a point that belongs to @xmath49 .
if @xmath29 is positively oriented , then @xmath49 is located to the left of @xmath29 . * lefschetz fixed point theorem . * _ let @xmath1 be a continuous self - map of the plane . _ 1 . if @xmath29 is a simple closed curve such that @xmath54 , then @xmath1 has a fixed point in the bounded component of @xmath55 .
2 . let @xmath3 be an annulus in the plane whose boundary components are positively oriented curves @xmath56 and @xmath57 , and assume that @xmath58 , then @xmath1 has a fixed point in @xmath3 .
it is our prupose to use the above result to prove existence of periodic points .
we will need some techniques for calculating indexes of curves .
the next assertion follows by continuity of @xmath48 on @xmath1 and @xmath29 whenever the index is defined .
[ l1 ] assume that @xmath29 is a closed curve and that @xmath59 is a homotopy .
if no @xmath59 has a fixed point in @xmath29 , then @xmath60 is constant .
moreover , if @xmath61 is a free homotopy and @xmath1 has no fixed points on any @xmath61 , then @xmath62 is constant .
the following is an immediate consequence : [ l3 ] let @xmath1 be fixed point free on a simple closed curve @xmath29 . if @xmath63 is a finite subset of @xmath29 , then there exist @xmath64 and @xmath65 such that the index of @xmath66 on @xmath43 is equal to the index of @xmath1 in @xmath29 whenever the following conditions hold : 1 .
@xmath43 is a simple closed curve such that @xmath67 is contained in @xmath68 , the @xmath69 neighborhood of @xmath63 , 2 .
@xmath66 is a continuous map such that @xmath70 outside @xmath68 , 3 .
the image under @xmath66 of @xmath68 is contained in @xmath71 .
the proof of this result follows from the fact that if @xmath72 and @xmath69 are sufficiently small , then both @xmath1 and @xmath66 and @xmath29 and @xmath43 are homotopic .
this previous corollary and the following two criteria is all that is needed to calculate indexes in the sequel .
assume that @xmath1 is constant , equal to @xmath27 and that @xmath29 is a positevely oriented simple closed curve .
the index @xmath48 is equal to @xmath52 if @xmath27 belongs to @xmath49 , equal to @xmath73 if @xmath27 belongs to @xmath50 , and not defined when @xmath74 .
we proceed to generalize this statement .
[ l2]let @xmath1 be defined on a positively oriented simple closed curve @xmath75\to { { \mathbb{r}}}^2 $ ] .
let @xmath76 be an open arc of @xmath29 and assume that the image of @xmath76 is compactly contained in @xmath49 . assume also that @xmath1 has no fixed points on @xmath29 .
then , there exists a homotopy @xmath77 beginning at @xmath78 , and a time @xmath79 such that the following conditions hold : 1 .
@xmath80 for every @xmath81 .
2 . @xmath82 for every @xmath83 3 .
4 . for every @xmath41 it holds that @xmath85 in @xmath86 where @xmath87 is an open arc compactly containing @xmath76 , and @xmath88 .
in addition , for a homotopy satisfying all these properties , it holds that @xmath89 .
let @xmath90 be a homotopy satisfying that @xmath91 , @xmath92 carries @xmath93 to a point @xmath27 contained in @xmath49 , and for every @xmath41 it holds that @xmath94 and @xmath95 is the identity in a neighborhood @xmath96 of the closure of @xmath50 .
then @xmath97 is a homotopy begining at @xmath1 , having no fixed points on @xmath29 , and such that the final map has @xmath98 .
+ so it can be assumed that @xmath1 is constant in @xmath76 , say @xmath99 . next ,
let @xmath100 be a curve defined in @xmath101 $ ] such that @xmath102 , @xmath103 if @xmath104 , and @xmath105 if @xmath106 and @xmath107 . take an arc @xmath87 containing @xmath76 in its interior , so that @xmath108 has two connected components whose images under @xmath1 are contained in the complement of @xmath96 .
let @xmath59 be a homotopy beginning at @xmath1 , such that following three conditions hold : @xmath109 , @xmath110 , and @xmath85 in @xmath86 .
then the conditions of the lemma hold with @xmath111 .
this proves the first assertion of the lemma .
+ it remains to prove that @xmath89 , the second assertion of the lemma .
it was explained above that @xmath59 can be assumed to be constant in @xmath76 for every @xmath41 .
so the problem reduces to estimate the jump of @xmath112 when @xmath41 passes through @xmath113 .
the construction made implies that no @xmath59 has fixed points in @xmath114 so that every @xmath59 can be assumed to be equal to @xmath1 in @xmath86 , where @xmath87 is a small neighborhood of @xmath76 whose image under @xmath59 is contained in @xmath49 for every @xmath81 .
then the jump can be calculated as if @xmath76 is the segment @xmath115 contained in the real axis , oriented from right to left , @xmath87 is the arc @xmath116 and the image of @xmath76 changes from the lower half plane to the upper one passing through the origin , while the intersection of @xmath117 with the real axis is contained in @xmath76 , and @xmath118 in the rest of @xmath29 .
moreover , it may also be assumed that @xmath119 is piecewise linear .
see figure [ ind ] . in this case
, the index change over @xmath87 can be calculated directly from the definition .
analogously , we have : [ meter ] let @xmath1 be defined on a positively oriented simple closed curve @xmath75\to { { \mathbb{r}}}^2 $ ] .
let @xmath76 be an open arc of @xmath29 and assume that the image of @xmath76 is compactly contained in @xmath49 . assume also that @xmath1 has no fixed points on @xmath29 .
then , there exists a homotopy @xmath77 beginning at @xmath78 , and a time @xmath79 such that the following conditions hold : 1 .
@xmath120 for every @xmath81 .
2 . @xmath121 for every @xmath83 3 .
4 . for every @xmath41 it holds that @xmath85 in @xmath86 where @xmath87 is an open arc compactly containing @xmath76 , and @xmath88 .
in addition , for a homotopy satisfying all these properties , it holds that @xmath122 .
[ rectangulo ] let @xmath123 be the square @xmath124 ^ 2 $ ] .
let @xmath1 be defined on @xmath125 such that : * @xmath126 , * @xmath127 , * @xmath128 , * @xmath129 .
then , @xmath130 , where @xmath29 is @xmath125 with the positive orientation .
let @xmath131 , so that @xmath132 is the identity and @xmath133 .
let @xmath134 .
then @xmath135 and the hypotheses imply that no @xmath59 has fixed points in @xmath125 ( for example , take any @xmath136 and let @xmath137 , then @xmath138 implies that @xmath139 and @xmath140 for every @xmath141 ) .
note that the image of @xmath125 under @xmath142 is contained in the line @xmath143 , and that the image of @xmath144 is contained in @xmath145 and the image of @xmath146 contained in @xmath147 .
it follows from corollary [ meter ] that one can obtain a map @xmath148 such that @xmath149 is contained in the interior of @xmath150 , so that @xmath151 . of course , the fact that @xmath150 is a square is not essential in the hypothesis , and we will use the lemma and remark above to calculate index of more general curves : [ torcido ] let @xmath152 and @xmath153 be disjoint simple proper lines in the plane , each one of which separate the plane .
let @xmath29 and @xmath69 be another pair of disjoint curves separating the plane .
assume also that each @xmath29 and @xmath69 intersect @xmath152 in one point and @xmath153 in one point .
now let @xmath154 be the simple closed curve determined by the four arcs of the curves delimited by the intersection points , with the positive orientation .
now let @xmath1 be a map defined on @xmath154 such that @xmath155 is contained in the component of the complement of @xmath152 that contains @xmath153 , @xmath156 is contained in the component of the complement of @xmath153 that contains @xmath152 , that @xmath157 is contained in the component of the complement of @xmath69 that does not contain @xmath29 and that @xmath158 is contained in the component of the complement of @xmath29 that does not contain @xmath69 .
then the index of @xmath1 in @xmath154 is equal to @xmath159 .
note that there exists a homeomorphism @xmath160 isotopic to the identity such that @xmath161 , @xmath162 , @xmath163 and @xmath164 .
then the map @xmath165 satisfies the hypothesis of lemma [ rectangulo ] so @xmath166 .
[ rk1 ] if the last two items are changed to @xmath167 and @xmath168 , then the conclusion is @xmath169 . to prove that apply the same homotopy and then observe that there exists
@xmath148 homotopic to @xmath142 such that @xmath149 is contained in the interior of @xmath150 , but now the homotopy from @xmath142 to @xmath148 has no fixed points in @xmath125 .
the conclusion follows immediately . as before ,
one obtains : [ torcido2 ] let @xmath152 and @xmath153 be disjoint simple proper lines in the plane , each one of which separate the plane .
let @xmath29 and @xmath69 be another pair of disjoint curves separating the plane . assume also that each @xmath29 and @xmath69 intersect @xmath152 in one point and @xmath153 in one point .
now let @xmath154 be the simple closed curve determined by the four arcs of the curves delimited by the intersection points , with the positive orientation .
now let @xmath1 be a map defined on @xmath154 such that @xmath155 is contained in the component of the complement of @xmath152 that contains @xmath153 , @xmath156 is contained in the component of the complement of @xmath153 that contains @xmath152 , that @xmath157 is contained in the component of the complement of @xmath29 that does not contain @xmath69 and that @xmath158 is contained in the component of the complement of @xmath69 that does not contain @xmath29 .
then the index of @xmath1 in @xmath154 is equal to @xmath52 .
* theorem [ t1 ] . * _ let @xmath19 be a surjective map with degree @xmath4 , @xmath170 an essential continuum such that @xmath25 . if @xmath0 , then @xmath1 is complete on @xmath22 . _ as was already explained , it is sufficient to prove that every lift of @xmath1 has a fixed point on the preimage of @xmath22 under the covering projection .
the proof will use indexes of curves , so it is convenient to interpret the annulus as a subset of the plane .
the map @xmath1 will be considered as a map from @xmath171 into itself , where @xmath172 is the origin ( called south ; the other end of the annulus is @xmath173 , called north ) .
the map @xmath1 is assumed to have degree @xmath4 , which is equivalent to impose that the index of the closed curve @xmath174\to f(\exp(2\pi i t))$ ] is equal to @xmath4 . as @xmath21 is essential , the complement of @xmath21 in the annulus @xmath175 has two distinguished connected components .
the unbounded one is denoted @xmath176 , while @xmath177 stands for the connected component of @xmath178 that accumulates on @xmath172 .
note that @xmath179 is simply connected , thus riemann s theorem implies that there exists a conformal bijection @xmath180 from @xmath179 to the unit disc @xmath181 that carries @xmath172 to the origin .
[ ends ] note that as @xmath21 is connected , then @xmath176 and @xmath177 are the unique components of @xmath178 that are essential in @xmath3 , while the others are simply connected .
as the degree of @xmath1 is nonzero , the image under @xmath1 of an essential open set is essential , and as @xmath182 then @xmath183 is a subset of @xmath177 or a subset of @xmath176 .
the same can be said about @xmath177 . on the other hand ,
as @xmath1 was assumed to be surjective , only two possibilities remain : either @xmath184 and @xmath185 or @xmath186 and @xmath187 .
we will consider the two cases separately .
we will assume first that each @xmath177 and @xmath176 are @xmath1-invariant .
we begin with some constructions of curves in the annulus , this will not give the desired conclusions , not before lifting the curves to the universal covering and calculating their index under a given lift @xmath188 of @xmath1 .
we will prove first that @xmath1 has fixed points in the annulus .
assume by contradiction that @xmath1 has no fixed point .
it will be convenient to recall some results from caratheodory s theory .
let @xmath189 be a simply connected region in the plane with a conformal bijection @xmath190 .
a crosscut in @xmath189 is a simple curve @xmath191 contained in @xmath189 except for its extreme points which belong to the boundary of @xmath189 .
if @xmath192 for every @xmath41 , then the complement of @xmath191 in @xmath189 has two connected components , one of which contains @xmath193 ; the other one is denoted @xmath194 .
a point @xmath195 in the boundary of @xmath189 is accesible if there exists an arc that is contained in @xmath189 except for one extreme point in @xmath195 .
it is well known that accesible points are dense in the boundary of @xmath189 .
moreover , given two accesible points there exists a crosscut in @xmath196 whose extremes are the given points .
assume that @xmath197 is compact , ( which in our case is obvious since @xmath179 will stand for @xmath189 ) .
under this assumption , given a positive number @xmath69 , there exists a finite set of accesible points , each one of them can be joined with another by a crosscut of diameter less than @xmath69 .
these crosscuts will be used to construct an essential simple closed curve .
let @xmath180 be a conformal bijection from @xmath198 to the unit disc carrying @xmath172 to the origin @xmath73 .
it is claimed that for every sufficiently small @xmath65 , there exist a finite sequence of crosscuts @xmath199 in @xmath177 such that 1 .
@xmath200 whenever @xmath201 .
@xmath202 is an essential positively oriented simple closed curve , 3 .
each @xmath203 has diameter less than @xmath69 , 4 .
@xmath204 .
( see figure 2 a ) .
the proofs of the first , second and third properties follow from the comments above . to prove the last one ,
note that the contrary assumption implies that for every sufficiently small positive @xmath205 there exists a point @xmath206 in some crosscut @xmath207 of diameter less than @xmath205 such that @xmath208 , which implies that the distance from @xmath209 to @xmath206 is less than @xmath205 .
an accumulation point of the sequence @xmath206 must belong to the boundary of @xmath177 , that is contained in @xmath21 , and must be a fixed point of @xmath1 , contrary to our assumption .
except for the extreme points of the crosscuts , the remaining points of @xmath29 are contained in the region @xmath177 . by assumption in remark
[ ends ] , @xmath1 carries @xmath177 to itself .
thus the curve @xmath210 is contained in @xmath177 excepting for the images of the extreme points of the crosscuts , that are also contained in @xmath21 .
note that the image of a crosscut is not necessarily a crosscut .
let @xmath63 be the intersection of @xmath29 with @xmath21 .
@xmath63 is finite and by corollary [ l3 ] there exists @xmath211 and @xmath212
such that one can modify @xmath1 in the @xmath69 neighborhood of @xmath63 arbitrarily , whenever the image of the perturbation is still contained in the @xmath72 neighborhood of @xmath213 .
using the fact that @xmath179 is conformally equivalent to the unit disc , it is possible to construct a curve in @xmath177 that is as close to the boundary of @xmath177 as wished . with this in mind , let @xmath152 be a simple closed curve contained in @xmath177 that is close to the boundary of @xmath177 in such a way that the intersection of @xmath152 with @xmath29 only occurs in @xmath68 and that the intersection of @xmath152 with @xmath210 is contained in @xmath71 .
in order to calculate the index of @xmath1 in @xmath29 , we apply corollary [ l3 ] and we may assume that @xmath29 and @xmath210 are both contained in the closure of @xmath214 , the connected component of the complement of @xmath152 that contains @xmath172 . therefore it can be assumed ( for the purposes of index calculation ) that the boundary of @xmath177 is a simple closed curve @xmath152 , of course , essential in the annulus @xmath3 .
the next step consists in changing @xmath1 to a homotopic map @xmath66 in order to obtain that @xmath215 in the process , the index will change , but by means of lemma [ l2 ] and lemma [ rectangulo ] the difference between @xmath48 and @xmath216 will be easily calculated .
+ using property ( c4 ) , there are two possibilities remaining for each @xmath203 : either @xmath217 , or @xmath218 .
assume that @xmath217 .
the image under @xmath1 of the arc @xmath203 is contained in @xmath50 ; a direct application of lemma [ l2 ] implies that the map @xmath1 is homotopic to a map @xmath219 satisfying that @xmath220 and @xmath221 as @xmath222 is positively oriented .
+ if there were exactly @xmath223 of the crosscuts @xmath203 satisfying @xmath217 , then change @xmath1 by a map @xmath66 such that @xmath224 and @xmath225 .
moreover @xmath66 is equal to @xmath1 outside a small neighborhood of each @xmath203 . + now use @xmath66 instead of @xmath1 .
note that , even if @xmath226 for every @xmath227 , it may very well happen that @xmath228 .
let @xmath229 be the connected component of @xmath230 containing @xmath172 .
let @xmath231 be a deformation retract of @xmath214 to @xmath232 be such that a point in @xmath233 is sent under application of @xmath234 to a point in the interior of the arc @xmath203 and such that @xmath235 for all @xmath41 .
note that @xmath236 can not have fixed points in @xmath29 because for a point @xmath237 in @xmath238 it is known that @xmath239 does not belong to the closure of @xmath240 .
it follows that the index of @xmath241 and the index of @xmath242 with respect to the curve @xmath29 are equal .
note that the image of @xmath29 under @xmath242 is contained in the closure of @xmath49 , so a small @xmath16 perturbation @xmath243 of @xmath242 does not change the index and satisfies @xmath244 and @xmath245 . in conclusion ,
the following assertion was proved .
[ c1 ] there exists a map @xmath66 homotopic to @xmath1 such that @xmath70 in @xmath21 and all the components of the complement of @xmath21 other than @xmath177 .
moreover , it holds that @xmath246where @xmath223 is the number of crosscuts @xmath203 satisfying @xmath217 .
now , we want to explain briefly what happens if the construction like that of @xmath29 and the map @xmath66 is done in @xmath176 , the unbounded component of the complement of @xmath21 containing @xmath247 .
now @xmath248 gives a simply connected region and there exists a riemann map carrying @xmath249 to the unit disc and @xmath247 to @xmath73 .
the construction of a curve @xmath250 by concatenation of crosscuts of @xmath176 satisfying the properties ( c1 ) to ( c4 ) is the same as above . now
the component @xmath233 of a crosscut @xmath203 is the bounded component of the complement of @xmath203 in @xmath176 .
observe in addition that if the curve @xmath29 is positively oriented , then a point in @xmath233 is located in @xmath49 , therefore by lemma [ l2 ] @xmath251 whenever @xmath203 is a crosscut such that @xmath217 and @xmath66 is homotopic to @xmath1 and constructed exactly as in the previous part .
the construction of the curve @xmath152 is analogous in this case .
the only difference appears when calculating @xmath216 : [ c2 ] there exists a map @xmath66 homotopic to @xmath1 such that @xmath70 in @xmath21 and all the components of the complement of @xmath21 other than @xmath176 .
moreover , it holds that @xmath252 where @xmath253 is the number of crosscuts @xmath203 such that @xmath254 is contained in @xmath233 and @xmath255 with the positive orientation .
it remains to see why @xmath256 , but this is a simple calculation based on the fact that @xmath29 is a simple closed curve nontrivial in the annulus , that @xmath257 has index @xmath4 and is contained in the unbounded component of @xmath29 .
assuming that @xmath258 , the conclusion is the following : [ c3 ] let @xmath1 be a map of the annulus of positive degree @xmath0 . if @xmath21 is an essential compact subset of the annulus such that @xmath25 and both unbounded components of the complement of @xmath21 are invariant under @xmath1 , then @xmath1 has a fixed point in @xmath22
. note first that the curves @xmath259 and @xmath260 ( the notation makes reference to the curve @xmath29 constructed in @xmath176 , resp .
@xmath177 ) are as close to the set @xmath21 as wished .
both curves are positively oriented , and their indexes differ , because @xmath261 now lefschetz fixed point theorem ( see section [ 3 ] ) shows that @xmath1 has a fixed point in the annulus bounded by @xmath259 and @xmath260 .
as these curves can be constructed arbitrarily close to the boundary of @xmath177 and @xmath176 , it comes that @xmath1 has a fixed point in @xmath22 .
this is not sufficient to conclude that @xmath1 is complete .
it just proves that @xmath1 has a fixed point , also the same argument shows that every iterate of @xmath1 has fixed points , but this contains no additional information .
we need to prove that every lift of @xmath1 has fixed points . adapting the proof of corollary 3
a well known theorem due to cartwright and littlewood follows : _ if an orientation preserving homeomorphism @xmath1 of the plane leaves invariant a continuum @xmath21 not separating the plane , then @xmath1 has a fixed point in @xmath21 . _ + indeed , take a curve like @xmath29 in the complement of @xmath21 and the arguments above show that the index @xmath48 is equal to @xmath262 , where @xmath263 ( as @xmath1 is a homeomorphism ) and @xmath264 .
we will lift the construction in the annulus and consider an arbitrary lift @xmath188 of @xmath1 .
we will prove that @xmath188 has a fixed point ( in order to prove completeness by proposition 1 ) .
assume by contradiction that @xmath188 is fixed point free .
it is not straightforward to lift the construction .
the difficulty comes at first sight , when the preimage of @xmath21 under the covering projection is not compact .
but there are others , as will become clear while calculating the index of a curve .
the universal covering of the annulus is given by the map @xmath265 , written in complex coordinates as @xmath266 .
note that @xmath267 so the group of deck transformations is generated by the translation @xmath268 .
let @xmath269 be the preimage of @xmath21 under the covering projection .
note that @xmath269 is not a compact set but is a closed subset of some horizontal strip @xmath270 ; we will assume @xmath271 to simplify notation
. moreover , @xmath269 separates the plane in at least two components , one contains @xmath272 , denoted @xmath273 and the other contains @xmath274 , denoted @xmath275 .
denote as usual @xmath276 for the compex number @xmath277 .
let @xmath29 and @xmath152 the curves in @xmath177 constructed above and denote by @xmath278 and @xmath279 the preimages of @xmath29 and @xmath152 under the covering projection .
these are embedded lines , each one of which separates the plane in two components .
choose @xmath280 and let @xmath281 be the lift of @xmath29 that begins at @xmath282 .
for @xmath283 , let @xmath284 we define analogously @xmath285 and @xmath286 .
the closed strip contained between the lines @xmath287 and @xmath288 is denoted by @xmath289 .
let @xmath290 be a simple curve from @xmath282 to @xmath291 contained in @xmath289 , and let @xmath292 .
let @xmath293 an embedded line that separates the plane leaving @xmath294 in the connected component of @xmath295 to the right of @xmath296 .
let @xmath297 .
clearly @xmath290 separates the horizontal strip @xmath289 .
note that as @xmath298 , and @xmath0 , then there exists @xmath299 such that the following conditions hold : 1 .
the image of @xmath300 under @xmath188 is contained in the connected component of @xmath301 to the right of @xmath302 .
the image of @xmath303 under @xmath188 is contained in the connected component of @xmath304 to the left of @xmath305 .
3 . @xmath306 and @xmath307 finally , let @xmath153 be the simple closed curve @xmath308 positively oriented .
recall we have assumed that @xmath188 has no fixed points .
the lift of @xmath29 is a concatenation of crosscuts , each one satisfiyng @xmath309 , where @xmath310 lifts @xmath203 .
there exists @xmath311 , homotopic to @xmath188 , such that @xmath312 .
let @xmath313 be the number of crosscuts contained in @xmath314 such that @xmath315 .
define @xmath316 analogously .
moreover , note that @xmath317 is contained in the region @xmath273 , as well as @xmath318 . by application of lemma [ l2 ]
it follows that the @xmath319 .
note that @xmath320 .
the fact that @xmath321 is contained in @xmath275 and that @xmath322 and that @xmath323 in @xmath324 and the choice of @xmath325 imply that @xmath326 as @xmath153 is contained in @xmath327 .
this implies that the index of @xmath328 and this is absurd since we have assumed that @xmath188 was fixed point free .
it follows that every lift of @xmath1 has fixed points .
recall that to finish the proof of this theorem , it remains to consider the case where the unbounded components of the complement of @xmath21 are not invariant , but two - periodic ( remark [ ends ] ) . that is : + _
assume that @xmath329 and @xmath330_. + let @xmath331 be an even number .
then @xmath35 satisfies the assumptions made above , ( that @xmath177 and @xmath176 are @xmath35-invariant ) , therefore any lift @xmath332 of @xmath35 has a fixed point . on the other hand ,
let @xmath331 be an odd number and @xmath332 a lift of @xmath35 .
note first that there exists a positive number @xmath150 such that @xmath333 is contained in @xmath334 and @xmath335 is contained in @xmath336 .
by assumption , @xmath0 , which implies that there exists a number @xmath325 such that @xmath337 and @xmath338 whenever @xmath339 .
let @xmath154 be the simple closed curve whose trace is the boundary of the rectangle with vertices @xmath340 , @xmath341 ; give @xmath154 the positive orientation .
then @xmath154 satisfies the hypothesis of lemma [ rectangulo ] .
thus the index of @xmath332 in @xmath154 is equal to @xmath159 .
it follows that @xmath332 has fixed points .
this finishes the proof of theorem [ t1 ] .
[ cambia ] 1 .
it was assumed that @xmath1 is surjective .
it can be seen from the proofs above that it is sufficient to assume that the invariant continuum @xmath21 is contained in the interior of @xmath342 .
2 . assume that @xmath329 and @xmath330 , then the following assertion is also true : + _ let @xmath1 be a map of the annulus that interchange the ends ( meaning that @xmath1 can be continuously extended to @xmath343 in such a way that @xmath344 is two - periodic ) . if the degree of @xmath1 has absolute value greater than @xmath52 , then for every odd @xmath331 , the map @xmath35 has at least @xmath36 fixed points .
_ for @xmath0 use the arguments above and for @xmath24 , the conclusion follows from remark [ rk1 ] .
in particular , the growth rate inequality @xmath345 holds for @xmath1 .
let @xmath20 be an essential locally connected continuum .
we will use the following properties : 1 .
there exists a simple essential closed curve contained in @xmath21
. see theorem 43 , p.193 in @xcite .
2 . @xmath346 is connected and locally connected .
3 . let @xmath96 be an unbounded connected component of @xmath347 .
for all @xmath211 there exists a neighborhood @xmath348 of @xmath21 in @xmath349 such that the radial retraction @xmath350 of @xmath348 onto @xmath21 satisfies @xmath351 for all @xmath352 ( here radial stands for the induced retraction on @xmath348 by the riemann map ) .
this is true because @xmath21 is locally connected and therefore the riemann map extends continuously to @xmath353
. * theorem [ t2 ] . * _ let @xmath19 , @xmath170 an essential continuum such that @xmath354 . if @xmath21 is locally connected and @xmath2 , then @xmath1 is complete on @xmath22 .
_ notations : @xmath355 is the universal covering of the annulus @xmath3 , the component of the complement of @xmath21 that contains @xmath247 is called @xmath176 and that containing @xmath172 is denoted @xmath177 .
moreover , @xmath356 is identified with the product @xmath357 , and define @xmath358 and @xmath359 .
note that for all @xmath64 there exists a simple closed curve @xmath360 such that @xmath361 for every @xmath362 .
analogously , there exists a simple closed curve @xmath363 satisfying the same properties .
let @xmath364 and @xmath365 be the preimages of @xmath152 and @xmath153 by the covering projection .
note that both @xmath366 and @xmath367 separate @xmath356 and as they do not intersect , then there is also a component of @xmath368 whose boundary is @xmath369 , and this component , denoted @xmath370 , contains @xmath269 .
let @xmath188 be any lift of @xmath1 .
suppose that @xmath371 ( otherwise we are done ) .
so , there exists @xmath211 such that @xmath372 for all @xmath373 . for this @xmath72 , choose @xmath152 and @xmath153 as above . by property
( 3 ) of locally connected sets at the beginning of this section , there exists a deformation retract @xmath374 such that @xmath375 , where @xmath348 is a neighborhood of @xmath269 .
so , @xmath376 is contained in @xmath269 .
analogously , @xmath377 is contained in @xmath269 .
besides , the set of fixed points of @xmath378 in @xmath379 are the same as those of @xmath188 . if @xmath0 , we may choose real numbers @xmath380 such that the intersection of the vertical segment @xmath381 with @xmath370 is mapped under @xmath378 strictly to its right , and the intersection of the vertical segment @xmath382 with @xmath370 is mapped under @xmath378 strictly to its left .
let @xmath29 be the simple closed curve constructed with the segments of @xmath366 and @xmath367 between @xmath383 and @xmath384 and the subarcs of the vertical segments @xmath382 and @xmath381 that are contained in @xmath370 .
note that if @xmath29 is positively oriented then corollary [ torcido ] ( of lemma [ rectangulo ] ) gives @xmath385 .
this was assuming @xmath0 . if @xmath24 , then the vertical segments at @xmath383 and @xmath384 can be taken so that the vertical at @xmath383 is mapped to the right of @xmath384 and the vertical at @xmath384 is mapped to the left of @xmath383 .
now apply corollary [ torcido2 ] to obtain that the index is equal to @xmath52 .
so , @xmath386 .
* example .
* we will show that if @xmath387 it may happen that @xmath1 is fixed point free on @xmath21 .
let @xmath21 be as in figure 1 ( a ) .
we will construct a map @xmath388 such that @xmath389 . take a fundamental domain @xmath390 of @xmath269 as in figure 1 ( b ) and we proceed to define @xmath188 on @xmath390 .
let @xmath391\to d $ ] a parametrization such that : @xmath392 for all @xmath393 , @xmath394 , and such that @xmath395 .
first define @xmath396\to [ 0,1]$ ] ( discontinuous ) as in figure [ figura3 ] ( c ) .
note that there exists @xmath397 such that @xmath398 .
moreover @xmath188 is continuous on @xmath390 , as the disconitnuity of @xmath399 can be avoided on @xmath400)$ ] because @xmath401 .
clearly , @xmath399 is fixed point free . if @xmath188 has a fixed point @xmath402 , then @xmath403 so , @xmath404 implying @xmath405 because @xmath41 and @xmath406 are different from @xmath195 and @xmath407 .
note that @xmath188 factors to a map @xmath19 of degree @xmath159 and that it is fixed point free by construction .
we have then proved that with @xmath408 one can avoid having a fixed point on @xmath21 . note that with exactly the same technique one can build a map @xmath1 , not complete in @xmath21 , of any given degree @xmath24 .
+ to assure the existence of periodic points in @xmath21 , and not just @xmath22 , we assume in the next theorem that @xmath21 is the boundary of one of the unbounded components of its complement and that @xmath0 .
[ t4 ] let @xmath19 be a map of degree @xmath4 , @xmath20 a locally connected essential continuum such that @xmath354 .
assume that @xmath21 is the boundary of one of the unbounded connected components of its complement .
if @xmath0 , then @xmath1 is complete on @xmath21 . by hypothesis @xmath269 is the boundary of a simply connected region @xmath409
+ the preimage under @xmath44 of the curve given by property ( 1 ) at the beginning of this section , gives a continuous @xmath52-periodic function @xmath410 with the following properties : 1 .
@xmath411 is a homeomorphism from @xmath412 onto @xmath413 .
2 . @xmath413 separates the plane . * claim .
* let @xmath414 be a connected component of @xmath415 .
then @xmath416 , the closure of @xmath414 , intersects @xmath413 in a one point set .
+ to prove this claim note first that the intersection is not empty because @xmath417 is connected as @xmath21 is locally connected .
now assume that there are points @xmath418 such that @xmath419 and @xmath420 belong to @xmath416 .
as @xmath413 is contained in @xmath417 and @xmath269 is the boundary of an unbounded region @xmath421 , for all @xmath422 there exists a simple embedded line @xmath423 landing at @xmath424 and separating @xmath425 in two connected components @xmath3 and @xmath426 .
as @xmath414 must have points in @xmath3 and in @xmath426 , this contradicts the fact that @xmath414 is connected .
+ from now on the unique point of @xmath427 is denoted by @xmath428 . define a map @xmath429 that is the identity on @xmath413 and to each point @xmath237 in a component @xmath414 of @xmath430 assigns the point @xmath428 . we claim that as @xmath417 is locally connected , @xmath431 is continuous .
to see this , note that if @xmath432 , then for all neighborhood @xmath433 of @xmath434 there exists a neighborhood @xmath96 of @xmath434 such that if @xmath435 and @xmath436 then @xmath437
. otherwise , there exist a sequence @xmath438 of different components accumulating both on @xmath434 and in a point @xmath439 in @xmath440 .
this contradicts local connectivity .
continuity of @xmath431 now follows .
define @xmath441 .
this is a continuous function from @xmath412 to @xmath412 .
the assumption @xmath0 implies that there is a point @xmath383 such that @xmath442 for every @xmath443 and a point @xmath384 such that @xmath444 for every @xmath445 .
let @xmath446 be the supremum of the set of points @xmath447 such that @xmath448 . as @xmath311 is continuous , @xmath449 and by definition of @xmath446 , @xmath450 for every @xmath451 .
then either @xmath452 is fixed for @xmath188 or @xmath453 belongs to a component @xmath414 of @xmath454 such that @xmath455 . in the latter case ,
let @xmath456 be a neighborhood of @xmath457 in @xmath269 such that @xmath458 .
now let @xmath451 be close to @xmath446 in such a way that @xmath459 .
so @xmath460 and it follows that @xmath461 .
this is a contradiction , so the unique conclusion left is @xmath462 .
* theorem [ t3]*. _ let @xmath19 be a degree @xmath4 map of the annulus , where @xmath2 .
each one of the following conditions imply that @xmath1 is complete .
_ 1 . both ends of @xmath3 are attracting . 2 .
both ends of @xmath3 are repelling .
the hypothesis of the first assertion implies that defining @xmath463 and @xmath464 gives a continuous extension of @xmath1 to @xmath465 , the compactification of the annulus with two points , in such a way that @xmath247 and @xmath172 become topological attractors . then there exists a simple closed essential curve @xmath152 whose image under @xmath1 is contained in @xmath466 , the connected component of @xmath465 that contains @xmath247 .
similarly , there is a simple closed essential curve @xmath153 whose image under @xmath1 is contained in @xmath467 .
now let @xmath188 be a lift of @xmath1 , @xmath366 and @xmath367 the preimages of @xmath152 and @xmath153 under the covering projection , so that there is a region @xmath468 whose boundary is the union of @xmath366 and @xmath367 .
let @xmath290 be a simple arc having one point in @xmath366 another point in @xmath367 and whose interior is contained in @xmath469 , and let @xmath470 for @xmath471 .
let @xmath472 be the simple closed curve formed with @xmath300 , @xmath303 and parts of @xmath366 and @xmath367 .
the fact that @xmath2 implies that there exists @xmath299 such that @xmath473 does not intersect @xmath472 and that @xmath474 , depending on @xmath258 or @xmath475 .
it follows that @xmath188 has a fixed point . as every iterate of @xmath1 satisfies the same hypothesis , then @xmath1 is complete .
for the case of repelling ends , there exists a curve @xmath152 , close to the end @xmath247 of the annulus , such that @xmath476 is contained in the component of the complement of @xmath152 that does not contain @xmath247 .
there will be as well a curve @xmath153 , close to @xmath172 , whose image under @xmath1 is contained in the component of the complement of @xmath153 that contains @xmath247 .
next consider a lift @xmath188 of @xmath1 , take the preimages of @xmath152 and @xmath153 in the universal covering and cut it with vertical arcs , in such a way to obtain a curve @xmath29 whose sides are @xmath366 , @xmath367 and vertical segments , and satisfies the properties of the curve @xmath29 of corollary [ torcido ] if @xmath0 . the corollary implies that the index of @xmath188 in @xmath29 is equal to @xmath159 . on the other hand , if @xmath24 , then the comments succeding corollary [ torcido ] imply that @xmath477 .
this implies that every lift of @xmath1 has a fixed point .
as every @xmath35 satisfies the same hypothesis , it follows that @xmath1 is complete .
the first result is a consequence of hagopian s theorem @xcite which states that an arcwise connected plane continuum @xmath21 that does not separate the plane has the fixed point property ( i.e. every continuous map from @xmath21 to @xmath21 has a fixed point in @xmath21 ) .
we will use this theorem on the universal covering , so we need that the lift of k is arcwise connected .
this may not happen even if the set k is arcwise connected ( see figure 6 ) .
however , note that in this example the set @xmath21 is simply connected .
if one assumes that @xmath21 has a nontrivial closed loop , this guarantees that the lift is arcwise connected .
+ by hypothesis , @xmath269 and @xmath479 are arcwise connected .
take @xmath188 any lift of @xmath35 .
as @xmath331 is odd , then @xmath35 has negative degree .
one can modify @xmath188 outside @xmath479 so that it can be extended to @xmath480 $ ] .
let @xmath390 be the compactification of @xmath480 $ ] with two points @xmath481 , which is a topological closed disk .
as @xmath475 , @xmath188 interchanges these two points and @xmath482 is compact . by hagopian s theorem @xcite , @xmath188 must have a fixed point in @xmath479 , and this point does not belong to @xmath483 .
it follows that @xmath188 has a fixed point in @xmath479 .
concluding , if @xmath24 then every lift of @xmath35 with odd @xmath331 has a fixed point .
this implies that @xmath35 has at least @xmath36 fixed points , so @xmath1 satisfies the growth rate inequality . for @xmath484 ,
just the existence of a fixed point in @xmath22 can be assured .
let @xmath485 be a map of degree @xmath0 .
assume that there exists two disjoint simply connected nonempty open sets @xmath490 and @xmath491 such that @xmath492 for @xmath493 .
then , @xmath1 has at least the same number of periodic points as @xmath13 on @xmath494 .
let @xmath21 be the complement of @xmath495 in @xmath15 .
then @xmath21 is a completely invariant continuum . for @xmath493 choose points @xmath496 .
let @xmath66 be a map homotopic to @xmath1 such that @xmath497 in a neighborhood of @xmath21 and such that @xmath498 for @xmath493 .
then @xmath66 is a degree @xmath4 map of the annulus @xmath499 and has a completely invariant essential continuum @xmath21 . by theorem [ t1 ] @xmath66 is complete on @xmath500 , so it has at least the same number of periodic points as @xmath13 on @xmath494 .
the conclusion follows because @xmath497 on @xmath21 . | consider a continuous surjective self map of the open annulus with degree @xmath0 .
it is proved that the number of nielsen classes of periodic points is maximum possible whenever @xmath1 has a completely invariant essential continuum .
the same result is obtained in negative degree @xmath2 and for just forward invariant essential continua , provided that the continuum is locally connected .
we also deal with the problem of wether there is a representative of each nielsen class in the filled set of the invariant continuum . moreover , if the map extends continuously to the boundary of the annulus and both boundary components are either attracting or repelling , the hypothesis on the existence of the invariant continuum is no longer needed for obtaining all the periodic points in the interior of the annulus . |
efforts to probe the universe at ever higher redshifts have continued to gather momentum in the last few years . the sloan digital sky survey ( sdss )
has led to the discovery of many quasars ( qsos ) at @xmath15 ( fan et al .
2003 and references therein ) and searches for normal star forming galaxies at these epochs have been equally successful , reaching @xmath16 ( hu et al . 2002 ; yan , windhorst , & cohen 2003 ; lehnert & bremer 2003 ; kodaira et al . 2003 ; stanway , bunker , & mcmahon 2003 ) .
one of the main goals of all of these studies is to identify the time when the baryons in the intergalactic medium ( igm ) were reionized by the light of the first stars and galaxies .
the spectra of sdss qsos at @xmath17 are essentially black at wavelengths below the lyman @xmath0 emission line , a finding which has been interpreted as the signature of the trailing edge of the cosmic reionization epoch ( e.g. becker et al .
2001 ; fan et al .
2002 ; cen & mcdonald 2002 ; white et al . 2003 ) .
yet , the recent detection by the _ wilkinson microwave anisotropy probe _ ( _ wmap _ ) of a large optical depth to thomson scattering , @xmath18 , suggests that the universe was reionized at higher redshifts , @xmath19 ( kogut et al .
2003 ; spergel et al .
if confirmed , this would be an indication of significant star - formation activity at very early times . beyond @xmath20 ,
the information provided by the lyman @xmath0 forest itself becomes progressively more difficult to interpret because of the severe line blending and rapidly increasing optical depth which leave little signal in qso spectra . in order to follow the evolution of the igm to earlier times we have to rely on the metal lines associated with the lyman @xmath0 forest ,
the most common of which is the c iv @xmath21 doublet .
the standard for this work has been set by the analysis of songaila ( 2001 ) . by bringing together measurements from the spectra of 32 qsos ,
songaila was able to follow the evolution of the column density distribution of c iv absorbers , @xmath22 , over the redshift interval @xmath23 . in today
s ` consensus ' cosmology , @xmath24 , @xmath25 , @xmath26 km s@xmath27 mpc@xmath27 , this redshift range corresponds to a time interval from 4.5 to 1.25gyr after the big bang .
the surprising result is that no evolution can be discerned in @xmath22 , nor in its integral which gives the mass density of c@xmath6 ions , @xmath28 ( expressed as a fraction of the critical density ) .
taken at face value , this finding may suggest that most of the igm metals were already in place at the highest currently observable redshifts , and may thus point to an _ early _ enrichment epoch by outflows from low - mass subgalactic systems ( madau , ferrara , & rees 2001 ) .
alternatively , the density of c@xmath6 ions may not reflect in a simple way the overall density of metals in the igm if , for example , _ late _ winds from massive lyman break galaxies are the source of the strongest c iv absorption systems ( haehnelt 1998 ; adelberger et al .
it then becomes extremely important to push the study of the c iv ` forest ' to higher redshifts not only to distinguish between these different chemical evolution scenarios , but also to understand the mechanisms by which metals are distributed from their stellar birthplaces and mixed within the igm .
the sample of qsos analyzed by songaila ( 2001 ) was assembled as the first results from the sdss were beginning to appear in the public domain , and therefore included relatively few objects at @xmath15 .
consequently , her statistics on @xmath22 and @xmath28 are least secure at the highest redshifts . in this paper , we add to songaila s work with deep observations of three recently discovered sdss qsos , all at @xmath15 , obtained with the aim of improving the statistics of the column density distribution by : ( a ) increasing the sample , ( b ) reaching to lower values @xmath29(c iv ) , and ( c ) considering the likely corrections due to sample incompleteness .
our main conclusion is that we confirm the previously reported lack of evolution in @xmath28 .
the observations are described in 2 , while in 3 we provide measurements of the c iv absorbers detected in the three qso spectra . in 4 and 5
we analyze our sample and compare our findings with those reported by songaila ( 2001 ) .
finally , in 6 we briefly discuss possible interpretations of these results and their implications for the origin of c iv absorption at high redshifts .
the spectra of the three qsos were recorded with the echelle spectrograph and imager ( esi ; sheinis et al .
2000 ) on the keck ii telescope in january and february of 2002 ; relevant details of the observations are collected in table 1 . with its combination of high efficiency at red and near - ir wavelengths , wide wavelength coverage ( from 4000 to 1@xmath30 m ) , and moderately high resolution ( @xmath31 )
esi is well suited to the aims of the present work .
the qsos sdss0231@xmath320728 ( anderson et al . 2001 ) at @xmath33 forest in the esi data presented here and therefore differ slightly from those given in the original discovery papers which were based on lower resolution spectra . ] and sdss0836 + 0054 ( fan et al .
2001b ) at @xmath34 were observed with a 0.75wide entrance slit , which projects to approximately four @xmath35 km s@xmath27 pixels in the dispersion direction . for sdss1030 + 0524 ( fan et al .
2001b ) at @xmath36 , a 1wide entrance slit was employed . for all observations
the slit was aligned at the parallactic angle , and the airmass was always less than 1.5 .
reference spectra of internal lamps were used for wavelength calibration and flat - fielding .
observations of the smooth spectrum white dwarf star g191b2b , obtained on each night , provided a template for dividing out the numerous telluric absorption lines which mar ground - based spectra in the far red and near - ir ; they were also used to place the qso spectra on an absolute flux scale ( massey et al 1988 ; massey & gronwall 1990 ) . the individual two - dimensional esi images ( recorded with exposure times of either 1200 , 1800 , or 2700s ) were processed using custom idl routines written by one of us with the specific aim of maximising the accuracy of background subtraction ; this is often the limiting factor in deep spectroscopy of faint sources at long optical wavelengths , where line emission from the night sky dominates the signal .
after flat - fielding , wavelength calibration and background subtraction , the individual one - dimensional spectra were mapped onto a common , vacuum heliocentric , wavelength scale before being added together to produce the final spectra , shown in figure 1 .
corresponding variance spectra are also produced by the data reduction software . in column
( 6 ) of table 1 we list typical values of the signal - to - noise ratio ( s / n ) in the qso continuum in the wavelength regions where we searched for c iv doublets ( see 3 ) . generally , the s / n is highest near the lyman @xmath0 emission line and decreases at longer wavelengths . in the two best observed qsos , sdss0836 + 0054 and sdss1030 + 0524 ,
the long exposure times ( see column ( 5 ) of table 1 ) resulted in s / n@xmath37 and 40 respectively at wavelengths between lyman @xmath0 emission and 9000 .
the resolution of the spectra , as measured from the widths of night sky emission lines , is @xmath38 fwhm ( @xmath39 km s@xmath27 ) , sampled with @xmath40 wavelength bins for the qsos sdss0231@xmath320728 and sdss0836 + 0054 ( and @xmath41% coarser for sdss1030 + 0524 ) . with their combination of high s / n and resolution ,
the spectra used in this study are some of the best published of very high redshift qsos , as can be appreciated from figure 1 and from the last column of table 1 which lists the @xmath42 detection limits for the rest frame equivalent widths of unresolved c iv lines . the final steps in the data reduction involved correcting for atmospheric absorption by diving the qso spectra by that of the smooth spectrum star ( suitably normalized ) , and fitting the qso continuum .
both steps were carried out using the starlink software package dipso . the end result
are normalised qso spectra which could then be searched for c iv absorption .
the spectra of the three qsos were visually inspected independently by two of us for pairs of absorption lines with the correct separation and relative strengths to be c iv doublets .
after various trials ( including the simulations described in 5 below ) , we decided to restrict the search to two wavelength regions .
the first region extends from the lyman @xmath0 emission line of the qso longward to 8940 , near the onset of the atmospheric a - band .
the second region is a relatively small gap , between 9200 and 9300 , which is free of strong atmospheric absorption and sky emission lines ( and , for this reason , is being exploited in narrow - band searches for high redshift lyman @xmath0 emitters see , for example , hu et al .
the reason for restricting ourselves to these wavelength intervals is that at other wavelengths strong atmospheric absorption reduces significantly the s / n achieved .
even with the best efforts to divide out the atmospheric lines , our sensitivity to c iv absorption is much reduced here ; simulations confirmed that we could only recover the strongest c iv doublets , and with only partial success .
although this choice effectively imposes a limit of @xmath43 to the redshift range over which we can study c iv , we nevertheless prefer to concentrate our analysis on regions where our detection limit is relatively uniform .
the simulations described in 5 show that in these regions we are essentially complete for c iv absorption lines with rest frame equivalent width @xmath44m .
once c iv doublets had been identified , their normalized line profiles were fitted with theoretical voigt profiles using the vpfit package .
vpfit deconvolves the composite absorption profiles into the minimum number of discrete components and returns for each the most likely values of redshift @xmath45 , doppler width @xmath46 ( km s@xmath27 ) , and column density @xmath29(c iv ) ( @xmath47 ) by minimizing the difference between observed and computed profiles .
the profile decomposition takes into account the instrumental point spread function of esi .
vacuum rest wavelengths and @xmath48-values of the c iv transitions are from the recent compilation by morton ( 2003 ) .
we now briefly discuss each qso in turn .
the peak of the lyman @xmath0 emission line in this qso , at 7806 , corresponds to @xmath45=4.0420 for c iv@xmath49 ; thus , we can search for c iv doublets in the redshift intervals @xmath50 ( 78068940 ) and @xmath51 ( 92009300 ) .
we find eight c iv systems , listed in table 2 and reproduced , together with their profile fits , in figure 2 .
the two systems labelled ` marginal ' in table 2 are cases where we do not feel confident of the identifications because the weaker member of the doublet is affected by residuals in the sky subtraction .
the higher redshift of this qso ( the peak of the lyman @xmath0 emission line is at 8270 ) means that we can only search for c iv doublets over a more restricted redshift range , from @xmath52 to 4.7745 ( 82708940 ) , as well as @xmath51 ( 92009300 ) .
we find seven c iv absorption systems , one of which consists of ( at least ) three separate components ( see table 3 and figure 3 ) .
this is the highest redshift qso among the three studied here ; with the peak of the lyman @xmath0 emission line occurring at 8880 , we have only 60 of clear continuum before the onset of the atmospheric a - band .
we find no c iv systems in this narrow redshift interval ( @xmath45= 4.7357 4.7745 ) , although one is detected between 9200 and 9300 , at @xmath45= 4.94866 ( see table 4 and figure 4 ) .
the errors quoted in tables 2 , 3 , and 4 are the @xmath53 estimates returned by vpfit on the basis of the error spectra provided to the fitting program .
these spectra are shown as a line near the zero level in each panel of figures 2 , 3 , and 4 , and can be seen to be a reasonable representation of the rms deviations from the continuum level away from strong sky lines ( the residuals from the subtraction of the sky lines can sometimes amount to many times the estimated value of sigma , reflecting systematic , rather than random , errors in the sky subtraction ) .
however , vpfit does not take into account the possibility that what we see as an individual c iv line is actually an unrecognized blend of more than one absorption component .
this is a common problem in the analysis of interstellar ( or in this case intergalactic ) absorption lines , made worse here by the relatively coarse resolution of esi .
specifically , the esi point spread function of fwhm=45 km s@xmath27 corresponds to an instrumental doppler parameter @xmath54 km s@xmath27 .
this is significantly larger than the typical @xmath46-values of c iv lines at lower redshifts ( @xmath55 ) ; for example rauch et al . (
1996 ) reported @xmath56(c iv)@xmath57 km s@xmath27 , while ellison et al .
( 2000 ) found @xmath56(c iv)@xmath58 km s@xmath27 , both from higher resolution ( @xmath59 km s@xmath27 fwhm ) spectra than those considered here .
there are several hints in our data that we may underestimating the complexity of the c iv absorption lines .
first , the values of @xmath46 we deduce are all larger than the median values at @xmath60 , as referenced above .
we consider it much more likely that this is an effect due to the limited spectral resolution of our data , rather than a genuine redshift evolution in the intrinsic line widths .
second , unrecognized velocity structure may be the reason why the fits to the some of our c iv systems are poor ( examples are the @xmath45 = 4.685 complex and the @xmath45 = 4.99695 system in figure 3 ) . in general , the result of blending several components into one unresolved feature is an _
under_estimate of the column density , because narrow saturated components , if they exist , could easily be masked by broader ones ( nachman & hobbs 1973 ) . the @xmath45 = 4.66874 system in figure 3 is an example where the c iv doublet ratio is indicative of saturation .
vpfit converges to a value of @xmath61 very much less than @xmath62 ; therefore , in this case @xmath61 remains undetermined and we quote in table 3 a lower limit to @xmath29(c iv ) based on the equivalent width of @xmath63 , the weaker member of the doublet , assuming no saturation . in conclusion , given the limited resolution of esi , the values of @xmath29(c iv ) derived here should strictly be considered as lower limits , a point which we shall keep in mind in the interpretation of the results .
the mass density of c iv ions , expressed as a fraction of the critical density today , @xmath64 g @xmath65 , can be calculated as ( e.g. lanzetta 1993 ) @xmath66 where @xmath67 km s@xmath27 mpc@xmath27 is the hubble constant , @xmath68 is the mass of a c iv ion , @xmath69 is the speed of light , and @xmath22 is the number of c iv absorbers per unit column density per unit absorption distance @xmath70 .
this last quantity is used to remove the redshift dependence in the sample : absorbers with constant comoving space density and constant proper size have a constant number density per unit absorption distance along a line of sight . in a flat friedmann universe with matter and vacuum density parameters today @xmath71 and @xmath72 , the absorption distance is given by @xmath73^{1/2}}= { 2\over 3\omega_m}\,\{[\omega_m(1+z)^3+\omega_\lambda]^{1/2}-1\}.\ ] ] in the case of an einstein - de sitter universe with ( @xmath74 , we recover the standard expression @xmath75/3 $ ] ( tytler 1987 ) .
our total sample consists of 16 or 12 c iv systems , depending on whether we include the marginal detections or not .
although these absorbers span a factor of @xmath41 in column density , from @xmath76(c iv ) = 12.50 to 13.98 ( see tables 2 , 3 and 4 ) , our completeness drops to below 50% for @xmath76(c iv)@xmath77 ( 5 ) .
it is not possible to determine @xmath22 on the basis of such limited statistics .
however , we can still deduce an estimate of @xmath28 using the approximation @xmath78 ( e.g. storrie - lombardi , mcmahon , & irwin 1996 ) , where @xmath79 is the total absorption distance covered by our spectra .
in the following we will adopt an einstein - de sitter cosmology for comparison with earlier work ; then @xmath79 is given by @xmath80.\ ] ] the summation in equation ( 4 ) is over the @xmath81 redshift intervals @xmath82 where we can detect c iv systems in our data ; as explained in 3 , we have two such intervals in each of the three sight - lines to the qsos in table 1 . with @xmath9 , equations ( 1 ) and ( 3 ) then lead to : @xmath83 for the error in @xmath28 , @xmath84 , we adopt the estimator proposed by storrie - lombardi et al . ( 1996 ) : @xmath85 ^ 2 } { \left [ \sum_{i } n_{i } { \rm ( c~iv ) } \right ] ^2}.\ ] ] for consistency with songaila ( 2001 ) , in what follows we quote errors which correspond to formal 90% confidence limits , that is @xmath86 ( assuming a gaussian distribution ) .
we also estimated errors using the bootstrap method ( efron & tibshirani 1993 ) and found them to be about 65% of the values obtained from eq.(6 ) .
we adopt the values from eq.(6 ) , as they are likely to be more realistic .
the results are collected in table 5 . over the redshift interval @xmath87 covered by our spectra ,
the total redshift path is @xmath4 . summing the column densities of all 16 absorption systems detected
, we obtain @xmath88 at a mean @xmath89 . excluding the four ` marginal ' systems in tables 2 , 3 , and 4 would have a very small effect on @xmath90 , and therefore @xmath28 , because relatively low column densities are associated with these uncertain detections . on the other hand
, @xmath28 may have been underestimated if some of the c iv doublet lines include strongly saturated components , as discussed above . in her analysis , songaila ( 2001 ) divided her sample in bins of @xmath91 .
the only such bin where we have sufficient statistics to compare with her results is between @xmath92 and 5.0 , which includes 11 out of the 16 c iv systems detected here . for this subsample of our data , @xmath93 , @xmath94 and @xmath95 . for comparison , songaila s ( 2001 )
survey in this redshift bin covered @xmath96 at a very similar mean redshift @xmath97 , and yielded @xmath98 ( see table 5 and figure 5 ) . once corrected for incompleteness
, we will see below that our estimate of @xmath28 is @xmath99% higher than songaila s earlier value , although the two measurements are mutually consistent within the errors .
our density of absorbers per unit redshift path is two times higher than that of songaila ( 2001 ) , presumably reflecting the fact that the high s / n ratios of our spectra allow us to reach further down the column density distribution of c iv .
songaila s best fit to the @xmath22 distribution in the redshift range @xmath100 is @xmath101 , where @xmath102(c iv ) is measured in units of @xmath103@xmath47 .
if this relatively steep slope of @xmath22 persists to the higher redshifts probed here , by integrating over the column density we would expect @xmath104 , that is about 5 absorbers with @xmath105 per unit absorption distance . over the redshift interval @xmath106 covered by our spectra , we have identified between 4 and 4.6 systems ( depending on whether marginal detections are included or not ) with @xmath105 per unit @xmath79 , in very good agreement with the expected value .
thus , within the limited statistics of our sample , it appears that there is little evolution in the density of c iv systems with @xmath107 .
it is evident from the results in tables 2 , 3 , and 4 , that our ability to detect c iv absorption systems decreases as @xmath76(c iv ) decreases .
most of our systems have @xmath76(c iv)=13.014.0 and only three have @xmath76(c iv)@xmath108 , whereas @xmath22 still rises at these column densities , at least at @xmath109 ( ellison et al .
2000 ) . in order to assess the completeness of our sample
, we have performed some checks , as follows .
one of us generated a number of fake c iv systems comparable to the number detected in our spectra of sdss0231@xmath320728 and sdss0836@xmath1100054 ( approximately nine systems per simulation ) .
the @xmath46-values of the fake c iv lines were drawn at random from the observed distribution of @xmath46-values in our data ; similarly their redshifts were chosen randomly from the redshift ranges where we could search for c iv systems .
these fake c iv doublets were then added to the real spectra and searched for by one of the two members of our team who had previously identified the real c iv lines . given the visual , rather than automatic , character of our searches , only a limited number of such trials could be performed . specifically , we performed three such trials for each of the following values of column density : @xmath76(c iv)= 13.3 , 13.5 , 13.7 , and 14.0 and six trials for @xmath76(c iv)= 13.0 .
thus , the results of these tests are only indicative ; they are collected in table 6 .
we found that we could recover essentially all c iv doublets so long as @xmath76(c iv)@xmath111 .
below this value , however , we quickly become incomplete because the equivalent width of the weaker member of the doublet is comparable to residuals from the subtraction of sky emission lines and from the correction for atmospheric absorption , and the chances of blending with these residuals are high .
specifically , we were only able to recover 21 out of 47 fake c iv doublets with @xmath76(c iv)=13.0 . assuming that this completeness fraction of 45% applies to all systems with @xmath112(c iv)@xmath113 in our data , the resulting correction factor to @xmath28 is + 22% .
the corresponding correction for systems with @xmath76(c iv)@xmath108 is more difficult to estimate because our incompleteness becomes severe at these low column densities .
if we are 10% complete in the range @xmath114(c iv)@xmath108 , the corresponding correction to @xmath28 from this interval alone is + 28% ; taking both corrections into account would raise the value of @xmath28 by 50% . more generally ,
if the steep slope of @xmath22 found by songaila ( 2001 ) continued below @xmath76(c iv)@xmath115 , we would expect roughly comparable contributions to @xmath28 from each decade in column density .
songaila ( 2001 ) estimated that at redshifts @xmath11 incompleteness effects become significant in her sample only for column densities @xmath76(c iv)@xmath108 .
the simulations described above then suggest that the values of @xmath28 from our study should be multiplied by a factor of 1.22 for a meaningful comparison with the values at @xmath11 .
the resulting @xmath116 derived here for the interval @xmath117 ( where the error does _ not _ take into account the uncertainty in the incompleteness correction ) can then be seen from figure 5 to be entirely consistent with the values measured at lower redshifts .
quantitatively , the mean of the five values of @xmath28 at @xmath11 from the survey by songaila ( 2001 ) is @xmath118 , with a standard deviation @xmath119 ; the value deduced here for @xmath117 is within @xmath53 of @xmath120 .
the principal conclusion from this work is that we confirm the findings by songaila ( 2001 ) that the column density distribution @xmath22 and its integral @xmath10 , which measures the mass density of c iv ions in the intergalactic medium , evidently remain approximately constant over an interval of time which stretches from @xmath12 to @xmath13gyr after the big bang ( @xmath121 ) .
a straightforward interpretation of these findings is that most of the igm metals were already in place at the highest currently observable redshifts ; it seems unlikely that an invariant distribution could be caused by compensating variations in the metallicity and ionization parameter . in currently popular hierarchical clustering scenarios for the formation of cosmic structures ,
the assembly of galaxies is a bottom up process in which large systems result from the merging of smaller subunits . in these theories subgalactic halos with masses comparable to those of present
day dwarf ellipticals form in large numbers at very early times .
their gas condensed rapidly due to atomic line cooling , and became self - gravitating : massive stars formed with some ( perhaps top - heavy ) initial mass function , synthesized heavy - elements , and exploded as supernovae ( sne ) after a few @xmath122yr , enriching the surrounding medium .
it is a simple expectation of the above scenario that the energy deposition by sne in shallow potential wells will disrupt the newly formed protogalaxies and blow away metal - enriched baryons from the host , causing the pollution of the igm at early times ( e.g. tegmark , silk , & evrard 1993 ; gnedin & ostriker 1997 ; madau , ferrara , & rees 2001 ; mori , ferrara , & madau 2002 ) .
these subgalactic stellar systems , possibly aided by a population of accreting black holes in their nuclei and/or by an earlier generation of stars in even smaller halos ( ` minihalos ' with virial temperatures of only a few hundred kelvins , where collisional excitation of molecular hydrogen is the main coolant ) , are believed to have generated the ultraviolet radiation and mechanical energy that reheated and reionized the universe ( e.g. haiman & holder 2003 ; loeb & barkana 2001 ) .
an alternative picture involves later enrichment from lyman break galaxies ( lbgs ) instead . in a study which combined qso absorption line spectroscopy with deep galaxy imaging and spectroscopy in the same fields , adelberger et al .
( 2003 ) have shown that the lyman @xmath0 forest and lbgs are more closely related than had been suspected previously .
of particular relevance to the present discussion is the spatial association of strong c iv systems with galaxies : essentially all of the systems with @xmath76(c iv)@xmath123 in the adelberger et al .
( 2003 ) sample are found within @xmath124 km s@xmath27 and @xmath125kpc ( proper distance ) of a lbg .
adelberger et al .
( 2003 ) show that these dimensions , both in space and velocity , are characteristic of the galactic - scale outflows driven by the star formation activity in lbgs .
although the correlation weakens as one moves to lower column densities of c iv , it remains significant over the full range of values of @xmath29(c iv ) sampled here ( @xmath126 ) .
and yet the apparent lack of evolution in @xmath28 in both scenarios is somewhat puzzling .
if these metals are truly intergalactic and due to early pollution , then one might expect the fraction of c which is triply ionized to change between @xmath127 and 1.5 , since the physical conditions in the igm are thought to have evolved between these epochs .
its large scale structure developed dramatically , so that a given optical depth in the lyman @xmath0 forest generally refers to condensations of lower overdensity ( relative to the mean ) at @xmath128 than at @xmath129 ( e.g. cen 2003 ) .
perhaps most importantly , the ionizing background may have changed in both intensity and shape , as the comoving density of bright qsos grew to a peak near @xmath130 ( fan et al .
2001a ) and if the universe became transparent at wavelengths below 228 following the reionization of helium at @xmath131 ( bernardi et al .
2003 ; vladilo et al .
photons with @xmath132have sufficient energy to ionize c@xmath6 and thereby reduce the c iv / c@xmath133 ratio ( dav et al .
1998 ) .
if , on the other hand , the metals are ejected from star - forming lbgs , the abundance and ionization fraction of c atoms are likely to depend more closely on local conditions , rather than those of the igm at large . as suggested by adelberger et al .
( 2003 ) , the approximately constant value of @xmath28 may then simply mirror the behaviour of the cosmic star formation rate density , which remains essentially flat over the redshift interval @xmath134 ( steidel et al .
1999 ) . in this picture , while the mean metallicity of universe grows with cosmic time , one has to assume that this growth is not reflected in the quantity @xmath135 , perhaps because the systems which make the larger contribution to this integral ( in current data sets ) are associated with outflowing interstellar gas .
the true intergalactic metals may be those at the low column density end of @xmath22 , below @xmath76(c iv)@xmath136 ( haehnelt 1998 ) , where data are still limited to a few sight - lines to the brightest qsos ( ellison et al . 2000 ) .
note also that the results of adelberger et al .
apply to galaxies and the igm at @xmath137 , and it remains to be established whether a similar picture holds at higher and lower redshifts . at @xmath138 chen , lanzetta , & webb ( 2001 ) do find that c iv absorption systems are clustered around galaxies on velocity scales of up to @xmath139 km s@xmath27 and linear scales of up to @xmath140kpc , but those systems are generally stronger than the ones considered here .
it is possible that both enrichment mechanisms are at work , and it is also perhaps conceivable that , by coincidence , all complicating effects described above might work in opposite directions and compensate each other to maintain the approximately invariant @xmath28 found by songaila ( 2001 ) and confirmed here .
these possibilities can only be assessed quantitatively with detailed calculations which are beyond the scope of this paper . from an observational point of view , improving the sensitivity of the spectroscopy presented in this paper to include weaker c iv systems is a very challenging task at @xmath141 , even with 8 - 10 m telescopes . on the other hand
, there is an incentive to extend this work to even higher redshifts . detecting even only the strongest c iv absorbers at @xmath15 ( which will require paying special attention to the problem of correcting for atmospheric absorption )
, would still provide an extremely important probe of the star formation activity at very early epochs . + support for this work
was provided by nsf grant ast-0205738 and by nasa grant nag5 - 11513 ( p.m. ) .
we are grateful to the esi team for developing the efficient instrument which made this study possible ; to the staff of the keck observatory for their competent assistance with the observations ; and to bob carswell for providing the vpfit software .
we are indebted to robert becker , richard white , and michael strauss who kindly contributed most of the esi exposures of sdss 1030 + 0524 .
we acknowledge valuable discussions with martin haehnelt .
max pettini thanks the instituto de astrofsica de canarias for their hospitality during a visit when this work was completed .
finally , we wish to extend special thanks to those of hawaiian ancestry on whose sacred mountain we are privileged to be guests . without their generous hospitality , the observations presented herein would not have been possible .
+ sdss0231@xmath320728 & 5.421 & 19.19 & 2002jan13 & & 10800 & 5020 & 1435 + & & & 2002feb07 & & & + sdss0836 + 0054 & 5.803 & 18.74 & 2002jan13 & & 32400 & 8040 & 817 + & & & 2002feb07 & & & + sdss1030 + 0524 & 6.305 & 20.05 & 2002jan10,11,12 & 28500 & 6015 & 1145 + 1 & @xmath142 & @xmath143 & @xmath144 & + 2 & @xmath145 & @xmath146 & @xmath147 & + 3 & @xmath148 & @xmath149 & @xmath150 & + 4 & @xmath151 & @xmath152 & @xmath153 & + 5 & @xmath154 & & @xmath155 & marginal + 6 & @xmath156 & @xmath157 & @xmath158 & + 7 & @xmath159 & @xmath160 & @xmath161 & + 8 & @xmath162 & @xmath163 & @xmath164 & marginal + 1 & @xmath165 & @xmath166 & @xmath167 & + 2 & @xmath168 & @xmath169 & @xmath170 & + 3 & @xmath171 & & @xmath172 & marginal + 4 & @xmath173 & & @xmath174 & lines are saturated + 5a & @xmath175 & @xmath176 & @xmath177 & poor fit + 5b & @xmath178 & @xmath179 & @xmath180 & poor fit + 5c & @xmath181 & @xmath182 & @xmath183 & poor fit + 6 & @xmath184 & @xmath185 & @xmath186 & marginal + 7 & @xmath187 & @xmath188 & @xmath189 & + | in order to search for metals in the lyman@xmath0 forest at redshifts @xmath1 , we have obtained spectra of high signal - to - noise ratio and moderately high resolution of three qsos at @xmath2 discovered by the sloan digital sky survey .
these data allow us to probe to metal enrichment of the intergalactic medium at early times with higher sensitivity than previous studies .
we find 16 c iv absorption systems with column densities @xmath3 over a total redshift path @xmath4 . in the redshift interval @xmath5 , where our statistics are most reliable
, we deduce a comoving mass density of c@xmath6 ions @xmath7 ( 90% confidence limits ) for absorption systems with @xmath8 ( for an einstein - de sitter cosmology with @xmath9 ) .
this value of @xmath10 is entirely consistent with those measured at @xmath11 ; we confirm the earlier finding by songaila ( 2001 ) that neither the column density distribution of c iv absorbers nor its integral show significant redshift evolution over a period of time which stretches from @xmath12 to @xmath13gyr after the big bang .
this somewhat surprising conclusion may be an indication that the intergalactic medium was enriched in metals at @xmath14 , perhaps by the sources responsible for its reionization .
alternatively , the c iv systems we see may be associated with outflows from massive star - forming galaxies at later times , while the truly intergalactic metals may reside in regions of the lyman@xmath0 forest of lower density than those probed up to now . |
the analysis of remote - sensing observations and in - situ measurements require very different expertise .
the physical processes that are remotely sensed must be analysed by first accounting for the mechanisms that generate or scatter the observed electromagnetic radiation .
these mechanisms will differ according to the wavelength detected and the height at which the solar atmosphere is observed . in the optically thin corona for instance
, each image pixel detects light integrated along the line of sight that passes through an extended region of the corona . at photospheric and chromospheric altitudes
the medium is more optically thick . in -
situ measurements made further out in the solar wind provide directly the physical parameters of the plasma at a single point but they are exact .
remote - sensing observations are our only source of information of the state of the low corona , solar imagery also provides a large - scale view of the physical processes at play . by combining both remote - sensing observations and
in - situ measurements we can gain a more complete picture of the sun - earth system over a wide range of spatial and temporal scales .
+ the very different nature of remote - sensing observations and in - situ measurements coupled with the large distances that separate the solar corona with current in - situ measurements , meant that solar and solar wind research evolved separately over many decades , all too often without those involved consulting each other when working on the same problem .
the databases used by researchers also evolved independently and little concerted effort was made to link them .
rare attempts have been made to connect some datasets by carrying out ballistic tracing of heliospheric structures between the sun and planets , or probes .
one example of such an undertaking was the first propagation tool developed by the helio fp7 project ( http://hfe.helio-vo.eu/helio/ ) .
these first attempts lacked the observational support to validate such ballistic tracing and the user was left uncertain about the validity of the tracing method .
the launch of the heliospheric imagers ( hi ) instruments onboard the solar - terrestrial relations observatory ( stereo ) in 2006 has provided imaging of the solar wind with sufficient resolution and cadence to enable plasma structures to be clearly identified and tracked as they propagate through the inner heliosphere ( see for example review by rouillard 2011a ) . the propagation tool developped at the institute of research in astrophysics and planetology ( irap ) , accessible at http://propagationtool.cdpp.eu/ , and presented in this paper exploits , in particular , heliospheric imagery as observational support to connect solar imagery with in - situ measurements .
+ the tool was created to support research in heliophysics by establishing connections between different databases that often hold complementary datasets on the sources , evolution and effects of propagating heliospheric structures .
the different databases that are currently connected to the propagation tool are shown in figure [ fig : database ] .
they include the medoc solar imagery datacenter ( https://idoc.ias.u-psud.fr/medoc ) , the cdpp in - situ datacenter ( http://cdpp.eu ) and the auroral planetary imaging and spectroscopy ( apis ) database ( http://apis.obspm.fr/ ) .
in addition the tool has access to the simple applications messaging protocol ( samp ) that allows the propagation tool to exchange data with other tools or users through a central hub and the virtual observatory for planetary science ( vespa ) which can be interrogated to access a number of other databases .
once the trajectory of a heliospheric structure is determined by the tool , the various databases can be interogated to determine if interesting data is available along its trajectory
. + [ [ structure - and - aim - of - the - paper ] ] structure and aim of the paper + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + after a brief introduction of the different types of propagation that can be carried out with the tool and the corresponding exploitation of heliospheric imaging ( section [ sec : overview ] ) , we describe the general layout of the components visible by default in the main interface ( section [ sec : maininterface ] ) .
each component is then described separately by illustrating some of its functionalities ( section [ sec : paraminterface ] ) to [ sec : vplotinterface ] ) .
we also show how different databases can be interogated from the propagation tool ( section [ sec : accessdatabases ] ) and how the tool has been used so far for various studies published in the litterature ( section [ sec : research ] ) .
the aim of the paper is to highlight the rich set of functionalities provided by the tool , a good understanding of how the tool works and its full potential can only be achieved by using it .
in order to establish connections between the effects of a particular solar wind structure ( cme , cir , streamer blob , jet , ... ) at different locations in the corona / heliosphere system , the propagation time between the start and end points , as well as the spatial extent of the solar wind structure , must be computed as accurately as possible . the propagation time is calculated once the kinematic and spatial properties of that propagating structure
are either specified by ( 1 ) the users manually , ( 2 ) selection from a catalogue of pre - identified heliospheric structures or ( 3 ) determination through a bespoke analysis undertaken with the tool .
heliospheric imagery , presented in the tool in the form of time - elongation maps ( sheeley et al .
1997 ; davies et al . 2009 ) of solar wind density variations , provides the crucial information needed to confirm the trajectory of the outflowing structure .
the tool provides a visualisation interface to these time - elongation maps ( often called j - maps ) .
the j - maps extend in elongation from the solar corona , imaged by coronagraphs , through the inner heliopshere , out to 1au and beyond imaged by heliospheric imagers .
the generation of these maps will be discussed in more detail in section [ sec : jmapinterface ] .
as the signal within a pixel integates along the line - of - sight , a specific elongation does not correspond to a unique height above the solar surface and analysis techniques must be applied to determine the 3-d trajectory of the heliospheric structure of interest .
these techniques are usually based on ballistic propagation techniques ( see section [ sec : jmapinterface ] ) .
+ the propagation tool offers three simple types of propagations that allow users to : * propagate radially - outflowing structures ( cmes ) sunward or anti - sunward ( radial propagation method ) , * propagate corotating structures ( cirs ) forwards in time ( corotation method ) , * propagate solar energetic particles ( seps ) along magnetic fields lines sunward or anti - sunward ( sep propagation method ) . for the radial propagation and corotation methods ,
the tool offers three different ways to use j - maps that are extremely useful to evaluate the accuracy of a particular propagation technique by comparing the model with real observations . + a summary of the different types of propagations and modes of j - map usage
are given in figure [ fig : proptech ] .
we provide examples of the different ways j - maps can be used in section [ sec : jmapinterface ] . for the three propagation methods
, we employ the most simple ballistic techniques so that navigation back and forth between imagery and in - situ data browsing remains intuitive .
the main interface is shown in figure [ fig : maininterface ] and displays seven components of the tool .
the set - up of the displayed interface is for the radial propagation method of a radially - outflowing structure such as a cme . for a user - defined time
( area a ) , the orbital positions of various probes , planets and comets are derived from their spice kernels and shown in the ecliptic plane ( area b ) .
area e is the interface through which the users defines the starting point of the transient , its speed at that point and its direction of propagation . in the example shown in figure [ fig : maininterface ]
, the transient is assumed to move outward at a constant speed from sun to earth . with the parameters entered in area e ,
the propagation of the transient is computed by the tool and the result is displayed on the ecliptic plane and if an end point is also entered in area e ( here the earth ) , the arrival time is computed at that end point . by default , the simplest geometry is assumed for the radially - outflowing transient ( area c ) , it is an arc of a sun - centered circle moving outward over time .
the color - code corresponds to the time , in hours , for the leading edge to reach a certain distance away from the sun ( see color bar ) .
the earth corresponds to the blue filled circle always situated below the sun .
the map appearing in the top right - hand side ( area d ) is a carrington map ; it shows a gray - scale map of extreme ultraviolet emission ( 195 ) of the corona from the imager on stereo - a .
the choice of the carrington rotation number is defined by the launch time of the radially - outflowing transient computed automatically by the tool once the parameters are defined in area e. the extent of the transient projected onto this map is marked by an ellipse which corresponds to the projection of the angular cone sustained by the transient s latitudinal and longitudinal extent .
access to the different databases is available in the bottom right - hand part of the interface ( area f ) . by default ,
the tool also loads a j - map built with stereo - a observations .
this j - map interface , shown in area g , replaces areas e and f when the button j - map interface is pressed in area e. the red curve superposed on the j - map tracks the angular position of the center of the radially outflowing cme seen in the ecliptic plane as viewed from stereo - a .
this will be explained in more detail in section [ sec : overview ] .
the field of view displayed on the ecliptic plane corresponds to the combined fields of view of the imagers on stereo - a that were used to construct the j - map shown in area g. + figure [ fig : maininterface ] gives the general layout of the interconnected components of the tool that allow , in this example , a simultaneous visualisation of a transient s radial outflow through the inner heliosphere .
to summarise , for a given propagation technique , here ` radial propagation ' combined with ` j - map : carrington / in situ ' selected in area c , a transient s launch time , kinematic properties , extent and direction of propagation can be defined in areas a and e. from these properties , the derived source region and extent of the transient in the low corona are projected on a carrington map ( area d ) and its propagation in the interplanetary medium can be visualised on the ecliptic plane ( area b ) and on a j - map ( area g ) . once a propagation has been carried out , the tool can interrogate different databases at the relevant times by selecting the buttons in area f. as we shall see , if ` j - map : catalogue of fits ' or ` j - map : click to fit ' are selected instead in area c , the user defines the kinematic properties and the trajectory of the propagating heliospheric structure ( cir , cme ) using the j - map interface and therefore these properties can no longer be entered manually in area e. in the next sections we illustrate the rich set of functions available through the individual components of the tool by considering each component individually . in order to define the start and end points as well as the properties of the cme , cir or sep event of interest , the user can enter parameters in area e of figure [ fig : maininterface ] .
the three different formats of that interface , corresponding to the three propagation methods , are shown in figure [ fig : paraminterface ] .
although adapted to the different propagation methods , they all share a similar layout .
they correspond to the interface displayed in the j - map : carrington / in situ mode which allows the parameters of the outflowing transients to be entered manually and not defined by the result of an analysis based on fitting tracks in the j - map ( j - map catalogue of fits and j - map : click to fit ) .
the start and end points of the propagation are input by selecting the corresponding button of the interface , which opens a scroll down menu with a list of selectable planets , probes and the sun .
those fields of the interface into which values can be entered manually are coloured blue fields .
fields that are not modifiable - because they display either results of the propagation ( end row ) , parameters from pre - defined catalogues ( ` j - map : catalogues of fits ' ) , or the results of j - map analysis ( ` j - map : click to fit')- are coloured orange . the solar source location and radial speed of the cme ( figure [ fig : paraminterface]a ) , of the cir ( figure [ fig : paraminterface]b ) and of the magnetic spiral along which the seps propagate ( figure [ fig : paraminterface]c ) are only selectable in the mode j - map carrington / in situ. this latter mode offers full flexibility to define the properties of the propagating structures and seps .
in addition , the radial propagation method ( figure [ fig : paraminterface]a ) provides an option to activate the drag - based model developped originally by vrnsak et al .. the drag coefficient and the height in the atmosphere where the user deems necessary to invoke the drag - based model must then be defined .
this is the only option available in the tool to vary the speed of a propagated structure .
the sep propagation mode requires specification of the type and energy of particles to be propagated ( figure [ fig : paraminterface]c ) .
errors in the arrival times are also computed .
they account for the uncertainty in the radial speed of the cme and cirs or the length of the spiral channeling seps .
each time a parameter is changed , the propagation is re - run and all components of the tool are updated .
the parameters necessary for a propagation to be carried out are : * radial propagation : launch time ( ut ) , radial speed ( km / s ) and source longitude of the cme ( in carrington or hae coordinates ) , * corotation : launch time ( ut ) , radial speed ( km / s ) , rotation period of the sun ( in days ) , source longitude of the cir ( in carrington or hae coordinates ) , * sep propagation : launch time ( ut ) , energy ( mev ) and source longitude ( in carrington or hae coordinates ) of the seps as well as the properties of the parker spiral defined by the speed of the solar wind ( km / s ) and the rotation period of the sun ( in days ) .
the propagation time of the particles along these spirals are computed by dividing the speed of the particle to the distance travelled along the parker spiral between the start and end points .
the energy of the particles are of course converted to speed by accounting for relativistic effects .
the interface labelled b in figure [ fig : maininterface ] is repeated in figure [ fig : ecliptic ] for the three different types of propagation .
this figure shows the geometries used to study radially - outflowing transients ( panels a , b ) , corotating structures ( c ) and beams of energetic particles ( d ) .
the horizontal scroll bar at the top of the ecliptic plane can be moved to let time flow , fowards or backwards , and enable the location of the modeled heliospheric structure or sep stream to be visualised in the ecliptic plane as a function of time . + * the radial propagation * method exploits two techniques based on different geometries .
the simplest geometry assumes that the cme is the arc of a circle with the sun at its center .
as the arc moves radially outwards , with the same speed along its entire front , it forms the color - coded sector visible in figure [ fig : ecliptic]a .
the second geometry assumes the cme is the intersection of the 3-d ice - cream cone model with the ecliptic plane , the ice - cream cone model is used routinely to model cmes in coronagraphic observations .
the resulting 2-d shape for the case is shown as the gray - shaded area in figure [ fig : ecliptic]b .
such a shape has been used to located cmes in coronagraphic in heliospheric images .
it has been called the self - similar expansion technique when used to located cmes in j - maps . due to the effect of the line of sight integration
, the angular distance away from the sun - observer line of these two cme shapes will be quite different ( even for the same speed , direction of propagation and longitudinal extent of the cme ) , this is discussed in section [ sec : jmapinterface ] .
the propagation times computed with the two techniques will be different when the end point is situated off the central axis of the outflowing transient because the curvatures of the two arcs are different .
the tool accounts for these differences and also considers the orbital motion of the target planet / spacecraft during the propagation time of the cme from the start to the end point . +
* the corotation * method assumes that the structure is a corotating archimedean spiral defined by the speed of the solar wind and the solar rotation period ( figure [ fig : ecliptic]c ) .
the computed propagation time between the start and end points folds in the necessary radial motion and corotation .
it also accounts for the orbital motion of probes and planets between the start and end times .
the latter is particularly important for the propagation of cirs over many days since the planets and probes can move several degrees in longitude along their orbits over these long time intervals . + * the sep propagation * assumes that the spatial extent of the beam of energetic particles is represented by a bundle of parker spirals ( figure [ fig : ecliptic]d ) .
the longitudinal extent of these spirals are defined by the user and should correspond to the extent of the particle source region , for instance of a triangulated shock in the low corona ( e.g. rouillard et al . 2016 ) .
solar energetic particles of interest are usually so fast that typically the spiral does not move greatly between the start and end times . even the low - energy suprathermal electrons with energies of 300ev take at least 3 - 4 hours to reach earth along a spiral propagating outwards with a speed of 300 @xmath0 .
nevertheless even during that transit time , the spiral will corotate by 2 - 3 degrees ; the tool will tell check that the earth has not moved out of the corotating parker spiral between the release and impact times of the particles .
the interface labelled d in figure [ fig : maininterface ] is repeated in figure [ fig : carrington_map ] .
it offers a visualisation of carrington maps of the solar surface in order to ease the definition / identification of the source region of propagating heliospheric structures .
the identifiable coronal features are typically coronal holes , active regions and flares .
the accessible database of carrington maps includes maps constructed from extreme ultraviolet and white - light images of the corona as well as maps of magnetic measurements at the photospheric level . in the near future
we will also provide results of numerical simulations of solar wind simulations at different altitudes in the atmosphere in order to locate likely source regions of the fast and slow solar winds . a right - click on
the carrington map opens up a menu which is adapted to the propagation method .
this menu is shown in figure [ fig : carrington_map ] below the carrington map for the radial propagation method .
this menu provides access to the different types of carrington maps that can be selected , as well as the the positions of c , m and x - class flares on the solar surface near the estimated launch time of the radially - outflowing transient .
this can be used to confirm the release time of a cme for instance or re - define the location of the cme source region by right - clicking on the carrington map ( in the ` j - map carrington / in - situ mode ' only ) .
the carrington coordinates of the planets ( circles ) and probes at the launch time of the cme can also be projected on the map for the purpose of comparing the longitudes of the cme and the planets / probes .
this projection will change when the user uses the time scroll bar in the interface of the ecliptic plane . for the other propagation methods ,
the projection method is different ; in the corotation tool the source location is defined by the longitude of the footpoint location of the cir spiral at the impact times of the cir at the different planets / probes . for a typical synodic rotation rate ,
all planets appear on the carrington map at roughly the same location with only slight relative displacement corresponding the orbital motion of the impacted objects . for the sep propagation
, the projected position on the carrington map corresponds to the footpoint location of the spiral at the launch time of the seps so that magnetic connectivity of each planet and probe to the low corona is evaluated at the single release time of the particles .
currently the magnetic connectivity assumes a parker spiral all the way down to the solar surface .
we plan , in the near future , to trace magnetic connectivity with more realistic models of the coronal and interplanetary medium . the extent in latitude and longitude of the cme , cir spiral and the beam of seps is shown in an identical way on the carrington map : they correspond to the same orange ellipse representing the intersection of the geometrical shapes assumed for the different heliospheric structures ( cme , cir spiral and the parker spirals channeling the seps ) .
this component , shown in figure [ fig : vplot ] is only accessible in the ` j - map : carrington - in situ ' mode .
the v - plot replaces the carrington map when the user selects v - plot and decides to start the propagation from a probe or planet where the radial speed of the heliospheric plasma has been directly measured in - situ or simulated numerically .
this functionality allows users to define the propagation speed of the heliospheric structure ( cme , cir , parker spiral ) with direct measurements instead of imagery .
the passage of a cme shock over earth ( in omni data ) is shown in figure [ fig : vplot ] and corresponds to the cme that emerged from the solar surface on 2010 april 3 ( rouillard et al .
once a speed has been selected on the v - plot , all components of the tool are automatically updated .
a cursor allows the user to specify the radial speed of the heliospheric structure propagated with the tool .
scroll down menus are accessible via this interface to select whether real or simulated data should be plotted in the interface .
in addition this component offers the possibility to access a large number of icmes , cir and sep catalogues , including those produced by the fp7 helcats project .
the j - map interface is probably the most interesting asset of the propagation tool since it offers an interactive visualisation and exploitation of maps of solar wind outflows obtained from the stereo spacecraft . the way j - maps are produced with hi-1/2 images
is described in e.g. davies et al .
a simple visualisation of the maps can not reveal along which heliocentric longitude a particular heliospheric structure is propagating .
indeed each image pixel used to construct a j - map results from the integration of the light scattered by solar wind electrons along the the line of sight defined by that pixel s pointing direction .
however the rate of change of the cme s elongation ( angular distance to the sun ) with respect to time is to a large extent defined by the longitudinal separation between the cme and the observer ( e.g. rouillard et al .
2008 ) . using simple geometrical shapes representing a cme ,
the observed time - varying elongation angle can be fitted to derive a cme s direction of propagation relative to the sun - observer line ( rouillard et al .
2008 , lugaz et al .
2010 , davies et al . 2012 ) .
the interface allows superposition on the j - maps of the theoretical trajectories obtained from either ( 1 ) cme properties ( direction of propagation , speed , extent ) defined manually by the user ( ` j - map : carrington - in situ ' mode ) , ( 2 ) from pre - defined catalogues of cme trajectories ( ` j - map : catalogue of fits ' mode ) or ( 3 ) directly fitted by the user ( ` j - map : click to fit ' mode ) .
these three modes are selected in the area c of the main interface ( figure [ fig : maininterface ] ) .
the two techniques exploited in the propagation tool to map a cme s heliocentric position into an angular position are the so - called fixed - phi techniques ( sheeley et al .
1999 ; rouillard et al .
2008 ) and the self - similar expansion technique ( davies et al .
2012 , moestl and davies 2013 ) .
the fixed - phi approach assumes that the cme is a single point ( hence sometimes also called fixed - point technique ) situated at the center of the cme arc shown in panel a of figure [ fig : ecliptic ] .
the self - similar approach assumes that the outermost position of the cme in the heliospheric images is defined by the intersection of the line passing by the observer and which is tangent to the semi - circular structure shown in panel b of figure [ fig : ecliptic ] .
+ we present in figure [ fig : jmap_interface_cme ] the three different ways in which the j - map can be used to visualise cme trajectories . at this stage ,
only ecliptic j - maps are visualised because the tool was devised to connect in - situ measurements , which up to now , have been primarily taken in the ecliptic plane .
extension to other latitudes and to a fully 3-d geometry is planned in preparation for science support of the solar orbiter mission ( see section [ sec : futuremissions ] ) .
results of using the fixed - phi technique are shown in the top and bottom j - maps shown in figure [ fig : jmap_interface_cme ] and the result of using the self - simiar expansion technique is shown in the middle j - map .
superposed on the j - map in the top panel is the apparent fixed - phi trajectory ( as a red line ) of the cme that erupted on 17 may 2012 ( rouillard et al . 2016 ) and whose properties were defined in this example by selecting manually the source region near the location of the m - class flare identified on the carrington map also shown in that panel ( it is the same rotation as in figure [ fig : carrington_map ] ) .
the modeled and real tracks match well for a radial speed of 862 km / s defined manually in the parameter interface and by comparing the observed with the theoretical tracks on the j - map .
this approach confirms the idea that a cme propagated outwards from the flare location for a reasonable speed value .
also shown are the time - varying elongations of the earth , mars and venus during the displayed time interval .
the ` j - map : carrington - in situ ' mode offers the flexibility to start from a spacecraft taking in - situ measurements ( through the v - plot interface ) to define the time of passage of the cme at that spacecraft . for such an approach the cme trajectory is defined by the longitude of the spacecraft and the speed selected in - situ via the v - plot interface ( figure [ fig : vplot ] ) .
+ the middle panel displays the ` j - map : catalogue of fits ' interface .
the user can overlay , onto the j - map , tracks from a pre - defined catalogue produced by the fp7 helcats project corresponding to either of the two geometries described above .
the catalogues are accessible by right - clicking on the j - map through the menu displayed thereon . in this example
, the blue lines correspond to the fitted elongation variations obtained with the self - similar expansion technique . to select a particular track , the user right - clicks on the map and
chooses the select track function and the selected track becomes red .
the tool is then updated assuming the catalogue s cme trajectory and its kinematic properties .
this includes of course the ecliptic plane shown on the right - hand sides .
the helcats catalogues were derived using j - maps generated along the cme central axis .
often this does not correspond to the ecliptic plane . since
the tool plots these theoretical tracks on j - maps constructed along the ecliptic plane , there can be a discrepancy between the theoretical and observed tracks due to this latitudinal difference .
the pas considered to fit the cme trajectories as well as the northernmost and southernmost pas at which the cme was observed in heliospheric images are given below the j - map for the purpose of comparison .
more information on these catalogues are given in harrison et al .
( 2016 ) . in order to gain some insight on how different the results of the fixed - phi and the self - similar technique can be , the user is refered to the paper by moestl and davies ( 2013 ) or by using the propagation tool .
+ the bottom panel shows the interface when ` j - map : click to fit ' is selected . in this mode ,
the user can right click on the j - map and selects the function ` enter points by click ' , to enter points situated on the observed track of interest . at this stage , the tool can only fit the selected points with the fixed - phi technique and all of the components of the tool are updated automatically once the trajectory is derived . the fitting technique accounts for the motion of the imaging instrument along its orbit over the duration of the fit .
once a cme trajectory is determined , the arrival times of the cme at any planet and many probes can be computed and is made available the interface table of arrival times ( not shown here ) .
these arrival times account for the geometry of the cme and the orbital motion of the spacecraft / planet . for long propagation times a planet or probe can move out of the cme s path , the tool will take that into account as well
. + the three j - map tools discussed above are also available for propagation by corotation . in j - maps constructed from stereo - a / stereo - b hi images , characteristic patterns of converging / diverging tracks
appear due to corotating density structures ( such as cirs ) as demonstrated by rouillard et al .
( 2008 , 2010a ) and sheeley et al .
( 2008b ) .
each track in any such pattern corresponds to the thomson - scattered white - light signature of a strong density inhomogeneity ( a so - called ` density blob ' ) moving radially outward from the sun . beyond about 0.5 au (
situated roughly in the overlap region between the hi-1 and hi-2 fields of view ) , these inhomogeneities become entrained ahead of corotating high - speed streams ( e.g. rouillard et al . , 2008 ) .
because these density structures are emitted by a spatially limited source region on the sun , they rapidly form a spiral of density inhomogeneities in the interplanetary medium .
this spiral is analogous to the parker spiral formed by the interplanetary magnetic field , as both trace approximately the locus of plasma emitted by a single corotating source region on the sun .
the three j - map tools available to analyse these corotating structures are displayed in figure [ fig : jmap_interface_cir ] . in these examples
, they all show the same passage of a cir that formed due to an extensive low - latitude coronal hole imaged by nasa s solar dynamics observatory between the 25 may 2013 ( appearance on the east limb ) and the 03 june 2013 ( disappearance behind the west limb ) .
the coronal hole was also imaged by the stereo - a spacecraft and the associated carrington map ( rotation 2137 ) is shown in the top panel situated above the j - map .
this cir has been analysed extensively by sanchez - diaz et al .
( 2016 ) because multi - point observations of the corona during this cir passage revealed , for the first time , the formation height of the density inhomogeneities ( blobs ) entrained by the cir .
the source region of a cir can often be located near the western edge of these low - latitude coronal hole , where we expect the fast and slow solar wind to form at adjacent longitudes and to become radially aligned in response to solar rotation at high altitudes . to reconstruct the cir pattern
, the tool assumes that at a single blob is emitted along a longitude near the western boundary of the coronal hole ( in the ecliptic ) , this produces the red track on the j - map . the cir pattern that results from a succession of blobs emitted every 8 hours from that same corotating
longitude is shown as the black curves in the top panel . a good match between the family of black curves and
the observed pattern of converging tracks confirms the tool s localisation of the cir .
plotnikov et al .
( 2016 ) showed that at solar minimum all cirs that eventually impact a spacecraft can be located using the heliospheric imagers .
hence , acatalogue of cirs could be derived for the fp7 helcats project and the cir described above is part of this catalogue ( plotnikov et al .
the catalogue gives an average speed for the modeled cir of 440 km / s and confirms the source region near the western boundary of the giant coronal hole already mentioned .
the selection of this catalogue is done in an analogous manner to the cme catalogue by right - clicking on the j - map .
the bottom panel shows the cir pattern left by the same cir a few days early in the stereo - b imager .
cirs are difficult to identify in stereo - b images because the pattern of diverging tracks is not always visible .
one feature that nearly always appears is however the track left by the cir when the tangent to the cir curve passes by stereo - b ( sheeley and rouillard 2010 ) .
this configuration occurs during an approximately 5-day long time - window .
the associated track is fitted in figure [ fig : jmap_interface_cir ] to the clicked blue crosses . the source longitude of the cir passing in stereo - b s imager
is confirmed to be situated near the western boundary of the coronal hole shown in the carrington map .
once a cir is located in the j - map , it can be propagated to any planet or probe in the clock - wise direction of solar rotation only .
analogous to the radial propagation , the orbital motion of the spacecraft and probes are accounted for when the tool calculates the impact times .
+ the j - maps can not directly provide information on sep propagation , hence the tool does not offer the possibility to access the j - map interfaces when the sep propagation is selected . as mentioned in the introduction
, the propagation tool was created to provide a simple way to propagate heliospheric structures in order to connect remote - sensing observations with in - situ measurements .
once a propagation has been carried out with the tool and a launch time has been estimated at the sun , it is possible to launch two versions of the esa / nasa helioviewer interface straight from the propagation tool ; these are the helioviewer webpage and the java webbased interface called jhelioviewer .
both versions were modified to read data stored at the medoc datacenter with the aim of providing a back - up site in case data becomes unavailable at the two other different servers running helioviewer ( goddard space flight center , royal observatory belgium ) .
jhelioviewer offers advanced viewing options , including running and base difference images , a side - by - side ( multiview ) mode , image projections such as orthographic , latitudinal and polar , timeline displays of 1d and 2d datasets , synchronized with image time series .
in addition to compare the launch times of the heliospheric structures propagated in the propagation tool , jhelioviewer can display features and events from the heliophysics events knowledgebase , and alerts from the comesep system .
the steps followed to launch jhelioviewer are shown in figure [ fig : databases2 ] .
they begin by selecting the black button labelled helioviewer. this opens a pop - up window where the user enters a time of interest , a spacecraft , an instrument and the desired type of observation ( euv , white light ... ) . depending on whether the http or java version of helioviewer is then selected
, the tool will either launch the webpage or a version of jhelioviewer ( preinstalled on the user s computer ) with the requested images / movies .
+ to browse in situ data at the impact times predicted by the propagation tool a similar procedure is followed by pressing on the black button labelled cdpp interface. in the launched pop - up , the times , spacecraft and the types of datasets are selected manually .
this generates a requests via webservice to the amda tool at cdpp , amda then creates a plots that is displayed as a page in the user s web browser .
the order of solar wind parameters displayed in the created plot are pre - defined . to carry out a more in depth analysis of the in - situ data ,
the user can launch the full data mining environment offered by the amda tool at cdpp ( amda.cdpp.eu ) .
amda is a multidataset visualisation and download software , it can do automated event search and data mining , it provides access to existing catalogues and allows users to produce their own catalogues . most of the omniweb database can be accessed via amda .
+ the interoperability of the propagation tool has been recently expanded with a direct connection to the apis database ( http://apis.obspm.fr/ ) of auroral images . the connection is made possible by a data exchange protocole called epn - tap , developed and promoted by the vespa infrastructure of europlanet 2020 ri ( http://www.europlanet-vespa.eu/ ) .
when a solar wind structure impacts a planet ( most appropriately a giant planet ) hitting the apis interface button launches a search in apis for the same target and for an interval of time bracketing the structure arrival time .
if the search returns results ( in the form of raw / processed / projection images ) they can be either browsed ( as thumbnails are available ) , or visualized ( in case of pdf or png ) , or selected for full display and further analysis in companion tools like aladin or topcat ( data are then sent via the ivoa samp protocole ) .
these new capabilities will be discussed in a future publication that presents in more detail the recent catalogues that have been produced by international teams and integrated in the propagation tool to further support research in planetary space weather .
this article has presented the core functionalities available in the tool .
the large number of links between the numerous components of the interface provide a very rich combination of possible ways in which imagery , modelling and in - situ data can be used to address a number of research topics .
one of the major advantages of this tool resides in the speed at which datasets normally accessed via many different websites can be compared in a few clicks only .
this is reflected in the variety of different studies that have been recently published that have carried out part of their analysis on results using obtained with the propagation tool .
they include studies investigating the long - term variations of corotating density structures ( plotnikov et al .
2016a ) , the relation between cmes and solar energetic particles ( salas - matamoros et al . 2016
; rouillard et al . 2016 ) , the relation between cmes and solar gamma - ray events ( plotnikov et al .
2016b ) as well as the effect of a cme at many planets on its way to outer heliosphere ( witasse et al .
+ the simple propagation techniques and wide range of connected datasets also make the tool useful for university lecturing and doctoral and post - doctoral summer training schools .
the propagation tool was used interactively at the course entitled heliospheric physical processes for understanding solar - terrestrial relations held in september 2015 in the international school on space science in lacquila , italy http://www.cifs-isss.org/pastcourses.asp ) .
the propagation tool exploits heliospheric imagery to connect imagery with in - situ measurements . the upcoming solar probe plus and solar orbiter missions will also carry heliospheric imagers , they are the wispr ( vourlidas et al .
2015 ) and solohi ( howard et al . 2010 ) instruments .
we plan to modify the propagation tool to include images of these two instruments . because the orbit of solar orbiter will eventually move out of the ecliptic planes
, all the components of the propagation tool will need to be extended to 3-d , this includes tracking techniques , presentation of the propagation in the ecliptic plane and the way we visualise j - maps .
we also hope to include ground - based imagery such as spectro - polarimetric measurements obtained by comp instrument in hawaii and soon the climso instrument at pic du midi which provide additional information on the topology of the coronal magnetic field .
we acknowledge support from the plasma physics data center ( centre de donnees de la physique des plasmas ; cdpp ; http://cdpp.eu/ ) , the virtual solar observatory ( vso ; http://sdac.virtualsolar.org ) , the multi experiment data @xmath1 operation center ( medoc ; https://idoc.ias.u-psud.fr/medoc ) , the french space agency ( centre national des etudes spatiales ; cnes ; https://cnes.fr/fr ) and the space weather team in toulouse ( solar - terrestrial observations and modelling service ; storms ; https://stormsweb.irap.omp.eu/ ) .
this includes the data mining tools amda ( http://amda.cdpp.eu/ ) and the propagation tool ( http://propagationtool.cdpp.eu ) .
, i.p . and
esd acknowledge financial support from the helcats project under the fp7 eu contract number 606692 .
europlanet 2020 ri has received funding from the european union s horizon 2020 research and innovation programme under grant agreement no 654208 .
the _ stereo _ _ secchi _ data are produced by a consortium of _ ral _ ( uk ) , _ nrl _ ( usa ) , _ lmsal _ ( usa ) , _ gsfc _ ( usa ) , _ mps _ ( germany ) , _ csl _ ( belgium ) , _ iota _ ( france ) and _ ias _ ( france ) .
+ j. a. davies , et al .
, a synoptic view of solar transient evolution in the inner heliosphere using the heliospheric imagers on stereo , geophys .
36 ( 2009 ) l02102 .
http://dx.doi.org/10.1029/2008gl036182 [ ] .
n. lugaz , accuracy and limitations of fitting and stereoscopic methods to determine the direction of coronal mass ejections from heliospheric imagers observations , sol .
( 2010 ) 411429 .
http://arxiv.org/abs/1010.1949 [ ] , http://dx.doi.org/10.1007/s11207-010-9654-9 [ ] .
c. moestl , j. a. davies , speeds and arrival times of solar transients approximated by self - similar expanding circular fronts , sol .
( 2013 ) 411423 .
http://arxiv.org/abs/1202.1299 [ ] , http://dx.doi.org/10.1007/s11207-012-9978-8 [ ] .
i. plotnikov , et al . ,
long - term tracking of corotating density structures using heliospheric imaging , solar physics 291 ( 2016 ) 18531875 .
http://arxiv.org/abs/1606.01127 [ ] , http://dx.doi.org/10.1007/s11207-016-0935-9 [ ] .
i. plotnikov , a. rouillard , g. share , far - side coronal mass ejections : shock front magnetic connectivity to the visible disk and the origin of long - duration gamma - ray flares , astronomy and astrophysics ( 2016 ) submitted .
a. p. rouillard , et al .
, deriving the properties of coronal pressure fronts in 3d : application to the 2012 may 17 ground level enhancement , astrophys . j. 833 ( 2016 ) 45 .
http://arxiv.org/abs/1605.05208 [ ] , http://dx.doi.org/10.3847/1538-4357/833/1/45 [ ] .
a. p. rouillard , et al . , interpreting the properties of solar energetic particle events by using combined imaging and modeling of interplanetary shocks , astrophys .
j. 735 ( 2011 ) 7 .
[ ] . a. p. rouillard , relating white light and in situ observations of coronal mass ejections : a review , journal of atmospheric and solar - terrestrial physics 73 ( 2011 ) 12011213 .
http://dx.doi.org/10.1016/j.jastp.2010.08.015 [ ] . c. salas - matamoros , k .-
klein , a. p. rouillard , coronal mass ejection - related particle acceleration regions during a simple eruptive event , a. and a. 590 ( 2016 ) a135 . http://dx.doi.org/10.1051/0004-6361/201528015 [ ] .
e. sanchez - diaz , et al . ,
observational evidence for the associated formation of blobs and raining inflows in the solar corona , astrophys .
j. lett . 835 ( 2017 )
http://arxiv.org/abs/1612.05487 [ ] , http://dx.doi.org/10.3847/2041-8213/835/1/l7 [ ] .
o. witasse , et al .
, interplanetary coronal mass ejection observed at stereo - a , mars , comet 67p / churyumov - gerasimenko , saturn , and new horizons en - route to pluto . comparison of its forbush decreases at 1.4 , 3,1 and 9.9 au , journal of geophys .
res . submitted . | the remoteness of the sun and the harsh conditions prevailing in the solar corona have so far limited the observational data used in the study of solar physics to remote - sensing observations taken either from the ground or from space .
in contrast , the ` solar wind laboratory ' is directly measured in situ by a fleet of spacecraft measuring the properties of the plasma and magnetic fields at specific points in space . since 2007
, the solar - terrestrial relations observatory ( stereo ) has been providing images of the solar wind that flows between the solar corona and spacecraft making in - situ measurements .
this has allowed scientists to directly connect processes imaged near the sun with the subsequent effects measured in the solar wind .
this new capability prompted the development of a series of tools and techniques to track heliospheric structures through space .
this article presents one of these tools , a web - based interface called the propagation tool that offers an integrated research environment to study the evolution of coronal and solar wind structures , such as coronal mass ejections ( cmes ) , corotating interaction regions ( cirs ) and solar energetic particles ( seps ) .
these structures can be propagated from the sun outwards to or alternatively inwards from planets and spacecraft situated in the inner and outer heliosphere . in this paper , we present the global architecture of the tool , discuss some of the assumptions made to simulate the evolution of the structures and show how the tool connects to different databases .
heliophysics , solar imaging , cmes , cirs |
the original motivation for this paper was to understand the results about a chaos decomposition for general lvy processes obtained by nualart and schoutens @xcite . a relation between multiple wiener integrals for the brownian motion and the symmetric fock space , and its role in the chaos decomposition ,
are well known . for more general lvy processes , a representation on a symmetric fock space was introduced by parthasarathy , see @xcite .
we will show that in this representation , the chaos decomposition has a very natural interpretation , and its proof becomes more immediate .
a recent preprint @xcite , which came to our attention when the paper was nearing completion , handles a number of related questions , although by quite different methods .
a number of preceding papers which deal with related topics are cited in the references .
our second goal was the extension of these results to certain non - commutative stochastic processes . here
the starting point is the @xmath0-fock space construction introduced by boejko and speicher @xcite .
they also introduced the @xmath0-brownian motion , which reduces to the usual brownian motion for the bosonic case @xmath1 , to the fermionic brownian motion for @xmath2 , and to the analog of the brownian motion in free probability for @xmath3 . in a previous paper @xcite , following the ideas of parthasarathy and schrmann @xcite
, we introduced lvy processes on the @xmath0-deformed fock spaces .
since we are able to express the chaos decomposition property in the fock space language , the same method gives the corresponding result for the @xmath0-deformed processes .
moreover , this decomposition can then be used to obtain a number of consequences .
combinatorial stochastic measures were introduced for the usual stochastic processes by rota and wallstrom @xcite . in this paper , we obtain an explicit formula for such measures for the @xmath0-lvy processes , unifying a number of previous results .
@xmath4th power of a process can be expressed as a sum over stochastic measures . on the level of operators ,
this turns out to be precisely the decomposition of the product of operators in terms of their wick products .
since the @xmath0-lvy processes consist of non - commuting operators , they generate non - commutative von neumann algebras , and any information about these algebras is of interest .
it was known that for the @xmath0-brownian motion , or for the free probability case @xmath3 , these algebras are ii@xmath5-factors . we show that in contrast , in all other cases these algebras have no normal tracial states , and thus are infinite algebras .
further information about these algebras awaits discovery .
there are three standard classes of infinitely divisible distributions ; see @xcite for one interpretation of these classes .
the intermediate one , with fourier transforms @xmath6\ ] ] for @xmath7 a finite measure , was considered by kolmogorov .
it consists of the distributions with mean @xmath8 and finite variance .
this is the class whose @xmath0-analogs are treated in this paper . for simplicity
we treat only one - dimensional such processes , although there is no real difficulty in extending the results to @xmath4 dimensions . the narrowest class , with fourier transforms @xmath9,\ ] ] for @xmath7 a finite measure , consists of the compound poisson distributions , considered by de finetti . in this case
there is a more natural construction in which the starting object is an algebra with a fixed state .
the analogs of some results of the paper for this class are treated in the appendix .
finally , all infinitely divisible distributions have fourier transforms of the form @xmath10\ ] ] for @xmath7 a finite measure , by the lvy - khinchine theorem .
representations of all of the corresponding lvy processes on the symmetric fock space are described in section 21 of @xcite .
there is no difficulty with extending the definitions to the general @xmath0 case , but since the usual treatment involves measures which are no longer finite , we do not treat this case in this paper . *
acknowledgments : * i thank ed effros , marius junge , and murad taqqu for a number of useful ( and enjoyable ) conversations .
this paper is a sequel to @xcite ; see that paper for all definitions and references not explicitly provided here .
let @xmath11 be a real hilbert space and @xmath12 its complexification .
let @xmath13 , for @xmath14 , be its `` @xmath15-particle space . ''
let its algebraic fock space be the vector space @xmath16 here @xmath17 is the generator of the @xmath8th component , traditionally called the vacuum vector .
define an inner product on @xmath18 by @xmath19 where @xmath20 is the inner product on @xmath12 .
fix @xmath21 .
define an operator @xmath22 on @xmath13 by @xmath23 where @xmath24 is the number of inversions of the permutation @xmath25 .
then all @xmath22 , and so the corresponding operator @xmath26 on the whole of @xmath18 , are strictly positive @xcite .
we define a new @xmath0-inner product on @xmath18 by @xmath27 this is a positive definite inner product . completing the algebraic fock space with respect to the corresponding norm
, we get the @xmath0-fock space @xmath28 . for @xmath29 ,
the inner product is only positive semi - definite . for @xmath1 ,
taking the quotient by its kernel gives the symmetric fock space ; for @xmath2 , it gives the anti - symmetric fock space .
for @xmath3 , we get the full fock space . for @xmath30 ,
define creation and annihilation operators on @xmath28 by @xmath31 here @xmath32 means `` omit @xmath33 '' .
these operators can be extended to bounded operators on the whole @xmath28 , so that @xmath34 and @xmath35 are adjoints of each other .
moreover , @xmath36 let @xmath37 be a bounded operator on @xmath11 .
we will also denote by @xmath37 its complexification , which is a bounded operator on @xmath12 .
define the gauge ( or preservation , or differential second quantization ) operator @xmath38 by @xmath39 by proposition 2.2 of @xcite , if @xmath37 is a self - adjoint operator , the operator @xmath38 is essentially self - adjoint with dense domain @xmath18 . the operator @xmath38 on the @xmath0-fock space is bounded .
denote by @xmath40 the norm of @xmath38 as an operator from @xmath28 to itself .
first we show that @xmath41 .
indeed , @xmath38 is a composition of two operators , @xmath42 followed by @xmath43 on @xmath44 , @xmath45 and @xmath46 from general tensor product considerations .
now we show that @xmath47 . for the remainder of the proof , we work with the @xmath8-inner product , the @xmath0-inner product being given by the positive definite density operator @xmath26 .
it follows from the proof of proposition 2.2 of @xcite that the adjoint of @xmath38 with respect to the @xmath0-inner product is @xmath48 .
this means that @xmath49 , where @xmath50 denotes the adjoint with respect to the @xmath8-inner product .
in particular , @xmath51 is a positive operator .
so from the inequality @xmath52 it follows that @xmath53 so @xmath54 since @xmath55 , we conclude that @xmath56 we conclude that for any pair @xmath57 , @xmath30 , the operator @xmath58 is a bounded operator on the @xmath0-fock space of @xmath12 , self - adjoint if @xmath37 is .
note that @xmath59 is linear as a function of the pair @xmath57 .
the following construction is inspired by @xcite .
let @xmath60 be an index set .
denote by @xmath61 the set of multi - indices ( finite sequences ) of elements of @xmath60 .
denote by @xmath62 the algebra of all real polynomials without a constant term , in a collection of non - commuting indeterminates @xmath63 .
it is an algebra without a unit , and even a star - algebra with the obvious involution .
let @xmath64 be a positive linear functional on @xmath62 ; equivalently , it is a conditionally positive linear functional on the full algebra @xmath65 .
assume that it is also both left- and right - bounded , in the sense that for all @xmath66 there exist constants @xmath67 such that @xmath68 \leq m_f \psi[g^\ast g ] ; \qquad \psi[f^\ast g^\ast g f ] \leq n_g \psi[f^\ast f].\ ] ] @xmath64 induces a positive semi - definite inner product on the space @xmath62 in the usual way , @xmath69 $ ] , as well as a semi - norm @xmath70 . taking a quotient by the subspace of seminorm - zero vectors and completing with respect to the induced norm , we obtain a real hilbert space @xmath11 with the induced inner product .
denote by @xmath71 the canonical mapping @xmath72 , let @xmath73 be its image , and for @xmath66 define the operator @xmath74 by @xmath75 .
put , for @xmath76 , @xmath77 , @xmath78 . more generally , for @xmath79 , denote @xmath80 , @xmath81 .
then each @xmath82 is essentially self - adjoint , with dense domain @xmath73 consisting of analytic vectors .
since @xmath64 is left - bounded , for each @xmath79 , @xmath83 , so each @xmath84 is bounded .
any family @xmath85 satisfying a certain compatibility condition arises in this way , see proposition 4.3 of @xcite and theorem 3 of @xcite .
let @xmath12 be the complexification of @xmath11 and @xmath86 the complexification of @xmath73 .
denote @xmath87 , and more generally @xmath88 for @xmath79
. these are bounded operators on @xmath28 .
note that the mapping @xmath89 is @xmath90-linear .
if @xmath91 , by definition @xmath92 .
moreover , since @xmath64 is right - bounded , @xmath93 as well . as a result
, @xmath94 depends only on the equivalence class of @xmath95 in @xmath12 , and so can be defined for @xmath96 .
denote by @xmath97 the complex algebra ( no closure ) generated by @xmath98 , and by @xmath99 the complex algebra generated by @xmath100 .
[ lemma : cyclic ] the vacuum vector @xmath17 is cyclic for @xmath99 . by definition , @xmath101 where @xmath102 .
it follows by induction that @xmath103 , which is dense in @xmath28 .
the result of the preceding lemma will be made more precise through the use of multiple stochastic integrals and kailath - segall polynomials .
define the maps @xmath104 , @xmath105 , inductively as follows : @xmath106 and @xmath107 by the discussion in the preceding subsection , @xmath108 is a multi - linear map which depends only on the equivalence classes of @xmath109 in @xmath12 .
so we can project it to a map on @xmath110 , and extend it to a @xmath111-linear map on @xmath112 . as a result
, we can define a linear _ wick map _ @xmath113 as such an extension @xmath114 with the extra condition @xmath115 .
clearly , @xmath116 and @xmath117 [ prop : separating ] the vacuum vector @xmath17 is separating for @xmath99 .
it is clear from the definition that @xmath118 is a polynomial in the @xmath119 operators ( and so lies in @xmath99 ) , and conversely that the mapping @xmath120 is onto .
in fact , @xmath121 where @xmath122 is a polynomial in the @xmath119 operators of degree at most @xmath123 . inverting this relation , @xmath124 where @xmath102 .
now let @xmath125 .
then @xmath126 for some @xmath127 .
if @xmath128 , then @xmath129 , and so @xmath130 .
[ remark : commutant ] the following ( possibly unbounded ) operators commute with @xmath99 on the dense domain @xmath131 : for @xmath79 , define the operator @xmath132 by @xmath133 explicitly , their values on the tensors of low order are @xmath134 where @xmath135\end{gathered}\ ] ] a set partition @xmath136 of a set @xmath137 is a collection of disjoint subsets of @xmath137 whose union equals @xmath137 .
let @xmath138 be the collection of all set partitions of the set @xmath139 .
let @xmath140 be a set partition , @xmath141 .
order the classes according to their first elements , that is , @xmath142 . for @xmath143 , call the pair @xmath144 an extended partition ; @xmath145 is to be thought of as the collection of classes `` open on the left '' .
see figure [ figure : extended ] for an example . for @xmath146 ,
define the restriction @xmath147 as follows : @xmath148 define the number of right restricted crossings of @xmath144 at the point @xmath4 as follows : @xmath149 let @xmath150 .
note that also ( see @xcite ) @xmath151 [ prop : product - wick ] let @xmath152 denote the single - element classes of a partition @xmath136 . then @xmath153 both sides of the expression are in @xmath99 .
evaluate them on @xmath17 .
we obtain @xmath154 for fixed @xmath144 , each term on the right - hand - side , possibly up to a power of @xmath0 , is a summand in the expansion of @xmath155 for a unique sequence of operators @xmath156 , where @xmath157 conversely , any sequence @xmath158 in the expansion of the left - hand - side of whose evaluation on @xmath17 is non - zero is of this form for some extended partition @xmath144 . it remains to show that its coefficient is exactly @xmath159 .
we show this by induction .
consider a word @xmath158 corresponding to an extended partition @xmath144 .
it is clear that @xmath160 is the element of @xmath161 corresponding to the restriction @xmath162 . by induction hypothesis
its coefficient is @xmath163 if @xmath164 , the coefficient in front of the tensor corresponding to @xmath165 is the same , and so is the number of restricted crossings of @xmath166 . on the other hand ,
suppose @xmath167 or @xmath168 .
suppose @xmath169 and @xmath170 .
then the degree of @xmath0 in the coefficient is incremented by @xmath171 but this is exactly the number of right restricted crossings of @xmath144 at @xmath172 .
let @xmath7 be a probability measure on @xmath90 with compact support , and in particular with finite moments @xmath173 ( note the shift in the index ) of all orders .
then the functional @xmath174 = r_k$ ] , @xmath175 = r_1 = 0 $ ] , @xmath176 = 0 $ ] on @xmath177 $ ] is conditionally positive definite and both left- and right - bounded . let @xmath178 . on the corresponding @xmath65 ,
define the functional @xmath179 = { \left\vert\bigcap_{i=1}^n i_i\right\vert } r_n.\ ] ] this functional is also conditionally positive and bounded .
the corresponding @xmath12 is naturally isomorphic to @xmath180 , where @xmath181 .
note that the hilbert space @xmath182 is isomorphic to @xmath183 via the map @xmath184 .
so @xmath185 denote @xmath186 .
note that @xmath187 corresponds to @xmath188 .
similarly , @xmath189 naturally corresponds to @xmath190 , where @xmath191 is the independent variable on @xmath192 .
finally , denote @xmath193 , @xmath194
. then all @xmath195 are @xmath0-lvy processes in the sense of @xcite .
we could have similarly defined multi - dimensional processes , but the notation gets heavier , while the phenomena are the same . note that @xmath97 is equal to the algebra generated by @xmath196 , and @xmath99 is equal to the algebra generated by @xmath197 . on @xmath97 , define the vacuum state @xmath198 } = { \left \langle \omega , a \omega \right \rangle}\ ] ] and an inner product @xmath199 } = { \left \langle a \omega , b \omega
\right \rangle}_q.\ ] ] the completion of @xmath97 with respect to this inner product will be denoted by @xmath200 .
[ lemma : diagonal ] for a subdivision @xmath201 of @xmath202 , denote @xmath203 .
define the diagonal measure @xmath204 of @xmath119 to be the limit @xmath205 then the limit exists in @xmath206 , the diagonal measures of @xmath119 are elements of @xmath200 , and @xmath207 in particular , the operators @xmath208 can in fact be identified with elements of @xmath200 .
see the appendix of @xcite .
note that @xmath209 , so the variance of @xmath119 at time one is @xmath210 . since the @xmath0-inner product is non - degenerate , @xmath211 with the usual and the @xmath0-inner product are isomorphic as vector spaces .
since @xmath212 , we can induce the @xmath0-inner product on this space and denote the resulting hilbert space by @xmath213 .
note also that @xmath214 can be naturally identified with a dense subset of @xmath215 .
let @xmath216 be centered @xmath0-lvy processes normalized to have variance @xmath210 at time @xmath210 .
let @xmath217 be an indicator function @xmath218 such that the intervals @xmath219 are disjoint .
define the multiple wiener - it stochastic integral of @xmath217 @xmath220 to be @xmath221 .
extend the definition to @xmath111-linear combinations of such functions in a linear way .
such an integral will frequently be denoted simply by @xmath222 .
let @xmath216 be as in the preceding definition .
the map @xmath223 extends to an isometry from @xmath224 into @xmath225 . here
@xmath226 is the algebra generated by @xmath227 .
let @xmath217 be a simple function , @xmath228 , such that the intervals @xmath229 are disjoint for each @xmath230 . by definition ,
@xmath231 let @xmath232 be a function of the same type . in the following expression
, @xmath233 denotes the pair partitions @xmath136 of @xmath234 such that for all @xmath235 , @xmath236 .
such a partition naturally induces a permutation @xmath237 by @xmath238 for @xmath239 .
it is easy to see that @xmath240 . @xmath241 } \\ & \quad = \sum a_{\vec{u } } b_{\vec{v } } \sum_{\pi \in \operatorname{\mathcal{p}}_2(k , k ) } q^{{\mathrm{rc } \left ( \pi \right ) } } \prod_{i=1}^k { \left\verti_{u(k+1-\min(b_i ) ) } \cap j_{v(\max(b_i)-k)}\right\vert } \\ & \quad = \sum a_{\vec{u } } b_{\vec{v } } \sum_{\sigma \in \operatorname{sym}(n ) } q^{i(\sigma ) } \prod_{i=1}^k { \left\verti_{u(i ) } \cap j_{v(\sigma(i))}\right\vert } \\ & \quad = { \left \langle f , g \right \rangle}_q.\end{aligned}\ ] ] since such simple functions are dense in @xmath224 , the stochastic integral map can be isometrically extended to the whole of this hilbert space . in particular ,
define the full stochastic measure @xmath242 if all @xmath243 s are equal to some @xmath119 , we will omit it from the notation , and write simply @xmath244 .
if all @xmath245 s are equal to @xmath246 , we write @xmath247 .
if the variances of the integrator processes are not normalized to @xmath210 , the multiple stochastic integral map differs from an isometry by a constant factor , and is again well - defined .
let @xmath119 be a @xmath0-lvy process .
let @xmath248 be the gram - schmidt orthogonalization of @xmath249 in @xmath250 .
note that @xmath251 } = t r_{k+j}$ ] and @xmath252 , where @xmath7 is a ( @xmath0-)canonical measure for this process .
thus the coefficients of @xmath253 in the expansion of @xmath254 are precisely those of @xmath255 in the orthogonal polynomials @xmath256 with respect to the measure @xmath257 .
equivalently , @xmath258 corresponds to @xmath259 .
[ prop : orthogonal ] for a multi - index @xmath230 , denote @xmath260 then these subspaces are orthogonal for different @xmath230 .
it suffices to prove the result for simple @xmath217 . in that case
the argument is similar to the proof of the preceding proposition .
it suffices to note that @xmath261 } = 0 $ ] for @xmath262 .
[ defn : st ] let @xmath201 be a subdivision of @xmath202 . for @xmath140 , define @xmath263^n_\pi } x(i_{u(1 ) } ) x(i_{u(2 ) } ) \ldots x(i_{u(n)}),\ ] ] where @xmath264^n_\pi = { \left\{\vec{u } \in { \left\{1 , \ldots , n\right\}}^n : u(i ) = u(j ) \leftrightarrow i \stackrel{\pi}{\sim } j\right\}}.\ ] ] define the partition - dependent stochastic measures @xmath265 if the limit exists .
also define @xmath266 . note that this notation differs from the one in @xcite .
in particular , @xmath267 and @xmath268 [ prop : st_pi ] partition - dependent stochastic measures of a @xmath0-lvy process are well - defined as limits in @xmath200 , and equal to @xmath269 let @xmath270 be a subdivision of the interval @xmath202 .
we will show that @xmath271 as @xmath272 .
since the @xmath273-limit of the second term of exists and equals the right - hand - side of , this will also show that the limit of the left - hand - side exists .
moreover , since the first term of is in @xmath97 , all the quantities involved are in @xmath200 since @xmath274 and the intervals in @xmath270 are disjoint , @xmath275^n_{\pi } } \sum_{\substack{\sigma \in \operatorname{\mathcal{p}}(n ) \\
\sigma \leq \pi } } \sum_{s \subset \sigma } a_{(s , \sigma ) } \prod_{b \not \in s } \bigl ( { \left\vert\bigcap_{i \in b } i_{u(i)}\right\vert } r_{{\left\vertb\right\vert } } \bigr ) \bigotimes_{b \in s } \bigl ( { \mathbf{1}_{\bigcap_{i \in b } i_{u(i ) } } } \otimes x^{{\left\vertb\right\vert } - 1 } \bigr),\ ] ] where the sum is over all refinements of @xmath136 , and each @xmath276 is a power of @xmath0 . since @xmath277^n_{\pi}$ ] and @xmath278 ,
we may write @xmath279 for any @xmath280 , so the preceding expression is equal to @xmath281^n_{\pi } } \sum_{\substack{\sigma \in \operatorname{\mathcal{p}}(n ) \\
\sigma \leq \pi } } \sum_{s \subset \sigma } a_{(s , \sigma ) } \prod_{b \not \in s } \bigl ( { \left\verti_{u(b)}\right\vert } r_{{\left\vertb\right\vert } } \bigr ) \bigotimes_{b \in s } \bigl ( { \mathbf{1}_{i_{u(b ) } } } \otimes x^{{\left\vertb\right\vert } - 1 } \bigr).\ ] ] also , @xmath282^{{\left\verts\right\vert}}_{\hat{0 } } } \bigotimes_{j=1}^{{\left\verts\right\vert } } \bigl ( { \mathbf{1}_{i_{v(j ) } } } \otimes x^{{\left\vertb_j\right\vert } - 1 } \bigr).\end{gathered}\ ] ] as @xmath272 , each term in equation for @xmath283 converges to @xmath284 this is also the limit of the corresponding term in equation , provided that we show @xmath285 .
this follows via an argument similar to the one in proposition [ prop : product - wick ] . on the other hand ,
the norm of the vector in is bounded by @xmath286^n_{\pi } } \sum_{\substack{\sigma \in \operatorname{\mathcal{p}}(n ) \\
\sigma \leq \pi } } \prod_{b \in \sigma } { \left\verti_{u(b)}\right\vert},\ ] ] where @xmath287 is a constant depending on @xmath288 but independent of @xmath289
. for each @xmath290 , some @xmath291 are in the same class of @xmath136 . as a result ,
@xmath292 for all @xmath277^n_{\pi}$ ] .
we may assume @xmath293
. then @xmath294^n_{\pi } } \prod_{b \in \sigma } { \left\verti_{u(b)}\right\vert } \leq \sum_{\vec{u } \in [ n]^n_{\pi } } { \left\verti_{u(c_1)}\right\vert } \prod_{b \in \pi } { \left\verti_{u(b)}\right\vert } \leq \delta({\mathcal{i } } ) \sum_{\vec{u } \in [ n]^n_{\pi } } \prod_{b \in \pi } { \left\verti_{u(b)}\right\vert } \leq \delta({\mathcal{i } } ) t^{{\left\vert\pi\right\vert}},\ ] ] which converges to @xmath8 as @xmath272 .
[ ex : classical ] processes with independent increments correspond to @xmath1 . in this case
the formula takes the form @xmath295 processes with freely independent increments correspond to @xmath3 . in this case
the formula takes the form @xmath296 for @xmath297 , and @xmath8 otherwise . here
@xmath298 are all the non - crossing partitions , and @xmath299 , @xmath300 are the inner , respectively , outer classes of @xmath136 .
this result is the main theorem of @xcite ( with a weaker mode of convergence ) .
finally , in the @xmath0-gaussian case , the formula takes the form @xmath301 for @xmath302 , and @xmath8 otherwise .
here @xmath303 , @xmath152 are @xmath304 , respectively , @xmath210-element classes of @xmath136 .
note that in this case , @xmath305 the singleton depth of @xmath136 , @xmath306 .
thus we recover proposition 6.12 of @xcite ( with a weaker mode of convergence ) . for closely related results ,
see @xcite . in the centered @xmath0-charlier case , @xmath307 as a consequence , @xmath308 here @xmath309 are the continuous big @xmath0-hermite polynomials , see for example @xcite .
[ lem : dense ] the image of the multiple stochastic integral map with respect to the processes @xmath310 contains @xmath99 . 1 .
[ part : poly ] polynomials in @xmath119 can be expressed as multiple stochastic integrals : @xmath311 2 .
[ part : cyclic ] for @xmath312 , @xmath313 the first part of the lemma is a basic , purely combinatorial , property of partition - dependent stochastic measures due to @xcite . for the second part
, we observe that the maps @xmath314 and @xmath315 are isometries .
therefore it suffices to show the property for simple functions @xmath316 where all @xmath317 are disjoint . but in this case @xmath318 [ cor : wick - integrals ] in particular , @xmath319 the vacuum vector @xmath17 is cyclic and separating for @xmath97 .
the representation @xmath315 of @xmath97 is faithful , and extends to an isomorphism between @xmath200 and @xmath28 .
the wick map extends to the inverse isomorphism .
we only need to show that @xmath17 is cyclic for @xmath97 . by proposition [ prop : st_pi ] , for a fixed @xmath136 @xmath320 so by induction , @xmath321 therefore , using the same proposition , each @xmath322 where this expression is defined in the obvious way .
since , similarly to lemma [ lem : dense ] , @xmath323 we conclude that @xmath324 .
but @xmath17 is cyclic for this algebra .
the following proposition is an analog of a result of @xcite .
any @xmath325 has a unique chaos decomposition @xmath326 where @xmath327 and @xmath328 conversely , any such series converges to an element of @xmath200 . by proposition [ prop : orthogonal ]
, it suffices to show that @xmath329 is dense in @xmath200 . using the isomorphism between @xmath200 and @xmath28 ,
this is guaranteed by lemma [ lem : dense ] .
[ remark : classical ] a few modifications are necessary for the classical case @xmath1 . from the point of view of @xmath0-deformations of the full fock space , this case is degenerate , since it involves the reduction to the symmetric fock space .
an easy way to modify the preceding arguments for this context is to work , instead of functions in @xmath330 , with square - integrable functions with support in the simplex @xmath331 in this case the von neumann algebras are commutative , and as a result instead of working with bounded self - adjoint operators we can work with essentially self - adjoint operators with an invariant dense domain consisting of analytic vectors . for the measure @xmath7
, this corresponds to dropping the requirement of compact support and instead requiring that it has a finite moment generating function , @xmath332 for @xmath333 small enough . note that this is exactly the hypothesis of @xcite
moreover , the vacuum vector @xmath17 is cyclic and separating for the von neumann algebra ( and not just the algebra ) generated by @xmath334 . with these modification ,
all the preceding statements about isometries and orthogonality remain true , and the result of @xcite follows . in this case
all the operators @xmath334 commute and are independent with respect to the expectation @xmath335 , and the corresponding convolution semigroup is given by @xmath336 note also that by combining lemma [ lem : dense ] with proposition [ prop : st_pi ] ( more specifically , with example [ ex : classical ] ) , we get an explicit formula @xmath337 it is to be compared with the results of section 3.1 of @xcite .
note also a number of previous results in this direction , such as @xcite .
[ remark : free ] in the free case @xmath3 , the von neumann algebras are no longer commutative , but the vacuum expectation is tracial ( which is not the case in general : see the next section ) . as a result
, we can again work with unbounded operators , and weaken the hypothesis to a measure with a finite moment - generating function . using the operators from remark [ remark : commutant ]
, we can again show that the vacuum vector is separating for the von neumann algebra of @xmath334 .
the processes have freely independent increments , and the corresponding free convolution semigroup is given by @xmath338 where @xmath339 is the @xmath340-transform of @xmath341 .
[ cor : step1 ] let @xmath342 .
then @xmath343 let @xmath344 be ( possibly non - commuting ) indeterminates .
for @xmath345 , define the polynomial @xmath346 of total degree @xmath347 in the variables @xmath348 by the recursion @xmath349 with initial conditions @xmath350 , @xmath351 .
these polynomials have apparently not been considered explicitly before ; they are in some weak sense analogs of the appell polynomials . because of @xcite , it is appropriate to call them kailath - segall polynomials . from equation and corollary [ cor : wick - integrals ] , @xmath352 and similarly ,
@xmath353 are polynomials in @xmath354 .
the following proposition is closely related to the results of @xcite in the classical case .
we use the notation @xmath355_q = 0 $ ] , @xmath356_q = \sum_{i=0}^{n-1 } q^k$ ] , @xmath355_q !
= 1 $ ] , @xmath356_q ! = \prod_{i=1}^n [ i]_q$ ] . . 1
denote @xmath357 .
for a polynomial of total degree @xmath358 , @xmath359_q!}{[n - k]_q ! } ( x_{j+k } + r_{j+k } ) a^{(n - k)}.\ ] ] 2 . since @xmath360 , @xmath361_q!}{[n - k]_q ! } ( x_{k+1 } + r_{k+1 } ) a^{(n - k)}.\ ] ] 3 . in particular , @xmath362_q!}{[n - k]_q !
} \delta_{k+1}(t ) \psi_{n - k}(t).\ ] ] for the first part , denote the polynomial @xmath363 of degree @xmath358 by @xmath364 .
the result follows by induction from the recursion : @xmath365 , and @xmath366_q r_{j+1 } a^{(n-1 ) } - [ n]_q a_{j+1}^{(n-1 ) } \\ & = x_{j } a^{(n ) } - [ n]_q r_{j+1 } a^{(n-1 ) } \\ & \quad - [ n]_q \left ( x_{j+1 } a^{(n-1 ) } + \sum_{k=0}^{n-1 } ( -1)^k \frac{[n-1]_q!}{[n-1-k]_q ! } ( x_{j+k+1 } + r_{j+k+1 } ) a^{(n - k-1 ) } \right ) \\ & = x_{j } a^{(n ) } + \sum_{k=1}^n ( -1)^k \frac{[n]_q!}{[n - k]_q ! } ( x_{j+k } + r_{j+k } ) a^{(n - k)}.\end{aligned}\ ] ] the second part follows from the first one , and the third one is an application of corollary [ cor : wick - integrals ] .
it can also be obtained using corollary [ cor : step1 ] . in the @xmath0-gaussian case , @xmath367 , @xmath368 for @xmath369 .
let @xmath370 , @xmath371 , @xmath372 for @xmath369 . then @xmath373_q a^{(n-1)}(x).\ ] ] so @xmath374 are the continuous ( rogers ) @xmath0-hermite polynomials .
in the @xmath0-poisson case , @xmath375 for @xmath376 .
let @xmath377 for @xmath378 .
then @xmath373_q a^{(n-1)}(x ) - [ n]_q a^{(n)}(x).\ ] ] so @xmath374 are the centered continuous big @xmath0-hermite polynomials , which in this context are @xmath0-analogs of the charlier polynomials .
see @xcite for further results in this direction .
let @xmath379 be the von neumann algebra generated by @xmath380 .
let @xmath381 be the orthogonal projection from @xmath382 onto @xmath383 .
let @xmath384 denote also by @xmath381 the induced projections from @xmath12 to @xmath385 and from @xmath18 to @xmath386 .
it extends to an orthogonal projection from @xmath28 to @xmath387 .
note that @xmath379 is also generated as a von neumann algebra by @xmath388 . denote @xmath389= p_t a p_t$ ] .
then @xmath390 is a norm- and strongly continuous projection . since is clearly maps @xmath391 onto @xmath388
, it maps @xmath99 onto @xmath392 and @xmath393 onto @xmath379 . as a result
, @xmath390 is a conditional expectation from @xmath393 to @xmath379 , which preserves the vacuum state . in particular , for @xmath394 , @xmath395 , @xmath396 = a_t e_t[z ] b_t$ ] .
a map @xmath397 will be called a process .
a simple process is piecewise constant and zero at infinity , @xmath398 , where without loss of generality all the intervals @xmath399 are disjoint .
a simple adapted process is a simple process with each @xmath400 .
denote by @xmath401 the completion of the set of simple adapted processes with respect to the norm coming from the inner product @xmath402 and call the elements of this completion adapted processes .
an simple algebraic process is @xmath398 with all @xmath403 , and simple adapted algebraic processes and adapted algebraic processes are defined similarly .
let @xmath119 be a @xmath0-lvy process and @xmath404 a simple adapted process , @xmath398 .
define the left and right it stochastic integrals @xmath405 and @xmath406 note that for all the arguments below , it is not necessary that the @xmath0-lvy process @xmath334 generate the filtration @xmath407 , but only that it be a martingale with respect to it , in other words that @xmath408 if @xmath409 . in particular , for processes adapted with respect to @xmath379 , all the integrals with respect to @xmath249 , @xmath410 are defined .
the stochastic integral map in the preceding definition is @xmath411 times an isometry from @xmath401 to @xmath412 .
therefore it can be extended to all adapted processes .
we consider the right integrals .
for two simple adapted processes , @xmath413 since @xmath414 are adapted , @xmath415 if , say , @xmath416 , this is @xmath8 . on the other hand , if @xmath417 , then the inner product is @xmath418 so for two simple adapted processes , @xmath419 the result follows .
for the left integrals , the proof is similar .
denote by @xmath420 the unitary second quantization of the operator @xmath421 , determined by @xmath422 it is a completely positive contraction .
the role of this operator has been emphasized by donati - martin in @xcite .
let @xmath423 .
define @xmath424 where @xmath425 is a subdivision of @xmath426 .
this limit exists in @xmath206 .
the definition extends linearly to simple adapted algebraic processes so that @xmath427 first we prove formula for an elementary process .
let @xmath404 be simple , @xmath428 , with all the vectors @xmath429 .
let @xmath270 be a subdivision of @xmath430 .
then by definition , @xmath431 note that for @xmath432 , @xmath433 therefore representing the right - hand - side of equation on @xmath17 , we obtain @xmath434 \\
& \quad= \sum_{i=1}^n ( { \mathbf{1}_{i_i } } \otimes 1 ) \otimes \eta_1 \otimes \eta_2 \otimes \ldots \otimes \eta_k \otimes ( { \mathbf{1}_{i_i } } \otimes 1 ) \\ & \quad\quad+ ( v - u ) r_2
q^k \eta_1 \otimes \eta_2 \otimes \ldots \otimes \eta_k + q^k ( { \mathbf{1}_{[u , v ) } } \otimes x ) \otimes \eta_1 \otimes \eta_2 \otimes \ldots \otimes \eta_k ] .\end{aligned}\ ] ] since the sum in the last term converges to @xmath8 , it follows that @xmath435 by linearity , the same result holds for simple algebraic processes .
the algebra @xmath393 has no normal tracial states except , possibly , the vacuum state .
suppose @xmath436 is a tracial state on @xmath393 . by the previous proposition , for @xmath437 @xmath438 = q^k \tau[\delta_2([u , v ) ) w(\eta_1 \otimes \eta_2 \otimes \ldots \otimes \eta_k)].\ ] ] @xmath436 is normal , so strongly continuous , so @xmath206 continuous . using this and the trace property
, the expression above is also equal to @xmath438 = \tau[\delta_2([u , v ) ) w(\eta_1 \otimes \eta_2 \otimes \ldots \otimes \eta_k)].\ ] ] therefore @xmath439 = 0 $ ] for @xmath440 . but note that the limit @xmath441 thus @xmath442 = 0 $ ] for @xmath440 . the following dichotomy holds . 1 . for the @xmath0-brownian motion , or for @xmath3 , the von neumann algebra is a @xmath443-factor .
[ part : non - bm ] for @xmath444 and all other @xmath0-lvy processes , the von neumann algebra has not normal tracial states , and thus is an infinite algebra .
only part is new .
let @xmath445 .
note that @xmath446 } = q^2 r_2 r_{2+k } { \left\verti\right\vert } \cdot { \left\vertj\right\vert},\ ] ] while @xmath447 } = q r_2 r_{2+k } { \left\verti\right\vert } \cdot { \left\vertj\right\vert}.\ ] ] thus if the vacuum state is tracial and @xmath444 , @xmath368 for @xmath369 .
this characterizes the @xmath0-brownian motion .
it is not known if these von neumann algebras are always factors .
using an idea of donati - martin , we can define two - sided stochastic integrals .
for @xmath448 simple algebraic processes , a simple algebraic bi - process is @xmath449 .
@xmath404 is adapted if each of @xmath450 is .
denote @xmath451}.\ ] ] this is a positive sesquilinear form ( cf .
@xcite , although our notation is slightly different ) .
denote the closure of the space of all adapted simple algebraic bi - processes with respect to the corresponding ( semi)norm by @xmath452 , and call its elements adapted algebraic bi - processes .
for a simple algebraic adapted bi - process @xmath449 , with @xmath453 , @xmath454 , define the stochastic integral of @xmath404 to be the operator @xmath455 the stochastic integral map is an isometry from @xmath452 to @xmath456 .
the proof is the same as in the @xmath0-brownian motion case , and relies on the key for @xmath457 and @xmath458 , @xmath459 = ( t - s ) \gamma_q(q)(z).\ ] ] let @xmath460 , with @xmath461
. then @xmath459 = p_s x([s , t ) ) z x([s , t ) ) p_s = p_s a([s , t ) ) z a^\ast([s , t ) ) p_s.\ ] ] evaluating this expression on @xmath17 , we get @xmath462 since the vacuum vector is separating for @xmath99 and the conditional expectation maps this algebra into itself , the result follows .
let @xmath463 be a unital star - algebra , given in a faithful star - representation ( by bounded operators ) on a hilbert space @xmath464 with a cyclic , separating vector @xmath17 . denote by @xmath465 the state on @xmath463 . construct @xmath161 and @xmath466 as in the beginning of section [ sec : q - fock ] , and let @xmath467 } = { \left \langle \omega , x
\omega \right \rangle}$ ] be the vacuum expectation on @xmath468 . note that now , unlike in the body of the paper , we identify @xmath469 with the vacuum vector in @xmath466
. we will also identify each @xmath470 with the corresponding vector @xmath471 . for @xmath1 ,
if @xmath341 is the distribution of @xmath94 with respect to @xmath335 and @xmath7 is the distribution of @xmath95 with respect to @xmath475 , then @xmath476 thus @xmath94 has a compound poisson distribution .
in particular , if @xmath95 is a projection , @xmath94 has a poisson distribution . for general @xmath0 ( see @xcite ) , @xmath479 } = \sum_{\pi \in \operatorname{\mathcal{p}}(n ) } q^{{\mathrm{rc } \left ( \pi \right ) } } \prod_{b \in \pi } { \left\langle f^{{\left\vertb\right\vert } } \right\rangle}.\ ] ] let @xmath480 be the algebra generated by @xmath481 , with the obvious involution . as in lemma [ lemma : cyclic ] and
proposition [ prop : separating ] , @xmath17 is a cyclic and separating vector for @xmath480 , so the vacuum state is faithful . define the wick product @xmath482 then @xmath483 , @xmath484 since both states @xmath335 and @xmath475 are faithful , @xmath487 thus @xmath488 the distribution of @xmath489 is the standard @xmath0-poisson distribution ( the orthogonality measure of the continuous big @xmath0-hermite polynomials ) . the distribution , and in particular its support , are known explicitly , see @xcite and their references .
its maximum is the indicated constant .
the following are some analogs of formulas of the main body of the paper for this context , mostly given without proof .
the kailath - segall formula takes the form @xmath490_q!}{[n - k]_q ! } x(f^{k+1 } ) w \left(f^{\otimes ( n - k ) } \right ) - { \left\langle f \right\rangle } w \left(f^{\otimes n } \right).\ ] ] evaluate both sides of the expression on @xmath17 .
we obtain @xmath492 each term on the right - hand - side , possibly up to a power of @xmath0 , is obtained by applying a unique sequence of operators @xmath493 to @xmath17 , where @xmath494 the rest of the proof proceeds as in proposition [ prop : product - wick ] . in particular , since @xmath495 } = 0 $ ] , @xmath496 } = \sum_{\pi \in \operatorname{\mathcal{p}}(n ) } q^{{\mathrm{rc } \left ( \pi \right ) } } \prod_{b \in \pi } { \left\langle \prod_{i \in b } f_i \right\rangle},\ ] ] a generalization of equation .
see @xcite for further results in this direction .
now let @xmath497 where @xmath7 is a compactly supported probability measure .
then polynomials are contained in and are dense in @xmath183 .
denote @xmath498 and @xmath499 note that we no longer have a shift in the index . as in lemma
[ lemma : diagonal ] , @xmath500 exists in @xmath501 , where @xmath201 is a subdivision of @xmath202 .
consequently , @xmath17 is cyclic and separating for the algebra generated by@xmath502 .
streater , _ fock space decomposition of levy processes _ , levy processes , theory and applications ( aarhus university , jan 2002 ) , miscellanea , vol .
22 , maphysto , 2002 , ` arxiv : math.pr/0307243 ` , pp . | we continue the investigation of the lvy processes on a @xmath0-deformed full fock space started in @xcite .
first , we show that the vacuum vector is cyclic and separating for the algebra generated by such a process .
next , we describe a chaotic representation property for it in terms of multiple integrals with respect to diagonal measures , in the style of nualart and schoutens .
we define stochastic integration with respect to these processes , and calculate their combinatorial stochastic measures .
finally , we show that they generate infinite von neumann algebras . |
the standard theory of cosmological structure formation tells us that rich galaxy clusters form through mergers and absorption of smaller galaxy clusters and groups .
cluster major mergers are the largest energy release events in the universe where the energy of @xmath8 erg is involved .
mergers have great impacts on cluster evolution .
indeed , numerical simulations @xcite show that mergers cause shocks and turbulence in the intracluster medium ( icm ) .
while a large part of the dissipated energy in the shocks and turbulence is transformed into the thermal icm , some are most likely transported into a non - thermal form such as cosmic - rays and magnetic fields @xcite .
in fact , some merging clusters have diffuse non - thermal synchrotron radio emissions , which is direct evidence of the existence of the cosmic - ray electrons whose energy is @xmath9 gev and the magnetic fields of @xmath10 g in the icm . such diffuse non - thermal radio sources are classified into two types owing to their morphology and location .
radio halos have similar morphology as the icm x - ray emission and are located in the central part of the cluster @xcite .
on the other hand , radio relics usually have an arc - like shape and are located in the periphery of the cluster @xcite . the recent status of this field is extensively reviewed in @xcite and @xcite .
because of their morphology and locations , it has been believed that radio relics trace outgoing shocks seen in the later phases of major mergers .
recently , in fact , discontinuities of the icm physical quantities such as temperature and density are found in the outer edge of some radio relics , which is direct evidence of the association of relics with shocks .
mach numbers of shocks related to the relics can be observationally estimated in two independent ways . from the synchrotron radio spectral index , we are able to obtain the cosmic - ray electron energy spectral index @xcite , which can be related with the shock mach numbers assuming a simple diffusive shock acceleration ( dsa ) theory @xcite . on the other hand ,
x - ray observations of the icm are able to determine directly the mach number through the detection of discontinuities of the icm physical quantities such as temperature and density with ranking - hugoniot jump conditions @xcite .
if the simple diffusive shock acceleration model is valid , both methods have to give us consistent results .
thus , comparison of the high quality x - ray data with excellent radio spectra is useful to get insights into the particle acceleration processes .
performed the first systematic study about this issue , though the sample size was not large enough to draw definite conclusions .
therefore , it is important to observe more relics in both radio and x - rays in order to reveal their origin .
the electron population attributed to the synchrotron radiation also emits non - thermal x - rays through the inverse compton processes of cosmic microwave background ( cmb ) photons .
comparison between synchrotron and inverse compton fluxes provides us with the magnetic field strength .
even if only the upper limit of the inverse compton flux is obtained , we are able to get the lower limit of the field strength . although a lot of efforts have been made to search for such non - thermal x - rays from cluster radio halos and relics @xcite , this is still challenging and any firm detections have not been reported unfortunately .
1rxs j0603.3 + 4214 is a merging cluster with radio relics and a halo at @xmath11 .
its north radio relic is well - known for a peculiar linear shape and nicknamed by `` toothbrush '' @xcite .
the radio spectrum at the outer edge of the relic is rather flat and its spectral index is @xmath12 , which means that the shock mach number is @xmath13 .
this value is rather high for cluster merger shocks .
an xmm - newton observation reveals that it has an elongated north - south morphology with two subcluster cores @xcite , which strongly suggests that this cluster is in the later phase of a major merger . from the x - ray surface brightness distribution analysis
, it is claimed that the shock is offsetted by 1 arc - minute outwards from the outer edge of the relic and extended in the west of the relic .
the mach number is estimated to be @xmath14 from the density modeling with a discontinuity , which is significantly lower than the value derived from the radio data . however , note that their results could be affected by the three dimensional density structure modeling and projection effects .
we conducted a suzaku x - ray observation of the field around the `` toothbrush '' relic .
the x - ray imaging spectrometer ( xis ) aboard the suzaku satellite @xcite is more appropriate for observing low surface brightness diffuse sources than chandra and xmm - newton , thanks to its low and stable background @xcite .
this feature is very useful to investigate the physical properties of the icm around cluster radio relics , which are generally located in the outerpart of the cluster .
indeed , icm temperature in the pre - shock region around radio relics can be determined practically only by suzaku , considering that the icm emission fill the field of view and in the hard band . in this paper
, we present a suzaku x - ray observation of galaxy cluster 1rxs j0603.3 + 4214 to investigate the physical status and particle acceleration processes at `` toothbrush '' radio relic .
the rest of this paper is organized as follows . in section 2
we describe the observation and data reduction . in section 3
we present spectral analysis results . in section 4
we discuss the results and their implications . in section 5
we summarize the results .
canonical cosmological parameters of @xmath15 mpc@xmath5 km s@xmath5 , @xmath16 , and @xmath17 are used in this paper . at the redshift of the cluster , 1
arcmin corresponds to 217 kpc .
the solar abundances are normalized to @xcite .
unless otherwise stated , all uncertainties are given at the 90% confidence level .
we observed a field around `` toothbrush '' radio relic in galaxy cluster 1rxs j0603.3 + 4214 with suzaku on 2012 october 7 - 10 .
in addition , a field offsetted by 1 degree in galactic longitude was observed on 2012 october 14 - 15 to estimate the background components .
the observations were performed at xis nominal pointing .
the xis was operated in the normal full - frame clocking mode .
the edit mode of the data format was @xmath18 and @xmath19 , and combined data of both modes were utilized .
the spaced - row charge injection was adopted for xis .
all data were processed with suzaku pipeline processing , version 2.8.16.34 .
the calibration data files ( 20130305 ) were adopted .
a summary of the observations is given in table [ tab : obs ] .
[ cols="^,^,^,^",options="header " , ] with the highest temperature ) , where the north and south directions are positive and negative , respectively , width=302 ] in this subsection , we investigate temperature structure around the candidate shock regions in more detail .
the spectral model used in the fitting is the same as in subsection [ ss : tempcolaxis ] .
figure [ fig : region ] shows regions used in the analysis .
we searched for point sources with xmm - newton data , whose spatial resolution is better than suzaku . to reduce contamination and
cxb systematic errors , @xmath20 radius circular regions centered by a position of a point source whose flux is more than @xmath21 erg @xmath4 s@xmath5 are excluded and shown in figure [ fig : region ] by green circles .
it is expected that an outgoing shock front exits at the outer edge of the radio relic .
we chose the regions as @xmath22 boxes , shown in red in figure [ fig : region ] , except for the outermost region , where a @xmath23 box is adopted to avoid the region illuminated by the calibration sources , in order to investigate temperature profile across the expected shock , assuming that the shock is located at the relic s outer edge .
a temperature gradient along the relic s long axis is suggested by @xcite .
region r3 is divided into two @xmath24 boxes as in figure [ fig : region ] to investigate how this affects the mach number estimation .
@xcite also suggests that the shock is elongated towards the west of the relic .
cyan annuli in figure [ fig : region ] are regions to investigate this expected west shock
. systematic errors of both cxb and nxb were estimated in a similar way in @xcite .
the cxb spatial fluctuations can be modeled as @xmath25 , where @xmath26 and @xmath27 are the effective solid angle and upper cutoff flux of a point source , respectively . from the heao - i
a2 results , @xcite reported that @xmath28 adopting an upper cutoff flux of @xmath29 erg @xmath4 s@xmath5 and the values of the regions used in the spectral fit for @xmath26 , we calculated the cxb fluctuations at the 90 % confidence level , which are shown in table [ tab : relic ] and [ tab : west ] for each region . on the other hand , it is reported that the reproducibility of nxb was 6.0 % and 12.5 % for xis fi and bi at the 90 % confidence level , respectively @xcite .
systematic errors of lhb and mwh were not considered because the icm temperature is sufficiently higher than those of the galactic background components and it is most likely that the effects of those errors are very limited .
figure [ fig : rpre - spec ] and [ fig : rpost - spec ] represent the xis spectra of the pre - shock ( r2 ) and post - shock ( r3 ) regions for the relic shock analysis , fitted with the above - mentioned model , respectively . even in the pre - shock region where the surface brightness is relatively low , the icm component is clearly detected over the cxb thanks to the low and stable nxb of the xis .
the fitting results are summarized in table [ tab : relic ] , where the first , second , and third errors are statistical , cxb systematic , and nxb systematic , respectively . even in the pre - shock region , where the icm emission is relatively faint , the systematic errors are not much more than the statistical ones . the temperature profile across the relic is shown in figure [ fig : relic_temp ] , where horizontal axis represents the angular distance from the relic outer edge . inward and
outward directions from the relic outer edge is positive and negative , respectively .
black crosses shows the results from region r1 , r2 , r3 , r4 , and r5 .
light gray crosses show the results when the region r3 ( post - shock ) is divided into the east and west .
only statistical errors are displayed .
the positions of the inner and outer edges of the relic are displayed by dark gray dotted lines .
a temperature drop is surely seen at the outer edge of the relic , and we also confirmed the temperature gradient in the relic region along the relic s longer axis . , but for the post - shock ( r3 ) region.,width=302 ] lcccc region number & @xmath30 ( kev ) & abundance & @xmath31 & @xmath32 ( % ) + r1 & @xmath33 & @xmath34(fixed ) & @xmath35 & 38 + r2 ( pre - shock ) & @xmath36 & @xmath34(fixed ) & @xmath37 & 33 + r3 ( post - shock ) & @xmath38 & @xmath34(fixed ) & @xmath39 & 35 + r4 & @xmath40 & @xmath41 & @xmath42 & 38 + r5 & @xmath43 & @xmath44 & @xmath45 & 32 + r3east & @xmath46 & @xmath34(fixed ) & @xmath47 & 59 + r3west & @xmath48 & @xmath34(fixed ) & @xmath49 & 41 + + [ tab : relic ] we performed similar analysis for the west shock regions . again , the icm component is clearly detected over cxb even in the pre - shock ( w1 ) region .
the fitting results are summarized in table [ tab : west ] , and temperature profile is shown in figure [ fig : west_temp ] .
again , we see a clear temperature drop across the expected shock front . lcccc region number & @xmath30 ( kev ) & abundance & @xmath31 & @xmath32 ( % ) + w1 ( pre - shock ) & @xmath50 & @xmath51 ( fixed ) & @xmath52 & 30 + w2 ( post - shock ) & @xmath53 & @xmath51 ( fixed ) & @xmath54 & 34 + + [ tab : west ] ) . only statistical errors are displayed . ,
width=302 ] it is expected that the non - thermal x - rays are emitted through the inverse compton scattering of the cmb photons by the same electron population attributed to the radio relic . to constrain this component , the spectra in the radio relic region ( r3 in figure [ fig : temp_plot ] )
are fitted by the following model ; @xmath55 \end{aligned}\ ] ] where @xmath56 represents the inverse compton component .
again , all the parameters in the background components were fixed to the values obtained from the background field results .
we fixed @xmath57 to be unity for xis1 and allowed it to vary freely for xis0 and xis3 .
the other parameters except @xmath57 are common among all xis sensors .
the temperature of the icm component was fixed to the value ( 6.10 kev ) obtained in the relic shock region analysis because the expected inverse compton component is much weaker than the thermal icm one .
we fixed the photon index of the inverse compton component to be 2.1 considering that @xcite reported that the integrated radio spectral index of the relic is @xmath58 .
systematic errors of the cxb and nxb were taken into account in the same way as in the candidate shock region analysis .
the fitting results are summarized in table [ tab : ic ] . because the lower limit of the normalization is consistent with zero ( taking into account of 3 @xmath59 errors ) ,
we conclude that the inverse compton component is not detected .
considering both the statistical and systematic errors , we derive an upper limit of the inverse compton component in 0.3 - 10 kev at the 90 % confidence level . the resultant upper limit on the flux
is @xmath3 erg s@xmath5 @xmath4 in a @xmath22 rectangular area .
the spectral fit in the rectangular region is done by excluding point source regions . to calculate the upper limit , we assume that the x - ray emission is uniformly distributed within this rectangular region so we can directly compare with the radio flux .
we tried to fit the data allowing the photon index of the @xmath56 component to vary in order to confirm the component is really consistent with the inverse compton interpretation .
however , the fit did not work well and we obtained no useful results .
ccc components & @xmath30 ( kev ) or @xmath60 & normalization + @xmath61 & @xmath62 ( fixed ) & @xmath63 + @xmath56 & @xmath64 ( fixed ) & @xmath65 + & @xmath31 & @xmath66 + + [ tab : ic ]
we compare the temperature decline in the outer part of the 1rxs j0603.3 + 4214 with that of typical relaxed clusters .
first , we compare our results with a scaled temperature profile proposed by @xcite for dynamically relaxed clusters , which is based on cosmological n - body + hydrodynamic simulation results . to get the scaled profile , the temperature and radius are normalized by the mean temperature and virial radius , respectively .
we adopt the mean temperature @xmath67 kev @xcite .
the virial radius is calculated from the following equation @xcite , @xmath68 where @xmath69 is the hubble constant normalized by the value @xmath70 mpc@xmath5 km s@xmath5 and @xmath71^{1/2}$ ] .
the resultant virial radius is @xmath72 mpc .
we applied these values to the results presented in table [ tab : temp ] in the north of the main x - ray peak , which is shown in figure [ fig : temp+burns ] by crosses .
on the other hand , the scaled temperature profile outside @xmath73 proposed by @xcite is , @xmath74^{\alpha},\ ] ] where the best fit parameters are @xmath75 , @xmath76 , @xmath77 .
this fitting function is also plotted in figure [ fig : temp+burns ] by dotted lines .
considering that the 1rxs j0603 shows clear signatures of a major merger , it is not so surprising that the observed profile does not seems to match with the fitting function .
the obtained temperatures are higher than the values of the fitting function around @xmath78 , which might be because of the heating by the merger shock , disturbed shape of the gravitational potential , and/or disturbed entropy distribution . , which might be because of the heating by the merger shock , disturbed shape of the gravitational potential , and/or disturbed entropy distribution . , width=302 ] next , we compare our results with a universal temperature profile obtained by @xcite from suzaku x - ray and subaru weak - lensing observations of four relaxed galaxy clusters . the proposed profile is @xmath79^{-(\frac{2}{5}\delta - \frac{2}{5}\gamma + \frac{3}{5 } \alpha)/\beta},\ ] ] where @xmath80 , @xmath81 , @xmath82 , @xmath83 , and @xmath84 .
the normalization factor of the temperature is @xmath85^{2/3},\ ] ] where @xmath86 is the total mass inside @xmath87 .
though both @xmath86 and @xmath87 should be determined through direct mass measurements with gravitational lensing technique ideally , we do not have available lensing data for this cluster .
thus , we utilize @xmath87 derived from equation ( [ eq : virial ] ) and calculate @xmath86 simply assuming that the mean density inside @xmath87 is 200 times of the critical density of the universe . as a result , @xmath88 and @xmath89 kev .
figure [ fig : temp+okabe ] shows a comparison of our results ( crosses ) with the profile of @xcite ( dotted lines ) .
it is clear that the measured temperatures are systematically higher than the universal profile in all radii .
this suggests that the normalization determination does not work very well , which is not so surprising considering that we did not use gravitational lensing data .
the central part and outermost data could be marginally consistent with the universal profile taking into account of the errors of @xmath90 .
note that the our data shows a convex profile whereas the universal profile is concave around @xmath78 . .
note that the our data shows a convex profile whereas the universal profile is concave around @xmath78 .
, width=302 ] here , we derive the mach number from the temperature difference of the shock candidate region using the a rankine - hugoniot relation as follows , @xmath91 where the @xmath92 and @xmath93 denote the temperature in the pre - shock and post - shock regions , respectively , and we assume that the specific heat ratio @xmath94 . applying the equation ( [ rankine - hugoniot ] ) to the region r2 and r3 of the relic shock results , we obtain the @xmath95 , where the first , second , and third errors are statistical , cxb systematic , and nxb systematic at the 90 % confidence level , respectively .
this means that @xmath96 at the 1 @xmath59 confidence level . on the other hand ,
the radio spectral index results indicate @xmath97 @xcite on the assumption of a simple dsa theory .
thus , our results show clear discrepancy between them even including both statistical and systematic errors , which strongly suggests that a simple dsa theory does not hold at least for this object and that there should be other parameters that control the shock acceleration .
figure [ fig : mach ] shows mach numbers derived from the radio spectral index ( @xmath98 ) plotted against those from the x - ray temperature measurements ( @xmath99 ) , which is an updated version of figure 8 in .
in addition to the results of toothbrush relic ( a black cross ) , those of ciza2242 north and south @xcite , a2256 @xcite , coma @xcite , a3667 north west @xcite and south east , and a3376 @xcite are displayed .
the error bars of @xmath99 is only statistical and at the one sigma level .
the toothbrush relic ( a black cross ) seems to be a rather extreme example . )
plotted against those from the x - ray temperature measurements ( @xmath99 ) , which is an updated version of figure 8 in .
the gray dotted lines represent those of @xmath100 .
the error bars of @xmath99 is only statistical and at the one sigma level .
the toothbrush relic ( a black cross ) seems to be a rather extreme example.,width=302 ] a temperature gradient across the relic long axis was suggested by @xcite and also confirmed in our data as shown in figure [ fig : relic_temp ] .
this means that the temperature of the region r3 could be underestimated , which lead the underestimation of the mach number . to estimate this effect , we derived the mach number using the region r2 and r3west temperatures .
the resultant mach number becomes @xmath101 .
therefore , this effect is quite limited and the discrepancy between the radio and our results still remains .
we searched for a corresponding temperature gradient in the pre - shock region , dividing r2 region into the east and west and performing the spectral analysis as in subsection [ ss : an_ca_sh_re ] .
however , no significant temperature difference is found because of the poor statistics .
considering suzaku s point spread function ( psf ) , whose half power diameter is @xmath102 , the temperature difference between the neighboring regions could be somewhat underestimated .
especially , the temperature of the fainter and cooler pre - shock region would be overestimated because of the contamination from the brighter and hotter post - shock region . to check this
, we measured the temperature of the region @xmath20 ahead of the relic outer edge .
the obtained results is @xmath103 kev .
this is @xmath9 1.2 kev lower than the pre - shock temperature .
this temperature decline is partly because of global temperature decline trend shown in figure [ fig : temp_plot ] , partly because of less contamination from the post - shock region . in other word
, the overestimation of pre - shock region temperature due to psf should be @xmath9 1 kev at most , which could be comparable to the statistical and cxb systematic errors at most .
if the icm components that are not directly associated with the shock but have different temperature is accidentally located along the line - of - sight , we would have incorrect temperature difference . to investigate this possibility
, we fit the spectra with a two - temperature icm model , but the fitting results did not improve significantly .
at least , there is no compelling evidence to introduce two - temperature icm model .
also , this unexpected temperature difference could be a result of complicated dynamical history of the cluster .
indeed , both a peculiar linear shape of the `` toothbrush '' and the temperature gradient along the longer axis of the relic strongly suggest that this cluster does not undergo a simple bimodal merger .
@xcite proposed a triple merger scenario , where interaction of multiple shocks generate a linear shape shock front like `` toothbrush '' relic . in a re - acceleration scenario @xcite , where the electrons in the relic have already accelerated once somewhere , the spectrum has a memory of the past acceleration history . in other words ,
the present shock mach number alone can not determine the relic spectrum .
if the past acceleration process is relevant to the higher mach number shocks such as virial shocks @xcite , relic shock mach number would be lower than what is expected from radio observations .
a non - linear acceleration model @xcite , where the shock structures are modified by the interactions between the thermal icm and accelerated non - thermal particles , is another possibility to explain this kind of discrepancy .
the relation between the radio spectrum index and mach number is not simple and will depend on the details of the modeling .
a similar analysis is performed also for the western shock region and the obtained mach number becomes @xmath104 at the 90 % confidence level , which is consistent with the xmm - newton results ( @xmath105 at the @xmath106 level confidence ) based on the x - ray surface brightness analysis @xcite . in general , the temperature measurement is less seriously affected by the line - of - sight structures .
note that some kind of three dimensional modeling is necessary to get the information of the density structure from the x - ray surface brightness distribution . on the other hand
, our result could be affected by suzaku s moderate spatial resolution as mentioned above .
again , it is probable that a temperature difference and mach number are underestimated to a certain extent .
nevertheless , we obtained results consistent with @xcite in an independent way .
we constrain the magnetic field strength from the comparison of the radio synchrotron flux and an inverse compton x - ray upper limit @xcite .
the following procedure is basically the same as in @xcite .
the typical energy , @xmath107 , of photons scattered through inverse compton processes by electrons with energy @xmath108 is @xmath109 , where @xmath110 is the photon energy before scattering . with the typical cmb photon energy at @xmath11 ( @xmath111 ) , the range of the electron s relativistic lorentz factor corresponding to the 0.3 - 10 kev inverse compton x - rays
is @xmath112 . on the other hand , the synchrotron critical frequency of emitted by electrons with magnetic field strength @xmath113
is @xmath114 assuming a homogeneous pitch angle distribution .
thus , the typical synchrotron frequency range emitted by the electrons in the energy range mentioned above is @xmath115 .
@xcite reported that the radio flux of the toothbrush relic is 319.5 mjy at 1.382 ghz and that the spectral index is @xmath116 .
therefore , the radio flux corresponding to the energy range of 0.3 - 10 kev inverse compton x - ray is @xmath117 erg s@xmath5 @xmath4 , with a monochromatic approximation for a single electron s spectrum and on the assumption of a single power - law electron energy distribution .
it is well - known that the flux of the inverse compton scattering of cmb photons and synchrotron radiation from the same electron population has the following relation : @xmath118 where @xmath119 and @xmath120 are the energy density of the cmb photons and magnetic field , respectively . with @xmath121 erg @xmath122 at @xmath11 , above - mentioned @xmath123 , and
obtained upper limit of @xmath124 , the lower limit of the magnetic field strength in the relic becomes @xmath125 .
this is consistent with the equipartition magnetic field energy , @xmath126 , @xcite .
note that our results are free from the assumption of energy equipartition between cosmic - ray electrons and magnetic field .
the upper limit on the inverse compton flux might be sensitive to the temperature determination of the thermal icm component .
we checked how the temperature changes of the icm component affect the lower limit on the inverse compton flux in the spectral analysis . as a result , a 1 kev increase and decrease of the icm temperature cause a 24 % increase and 27 % decrease of the upper limit flux , respectively . these also result in an 12 % decrease and 12 % increase of the lower limit on magnetic field strength .
it seems odd at first glance that the temperature increase ( or decrease ) of the icm causes increase ( or decrease ) of the inverse compton component .
the best fit values and statistical errors of the inverse compton component normalization are less sensitive to the icm temperature change . on the other hand ,
the systematic errors are more sensitive and becomes larger ( or smaller ) when the icm temperature becomes higher ( or lower ) , which is the main factor governing the resultant upper limit flux . in any cases
, magnetic fields of @xmath10 g level is still inferred .
we also investigated how changes in the radio spectral index affect the results .
the 0.1 increase and decrease of the absolute value of the radio spectral ( and x - ray photon ) index cause a 31 % and 62 % decrease of the upper limit flux , respectively .
these cause a 50 % increase and 19 % decrease of the lower limit on the magnetic field strength , respectively .
note that the energy range of electrons attributed to the observed synchrotron radiation is most likely higher than that to the inverse compton component in our analysis . while the range of electron s lorentz factor attributed to 100 mhz - 1 ghz
synchrotron radiation is @xmath127 , that to the 0.3 - 10.0 kev inverse compton component is @xmath128 as discussed in the above .
fixing the normalization in the radio observation range , therefore , increase ( or decrease ) of the absolute value of the radio spectral index causes increase ( or decrease ) of synchrotron flux corresponding to the inverse compton component , which results in increase ( or decrease ) of the lower limit on the magnetic field strength .
generally thought , the above results all indicate a magnetic field at the @xmath129 level
. however , this might not be the case if the radio spectrum is not a single power - law because we rely on the extrapolation of the synchrotron spectrum .
lower frequency radio and higher energy x - ray observations are crucial in this regard .
lofar and astro - h will make a significant contribution to resolve this problem .
the energy densities of the thermal icm , non - thermal electrons , and magnetic field in the radio relic region can be estimated from our results , which are basic and crucial parameters to investigate the origin and physical states of the radio relic . for simplicity ,
the radio relic region is assumed to be a cylinder whose radius and height are 868 kpc and 434 kpc , respectively . from the normalization of the apec model in the spectral analysis ,
the electron number density of the thermal icm in the relic region is estimated to be @xmath130 @xmath122 . with this value and the temperature of this region ( @xmath131 kev ) ,
the energy density of the thermal icm is @xmath132 erg @xmath122 , assuming that the mean molecular weight is 0.6 . with the lower limit of the magnetic field strength ( @xmath125 ) ,
the energy density of the magnetic field is @xmath133 erg @xmath122 . as a result , @xmath134 , which means that the magnetic energy could be more than a few % of the thermal one and that the icm evolution and structures could be somewhat affected by the magnetic field . from the upper limit of the inverse compton component ,
the energy density of the relativistic electrons corresponding to 0.3 - 10 kev x - ray band ( or , @xmath128 ) becomes @xmath135 erg @xmath4 .
this indicates @xmath136 , though the contribution from lower energy electrons is not included in this calculation , which could be dominant in the energy density of the non - thermal electron populations .
we observed the field around the `` toothbrush '' radio relic in the galaxy cluster 1rxs j0603.3 + 4214 with suzaku . from the xis analysis
, we found that the mach number at the outer edge of the relic derived from the temperature difference is significantly lower than the value derived from the radio spectral index assuming a simple dsa theory @xcite .
this suggests that a simple dsa theory does not hold at least for this object .
the re - acceleration scenario could explain this discrepancy .
alternatively , interactions of multiple shocks owing to complicated dynamical history would resolve both this problem and the peculiar linear shape of the relic .
we also confirmed that the mach number of the possible shock , which is elongated towards the west of the relic , is consistent with the value obtained from xmm - newton data @xcite in an independent way .
we searched for the non - thermal inverse compton component in the relic region and set an upper limit on the flux of @xmath3 erg @xmath4 s@xmath5 . comparing the synchrotron radio flux with this upper limit , we obtained a lower limit on the magnetic field strength of @xmath137 , which means that the magnetic energy could be more than a few % of the thermal energy .
the authors would like to thank t. akahori , k. sato , s. shibata , and h. ohno for helpful comments .
we are also grateful to the suzaku operation team for their support in planning and executing this observation .
mt is supported in part by japan society for the promotion of science ( jsps ) kakenhi grant number 26400218 .
ha is supported by a grant - in - aid for jsps fellows ( 22 - 606 ) .
sron is supported financially by nwo , the netherlands organization for scientific research .
rjw is supported by nasa through einstein postdoctoral grant number pf2 - 130104 awarded by the chandra x - ray center , which is operated by the smithsonian astrophysical observatory for nasa under contract nas8 - 03060 .
99 ajello , m. et al .
2009 , , 690 , 367 ajello , m. , rebusco , p. , cappelluti , n. , reimer , o. , bhringer , h. , la parola , v. , & cusumano , g. 2010 , , 725 , 1688 akahori , t. , & yoshikawa , k. 2010 , , 62 , 335 akamatsu , h. , de plaa , j. , kaastra , j. , ishisaki , y. , ohashi , t. , kawaharada , m. , & nakazawa , k. 2012 , , 64 , 49 akamatsu , h. , takizawa , m. , nakazawa , k. , fukazawa , y. , ishisaki , y. , & ohashi , t. 2012 , , 64 , 67 akamatsu , h. , & kawahara , h. 2013 , , 65 , 16 akamatsu , h. , inoue , s. , sato , t. , matsusita , k. , ishisaki , y. , & sarazin , c. l. 2013 , , 65 , 89 akamatsu , h. et al .
2015 , accepted for publication in ( arxiv:1507.02285 ) anders , e. , & grevesse , n. 1989 , geochim .
acta , 53 , 197 blandford , r. , & eichler , d. 1987 , physics reports , 154 , 1 bonafede , a. , giovannini , g. , feretti , l. , govoni , f. , & murgia , m. 2009 , , 494 , 429 bourdin , h. , mazzotta , p. , markevitch , m. , giacintucci , s. , & brunetti , g. 2013 , , 764 , 82 brggen , m. , van weeren , r. j. , & rttgering , h. j. a. 2012 . , 425 , l76 brunetti , g. , setti , g. , feretti , l. , & giovannini , g. 2001 , , 320 , 365 brunetti , g. , & jones , t. w. 2014 , international journal of modern physics d , 23 , 1430007 burns , j. o. , skillman , s. w. , & oshea , b. w. 2010 , , 721 , 1105 dickey , j. , m. , & lockman , f. , j. 1990 , 28 , 215 donnert , j. , dolag , k. , brunetti , g. , & cassano , r. 2013 , , 429 , 3564 drury , l. o. 1983 , reports on progress in physics , 46 , 973 finoguenov , a. , sarazin , c. l. , nakazawa , k. , wik , d. r. , & clarke , t. e. 2010 , apj , 715 , 1143 feretti , l. , giovannini , g. , & bhringer , h. 1997 , new astron .
, 2 , 501 feretti , l. , giovannini , g. , govoni , f. , & murgia , m. 2012 , 20 , 54 govoni , f. , markevitch , m. , vikhlinin , a. , van speybroeck , l. , feretti , l. , & giovannini , g. 2004 , , 605 , 695 henry , j. p. , evrard , a. e. , hoekstra , h. , babul , a. , & mahdavi , a. 2009 , , 691 , 1307 hindson , l , et al .
2014 , , 445 , 330 ishisaki , y. et al .
2007 , , 59 , s113 kale , r , dwarakanath , k. s. , bagchi , j. , & paul , s. 2012 , , 426 , 1204 kawano , n. et al .
2009 , , 61s , 377 koyama , k. , et al .
2007 , , 59 , s23 kushino , a. , ishisaki , y. , morita , u. , yamasaki , n. y. , ishida , m. , ohashi , t. , & ueda , y. 2002 , , 54 , 327 landau , l. d. , & lifshitz , e. m. 1959 , fluid mechanics ( oxford : pergamon press ) liang , h. , hunstead , r. w. , birkinshaw , m. , & andreani , p. 2000 , , 544 , 686 macario , g. , markevitch , m. , giacintucci , s. , brunetti , g. , venturi , t. , & murray , s. s. 2011 , , 728 , 82 malkov , m. a. , & drury , l. oc .
, 2001 , reports on progress in physics , 64 , 429 mitsuda , k. , et al .
2007 , , 59 , s1 nakazawa , k. et al .
2009 , , 61 , 339 ogrean , g. a. , brggen , m. , van weeren , r. j. , rttgering , h. , croston , j. h. , hoeft , m. 2013 , , 433 , 8120 ohno , h. , takizawa , m. , & shibata , s. 2002 , , 577 , 658 okabe , n. , et al .
2014 , , 66 , 99 ricker , p. , m. , & sarazin , c. , l. , 2001 , apj,561,621 .
rttgering , h. j. a. , wieringa , m. h. , hunstead , r. w. , & ekers , r. d. 1997 , , 290 , 577 ryu , d. , kang , h. , hallman , e. , & jones , t. w. 2003 , , 593 , 599 rybicki , g. b. , & lightman , a. p. 1979
, radiative processes in astrophysics ( new york : john wiley & sons ) sekiya , n. , yamasaki , n. y. , mitsuda , k. , & takei , y. 2014 , , 66 , l3 shafer , r. , a. 1983 , phd .
thesis , university maryland bibitem[shimwell et al.(2015)]shim15 shimwell , t. w. , markevitch , m. , brown , s. , feretti , l. , gaensler , b. m. , johnston - hollitt , m. , lage , c. , & srinivasan , r. 2015 , , 449 , 1486 shu , f. , h. 1992 , the physics of astrophysics .
volume ii : gas dynamics ( mill valley : university science books ) stroe , a. , harwood , j. j. , hardcastle , m. j. , rttgering , h. j. a. 2014 , , 445 , 1213 sugawara , c. , takizawa , m. , & nakazawa , k. , 2009 , , 61 , 1293 takizawa , m. 2005 , apj , 629 , 791 takizawa , m .
2008 , apj , 687 , 951 takizawa , m. , nagino , r. , & matsushita , k. 2010 , , 62 , 951 tawa , n. et al .
2008 , pasj , 60 , 11 thierbach , m. , klein , u. , & wielebinski , r. 2003 , , 397 , 53 trasatti , m. , akamatsu , h. , lovisari , l. , klein , u. , bonafede , a. , brggen , m. , dallacasa , d. , & clarke , t. 2015 , , 575 , 45 van weeren , r. j. , rttgering , h. j. a. , brggen , m. , & hoeft , m. 2010 , science , 330 , 347 van weeren , r. j. , brggen , m. , rttgering , h. j. a. , hoeft , m. , nuza , s. e. , & intema , h. t. 2011 , , 533 , 35 van weeren , r. j. , rttgering , h. j. a. , intema , h. t. , rudnick , l. , brggen , m. , hoeft , m. , oonk , j. b. r. 2012 , , 546 , a124 vazza , f. , brunetti , g. , kritsuk , a. , wagner , r. , gheller , c. , & norman , m. 2009 , , 504 , 33 watanabe , e. , takizawa , m. , nakazawa , k. , okabe , n. , kawaharada , m. , babul , a. , finoguenov , a. , smith , g. p. , & taylor , j. e. 2011 , , 63 , 357 wik , d. r. , et al .
2011 , , 727 , 119 yoshino , t. , et al .
2009 , , 61 , 805 zuhone , j. a. , markevitch , m. , & lee , d. 2011 , , 743 , 16 | we present the results of suzaku observations of the galaxy cluster 1rxs j0603.3 + 4214 with `` toothbrush '' radio relic .
although a shock with mach number @xmath0 is expected at the outer edge of the relic from the radio observation , our temperature measurements of the intracluster medium indicate a weaker temperature difference than what is expected .
the mach number estimated from the temperature difference at the outer edge of the relic is @xmath1 , which is significantly lower than the value estimated from the radio data even considering both statistical and systematic errors .
this suggests that a diffusive shock acceleration theory in the linear test particle regime , which is commonly used to link the radio spectral index to the mach number , is invalid for this relic .
we also measured the temperature difference across the western part of the relic , where a shock with @xmath2 is suggested from the x - ray surface brightness analysis of the xmm - newton data , and obtained consistent results in an independent way .
we searched for the non - thermal inverse compton component in the relic region and the resultant upper limit on the flux is @xmath3 erg @xmath4 s@xmath5 in the 0.3 - 10 kev band .
the lower limit of the magnetic field strength becomes 1.6 @xmath6 g , which means that magnetic energy density could be more than a few @xmath7 of the thermal energy . |
quasars represent episodes of rapid supermassive black hole ( smbh ) growth and probably a unique period in the early evolution of galaxies .
they may directly follow a major merger @xcite or a big blowout of gas and dust .
however , the nature of the relationship between smbh growth and galaxy formation is not well understood .
feedback from quasar outflows may play an important role in the evolution of this relationship .
we are using narrow absorption lines ( nals ) in quasar spectra to study quasar outflows and environments across a range of scales .
nals have full widths at half minimum ( fwhms ) less than several hundred km s@xmath1 , and they appear in a variety of ultraviolet ( uv ) resonance transitions , including c iv @xmath01548 and @xmath01551 , n v @xmath01239 and @xmath01243 , c iii @xmath0977 , si iv @xmath01394 and @xmath01402 and ly @xmath9 @xmath01216 , ly@xmath10 @xmath01026 , ly@xmath11 @xmath0973 , ly@xmath12 @xmath0950 , and ly@xmath13 @xmath0938 @xcite .
the first step in using these absorption lines to study the quasar environment is to determine simply where the lines form .
there are several possibilities , including unrelated ( intervening ) clouds and galaxies which happen to lie in the line - of - sight , nearby cluster galaxies and intrinsic clouds within the quasar host galaxy and its extended environment @xcite .
the statistical excess of c iv nals at velocity shifts v @xmath14 km s@xmath1 , where negative v indicates motion towards the observer , indicates that many of these absorbers are directly related to quasar environments @xcite .
the excess is largest at v @xmath15 km s@xmath1 , where roughly 80% of c iv systems with rest equivalent width rew(1548 ) @xmath16 have a quasar related origin @xcite . this intrinsic gas can form in quasar driven outflows , starburst driven outflows , merger remnants or ambient gas in the host halos , or in other galaxy halos in the same galaxy cluster as the quasar . at higher velocities , the excess can be attributed directly to quasar driven outflows . in the velocity range -1000 to -12,000 km s@xmath1 ,
@xcite estimate that @xmath17 43% of c iv nals with rew(1548 ) @xmath18 0.3 originate in a quasar outflow .
nals encompass a wealth of information about the basic properties of quasar outflows , which we use to gain insights into the outflow physics , acceleration mechanisms and geometry of the near quasar environment .
nals represent a very different type of quasar outflow compared to the well - studied much broader and higher velocity broad absorption lines ( bals ) , but they might simply be different manifestations of a single outflow phenomenon viewed at different angles @xcite .
the nals that do not form in outflows probe the gaseous environments of quasars more generally , and can be used to examine conditions in different host galaxies environments such as starburst driven outflows or larger - scale gas distributions produced by messy mergers .
we are involved in a program to study the location , origin and abundance information for absorbers in a sample of high redshift quasars .
we are particularly interested in high redshift quasars because z @xmath19 is the cosmic era when host galaxies are thought to grow rapidly and form most of their stars , possibly through merger events @xcite .
choosing redshifts above z @xmath20 2.73 also allows us to measure lines at shorter rest - frame wavelengths with ground based telescopes , including importantly , the h i lyman series , in order to obtain more and better constraints on the absorber ionisations , column densities and metal abundance .
we are interested in using abundances to discern rough star formation histories of quasar host galaxies in order to make inferences about the relationship between the quasar , the growth of the central black hole , and the evolution of the host galaxy .
broad emission lines ( bels ) have been used most often to study quasar abundances .
the most reliable results suggest metallicities of at least solar , and up to a few times solar , which requires significant previous star formation in the host @xcite . the metal - rich bel result is true even for
the highest redshifts studied , e.g. @xcite , with redshifts out to 6.4 .
the most reliable results based on bal column densities suggest metallicity ranges between solar and ten times solar @xcite .
previous studies of nals in low redshift samples have found super - solar metallicities and highly ionised gas , and have successfully probed several other nal outflow characteristics @xcite .
these lower redshift samples cover a wide range of luminosities , observed in the ultraviolet ( uv ) spectral range , where many useful metal and hydrogen lines occur .
nals offer certain advantages in the study of metallicities and other gas characteristics in the near - quasar environment .
their narrow widths mean the c iv doublets , separated by 500 km s@xmath1 , are resolved .
we use resolved absorption line doublets to disentangle saturation effects , and to obtain accurate line optical depth and column density measurements .
nals also form in a range of physical locations , providing a more complete picture of the regions near quasars . because the nal methods are completely independent from the bel methods , requiring only column densities and ionisation corrections , the nal metallicities provide an independent test of the bel results . here
we present results for the particular luminous quasar j102325.31 + 514251.0 ( hereafter j1023 + 5142 ) at a redshift of @xmath21 , which is during the peak in quasar activity and the epoch of rapid galaxy formation .
this quasar contains a rich complex of nine distinct c iv nal systems at velocities from -1400 to -6200 km s@xmath1 . the density and diversity of lines in this complex merits special attention .
we will argue below that most ( or all ) of these systems form in a highly structured quasar
driven outflow . to interprete the metallicities and other data provided by these nals , we examine several diagnostics that can identify intrinsic nals that form in quasar
driven outflows @xcite . in particular , 1 ) variability studies have found intrinsic absorbers varying on relatively short timescales of months to years , providing strong evidence for these absorbers belonging to outflows either crossing the line of sight to the quasar or experiencing changing ionisation with the variations in the continuum emission @xcite .
2 ) detection of partial coverage of the background light source along the line of sight strongly implies gas forming very near the source .
this phenomenon occurs when the absorbing clouds are smaller than the background source , allowing part of the light from the source to reach the observer unabsorbed .
this partial covering is easily detected in multiplets like the c iv doublet where the optical depth ratio between the two lines is fixed by the oscillator strengths .
when the source is partially covered , some light fills in the bottom of the absorption line , and makes the apparent optical depth ratio appear different than the real optical depth ratio .
3 ) outflow lines tend to have profiles that are broad and smooth compared to thermal widths @xcite . in well studied nals , these three indicators ( variability , partial covering and broad profiles ) tend to appear together , which further increases the probability that the occurrence of an individual indicator accurately predicts an outflow origin very near the quasar for a given absorption line ( see also @xcite and references therein ) .
we also note that super
solar metallicities are consistent with an intrinsic origin for the gas .
there are examples of high metallicity gas in intervening systems , but not of low
metallicity intrinsic gas @xcite .
we describe the data acquisition and reduction in [ sec : obs ] , the identification and fitting of the absorption lines in [ subsec : id ] and [ subsec : fit ] and the abundance and ionisation analysis in [ sec : ion ] .
we briefly describe individual absorption line systems in [ sec : sys ] .
we discuss the arguments for the locations , probable intrinsic origins and quasar driven outflow properties of the gas in
[ sec : disc ] and conclude with a summary in [ sec : sum ] .
we observed the quasar j1023 + 5142 on march 29 , 2007 with the keck i hiresr echelle spectrograph as part of an observing campaign to measure spectra of high redshift quasars with known narrow associated absorption lines .
we used an 0.86 wide slit for a spectral resolution of r @xmath22 or velocity resolution of @xmath23 km s@xmath1 .
our data span the wavelength range from 3700 to 8100 corresponding to 830 to 1820 in the quasar rest frame .
this spectral range covers a variety of interesting lines , including rest - frame ly @xmath24 , c iv 1548 , 1551 and several other lines in the h i lyman series down to the lyman limit at 912 .
we use four exposures totalling 2 hours on the source .
the spectral region from @xmath25 to @xmath26 is well - covered with considerable overlap between echelle orders at some wavelengths , but above 6540 there are small 15 - 40 gaps between orders ( one such gap is apparent in figure [ fig : civ ] ) , and two larger gaps at 4980 - 5070 and 6575 - 6670 where the spectrum falls into a physical gap between detectors .
we reduce all the data using the makee hires data reduction package .
the spectra are sky background - subtracted and extracted from the 2d frame with a low order polynomial trace using a white dwarf standard observed the previous night with similar seeing conditions .
we use a thorium argon ( thar ) lamp spectrum for wavelength calibration .
the resulting spectra are on a vacuum and heliocentric wavelength scale .
the spectra are not absolute flux calibrated .
we normalise the spectra to unity by fitting a pseudo - continuum to all of the quasar emission , including the emission lines .
the pseudo - continuum is defined as follows : for regions with few absorption lines , we apply a polynomial fit to the local continuum in each echelle order .
we accomplish the continuum normalisation in crowded regions where the continuum is affected by significant absorption , e.g. the ly @xmath9 forest , by first averaging together several adjacent spectral orders into a single spectrum .
then , we visually inspect the region for small sections of continuum not affected by absorption or obvious noise spikes , and interpolate between these sections , fitting the entire region with a low order polynomial .
our continuum fit for a region of the ly @xmath9 forest containing the ly@xmath10 and o vi nals is shown in figure [ fig : contin ] .
the continuum placement has an uncertainty of @xmath27% in the forest and 2 - 3% at other wavelengths .
forest spectrum with the continuum fit over - plotted .
the regions spanning the ly @xmath10 and o vi nals are labelled above the spectrum .
, width=302 ]
the broad , flat shape of the emission features in the spectrum of j1023 + 5142 make an accurate emission redshift difficult to determine . the redshift provided by the sdss spectrum is @xmath21 .
we estimate the reliability of this value by measuring the redshifts of the c iv @xmath01549 , c iii ] @xmath01909 and si iv+o iv ] @xmath01398 emission lines using measurements of their centroids .
we shift each centroid respectively by -824 , -730 and + 36 km s@xmath1 to correct for known offsets from the nominal quasar redshift ( [ o iii ] @xmath05007 emission ) , based on measurements by @xcite and @xcite for average quasars , where negative values are blueshifts .
the average redshift obtained from these emission lines is @xmath28 , which is offset from the sdss value by @xmath29 or @xmath30 km s@xmath1 .
these results and our efforts to measure the inherently uncertain emission line centroids suggest that the uncertainty of the redshift measured by sdss is not more than @xmath31 , corresponding to @xmath32 km s@xmath1 .
we adopt the sdss value throughout the remainder of this paper .
we identify nine distinct c iv absorption line systems within 6200 km s@xmath1 of the quasar redshift .
we will refer to these as systems 19 , as indicated in figure [ fig : civ ] and table [ tab : lines ] below .
other c iv systems are present at -16,800 and -33,800 km s@xmath1 in the spectrum , but they have narrow widths , complete covering , and blending problems in the ly @xmath9 forest , which , along with their high velocities , make them likely candidates for intervening gas and exclude them from further analysis in this work . after identifying the c iv doublets , we search the spectrum for other common nals such as si iv , n v , c iii , o vi , and h i lyman series lines at the same redshift
. we also search for lower ionisation species , such as c ii and si ii , but find none .
all of the systems , except possibly system 1 , appear to have relatively high ionisations based on the presence and absence of high and low ionisation species respectively .
each set of absorption lines at one redshift is considered a system , as labelled in figure [ fig : civ ] .
several of these systems are blends of two or three components , which are not individually labelled in the figure .
systems 1 and 2 ( figures [ fig:1 ] and [ fig:2 ] ) appear to be line - locked in c iv .
the velocity offset between the @xmath01548 line in system 1 and the @xmath01551 line in system 2 is remarkably small ( @xmath33 km s@xmath1 ) compared to the fwhms of these lines ( @xmath34 km s@xmath1 ) and the velocity shifts from the quasar systemic , @xmath35 and @xmath36 km s@xmath1 .
if this overlap between the c iv lines in systems 1 and 2 represents a physical line - lock , where the velocities of the two systems are actually separated by exactly their doublet separation , and not a chance alignment in the spectrum ( see @xcite and @xcite for full discussions of the possible physical nature of line - locking ) , then it provides evidence for these lines forming in a quasar outflow driven by radiation pressure ( see [ sec : loc ] below ) .
we fit each nal system with a gaussian optical depth profile .
the narrowest absorption lines are at least 1.5 times broader than the spectral resolution and the other lines are significantly broader than this .
the absorption lines are , therefore , fully resolved , and such gaussian optical depth profile fits are sufficient to determine accurate optical depths and covering fractions .
the optical depths and covering fractions are held constant across the width of each line profile .
gaussian fits are actually essential to distinguish individual absorption features in the crowded ly @xmath9 forest , and also useful to disentangle blended absorption in other areas of the spectrum .
furthermore , gaussian fits smooth over noise spikes and large optical depth and covering fraction uncertainties in the wings of the lines .
we also use gaussian fits to simultaneously fit and lock together various parameters including redshift , doppler @xmath37 parameter , covering fraction and a 2:1 optical depth ratio based on oscillator strength ratios for doublets such as the c iv , si iv , n v , and o vi . to measure accurate optical depths
, we must consider the possible effects of partial coverage of the emission source by the absorbing gas .
the line of sight covering fraction affects the observed line intensity as follows : @xmath38 where @xmath39 is the velocity dependant line of sight covering fraction , @xmath40 is the emitted ( unabsorbed ) intensity and @xmath41 and @xmath42 are the observed intensity and line optical depth at each velocity shift v. this equation assumes that the background light source has a uniform brightness given by i@xmath43 and the foreground absorber is homogeneous with a single value of @xmath42 .
the viability of this assumption is discussed by @xcite and @xcite .
we assume that all lines in a given multiplet have the same c@xmath44 at a given velocity . we do not explicitly attempt to distinguish between partial covering of the continuum source and of the bel region as discussed by @xcite
. however , we estimate from the sdss spectrum that the c iv bel peaks 20% above the continuum , which implies that the bel can only account for partial covering of 0.8 or higher .
we attempt to fit each system with the smallest possible number of gaussian components .
this minimizes the number of free parameters and provides a more robust characterization of column densities , ionisations and abundances in absorbing regions whose internal velocities might be more complex than simple gaussians ( e.g. , in outflows ) .
we fit each absorption line with a single gaussian unless 1 ) the system clearly has multiple components distinguished by inflection points that stand out significantly above the noise fluctuations in the spectrum ( e.g. system 7 ) , or 2 ) a single gaussian would miss a significant fraction , @xmath45% , of the absorption line strength ( e.g. system 4 ) .
the @xmath46% threshold is somewhat arbitrary , but it ensures that we achieve a good fit to the observed line and that significant portions of absorption ( i.e. , large enough to change the column density measurements ) are not missed . for these exceptional cases we use the minimum number of gaussians possible to achieve an accurate fit to the data . if a system is fit with two or more gaussians , each gaussian is labelled as a component .
we assume that the covering fraction is the same for all components in a given system , such that the optical depths in equation [ eq : cf ] simply add together in regions of component overlap ( see hamann et al . in preparation for further discussion ) .
this simplifying assumption is well justified by the excellent fits to all the systems , with the possible exception of system 6 , which we discuss in more detail in [ sec : sys ] .
all ions with gaussian fits are shown along with their gaussian optical depth profiles in figures [ fig:1]-[fig:9 ] .
badly blended members of , e.g. , the lyman - series lines are not used to constrain ionisation or abundance in [ sec : ion ] and are not shown .
we check the gaussian results using direct integration and point - by - point measurements of @xmath42 and @xmath47 across the line profiles for several systems with either non - gaussian profiles ( system 8) or @xmath48 ( systems 5 and 6 ) as discussed later in this section .
the free parameters in the gaussian fits are @xmath47 , central optical depth , @xmath49 , the doppler @xmath37 parameter , and redshift for each component in each system .
we fit the c iv absorption lines first , then base the fit for other absorption lines on the c iv fit parameters .
to ensure we are analysing the same gas in different ions , we fix the redshift for all absorption lines in a system to the c iv redshift .
we further exclude unwanted contributions from blends or complex multi - phase gas by fixing the @xmath37 parameter of all ions with ionisations lower than c iv to that of c iv .
higher ionisation lines , such as n v and o vi , are allowed to be broader .
however , we cap the @xmath37-values of the h i profiles at 140% of c iv .
this cap is important for the abundance analysis below ( [ sec : ion ] ) because it ensures that the derived h i column densities do not include gas with dramatically different kinematics than c iv .
we choose to cap the h i @xmath37-values at 140% of the c iv @xmath37-values instead of the much higher percentage expected for purely thermal broadening because the widths of the c iv lines exceed the thermal widths expected for a gas photoionised by either the quasar or the inter - galactic uv spectrum .
therefore , we assume the @xmath37-values are dominated by non - thermal broadening effects . on the other hand , setting the cap at 140% instead of something smaller , such as 100% , allows for some contribution of thermal broadening to @xmath37 in the narrower systems ( which would affect h i more than c iv ) .
overall , our fits to the lyman lines should lead to reasonable but generously large estimates of the amount of h i gas that coexists with c iv , and therefore , to conservatively low estimates of the c / h abundance . as stated above , the covering fraction
is a free parameter in the gaussian optical depth fits of each doublet . in cases where the best fit profile has @xmath48 , we repeat the fit with @xmath50 to test the robustness of the @xmath48 result .
we then compare how well each of the two fits with different values of @xmath47 match the data . in cases
where the fits are comparable , we assume @xmath50 , otherwise , the best fit is chosen .
for example , we confirm that the @xmath48 fit follows the data in systems 5 and 6 , as shown in figure [ fig:5 + 6 ] , whereas the @xmath50 fit does not match the observed doublet ratios in c iv and n v. we assume @xmath50 for all singlet lines .
the covering fraction for the lyman lines is fixed at the c iv doublet covering fraction .
this is necessary because the observed line ratios within the lyman series are too severely affected by blending in the ly @xmath9 forest to yield their own independent measures of @xmath47 .
table [ tab : lines ] lists fit parameters for all of the absorption lines that yield useful constraints for the ionisation and abundance analysis described in [ sec : ion ] .
absorption lines that are badly blended are not used in subsequent analysis , and are not listed in the table .
each system is listed separately by redshift and velocity shift , where negative velocities denote gas moving towards the observer .
only the stronger member of the o vi doublet is listed , as o vi is never used in the abundance analysis because of either line saturation or strong blending in these lines in all systems . however , the strength of o vi is still useful as an indicator of the ionisation of the gas . in systems where n v is not present , we list upper limits for the stronger member of the n v doublet for completeness .
table [ tab : lines ] lists the central wavelength ( @xmath0 ) and doppler @xmath37 parameter values along with column densities and rest equivalent widths ( rew ) derived from the gaussian optical depth profile fits .
systems 4 and 7 each have two blended components .
the values of @xmath51 , and @xmath52 are listed separately for these components in table [ tab : lines ] , but the rews , listed only with the first component data , apply to the entire blend .
we measure upper limits on h i column densities in all cases where all the lyman series absorption lines are blended with intervening absorption lines in the ly @xmath9 forest .
the same is true for singlet ions with upper limits on the column densities .
we estimate uncertainties for the column densities by placing the continuum at the top and bottom of the noise around the fitted continuum , corresponding to the reasonable maximum / minimum values ( @xmath53 uncertainties ) for continuum placement .
we measure @xmath54 uncertainties for the h i column densities of 0.18 dex on average .
the covering fraction is @xmath50 for all the systems , unless otherwise noted in the footnotes of table [ tab : lines ] , e.g. systems 5 and 6 .
we estimate the uncertainty in the covering fraction derived from the gaussian profile fits both formally and informally .
the formal uncertainties are estimated by propagating the error spectrum through the calculation of @xmath47 . however , these uncertainties are much smaller than the informal uncertainties , which are dominated by uncertainty in continuum placement .
we estimate covering fraction uncertainties due to continuum placement uncertainties by first shifting the continuum near each red doublet member up and down by the @xmath55 continuum uncertainty , and then fitting the doublets with this new continuum .
the actual uncertainties are probably smaller than the uncertainties we derive in this way , because a similar shift in the continuum around both doublet members ( a more likely occurrence ) produces smaller changes in @xmath47 .
we find @xmath56 for system 5 and @xmath57 for system 6 .
if we fix the covering fraction in systems 5 and 6 at @xmath50 instead of at the measured values , the column densities in all ions decrease by an average of 0.25 dex .
llccccclc [ tab : lines ] & @xmath58 & z@xmath59 & i d @xmath60 & @xmath0 & rew & @xmath37 & @xmath52 + & & v ( km s@xmath1 ) & & & & km s@xmath1 & @xmath5 & + & 1 & 3.42865 & ly @xmath11 973 & 4307.03 & 0.162 & 27.8 & 15.05 + & & -1442 & c iii 977 & 4326.88 & 0.076 & 10.8 & @xmath6113.37 + & & & ly @xmath9 1216 & 5383.78 & 0.445 & 27.8 & 15.05 + & & & n v 1239 & 5486.30 & @xmath610.002 & 10.8 & @xmath6112.57 + & & & c iv 1548 & 6856.43 & 0.034 & 10.8 & 12.99 + & & & c iv 1551 & 6867.81 & 0.018 & 10.8 & 12.99 + + & 2 & 3.42133 & c iii 977 & 4319.73 & 0.015 & 12.0 & @xmath6112.39 + & & -1938 & ly @xmath10 1026 & 4535.05 & 0.011 & 14.0 & @xmath6113.23 + & & & ly @xmath9 1216 & 5374.88 & 0.069 & 14.0 & @xmath6113.23 + & & & n v 1239 & 5477.24 & 0.034 & 20.5 & 13.24 + & & & n v 1243 & 5494.83 & 0.018 & 20.5 & 13.24 + & & & c iv 1548 & 6845.11 & 0.017 & 10.0 & 12.66 + & & & c iv 1551 & 6856.47 & 0.009 & 10.0 & 12.66 + + & 3 & 3.41864 & ly @xmath13 938 & 4143.82 & 0.011 & 13.9 & @xmath6114.26 + & & -2120 & ly @xmath11 973 & 4297.29 & 0.035 & 13.9 & @xmath62 + & & & ly @xmath9 1216 & 5371.61 & 0.186 & 13.9 & @xmath6114.26 + & & & n v 1239 & 5473.90 & @xmath610.006 & 9.5 & @xmath6112.45 + & & & si iv 1394 & 6158.52 & 0.023 & 7.2 & 12.48 + & & & si iv 1403 & 6198.30 & 0.013 & 7.2 & 12.48 + & & & c iv 1548 & 6840.94 & 0.096 & 9.5 & 13.62 + & & & c iv 1551 & 6852.29 & 0.062 & 9.5 & 13.62 + + & 4 & 3.41775 & ly @xmath13 938 & 4142.98 & 0.059 & 15.4 & @xmath6114.86 + & & -2182 & ly @xmath11 973 & 4296.43 & 0.164 & 15.4 & @xmath6114.86 + & & & ly @xmath9 1216 & 5370.53 & 0.589 & 15.4 & @xmath6114.86 + & & & n v 1239 & 5472.79 & @xmath610.055 & 11.0 & @xmath6112.28 + & & & si iv 1394 & 6157.28 & 0.081 & 9.1 & 12.76 + & & & si iv 1403 & 6197.09 & 0.046 & 9.1 & 12.76 + & & & c iv 1548 & 6839.56 & 0.215 & 11.0 & 13.56 + & & & c iv 1551 & 6850.91 & 0.132 & 11.0 & 13.56 + & & 3.41747 & ly @xmath13 938 & 4142.72 & * & 46.8 & @xmath6114.64 + & & -2200 & ly @xmath11 973 & 4296.15 & * & 46.8 & @xmath6114.64 + & & & ly @xmath9 1216 & 5370.19 & * & 46.8 & @xmath6114.64 + & & & n v 1239 & 5472.44 & * & 33.4 & @xmath6113.45 + & & & si iv 1394 & 6156.89 & * & 14.0 & 12.75 + & & & si iv 1403 & 6196.66 & * & 14.0 & 12.75 + & & & c iv 1548 & 6839.12 & * & 33.4 & 13.67 + & & & c iv 1551 & 6850.47 & * & 33.4 & 13.67 + + llccccclc + & 5@xmath63 & 3.40391 & c iii 977 & 4302.71 & 0.059 & 19.7 & @xmath6113.05 + & & -3121 & o vi 1032 & 4544.53 & 0.275 & 25.3 & @xmath6114.99 + & & & ly @xmath9 1216 & 5353.70 & 0.176 & 32.2 & 13.91 + & & & n v 1239 & 5455.65 & 0.098 & 13.0 & 14.30 + & & & n v 1243 & 5473.18 & 0.075 & 13.0 & 14.30 + & & & c iv 1548 & 6818.13 & 0.141 & 15.0 & 14.12 + & & & c iv 1551 & 6829.45 & 0.105 & 15.0 & 14.12 + + & 6@xmath64 & 3.40196 & ly @xmath11 973 & 4281.07 & 0.055 & 31.7 & 14.41 + & & -3254 & ly @xmath10 1026 & 4515.18 & 0.137 & 31.7 & 14.41 + & & & o vi 1032 & 4542.51 & 0.739 & 63.5 & @xmath6115.54 + & & & ly @xmath9 1216 & 5351.33 & 0.382 & 31.7 & 14.41 + & & & n v 1239 & 5453.24 & 0.284 & 50.0 & 14.57 + & & & n v 1243 & 5470.76 & 0.188 & 50.0 & 14.57 + & & & c iv 1548 & 6815.11 & 0.220 & 45.0 & 14.02 + & & & c iv 1551 & 6826.42 & 0.128 & 45.0 & 14.02 + + & 7 & 3.39936 & ly @xmath12 950 & 4178.26 & 0.008 & 27.3 & @xmath6113.19 + & & -3430 & c iii 977 & 4298.26 & 0.102 & 19.8 & @xmath6112.36 + & & & n iii 990 & 4354.49 & 0.023 & 19.8 & @xmath6113.04 + & & & ly @xmath9 1216 & 5348.17 & 0.295 & 27.3 & @xmath6113.19 + & & & n v 1239 & 5450.02 & 0.310 & 18.0 & 13.50 + & & & n v 1243 & 5467.53 & 0.180 & 18.0 & 13.50 + & & & c iv 1548 & 6811.09 & 0.187 & 19.9 & 13.09 + & & & c iv 1551 & 6822.40 & 0.101 & 19.9 & 13.09 + & & 3.39835 & ly @xmath12 950 & 4177.31 & * & 60.0 & @xmath6113.74 + & & -3496 & c iii 977 & 4297.28 & * & 43.7 & @xmath6113.19 + & & & n iii 990 & 4353.49 & * & 43.7 & @xmath6113.04 + & & & ly @xmath10 1026 & 4511.48 & * & 60.0 & @xmath6113.74 + & & & ly @xmath9 1216 & 5346.95 & * & 60.0 & @xmath6113.74 + & & & n v 1239 & 5448.77 & * & 52.4 & 14.22 + & & & n v 1243 & 5466.27 & * & 52.4 & 14.22 + & & & c iv 1548 & 6809.53 & * & 43.7 & 13.62 + & & & c iv 1551 & 6820.83 & * & 43.7 & 13.61 + + & 8 & 3.37983 & ly @xmath10 1026 & 4492.48 & 0.135 & 149.0 & @xmath6114.29 + & & -4763 & o vi 1032 & 4519.68 & 1.705 & 149.8 & 15.86 + & & & ly @xmath9 1216 & 5324.43 & 0.766 & 149.0 & @xmath6114.29 + & & & n v 1239 & 5425.82 & 1.045 & 155.5 & 14.88 + & & & n v1243 & 5443.26 & 0.643 & 155.5 & 14.88 + & & & c iv 1548 & 6780.85 & 0.807 & 155.5 & 14.41 + + & 9 & 3.35922 & ly @xmath11 973 & 4239.50 & 0.039 & 58.4 & @xmath6114.02 + & & -6186 & c iii 977 & 4258.94 & @xmath610.158 & 50.0 & @xmath6113.35 + & & & ly @xmath10 1026 & 4471.34 & 0.112 & 58.4 & @xmath6114.02 + & & & o vi 1032 & 4498.41 & 0.413 & 42.2 & 14.51 + & & & ly @xmath9 1216 & 5299.37 & 0.513 & 58.4 & @xmath6114.02 + & & & n v 1239 & 5400.29 & 0.212 & 40.0 & 13.84 + & & & n v 1243 & 5417.64 & 0.115 & 40.0 & 13.84 + & & & c iv 1548 & 6748.94 & 0.156 & 40.4 & 13.35 + & & & c iv 1551 & 6760.14 & 0.083 & 40.4 & 13.35 + + + + + we perform a simple test to determine the reliability of the @xmath48 result from the gaussian profile fits for systems 5 and 6 .
we predict the shape of the longer wavelength c iv and n v doublet members based on the intensity of the shorter wavelength member , combined with the 2:1 @xmath49-ratio derived from the oscillator strengths of each line .
the predicted shorter wavelength member will only match the data if @xmath50 .
these predictions are shown in figure [ fig : tau ] .
the observed data for both doublet members are plotted with solid curves and the predicted longer wavelength doublet member is plotted with a dot - dashed curve .
the predicted shape of the longer wavelength member of c iv and especially of n v is much weaker than the observed shape for system 5 . in system 6 , the line centres of c iv , and more clearly n v , are stronger in the observed data for the longer wavelength members than in the predictions .
we conclude that @xmath48 for at least some portion of each of these lines in systems 5 and 6 .
the gaussian fitting technique described above assumes a single covering fraction across the entire line .
however , we demonstrate with the above @xmath49-ratio analysis that covering fractions are not always constant across a single line profile . to better account for this , and to determine if the guassian fits find reasonable average values for @xmath47 , particularly for lines with very non - gaussian shapes , we use a point - by - point method in addition to the gaussian fitting method to determine @xmath42 and @xmath47 across the line profiles in several systems .
we fit systems 5 and 6 , the two systems with @xmath48 , along with system 8 , which has a very non - gaussian profile .
we step across the absorption line , calculating average intensity in each small ( a few times the resolution ) regularly spaced sections of the spectrum , using the ratio of the intensities in the doublet in equation [ eq : cf ] to measure @xmath47 and @xmath42 at each step .
the point - by - point fits are shown in figures [ fig : cf5 ] and [ fig : cf8 ] .
the solid curve shows the shorter wavelength doublet member , while the dot - dashed curve shows the longer wavelength doublet member .
the covering fraction at each point is represented by the filled circles .
the steps used for system 6 are three resolution elements wide , which is wide enough to smooth over the noise but narrow enough to avoid blending the wings and core of the line .
however , system 5 is narrow enough that using bins three resolution elements wide , or wider , across the wings of the line would blend too much information from the core and the continuum .
furthermore , the spectrograph resolution could be blending the covering fraction in the wings of the line with the continuum .
thus , for the narrow system 5 , we measure only the 3 resolution element bin at the line core .
the step size for system 8 is four resolution elements .
this larger step size further smooths over noise , and can be used because the line is much broader than the other systems , lessening the impact of blending of the core and wings of the line .
we derive formal covering fraction uncertainties ( @xmath65 ) , represented as error bars at each point in figures [ fig : cf5 ] and [ fig : cf8 ] . the average c iv and n v central covering fraction from the point - by - point method matches the c iv and n v covering fraction derived from the gaussian fitting method to within 10% in systems 5 and 6 . based on this result , the results of the @xmath49-derived doublet ratio analysis , and uncertainties derived from the gaussian fitting method , we are confident of the accuracy of the @xmath48 measurements for systems 5 and 6 , although the exact value of the covering fraction remains unknown beyond the fact that it is below 1 .
the c iv and n v column densities found in system 6 using the point - by - point method match the c iv and n v column densities derived from the gaussian fitting method to within 0.14 dex and 0.06 dex respectively .
the same comparison for n v in system 8 yields a difference of 0.18 dex between the two methods .
we conclude that the gaussian technique is sufficient for comparing different systems and generally provides accurate column density and covering fraction results for the purposes of this work .
the abundance ratios can be derived from the ratio of measured column densities corrected for the degree of ionisation in the gas .
for example , the relative carbon to hydrogen abundance normalised to solar is given by : @xmath66=\log\left(\frac{n({c~\textsc{iv}})}{n(h i)}\right ) + \log\left(\frac{f(h i)}{f({c~\textsc{iv}})}\right ) + \left[\frac{h}{c}\right]_{\sun } \label{eq : c / h}\ ] ] where @xmath67 is the ionisation fraction of a given ion , @xmath68 is the column density and the final term on the right - hand side is the logarithmic solar abundance ratio of hydrogen to carbon listed in @xcite .
the second term on the right is the ionisation correction ( ic ) .
these correction factors can be large when comparing a highly ionised metal like c iv to h i. the exact values depend on the ionisation mechanism .
photoionisation by the quasar spectrum is by far the most likely scenario based on the arguments in
[ sec : loc ] that all of the systems are likely to be intrinsic to the quasar environment .
we derive values of ic using the photoionisation calculations shown in @xcite .
their calculations adopt a nominal quasar spectrum consistent with recent observational estimates at the critical ionising ( far - uv ) photon energies .
the calculations also assume that the absorbing gas is optically thin in the lyman continuum , which is appropriate for the column densities we measure in the absorption lines of j1023 + 5142 ( table [ tab : lines ] ) .
ideally , we would constrain the absorber ionisations by comparing the ratios of observed column densities in different ions of the same element , such as @xmath69 or @xmath70 , to the theoretical results in @xcite . however , these constraints are only marginally usable in our data because @xmath71 and @xmath72 are always blended in the ly @xmath9 forest and are therefore only ever constrained as upper limits .
therefore we estimate the ic from ratios such as @xmath73 or @xmath74 , with the additional assumption that the relative metal abundances are approximately solar .
the specific ionisation constraints used for each system sometimes lead to upper limits , lower limits or specific values for the abundance ratios , and are described in more detail for individual systems in
[ sec : sys ] below .
our best estimates for the @xmath75 abundances based on these constraints are all super solar , except in system 1 .
table [ tab : abs ] , which contains several different abundance indicators for each nal system , lists these estimated best abundances in column 3 , titled @xmath76_{best}$ ] , for the nine systems .
we also calculate robust lower limits on the metal to hydrogen abundance ratios by applying minimum values of the ionisation correction ( ic@xmath77 , @xcite ) to the measured c iv , si iv and n v column densities , when available .
each metal ion has a unique global ic@xmath77 that occurs near the peak of its own ionisation fraction .
for example , @xmath78 peaks approximately where @xmath79 is largest .
we use the values of ic@xmath77 listed in @xcite . applying these minimum correction factors to the observed column density ratios ( equation [ eq : c / h ] ) leads to the firm lower limits listed for @xmath76_{min } , [ si / h]_{min}$ ] and @xmath80_{min}$ ] abundances in columns 46 of table [ tab : abs ] .
the minimum ionisation corrections provide firm lower limits on the abundances that do not depend on the ionisation uncertainties or the possibility of a multi phase gas .
in particular , any gas components not at an ionisation corresponding to ic@xmath77 would have the effect of raising the actual value of ic and thus also the actual abundance .
we derive total h column densities for each nal system from the h i column densities listed in column 7 of table [ tab : lines ] and the best ionisation correction described above .
we use @xmath81 where @xmath82 is the total h column density , @xmath83 is the column density of h i and @xmath84 is the h i fraction used to obtain ic .
@xmath82 for each system is listed in column 7 of table [ tab : abs ] .
the uncertainties in these results are dominated by uncertainties in the ic .
in addition to the limited constraints provided by the data , a few well - studied cases have shown that individual absorbers can span a range of ionisations and have a range of ic values ( e.g. @xcite ) .
we assume a single ionisation state for each absorption line system .
we discuss the individual systems briefly in
[ sec : sys ] .
.metal abundance and total h column density [ cols="<,^,^,^,^,^,^,^",options="header " , ]
_ system 1 , @xmath85 km s@xmath1 _ km s@xmath1 .
the velocity scale is with respect to the rest frame of the quasar based on @xmath86 , where negative velocities denote motion towards the observer and away from the quasar .
the velocity range is 400 km s@xmath1 for this and figures [ fig:2 ] through [ fig:7 ] and [ fig:9 ] .
the solid curve in each panel is the gaussian optical depth fit to individual lines .
the dashed vertical line is the central velocity of the system .
all of the lines used to derive or constrain column densities with gaussian fits are shown in the figure .
, width=264 ] the c iv in system 1 appears line - locked with the c iv in system 2 , as discussed further in [ sec : loc ] .
o vi is not present , or is very weak , implying that the ionisation is low .
further evidence for low ionisation is the weak c iv combined with strong h i measured in ly @xmath9 and lyman @xmath11 , as seen in figure [ fig:1 ] .
our best ionisation constraint comes from an upper limit on c iii , which means the best @xmath75 abundance is an upper limit as well .
this gas is a likely candidate for host galaxy halo gas based on the weakness of the metal lines and the low abundances .
km s@xmath1 .
the symbols and ranges are the same as in figure [ fig:1].,width=264 ] the c iv in system 2 appears to be line - locked with system 1 , as mentioned above and discussed further in [ sec : loc ] .
the h i could be shifted to a lower velocity by as much as 30 km s@xmath1 from the metal lines in this system , indicating a multi phase gas , but heavy blending obscures the precise shift of the lines as can be seen in figure [ fig:2 ] .
the ly @xmath9 absorption line is poorly constrained .
the resulting h i optical depth and doppler @xmath37 parameter are upper limits , resulting in lower limits for the best estimate of @xmath75 abundance .
we constrain the ionisation by the relative strengths of c iv and n v , assuming solar abundance ratios .
km s@xmath1 for system 3 and @xmath87 km s@xmath1 for system 4 .
the symbols and ranges are the same as in figure [ fig:1 ] . , width=264 ] the h i lines in system 3 are blended with those from system 4 , but appear consistent with the metal lines , shown in figure [ fig:3 + 4 ] .
because of the relatively poor constraints on the h i absorption lines , the h i optical depth and doppler @xmath37 parameter are upper limits , resulting in lower limits for the best estimate of @xmath75 abundance .
we constrain the ionisation by the relative strengths of c iv and si iv , assuming solar abundance ratios . the c iv and si iv doublets in system 4 are fit with two blended gaussian components to accommodate the asymmetric profile .
we use the central velocity of each component to identify the system .
the h i absorption lines are poorly constrained due to blending with system 3 .
the resulting h i optical depth and doppler @xmath37 parameter are upper limits , resulting in lower limits for the best estimate of @xmath75 abundance .
we constrain the ionisation by the relative strengths of c iv and si iv , assuming solar abundance ratios .
this system is broad and asymmetric , which is indicative of a wind or outflow feature ( see [ sec : loc ] ) .
km s@xmath1 for system 5 and @xmath88 km s@xmath1 for system 6 .
the solid curve is the @xmath48 gaussian fit for the c iv , n v and h i absorption lines . system 5 is the narrower system .
the ranges are the same as in figure [ fig:1 ] . , width=264 ] the h i in system 5 is well constrained by ly @xmath9 .
the o vi is strongly blended with that of system 6 as shown in figure [ fig:5 + 6 ] .
the covering fraction in the doublet is @xmath89 .
the @xmath90 gaussian fits are shown as solid curves in figure [ fig:5 + 6 ] .
system 5 appears to have 2 components ; a narrow , optically thick component sitting directly on top of a broader , optically thin one .
this is most clearly seen in figure [ fig:5 + 6 ] in the longer wavelength members of the c iv and n v doublets , which have a much sharper central feature than their shorter wavelength counterparts .
we constrain the ionisation by the relative strengths of c iv and n v , assuming solar abundance ratios .
the partial coverage in this system indicates that it is intrinsic to the quasar .
the partial coverage in this system and in system 6 are examined qualitatively with the @xmath49-ratio predicted doublets , shown in figure [ fig : tau ] , and further with the point - by - point analysis , illustrated in figure [ fig : cf5 ] .
both analyses confirm similar @xmath48 results in both systems ( see [ subsec : fit ] for details ) .
h i is well - constrained in system 6 , with three mostly blend - free lyman lines .
the covering fraction in the doublets is @xmath91 , similar to system 5 .
the solid curves in figure [ fig:5 + 6 ] represent the @xmath48 gaussian fits , as for system 5 .
the o vi lines are blended with the o vi lines in system 5 .
the covering fraction in h i appears to be @xmath50 because ly @xmath9 reaches zero intensity .
we constrain the ionisation by the relative strengths of c iv and n v , assuming solar abundance ratios .
the broad smooth shape , along with the partial coverage indicate that this system is part of an outflow .
this system appears somewhat asymmetric and the @xmath49-ratio analysis in figure [ fig : tau ] suggests further that there may be two components , one with partial covering near the line - centre , and a second broader component with complete covering in the blue wing .
although one component does not provide the best possible fit to all the lines in system 6 , it is not clear that adding a second distinct component would provide a better characterisation of the actual conditions in the absorber .
we test this by fitting the system with one and two gaussian components , where the two component fit still assumes the same covering fraction in both components .
both fits produce similar column densities in all ions , @xmath92 dex , @xmath93 dex and @xmath94 dex in the same direction , therefore we prefer the single gaussian fit in keeping with our prescription to minimise free parameters in the fits .
also , by using the single gaussian fit , we ignore parts of the ly @xmath9 absorption which do not correspond directly to c iv absorbing gas , and therefore retain the ability to directly compare h i and c iv column densities for the abundance analysis . and -3496 km s@xmath1 .
the symbols and ranges are the same as in figure [ fig:1 ] .
, width=264 ] the h i column density is constrained as an upper limit in system 7 because of blending in the lyman lines , shown in figure [ fig:7 ] , resulting in lower limits for the best estimate of @xmath75 abundance .
we fit this broad system with two gaussian components to better match the absorber shapes , and identify the system by the central velocities of the two components .
the ionisation is constrained by the relative strengths of c iv and n v , assuming solar abundance ratios .
km s@xmath1 .
the velocity range is 1100 km s@xmath1 .
the symbols and ranges are the same as in figure [ fig:1].,width=264 ] the longer wavelength member of the c iv doublet in system 8 falls on a gap between orders of the spectrograph between 6785 and 6795 , but the n v doublet is present in the spectrum , as is the shorter wavelength member of the c iv doublet .
the n v doublet is used to determine the @xmath47 and the doppler @xmath37 parameter for both doublets .
this system is almost broad enough to be a mini - bal , and is likely an outflow system based on the shape and strength of the line profile , shown in figure [ fig:8 ] .
the h i appears to be relatively weak in this system compared to the metal lines , although there is severe blending in the ly @xmath9 forest .
this blending means the h i absorption is poorly constrained with an upper limit , and therefore the best estimate for @xmath75 abundance is a lower limit .
we constrain the ionisation by the relative strengths of c iv and n v , assuming solar abundance ratios .
we use the gaussian fit to compare system 8 to other systems , but the profile of system 8 is distinctly non - gaussian
. therefore we also fit the central trough of the line with a point - by - point analysis , shown in figure [ fig : cf8 ] .
the @xmath75 abundance found by the gaussian fit is consistent within 10% of the @xmath75 abundance found using the point - by - point method .
, -6186 and -6298 km s@xmath1 .
although all three components are fit with gaussian profiles , only the central component is considered in the abundance analysis .
the symbols and ranges are the same as in figure [ fig:1].,width=264 ] system 9 has three components , but we chose to analyse only the central component for abundances , as the two outer components are very poorly constrained , as shown in figure [ fig:9 ] .
this system has the highest velocity shift out of the group of narrow absorption lines , and lies just nominally outside of the velocity shift region for associated lines ( v @xmath95 km s@xmath1 ) , at @xmath96 km s@xmath1 .
the lyman lines could be shifted up to 20 km s@xmath1 from the metal lines , indicating a possible multiphase gas , but the line are too weak to determine their precise centroids .
the weakness of the lyman lines , along with blending in the ly @xmath9 forest mean the h i column densities are upper limits , so the best estimate of the @xmath75 abundance is a lower limit .
we constrain the ionisation by the relative strengths of c iv and n v , assuming solar abundance ratios .
-predicted line profiles for systems 5 and 6 .
system 5 is the narrower system .
the dot - dashed curve shows the smoothed c iv and n v predicted long wavelength doublet member , based on doublet optical depth ratio from short wavelength member , assuming @xmath50 .
the actual smoothed data for the shorter and longer wavelength doublets are shown as the bold and thin solid curves , respectively .
the longer wavelength data are stronger than the predictions , indicating partial covering , especially for the n v doublet in system 5 and in the centre of the n v doublet in system 6 . , width=264 ] at each step so that a point at zero flux has complete coverage , and a point at the continuum flux of one has no coverage .
the circles are located at the centre of the average velocity steps .
, width=264 ] .,width=264 ]
j1023 + 5142 has nine nal systems with a range of column densities from @xmath97 to @xmath4 @xmath5 , velocities from 1400 to 6200 km s@xmath1 , c iv doppler @xmath37 values from 7 to 150 km s@xmath1 , c iv rew(1548 ) from 0.02 to 0.81 and two systems with partial covering of either the continuum source or the broad emission line region ( blr ) , @xmath98 , which imply absorber diameters of @xmath99 pc or @xmath100 pc ( discussed below in [ sec : out ] ) .
these systems are generally much weaker than those studied in larger statistical surveys of nals , such as @xcite , which use lower resolution data , and measure c iv rew integrated across the doublet , with completeness limits of 0.30.5 .
the nal systems all appear to be highly ionised ; none of the systems exhibit low ionisation species such as si ii , c ii or si iii , whereas all contain c iv and some contain higher ionisation species such as o vi and n v. systems 5 , 6 , 8 and 9 exhibit high ionisation ( o vi ) absorption , and others may also have absorption at these wavelengths that is not observable due to blending in the ly @xmath9 forest .
systems 29 exhibit supersolar metallicities ranging from @xmath101 to @xmath102 .
system 1 has a slightly lower metallicity of @xmath103 .
we examine several diagnostics to estimate directly the location of each system .
the tight grouping and similar high metallicities ( see [ sec : metal ] for further discussion ) for all but one ( systems 29 ) of the nine c iv absorption line systems in j1023 + 5142 suggest a possible physical connection between the absorbers .
the proximity of this nal complex to the quasar redshift suggests further that the physical relationship includes the quasar itself .
the velocity span across the group is too large to be explained by a single galaxy or even a large cluster of galaxies
. it might be consistent with some larger cosmic structure connected to the quasar , but then we would expect the velocity distribution to include the red side of the quasar systemic .
a more likely explanation is that the nal complex formed in a multi - component outflow from the quasar .
there are several indirect arguments for an intrinsic origin for the gas in this nal complex .
i ) 8 of the 9 systems have super solar metallicities , discussed in detail in [ sec : metal ] below .
ii ) some authors have argued that strong o vi absorption may indicate intrinsic gas near the quasar .
@xcite carry out a detailed study of o vi absorption in @xmath104 quasars and argue that @xmath105 log(@xmath5 ) indicates an intrinsic origin , supported by evidence for partial covering in most of these systems .
we measure @xmath106 in four systems in j1023 + 5142 .
two of them ( 6 and 8) are above the intrinsic thereshold defined by fox et al . , while the other two ( 5 and 9 ) are very near this threshold at @xmath107 .
iii ) all systems with o vi , that is systems 5 , 6 , 8 and 9 , also have strong n v compared to c iv .
strong n v , especially compared to c iv , is often ( though not always ) present in intrinsic gas ( e.g. @xcite ) .
iv ) the presence of strong o vi and n v , especially with the absence of low ionisation species such as c ii in these absorption systems is consistent with gas exposed to the intense ionising radiation field near a quasar . if the nals in j1023 + 5142 are intrinsic to the quasar environment , the most likely origin is in a quasar driven outflow .
other possible intrinsic origins all have lower velocities : i ) starburst - driven outflows typically have @xmath108 km s@xmath1 @xcite , and in seyfert galaxies have maximum outflow speeds of 600 km s@xmath1 , and more typical speeds of 100 - 200 km s@xmath1 @xcite , ii ) other galactic / halo gas should have velocities near the typical velocity dispersion for such galaxies ( @xmath109 km s@xmath1 ) , iii ) gas in the narrow line region of the quasar has typical velocities of @xmath110 km s@xmath1 , and maximum velocities of @xmath111 km s@xmath1 @xcite , and iv ) intra - cluster galaxy motions are shown by @xcite to generally have velocity dispersions @xmath112 km s@xmath1 or less for clusters with higher numbers of active galactic nuclei @xcite .
statistical studies , @xcite ( see also @xcite ) , have shown that @xmath113% of nals at @xmath114 km s@xmath1 with rew(1548 ) @xmath115 form in high - velocity quasar outflows .
this percentage increases to @xmath2057% for the narrower range of @xmath116 km s@xmath1 , spanned by the nals in j1023 + 5142 .
the percentage reaches @xmath2072% for the narrow range of @xmath117 km s@xmath1 , which encompasses systems 1 through 4 in j1023 + 5142 .
these percentages are probably lower for weaker lines ( @xcite and private communication ) .
@xcite also find that for c iv nals with rew(1548 ) @xmath118 at velocities @xmath119 km s@xmath1 the intrinsic ( outflow ) fraction is @xmath120% and at higher velocities , @xmath121 km s@xmath1 , the intrinsic fraction is @xmath122% .
nonetheless , these outflow fractions support the idea that most or all of the systems in this group of nine nals in j1023 + 5142 form in a quasar outflow .
we search for direct signatures of quasar outflow origin via 1 ) line variability , 2 ) partial covering and 3 ) broad profiles ( see @xcite and references therein for more discussion ) .
\1 ) we have only very poor constraints on the variability .
we compare c iv and n v rew results measured from the guassian fits to the sdss and keck spectra ( @xmath123 months ) in search of variability in the absorption lines .
system 8 is the only individual c iv and n v system resolved in the sdss spectrum , while the weaker lines are not detected in the sdss spectrum .
system 8 is the strongest of the nine systems , and did not vary in rew by more than 15% in c iv and n v between the sdss and keck observations .
for the eight weaker systems , we conclude only that variability greater than a factor of 2 to 3 did not occur .
\2 ) there is partial covering in two ( systems 5 and 6 , figures [ fig:5 + 6 ] and [ fig : cf5 ] ) out of the nine systems .
absorption lines with partial covering of the luminosity source are attributed to gas near the quasar because partial covering is not expected to occur in intervening clouds or galaxies @xcite .
the presence of partial covering in these lines strongly suggests that the gas is intrinsic and located in the near quasar environment .
\3 ) the profiles of systems 4 , 6 , 7 , 8 , and 9 shown in figures [ fig:3 + 4 ] , [ fig:5 + 6 ] , [ fig:7 ] , [ fig:8 ] and [ fig:9 ] have c iv and n v @xmath37 values between 33 and 155 km s@xmath1and o vi @xmath37 values between 42 and 150 km s@xmath1 .
these @xmath37 values are broad and smooth compared to the thermal widths for gas at the highest expected temperature near @xmath124 k @xcite for a photoionised gas near a quasar ( 33 km s@xmath1 for h and less than 10 km s@xmath1 for c and n ) .
they are also broader than typical non - damped ly @xmath9 intervening c iv , n v and o vi absorption lines , which have on average @xmath125 km s@xmath1 for o vi , and @xmath126 km s@xmath1 for c iv and n v @xcite .
these profiles , therefore , exhibit morphologies consistent with formation in an outflow .
these three characteristics , variability , partial covering , and broad profiles are often found together in a single object , further supporting the idea that each individual characteristic likely indicates an outflow .
one well studied nal outflow in j2123 - 0050 @xcite is a prime example of all three , exhibiting variability , partial covering , and broad profiles that still have fwhms that are as narrow or narrower than many of the systems in j1023 + 5142 .
there is more tentative evidence for a quasar outflow origin in the apparent line lock between the c iv doublets in systems 1 and 2 .
line locking , where the difference in outflow velocities of two systems is exactly the velocity separation of the doublet , means that the lines are being radiatively accelerated directly towards the observer .
the reality of the line locking in this case is unclear , due to the difference in derived metallicities between the two systems .
nevertheless , the incredibly small velocity offset ( [ subsec : id ] ) , along with the very small probability for chance alignments @xcite suggest that the phenomenon may be real . the possible line
lock in c iv in systems 1 and 2 suggests that they are both part of an outflow and that these weak c iv lines play a significant role in radiatively driving the flow .
if this is really the case here , the gas probably originated near the source of radiative acceleration , i.e. the quasar .
finally , we note a trend in line width with velocity shift away from the quasar .
the narrowest lines , with fwhm @xmath127 km s@xmath1 are closest to the quasar redshift .
the lines appear progressively broader as the velocity shift increases , with the broadest system described as a ( narrow ) mini - bal with fwhm @xmath128 km s@xmath1 , shown in figure [ fig : civ ] .
a similar phenomenon has been observed before in other quasars with multiple c iv absorption lines clearly forming in outflows @xcite .
although it provides no direct information on the absorber locations , the appearance of this pattern in j1023 + 5142 supports the idea that at least some of the systems form in a quasar outflow .
the tight grouping of all nine of the systems also suggests a relationship between them .
determine that the probability of six similarly grouped nals in the quasar rx j1230.8 + 0115 all forming in intervening ( uncorrelated ) gas at similar velocity shifts is extremely small .
the similarities between those nals and the nal complex in j1023 + 1542 implies a similarly small probability for all nine nals in this complex forming independently in intervening gas .
although there could be up to several interlopers in the nal complex of j1023 + 5142 that might form in nearby galaxies in the line of sight , the density of these galaxies required to form all of the absorbers in the complex is beyond any expectations of cluster density at this redshift .
overall , we conclude that at least six out of the nine systems originate in a highly structured outflow driven by the quasar , because they exhibit one or more of the following properties : partial covering , broad profile shapes , large line strengths , tight grouping with other systems , and proximity to the quasar redshift .
systems 5 and 6 are the most likely outflow candidates because they exhibit partial covering as well as several of the other properties listed above .
systems 4 and 8 are likely outflows because of their strong , broad , asymmetric and smooth profiles and systems 7 and 9 are probably outflows because of their broad and smooth shapes .
systems 1 through 3 are more ambiguous in origin , with narrow widths , complete covering , lower velocity shifts and smaller strengths .
however systems 1 and 2 exhibit line - locking , which could be evidence of an outflow .
ultimately , we find strong evidence that systems 4 , 5 , 6 , 7 , 8 and 9 are part of a quasar outflow , whereas systems 1 , 2 , and 3 could consist of intervening gas from the igm or other galaxies in the line of sight .
as described in
[ sec : loc ] , the evidence suggests that the majority of absorption lines in this grouping are part of a complex quasar outflow .
this flow must be highly structured , with at least 6 and as many as 9 distinct absorbing structures along the line of sight .
the velocities in the 6 most secure outflow systems range from -2120 to -4760 km s@xmath1 .
several of these systems ( 4 , 6 , 7 and 8) also have super thermal line widths , indicative of large turbulence or strong radial velocity sheer across the outflow structure .
some of the outflow structures , represented by systems 5 and 6 , must be spatially small to produce partial covering of the background emission source .
these lines lie on top of the very weak c iv broad emission line .
therefore , nearly all ( @xmath12980% ) of the flux beneath these lines is continuum emission and any partial covering below @xmath130 can be ascribed to the continuum source and not the much larger blr .
we estimate the diameter of the accretion disk continuum source at 1550 to be @xmath131 pc and the diameter of the c iv broad line region to be @xmath132 pc , based on the scaling relationsl@xmath133(1450 ) , gives @xmath0l@xmath134(1450 ) .
we measure the c iv emission line fhwm in the sdss spectrum , and using equation 7 in @xcite , derive a black hole mass of @xmath135 . based on these values
, we calculate an eddington luminosity fraction of 0.8 .
the black hole mass and eddington luminosity fraction are then used in the scaling relations in @xcite to calculate the size of the continuum and broad line emission regions . ] in @xcite . to partially cover the emission source
, the absorbing clouds should have characteristic sizes similar to or less than the blr diameter , and possibly even less than the accretion disk diameter
. if the absorbers are discrete clouds , their small sizes and substantial velocity dispersions should lead to fairly rapid dissipation in the absence of an external pressure ( see @xcite ) .
in particular , the characteristic size of @xmath136 pc or @xmath137 pc combined with @xmath138 km s@xmath1 in the partial covering system 6 indicate a dissipation time of roughly @xmath139 yrs or @xmath140 yrs . at the measured velocity of v @xmath141 km s@xmath1 , this gas component would travel just @xmath142 pc or @xmath143 pc before dissipating .
a thorough discussion of the creation and survival of these absorbing structures is beyond the scope of this paper .
however , these simple arguments suggest that at least some of the outflow components we measure are very near their point of creation . it is useful to compare the basic properties of this nal outflow to bals .
the ionisations in both types of outflows are similar , with very little low ionisation gas ( e.g. c iii ) .
the outflow velocities of the nals in j1023 + 5142 are lower , @xmath144 km s@xmath1 , than the typical outflow velocities in bals , which can reach up to @xmath145 km s@xmath1 @xcite , but they do overlap .
the nals have total h column densities ( @xmath97 to @xmath4 @xmath5 , individual values listed in table [ tab : abs ] ) , more than 1000 times lower than typical bal h column densities ( @xmath146 @xmath5 , and probably higher ) . by definition ,
these nals also have fwhms around 1000 times narrower than typical bals , and much smaller rews as well .
nonetheless , nals like this might be part of the same general outflow phenomenon as bals , viewed at different angles @xcite .
this complex of weak nal outflows appears to be dramatically different from typical bal outflows , and constitutes a nearly unexplored part of the quasar outflow phenomenon , with a range of physical parameters and kinematics more complex and varied than previously thought .
it is well - known that nals are a common feature of quasar spectra .
previous surveys have found that 40% of quasars have c iv nals , in particular 25% have strong c iv nals within v @xmath147 km s@xmath1 @xcite , 60% have quasar - driven outflows in some form , either bals , nals , or something in - between @xcite , and including high - velocity outflows raises the percentage to 70% @xcite .
if the coverage fraction of these outflows is less than 100% , which is likely , they could be ubiquitous in the near quasar environment , and could potentially play an influential role in the physical processes occurring therein .
finally , we would like to understand what role the nal outflow in j1023 + 5142 might have in feedback to galaxy evolution .
the low speeds and small column densities , e.g. , compared to bal flows , suggest that its feedback contribution is negligible . however , there are large uncertainties relating to the outflow location and geometry .
for one particular nal outflow at a derived radial distance of @xmath148 pc , @xcite estimate that the kinetic energy yield is several orders of magnitude smaller than that necessary to influence feedback . at the opposite extreme ,
@xcite argue that the feedback contribution is significant for another nal outflow at a derived distance of @xmath149 kpc . the location of the nal outflow in j1023 + 5142 is not known well enough to make these estimates .
a more sensitive search for variability in these nals could be very helpful for refining both the location and the total energy yield ( see @xcite and references therein ) .
we find greater than or consistent with solar abundances in all of the absorption systems in j1023 + 5142 except in system 1 , in agreement with previous studies of narrow associated absorption at lower redshifts @xcite .
these high metallicities are consistent with the results of other studies of intrinsic gas as well , including bel gas @xcite and therefore consistent with our interpretation that the gas is intrinsic to the quasar .
intervening absorbers generally have very low metallicities , with @xmath150 no more than a few hundredths solar , although there are cases where high metallicity intervening gas has been observed @xcite .
we argue that the high metallicities found in 8 of the 9 systems in this quasar are consistent with locations near the quasar , however , we do not rely solely on this argument to determine the gas location .
instead , we consider that high metallicities could be a general phenomenon found in all gas in the quasar host environment @xcite .
the high metallicities of the nal systems in j1023 + 5142 require that its host galaxy had vigorous star formation in the epoch before the quasar was observable , leading to metal - rich gas in the quasar outflows @xcite .
this evidence , along with previous studies of bels lead us to conclude that the generally accepted paradigm of quasar - host galaxy evolution is correct , where a major merger leads to a vigorous burst of star formation , which then funnels gas to the centre of the galaxy and ignites a quasar that eventually blows out obscuring gas and dust to become visibly luminous @xcite . however , larger samples are needed to examine the full range of nal properties and study their relationships to quasar outflows and host galaxy environments .
measurements at high redshifts are particularly valuable because this is the main epoch of host / massive galaxy formation when the nal gas might have a close relationship to ongoing or recent star formation in the hosts .
we use nals to improve our understanding of the evolutionary relationship between the central black hole and its host galaxy through the study of their location , origin and abundance information in high redshift quasars . here
, we examine the properties of nine nal systems in the quasar j1023 + 5142 and find @xmath151 from @xmath152 to @xmath4 @xmath5 , velocities from 1400 to 6200 km s@xmath1 , c iv doppler @xmath37 values from 7 to 150 km s@xmath1 , c iv rew from 0.02 to 0.81 and two systems with partial covering of either the continuum or the blr , at the level of @xmath98 , which imply absorber diameters of @xmath99 pc or @xmath100 pc .
the nal systems all appear to be highly ionised ; none of the systems exhibit low ionisation species such as si ii , c ii or si iii , whereas all contain c iv and some contain higher ionisation species such as o vi and n v. the c iv absorption nals are tightly grouped , suggesting that they have a physical relationship to one - another , and the proximity of the nal complex to the quasar redshift suggests that the physical relationship includes the quasar itself .
the range in velocity across the complex is larger than can be easily explained by a single galaxy or even by a large cluster of galaxies .
a more likely explanation is that the nal complex formed in a multi - component quasar - driven outflow .
we estimate directly the location of each system in j1023 + 5142 through the use of several diagnostics and find strong evidence ( partial covering , broad and smooth profiles compared to thermal widths , velocities greater than galaxy dispersion velocities , supersolar metallicities ) that systems 4 , 5 , 6 , 7 , 8 and 9 are part of a quasar outflow .
systems 1 , 2 , and 3 have more ambiguous origins because they exhibit narrow widths , lower velocity shifts , and system 1 has a lower metallicity , so these systems could consist of intervening gas from the igm or other galaxies in the line of sight . systems 29 in j1023 + 5142 exhibit supersolar metallicities ranging from @xmath101 to @xmath102 .
system 1 has a lower metallicity of @xmath103 .
the high metallicities are consistent with scenarios of galaxy and black hole formation and evolution .
the nals in outflows appear to be part of a related outflow complex , which is very different than other known outflow regions such as bal outflows , and constitutes a relatively unexplored part of the quasar outflow phenomenon .
the outflows in j1023 + 5142 could be important for feedback between the black hole and the host galaxy , depending on the radial distance of the gas from the quasar .
the narrow widths of nals mean that detailed studies of individual objects like this are the only way to make progress in understanding this type of outflow .
variability studies could be useful to add more examples of nal outflows to the current sample available for similar detailed analysis .
we will add significantly to this sample in future work , including detailed studies of the full range of nal properties in 24 quasars at high redshift , during the main epoch of host / massive galaxy formation .
this work was supported in part by a nasa chandra award ( tm9 - 0005x ) and a grant from the space telescope science institute ( hst - go-11705 ) .
t. , bechtold j. , foltz c. , 1997 , in arav n. , shlosman i. , weymann r. j. , eds , mass ejection from active galactic nuclei vol .
128 of astronomical society of the pacific conference series , variability in radio - loud quasar associated absorption lines .
pp 25+ t. a. , hamann f. , sargent w. l. w. , 1997 , in arav n. , shlosman i. , weymann r. j. , eds , mass ejection from active galactic nuclei vol . 128 of astronomical society of the pacific conference series , partial coverage and time variability of narrow - line intrinsic qso absorption systems .
pp 13+ f. , sabra b. , 2004 , in richards g. t. , hall p. b. , eds , agn physics with the sloan digital sky survey vol .
311 of astronomical society of the pacific conference series , the diverse nature of intrinsic absorbers in agns .
pp 203+ y. , strauss m. a. , oguri m. , hennawi j. f. , fan x. , richards g. t. , hall p. b. , gunn j. e. , schneider d. p. , szalay a. s. , thakar a. r. , vanden berk d. e. , anderson s. f. , bahcall n. a. , connolly a. j. , knapp g. r. , 2007 , aj , 133 , 2222 | we discuss the nature and origin of a rich complex of narrow absorption lines in the quasar j102325.31 + 514251.0 at redshift 3.447 .
we measure nine c iv(@xmath01548,1551 ) absorption line systems with velocities from 1400 to 6200 km s@xmath1 , and full widths at half minimum ranging from 16 to 350 km s@xmath1 .
we also detect other absorption lines in these systems , including h i , c iii , n v , o vi , and si iv .
lower ionisation lines are not present , indicating a generally high degree of ionisation in all nine systems .
the total hydrogen column densities range from @xmath2@xmath3 to @xmath4 @xmath5 .
the tight grouping of these lines in the quasar spectrum suggests that most or all of the absorbing regions are physically related .
we examine several diagnostics to estimate more directly the location and origin of each absorber .
four of the systems can be attributed to a quasar - driven outflow based on line profiles that are smooth and broad compared to thermal line widths and to the typical absorption lines formed in intergalactic gas or galaxy halos .
several systems also have other indicators of a quasar outflow origin , including partial covering of the quasar emission source ( e.g. , in systems with speeds too high for a starburst - driven flow ) , o vi column densities above @xmath6 @xmath5 and an apparent line - lock in c iv ( in two of the narrow profile systems ) . a search for line variability yielded null results , although with very poor constraints because the comparison spectra have much lower resolution . altogether ( but not including the tentative line - lock )
there is direct evidence for 6 of the 9 systems forming in a quasar outflow .
consistent with a near - quasar origin , eight of the systems have metallicity values or lower limits in the range @xmath7 . the lowest velocity system , which has an ambiguous location based on the diagnostics mentioned above , also has the lowest metallicity , @xmath8 , and might form in a non - outflow environment farther from the quasar .
overall , however , this complex of narrow absorption lines can be identified with a highly structured , multi - component outflow from the quasar .
the high metallicities are similar to those derived for other quasars at similar redshifts and luminosities , and are consistent with evolution scenarios wherein quasars appear after the main episodes of star formation and metal enrichment in the host galaxies .
quasars : general quasars : individual quasars : absorption lines galaxies : evolution |
the nature of dark matter ( dm ) is one of the most relevant topics in physics today .
even if the list of experimental evidence is large and variate , we still do not know what is made of .
given the fact that many of the properties of the particle ( or particles ) forming the dark matter in the universe are unknown , it is important to use different approaches .
as it will be explained in the following , neutrino telescopes have specific advantages and can explore models not accessible to other searches .
the antares neutrino telescope @xcite is installed in the mediterranean sea , about 40 km off the french coast , near toulon , at a depth of 2500 m. it is made of 885 photomultimpliers ( pmts ) housed in glass spheres , forming the so - called optical modules ( oms ) .
the oms are installed in 12 lines anchored at the sea bottom and kept vertical by buoys .
the operation principle is based on the detection of the cherenkov light induced in water by the leptons produced after the weak interaction of neutrinos in the surroundings of the detector .
the main channel , given its best angular resolution , are the muon tracks from the cc interactions of muon neutrinos .
this is the only channel used in this analysis . for a review of recent results of the antares experiment see @xcite
neutrino telescopes are an important and complementary experimental approach to detect dark matter .
this technique offers some unique advantages .
the search for dark matter with neutrino telescopes is based on the idea that dark matter particles would self - annihilate . in most models ,
direct neutrino production is not expected , but the decay of the secondary particles produced in the annihilations would yield neutrinos that can reach the detector .
many models assume that the dm particle is a weakly interacting massive particle ( wimp ) .
there are several sources interesting for dark matter searches . in the case of the sun or the earth ,
the dm particles are scattered off with the nuclei in these astrophysical bodies and lose energy , so they can become gravitationally bound , accumulate and self - annihilate . in this case , the results on the neutrino flux can be related to the wimp - nucleon scattering cross section , analogously to direct searches . for this
, equilibrium between capture and annihilation must be assumed , which is a reasonable assumption for the sun , but not in the case of the earth .
the second kind of searches is the case of astrophysical bodies where dark matter gathers by gravitational accumulation along the process of structure formation .
the candidate sources are the galactic centre , dwarf galaxies and galaxy clusters . for these
, the dm annihilation cross section can be measured or constrained .
given its proximity , the best results with neutrino telescopes are obtained for the galactic centre .
the sun is one of the most interesting sources to be looked at with neutrino telescopes .
the main advantage of this analysis is that it is free from astrophysical uncertainties , contrary to cases like searches with gamma - rays or cosmic rays . for the latter ,
recent hints are very difficult to interpret as a genuine dark matter signal since other explanations ( pulsars , supernova remnants , etc . )
are also plausible . for neutrino telescopes , on the contrary
, high energy neutrinos are only expected from dark matter scenarios .
the background comes only from atmospheric muons and atmospheric neutrinos , the rate of which can be robustly estimated from data .
a contamination from high energy cosmic rays interacting in the sun s corona and producing high energy neutrinos is negligible .
the results of the search for dark matter in the sun using antares data of 2007 - 2012 are presented in figure [ sdcs ] .
it shows the limits on the spin dependent wimp - proton cross section . for this case
neutrino telescopes ( antares , superkamiokande , icecube ) obtain limits which are better than any direct search . for the spin - independent cross section
this is not the case .
limits set by antares on the wimp - p scattering cross section ( spin - dependent ) as a function of the wimp mass , compared to other experiments . ]
the galactic centre is also a major candidate for dm searches with neutrino telescopes . as mentioned above the results
are interpreted in terms of annihilation cross section , so in some sense they are complementary to the analysis of the sun . moreover , in this case the high energy component of the neutrino spectrum arrives unaffected to the detector , while in the sun part of these neutrinos are absorbed .
figure [ gc ] shows the results of the search for neutrinos in the direction of the galactic centre .
these limits are better than those of its main competitor , icecube , since the visibility of the galactic centre is better for antares due to their respective geographical locations . for masses above 10 tev , antares provides the best limits in the world . left : distribution of events in the sky around the galactic centre in the antares data of 2007 - 2013 . right : limits set by antares on the wimp annihilation cross section as a function of the wimp mass , compared to other experimental searches , for the @xmath0-@xmath0 channel and the nfw halo profile . ]
left : distribution of events in the sky around the galactic centre in the antares data of 2007 - 2013 .
right : limits set by antares on the wimp annihilation cross section as a function of the wimp mass , compared to other experimental searches , for the @xmath0-@xmath0 channel and the nfw halo profile . ]
we have presented the results of several dark matter searches done with antares . for the sun ,
the results of neutrino telescopes , including those of antares ( with data of 2007 - 2012 ) , offer the best limits for spin - dependent wimp - p cross section and a case free of astrophysical uncertainties . for the galactic centre , antares ( with data of 2007 - 2013 ) set the best limits among neutrino telescopes , which are also best of any technique for wimp masses above 10 tev .
the authors thank the support of plan estatal de investigacin ( refs .
fpa2015 - 65150-c3 - 1-p , -2-p and -3-p , ( mineco / feder ) ) , severo ochoa centre of excellence and multidark consolider ( mineco ) , and prometeo and grisola programs ( generalitat valenciana ) , spain .
9 m. ageron _ et al . _ ( antares collaboration ) , nucl . instrum . meth . a 656 , 11 - 38 ( 2011 ) high - energy neutrino searches in the mediterranean sea : probing the universe with antares and km3net / arca , a. kouchner on behalf of the antares collaboration , proc . of this conference s. adrin
et al . , antares collaboartion , physics letters b , volume 759 ( 2016 ) | neutrino telescopes have a wide scientific scope .
one of their main goals is the detection of dark matter , for which they have specific advantages .
neutrino telescopes offer the possibility of looking at several kinds of sources , not all of them available to other indirect searches . in this work
we provide an overview of the results obtained by the antares neutrino telescope , which has been taking data for almost ten years .
one of the most interesting ones is the sun , since a detection of high energy neutrinos from it would be a very clean indication of dark matter , given that no significant astrophysical backgrounds are expected , contrary to other indirect searches .
moreover , the limits from neutrino telescopes for spin - dependent cross section are the most restrictive ones .
another interesting source is the galactic centre , for which antares has a better visibility than icecube , due to its geographical location .
this search gives limits on the annihilation cross section .
other dark matter searches carried out in antares include the earth and dwarf galaxies . |
many systems in nature or society can be well described by small world networks ( swn ) , which were first proposed by watts and strogatz in ref .
swns are those intermediate between a regular lattice and a random network
. they can be realized by introducing a very small portion of long range links to a regular lattice .
such networks appear as small - worlds like random graphs , i.e. , with a small average node - node distance that grows logarithmically with the network size . meanwhile they also have usually large clustering coefficients @xcite .
since the first model system was proposed@xcite , a large literature has appeared and the properties of various models and processes on swns have been extensively studied , including percolation , coloring , coupled oscillators , iterated games , diffusion processes , epidemic processes , and spin models ( see ref . @xcite and references therein ) . for example , it has been recently found that some spin - lattice models belonging to different universality classes now show mean - field behavior on swns@xcite .
furthermore , recently zhu _ et al .
_ introduced swn effect to critical dynamics @xcite , and the investigation has been extended to the kinetic properties of the spin models . according to glauber s and kawasaki s theory @xcite , the time evolution of the order parameters
can be described by a markov process with a specific choice of the dynamic mechanism .
two mechanisms have been extensively studied : glauber s single - spin flip mechanism@xcite with the order parameter nonconserved and kawasaki s spin - pair exchange mechanism@xcite with the order parameter conserved . in the past decade , an interesting problem has been attracting much attention , i.e. , the competing glauber - type and kawasaki - type dynamics .
we briefly explain the physical ground : glauber s mechanism with probability @xmath0 is used to simulate the contact of the system with a heat bath and favors a low energy state
. meanwhile kawasaki s mechanism with probability @xmath1 simulates an external energy flux and , naturally , favors a high energy state
. then there will a competition between the two dynamics .
this competing mechanism has been applied to the spin models@xcite , and interesting self - organization has been reported .
all these works have considered regular lattice , and to extend the investigation to the small - world networks should be of interest . at the same time , such a study may also have theoretical meaning . in zhu s work@xcite
, it has been found that the swn effect may have different types of influence on the glauber - type dynamics and the kawasaki - type dynamics . by focusing on the competition of the dynamics , it is easy for us to further understand the influence of the swn effect and highlight the disparities between the dynamics . in this article
we investigate the swn effect on the competing dynamics , and we demonstrate the conclusions by comparing the phase diagrams obtained with and without the swn effect . before we present the calculations , we first describe the prototypes of swns that are used in the present study , and then briefly review the generalized competing mechanism . _ the swn models _ : after the first prototype of swn was proposed by watts and strogatz@xcite , there appear a variety of model systems that can be categorized into the two groups , i.e. , with the total number of links increased or conserved
. correspondingly , in the present research we study two specific , while representative models described below : ( 1 ) in a one - dimensional loop , for example , each randomly selected pair of vertices are additionally connected with probability @xmath2 ; ( 2 ) the vertices are visited one after another , and each of its links in the clockwise sense is left in place with probability @xmath3 and is reconnected to a randomly selected other node with probability @xmath4 .
networks of higher dimensions can be similarly built .
we call the first model adding - type small - world network ( a - swn ) and the second one rewiring - type network ( r - swn ) . _ the generalized competing mechanism _ : glauber s flip mechanism and kawasaki s exchange mechanism were originally presented for the ising model .
recently they have been generalized to single - spin transition mechanism@xcite and spin - pair redistribution mechanism@xcite , respectively , which can be applied to arbitrary spin systems .
correspondingly there is the generalized competing dynamics@xcite , which provides a basis for the discussions in this article . with the competing mechanism
, the master equation can be written as @xmath5 where @xmath6 denotes the single - spin transition with probability @xmath0 and @xmath7 denotes the spin - pair redistribution with probability @xmath1 . for your convenience
we list in sec .
ii and iii some important equations of the transition mechanism and the redistribution mechanism , the details of which can be found in respective references .
in this article we study the generalized competing dynamics in the framework of the kinetic gaussian model , which is mathematically tractable .
it has been revealed in ref .
@xcite that the swn effect on the gaussian model is of the mean - field nature . considering this , in the present study a simplified method
is adopted , although a rigorous treatment is possible . in sec .
[ sec.2 ] we study the competing mechanics on a - swn and in sec .
[ sec.3 ] , we extend the discussions to r - swn .
section [ sec.4 ] is the summarization with some discussions .
in this section we present our discussion of the competing dynamics on adding - type small - world networks ( the definition see the introduction ) . in subsection [ subsec.2.1 ] through [ subsec.2.3 ]
we give the formulation of the glauber - type mechanism , the kawasaki - type mechanism and the competing mechanism , and then in subsection [ subsec.2.4 ] the competing mechanism is applied to the three - dimensional gaussian model .
with the glauber - type single - spin transition mechanism@xcite on a - swn , the master equation can be written as @xmath8 , \label{3}\ ] ] where the transition probability @xmath9 , q_{i}=\sum\limits_{\hat{\sigma}_{i}}\exp \left [ -\beta { \cal h}\left ( \left\ { \sigma _ { j\neq i}\right\ } , \hat{\sigma}% _ { i}\right ) \right ] .\ ] ] with equation ( [ 3 ] ) , one can continue to write that@xcite , @xmath10 p\left ( \{\sigma \},t\right ) , \label{q - glauber}\ ] ] where @xmath11 with the kawasaki - type spin - pair redistribution mechanism on a - swn , the master equation can be written as@xcite @xmath12 \\ & & + \frac{1}{2}p_{a}\sum\limits_{j}\sum\limits_{l\neq j}\sum\limits_{\hat{% \sigma}_{j}\hat{\sigma}_{l}}\left [ -w_{jl}(\sigma _ { j}\sigma _ { l}\rightarrow \hat{\sigma}_{j}\hat{\sigma}_{l})p(\{\sigma \},t)\right . \\ & & \left .
+ w_{jl}(\hat{\sigma}_{j}\hat{\sigma}_{l}\rightarrow \sigma
_ { j}\sigma _ { l})p(\{\sigma _ { j\neq i},\sigma _ { l\neq k}\},\hat{\sigma}_{j},% \hat{\sigma}_{l},t)\right ] , \end{aligned}\ ] ] where the redistribution mechanism @xmath13 , \label{5}\ ] ] @xmath14.\ ] ] with the master equation above , we can further get that @xmath15 p(\{\sigma \},t ) \nonumber \\ & & + p_{a}\{-(n-1)q_{k}(t)+\sum\limits_{\{\sigma \}}\left [ \sum\limits_{l\neq k}\sum\limits_{\hat{\sigma}_{k},\hat{\sigma}_{l}}\hat{\sigma}% _ { k}w_{kl}\left ( \sigma _ { k}\sigma _ { l}\rightarrow \hat{\sigma}_{k}\hat{% \sigma}_{l}\right ) \right ] p(\{\sigma \},t)\ } , \label{q - kawasaki}\end{aligned}\ ] ] where @xmath16 denotes the summation taken over nearest neighbors . naturally , both glauber s dynamics and kawasaki s dynamics favor a lower energy state .
however , when the system under study is in contact with a heat bath while exposed to an external energy flux , one requires a competition between a process that favors lower system energy and another process that favors higher system energy .
usually , glauber s mechanism is used to simulate the contact of the system with a heat bath and favors a lower energy state .
meanwhile kawasaki s mechanism can be modified in order to simulate an external energy flux that drives the system towards higher energy .
this can be achieved by switching @xmath17 to @xmath18 , or @xmath19 to @xmath20 , and modifying the redistribution probability accordingly , i.e. @xmath21\rightarrow w_{jl}\propto \exp [ + \beta h]\ ] ] this means that the competition between glauber s mechanism and kawasaki s mechanism is actually that between ferromagnetism and antiferromagnetism .
based on the above considerations , we use the competing mechanism to simulate the dynamics of a system in contact with a heat bath and exposed to an external energy flux simultaneously .
the master equation can be written as ( eq . ( [ 1 ] ) ) , @xmath22 where @xmath23 and @xmath24 \\ & & + \frac{1}{2}p_{a}\sum\limits_{j}\sum\limits_{l\neq j}\sum\limits_{\hat{% \sigma}_{j}\hat{\sigma}_{l}}\left [ -w_{jl}(\sigma _ { j}\sigma _ { l}\rightarrow \hat{\sigma}_{j}\hat{\sigma}_{l})p(\{\sigma \},t)\right .
\\ & & \left
. + w_{jl}(\hat{\sigma}_{j}\hat{\sigma}_{l}\rightarrow \sigma
_ { j}\sigma _ { l})p(\{\sigma _ { j\neq i},\sigma _ { l\neq k}\},\hat{\sigma}_{j,}% \hat{\sigma}_{l},t)\right ] .\end{aligned}\ ] ] from the master equation , we can obtain the evolving equation of single spins , @xmath25 the first term describes the glauber - type dynamics which is used to simulate the influence of the energy flux and @xmath26 is given by eq .
( [ q - glauber ] ) .
the second term describes the kawasaki - type dynamics which is used to simulate the influence of the energy flux .
as mentioned above , we require it to favor higher energy state . however , the redistribution probability , eq .
( [ 5 ] ) , clearly favors a lower energy state . considering this
, we can use the expression of eq .
( [ q - kawasaki ] ) for @xmath27 , if we simply switch the sign before @xmath17 in the redistribution probability .
now we turn to the gaussian model built on an a - swn , of which the hamiltonian can be written as @xmath28 where @xmath29 and @xmath30 .
we only discuss the case of @xmath31 , namely , @xmath32 .
this case corresponds to the ferromagnetic system .
the expression ( [ a - swn ] ) is obtained by taking into consideration the fact that the influence of the system as a whole on individual spins is of the mean - field nature@xcite ( for related discussions see ref .
the spins can take any real value from @xmath33 to @xmath34 .
the probability of finding a given spin between @xmath35 and @xmath36 is assumed to be the gaussian - type distribution , @xmath37 @xmath38 , where @xmath39 is a distribution constant independent of temperature .
therefore , the summation of the spin value turns into an integration , and we can further obtain from eqs .
( [ q - glauber ] ) and ( [ q - kawasaki ] ) , @xmath40 where @xmath41 , and @xmath42 } b\left\ { \left [ ( q_{i+1,j , k}-q_{i , j , k})-(q_{i , j , k}-q_{i-1,j , k})\right ] \right . \nonumber \\ & & \left .
+ \left [ ( q_{i , j+1,k}-q_{i , j , k})-(q_{i , j , k}-q_{i , j-1,k})\right ] + \left [ ( q_{i , j , k+1}-q_{i , j , k})-(q_{i , j , k}-q_{i , j , k-1})\right ] \right\ } \nonumber \\ & & + \frac{\left ( -k\right ) } { 2\left [ b+(-k)\right ] } \left
[ 2(2q_{i-1,j , k}-q_{i-1,j+1,k}-q_{i-1,j-1,k})+(2q_{i-1,j , k}-q_{i , j , k}-q_{i-2,j , k})\right .
\nonumber \\ & & + 2(2q_{i+1,j , k}-q_{i+1,j+1,k}-q_{i+1,j-1,k})+(2q_{i+1,j , k}-q_{i , j , k}-q_{i+2,j , k } ) \nonumber \\ & & + 2(2q_{i , j-1,k}-q_{i , j-1,k+1}-q_{i , j-1,k-1})+(2q_{i , j-1,k}-q_{i , j , k}-q_{i , j-2,k } ) \nonumber \\ & & + 2(2q_{i , j+1,k}-q_{i , j+1,k+1}-q_{i , j+1,k-1})+(2q_{i , j+1,k}-q_{i , j , k}-q_{i , j+2,k } ) \nonumber \\ & & + 2(2q_{i , j , k-1}-q_{i+1,j , k-1}-q_{i,-1j , k-1})+(2q_{i , j,-1k}-q_{i , j , k}-q_{i , j-2,k } ) \nonumber \\ & & \left .
+ 2(2q_{i , j , k+1}-q_{i+1,j , k+1}-q_{i-1,j , k+1})+(2q_{i , j , k+1}-q_{i , j , k}-q_{i , j , k+2})\right ] \nonumber \\ & & -\frac{p_a}2(n-1)\left [ q_{ijk}(t)-m(t)\right ] + p_a(n-1)\frac{\left ( -k\right ) } { 2b } \nonumber \\ & & \times \sum\limits_{\omega = \pm 1}\left [ \left ( q_{i+\omega , j , k}-m(t)\right ) + \left ( q_{i , j+\omega , k}-m(t)\right ) + \left ( q_{i , j , k+\omega } -m(t\right ) \right ] , \ ( b > k>0 ) \label{7}\end{aligned}\ ] ] where @xmath43 is required by the convergence of the integration
. now we determine the system behavior by studying the following order parameters .
first , from eq .
( [ 6 ] ) we obtain @xmath44 m(t ) , \label{m}\end{aligned}\ ] ] @xmath45 means that the kawasaki - type dynamics does not change the value of @xmath46 .
second , we define @xmath47 and similarly we obtain @xmath48 where @xmath49 + \frac{k}{b}% ( n-1)p_{a}(-1)^{i+j+k}m(t),\ ] ] and @xmath50 \right . \\ & & + \left [ ( -q_{i , j+1,k}^{\prime } -q_{i , j , k}^{\prime } ) -(q_{i , j , k}^{\prime } + q_{i , j-1,k}^{\prime } ) \right ] + \left [ ( -q_{i , j , k+1}^{\prime } -q_{i , j , k}^{\prime } ) -(q_{i , j , k}^{\prime } + q_{i , j , k-1}^{\prime } ) \right ] \ } \\ & & -\frac{k}{2(b - k)}\left [ 2(-2q_{i-1,j , k}^{\prime } -q_{i-1,j+1,k}^{\prime } -q_{i-1,j-1,k}^{\prime } ) + ( -2q_{i-1,j , k}^{\prime } -q_{i , j , k}^{\prime } -q_{i-2,j , k}^{\prime } ) \right .
\\ & & + 2(-2q_{i+1,j , k}^{\prime } -q_{i+1,j+1,k}^{\prime } -q_{i+1,j-1,k}^{\prime } ) + ( -2q_{i+1,j , k}^{\prime } -q_{i , j , k}^{\prime } -q_{i+2,j , k}^{\prime } ) \\ & & + 2(-2q_{i , j-1,k}^{\prime } -q_{i , j-1,k+1}^{\prime } -q_{i , j-1,k-1}^{\prime } ) + ( -2q_{i , j-1,k}^{\prime } -q_{i , j , k}^{\prime } -q_{i , j-2,k}^{\prime } ) \\ & & + 2(-2q_{i , j+1,k}^{\prime } -q_{i , j+1,k+1}^{\prime } -q_{i , j+1,k-1}^{\prime } ) + ( -2q_{i , j+1,k}^{\prime } -q_{i , j , k}^{\prime } -q_{i , j+2,k}^{\prime } ) \\ & & + 2(-2q_{i , j , k-1}^{\prime } -q_{i+1,j , k-1}^{\prime } -q_{i,-1j , k-1})+(-2q_{i , j,-1k}^{\prime } -q_{i , j , k}^{\prime } -q_{i , j-2,k}^{\prime } ) \\ & & \left .
+ 2(-2q_{i , j , k+1}^{\prime } -q_{i+1,j , k+1}^{\prime } -q_{i-1,j , k+1}^{\prime } ) + ( -2q_{i , j , k+1}^{\prime } -q_{i , j , k}^{\prime } -q_{i , j , k+2}^{\prime } ) \right ] \\ & & -\frac{p_{a}(n-1)}{2}\left\ { q_{ijk}^{^{\prime } } ( t)-(-1)^{i+j+k}m(t)\right . \\ & & \left .
( q_{i+\omega , j , k}^{^{\prime } } ( t)+q_{i , j+\omega , k}^{^{\prime } } ( t)+q_{i , j , k+\omega } ^{^{\prime } } ( t))-% \frac{6k}{b}(-1)^{i+j+k}m(t)\right\ } .\end{aligned}\ ] ] performing the summation over the indices @xmath51 ,
@xmath52 , and @xmath53 , we get @xmath54 \right\ } m^{\prime } ( t ) .
\label{m'}\ ] ] the solutions of equations ( [ m ] ) and ( [ m ] ) are @xmath55 t\right\ } , \left ( k>0\right ) \label{m - s}\ ] ] and @xmath56 \left ( 1-\frac{k}{k_{c}^{0}}\right ) \right ) t\right\ } , \nonumber \\ & & \left ( \begin{array}{c } if\ p=1,then\ k>0 \\ if\ p\neq 1,then\ ( b > k>0 ) \end{array } \right ) \label{m'-s}\end{aligned}\ ] ] where @xmath57 is the critical point of the three - dimensional gaussian model without the swn effect .
\(1 ) when the swn effect does not exist and @xmath58 , @xmath59 , \label{m - regular}\ ] ] @xmath60 t\right\ } . \label{m'-regular}\ ] ] analyzing the long - time system behavior , we find that : ( 1.a ) for the case of @xmath61 we have both vanishing @xmath46 and @xmath62 , which correspond to the paramagnetic phase .
( 1.b ) for the case of @xmath63 @xmath64 , and @xmath65 , \tau = \frac{1}{% 1-k / k_{c}^{0}}\rightarrow \infty .\ ] ] the critical slowing down of the order parameter @xmath66 will appear at the critical point @xmath67 .
( 1.c ) for the case of @xmath68 and @xmath69 we have nonvanishing @xmath46 and vanishing @xmath70 , which correspond to the ferromagnetic phase .
( 1.d ) for the case of @xmath71 we have both nonvanishing @xmath46 and @xmath70 . with both of the order parameters nonvanishing
, this phase can not be simply identified as ferromagnetic or antiferromagnetic .
we name it as heterophase , and from eqs .
( [ m - regular ] ) and ( [ m-regular ] ) we can see that the system behavior strongly depends on the initial condition and the temperature .
( 1.e ) for the case of @xmath72 we could obtain @xmath73 , and when @xmath74 ( with pure glauber - type dynamics simulating the contact with a heat bath ) , we have @xmath75 .
however , with @xmath76 ( with an external energy flow ) , because the antiferromagnetic kawasaki - type dynamics is limited by the condition ( @xmath43 ) to ensure the convergence of the integration , we can not obtain the value of @xmath62 . as a result ,
we can not theoretically obtain the system behavior in the temperature region @xmath77 . the phase diagram is shown in fig .
\(2 ) now the a - swn effect is introduced and @xmath78 .
we suppose @xmath79 , and then we have @xmath80 t\right\ } , \ ] ] and @xmath81 \left ( 1-\frac{k}{k_{c}^{0}}\right ) \right ) t\right\ } .\ ] ] analyzing the long - time asymptotic behavior , we can similarly get : ( 2.a ) for the case of @xmath82 the condition leads to vanishing @xmath46 and vanishing @xmath83 , which correspond to the paramagnetic phase .
( 2.b ) for the case of @xmath84 the critical slowing down of the system will appear .
( 2.c ) for the case of @xmath85 we have nonvanishing @xmath46 and vanishing @xmath70 , which correspond to the ferromagnetic phase .
( 2.d ) for the case of @xmath86 we have both nonvanishing @xmath46 and @xmath70 , which correspond to the heterophase .
( 2.e ) for the case of @xmath72 for the same reason ( see ( 1.e ) ) , we can not obtain theoretically the system behavior . the phase diagram is shown in fig .
now we investigate the competing dynamics on rewiring - type small - world networks ( the definition see the introduction ) , and the sequence of the content is the same as that in the previous section . with the glauber - type mechanism , the master equation has the same form as eq .
( [ 3 ] ) , and the single - spin evolving equation is also given by eq .
( [ q - glauber ] ) . with the kawasaki - type mechanism ,
the master equation can be written as @xmath87 \\ & & + \frac{dp_r}{n-1}\sum\limits_j\sum\limits_{l\neq j}\sum\limits_{\hat{\sigma% } _ j\hat{\sigma}_l}\left [ -w_{jl}(\sigma _ j\sigma _ l\rightarrow
\hat{\sigma}_j% \hat{\sigma}_l)p(\{\sigma \},t)\right . \\ & & \left . + w_{jl}(\hat{\sigma}_j\hat{\sigma}_l\rightarrow \sigma _ j\sigma _ l)p(\{\sigma _ { j\neq i},\sigma _ { l\neq k}\},\hat{\sigma}_{j,}\hat{\sigma}% _ l , t)\right ] .\end{aligned}\ ] ] the redistribution probability @xmath88 is given by eq .
( [ 5 ] ) , and the single - spin evolving equation @xmath89 p(\{\sigma \},t)\right\ } \\ & & + \frac{dp_r}{n-1}\left\ { -(n-1)q_k(t)+\sum\limits_{\{\sigma \}}\left [ \sum\limits_{l\neq k}\sum\limits_{\hat{\sigma}_k,\hat{\sigma}_l}\hat{\sigma}% _ kw_{kl}(\sigma _ k\sigma _ l\rightarrow \hat{\sigma}_k\hat{\sigma}_l)\right ] p(\{\sigma \},t)\right\ } .\end{aligned}\ ] ] with the competing mechanism , the evolution of @xmath90 is once again described by an equation having the same form as eq .
( [ 6 ] ) @xmath91 which consists of two terms , one corresponding to the glauber - type dynamics and the other corresponding to the kawasaki - type dynamics . in the following we turn to the gaussian model . for the three - dimensional gaussian model built on a r - swn , the hamiltonian can be written as @xmath92 in the single - spin evolving equation with the competing mechanism , the term that corresponds to the glauber - type dynamics is now given by @xmath93 and the term that corresponds to the kawasaki - type dynamics can be written as ( similarly , we switch @xmath17 to -@xmath17 , and @xmath94 to @xmath20 ) @xmath95}b\left\ { \left [ ( q_{i+1,j , k}-q_{i , j , k})-(q_{i , j , k}-q_{i-1,j , k})\right ] \right . \\ & & + \left . \left
[ ( q_{i , j+1,k}-q_{i , j , k})-(q_{i , j , k}-q_{i , j-1,k})\right ] + \left [ ( q_{i , j , k+1}-q_{i , j , k})-(q_{i , j , k}-q_{i , j , k-1})\right ] \right\ } \\ & & -\frac{k(1-p_r)}{2\left [ b - k(1-p_r)\right ] } \left [ 2\left ( 2q_{i-1,j , k}-q_{i-1,j+1,k}-q_{i-1,j-1,k}\right ) + \left ( 2q_{i-1,j , k}-q_{i , j , k}-q_{i-2,j , k}\right ) \right .
\\ & & + 2\left ( 2q_{i+1,j , k}-q_{i+1,j+1,k}-q_{i+1,j-1,k}\right ) + \left ( 2q_{i+1,j , k}-q_{i , j , k}-q_{i+2,j , k}\right ) \\ & & + 2\left ( 2q_{i , j-1,k}-q_{i , j-1,k+1}-q_{i , j-1,k-1}\right ) + \left ( 2q_{i , j-1,k}-q_{i , j , k}-q_{i , j-2,k}\right ) \\ & & + 2\left ( 2q_{i , j+1,k}-q_{i , j+1,k+1}-q_{i , j+1,k-1}\right ) + \left ( 2q_{i , j+1,k}-q_{i , j , k}-q_{i , j+2,k}\right ) \\ & & + 2\left ( 2q_{i , j , k-1}-q_{i+1,j , k-1}-q_{i,-1j ,
k-1}\right ) + \left ( 2q_{i , j,-1k}-q_{i , j , k}-q_{i , j-2,k}\right ) \\ & & + 2\left ( 2q_{i , j , k+1}-q_{i+1,j , k+1}-q_{i-1,j ,
k+1}\right ) + \left ( 2q_{i , j , k+1}-q_{i , j , k}-q_{i , j , k+2}\right ) \\ & & -\frac 32p_r\left\ { q_{ijk}(t)-m(t)+\frac{k(1-p_r)}b\left [ \sum\limits_{\omega = \pm 1}\left ( q_{i+\omega , j , k}+q_{i , j+\omega , k}+q_{i , j , k+\omega } \right ) -6m\left ( t\right ) \right ] \right\ } , \\ & & \left ( b > k(1-p_r)>0\right)\end{aligned}\ ] ] where @xmath96 is required by the convergency of the integration . from the single - spin evolving equation with the competing mechanism ,
we can obtain @xmath97 and @xmath98 \right .
\nonumber \\ & & \left .
+ ( 1-p)\left\ { 6\frac{6k\left ( 1-p_r\right ) -b}{b - k\left ( 1-p_r\right ) } -\frac{3p_r}2\left [ 1-\frac{6k}b\left ( 1-p_r\right ) \right ] \right\ } \right ) m^{\prime } ( t ) .
\label{12_1}\end{aligned}\ ] ] the solutions of eqs .
( [ 11 ] ) and ( [ 12_1 ] ) are , respectively , @xmath99 , \
\left ( k>0\right ) \label{13}\ ] ] and @xmath100 \right .
\nonumber \\ & & \left .
+ ( 1-p)\left [ \frac{36}{6-k(1-p_r)/k_c^0}+\frac{3p_r}2\right ] \left [ 1-(1-p_r)\frac k{k_c^0}\right ] \right\ } \right ) , \nonumber \\ & & \left ( \begin{array}{c } if\ p=1,then\ k>0 \\
if\ p\neq 1,then\ \left ( b > k(1-p_r)>0\right ) \end{array } \right ) , \label{14}\end{aligned}\ ] ] where @xmath101 is the critical point without the swn effect .
\(1 ) when no swn effect is considered and @xmath102 , the phase diagram is given by fig .
1 ( a ) .
\(2 ) now we introduce the r - swn effect and set @xmath103 , and then eqs .
( [ 13 ] ) and ( [ 14 ] ) become @xmath104 , \label{m - solution - rswn}\ ] ] and @xmath105 \right\ } .
\label{15}\ ] ] analyzing the long - time asymptotic behavior , we find that : ( 2.a ) for the case of @xmath106 we have both vanishing @xmath46 and @xmath62 , and this corresponds to the paramagnetic phase .
( 2.b ) for the case of @xmath107 we shall observe the critical slowing down of the system .
( 2.c ) for the case of @xmath108 and @xmath109 the condition leads to nonvanishing @xmath46 and vanishing @xmath83 , which correspond to the ferromagnetic phase .
( 2.d ) for the case of @xmath110 we have both nonvanishing @xmath46 and @xmath70 , which correspond to the heterophase .
( 2.e ) for the case of @xmath111 for the same reason ( see sec .
( 1.e ) ) , we can not give theoretically the behavior of the system in this region . the phase diagram is shown in fig .
comparing fig .
1(a ) , ( b ) and ( c ) , we find both similarities and differences .
the phase diagrams have similar structures while the boundaries may be shifted by the swn effect . in the following
we briefly explain our observations and discuss the nature of the competing mechanism and the swn effect .
first we notice that the phase diagrams are all separated into four regions , with two special temperatures : one is the critical temperature @xmath112 , another is the limit temperature @xmath113 required by the convergence of the integration .
when the swn effect does not exist , @xmath114 is the critical point of the gaussian model , where @xmath39 is the gaussian distribution constant and @xmath115 is the space dimension , and @xmath116 ( since @xmath117 * * is the region where we can assure the convergence of the integration in the kawasaki - type dynamics of the antiferromagnetic system ) . when the a - swn effect is introduced , @xmath118 and @xmath119 ( in the example studied @xmath120 , @xmath121 , and @xmath122 ) . * * when the r - swn effect is introduced , @xmath123and @xmath124 ( in the example studied @xmath120 , @xmath125 , and @xmath126 ) .
now we describe the phase behavior : above the critical temperature @xmath127 , one can only observe a disordered state , namely the paramagnetic phase , because of the dominating heat noise .
below critical temperature @xmath128 , the system begins to show some kind of order , and the system behavior is determined by the competition between the glauber - type mechanism which favors a ferromagnetic state , and the kawasaki - type mechanism which favors an antiferromagnetic state .
the result of the competition is subject to two factors , the probability of each mechanism and the temperature .
however , the kawasaki - type dynamics describing an antiferromagnetic gaussian system is limited by the condition @xmath129 , which is necessary for the convergence of the integration . as a result , in the region @xmath130 we can not theoretically obtain the system behavior .
now we discuss how and why the two special temperatures @xmath131 and @xmath132 may be affected by the swn effect .
when the gaussian model is built on a r - swn , the temperature @xmath133 remains the same .
this is because , as is clear in all of the phase diagrams , @xmath131 is determined by the competition between the heat noise and the glauber - type mechanism , which favors a lower energy state
. one important characteristic of the gaussian model is that this temperature can be further determined by the average coordination number .
for example , without the swn effect this temperature is given by @xmath134 , where @xmath135 is the average coordination number of a @xmath115-dimensional lattice . on a r - swn ,
the average coordination number is not changed when a portion of the regular links are rewired , and thus @xmath131 also remains unchanged .
on the other hand the temperature @xmath136 is lower than the value on a regular network . on r - swn long range links
have made long range spin - pair redistribution possible , and therefore the influence of the kawasaki - type mechanism is enhanced .
thus it is not difficult to understand why on r - swn there is a lower value of the temperature @xmath136 .
when the gaussian model is built on an a - swn , the temperature @xmath137 becomes higher .
as is explained above , this temperature is determined by the average coordination number , which is @xmath138 on a three dimensional a - swn with @xmath139 . at the same time
the temperature @xmath136 is unchanged .
we may explain that it is because the two dynamics are both strengthened , i.e. , by the larger coordination number and the long range spin - pair redistribution , and the temperature that characterizes a certain counterbalance remains the same .
to summarize , in this article we investigate the competing glauber - type and kawasaki - type dynamics on two typical three - dimensional small - world networks , adding - type ( a - swn ) and rewiring -type ( r - swn ) , in the framework of the gaussian model .
we get the evolution of the order parameters , @xmath66 and @xmath140 , and by analyzing the long time asymptotic behavior we draw the phase diagrams . with the competing mechanism ,
there exist two easily identifiable special temperatures .
the influence of the long range links is analyzed : for the glauber - type dynamics , more important is the altered average coordination number while the kawasaki - type dynamics is enhanced by the long range spin interaction and redistribution .
1 the phase diagrams of the three - dimensional gaussian model on ( a ) regular lattice , ( b ) a - swn , and ( c ) r - swn .
the regions para , ferro , and hetero correspond to the paramagnetic , ferromagnetic , and heterophase phase , respectively , while the part of oblique lines belong to unknown region . | in this article , we investigate the competing glauber - type and kawasaki - type dynamics with small - world network ( swn ) effect , in the framework of the gaussian model . the glauber - type single - spin transition mechanism with probability @xmath0
simulates the contact of the system with a heat bath and the kawasaki - type dynamics with probability @xmath1 simulates an external energy flux .
two different types of swn effect are studied , one with the total number of links increased and the other with it conserved .
the competition of the dynamics leads to an interesting self - organization process that can be characterized by a phase diagram with two identifiable temperatures . by studying the modification of the phase diagrams , the swn effect on the two dynamics
is analyzed .
for the glauber - type dynamics , more important is the altered average coordination number while the kawasaki - type dynamics is enhanced by the long range spin interaction and redistribution .
keywords : small - world network effect ; classical spin model ; competing glauber- and kawasaki - type dynamics ; phase diagram |
in a fixed homology class of a complex surface , there are at most finitely many complex curves up to smooth isotopy . in contrast , there exist examples of homology classes in symplectic @xmath5-manifolds that are represented by infinitely many non - isotopic connected symplectic surfaces . in @xcite fintushel and
stern proved that @xmath2\in h_2(x ; { \mathbb{z}})$ ] is such a class provided that @xmath6 is an even integer and @xmath7 is a symplectic @xmath8-embedded torus , i.e. a homologically nontrivial torus of self - intersection @xmath9 with first homology generated by two circles which bound disks of self - intersection @xmath10 in @xmath11 , in a simply - connected symplectic @xmath5-manifold @xmath11 ( e.g. we can take @xmath11 to be an elliptic surface @xmath3 and @xmath7 to be a regular fiber ) . in this paper
we generalize their result by constructing infinitely many non - isotopic symplectic tori in @xmath2 \in h_2(x ; { \mathbb{z}})$ ] for each @xmath12 ( even or odd ) in case when @xmath7 is an essentially embedded ( cf.@xmath13definition [ def : essential ] ) symplectic torus in a symplectic @xmath5-manifold @xmath11 satisfying @xmath4 ^ 2 = 0 $ ] and @xmath14 .
moreover , as a consequence of our calculation , it will be apparent that no two of these tori are equivalent under the action of _
diff_@xmath15 , the group of self - diffeomorphisms of @xmath11 . in some special cases
our result can be slightly improved to include @xmath16 $ ] as in the following theorem .
[ theorem:1 ] for any pair of integers @xmath0 and @xmath1 , there exists an infinite family of mutually non - isotopic symplectic tori representing the homology class @xmath2 $ ] of an elliptic surface @xmath3 , where @xmath4 $ ] is the homology class of the fiber .
it should be noted that in @xcite , vidussi has constructed such families for every positive multiple of the fiber class in @xmath3 provided that @xmath17 .
however , it still remains an interesting open problem to construct infinite families of non - isotopic symplectic tori in the fiber class @xmath4 $ ] of @xmath18 and @xmath19 .
we should also note that our construction can be carried out for a more general class of 4-manifolds than those in @xcite . in all of the above examples , families of braids
are used to construct symplectic tori and seiberg - witten theory is used to distinguish them .
the family of braids we use in this paper are , in some sense , the simplest possible and this allowed us not only to calculate the seiberg - witten invariants of the corresponding @xmath5-manifolds completely , but also to reprove and extend the results in @xcite and @xcite ( except for the case of fiber class itself in @xmath3 for @xmath17 ) . from our construction
, it will be apparent that any other collection of reasonably complicated braids will also give rise to many more new examples of non - isotopic symplectic tori . in the next section
we construct a family of tori , @xmath20 , in @xmath21 which could be identified with a regular tubular neighborhood of a torus of self - intersection @xmath9 in a @xmath5-manifold .
then we show how these tori could be symplectically embedded in @xmath3 . in section [ section : sw ]
we calculate the seiberg - witten invariants of symplectic @xmath5-manifolds obtained as the fiber sum of @xmath3 and @xmath22 along @xmath23 and a fiber in @xmath22 . as a consequence of these computations
we prove theorem [ theorem:1 ] . in section [ section : generalization ] we explain how theorem [ theorem:1 ] could be generalized to a larger class of @xmath5-manifolds ( see theorem [ theorem:2 ] ) .
we first construct an infinite family of non - isotopic symplectic tori @xmath24 representing @xmath2 $ ] in @xmath18 , and show that these remain non - isotopic after fiber - summing with @xmath25 .
i.e. we will construct an infinite family of tori that embed non - isotopically into @xmath3 for any @xmath0 .
even though we are unable to show that @xmath26 is not isotopic to any torus in @xmath27 in general , we will still include it in our construction and computations for the sake of completeness .
for any pair of integers @xmath28 and @xmath1 , consider the @xmath29-strand braid @xmath30 in figure [ fig : braid ] .
the closed braid @xmath31 is the ` simplest ' knot in @xmath32 with @xmath33 crossings , which is usually denoted by @xmath34 in the literature ( see e.g. @xcite or @xcite ) . for @xmath28 and @xmath1 ]
let @xmath36 denote the two - component link in @xmath32 which is the union of the closed braid @xmath31 and its axis @xmath37 .
see figure [ fig : l24 ] for a picture of @xmath38 . ]
next we review the generalization of the link surgery construction of fintushel and stern @xcite by vidussi @xcite . for an @xmath39-component link @xmath40 ,
choose a homology basis of simple curves @xmath41 such that the intersection of @xmath42 and @xmath43 is 1 in the boundary of the link exterior .
we define the link surgery manifold @xmath44\hspace{-20pt}\bigcup_{f_i\times\partial d^2=(s^1\times \alpha_i)\times\beta_i}\hspace{-20pt } [ s^1\times(s^3\setminus \nu l)]\ , , \ ] ] where @xmath45 denotes the tubular neighborhoods . here
, the gluing diffeomorphisms between the boundary 3-tori identify the torus fiber @xmath46 of @xmath47 with @xmath48 , and act as complex conjugation on the last remaining @xmath49 factor .
strictly speaking , our gluing construction depends on our choice of basis @xmath50 , but we will suppress this dependence in our notation . as before let @xmath37 denote the unknot which is the axis of the closed braid @xmath51 . choose the homology basis @xmath52 for the complement @xmath53 , where @xmath54 and @xmath55 denote the meridian and longitude , respectively .
[ lemma : e(1 ) ] @xmath56 .
note that the exterior of the unknot @xmath53 is diffeomorphic to a solid torus @xmath57 .
hence there is a diffeomorphism between the tubular neighborhood of a regular fiber @xmath58 and the cartesian product @xmath59 $ ] .
now consider @xmath31 as a knot inside the solid torus @xmath53 .
define @xmath60 , which is a torus embedded in @xmath61 \approx \nu f \subset e(1)_a\,$ ] .
[ lemma : symplectic ] the torus @xmath62 is a symplectic submanifold and represents the homology class @xmath2 $ ] . first note that @xmath31 is homologous to @xmath63 in @xmath64 .
since the fiber @xmath7 is identified with @xmath65 in @xmath66 , the torus @xmath67 is homologous to @xmath68 in @xmath69 . to show that @xmath67 is symplectic
, we can proceed as in @xcite and @xcite .
the key step is once again verifying that the knot @xmath70 is transverse to the disks @xmath71 .
next we show that the torus @xmath67 can be embedded in @xmath3 for any @xmath0 . for @xmath72 , we can write @xmath3 as the fiber sum , @xmath73\hspace{-20pt } \bigcup_{f'_1\times \partial d^2 = f_2\times \partial d^2 } \hspace{-20pt } [ e(n-1)\setminus \nu f_2]\;,\ ] ] where @xmath74 is a regular torus fiber lying outside @xmath75 $ ] .
we immediately see that @xmath67 can be embedded into the first fiber summand and hence into @xmath3 . without causing too much confusion
, we will continue to denote this embedded torus in @xmath3 by @xmath76 .
note that lemma [ lemma : symplectic ] continues to hold for the torus @xmath77 .
since the torus @xmath77 has self - intersection zero , we can form the fiber sum , @xmath78 \hspace{-20pt } \bigcup_{t_{p , q}\times\partial d^2=f\times \partial d^2 } \hspace{-20pt } [ e(r)\setminus \nu f]\ ; .\ ] ] the gluing diffeomorphism @xmath79 identifies @xmath67 with @xmath7 and acts as complex conjugation on the @xmath80 factor .
note that when @xmath72 there is a canonical framing of @xmath67 in @xmath3 that is inherited from a fixed framing of @xmath67 in @xmath18 by our fiber sum description ( [ fiber - sum ] ) .
we remark that the gluing formula in @xcite can not be used to compute the seiberg - witten invariant of the fiber sum ( [ fiber - sum2 ] ) , since the triple @xmath81,[e(r)\setminus \nu f],\varphi)$ ] is not admissible in the sense of definition 8 in @xcite . however , the key observation is that this fiber sum can be expressed as a link surgery manifold .
[ lemma : main ] @xmath82 . for the construction of @xmath83 ,
we choose the homology basis @xmath84 , and @xmath85 .
the identification with the fiber sum is now immediate .
to prove theorem [ theorem:1 ] , we need to show that the family of tori @xmath86 we constructed are mutually non - isotopic for fixed @xmath29 .
it is enough to show that the corresponding family of 4-manifolds @xmath87 are mutually non - diffeomorphic , since any isotopy between the tori @xmath88 will translate into diffeomorphism between corresponding 4-manifolds @xmath89 .
we shall compute their seiberg - witten invariants to show that @xmath90 are mutually non - diffeomorphic .
let @xmath91 denote the alexander polynomial of the link @xmath92 , where the variable @xmath93 corresponds to the axis @xmath37 .
[ lemma : alexander ] @xmath94 and when @xmath95 @xmath96 the braid group on @xmath29 strands is generated by the elementary braid transpositions @xmath97 , where @xmath98 denotes the crossing of the @xmath99st strand over the @xmath100th .
note that @xmath101 by theorem 1 in @xcite , we have @xmath102 where @xmath103 denotes the following @xmath104 matrix which differs from the identity matrix @xmath105 only in the three places shown on the @xmath100th row .
@xmath106 when @xmath107 or @xmath108 , the matrix is truncated appropriately to give two non - zero entries in row @xmath109 .
the statement for @xmath110 case is obvious .
we are going to prove the case @xmath12 by induction on @xmath29 .
the main step of this induction is proving that @xmath111 , where @xmath112 .
then we prove that @xmath113 and this implies @xmath114 and hence @xmath115 .
finally , by using the induction assumption and @xmath116 we finish the proof of the lemma .
@xmath117 , where @xmath118 .
note that @xmath119 for @xmath120 so we must have @xmath121 hence it follows that @xmath122 and @xmath123 where @xmath124 denotes the last row of @xmath125 .
when we calculate the determinant of the matrix @xmath126 by expanding along its last column we get the following equality : @xmath127\ , .\ ] ] to prove the above equality , observe that all but the last row of the minor of the matrix @xmath128 corresponding to the entry @xmath129 in the last column are the same as the rows of @xmath130 , and the last row of the minor is @xmath131 times the last row of @xmath132 except for the last entry . in the minor , this entry is @xmath9 , whereas in @xmath130 this entry is @xmath133 ( since equation ( [ eq : one ] ) shows that the last diagonal entry of @xmath134 is @xmath9 ) is @xmath9 , only if @xmath6 .
nevertheless , for @xmath135 , equation ( [ eq : two ] ) could be confirmed by a direct calculation of @xmath136 along the same lines as the calculation of @xmath137 . ] .
this observation is why the determinant of the minor corresponding to @xmath129 is @xmath131 times the difference between the determinant of @xmath132 and the determinant of the minor of @xmath132 obtained by deleting the last row and the last column ( and this minor is nothing but @xmath138 ) .
@xmath139 now follows trivially from equation ( [ eq : two ] ) .
the next step is to calculate @xmath137 . by equation ( [ eq : zero ] )
, we have @xmath140\!\!\begin{array}{l } \\ . \end{array}\ ] ] an easy induction argument shows that @xmath141 and then an explicit calculation gives @xmath142 . as we discussed before , to finish the proof one only needs to check that @xmath143 & + & { } \!\ ! x^{q-1}t^q ( 1-t+t^2- \cdots - t^{2p-3 } ) \;+\ ; x^qt^{2p+q-2}\end{aligned}\ ] ] @xmath144 recall that the seiberg - witten invariant @xmath145 of a 4-manifold @xmath11 ( with @xmath146 ) can be thought of as an element of the group ring of @xmath147 , i.e. @xmath148 $ ] . if we write @xmath149 , then we say that @xmath150 is a seiberg - witten _ basic class _ of @xmath11 if @xmath151 .
[ lemma : sw ] let @xmath152\rightarrow e(n , r)_{l_{p , q}}$ ] be the inclusion map .
let @xmath153,$ ] @xmath154\in h_2(e(n , r)_{l_{p , q } } ; { \mathbb{z}}).$ ] if @xmath155 and @xmath156 , then the seiberg - witten invariant of @xmath157 is @xmath158\ , .\end{aligned}\ ] ] when @xmath159 and @xmath160 , we have @xmath161\ , .\ ] ] when @xmath110 and @xmath162 , we have @xmath163 .\ ] ] recall from @xcite that @xmath164^{-1}-[f])^{n-2}$ ] . from the formulas in @xcite and
, we know that @xmath165^{-1}-[f])\cdot\overline{sw}_{\!\!e(n ) } = \ : ( [ f]^{-1}-[f])^{n-1}\ ] ] from the formulas in @xcite and , we conclude that @xmath166 we also need to identify @xmath167 $ ] with @xmath168 and @xmath169 $ ] with @xmath170 .
this last identification is really necessary since we specifically chose to have @xmath171 in the proof of lemma [ lemma : main ] .
hence we must have @xmath172=q[s^1\times\alpha_1]\in h_2(s^1\times(s^3\setminus \nu l_{p , q}))$ ] .
finally we note that @xmath173 where @xmath91 is presented as in lemma [ lemma : alexander ] . the rest of the proof is an easy exercise which we shall leave to the reader .
[ cor : non - diffeo ] for fixed @xmath156 and @xmath0 , the @xmath5-manifolds @xmath174 are mutually non - diffeomorphic .
we note that the total number of the @xmath175 basic classes are invariant under any diffeomorphism . as a consequence of lemma [ lemma : sw ] , the total number of basic classes of our @xmath5-manifolds depend on @xmath176 .
in fact , for @xmath156 and @xmath155 , the total number of basic classes of @xmath177 is @xmath178 .
one could check the validity of this formula by the following elementary argument which occupies the rest of the proof .
we first recall that the homology classes @xmath168 and @xmath179 are linearly independent in @xmath180 by proposition 3.2 in @xcite .
let @xmath181 be the number of basic classes of @xmath177 .
then @xmath181 is the number of nonzero coefficients of @xmath182 \,\big[\ : \xi^{-(q-1)}\tau^{-(2p+q-3 ) } + \ : \xi^{q-1}\tau^{2p+q-3 } \\ & + & \hspace{-5pt } \big(\tau^{-(2p-4 ) } \sum_{i=0}^{2p-4}(-\tau^2)^{i } \big)\ , \big ( \,\sum_{j=0}^{q-3 } ( \xi\tau)^{-(q-3)+2j } \big)\:\big ] \\ & = & \big [ \sum_{k=0}^{n-1 } ( -1)^k \left(\hspace{-5pt}\begin{array}{c } n-1 \\
k\end{array}\hspace{-5pt}\right ) ( \xi^{n-2k - q}\tau^{3 - 2p - q } + \xi^{n+q-2k-2}\tau^{2p+q-3 } ) \big ] \\ &
+ & \big [ \sum_{k=0}^{n-1 } \sum_{i=0}^{2p-4}\sum_{j=0}^{q-3 } ( -1)^{i+k } \left(\hspace{-5pt}\begin{array}{c } n-1 \\ k\end{array}\hspace{-5pt}\right ) \xi^{n+2j+2 - 2k - q } \tau^{2i+2j+7 - 2p - q } \big ] \ , .\end{aligned}\ ] ] note that , if @xmath183 and @xmath184 , then @xmath185 hence @xmath186 is equal to the number of elements in the set @xmath187 on the other hand , the number of elements in @xmath188 is @xmath189 to see this , consider the linear map @xmath190 given by @xmath191 .
the kernel of @xmath192 is the line generated by the vector @xmath193 .
hence @xmath194 is equal to the number of lines in @xmath195 that are parallel to the vector @xmath193 and meet an integer point in the rectangular parallelepiped @xmath196 we see easily that the set of such lines is parameterized by the integer points on the three sides of @xmath197 lying on the @xmath198-plane , @xmath199-plane , and the plane @xmath200 .
the number of integer points on these three sides of @xmath197 is equal to the total number of integer points in @xmath197 minus the number of integer points in a smaller parallelepiped of dimensions one less .
see figure [ fig : cube ] . in @xmath195 ]
this proves the formula for @xmath194 .
now an easy calculation gives @xmath202 it follows that the symplectic tori @xmath86 are non - isotopic in @xmath3 , which completes the proof of theorem [ theorem:1 ] for @xmath203 .
we should note that since the number of basic classes of @xmath204 is @xmath205 , which is the same as the number of basic classes of @xmath206 , the above argument can not be used to conclude that no torus in @xmath86 is isotopic to @xmath207 .
it is also insufficient to show that the @xmath5-manifolds @xmath208 are mutually non - diffeomorphic , since the number of their basic classes are independent of @xmath176 .
nevertheless , it is possible to prove the following corollary by comparing the divisibilities of certain basic classes of @xmath209 .
[ cor : q=2 ] there are infinitely many mutually non - isotopic tori in the set of symplectic tori @xmath210 in @xmath3 . first note that by an easy mayer - vietoris argument we can extend the set @xmath211 to an integral basis of @xmath212 ( see @xcite pp.@xmath2137274 for the description of a basis of @xmath214\hspace{1pt};{\mathbb{z}})\,$ ] ) .
to distinguish @xmath215 and @xmath216 in @xmath3 we could choose a suitable @xmath217 and compare the divisibilities of basic classes with seiberg - witten invariant @xmath218 for @xmath219 .
when @xmath220 is odd and @xmath221 , a comparison of the divisibilities of basic classes for @xmath222 shows that these @xmath223manifolds are not diffeomorphic and hence @xmath215 and @xmath216 are non - isotopic in @xmath3 . in
case @xmath220 is even @xmath217 could be chosen as @xmath224 to distinguish @xmath215 and @xmath216 in @xmath3 for sufficiently large @xmath225 .
it is not hard to show that our family @xmath226 embed non - isotopically into a more general class of 4-manifolds than @xmath3 .
one way to generalize theorem [ theorem:1 ] is to use the formulas in @xcite and show that there is an infinite family of non - isotopic symplectic tori in @xmath2 $ ] for @xmath203 provided that @xmath7 is a @xmath8-embedded symplectic torus of self - intersection @xmath9 in a symplectic @xmath5-manifold ( cf.@xmath13theorem 5.2 in @xcite ) .
another way is to utilize the product formulas for seiberg - witten invariants given in @xcite and to generalize theorem [ theorem:1 ] to an even larger class of @xmath5-manifolds .
the latter will be our approach here . in the rest of this section
let @xmath7 denote a 2-torus of self - intersection @xmath9 , which is a symplectic submanifold of a symplectic 4-manifold @xmath11 .
we shall always assume that @xmath4\in h_2(x;{\mathbb{z}})$ ] is a primitive class , i.e. @xmath4\neq m \gamma$ ] for some @xmath227 and @xmath228 .
just as before , we may identify a tubular neighborhood @xmath229 with @xmath230 $ ] , and we immediately obtain an embedding of the tori @xmath226 into @xmath231 .
it is easy to check that lemma [ lemma : symplectic ] continues to hold in this more general setting , i.e. @xmath232 is a symplectic submanifold representing the homology class @xmath2 $ ] . to show that the family @xmath226 are mutually non - isotopic in @xmath11
, we need to again show that the corresponding 4-manifolds , @xmath233 \hspace{-20pt } \bigcup_{t_{p , q}\times\partial d^2=f_2\times \partial d^2 } \hspace{-20pt } [ e(r)\setminus \nu f_2]\ ] ] @xmath234\hspace{-20pt}\bigcup_{f\times\partial d^2=(s^1\times \alpha_1)\times\beta_1}\hspace{-20pt } [ s^1\times(s^3\setminus \nu l_{p , q})]\hspace{-20pt}\bigcup_{(s^1\times \alpha_2)\times\beta_2=f_2\times\partial d^2}\hspace{-20pt } [ e(r)\setminus\nu f_2]\ ; , \ ] ] are mutually non - diffeomorphic . here
we once again choose the homology basis @xmath52 , and @xmath235 when we do link surgery .
to compute the seiberg - witten invariants , we will need the following definition .
[ lemma : sw_x ] suppose @xmath236 is an essentially embedded @xmath240-torus of self - intersection @xmath9 . if @xmath241 , then @xmath242^{-1}-[f])\ , ( [ f]^{-q}-[f]^q)^{r-1 } \ , \delta_{l_{p , q}}^{sym } ( [ f]^2 , \tau^2 ) \ , , \ ] ] where @xmath243\in h_2(x\#_{\varphi_{p , q}}e(r)\hspace{1pt};{\mathbb{z}})$ ] is defined as in lemma @xmath244 .
corollary 20 in @xcite gives @xmath245^{-1}-[f]).\ ] ] the rest of the proof goes exactly the same way as in the proof of lemma [ lemma : sw ] .
again we need to identify @xmath169=q[f]$ ] for the same reasons as in the proof of lemma [ lemma : sw ] .
[ cor : x_pq ] suppose that @xmath7 is an essentially embedded symplectic @xmath240-torus in a symplectic @xmath5-manifold @xmath11 with @xmath146 .
also assume that @xmath4\in h_2(x;{\mathbb{z}})$ ] is primitive , @xmath4\cdot[f]=0 $ ] , and @xmath246 .
then for fixed @xmath156 , there are infinitely many manifolds in the set of symplectic @xmath5-manifolds @xmath247 that are mutually non - diffeomorphic .
when @xmath146 , we always have @xmath248 by taubes result @xcite .
note that the homology classes @xmath4 $ ] and @xmath179 are linearly independent in @xmath249 by proposition 3.2 in @xcite .
the rest of the proof is an easier analogue of the proof of corollary [ cor : non - diffeo ] .
we formally set @xmath250 $ ] , and note that as @xmath251 , the total number of terms in @xmath252 } \,=\ ; \overline{sw}_{\!\ ! x } \cdot ( [ f]^{-1}-[f ] ) \ ,
\delta_{l_{p , q}}^{sym } ( [ f]^2 , [ f]^2 ) \ ] ] goes to @xmath253 as well .
this immediately implies that the number of @xmath175 basic classes of @xmath254 also goes to @xmath253 as @xmath251 .
[ theorem:2 ] suppose that @xmath7 is an essentially embedded symplectic @xmath240-torus in a symplectic @xmath5-manifold @xmath11 with @xmath146 .
also assume that @xmath4\in h_2(x;{\mathbb{z}})$ ] is primitive , @xmath4\cdot[f]=0 $ ] , and @xmath246 .
then for any integer @xmath203 , there exists an infinite family of mutually non - isotopic symplectic tori representing the homology class @xmath2 $ ]
. one must take care and define @xmath257 when @xmath258 ( see @xcite and @xcite ) .
when @xmath258 , it is not automatic that @xmath145 is a finite sum and @xmath259 for a symplectic @xmath11 .
if indeed @xmath259 and is a finite sum , then theorem [ theorem:2 ] will still be valid for such pair @xmath255
. however if @xmath260 or is an infinite sum , then there seems to be no method currently available to check whether the tori in our family are mutually non - isotopic in @xmath11 . along this line
, it has been conjectured that an infinite family of homologous but non - isotopic symplectic tori can not occur in @xmath11 , when @xmath11 is a rational ruled surface with @xmath261 .
in fact , sikorav has already proved this conjecture for @xmath262 in @xcite and siebert and tian announced a proof for @xmath263 and @xmath264 ( see @xcite ) .
we should also point out that since our proof uses the product formula to calculate the seiberg - witten invariants of @xmath5-manifolds obtained by gluing along the boundary of a regular neighborhood of a torus and since such product formulas do nt give the complete picture in case the gluing occurs along higher genus surfaces , we were able to construct infinite families of non - isotopic tori only .
in fact , there is an ongoing research to determine the number of symplectic representatives ( up to smooth isotopy ) in the homology class of symplectic surfaces of higher genus .
for example in @xcite , by using purely topological techniques , smith was able to prove that for every odd number @xmath265 there exists a symplectic @xmath5-manifold @xmath266 that contains an infinite family of homologous but non - isotopic connected symplectic surfaces of genus @xmath267 .
on the other hand , siebert and tian announced in @xcite that there is a unique symplectic surface in the homology class of a degree @xmath268 complex curve in @xmath269 provided that @xmath270 and similarly in the homology class of a complex curve of bidegree @xmath271 in a hirzebruch surface provided that @xmath272 .
morton : the multivariable alexander polynomial for a closed braid , in _ low - dimensional topology _ , ed .
hanna nencka , _ contemporary mathematics _ * 233 * , amer . math .
( 1999 ) , 167172 . also available at arxiv : math.gt/9803138 . | for any pair of integers @xmath0 and @xmath1 , we construct an infinite family of mutually non - isotopic symplectic tori representing the homology class @xmath2 $ ] of an elliptic surface @xmath3 , where @xmath4 $ ] is the homology class of the fiber .
we also show how such families can be non - isotopically and symplectically embedded into a more general class of symplectic @xmath5-manifolds . |
non - compact non - linear sigma models are an important and extensively used tool in the study of disordered electron systems .
the relevant formalism was pioneered by wegner @xcite , schfer & wegner @xcite , and pruisken & schfer @xcite .
efetov @xcite improved the formalism by developing the supersymmetry method to derive non - linear sigma models .
many applications of the supersymmetry method can be found in the textbook by efetov @xcite .
there exist different ways to derive non - linear sigma models from microscopic models ; for an introduction see @xcite .
one step in the traditional approach uses a hubbard - stratonovich transformation , i.e. , a transformation of the form @xmath2 where @xmath3 and the domain of integration d is left unspecified for now .
@xmath4 denotes lebesgue measure of a normed vector space . for the case of compact symmetries
the transformation is just a trivial gaussian integral . to give an indication of the difficulty which arises in the case of a non - compact symmetry ( also known as the boson - boson sector of efetov s supersymmetry formalism )
let us briefly discuss the example of orthogonal symmetry @xmath1 .
there , @xmath5 is given by @xmath6 with @xmath7 and @xmath8 .
the @xmath9 represent the microscopic degrees of freedom .
using equation and integrating out @xmath10 gives a description in terms of the effective degrees of freedom @xmath11 .
the task is to find a domain of integration d for which identity holds and the term @xmath12 stays bounded .
the latter condition is imposed in order for fubini s theorem to apply , as further execution of the wegner - efetov formalism calls for the @xmath10 and @xmath11 integrals to be interchanged .
note that the real matrices @xmath5 obey the symmetry relation @xmath13 . a naive choice of integration domain d keeping the term @xmath12 bounded would be the domain of all real matrices satisfying @xmath14 .
unfortunately , this choice of d is not a valid choice in the context of the integral as it renders the quadratic form @xmath15 of indefinite sign .
schfer and wegner ( sw ) @xcite suggested a domain and showed that it solves the difficulty .
yet , a different domain was proposed in later work by pruisken and schfer ( ps ) @xcite . until recently the mathematical status of identity for the ps domain was unclear .
the main obstacle in proving for the ps domain is the existence of a boundary .
this precludes an easy proof by completing the square and shifting the contour ( as is possible for the standard gauss integral and for the sw domain ) .
nevertheless , the ps domain was used in most applications worked out by the mesoscopic and disordered physics community ; an early and influential paper of this kind is @xcite .
most likely , the reason is that it is easier to do calculations with , as it is invariant under the full symmetry group of the domain of matrices @xmath5 .
recently fyodorov , wei and zirnbauer in a series of papers @xcite proved the ps variant of the hs transformation for the special cases of unitary and orthogonal symmetry . in this article
we extend the results to more general symmetry groups .
moreover , our proof clarifies the relation between the ps transformation , the sw transformation and the standard gaussian integrals .
it is shown that the different integrals can be transformed into each other by deforming the domain of integration without changing the value of the integral .
here is a guide to reading : in section [ results ] we define the setting and state our main result in the form of a theorem .
in addition , we give two corollaries which relate more directly to previous results . in section [ examples ] we apply our results to three different symmetry classes . in particular ,
previous results concerning the cases of unitary and orthogonal symmetry are reproduced .
the proof of the theorem is contained in section [ proof ] , which is divided into three subsections .
for the convenience of the reader each subsection is preceded by a short introduction of notation , essential structures , and a lemma containing the results of the pertinent part of the proof .
the last subsection of section [ proof ] deals with the two corollaries .
all constructions take place in @xmath16 , the lie algebra of complex @xmath17 matrices .
[ please be advised however that the following results also apply to the case where @xmath18 is replaced by a complex reductive lie subalgebra of @xmath18 .
] let @xmath19 be hermitian with the property @xmath20 .
this matrix @xmath21 gives rise to two involutions @xmath22 and @xmath23 on @xmath16 . `
involution ' here means an involutive lie algebra automorphism . for greater generality
we allow for further involutions @xmath24 to be present on @xmath16 .
two requirements have to be fulfilled : firstly , all involutions have to commute with each other and secondly , @xmath21 has to be in the plus or minus eigenspace of each @xmath24 , i.e. @xmath25 with @xmath26 .
the fixed point set of @xmath27 and the @xmath24 s is the real lie algebra @xmath28 we also introduce the real vector space @xmath29 which is an @xmath30-module for the adjoint ( or commutator ) action by @xmath31 . due to @xmath32 ,
the decompositions of @xmath31 and @xmath33 into the plus and minus one eigenspaces of @xmath34 are decompositions into hermitian and antihermitian parts .
we write these decompositions as @xmath35 and @xmath36 , where @xmath37 and @xmath38 are in the plus one eigenspace and @xmath39 and @xmath40 are in the minus one eigenspace . @xmath37 and @xmath40 consist of antihermitian matrices whereas @xmath39 and @xmath38 consist of hermitian matrices .
the commutation relations among all these spaces , @xmath41 [ \mathfrak{k } , \mathfrak{k } ] \subset \mathfrak{k } , & [ \mathfrak{k } , \mathfrak{p } ] \subset \mathfrak{p } , & [ \mathfrak{p } , \mathfrak{p } ] \subset \mathfrak{k } , & [ \mathcal{q}_+ \ , , \mathcal{q}_- ] \subset \mathfrak{p } , \\ [ \mathcal{q}_{\pm } \ , , \mathcal{q}_{\pm } ] \subset \mathfrak{k } , & [ \mathfrak{k } , \mathcal{q}_{\pm } ] \subset \mathcal{q}_{\pm } \ , , & [ \mathfrak{p } , \mathcal{q}_{\pm } ] \subset \mathcal{q}_{\mp } \ , , \end{matrix*}\end{aligned}\ ] ] imply that @xmath42 is a lie algebra .
( this lie subalgebra of @xmath16 could have served as the starting point of our setting . ) by the definition of @xmath43 the matrix @xmath44 is hermitian for all @xmath45 .
note that @xmath46 . to preclude any pathologies that might otherwise occur , we demand that the lie group @xmath47 be closed .
the parametrization of the pruisken - schfer domain is given by @xmath48 the standard domain for a gaussian integral is called ` euclidean ' in the following .
it is parametrized by @xmath49 where @xmath50 .
finally , the parametrization of the one - parameter family of schfer - wegner domains is given by @xmath51 where @xmath52 is any positive real number .
the following statement relies on making a choice of orientation for the ps domain .
( note that no such choice is made for d in . )
once and for all we now fix an orientation for each of the vector spaces @xmath38 , @xmath39 , and @xmath40 .
by viewing @xmath53 , @xmath54 , and @xmath55 as orientation - preserving maps , we then have orientations on the corresponding domains of integration .
[ thm_2_1 ] let @xmath45 in the setting above .
if @xmath56 one has @xmath57 here , @xmath58 ^ 2 ) \leq 1 $ ] is a regulating function ( @xmath59 ) and @xmath60 denotes a constant volume form ( i.e. a constant differential form of top degree ) on @xmath33 . the normalization constant @xmath61 does not depend on @xmath5 .
the main idea of the proof is to show that the ps domain can be extended by a nulldomain ( of holomorphic continuation of @xmath60 ) and then deformed into a euclidean domain without changing the value of the integral .
appendix [ app_sw_euclid ] shows that one can also deform the sw domain into this euclidean domain .
thus the ps and sw domains are deformations of the same euclidean domain .
now we formulate two corollaries . for that purpose
let @xmath62 be a maximal abelian subalgebra @xmath63 .
we require that @xmath64 .
let @xmath65 denote a set of positive roots of the adjoint action of @xmath62 on @xmath66 .
similarly @xmath67 denotes a set of positive roots of the adjoint action of @xmath62 on @xmath68 .
the multiplicity of a root @xmath69 is denoted by @xmath70 .
the following corollary is the analogue of corollary 1 in @xcite .
[ cor_2_1 ] let @xmath71 denote haar measure of the closed analytic subgroup @xmath72 with lie algebra @xmath31 .
we then have @xmath73 where @xmath74 denotes lebesgue measure on the vector space @xmath62 and @xmath75 the constant @xmath76 does not depend on @xmath5 .
it is particularly noteworthy that for odd multiplicities @xmath70 of roots @xmath77 the ` jacobian ' @xmath78 is not positive but has alternating sign .
the following corollary is the analogue of theorem 1 in @xcite : [ cor_2_2 ] if the parametrization @xmath53 is nearly everywhere injective and regular , then @xmath79 here @xmath80 denotes the non - oriented image of @xmath53 .
the mapping from @xmath81 to @xmath62 sending @xmath11 to @xmath82 is well defined up to a set of measure zero .
@xmath4 denotes lebesgue measure on @xmath33 , and @xmath83 is a constant which does not depend on @xmath5 .
while we believe that the assumptions on @xmath53 in corollary [ cor_2_2 ] follow from the general setting , we have not been able to find a proof thereof .
first we reproduce the examples of unitary and orthogonal symmetry . for this
we calculate @xmath84 and apply corollary [ cor_2_1 ] .
this case has been handled by fyodorov @xcite using different methods .
to apply the general theorem we work in the complex lie algebra @xmath85 and define @xmath86 .
no additional involutions @xmath24 are needed .
we have @xmath87 the maximal abelian subalgebra @xmath88 is spanned by the real diagonal matrices .
let @xmath89 be such a matrix .
the roots @xmath90 are given by @xmath91 where @xmath92 and @xmath93 .
for @xmath94 the roots @xmath91 are elements of @xmath95 , otherwise they are elements of @xmath96 .
the root space corresponding to @xmath97 is @xmath98 where @xmath99 is the matrix with unity in the @xmath100 position and zero elsewhere .
thus every root space has complex dimension one , or real dimension two .
hence @xmath101 with this expression for @xmath78 the formula of corollary [ cor_2_1 ] agrees with that of fyodorov @xcite . this case has been dealt with by fyodorov , wei and zirnbauer @xcite .
in addition to the involutions of the unitary setting we need an involution @xmath102 and @xmath103 .
the additional presence of this involution requires all matrices to be real . in consequence ,
all root spaces are now one dimensional , and they give rise to non - trivial signs : @xmath104 which is precisely corollary 1 in @xcite .
now we consider the case of symplectic symmetry which arises for random matrix ensembles of class @xmath5ii in the language of @xcite .
let @xmath106 @xmath107 denote the three pauli matrices and let @xmath108 . introducing @xmath109 , we choose @xmath110 and define @xmath111 .
the involution @xmath112 together with @xmath113 leads to @xmath114 a maximal abelian subalgebra of @xmath88 is @xmath115 note @xmath116 .
let @xmath117 and let @xmath118 be defined by @xmath119 .
then we have @xmath120 to determine the root multiplicities we note that the quaternions @xmath121 constitute a basis of the space @xmath122 .
the root spaces corresponding to @xmath123 then are @xmath124 thus all root spaces have dimension four and @xmath84 is given by @xmath125
in the proof we use some standard results of lie theory , all of which can be found in the literature , e.g. in @xcite . since @xmath31 is closed under hermitian conjugation ( @xmath126 ) we know that @xmath31 is reductive , i.e. the direct sum of an abelian and a semisimple lie algebra . for simplicity we first restrict ourselves to the case where @xmath31 is semisimple .
the extension to the reductive case will be straightforward .
the proof of the theorem is divided into three parts .
the first part , in section [ ps_domain ] , contains the derivation of a new parametrization of the ps domain , which makes it possible to deal with its boundary . the second part , in [ ext_ps ] ,
is concerned with the extension of the ps domain to a domain without boundary .
first we identify good directions into which to extend the ps domain .
then we give an extension of ps which does not change the value of the integral .
although much of it is unnecessary for the formal proof , section [ ext_ps ] is an important prerequisite to understanding the third part , [ equiv_ps_euclid ] , where we give a homotopy @xmath127 connecting the extended ps domain to the euclidean domain .
the main point is to make rigorous the following schematic application of stokes theorem : @xmath128 where we have introduced @xmath129 .
the first term on the right hand side is identically zero because @xmath130 is holomorphic in @xmath11 . in the final subsection [ sect:4.4 ]
we deduce the corollaries [ cor_2_1 ] and [ cor_2_2 ] . at this point
a warning is in order . in the given form
the expressions above do not make sense . in order for the integrals over @xmath53 and @xmath127 to exist
we have to include some regularization .
this delicate issue is discussed in detail in the last part of subsection [ equiv_ps_euclid ] .
that discussion also entails that the extension of @xmath53 does not contribute to the left hand side of the equation .
we now invest some effort in order to derive a parametrization of the domain of integration which gives full control over its boundary . to guide the reader , we first define and explain all objects that are necessary to formulate a lemma stating the parametrization .
in order to evaluate @xmath131 in explicitly , we need to compute multiple commutators of @xmath132 with @xmath133 .
therefore we now choose a maximal abelian subalgebra @xmath134 in @xmath39 and diagonalize the commutator action of @xmath134 on @xmath33 .
this diagonalization process gives rise to a root space decomposition @xmath135 where @xmath136 denotes a set of positive roots .
each root space in turn is decomposed into a hermitian ( @xmath38 ) and an antihermitian ( @xmath40 ) part : @xmath137 we also let @xmath138 .
hence we have the decompositions @xmath139 for future reference we observe that @xmath140 , \label{ad_s_isom}\end{aligned}\ ] ] is an isomorphism .
this fact will be used several times in the proof . for the following constructions we review the notion of pointed polyhedral cone and triangulations thereof @xcite .
a pointed polyhedral cone is a subset of a vector space . by definition
it is an intersection of finitely many half spaces where the intersection of all hyperplanes bounding the half spaces contains only the zero vector .
the word pointed reflects the fact that there exists a hyperplane which intersects the cone only at zero , with the rest of the cone lying strictly on one side of that hyperplane .
for example , if @xmath141 denotes a system of positive roots for the adjoint action of @xmath134 on @xmath31 , the positive weyl chamber @xmath142 is a pointed polyhedral cone . in the following we refer to a pointed polyhedral cone as a cone for short .
let @xmath143 be a vector space of codimension one such that @xmath144 lies entirely on one side of @xmath145 .
a face of @xmath144 is a set of the form @xmath146 .
the zero vector is the unique zero dimensional face .
it is convenient also to include the empty set as a face .
the one dimensional faces are called edges .
note that each nontrivial face is again a cone .
it is a fact @xcite that any cone @xmath144 admits a different representation : there exist @xmath147 elements @xmath148 such that @xmath149 the @xmath150 are called generators of the cone . they can be chosen in such a way that each @xmath150 generates an edge of the cone . in the case of a positive weyl chamber @xmath144 a set of generators
is furnished by the simple co - roots .
a cone is called simplicial if its generators are linearly independent .
a @xmath151-cone is a cone of dimension @xmath151 .
it is a known fact of lie theory that @xmath144 is a simplicial @xmath152-cone .
a finite collection @xmath153 of @xmath154-cones is called a subdivision of @xmath144 if @xmath155 and @xmath156 is a face of both @xmath157 and @xmath158 for all @xmath159 .
if each cone in a subdivision @xmath153 is simplicial , then @xmath153 is called a triangulation .
bearing these facts in mind we proceed to describe a decomposition of @xmath144 which , as we shall see below , is directly related to the boundary of the ps domain .
note , first of all , that a root @xmath160 may change sign on @xmath144 since @xmath144 is defined with respect to the root system @xmath141 .
the closures of the connected components of @xmath161 can be obtained as appropriate intersections of half spaces and hence are again cones .
let @xmath162 denote the collection of generators of these cones [ the cardinality @xmath163 exceeds @xmath147 if @xmath164 . by construction
the intersection of two such cones is a face common to both . put differently ,
the generators common to two such cones generate a joint face .
thus the decomposition we have just described yields a subdivision of @xmath144 . it is a fact @xcite that every subdivision of a cone can be refined to a triangulation without introducing any new generators . for the rest of the article we fix a triangulation @xmath165 which refines the subdivision of @xmath144 described above .
let @xmath166 be such that @xmath167 is the set of generators for the simplicial cone indexed by @xmath168 , i.e. , let @xmath169 note that @xmath170 and that the generators @xmath167 form a basis of @xmath134 .
the latter fact implies that each @xmath171 is represented uniquely as @xmath172 with coefficients @xmath173 .
the intersection @xmath174 of two simplicial cones is again a simplicial cone ; indeed , the set of generators of the latter is @xmath175 .
a key property of the decomposition is that the sign of each @xmath176 stays constant on any given simplicial cone @xmath177 .
however it may still happen that @xmath69 vanishes on the boundary of @xmath177 .
next we introduce a subdecomposition of each cone @xmath177 .
let @xmath178 and define @xmath179 an example of this decomposition is shown in figure [ eps_domains ] .
it may be a helpful observation to note that the decomposition @xmath180 carries the structure of a simplicial complex . for @xmath181 .
the index @xmath182 has been omitted since there exists only one simplicial cone in this case .
the possible @xmath183 s are subsets of @xmath184 . ] with these definitions understood we introduce for each index pair @xmath185 a function on @xmath134 by @xmath186 where @xmath187 is meant in the sense of with coefficients @xmath188 . in order for this function @xmath189 to be well - defined
it is crucial that the decomposition of @xmath144 into simplicial cones is such that for fixed @xmath182 and fixed @xmath69 the sign of @xmath190 is the same for all @xmath191 with @xmath192 .
we are now going to formulate a lemma which summarizes what we are aiming at in this section .
for that purpose we introduce @xmath193 and let @xmath194 be the centralizer of @xmath195 in @xmath196 . fixing some @xmath197 with @xmath198 for all @xmath199 we define @xmath200 = z - z ' .\end{aligned}\ ] ] note that @xmath201 satisfies @xmath202 in addition we define orthogonal projections @xmath203 the following lemma contains a parametrization of the domain of integration which gives direct control over its boundary . [ lem_parametrisation ] the mappings @xmath204 ) \mapsto \operatorname{ad}(k ) \big(x_0 + \sum_{\alpha \in \sigma_+(\mathcal{q},\mathfrak{a } ) } ( x_{\alpha } + t_{\alpha , c}(h ) \ , \phi ( x_{\alpha } ) ) \big ) , \notag\end{aligned}\ ] ] with @xmath205 and @xmath206= [ kz^{-1};zxz^{-1}]$ ] for @xmath207 have the following properties : 1 .
[ lem_par_it_boundary ] the boundary @xmath208 ( in the sense of integration chains ) is obtained by applying the boundary operator @xmath209 to @xmath210 .
[ lem_par_it_orientation ] a choice of orientation on @xmath211 induces an orientation for each @xmath212 .
there exists a particular choice of orientation for which @xmath213 holds , where the equality sign is meant in the sense of integration chains .
[ lem_par_it_codim]the contributions to the boundary of @xmath212 which come from @xmath214 are of codimension at least two and can be neglected . to prove lemma [ lem_parametrisation ] we perform a sequence of four reparametrizations of the original parametrization @xmath53 .
the first three reparametrizations are preparatory and do not relate directly to lemma [ lem_parametrisation ] .
each reparametrization is discussed in a separate subsection for clarity .
the goal of the next three reparametrizations is to evaluate @xmath215 in more detail .
key to this is a choice of maximal abelian subalgebra @xmath216 whose adjoint action on @xmath217 is diagonalizable . to get started , we parametrize @xmath39 using @xmath218 and the interior @xmath219 of @xmath144 : @xmath220 ) & \mapsto k h k^{-1 } .\end{aligned}\ ] ] @xmath221 is obviously well defined , and it is a standard fact that @xmath221 is injective for semisimple lie algebras with cartan decomposition @xmath222 .
hence @xmath221 is a diffeomorphism onto @xmath223 .
note that @xmath224 is a set of measure zero since @xmath225 ( see e.g. @xcite ) and @xmath226 where @xmath69 runs over the roots in @xmath227 .
precisely speaking , we are going to use the parametrization @xmath228 , x ) \mapsto \ ; & e^{khk^{-1 } } x e^{-k hk^{-1 } } .\end{aligned}\ ] ] recall that the ps domain is oriented by an orientation of @xmath229 .
declaring @xmath221 to be orientation preserving induces an orientation on @xmath230 .
further reparametrizations of the ps domain are introduced below . to avoid an overload of notation
, we will denote each new parametrization still by @xmath53 . in this section
we prepare the further evaluation of the @xmath232 action in the next subsection .
consider the reparametrization @xmath233&\mapsto ( [ k ] , kxk^{-1 } ) , \end{aligned}\ ] ] where the expression @xmath234\equiv [ k z^{-1 } ; z x z^{-1}]$ ] ( for @xmath235 ) stands for an equivalence class of the group action of @xmath194 on @xmath236 .
this group action defines the trivial bundle @xmath237 .
the inverse of @xmath238 is @xmath239,x ) & \mapsto [ k ; k^{-1}xk ] .\end{aligned}\ ] ] @xmath238 is a diffeomorphism and can therefore be used as a reparametrization to obtain the new parametrization @xmath240 ) \mapsto \ ; & e^{k h k^{-1 } } k x k^{-1 } e^{-kh k^{-1 } } \notag \\ & = k e^{h } x e^{-h } k^{-1 } = \operatorname{ad}(k ) ( e^{\operatorname{ad}(h ) } x ) .
\label{psii}\end{aligned}\ ] ] since @xmath241 is a subgroup of @xmath196 and , by definition , commutes with the @xmath232 action on @xmath33 , the parametrization is compatible with the decomposition of @xmath33 .
in particular we have @xmath242 for @xmath207 and @xmath199 .
a short calculation for @xmath243 gives @xmath244 hence the parametrization can be rewritten as @xmath245 ) \mapsto \operatorname{ad}(k ) \big ( x_0 + \sum_{\mathclap{\alpha \in \sigma_+(\mathcal{q},\mathfrak{a } ) } } \big(\cosh(\alpha(h ) ) x_{\alpha } + \sinh(\alpha(h ) ) \phi(x_{\alpha})\big ) \big ) , \label{ps_decomp}\end{aligned}\ ] ] where @xmath246 . as a motivation for the next reparametrization
we note that @xmath247 is an @xmath248-invariant scalar product on @xmath33 and that all the different spaces @xmath249 are orthogonal to each other . for the moment
, we fix @xmath250 and consider only the part @xmath251 of the parametrization . the corresponding two - dimensional picture is shown in figure [ rep_3 ] , where we see the image of a straight line through the origin in @xmath144 as a hyperbola . before ( left ) and after ( right ) the reparametrization @xmath252 . ]
we are going to change the parametrization in such a way that the hyperbola is rectified to a straight line ; see figure [ rep_3 ] .
such a reparametrization gives us a handle on the boundary of the ps domain , as is discussed in the next subsection .
accordingly , the third reparametrization we use is given by @xmath253 ) & \mapsto \big(h,\big[k;x_0 + \sum_{\alpha \in
\sigma_+(\mathcal{q},\mathfrak{a})}\frac{1}{\cosh(\alpha(h ) ) } x_{\alpha}\big]\big ) .\end{aligned}\ ] ] this is another orientation preserving diffeomorphism .
we thus obtain @xmath254 ) \mapsto \operatorname{ad}(k ) \big ( x_0 + \sum_{\mathclap{\alpha \in \sigma_+(\mathcal{q},\mathfrak{a } ) } } \big ( x_{\alpha } + \tanh(\alpha(h ) ) \phi(x_{\alpha})\big ) \big ) , \label{ps_param}\end{aligned}\ ] ] which is renamed to @xmath53 in the following . from the parametrization
( see also figure [ rep_3 ] ) it is clear that the boundary is reached when some @xmath255 goes to @xmath256 and hence @xmath257 goes to @xmath258 .
put differently , the boundary of the domain of integration can be reached through a limit in the parameter space @xmath144 . to obtain control over the boundary we have to make sense of the expression @xmath259 .
this limit is encoded in the functions @xmath260 .
recall that @xmath261 where the index @xmath182 refers to the decomposition of @xmath144 into the simplicial cones @xmath262 .
fix @xmath160 and @xmath263 such that @xmath264 .
then @xmath265 this shows that @xmath189 is continuous .
@xmath189 is also differentiable since @xmath266 generalizes to all higher ( and mixed ) partial derivatives . to put the functions @xmath260 to use we define for each cone
@xmath168 the mapping @xmath267
\right)&\mapsto \left(\sum_{i\in i_c } \frac{h^i}{1-h^i } h_{i},[k;x ] \right ) , \end{aligned}\ ] ] where @xmath268 denotes the interior of @xmath269 . for each simplicial cone , this is an ( orientation preserving ) diffeomorphism onto its image .
the mapping is visualized in figure [ rep_4 ] .
we use it to reparametrize @xmath53 on each cone .
we thus obtain @xmath270 , which is the parametrization defined in lemma [ lem_parametrisation ] , eq .
@xmath271 . in the following
we want to give the notion ` boundary of the ps domain ' a precise meaning . in the case of integration cells , i.e. , differentiable mappings defined on a cube , the boundary operator @xmath209 is defined as usual .
@xmath209 can also be applied to integration chains , i.e. formal linear combinations of cells . in principle
the correct procedure would be to decompose each @xmath212 into cells in order to apply @xmath209 .
however , in the following we argue that we can treat each @xmath212 effectively as single cell with the boundary operator @xmath209 acting just on the @xmath272 part of the domain of definition .
note that @xmath269 by the decomposition is diffeomorphic to an @xmath273-dimensional cube .
first note that @xmath212 extends as a differentiable mapping to a neighborhood of @xmath269 since the @xmath189 are differentiable functions defined on @xmath134 .
thus it is possible to define the orientation of the boundary .
furthermore , since @xmath196 is a closed compact manifold it suffices to discuss boundary contributions arising from a decomposition of @xmath38 into cells . by inspecting our parametrization
we see that going to infinity on @xmath38 implies going to infinity in the domain of integration . in section [ equiv_ps_euclid ]
we show that the integrand goes to zero exponentially on this domain and hence all possible boundary contributions vanish .
thus we obtain part [ lem_par_it_boundary ] of lemma [ lem_parametrisation ] part [ lem_par_it_orientation ] of lemma [ lem_parametrisation ] follows immediately by noting that the mappings @xmath274 are diffeomorphisms onto . since
similar arguments are needed in several other parts of section [ proof ] , the proof of part [ lem_par_it_codim ] of lemma [ lem_parametrisation ] is presented in appendix [ app_contributions ] .
example for @xmath275 mapping @xmath276 on the left hand side to @xmath277 on the right side .
in this example there is only one simplicial cone . ] in this section we construct an extension of the ps domain which has no boundary other than the irrelevant boundary at infinity .
the idea is to connect each boundary point to infinity by attaching one halfline .
the directions of these halflines should be such that the attached domain does not contribute to the integral of @xmath278 .
we first determine such a direction for each boundary point , and then give a parametrization of the attached domains . in the following let
@xmath279 denote the trace form on @xmath280 .
some care must be exercised in order to guarantee the convergence of the integral on the extended ps domain .
the positivity requirement @xmath281 in the theorem already gives a hint that the matrix @xmath21 plays a prominent role in the discussion of convergence .
owing to @xmath282 we have the decomposition @xmath283 where @xmath284 and @xmath285 .
the next lemma introduces the convergent directions which are used to extend @xmath53 in such a way that convergence is maintained .
[ e_lem ] the matrices @xmath286 are well defined and non - zero .
the following properties hold for all @xmath287 : 1 .
[ inequality]@xmath288 .
2 . there exist numbers @xmath289 such that the matrices @xmath290 decompose as @xmath291 3 . for @xmath292 one
has @xmath293 . in particular ,
@xmath294 .
before we come to the proof of lemma [ e_lem ] , we formulate another lemma which suggests how to extend the ps domain .
note however that an integral over @xmath53 needs a regulating function and that we postpone the discussion of convergence to section [ equiv_ps_euclid ] .
strictly speaking , the next lemma is not necessary for the proof of theorem [ thm_2_1 ] .
it is included as a preparation for the more involved definition of the homotopy introduced in section [ equiv_ps_euclid ] .
[ extension_lem ] for @xmath168 and @xmath295 the mappings @xmath296 ) \mapsto \operatorname{ad}(k ) \big ( x_0 + \sum_{\mathclap{\alpha \in \sigma_{+}(\mathcal{q},\mathfrak{a } ) } } \big(x_{\alpha } + t_{\alpha , c}(h )
\phi(x_{\alpha})\big ) -i \sum_{j \in l } ( h^j-1 ) e_{j } \big ) , \end{aligned}\ ] ] are well defined as integration chains and one has @xmath297 as long as the sum of chains is integrated against forms with sufficiently rapid decay at infinity . to see that the matrices @xmath298 are well defined , we express them in a more explicit fashion .
a short calculation using gives @xmath299 this shows that the limit in exists and that the matrices @xmath300 decompose as shown in equation .
thus we obtain @xmath301 . recalling that @xmath302 is injective
we conclude that each of the matrices @xmath300 is non - zero .
for property @xmath303 we note that @xmath56 and @xmath304 thus @xmath305 is the trace of the product of the non - zero positive semidefinite hermitian matrix @xmath306 and the positive hermitian matrix @xmath44 . by inserting the eigenvalue representation @xmath307 of @xmath306
we get @xmath308 since @xmath44 is positive we have @xmath309 .
property @xmath303 then follows because @xmath310 implies that there exists some @xmath311 . to prove property @xmath312 we use @xmath301 and
note that for all @xmath313 we have the orthogonality relations @xmath314 and @xmath315 .
we also recall that a fixed root @xmath69 does not change sign on a fixed simplicial cone @xmath177 .
the desired result @xmath316 for @xmath317 then follows directly . in this section
we show that the mappings @xmath318 in lemma [ extension_lem ] have the stated properties .
first of all , the mappings are well defined since @xmath241 acts trivially on the matrices @xmath298 .
we recall that the situation is visualized in figure [ eps_domains ] .
@xmath318 can be extended as a mapping to @xmath134 since @xmath189 makes sense on @xmath134 and so does @xmath319 .
thus @xmath318 is well defined as an integration chain . by the same argument as for the @xmath320 case ,
a boundary can arise only by the action of the boundary operator @xmath209 on the factor @xmath321 . in the following we always neglect possible boundary contributions from infinity , since the integrals under consideration are convergent by assumption .
to see that the different integration cells @xmath318 and @xmath322 fit together in a seamless way , we recall that @xmath323 is again a simplicial cone which is generated by the set @xmath175 .
in particular each @xmath324 can be represented in the form @xmath325 , which implies that @xmath326 .
we have @xmath327 for @xmath328 and @xmath329 for @xmath330 .
together with similar conditions from @xmath331 this yields @xmath332 on the intersection @xmath333 .
hence we obtain the equality @xmath334 ) = ps_{l',c'}(h , [ k;x])$ ] on the joint domain of definition .
moreover , the induced orientations on the boundaries between two neighboring cells are opposites of each other .
together with the fact ( shown in appendix [ app_contributions ] ) that the contributions from @xmath335 are of codimension no less than two , these results yield .
to get some intuition for the situation it is useful to observe that the halflines which are glued to boundary points of the ps domain , point into directions within @xmath336 , and hence can not coincide with vectors tangent to ps , which live in @xmath33 .
finally , we show that the integral over @xmath53 equals the integral over @xmath54 .
the idea is to deform the extended ps domain into the subspace @xmath337 $ ] of @xmath280 where @xmath338 is positive .
recall that @xmath339=\mathcal{q}_-$ ] .
we have to show that the integral remains convergent along the path of deformation and no boundary terms at infinity are generated .
to that end we prefer to proceed in the reverse order and deform @xmath54 into @xmath53 to the extent that this is allowed by convergence of the integral .
recall that @xmath194 acts trivially on @xmath340 and the matrices @xmath298 .
it also acts trivially on @xmath341 $ ] for all @xmath342 because @xmath343 is fixed by conjugation with @xmath21 . for these reasons the mapping defined as @xmath344\times&\mathfrak{a}_{l , c}^+ \times k \times_{z_k(\mathfrak{a } ) } \mathcal{q}_+ \rightarrow \mathcal{q}^{\mathbb{c } } , \notag\\ ( t , h,[k ; x ] ) \mapsto & \operatorname{ad}(k ) \big ( x_0 + \sum_{\mathclap{\alpha \in \sigma_+(\mathcal{q } , \mathfrak{a } ) } } \big(x_{\alpha } + ( 1-t ) t_{\alpha , c}(h ) \phi(x_{\alpha})\big ) \big ) \notag\\ & -i \operatorname{ad}(k ) \sum_{j \in l } ( h^j-1 ) \big((1-t)e_{j } + 2 t [ h_j , s ] \big)\end{aligned}\ ] ]
is well defined . by reasoning similar to that for @xmath318 each parametrization @xmath345 ( @xmath346 )
can be seen as an integration cell with boundary coming only from the @xmath347 \times\mathfrak{a}_{l , c}^+$ ] part . to simplify the notation we define @xmath348 @xmath349 and similarly @xmath350 .
[ hom_conv_lem ] the mappings @xmath345 have the following properties : 1 .
[ partial_eps_eps ] the boundary of the sum @xmath351 is given by @xmath352 2 .
[ eps_conv ] let @xmath129 .
then the integrals @xmath353 exist for @xmath59 .
[ eps_limit_zero ] for each @xmath346 and @xmath354 with cardinality @xmath355 we have @xmath356 4 . [ ps_limit]let @xmath357 .
the integral over @xmath358 in the limit @xmath359 may be computed as an integral over @xmath53 with regularized integrand : @xmath360 the proof of the lemma is spelled out in the next four subsections . here
we anticipate that once the lemma has been established , we can do the following series of manipulations : @xmath361 which yields the statement of our theorem [ thm_2_1 ] .
next we prove statement @xmath303 of lemma [ hom_conv_lem ] . by an argument similar to that in the proof of lemma [ extension_lem ] we obtain @xmath362 we will now deal with the summand @xmath363 .
for @xmath364 the mapping @xmath365 degenerates in the limit @xmath366 .
more precisely , for @xmath367 we have @xmath368 and thus a reduction in dimension .
hence we have the following identity relating integration chains : @xmath369 in the following we establish the connection between @xmath363 and @xmath54 . for @xmath346 and @xmath370
we have @xmath371 ) & = \operatorname{ad}(k ) \big ( x - 2 i \sum_{j \in i_c } ( h^j -1 ) [ h_j , s ] \big ) .\end{aligned}\ ] ] to facilitate the interpretation of the expression on the right hand side , we now change the left factor of the domain of definition of @xmath372 from @xmath373 to @xmath177 .
this is done by introducing the diffeomorphism @xmath374 by inserting it into the previous formula we get @xmath375 ) & = \operatorname{ad}(k ) \big ( x - 2 i[h , s ] \big ) .\end{aligned}\ ] ] note that while the composition with @xmath376 does alter the mapping @xmath372 , the effect is not a change of image but only a reparametrization .
the right hand side of the expression above does not have any explicit dependence on the simplicial cone @xmath182 any more .
therefore , by recalling @xmath377 and noting that the diffeomorphisms @xmath378 are orientation preserving , it is clear that our mappings @xmath379 combine to a smooth mapping @xmath380 ) \mapsto \operatorname{ad}(k ) \big ( x - 2 i [ h , s ] \big ) , \end{aligned}\ ] ] where @xmath381 stands for the identity on @xmath237 . the final step is to undo the reparametrizations @xmath238 and @xmath221 to obtain @xmath382 , \end{aligned}\ ] ] where @xmath133 and @xmath383 .
since @xmath339 = \mathcal{q}_-$ ] we conclude that @xmath363 is the same as euclid as an integration chain . in this subsection
we prove statement @xmath301 of lemma [ hom_conv_lem ] .
let us first make some general remarks and definitions which allow a simpler discussion of the integrals to be considered .
for this purpose let @xmath385 be the identity on @xmath321 and recall that @xmath238 yields a global factorization of the bundle @xmath386 as @xmath387 .
let @xmath388)$ ] be a left invariant volume form on @xmath389 and let @xmath390 and @xmath391 denote constant volume forms on @xmath144 and @xmath38 respectively .
then there exist functions @xmath392 such that @xmath393 ) \wedge dq_+ \ , .\end{aligned}\ ] ] by inspection of @xmath394 and @xmath395 we see that @xmath392 depends polynomially on @xmath396 , @xmath397 $ ] , the matrix entries of @xmath133 , @xmath398 , and on @xmath399 , where @xmath400 represents any number of partial derivatives with respect to @xmath398 . for the rest of the proof it is more convenient to switch to a formulation in terms of measures instead of volume forms . in that respect
we have @xmath401)| |dq_+| , \label{l_c_lebesgue_integral}\end{aligned}\ ] ] where @xmath402 and @xmath403 are lebesgue measures on @xmath144 and on @xmath38 and @xmath404)|$ ] denotes haar measure on @xmath405 . to prove statement @xmath301 it is enough to show the existence of @xmath406)| . \label{fubini_int}\end{aligned}\ ] ] indeed , the fubini - tonelli theorem then asserts that the original integral exists and that fubini s theorem can be applied to
. we now deal with the integral ( [ fubini_int ] ) .
note that replacing @xmath405 by @xmath196 in introduces only a constant factor which can be absorbed in the polynomial @xmath392 .
the mapping @xmath395 extends naturally from @xmath407 to @xmath408 and , similarly , @xmath365 extends from @xmath409 to @xmath410 .
furthermore we can apply for @xmath343 the transformation @xmath411 in the inner integral over @xmath38 .
the corresponding jacobian is unity .
hence equals @xmath412 where @xmath413 is a haar measure on @xmath196 .
let us now concentrate on the exponential function @xmath414 which is responsible for the convergence of the integral . referring to the second and third lines in
, we define @xmath415\big ) , \end{aligned}\ ] ] which lets us write the integrand in the form @xmath416 due to @xmath417 and @xmath418 , the terms @xmath419 and @xmath420 are imaginary .
therefore , they do not contribute to . to evaluate the remaining terms we note some useful relations .
for @xmath421 and @xmath422 we have @xmath423 @xmath424 can be re - expressed as @xmath425 @xmath338 is positive on @xmath38 , and for all @xmath426 and @xmath427 we have @xmath428 thus @xmath424 is positive definite .
since all dependence on @xmath429 occurs in @xmath424 this guarantees the convergence of the inner integral in for @xmath384 .
we turn to @xmath430 .
statement @xmath312 of lemma [ e_lem ] asserts that @xmath431 for @xmath432 .
hence we have @xmath433 , e_i ) , \end{aligned}\ ] ] where @xmath434 $ ] .
note that @xmath435 since @xmath436 .
the remaining terms of @xmath430 are non - negative since @xmath437,e_i ) & \underset{\mathclap{\eqref{s_decomposition}}}{= } b\big(- \sum_{\mathclap{\alpha \in \sigma_+(\mathcal{q } , \mathfrak{a } ) } } \alpha(h_{j } ) \phi(s_{\alpha } ) , e_{i}\big ) \\ & \underset{\mathclap{\eqref{e_{i , c}_explicit_decomposition}}}{= } \;\ ; -\sum_{\mathclap{\alpha,\beta \in \sigma_+(\mathcal{q } , \mathfrak{a } ) } } e_{i}^{\beta } \operatorname{sgn}(\beta(h_{i } ) ) \alpha(h_{j } ) b(\phi(s_{\alpha}),\phi(s_{\beta } ) ) \\ & = \;\ ; -\sum_{\mathclap{\alpha \in \sigma_+(\mathcal{q } , \mathfrak{a } ) } } e_{i}^{\alpha } |\alpha(h_{j})| b(\phi(s_{\alpha}),\phi(s_{\alpha } ) ) \geq 0 .\end{aligned}\ ] ] since @xmath438 is non - zero , identity @xmath439 ensures that there exists some @xmath250 such that @xmath440 , @xmath441 and @xmath442 .
thus we have @xmath443,e_i ) > 0 .\end{aligned}\ ] ] together with this shows that the second ( or middle ) integral in exists .
this conclusion is not changed by the factor @xmath444 as @xmath445 is linear in the variables @xmath446 whereas @xmath447 is quadratic .
the result depends continuously on @xmath448 and we hence conclude that the outer integral over the compact group @xmath196 exists .
it follows that all integrals @xmath449 exist for @xmath59 .
recall from section [ proof_existence ] that fubini s theorem applies to the integral .
we repeat the steps which led to except that now we do not take the absolute value of the integrand .
thus we obtain @xmath451 the reason why statement @xmath312 of lemma [ hom_conv_lem ] holds true is a very general one : the convergence of the integral to zero is brought about by cancelations due to an oscillatory term . more specifically , by integrating along one special direction in @xmath38
we obtain essentially a regularized delta distribution .
our parametrization is well suited to exhibit this mechanism explicitly .
we will show that it is possible to perform the limit @xmath359 after doing the inner gaussian integrations . in the following ,
einstein s summation convention is in place .
we can choose a basis of @xmath38 with coordinates @xmath452 such that the quadratic form @xmath453 is diagonal .
the choice of basis will be made explicit below .
schematically speaking , the gaussian integrations over @xmath38 in are of the form @xmath454 where @xmath455 , @xmath456 and @xmath457 are functions of @xmath343 , @xmath458 $ ] and @xmath459 .
these functions will be specified as we go along .
now it is possible to introduce sources @xmath460 for @xmath452 and perform the integral : @xmath461 where primes just indicate that these are different polynomials , and the dots represent a dependence on @xmath396 , @xmath398 , @xmath462 and @xmath463 . assuming that @xmath464
, we will show that for @xmath359 we have @xmath465 and @xmath466 for a suitable choice of basis of @xmath38 .
we also show that @xmath467 and @xmath468 for all @xmath469 and @xmath470 $ ] .
the exponential then dominates the polynomial and converges to zero in the limit of @xmath471 .
in addition , we show that @xmath472 , which has the consequence that the remaining integrals over @xmath473 are convergent ( since @xmath474 ) .
thus the issue of convergence is reduced to a discussion of the functions @xmath475 , @xmath476 and @xmath477 .
we start with @xmath475 . reading it off from its definition by @xmath478
we find that it has the expression @xmath479 ) , a ) .\end{aligned}\ ] ] from inequality @xmath480 of lemma [ e_lem ] we infer that @xmath481 , since @xmath343 and @xmath196 is compact .
( recall that @xmath5 is fixed . )
we thus see that @xmath482 for small enough @xmath396 . to discuss the functions @xmath483 and @xmath484 we have to choose a basis of @xmath38 .
for this purpose we fix some @xmath485 , recalling that @xmath486 in the situation at hand .
we then consider the decomposition @xmath487 we define @xmath488 and @xmath489 . denoting by @xmath490 the orthogonal projection onto @xmath38 we introduce @xmath491 we extend @xmath492 to an orthogonal basis @xmath493 of @xmath494 .
we also fix an orthogonal basis @xmath495 of @xmath496 which respects the root decomposition . for the basis vectors
@xmath497 we have identities like those in and .
recall that on @xmath321 we have @xmath498 and therefore @xmath499 if @xmath500 . as is shown by @xmath501
our choice of basis diagonalizes the quadratic form @xmath424 .
we also see that @xmath467 .
in particular , for @xmath502 we have @xmath503 note that @xmath504 for @xmath505 .
it is easy to check that the coefficients @xmath477 defined by @xmath506 are real . by using the statements @xmath301 and @xmath303 of lemma [ e_lem ] we have @xmath507
thus we obtain the result @xmath508 since the dependence on @xmath509 in is governed by the exponential function @xmath510 with @xmath482 and the dependence on @xmath343 is continuous , the limit @xmath359 is uniform .
thus , taking the limit commutes with the outer integrals and we obtain the third part of lemma [ hom_conv_lem ] .
it remains to prove statement @xmath512 of lemma [ hom_conv_lem ] . to that end , for any @xmath346 we introduce on @xmath513 the two functions @xmath514 to prove the desired statement , it is sufficient to show that @xmath515 holds for all @xmath168 .
we will do so by using lebesgue s dominated convergence theorem on both sides of .
let us first establish that the two functions defined in converge pointwise to the same function in the limit of @xmath359 .
for that , we have to distinguish between two situations for @xmath516 : there either exists a non - trivial @xmath295 such that @xmath517 , or there does not . in the first situation
we can apply the result of the previous section to see that @xmath518 for all @xmath343 .
a similar argument yields the same result for the function @xmath519 . in the second situation
there are no problems of convergence with the @xmath38-integral ( see ) and we can directly set @xmath520 , in which case the two functions coincide by definition .
thus we always have @xmath521 . from now on , for brevity ,
we discuss only the left hand side of , as the discussion of the right hand side is completely analogous .
our strategy is to show that the function @xmath522 on the compact domain @xmath523 \times \mathfrak{a}_{\emptyset , c}^+ \times k$ ] is continuous and hence integrable .
the property of continuity on a compact domain implies that the function @xmath524 is dominated by a constant function , which in turn is integrable as well
. thus we will be able to draw the desired conclusion by applying lebesgue s dominated convergence theorem .
following the line of reasoning of the last subsection we choose for @xmath38 an orthogonal basis @xmath525 compatible with the decomposition @xmath526 .
this means that for each @xmath469 there exists a unique root @xmath527 ( possibly the zero root ) such that @xmath528 .
moreover , whenever a root @xmath69 is such that @xmath21 has non - zero projection @xmath529 , then we take @xmath530 to be an element of our basis set @xmath531 .
we arrange for these non - zero vectors @xmath532 to be the first @xmath533 vectors of the set @xmath534 . by adaptation ( due to the change of basis @xmath535 ) of the definitions and we obtain new coefficient functions @xmath536 superseding the earlier functions @xmath483 and @xmath484 . if @xmath527 is the zero root we set @xmath537 .
note that @xmath538 is still real and @xmath539 . by recalling the dependence on @xmath540 of the functions
@xmath189 defined in section [ ps_domain ] , we see that if @xmath359 and if @xmath541 for at least one index @xmath542 , then we have @xmath543 for all @xmath469 with @xmath544 . in view of this behavior ,
the set of problematic points where continuity of the function @xmath545 is not obvious is the set @xmath546 as will be clear presently . to be precise ,
the limit function @xmath547 _ is not even defined _ on this set .
our main work in the rest of this subsection will be to show that it extends continuously as zero .
we will do so by constructing a continuous function which dominates @xmath548 and is zero at the problematic points . in the following
we restrict the discussion to the case of only one summand of the polynomial in .
its modulus is certainly smaller than @xmath549 with a constant @xmath550 and natural numbers @xmath551 .
here we have dropped the factors corresponding to zero roots , as these are of no relevance for our present purpose . fixing some regular element @xmath552 , so that @xmath553 for all @xmath199
, we introduce the functions @xmath554 by a short computation , these have the convenient property that @xmath555 is a continuous function on the compact space @xmath523 \times \mathfrak{a}_{\emptyset , c}^+ \times k$ ] and there exists a constant @xmath556 for which we have the upper bound @xmath557 for the next step , we recall the constancy on @xmath272 of the sign function @xmath558 for @xmath559 . by parts
@xmath303 and @xmath301 of lemma [ e_lem ] and our choice of basis elements @xmath560 for @xmath561 we then have @xmath562 for every @xmath542 . for the following discussion we let @xmath343 be arbitrary but fixed .
inequality guarantees that there exists a neighborhood @xmath563 of @xmath462 so that for each @xmath564 there exists an @xmath565 with the property that @xmath566 on @xmath563 . by inspection of the right hand side of one can see that its behavior ( in the limit @xmath567 and close to the set of problematic points ) is very similar to that of @xmath568 for @xmath569 and positive exponent @xmath570 . to make this observation more tangible
we now show how to simplify the dependence of @xmath571 on @xmath572 . as a first step , we note that only the first factor on the right hand side of @xmath573 is relevant for the discussion of the limit behavior .
now let @xmath574 and write @xmath575 and observing that this formula carries over to our functions @xmath189 , we obtain @xmath576 we claim that this identity yields the following bounds : @xmath577 of which the left one is immediate . to verify the right inequality
we observe that , by the identity preceding it , a stronger statement is @xmath578 owing to @xmath579 this inequality is obviously true if @xmath580 .
so let @xmath581 .
then the two square root factors on the right hand side never vanish and we may divide by them . since the resulting first term on the left hand side is never greater than one ,
the remaining job is to show that @xmath582 this follows from @xmath583 and @xmath584 , which concludes our proof of . as an easy consequence of we
have @xmath585 we now use these bounds to simplify the dependence on @xmath586 on the right hand side of .
iteration gives @xmath587 on the right hand side @xmath588 is meant without summation convention and @xmath589 and @xmath590 are positive constants . by and it follows that @xmath591 the exponential part of the right hand side is continuous and hence we have the bound @xmath592 where @xmath593 and @xmath594 are positive constants . now it is easy to see that the right hand side is essentially a product of continuous functions of the form @xmath595 ( with @xmath596 ) which are composed with continuous functions of the form @xmath597 . thus we obtain a dominating function for @xmath598 on a neighborhood of @xmath462 . in particular
this yields continuity of @xmath598 in each point of the set @xmath599 .
since @xmath462 was taken to be arbitrary we obtain that @xmath598 is a continuous function on @xmath523 \times \mathfrak{a}_{\emptyset , c}^+ \times k$ ] .
thus @xmath524 attains a maximum .
the maximum is a dominating function and hence lebesgue s dominated convergence theorem can be applied .
this finishes the proof of statement @xmath511 in lemma [ hom_conv_lem ] , which was the last step needed to complete the proof of theorem [ thm_2_1 ] . to obtain the theorem when @xmath600 is the direct sum of an abelian and a semisimple lie algebra , let @xmath601 denote the corresponding decomposition of a maximal abelian subalgebra of @xmath39 and replace @xmath144 by @xmath602 and @xmath603 by @xmath604 everywhere in the proof .
in addition let @xmath605 denote the semisimple part of @xmath37 and replace @xmath37 by @xmath605 everywhere in the proof .
it is possible to choose different regularization functions @xmath606 .
the choice made here seems natural , as it has the highest invariance possible and was also used in earlier work .
the convergence properties can be seen quite clearly in the discussion of @xmath607)$ ] .
the convergence is not uniform in @xmath5 . to have uniform convergence , we need @xmath608 . in applications with @xmath609 ,
one has to replace @xmath5 by @xmath610 . for fixed @xmath611
this gives uniform convergence in @xmath5 . in this section
we establish two different representations of the integral over the ps domain .
these are stated in corollaries [ cor_2_1 ] and [ cor_2_2 ] .
recall that the ` jacobian ' @xmath78 which appears in both representations may have alternating sign . in the following proof of corollary [ cor_2_1 ]
we pinpoint the origin of these surprising signs .
first we review the setting .
recall that the elements of @xmath37 are antihermitian and those of @xmath38 hermitian .
since @xmath612 is closed under hermitian conjugation , it is the direct sum of an abelian and a semisimple lie algebra .
the semisimple part of @xmath613 is denoted by @xmath614 .
we choose a maximal abelian subalgebra @xmath62 of @xmath38 containing @xmath21 .
the decomposition of @xmath613 into an abelian and a semisimple part induces a decomposition of @xmath615 and @xmath616 . here
@xmath617 and @xmath618 lie in the abelian part of @xmath613 while @xmath619 and @xmath620 lie in the semisimple part of @xmath621 .
let @xmath622 denote a positive open weyl chamber in @xmath619 with respect to the semisimple lie algebra @xmath614 .
we also define @xmath623
. then we have the following reparametrization : @xmath624,h_s , h_a ) & \mapsto ( y , k ( h_s + h_a)k^{-1 } ) .\end{aligned}\ ] ] recall that @xmath625 is closed by assumption and @xmath626 denotes the closed and analytic subgroup of @xmath627 with lie algebra @xmath628 .
the subgroup @xmath629 is central and closed . by the diffeomorphism @xmath630 and the isomorphism @xmath631 we have the reparametrization @xmath632 , h_s , h_a ) & \mapsto(y , k ( h_s + h_a ) k^{-1 } ) .\end{aligned}\ ] ] by @xmath633 and the cartan decomposition @xmath634 ( see @xcite ) we obtain yet another parametrization of the ps domain , @xmath635 , h_s , h_a ) & \mapsto g ( h_s + h_a ) g^{-1 } , \end{aligned}\ ] ] which is the one most frequently used in the literature . to proceed with the proof of corollary [ cor_2_1 ]
, we have to diagonalize the commutator action of @xmath62 on @xmath613 and on @xmath68 . for this purpose
we note that the values @xmath636 of the roots @xmath69 are real since @xmath637 $ ] is hermitian with respect to the hermitian form @xmath638 .
moreover @xmath639 for @xmath640 .
the pullback of @xmath60 by @xmath641 is then @xmath642 ) \wedge dh , \end{aligned}\ ] ] where @xmath643)$ ] is a left invariant volume form on @xmath644 and @xmath390 is a constant volume form on @xmath62 .
denoting by @xmath70 the dimension of the root space corresponding to @xmath69 , we get @xmath645 note that @xmath646 differs from @xmath84 in corollary [ cor_2_1 ] only by taking the modulus of the roots in @xmath647 .
but these roots @xmath648 are positive when evaluated on @xmath622
. therefore we have the following equality : @xmath649 ) \wedge dh , \end{aligned}\ ] ] where @xmath381 denotes the identity on @xmath650 . at this point a crucial difference between the roots in @xmath651 and those in @xmath67 is detected : since the definition of the weyl chamber @xmath652 refers only to the former roots , it is possible for the latter roots to change sign on @xmath652 .
these sign changes are particulary evident in our approach as we are integrating differential forms instead of densities ( or measures ) .
now it is convenient to replace the volume form @xmath643)$ ] by the left invariant measure @xmath653)|$ ] and @xmath390 by lebesgue measure @xmath402 on @xmath654 @xmath655)| |dh | .\end{aligned}\ ] ] the replacement of @xmath656 by @xmath626 simply leads to a change of normalization constant @xmath657 : @xmath658 where @xmath659 denotes haar measure on @xmath626 . let @xmath660 denote the normalizer of @xmath661 in @xmath662 . in the following we make use of the weyl group @xmath663 .
this weyl group acts on @xmath619 and generates @xmath619 from @xmath664 . to exploit this property
we need that @xmath84 is invariant under the action of the weyl group . recall that @xmath84 is given by @xmath665
the first factor is trivially invariant , whereas for the second factor an additional argument is needed . for that purpose
we define a root @xmath666 to be positive if @xmath667 .
this definition makes sense because @xmath668 for all roots @xmath669 .
since @xmath21 is @xmath248 invariant we conclude that the action of the weyl group does no more than permute the roots in @xmath670 .
hence @xmath671 is weyl - invariant and so is @xmath84 .
now the haar measure @xmath659 is @xmath626-bi - invariant and hence weyl - invariant . therefore , introducing another normalization constant @xmath672 we have @xmath673 by setting @xmath674 we obtain corollary [ cor_2_1 ] . since @xmath53 is nearly everywhere injective and regular by assumption , so is @xmath641 .
application of the change of variable theorem then yields corollary [ cor_2_2 ] .
here we give the detailed argument showing that for our purpose of integrating over @xmath53 and @xmath675 the contributions from the boundary @xmath335 of the weyl chamber @xmath144 are irrelevant , as they are of codimension at least two . without loss , we fix any @xmath346 and let @xmath676 be any one of the generators of @xmath177 which also lie in @xmath677 ( if there is no such generator then there is nothing to prove ) . by removing this generator
we get a boundary component @xmath678 next recall the definition of @xmath212 in .
we now show that by restricting to @xmath679 in the leftmost factor of the domain of definition of @xmath212 we get a domain of codimension at least two . here
the main observation is that the dimension of the isotropy group of @xmath134 changes at the boundary of @xmath144 and , in particular , @xmath680 .
this is seen as follows .
each face of @xmath681 lies in the zero locus @xmath682 of some root @xmath683 and we can arrange for @xmath684 . if @xmath685 is the root space of @xmath69 , the group generated by @xmath686 leaves the face @xmath687 invariant . when restricting @xmath688 in the first factor to @xmath679 we may replace the second factor @xmath689 by the lower dimensional space @xmath690 without changing the image of the parametrization .
thus the reduction @xmath691 is accompanied by a reduction of dimension of the @xmath196-orbits on @xmath692 .
altogether , the dimension is reduced by no less than two . moreover ,
the eigenspace decomposition of @xmath33 with respect to @xmath134 is a refinement of the eigenspace decomposition w.r.t . the smaller abelian algebra @xmath693 .
hence our further reparametrizations of ps ( by @xmath252 and @xmath275 , which rely on an eigenspace decomposition of @xmath33 ) are compatible with the restriction of @xmath134 to @xmath694 .
this completes the argument for @xmath212 .
turning to @xmath351 , we have to argue that the analogous restriction is still well defined . for that
, it is enough to note that for @xmath695 we have @xmath696 = 0 $ ] if @xmath697 . by this token
we see that also for @xmath351 the contributions from @xmath335 are of codimension at least two .
a detailed discussion of the sw domain and the validity of the pertinent hubbard - stratonovich transformation can be found in @xcite .
here we give another proof by deforming @xmath55 into @xmath54 . by using some of the constructions of the proof for the ps domain
, this deformation can be stated very explicitly .
we start with a brief discussion of the convergence of the gaussian integral ( [ motiv_identity ] ) over @xmath698 for @xmath133 and @xmath383 we have that @xmath699 and @xmath700 is constant .
the cross term @xmath701 is purely imaginary , and @xmath702 for @xmath703 yields convergence in the @xmath39 directions . to see the properties of @xmath55 more explicitly
, we use the reparametrization @xmath221 and the decomposition of @xmath21 to obtain @xmath704,x ) \mapsto x - i b\operatorname{ad}(k ) \big ( s_0 + \sum_{\mathclap{\alpha \in \sigma_+(\mathcal{q } , \mathfrak{a } ) } } \big(\cosh(\alpha(h ) ) s_{\alpha } + \sinh(\alpha(h ) ) \phi(s_{\alpha } ) \big ) \big ) .\end{aligned}\ ] ] from this parametrization we see that the image of the boundary of @xmath144 is again of codimension at least two , which clearly shows that @xmath705 .
a homotopy from @xmath55 to @xmath54 is given by @xmath706 & \times \mathfrak{a}^+ \times k / z_{k}(\mathfrak{a } ) \times \mathcal{q}_+ \rightarrow \mathcal{q}^{\mathbb{c } } , \\
( t , h,[k],x ) & \mapsto x - ib\operatorname{ad}(k ) \big [ ( 1-t)s_0 + \\ \sum_{\mathclap{\alpha \in \sigma_+(\mathcal{q } , \mathfrak{a } ) } } & [ ( \cosh((1-t)\alpha(h))-t ) s_{\alpha } + \frac{\sinh((1-t ) \alpha(h))}{1-t } \ , \phi(s_{\alpha})]\ , \big ] .\end{aligned}\ ] ] note that @xmath707 and @xmath708,x ) = x - i b[khk^{-1 } , s]$ ] . since @xmath339 = \mathcal{q}_-$ ]
we obtain @xmath709 . to complete the argument
we show that the integral over @xmath710 is convergent . for this
we note that for @xmath711,x)$ ] we have @xmath712 where the dots represent unimportant terms ; these are terms which are purely imaginary , terms which are linear in @xmath713 and all terms containing @xmath714 . owing to @xmath699 and @xmath715
we obtain convergence for @xmath716 . for @xmath717 convergence
is ensured by the @xmath718 term , as was discussed above for the @xmath55 parametrization .
10 f. wegner , _ the mobility edge problem : continuous symmetry and a conjecture _
, z. phys .
b * 36 * , 207 ( 1979 ) . l. schfer and f. wegner , _ disordered systems with n orbitals per site : lagrange formulation , hyperbolic symmetry , and goldstone modes _ , z. phys .
b * 38 * , 113 ( 1980 ) .
a.m.m . pruisken and l. schfer , _ the anderson model for electron localization nonlinear sigma model , asymptotic gauge invariance _ , nucl .
b * 200 * , 20 ( 1982 ) .
efetov , _ supersymmetry and the theory of disordered metals _ , adv . phys . * 32 * , 53 ( 1983 ) .
efetov , _ supersymmetry and the theory of disordered metals _ ( cambridge university press , cambridge , 1997 ) .
zirnbauer , _ supersymmetry methods in random matrix theory _ , encyclop . of math .
vol . 5 , p.151 - 160 ( elsevier , amsterdam , 2006 ) .
fyodorov , _ on hubbard - stratonovich transformations over hyperbolic domains _ , j. phys . :
condensed matter * 17 * , s1915 ( 2005 ) .
y. wei and y.v .
fyodorov , _ a conjecture on hubbard - stratonovich transformations for the pruisken - schfer parametrizations of real hyperbolic domains _ , j. phys . a * 40 * , 13587 ( 2007 ) .
fyodorov , y. wei , and m.r .
zirnbauer , _ hyperbolic hubbard - stratonovich transformation made rigorous _
, j. math .
* 9 * , 53507 ( 2008 ) .
verbaarschot , h.a .
weidenmller , and m.r .
zirnbauer , _ grassmann integration in stochastic quantum physics - the case of compound nucleus scattering _ , phys
* 129 * , 367 ( 1985 ) .
knapp , _ lie groups beyond an introduction _
( birkhuser , boston , 2005 , second edition , third printing ) . g.m .
ziegler , _ lectures on polytopes _
( springer - verlag , new york , 1994 ) .
de loera , j. rambau and f. santos , _ triangulations _ ( spinger - verlag , berlin , 2010 ) .
zirnbauer , _ riemannian symmetric superspaces and their origin in random - matrix theory _ , j. math
* 37 * , 4986 ( 1996 ) . | we settle a long standing issue concerning the traditional derivation of non - compact non - linear sigma models in the theory of disordered electron systems : the hyperbolic hubbard - stratonovich ( hs ) transformation of pruisken - schfer type .
only recently the validity of such transformations was proved in the case of @xmath0 ( non - compact unitary ) and @xmath1 ( non - compact orthogonal ) symmetry . in this article
we give a proof for general non - compact symmetry groups .
moreover , we show that the pruisken - schfer type transformations are related to other variants of the hs transformation by deformation of the domain of integration .
in particular we clarify the origin of surprising sign factors which were recently discovered in the case of orthogonal symmetry . |
supernovae of type b / c constitute about one third of the death rate of massive stars @xcite .
their spectra lack hydrogen @xcite , suggesting progenitors stripped of their hydrogen envelopes either due to mass transfer in a binary system or via copious stellar winds . on theoretical grounds ,
the anticipated progenitors are wolf rayet ( wr ) stars or massive helium stars ( e.g. , * ? ? ?
* ; * ? ? ?
the most direct way to test the above theoretical picture is direct detection of the progenitor .
this is possible for supernovae occurring in nearby galaxies which have deep pre - explosion images ( which means , in practice , _ hst _ ) .
@xcite comprehensively summarize searches for twelve b / c progenitors , which are all non - detections .
the deepest upper limit to date is @xmath11mag for sn2002ap , a type c supernova in the very nearby galaxy messier 74 @xcite .
an indirect way to infer the progenitor properties is the early light curve of a supernova , which is sensitive to the progenitor size , explosion energy and the composition of the outer layers ( e.g. , * ? ? ?
? * ; * ? ? ?
* ; * ? ? ?
* ) , and in some propitious cases can even diagnose the presence of a binary companion @xcite .
no detection of shock cooling constrains the radius of ptf10vgv ( type c ) to @xmath12 @xcite . for sn2008d ( type b )
, the shock cooling phase lasted about five days and the inferred radius is @xmath13 @xcite .
radio measurements serve as another indirect probe of the progenitor system by characterizing the mass - loss history .
radio emission is produced by relativistic electrons accelerated in the supernova shock as they gyrate in the amplified magnetic field when the shock expands freely .
the radio spectral evolution implies a mass loading parameter @xmath14 where @xmath15 is a constant mass - loss rate and @xmath16 is the wind velocity from the progenitor @xcite .
type b / c supernovae typically have @xmath17 of the order of 10@xmath6 g @xmath7 @xcite . in this letter
, we report the discovery of iptf13bvn , a young type b supernova in the nearby galaxy ngc5806 .
hst _ images of ngc5806 allow a direct progenitor search
. we also present early photometric , spectroscopic and wide - band radio observations and discuss implications on the progenitor .
on 2013 jun 16.238 ut , the automated real - time discovery and classification pipeline of the intermediate palomar transient factory ( iptf ; * ? ? ? * nugent et al . in prep ) identified a new transient source with @xmath18=18.6mag in the vicinity of ngc5806 ( see figure [ fig : discovery ] , @xcite ) .
no source was detected at the same location to @xmath19mag ( 3@xmath0 ) on jun 15.240 ( figure [ fig : lc ] ) .
also , there was no evidence for pre - outburst activity in 492 ptf images taken since 2009 jun 29 to a similar depth .
our duty astronomer saved this source as iptf13bvn and initiated rapid , multi - wavelength follow - up .
on jun 20.276 , we observed iptf13bvn in @xmath20-band with osiris @xcite and the laser guide star adaptive optics ( lgs - ao ) system @xcite mounted on the 10 m keck telescope . registering the ao image to the archival
_ hst_/acs image , we get a 1@xmath0 uncertainty of @xmath2140mas .
we find one source coincident with the supernova within the 2@xmath0 ( equivalently 8.7pc projection distance ) error circle ( figure [ fig : discovery ] ; @xcite ) .
we perform psf photometry on the _
hst_/acs images with dolphot @xcite .
the photometry of the progenitor candidate is @xmath22mag in f435w , @xmath23mag in f555w and @xmath24mag in f814w . to correct for extinction ,
we obtain a high resolution spectrum ( [ sec : obs ] ) and measure the equivalent widths of na d lines .
we find local extinction of @xmath25 and foreground extinction of 0.0278 @xcite . assuming @xmath26 @xcite and adopting a distance modulus of 31.76@xmath27 ( 22.5mpc ; @xcite ) , we find @xmath28mag , @xmath29mag and @xmath30mag . thus the progenitor , if single , is no brighter than these values
. the luminosity and colors are consistent with the compilation of wn and wc wolf rayet stars in @xcite . however , wr stars are often in binaries @xcite .
thus , the progenitor is possibly in a binary system and the light is dominated by the companion .
we further note that color alone can not be used as a discriminant .
the b , v and i filters are in the rayleigh - jeans tail of hot stars , including o - stars , wr stars and blue supergiants .
moreover , given the @xmath31 pixel size of _
hst_/acs , or equivalently 5.45pc at the distance of ngc5806 , the candidate can also be an unresolved young star cluster whose color is dominated by ob stars .
finally , we caution that the progenitor candidate may even be unrelated to the supernova
. the litmus test of whether this candidate is the progenitor or part of the progenitor system can only be undertaken by _
hst _ imaging after the supernova fades .
as part of regular iptf operations , the field of iptf13bvn was imaged twice a night every night by the palomar 48-inch oschin telescope ( p48 ) with a mould @xmath32-band filter @xcite during the spring quarter . upon discovery of iptf13bvn , the robotic palomar 60-inch telescope ( p60 ; * ? ? ?
* ) was triggered for follow - up in @xmath33@xmath34@xmath35@xmath36-bands .
we obtained photometry with the las cumbres observatory global telescope ( lcogt ; @xcite ) network in @xmath37-bands using the 1-m telescopes from cerro tololo ( chile ) , mcdonald observatory ( usa ) and sutherland ( south africa ) , along with the 2 m faulkes telescope south ( siding springs ) .
as part of our ongoing iptf-_swift _ program , we triggered target - of - opportunity observations beginning on 2013 jun 17 @xcite .
in the p48 , p60 and _ swift _ images , the host background is subtracted by using pre - explosion reference images , while in the lcogt images a low - order polynomial fit is used to remove the background .
psf photometry is then performed in all the images .
photometry in @xmath38-bands is calibrated to sdss stars and that in @xmath39-bands is calibrated with landolt standard stars .
the multi - color light curve of iptf13bvn is illustrated in figure [ fig : lc ] .
low - resolution spectroscopic follow - up of iptf13bvn was undertaken with the dolores low - resolution spectrograph on telescopio nazionale galileo ( tng ) , the marcario low - resolution spectrograph ( lrs ; @xcite ) on the hobby - eberly telescope ( het ) , the low - resolution , cross - dispersed spectrograph floyds ( sand et al . in prep ) on the robotic faulkes telescope ( ft ) , the dual imaging spectrograph ( dis ) on the arc 3.5-m telescope , the folded - port infrared echellete ( fire , * ? ? ?
* ) on the 6.5-m magellan telescope , and spex @xcite on the nasa infrared telescope facility ( irtf ) .
we also obtained a high - resolution spectrum with high resolution echelle spectrometer ( hires ; * ? ? ? * ) on the 10 m keck i telescope .
the spectroscopic series of iptf13bvn is displayed in figure [ fig : spec_seq ] .
all photometry tables and spectroscopy data will be made public via wiserep @xcite . in order to better constrain the explosion date
, we fit a power - law model @xmath40 to the p48 data of iptf13bvn taken in the first five nights after explosion and constrain the parameters with the upper limits preceding the discovery . the best fit results in an explosion date @xmath41 jun 15.67 and a power law index @xmath42 ( figure [ fig : spec_comp ] ) .
the @xmath43 confidence levels for @xmath44 and @xmath45 are @xmath46 $ ] and @xmath47 $ ] , respectively .
note that the color evolution is minimal , suggesting that r - band is a reasonable proxy for the bolometric light curve .
iptf13bvn peaked at @xmath4816.6mag at about @xmath4918days .
a direct comparison of the iptf13bvn @xmath18-band light curve with sn1994i @xcite , sn2008d @xcite , ptf10vgv
@xcite and ptf12gzk @xcite is also shown in figure [ fig : lc ] .
the type c sn1994i is much more luminous than iptf13bvn while its rise rate is less than that of iptf13bvn .
the type b sn2008d has a five day cooling phase after its shock breakout before it slowly rises again , while we do not detect any shock cooling signature from iptf13bvn .
the energetic type ic supernova ptf12gzk rises much faster than iptf13bvn .
surprisingly , the closest match to the rise rate of iptf13bvn through 8days is the high velocity type ic ptf10vgv ( albeit about two magnitudes fainter ) .
following @xcite , we derive a constraint on the progenitor radius based on our detection limits on shock cooling .
the inferred radius is sensitive to the `` mean '' velocity of the ejecta @xmath50 where @xmath51 is the explosion energy and @xmath52 is the mass of the ejecta .
based on the measured photospheric velocities and line velocities ( [ sec : spec ] ) , we conservatively adopt the lowest velocity of 8000kms@xmath9 .
if the velocity was higher , the radius of the progenitor would be constrained to be even smaller @xcite .
the model also depends weakly on @xmath51 and thus we assume @xmath53erg . in the right bottom panel of figure
[ fig : lc ] , the predicted @xmath18-band light curves ( gray lines ) of shock cooling with various explosion times are also plotted .
we find : if the explosion had happened earlier than t@xmath54 , the progenitor radius would have to be unreasonably small ( @xmath55 ) because the luminosity of the shock breakout is tightly constrained by non - detection on jun 14 .
if the explosion happened at t@xmath54 , the progenitor radius @xmath56 .
if the explosion happened at t@xmath54 + 0.5day , the progenitor radius @xmath57 .
we conclude that the progenitor radius was no larger than a few solar radii , suggesting a stripped core such as wr stars .
the strongest features in the early spectra are the ca h@xmath58k and ca near - ir triplet absorption .
the blends of fe lines at @xmath59 and fe , mg and ti at @xmath60 are also prominent .
the local minimum at about @xmath61 may be either he or na and that at about @xmath62 may be interpreted as either si or ne . at + 2.6days after explosion , a synow fit gives a photospheric velocity of 10000kms@xmath9 and a ca ii line velocity of 14000kms@xmath9 . by + 11.5days
, these velocities evolve to 8000 and 10000kms@xmath9 respectively .
helium lines are not expected to be prominent in early spectra of type ib supernovae @xcite .
we begin to see weak wiggles at the location of he i ( 5876 6678 and 7065 ) in optical spectra after + 10days .
however , synow can not fit a single expansion velocity to all the three lines simultaneously . in the ir spectra
, we see a prominent 10200 absorption feature , which may be explained by he 10830 line blended with metal lines .
the 20581 doublet is often used to confirm the presence of he @xcite . by + 16.7days
, we unambiguously detect this feature ( see the inset of figure [ fig : spec_seq ] ) .
therefore , we classify iptf13bvn as a type b supernova @xcite .
promptly after discovery , we initiated radio follow - up observations at the combined array for research in millimeter - wave astronomy ( carma ) and requested director s discretionary time at the jansky very large array ( jvla ) .
our jvla and carma data showed that the radio counterpart had a synchrotron self - absorption ( ssa ) spectrum . on being denied our request for additional jvla observations , our colleagues were able to initiate australia telescope compact array ( atca ) observations .
the jvla data were reduced using aips with j@xmath63 as a phase calibrator and 3c286 as a flux calibrator .
the carma data were reduced using miriad @xcite with j@xmath64 as a phase calibrator and mwc349 as a flux calibrator .
the atca data were reduced using miriad with pksb1921 - 293 as a bandpass calibrator , pksb1508 - 055 as a phase calibrator and pksb1934 - 638 as a flux calibrator .
all the radio data is presented in figure [ fig : radio ] .
we fit the multi - frequency multi - epoch data to the ssa model following @xcite .
we assume that the electron energy distribution can be described by a power law @xmath65 .
the fitted models are also presented in figure [ fig : radio ] .
we note that the ssa model alone is consistent with all our data including the null - detection in c band on jun 18 .
however , on jun 19.1 , @xcite observed iptf13bvn with the jvla in c band and announced a null - detection with an rms of @xmath66jy .
based on our ssa model , we estimate that kamble & soderberg should have detected iptf13bvn at a level of @xmath67 .
their null detection therefore may require additional free - free absorption at low frequencies .
based on inverse compton scattering , we estimate the x - ray emission to be @xmath68 .
this is consistent with our _ swift_/xrt upper limit of @xmath69 on jun 17 ( assuming a power - law spectrum with a photon index of 2.0 ) .
we note that @xcite reported an x - ray detection by co - adding the xrt data from jun 17 to jul 13 .
we reprocessed the xrt data from the first ten days after the supernova explosion and resulted in a 3-@xmath0 upper limit of @xmath70 .
this is also consistent with our expectation from inverse compton scattering .
we find that the radio emission peaks at @xmath71ghz with fluxes of @xmath72mjy on jun 18 , 21 and 24 , respectively .
following @xcite , we derive a shockwave radius @xmath73 and a magnetic field strength @xmath74 . using the explosion date determined in the optical data ,
we find a shock wave velocity of @xmath75 , which is typical for sneb / c @xcite .
we derive a mass - loading parameter @xmath76g@xmath7 ( see * ? ? ?
* for equations ) .
assuming a typical wind velocity of @xmath771000 km s@xmath9 from a wr star @xcite , the progenitor mass loss rate @xmath78 .
this is consistent with the wr mass loss rate in @xcite .
iptf13bvn is a type b supernova that exploded on jun 15.67 and rose to maximum luminosity of [email protected] in about 18days .
the luminosity of the supernova in the first ten days is approximately proportional to @xmath80 .
we identify a single progenitor candidate within a 2@xmath0 error radius of 8.7pc in pre - explosion _ hst _ imaging .
the candidate has a m@xmath1 luminosity of @[email protected] and a @xmath4 color of [email protected] .
hst _ imaging , after the supernova fades away , will determine whether this is a single star , a binary or a star cluster .
the non - detection of shock cooling in our light curve constrains the progenitor radius to smaller than a few solar radii .
our radio detections suggest a shock wave with velocity of @xmath81 km s@xmath9 and a progenitor mass loss rate of @xmath82 .
we conclude that the pre - explosion photometry of the detected candidate , the radius constraint based on absence of shock cooling , and the mass loss rate derived from radio are all consistent with a wolf rayet progenitor .
we thank a. l. piro for valuable discussions .
we thank the following people for co - operating with our target of opportunity or queue observations : m. roth ( magellan ) , a. hartuynan ( tng ) , j. johnson ( keck ) , j. caldwell ( het ) .
we thank a. howard and h. isaacson for hires data reduction .
we thank r. campbell , hien tran and s. tendulkar for helping with osiris lgs - ao observation and data reduction .
we thank p. vreeswijk for assisting with _ hst _
image registration .
we thank j. vinko , r. foley , b. kirshner , d. perley , a. corsi and k. mooley as proposal co - is .
we thank j. swift , b. montet , m. bryan , r. jensen - clem , d. polishook , s. tinyanont for assisting with observations .
m. m. k. acknowledges generous support from the hubble fellowship and carnegie - princeton fellowship .
jms is supported by an nsf astronomy and astrophysics postdoctoral fellowship under award ast-1302771 .
n. d. acknowledges the hubble fellowship .
research by agy and his group was supported by grants from the isf , bsf , gif , minerva , the eu / fp7 via an erc grant and the kimmel award .
the research of jcw is supported by nsf grant ast 11 - 09801 .
the national energy research scientific computing center , supported by the office of science of the u.s .
department of energy , provided staff , computational resources , and data storage for this project . the australia telescope is funded by the commonwealth of australia for operation as a national facility managed by csiro .
this research has been supported by the australian research council through super science fellowship grant fs100100033 .
the centre for all - sky astrophysics is an australian research council centre of excellence , funded by grant ce110001020 .
the national radio astronomy observatory is a facility of the national science foundation operated under cooperative agreement by associated universities , inc . ongoing carma development and operations
are supported by the national science foundation under a cooperative agreement , and by the carma partner universities . ,
g. j. , nicklas , h. e. , macqueen , p. j. , tejada , c. , cobos duenas , f. j. , & mitsch , w. 1998 , in society of photo - optical instrumentation engineers ( spie ) conference series , vol . 3355 , society of photo - optical instrumentation engineers ( spie ) conference series , ed . s. dodorico , 375386 , r. j. , teuben , p. j. , & wright , m. c. h. 1995 , in astronomical society of the pacific conference series , vol .
77 , astronomical data analysis software and systems iv , ed .
r. a. shaw , h. e. payne , & j. j. e. hayes , 433 , s. s. , et al .
1994 , in society of photo - optical instrumentation engineers ( spie ) conference series , vol . 2198 , society of photo - optical instrumentation engineers ( spie ) conference series , ed .
d. l. crawford & e. r. craine , 362 , @xmath83 ( j2000 ) .
panel ( a ) shows the _ hst _ image of ngc5806 , taken on 2005 - 03 - 10 ut ( _ hst _ proposal 10187 , pi smartt ) .
panel ( b ) is a zoom - in of the _ hst _ image near the site of iptf13bvn .
panel ( c ) is the composite image of the supernova taken with osiris and lgs - ao system .
this is made by stacking sixty 15s exposures .
we align this image with the _
hst_/acs images with 15 , 23 , 25 registration stars in the f435w , f555w and f814w filters and acquire registration uncertainties of 0.6 , 0.9 and 0.8 _ hst _ pixel , respectively .
panel ( d ) , ( e ) and ( f ) show the _ hst _ images at the position of the supernova in f435w , f555w and f814w filters .
the position of iptf13bvn is marked with 1-@xmath0 and 3-@xmath0 error circles .
the progenitor candidate is @xmath22mag in f435w ( @xmath84 ) , @xmath23mag in f555w ( @xmath85 ) and @xmath24mag in f814w ( @xmath86 ) .
[ fig : discovery],scaledwidth=80.0% ] is set to the explosion date derived from the best power - law fit ( solid gray curve ) . for comparison ,
@xmath18-band light curves of sn1994i , sn2008d , ptf10vgv and ptf12gzk are also shown in cyan with different line styles .
on the top axis , the epochs of spectroscopic follow - up are indicated by s " . in the right panel ,
the p48 @xmath18-band light curves are plotted against predicted light curves of shock cooling ( gray dash curves ) from @xcite for explosion starting at @xmath87days with a `` mean '' velocity of @xmath88kms@xmath9 .
the radius of the progenitor in each of the models is besides its corresponding light curve .
[ fig : lc ] ] ) in c band ( 6ghz ) at + 3d .
the purple triangle is the jvla detection limit ( @xmath89 ) on + 4d obtained by @xcite .
the solid lines show our fits using synchrotron self - absorption models .
additional carma measurements on + 4d , and + 8d ( black stars ) and an upper limit on + 10d ( cyan triangle ) are shown .
[ fig : radio],scaledwidth=90.0% ] | the intermediate palomar transient factory reports our discovery of a young supernova , iptf13bvn , in the nearby galaxy , ngc5806 ( 22.5mpc ) .
our spectral sequence in the optical and infrared suggests a type b classification .
we identify a blue progenitor candidate in deep pre - explosion imaging within a 2@xmath0 error circle of 80mas ( 8.7pc ) .
the candidate has a m@xmath1 luminosity of @[email protected] and a @xmath4 color of [email protected] .
if confirmed by future observations , this would be the first direct detection for a progenitor of a type b. fitting a power law to the early light curve , we find an extrapolated explosion date around 0.6days before our first detection .
we see no evidence of shock cooling .
the pre - explosion detection limits constrain the radius of the progenitor to be smaller than a few solar radii .
iptf13bvn is also detected in cm and mm - wavelengths .
fitting a synchrotron self - absorption model to our radio data , we find a mass loading parameter of 1.3@xmath510@xmath6 g @xmath7 . assuming a wind velocity of @xmath8kms@xmath9
, we derive a progenitor mass loss rate of @xmath10 yr@xmath9 .
our observations , taken as a whole , are consistent with a wolf rayet progenitor of the supernova iptf13bvn . |
the process of accretion onto compact objects is still not fully understood . how accretion leads to the production of relativistic , collimated jets , how much power
is contained within these jets and how that power is distributed internally , are some of the major open questions in astronomy .
recently it has been established that stellar - mass black holes ( bhs ) can channel a large fraction of their accretion energy into these jets @xcite .
these systems are x - ray binaries ( xbs ) , in which the accreting matter is fed from a companion star via an accretion disc towards the bh . a large scale height , poloidal magnetic field in the inner accretion flow is thought to launch the jet through magnetohydrodynamical processes , but it remains unclear how jets from accreting objects ( active galactic nuclei [ agn ] , xbs , @xmath0-ray bursts ) are formed and accelerated to relativistic velocities ( likely by extraction of either the spin energy of the compact object , or the accretion energy from the accretion disc , or a combination ; e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . for bhxbs , classical ` flat spectrum '
radio jets ( like those seen in agn ) are commonly observed during hard x - ray states ( hereafter the hard state ; for descriptions of x - ray states see * ? ? ?
* ; * ? ? ?
* ) , when the accretion flow structure likely permits the existence of a large , vertical magnetic field . in softer x - ray states jets
are observed to be quenched at radio frequencies ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) which may result from a suppression of the poloidal field by the geometrically thin disc which exists in the soft state ( e.g. * ? ? ?
the flat , or slightly inverted ( @xmath90.5 where @xmath10 ) radio spectrum seen in the hard state extends to at least the millimetre regime @xcite , which can be explained by self - absorbed synchrotron emission from lepton populations at different radii from the bh ; the signature of a conical , collimated jet @xcite . at near - infrared ( nir ) and higher frequencies , optically thin synchrotron emission has been detected ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , usually approximating a power law with spectral index @xmath11 .
this is expected from a compact jet spectrum at higher frequencies than the self - absorption _ break _ , or _
turnover _ in the spectral energy distribution ( sed ; * ? ? ?
* ; * ? ? ?
the jet origin of this optically thin synchrotron emission is supported by the extrapolation of the radio spectrum to nir frequencies ( e.g. * ? ? ?
? * ; * ? ? ?
* ) , correlated radio and ir variability ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , correlations with radio emission in the hard state @xcite , quenching during the soft state @xcite and the recent detection of linear polarization @xcite .
the majority of the fast ( @xmath12 seconds or less ) variability reported at optical / nir wavelengths is also likely to have a jet origin , since the variability has a spectrum consistent with optically thin synchrotron emission ( e.g. * ? ? ?
* ; * ? ? ?
* ) , and is stronger at lower frequencies ( e.g. * ? ? ?
* ; * ? ? ?
* ) which is inconsistent with both thermal emission and a nonthermal corona origin ( for a recent nonthermal model see * ? ? ?
* ) . physically , the size scale of the emitting region in the jet scales inversely with frequency .
it is thought that the jet break frequency , @xmath13 marks the start of the particle acceleration in the jet ( e.g. * ? ? ?
* ) , at distances only @xmath14 ( gravitational radii ) from the bh ( e.g. * ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
radiatively , the energetic output of the jet is dominated by the higher frequencies , and the peak flux density of the jet spectrum is at @xmath13 . in order to estimate the total power contained in the jet , to infer the fraction of accretion energy being channelled into these outflows , it is therefore necessary to identify the spectral break , and also the high energy synchrotron cooling break .
the total radiative luminosity of the jet can only be inferred via @xmath13 , the luminosity at the break , @xmath15 and the radiative efficiency , @xmath16 .
measuring as accurately as possible the power contained in the jets at different luminosities is key to understanding the process of jet formation and the overall physics of accretion and the matter and energy xbs input into the interstellar medium ( ism ) . during outbursts ,
xbs can vary by eight orders of magnitude in luminosity , so it is possible to study how the accretion process and jet properties vary with mass accretion rate from luminosities @xmath171 @xmath18 .
in addition to the total kinetic energy , other physical properties of the jet can be inferred by identifying the jet break . the cross section radius of the first acceleration zone ( hereafter faz ; i.e. the size of the region where the synchrotron power law starts ) and magnetic field strength can be estimated directly @xcite , while the velocity and opening angle can be inferred from fast timing fluctuations at the faz @xcite and magnetic field ordering and orientation , via polarization at the faz @xcite . for jet models , observables like these
are highly sought after .
constraining these parameters and how they vary with luminosity will dramatically improve attempts to model jets and simulate their production . unlike in agn
, the rapid time dependency of xbs allows us to probe jet evolution , adding a further dimension to models ( e.g. * ? ? ?
despite its importance , the jet break has only yet been observed directly in the spectrum of one bhxb ( gx 3394 ; @xcite ; see also @xcite ) and one neutron star xb ( 4u 0614 + 09 ; * ? ? ?
* ; * ? ? ?
* ) and inferred via spectral fitting in one bhxb ( cyg x1 ; * ? ? ?
in all three objects , the break is inferred to exist in the mid - ir , at @xmath19 hz , with gx 3394 displaying large variability in @xmath13 ( by @xmath20 order of magnitude ) on hour - timescales @xcite .
@xcite show that the mid - ir spectrum of a062000 in quiescence is flat , and consistent with being self - absorbed synchrotron emission , while the nir has a different power law , and is consistent with being optically thin synchrotron , which strongly favours a jet break between the two .
the reasons why few jet breaks in xbs have been discovered to date are because the companion star or accretion disc can simply dominate the emission at these frequencies in some objects ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , and few mid - ir data exist in the literature . until recently ,
the only mid - ir detections of bhxbs in outburst at wavelengths @xmath21 8@xmath22 m were of gro j0422 + 32 , in which the source was detected at a level of 50 mjy at 11@xmath22 m during a bright hard state @xcite .
standard jet models predict the jet break frequency to depend on the mass accretion rate , black hole mass , location of the particle acceleration and magnetic field strength @xcite .
if the latter three parameters are unchanged , a positive relation should exist between jet break frequency and mass accretion rate .
observational confirmation via any apparent correlation between break frequency and luminosity would provide strict constraints for jet models . here ,
we perform a comprehensive literature search for quasi - simultaneous multiwavelength ( radio to optical ) data of bhxbs in the hard state and quiescence ( we define quiescence as @xmath23 erg s@xmath24 ; * ? ? ?
* ) , in order to identify the jet break and test for a relation with luminosity .
specifically , we only gather data where there is evidence for synchrotron emission at nir / optical frequencies .
the data collection , sed construction and spectral fitting are described in section 2 .
each sed from every source is discussed , and we detail the method we use for isolating the jet emission in each case .
the jet break is constrained in several sources , and the results are analysed in section 3 .
the distribution of jet breaks and the global relation between jet break frequency and luminosity are analysed , and compared to relations expected theoretically . a large scatter in jet break frequency
is found , and we discuss the possible origins of this scatter . a correlation between jet break luminosity and x - ray luminosity is also presented in section 3 , and we assess the likely contribution of the jet to the x - ray luminosity .
a summary of the results are provided in section 4 .
[ cols= " < , < , < , < , < , < , < , < , < " , ] the columns are : source name , dates of observations , optically thick synchrotron spectral index , optically thin synchrotron spectral index , log(jet break frequency ) , monochromatic luminosity at the jet break , bolometric luminosity as a fraction of the eddington luminosity , and references .
@xmath25these refer to wise observations 12 , 13 , 4 and 8 respectively , as given in @xcite .
@xmath26the distance to maxi j1836194 is unconstrained at this time , so luminosities can not yet be calculated .
@xmath27the optically thick spectrum of v404 cyg was best fit by a broken power law ; @xmath28 at radio frequencies and @xmath29 at infrared .
references : ( 1 ) this paper ; ( 2 ) @xcite ; ( 3 ) @xcite ; ( 4 ) @xcite ; ( 5 ) @xcite ; ( 6 ) @xcite ; ( 7 ) @xcite ; ( 8) @xcite ; ( 9 ) @xcite .
we have measured directly , constrained , or inferred upper / lower limits on the jet break frequency in nine bhxbs . together with constraints from three additional bhxbs reported in the literature and one neutron star source , we can for the first time study the distribution of jet break frequencies and test for a relation between jet break frequency and luminosity . from our spectral fits
we infer the range of possible peak jet fluxes ( i.e. the flux at the jet break ) and we take published quasi - simultaneous x - ray luminosities ( for data references see table 1 ) . in table 2
we present our results ; all jet break frequencies , luminosities , optically thick and thin power law indices , and x - ray luminosities are tabulated . in the following subsections
we compare the distribution of jet breaks and the global relation between jet break frequency and luminosity in the context of the relations expected theoretically .
we also test for a global correlation between the luminosity at the jet break and the x - ray luminosity , and constrain the likely contribution the synchrotron jet has to the observed x - ray luminosity . in fig .
3 the jet break frequency @xmath13 is plotted against peak jet monochromatic luminosity ( left panel ) and bolometric luminosity ( in eddington units ; right panel ) for each constraint discussed in section 2 .
we define the monochromatic luminosity as the luminosity in erg s@xmath24 divided by the frequency ( i.e. flux density scaled for distance ) ; @xmath30 .
the fairly large errors in @xmath31 and @xmath32 for some sources are a result of the uncertainties in the distance ( and bh mass which is required to calculate @xmath18 ) but are necessary to include @xmath31 to compare between sources .
4u 0614 + 09 is also included in fig . 3 in order to compare bhxbs to a neutron star source .
the range of jet break frequencies spans more than two orders of magnitude , even at similar jet ( and bolometric ) luminosities .
the jet break with the lowest frequency is that of xte j1118 + 480 during its 2005 outburst ; @xmath3 hz at @xmath33 .
v4641 sgr on the other hand , has a jet break at a frequency of @xmath4 hz , at @xmath34 .
the jet break can also shift in frequency by more than @xmath35 order of magnitude for a single bhxb .
it was established that this occurs on timescales of hours in gx 3394 @xcite . here
we report a shift in the jet break in xte j1118 + 480 from @xmath36 hz at one epoch in its 2000 outburst , to @xmath3 hz during its 2005 outburst at a very similar x - ray luminosity . the distribution of jet breaks peaks in the nir , with @xmath37 out of @xmath38 data points in the range @xmath39 14.5 ( these values are approximate because of the uncertainties from the sources with upper or lower limits on @xmath13 ) .
it is also worth noting that more than half of the 13 epochs for gx 3394 were red ( optically thin ) with wise in 2010 @xcite , suggesting a break in the mid - ir , at frequencies probably around or below @xmath40 hz .
the one neutron star xb has a jet break in amongst the distribution , but at slightly lower frequencies than the average bhxb . due to small number statistics we can not say if neutron star xb jet breaks are at lower frequencies than those of bhxbs .
if jet breaks are found in more neutron star xbs then a statistical comparison can be conducted .
standard theory of compact jets , applicable to both xbs and agn ( e.g. * ? ? ?
* ) , predicts a positive relation between the jet break frequency and luminosity , and a negative relation between jet break frequency and black hole mass @xcite if all other parameters are unchanged .
analytically , under some assumptions ( see below ) the jet break frequency is expected to scale with jet power ( i.e. , total power contained in the jet , including radiative and kinetic power ) as a power law relation ; @xmath41 and the jet luminosity in the flat , self - absorbed part of the spectrum seen by the observer scales with the jet power as @xmath42 @xcite , resulting in the following relation : @xmath43 an illustration of how the jet break changes with luminosity and with bh mass is shown in fig . 2 of @xcite .
these relations result from the derived dependency of the jet flux and jet break frequency on the mass accretion rate @xmath44 .
a constant fraction of the accreted mass is assumed to be channelled into the jets ( @xmath45 ; see also * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) : @xmath46 and @xmath47 in these models the radiative efficiency of the jet is assumed not to vary with luminosity , and the hard state is considered to be radiatively inefficient , with @xmath48 ( this is appropriate for the direct jet synchrotron model and original advection dominated accretion flow model ; see section 3.23 for alternatives ) .
thus , @xmath49 it is important to note at this stage that @xmath13 is also dependent on the magnetic field strength and the radius of the faz , both of which may differ between sources .
this is discussed further in section 3.2.3 .
a detailed analytical model is presented by @xcite .
the jet break frequency depends on both the mass accretion rate and the lepton energy distribution in the flow , @xmath50 ( see their equation 14 and their following discussion ) : @xmath51 similarly , at a constant @xmath44 , @xmath13 depends on both bh mass and @xmath50 : @xmath52 from equation 13 of @xcite , if the jet power is a constant fraction of the accretion power then : @xmath53 where @xmath54 is the spectral index of the self - absorbed ( optically thick ) synchrotron spectrum .
note that here we retain the @xmath10 nomenclature , which is not to be confused with that in @xcite in which they adopt @xmath55 . from equations ( 5 ) , ( 6 ) and ( 7 )
the following relations can be derived ( see also * ? ? ?
at a constant black hole mass but changing mass accretion rate : @xmath56 and at a constant mass accretion rate but a changing black hole mass : @xmath57 the lepton energy distribution can be measured directly from observations , as the spectral index of the optically thin synchrotron emission is directly dependent on @xmath50 ; @xmath58 . in most of the works mentioned above ,
@xmath59 is assumed which , when substituted into equation ( 5 ) recovers the relation of equation ( 3 ) .
observationally , @xmath60 has been measured accurately in some x - ray binary jets ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* see also sections 1 and 2 ) and generally resides in the range @xmath61 @xmath62 , which implies @xmath63 .
likewise , for a purely flat spectrum , @xmath64 ( which is normally assumed ) but observationally , radio seds of bhxb compact jets have spectral indices @xmath65 + 0.5 ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* see also section 2 ) .
we arrive at the following ranges of power law indices for the scaling relations , where the ranges encompass all possible values adopting these observational ranges of @xmath60 and @xmath54 . at a constant black hole mass but changing mass accretion rate : @xmath66 and at a constant mass accretion rate but a changing black hole mass : @xmath67 the famous radio x - ray correlation for hard state bhxbs ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) can be recovered from equations ( 4 ) and ( 10 ) : @xmath68 this is consistent with the empirical global correlations of bhxbs , but there are also radio - faint systems that appear not to follow the same relation ( e.g. * ? ? ?
* ; * ? ? ?
in an independent model , @xcite arrived at a very similar correlation as equation ( 10 ) . here
, simulations of internal shocks in jets composed of discrete ejections , which take into account adiabatic energy losses in the jet , were able to reproduce the flat , self - absorbed radio to ir jet spectrum .
the time - averaged sed produced by the simulations had a jet break frequency which scales with luminosity as @xmath69 .
we can compare equations ( 10 ) and ( 4 ) directly with the observations of jet breaks as shown in the left and right panels of fig .
3 , respectively . one would only expect these relations to explain the data of all sources if other parameters that affect the jet break are the same for each source ( see below ) .
this is unlikely , as it is known that some variable parameters of the inflow ( e.g. the disc temperature , disc inner radius ) differ between sources at the same luminosity , and some fundamental parameters ( e.g. the black hole mass ) also differ between sources .
nevertheless , the large range of jet breaks at similar luminosities prevents there being one single relation between jet break frequency and luminosity .
in addition , since two sources at @xmath70 possess jet break frequencies similar to the broad range seen at @xmath71 ( right panel of fig .
3 ) , the predicted positive power law relation between luminosity and jet break frequency appears to be weak in the global sample ( although we do not have more than a few data points for each source ) .
we fit a power law as opposed to @xmath72 ) . ] to the bhxb data in the left panel of fig .
3 ( neglecting the data which represent only upper / lower limits on @xmath13 ) and arrive at a best fit empirical relation : @xmath73 this is a poor fit due to the two orders of magnitude of scatter in @xmath13 at a similar luminosity .
however this best fit to the sample implies the jet break frequency could be independent of luminosity altogether ( i.e. , a power law index of zero ) , and the power law index is @xmath74 away from the expected relation of @xmath75 . the best fit is shown as a black line in the left panel of fig .
3 ( with errors in the slope shown as dotted lines ) . in both panels of the figure
the theoretical relations are shown in grey .
since we are fitting a compilation of sources with typically one or two data points per source , we can not rule out the theoretical predicted relation being true for each source , but we can rule out a global relation applicable to all sources
. it may be that some sources could obey the theoretical relation , but have different normalizations .
this could be due to different values of the magnetic field strength or the radius of the faz at the same luminosity between sources .
other parameters are likely to be changing on short timescales , and the global expected relation appears to be lost in the short term changes .
this may be why the jet break of xte j1118 + 480 was seen to differ by a factor of 10 in frequency on two dates with very similar luminosities ( fig .
3 ) . in addition
, @xcite showed that @xmath13 shifted by a factor of 10 in frequency on hour timescales in gx 3394 while the x - ray luminosity remained largely unchanged on the same timescale .
we note that the best power law fit to the global compilation of sources is sensitive to the two data points at low luminosity .
while the jet break of a062000 in quiescence has been observed directly , the jet break of gro j0422 + 32 is inferred via interpolation , and its error bars are large because we can only infer limits on the spectral index and normalization of the optically thick spectrum .
the large errors in @xmath13 for gro j0422 + 32 are not 1@xmath76 however ( they do not represent one standard deviation assuming a gaussian distribution of possible values of @xmath13 ) , and the break must reside in the frequency range indicated , for all possible values of @xmath54 and radio luminosity .
more jet breaks identified at low luminosities would be beneficial to better constrain the best fit relation .
jet models predict the jet break frequency to depend not only on the mass accretion rate , but also on the bh mass , magnetic field strength in the flow ( @xmath77 ) and the scale height of the launching region ( the faz ) , @xmath78 . in @xcite
the hour - timescale variability of the jet break frequency was interpreted as dramatic changes in one or both of these two latter parameters .
the exact relations are model - dependent , but are derived from standard jet theory .
the relations derived are @xmath79 and @xmath80 ( see e.g. equations ( 1 ) and ( 2 ) of * ? ? ?
* ) . the magnetic field and the acceleration zone
are both also expected to vary with mass accretion rate . here
, we explore a number of possibilities that could change the theoretical global relation . _ bhxbs with different bh masses _ : the bhxbs in our sample will have bhs of differing masses . due to equation
( 11 ) , we expect some scatter of the jet break frequency between different sources at the same luminosity if they have different bh masses . here
we assess whether this could explain some of the observed scatter in fig .
all bhxbs in our sample have bh masses between 3 and 30 @xmath81 ( a conservative range which encompasses all error bars in all bhs in table 1 ) .
we can calculate the maximum scatter due to these different bh masses thus .
taking all likely electron energy distribution values ( @xmath59 to @xmath82 ) , equation ( 6 ) gives the maximum difference in @xmath13 between one bhxb with @xmath83 and one with @xmath84 , of a factor of 5.5 .
so the different bh masses in our sample can theoretically shift the jet break frequency by no more than a factor of 5.5 ( 0.7 dex ) between two bhs at the same luminosity .
this is significant , and some of the scatter may be due to the different bh masses , but this can not explain the much broader range of jet break frequencies observed at the same luminosity .
_ individual ejections and internal shocks _ : in the hard state , there is strong variability in the accretion flow from hours down to less than seconds ( typically @xmath85 per cent rms variability in the x - ray luminosity ; * ? ? ?
* ) , which is well correlated with the jet ir variability @xcite .
it has been shown that the jet break shifts on hour timescales in gx 3394 @xcite , which is not as rapid as the accretion rate changes that cause the fast variability . since
the size of the emitting region at the jet break is likely to be light seconds across , these hour - long variations can not be due to individual plasma ejections .
discrete jet ` shells ' with different velocities in the flow produce additional synchrotron emission when they collide ( aka ` internal shocks ' ) and this could be on timescales of hours , but are likely to produce emission at lower frequencies than the jet break frequency , as the plasma cools ( see * ? ? ? * for simulations of individual ejections and the broadband seds they produce ) .
sub - second variability has been detected from the optically thin emission ( e.g. * ? ? ?
* ) , which can only come from regions close to the bh , not from collisions downstream .
individual ejections and internal shocks are therefore unlikely to be responsible for the hour - long variability of the jet break seen in gx 3394 , nor the longer timescale global scatter in the jet break frequency . _
@xmath13 dependent on @xmath77 _ :
most jet models explored predict the magnetic field strength to decrease with distance @xmath86 along the length of the jet , due to the lateral expansion of the jet along its axis , and magnetic flux conservation . in the seminal model of compact , self - absorbed jets explored by @xcite , a conical jet with constant flow velocity
has radius @xmath87 linearly proportional to @xmath86 , and thus @xmath88 .
the self - absorbed synchrotron spectrum from a given segment of this ideal jet will peak at a frequency that is inversely proportional to @xmath86 ( e.g. * ? ? ? * ) .
the magnetic field strength will therefore scale linearly with the frequency of the jet break , which represents the peak emission from the smallest radiating scale in the jets .
furthermore , if the internal pressures depend linearly on the accretion rate , then this break frequency would be expected to scale as @xmath89 for sources with the same mass , again for the idealized case ( see , e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
using such ideal assumptions , @xcite derived the relation @xmath90 ( their equation ( 1 ) ; see also * ? ? ?
this analytical solution was found by solving for the flux of the synchrotron emission at the jet break as a function of the synchrotron absorption coefficient , the optically thin emissivity and the size of the emitting region @xcite , assumed to be a homogeneous , cylindrical jet seen sideways . using such relations , the magnetic field strength can be estimated to within a rough factor based on direct measurement of the jet break . in reality , the exact dependence of @xmath87 will depend on the balance of the internal and external pressures , and interpretation will thus be model dependent .
for instance , a strongly externally confined jet would be expected to have smaller jet opening angles resulting in a less rapid decrease of @xmath77 with @xmath86 ( e.g. * ? ? ? * ) .
if internal or external pressures vary between sources or in time for a particular source , this could also vary the magnetic field strength and configuration .
it is thus not too surprising that the idealized relation @xmath90 is hidden in the scatter between sources as shown in fig .
3 , which implies that these sources are not strictly self - similar . in order to unearth meaningful trends
, these data sets must be fit by the same model , to determine whether a consistent interpretation linking the breaks , magnetic fields and source characteristics can be found .
we are planning to carry this out in future work .
essentially , we find that different magnetic field strengths in different sources could reproduce the observed scatter in jet break frequencies seen in fig .
it is worth noting that in a related work , @xcite and @xcite consider a jet which is accelerated once at its base ; here it was found that above a critical magnetic field strength , @xmath91 g ( this varies depending on other parameters ) , electrons rapidly cool , producing suppressed radio emission but enhanced optical / ir emission . from constraints on the jet breaks in gx 3394 and xte j1550564 , magnetic field strengths of @xmath35@xmath92 g were reported @xcite , slightly below , but close to , this critical value .
the scatter in fig .
3 implies that there is likely to be a large range in magnetic field strengths between sources .
for some bhxbs with the higher jet break frequencies the critical value @xcite may be reached , implying that a flux enhancement could exist in the optical / ir regime .
while no enhancement is seen in our sample at these wavelengths ( since we measure ir spectral indices which are negative ; see table 2 ) , this scenario can not be ruled out if the enhancement is at higher or lower frequencies than those sampled in each sed ( see below regarding more complex seds ) . _
@xmath13 dependent on @xmath78 _ :
@xcite point out that the jet break frequency scales as @xmath93 ( a revised version of equation ( 3 ) ) and that the radius of the faz is dependent on the inner radius of the accretion disc ( see also * ? ?
* ; * ? ? ?
the inner radius of the disc ( @xmath94 ) can be constrained from x - ray observations , and it is known that at lower luminosities in the hard state the disc becomes more truncated . using data from a wealth of bhxbs , @xcite
derive an empirical relation between the inner disc radius and the x - ray luminosity in the hard state : @xmath95 .
if the radius of the faz is directly anchored to the inner edge of the disc then one expects @xmath96 ( see also * ? ? ?
* ) , and hence from equation ( 4 ) , @xmath97 .
taking this dependency of @xmath78 on @xmath44 into account results in a steeper relation between jet break frequency and luminosity , and so is unlikely to be the case .
the opening angle of the jet plasma entering the faz also affects the radius of the faz .
this opening angle , and how it varies with mass accretion rate , is unknown , but has been seen to be quite wide ( @xmath98 ) in one agn @xcite . _ an alternative relation proposed for gx 3394 _ : @xcite perform broken power law fits to quasi - simultaneous radio and x - ray data of gx 3394 in its hard state
. the jet break frequency is constrained here by the interpolation of the two power laws , and an empirical correlation is found ; @xmath99 ( see their fig .
3 ) . this relation is steeper than the theoretical relation in equation ( 4 ) .
@xcite adopt the assumption that the optically thin jet synchrotron emission is the origin of the x - ray power law .
this may be the case at some x - ray luminosities in the hard state , but may not be the case for all luminosities .
such a steep relationship extrapolated to low luminosities would imply a jet break in the radio domain or even lower frequency in quiescence ; much lower frequencies than observed .
correlations between ir and x - ray luminosities favour a different relationship between jet break frequency and luminosity in gx 3394 @xcite .
multiple detections of the jet break in gx 3394 over an outburst cycle would be useful to test these , and the theoretical relations .
_ relaxing the assumption of a radiatively inefficient hard state _
: some bhxbs appear too faint in radio compared to the normal radio x - ray correlation for hard state bhxbs ( e.g. * ? ? ?
* ; * ? ? ?
data from several outbursts of the bhxb h1743322 indicated that at moderate luminosities this is a radio - faint bhxb , but that the radio x - ray correlation steepened at high luminosities in the hard state to a correlation expected for a radiatively efficient accretion flow @xcite .
we may therefore consider radiatively efficient accretion flows in the hard state here . in this scenario , @xmath100 , which leads to @xmath101 .
this is a much steeper correlation than the observations suggest .
_ relaxing the assumption of a constant jet radiative efficiency _
: if the fraction of jet kinetic energy which is radiated away ( @xmath16 ) varies with accretion rate , this will change the correlations and scaling relations . specifically
, the @xmath42 relation breaks down in this scenario .
let us consider a jet radiative efficiency which scales as a power law relation with mass accretion rate ; @xmath102 .
we therefore have @xmath48 ( unchanged ) and @xmath103 .
the revised relations between jet break frequency and luminosity would then be @xmath104 ( as before ) and @xmath105 .
we can immediately see that since the @xmath13@xmath106 relationship is unchanged , the distribution of jet breaks in the right panel of fig .
3 is not affected by a changing jet radiative efficiency . if we consider only the @xmath13@xmath107 relation , using the best fit to the data ( equation ( 13 ) ) we find @xmath108 , hence @xmath109 .
this implies an extremely strong dependency of radiative efficiency on mass accretion rate globally , which is likely to be unphysical .
it also predicts a very steep correlation between jet and x - ray luminosities , of the form @xmath110 which , as we will see below , is inconsistent with the observations . different jet radiative efficiencies in different sources could introduce scatter in the @xmath13@xmath107 plot , but not the @xmath13@xmath106 plot . since the observed scatter is similar in both plots ( fig .
3 ) , it is unlikely that this is a main source of the scatter .
_ more complex seds _ : in order to identify the jet break , for many sources it was necessary to interpolate radio and ir / optical power laws .
this assumes the jet spectrum can be described approximately by a broken power law ( see section 2 ) .
this is the classical model for compact jets @xcite but more complex models have been developed .
some models predict an extra excess of emission in the sed at a frequency near the jet break @xcite , whereas some argue the flat optically thick spectrum is difficult ( but not impossible ) to reproduce theoretically , if adiabatic energy losses are taken into account @xcite .
we note that the jet break directly observed in v404 cyg actually requires two power laws ( or a curved spectrum ) between radio frequencies and the jet break . without the ir data
, we would have interpolated the radio and optical power laws assuming one break , and arrived at a jet break @xmath111 dex lower in frequency than the one observed .
this demonstrates some scatter in fig .
3 may be due to more complex seds .
we conclude that variations of the magnetic field strength and/or the radius of the faz on timescales generally shorter than the outburst timescale likely introduce some of the scatter in the observed relation between jet break frequency and luminosity .
the inner disc radius does not change rapidly so if the faz is anchored to the inner disc radius , then it is more likely that the magnetic field is the quantity that is varying rapidly
. it may also be that the flat optically thick regime may not be well approximated by a single power law in some cases . in this scenario ,
interpolation of the optically thick and optically thin power laws may introduce some inaccuracies in our estimations of the jet break , which could introduce more scatter in fig .
are included ; this is the reason for fewer data points appearing in the lower panel ) .
the bhxbs that follow the ` normal ' radio x - ray correlation ( ` radio - bright ' bhxbs ) are shown as separate symbols to bhxbs that appear to be radio - faint compared to this correlation , or appear to lie ( or switch ) between the two luminosity tracks ( e.g. * ? ? ?
* ) .,title="fig:",width=226 ] + are included ; this is the reason for fewer data points appearing in the lower panel ) .
the bhxbs that follow the ` normal ' radio x - ray correlation ( ` radio - bright ' bhxbs ) are shown as separate symbols to bhxbs that appear to be radio - faint compared to this correlation , or appear to lie ( or switch ) between the two luminosity tracks ( e.g. * ? ? ?
* ) .,title="fig:",width=226 ] the radio x - ray correlation becomes steeper for a radiatively efficient flow ; @xmath112 ( where these errors are propagated from those in equation ( 10 ) ) .
if a radiatively efficient flow could exist in the hard state , this relation can be tested directly against the observations ( see e.g. * ? ? ?
* ; * ? ? ?
* for different radio x - ray slopes found in some bhxbs ) . in fig .
4 the observed jet luminosity at the jet break versus the quasi - simultaneous x - ray luminosity is plotted . in the upper panel ,
monochromatic jet luminosities are used against x - ray luminosities in eddington units , for all the data considered in this paper . the best fit power law correlation to the bhxb data is @xmath8 , clearly inconsistent with a radiatively efficient flow .
the flow is consistent with being radiatively inefficient , as the correlation slope lies within the error bars of the expected slope in equation ( 12 ) .
we can also conclude that since @xmath113 for the bhxbs that follow the classical radio x - ray correlation ( e.g. * ? ? ?
* ) , the radio luminosity of the jet must scale approximately linearly with the jet luminosity at the jet break ; @xmath114 .
therefore , the spectral index of the self - absorbed synchrotron spectrum , spanning from radio to the jet break , does not vary as a function of luminosity in the hard state .
it is interesting to note that three sources appear too faint in @xmath107 in the upper panel of fig . 4 compared to the correlation .
the white circle is the neutron star xb 4u 0614 + 09 .
its jet break luminosity is @xmath115 orders of magnitude less luminous than most bhxbs at the same x - ray luminosity .
radio jets in neutron stars are less luminous than radio jets in bhxbs @xcite .
this confirms what was found by @xcite ; that the same appears to be the case for ir jets ; ir jets in neutron stars are fainter than ir jets in bhxbs ( see also * ? ? ?
the two bhxbs that appear to have fainter jets than the correlation ( both are around @xmath1160.01 @xmath18 ) are xte j1752223 and cyg x1 .
it is relevant to mention that both sources are ( slightly ) radio - faint bhxbs @xcite , so it is not unexpected that their jet breaks are also fainter than most bhxbs . in the lower panel of fig .
4 , the flux density of the jet at the jet break , scaled to distance ( to compare between sources , and approximated by @xmath117 ; which requires a direct detection of the jet break , not an upper or lower limit on its frequency ) is plotted against the bolometric luminosity , both in erg s@xmath24 .
this is a different relation , as @xmath107 is not measured at a fixed frequency , but this time is the total luminosity at the jet break , which may be a good indication of the total jet power ( at least the total relative jet power between sources ) . a correlation with a very similar slope as measured in the upper panel of fig . 4 ,
@xmath118 is measured from the population of bhxbs that produce the well known radio x - ray correlation in bhxbs , with very little scatter ( black filled circles in the lower panel of fig .
the dotted line in this figure indicates where the jet luminosity equals the x - ray luminosity .
it implies that the jet luminosity exceeds the x - ray luminosity below @xmath119 erg s@xmath24 , which is consistent with jet dominated sources existing at low luminosities ( see e.g. * ? ? ?
* ; * ? ? ?
the bhxbs that appear radio - faint compared to the originally defined radio x - ray correlation , or sources that appear to move between two luminosity tracks ( crosses in fig .
4 ) , here have fainter jet break luminosities than this correlation ( we use the radio x - ray diagrams of * ? ? ? * ; * ? ? ?
* to define which sources are radio - bright and which are radio - faint ) .
this shows that the two luminosity tracks in the radio x - ray diagram are actually reproduced in the jet break luminosity x - ray diagram .
the bhxbs which are radio - faint tend to also be ir - faint . ) versus the observed x - ray luminosity.,width=230 ] since we have measured the jet break frequency and its luminosity , we can directly infer the x - ray luminosity of the jet , under two assumptions .
the first is that the optically thin power law extends to x - ray energies ( the high energy cutoff , or the cooling break , resides at energies @xmath120 kev ) .
this cooling break is hard to detect , but observations have favoured a cooling break in the energy range @xmath12150 kev ( see * ? ? ?
* for a recent discussion ) .
the second assumption is that of the spectral index of the optically thin power law . for each measurement of the jet break ,
we compute the x - ray ( 210 kev ) luminosity of the extrapolated optically thin jet spectrum assuming its spectral index is @xmath122 ( a fairly typical spectral index ) . adopting different values of @xmath60 would shift each predicted x - ray luminosity systematically by the same amount in log(@xmath106 ) .
5 shows the predicted x - ray luminosity of the optically thin jet power law plotted against the observed x - ray luminosity . for all bhxbs except those in quiescence ,
the predicted jet luminosity is approximately equal to , or less than the observed x - ray luminosity . above the quiescent luminosity ,
all bhxbs have x - ray luminosities that are no brighter than @xmath123 times the predicted jet luminosity .
the only exceptions are cyg x1 , which is @xmath124 times more luminous , and the neutron star xb 4u 0614 + 09 .
it has been shown that in both of these objects , the x - ray emission is very unlikely to originate from synchrotron emission from the jet @xcite .
5 confirms this for these two sources . for most bhxbs , the jet luminosity underpredicts the observed x - ray luminosity by a factor of up to ten , implying that the jet does not dominate the x - ray emission .
this is consistent with the standard picture of the comptonized corona producing the x - ray power law in the hard state ( see e.g. * ? ? ?
* for a review ) .
we have here showed however that the jet emission could typically contribute more than 10 per cent of the x - ray flux , if the cooling break does not reside at lower energies than the x - ray band . at @xmath125 , three sources (
xte j1118 + 480 , 4u 154347 and xte j1550564 ) have observed x - ray luminosities within a factor of two of the predicted jet luminosities .
empirical evidence has favoured a jet origin to the x - ray emission at this luminosity in xte j1550564 @xcite .
xte j1118 + 480 was the first bhxb that was proposed to have a jet dominating the x - ray luminosity , and this was at a similar x - ray luminosity @xcite .
5 suggests that the jet of 4u 154347 may also have dominated its x - ray luminosity at @xmath126 .
the data are broadly consistent with the suggestion that the jet does not dominate the x - ray luminosity above @xmath127 @xmath18 , but may do below that luminosity @xcite .
the two quiescent bhxbs have observed x - ray luminosities that are several orders of magnitude ( three in the case of a062000 ) fainter than the predicted jet luminosity . for this to occur ,
the cooling break in the jet spectrum must lie at energies much lower than the x - ray regime ( @xmath128 ) .
it is possible that the optically thin spectral index becomes steeper ( e.g. @xmath129 ) at lower luminosities .
a correlation was found between @xmath60 and luminosity for xte j1550564 , in which it was as steep as @xmath130 at the lowest luminosities as the jet first appeared after the soft - to - hard state transition @xcite .
it is also interesting that the measured x - ray spectral index of quiescent bhxbs is generally steeper ( @xmath131 ) than is typical for the hard state ( e.g. * ? ? ?
indeed , the observed spectral indices in quiescence are consistent with that expected from the jet at energies higher than the cooling break ( e.g. * ? ? ?
perhaps this implies that the origin of the x - ray emission in quiescent bhxbs is the jet ( as predicted by , e.g. * ? ? ?
* ) , but this is speculative . we can conclude that it is likely that the high energy break in the jet spectrum shifts from x - ray energies at @xmath7 to uv energies at @xmath132 .
we have collected multiwavelength seds of bhxbs in their hard x - ray states .
synchrotron emission from the jets launched in these systems is identified spectrally , and constraints on the frequency and luminosity of the characteristic break in the jet spectrum have been achieved for a total of eight bhxbs .
we detect the jet break directly in the sed of v404 cyg during its 1989 outburst , at @xmath1 hz ( @xmath2 ) .
jet break frequencies span more than two orders of magnitude , and there appears to be no strong global relation between jet break frequency and luminosity from @xmath1331 @xmath18 . in two sources , gx 3394 and xte j1118 + 480 ,
the jet break frequency varies by more than one order of magnitude while the change in luminosity is negligible ( seen on hour - timescales by * ? ? ? * in the case of gx 3394 ) .
the wide scatter in the relation between jet break frequency and luminosity may be due to the magnetic field strength and/or the radius of the launching region in the jet varying dramatically , causing large shifts in the jet break frequency , and suppressing the visibility of any global relation with luminosity .
different bh masses in different sources are unlikely to cause much of the scatter , but differences in the magnetic field strength , radius of the faz , or the jet radiative efficiency could contribute to the scatter in jet break frequencies .
the high energy ( cooling ) break in the jet spectrum appears to shift in frequency , from uv energies at @xmath132 ( quiescence ) to x - ray energies at @xmath7 .
we find a power law correlation between the jet peak flux ( i.e. , at the jet break frequency ) and the x - ray luminosity for hard state bhxbs ; @xmath8 ; very similar to the well documented radio x - ray correlation .
this implies a radiatively inefficient accretion flow in the hard state ; @xmath134 .
the bhxbs that are radio - faint in the radio x - ray correlation ( or perform transitions between being radio - faint and radio - bright ) are also ir - faint , and the radio to ir spectral index of the jet is independent of luminosity .
dmr acknowledges support partly from a netherlands organisation for scientific research ( nwo ) veni fellowship and partly from a marie curie intra european fellowship within the 7th european community framework programme under contract no .
ief 274805 .
dmr and ts acknowledge support from the spanish ministry of science and innovation ( micinn ) under the grant aya2010 - 18080 .
sm gratefully acknowledges support from a netherlands organisation for scientific research ( nwo ) vidi fellowship , and from the european community s seventh framework program ( fp7/2007 - 2013 ) under grant agreement number itn 215212 ` black hole universe ' .
99 belloni t. m. , 2010 , in ` the jet paradigm - from microquasars to quasars ' , ed .
t. belloni , lect .
notes phys .
794 blandford r. d. , konigl a. , 1979 , apj , 232 , 34 blandford r. d. , payne d. g. , 1982 , mnras , 199 , 883 blandford r. d. , znajek r. l. , 1977 , mnras , 179 , 433 brocksopp c. , jonker p. g. , fender r. p. , groot p. j. , van der klis m. , tingay s. j. , 2001 , mnras , 323 , 517 brocksopp , c. , bandyopadhyay , r. m. , fender , r. p. 2004
, newa , 9 , 249 brocksopp c. , jonker p. g. , maitra d. , krimm h. a. , pooley g. g. , ramsay g. , zurita c. , 2010 , mnras , 404 , 908 buxton , m. m. , bailyn , c. d. 2004 , apj , 615 , 880 buxton m. m. , bailyn c. d. , capelo h. l. , chatterjee r. , diner t. , kalemci e. , tomsick j. a. , 2012 , aj , 143 , 130 cabanac c. , fender r. p. , dunn r. j. h. , krding e. g. , 2009 , mnras , 396 , 1415 cadolle bel m. , et al . , 2011 , a&a , 534 , a119 cantrell a. g. , et al . , 2010 ,
apj , 710 , 1127 cardelli j. a. , clayton g. c. , mathis j. s. , 1989 , apj , 345 , 245 casares j. , charles p. a. , jones d. h. p. , rutten r. g. m. , callanan p. j. , 1991 , mnras , 250 , 712 casares j. , zurita c. , shahbaz t. , charles p. a. , fender r. p. 2004 , apj , 613 , l133 casares j. , et al . , 2009 ,
apjs , 181 , 238 casella p. , peer , a. , 2009 , apj , 703 , l63 casella p. , et al . , 2010 , mnras , 404 , l21 castro - tirado a. j. , ilovaisky s. , pedersen h. , gonzalez j .- f . , pizarro m. , miranda j. , boehnhardt h. , 1997 , iauc , 6775 chaty s. , mirabel i. f. , goldoni p. , mereghetti s. , duc p .- a . , mart@xmath135 , j. , mignani r. p. , 2002 ,
mnras , 331 , 1065 chaty s. , haswell c. a. , malzac j. , hynes r. i. , shrader c. r. , cui w. , 2003a , mnras , 346 , 689 chaty s. , charles p. a. , mart j. , mirabel i. f. , rodrguez l. f. , shahbaz t. , 2003b , mnras , 343 , 169 chaty s. , dubus g. , raichoor a. , 2011 , a&a , 529 , 3 corbel s. , fender r. p. , 2002
, apj , 573 , l35 corbel s. , fender r. p. , tzioumis a.k . , nowak m. , mcintyre v. , durouchoux p. , sood r. , 2000 , a&a , 359 , 251 corbel s. , et al . , 2001 , apj , 554 , 43 corbel s. , tomsick j. a. , kaaret p. , 2006
, apj , 636 , 971 coriat m. , corbel s. , buxton m. m. , bailyn c. d. , tomsick j. a. , krding e. , kalemci e. , 2009 , mnras , 400 , 123 coriat m. , et al . , 2011 , mnras , 414 , 677 curran p. a. , maccarone t. j. , casella p. , evans p. a. , landsman w. , krimm h. a. , brocksopp c. , still m. , 2011 , mnras , 410 , 541 diner t. , kalemci e. , buxton m. m. , bailyn c. d. , tomsick j. a. , corbel s. , 2012 , apj , 753 , 55 eikenberry s. s. , matthews k. , morgan e. h. , remillard r. a. , nelson r. w. , 1998 , apj , 494 , l61 falcke h. , biermann p. l. , 1995 , a&a , 293 , 665 falcke h. , krding e. , markoff s. , 2004 , a&a , 414 , 895 fender r. p. , 2001 ,
mnras , 322 , 31 fender r. p. , 2006
, in _ compact stellar x - ray sources _ , cambridge university press , cambridge , u.k .
, pp 381 ( eds .
lewin w. h. g. , van der klis m. ) fender r. p. , pooley g. g. , brocksopp c. , newell s. j. , 1997 , mnras , 290 , l65 fender r. p. , pooley g. g. , durouchoux p. , tilanus r. p. j. , brocksopp c. , 2000 , mnras , 312 , 853 fender r. p. , hjellming r. m. , tilanus r. p. j. , pooley g. g. , deane j. r. , ogley r. n. , spencer r. e. , 2001 , mnras , 322 , l23 fender r. p. , gallo e. , jonker p. g. , 2003 , mnras , 343 , l99 fender r. p. , russell d. m. , knigge c. , soria r. , hynes r. i. , goad m. , 2009 , mnras , 393 , 1608 gallo , e. , fender , r. p. , pooley , g. g. 2003 , mnras , 344 , 60 gallo e. , fender r. p. , kaiser c. , russell d. , morganti r. , oosterloo t. , heinz s. , 2005 , nature , 436 , 819 gallo e. , et al .
, 2006 , mnras , 370 , 1351 gallo e. , migliari s. , markoff s. , tomsick j. a. , bailyn c. d. , berta s. , fender r. , miller - jones j. c. a. , 2007 , apj , 670 , 600 gallo e. , miller b. p. , fender r. , 2012 , mnras , 23 , 590 gandhi p. , et al .
, 2010 , mnras , 407 , 2166 gandhi p. , et al . , 2011 , apj , 740 , l13 gehrz r. d. , johnson j. , harrison t. , 1989 , iauc , 4816 gelino d. m. , harrison t. e. , 2003 , apj , 599 , 1254 gelino d. m. , balman . , kizilolu . , yilmaz a. , kalemci e. , tomsick j. a. , 2006 , apj , 642 , 438 gelino d. m. , gelino c. r. , harrison t. e. , 2010 , apj , 718 , 1 gilfanov m. , 2009 , in ` the jet paradigm - from microquasars to quasars ' , ed .
t. belloni , lect .
notes phys .
794 ( arxiv:0909.2567 ) goranskii v. p. , karitskaya e. a. , kurochkin n. e. , trunkovskii e. m. , 1996 , astl , 22 , 371 han x. , hjellming r. m. , 1992 , apj , 400 , 304 heinz s. , sunyaev r. a. , 2003 , mnras , 343 , l59 hjellming r. m. , johnston k. j. , 1988 , apj , 328 , 600 hjellming r. m. , et al . ,
2000 , apj , 2000 , 544 , 977 homan j. , buxton m. , markoff s. , bailyn c. d. , nespoli e. , belloni t. , 2005 , apj , 624 , 295 hynes r. i. , 2005 , apj , 623 , 1026 hynes r. i. , haswell c. a. , chaty s. , shrader c. r. , cui w. , 2002 , mnras , 331 , 169 hynes r. i. , et al . , 2003 , mnras , 345 , 292 hynes r. i. , steeghs d. , casares j. , charles p. a. , obrien k. , 2004 , apj , 609 , 317 hynes r. i. , et al . , 2006 , apj , 651 , 401 hynes r. i. , bradley c. k. , rupen m. , gallo e. , fender r. p. , casares j. , zurita c. , 2009 , mnras , 399 , 2239 jain r. k. , bailyn c. d. , orosz j. a. , mcclintock j. e. , remillard r. a. , 2001 , apj , 554 , l181 jamil o. , fender r. p. , kaiser c. r. , 2010 , mnras , 401 , 394 johnson j. , harrison t. , gehrz r. d. , 1989 , iauc , 4786 junor w. , biretta j. a. , livio m. , 1999 , nat , 401 , 891 kaaret p. , corbel s. , tomsick j. a. , fender r. , miller j. m. , orosz j. a. , tzioumis a. k. , wijnands r. , 2003 , apj , 582 , 945 kaiser c. r. , 2006 , mnras , 367 , 1083 kalemci e. , et al .
, 2005 , apj , 622 , 508 khargharia j. , froning c. s. , robinson e. l. , 2010 , apj , 716 , 1105 king n. l. , harrison t. e. , mcnamara b. j. , 1996 , aj , 111 , 1675 kleinmann s. g. , brecher k. , ingham w. h. , 1976 , apj , 207 , 532 krding e. , fender r. p. , migliari s. , 2006 , mnras , 369 , 1451 kuulkers e. , fender r. p. , spencer r. e. , davis r. j. , morison i. , 1999 , mnras , 306 , 919 kuulkers e. , et al .
, 2010 , a&a , 514 , a65 leibowitz e. m. , ney a. , drissen l. , grandchamps a. , moffat a. f. j. , 1991 , mnras , 250 , 385 livio m. , ogilvie g. i. , pringle j. e. , 1999 , apj , 512 , 100 maitra d. , markoff s. , brocksopp c. , noble m. , nowak m. , wilms j. , 2009a , mnras , 398 , 1638 maitra d. , markoff s. , falcke h. , 2009b , a&a , 508 , l13 maitra d. , cantrell a. , markoff s. , falcke h. , miller j. , bailyn c. , 2011 , proceedings of iau symposium 275 `` jets at all scales '' , 13 - 17
september 2010 , buenos aires , argentina ( arxiv:1010.4296 ) malzac j. , belmont r. , fabian a. c. , 2009 , mnras , 400 , 1512 markoff s. , falcke h. , fender r. , 2001 , a&a , 372 , l25 markoff , s. , nowak , m. , corbel , s. , fender , r. , falcke , h. 2003 , a&a , 397 , 645 markoff s. , nowak m. a. , wilms j. , 2005 , apj , 635 , 1203 mcclintock j. e. , remillard r. a. , 2006 , in compact stellar x - ray sources , eds .
lewin w. h. g. , van der klis m. , cambridge university press , p. 157
mckinney j. c. , tchekhovskoy a. , blandford r. d. , 2012 , mnras , 423 , 3083 meier d. l. , 2001 , apj , 548 , l9 migliari s. , fender r. p. , 2006 , mnras , 366 , 79 migliari s. , tomsick j. a. , maccarone t. j. , gallo e. , fender r. p. , nelemans g. , russell d. m. , 2006 , apj , 643 , l41 migliari s. , et al . , 2007 , apj , 670 , 610 migliari s. , et al . , 2010 , apj , 710 , 117 migliari s. , miller - jones j. c. a. , russell d. m. , 2011 , mnras , 415 , 2407 miller - jones j. c. a. , jonker p. g. , dhawan v. , brisken w. , rupen m. p. , nelemans g. , gallo e. , 2009 , apj , 706 , l230 miller - jones j. c. a. , jonker p. g. , maccarone t. j. , nelemans g. , calvelo d. e. , 2011 , apj , 739 , l18 mirabel i. f. , dhawan v. , chaty s. , rodriguez l. f. , marti j. , robinson c. r. , swank j. , geballe t. , 1998 , a&a , 330 , l9 muno m. p. , mauerhan j.,2006 , apj , 648 , l135 muoz - darias t. , motta s. , pawar d. , belloni t. m. , campana s. , bhattacharya d. , 2010 , mnras , 404 , l94 neilsen j. , lee j. c. , 2009 , nat , 458 , 481 nowak m. a. , wilms j. , heinz s. , pooley g. , pottschmidt k. , corbel s , 2005 , apj , 626 , 1006 orosz j. a. , jain r. k. , bailyn c. d. , mcclintock j. e. , remillard r. a. , 1998 , apj , 499 , 375 orosz j. a. , et al . , 2001 ,
apj , 555 , 489 orosz j. a. , polisensky e. j. , bailyn c. d. , tourtellotte s. w. , mcclintock j. e. , remillard r. a. , 2002 , baas , 34 , 1124 orosz j. a. , steiner j. f. , mcclintock j. e. , torres m. a. p. , remillard r. a. , bailyn c. d. , miller j. m. , 2011a , apj , 730 , 75 orosz j. a. , mcclintock j. e. , aufdenberg j. p. , remillard r. a. , reid m. j. , narayan r. , gou l. , 2011 , apj , 742 , 84 pavlenko e. p. , dmitrienko e. s. , shakhovskoi n. m. , shugarov s. y. , katysheva n. a. , volkov i. m. , 2001 , apsss , 276 , 63 peer a. , casella p. , 2009 ,
apj , 699 , 1919 peer a. , markoff s. , 2012 , apj , 753 , 177 polko p. , meier d. l. , markoff s. , 2010 , apj , 723 , 1343 ponti g. , fender r. p. , begelman m. c. , dunn r. j. h. , neilsen j. , coriat m. , 2012 , mnras , 422 , l11 ratti e. m. , et al .
2012 , mnras , 423 , 2656 rahoui f. , lee j. c. , heinz s. , hines d. c. , pottschmidt k. , wilms j. , grinberg v. , 2011 , apj , 736 , 63 rahoui f. , et al . , 2012 , mnras , 422 , 2202 reid m. j. , mcclintock j. e. , narayan r. , gou l. , remillard r. a. , orosz j. a. , 2011 , apj , 742 , 83 robertson b. s. c. , warren p. r. , bywater r. a. , 1976 , ibvs , 1173 , 1 rupen m. p. , mioduszewski a. j. , dhawan v. , 2003 , astronomer s telegram , 172 russell d. m. , fender r. p. , hynes r. i. , brocksopp c. , homan j. , jonker p. g. , buxton m. m. , 2006 , mnras , 371 , 1334 russell d. m. , fender r. p. , gallo e. , kaiser c. r. , 2007a , mnras , 376 , 1341 russell d. m. , maccarone t. j. , krding e. g. , homan j. , 2007b , mnras , 379 , 1401 russell d. m. , fender r. p. , jonker p. g. , 2007c , mnras , 379 , 1108 russell d. m. , fender r. p. , 2008 ,
mnras , 387 , 713 russell d. m. , maitra d. , dunn r. j. h. , markoff s. , 2010 , mnras , 405 , 1759 russell d. m. , miller - jones j. c. a. , maccarone t. j. , yang y. j. , fender r. p. , lewis f. , 2011a , apj , 739 , l19 russell d. m. , casella p. , fender r. , soleri p. , pretorius m. l. , lewis f. , van der klis m. , 2011b , in proceedings of ` high time resolution astrophysics iv the era of extremely large telescopes htra - iv ' ,
57 may 2010 , agios nikolaos , crete , greece , proceedings of science ( arxiv:1104.0837 ) russell d. m. , et al .
, 2011c , astronomer s telegram , 3689 russell d. m. , et al . , 2012 , mnras , 419 , 1740 rybicki g. b. , lightman a. p. , 1979
, radiative processes in astrophysics .
wiley , new york shahbaz t. , fender r. p. , watson c. a. , obrien k. , 2008 , apj , 672 , 510 shidatsu m. , et al . , 2011 , pasj , 63 , 785 shaposhnikov n. , markwardt c. , swank j. , krimm h. , 2010 , apj , 723 , 1817 shrader c. r. , wagner r. m. , hjellming r. m. , han x. h. , starrfield s. g. , 1994 , apj , 434 , 698 soleri p. , fender r , 2011 , mnras , 413 , 2269 soleri p. , et al . , 2010 , mnras , 406 , 1471 soria r. , bessell m. s. , wood p. , 1997 , iauc , 6781 taranova o. g. , shenavrin v. i. , 2001 , astl , 27 , 25 tomsick j. a. , corbel s. , kaaret p. , 2001
, apj , 563 , 229 tomsick j. a. , corbel s. , fender r. , miller j. m. , orosz j. a. , tzioumis t. , wijnands r. , kaaret p. , 2003 , apj , 582 , 933 tudose v. , fender r. p. , kaiser c. r. , tzioumis a. k. , van der klis m. , spencer r. , 2006 , mnras , 372 , 417 uemura m. , et al . , 2004a , pasj , 56 , 61 uemura m. , et al . , 2004b , pasj , 56 , 823 van paradijs j. , telesco c. m. , kouveliotou c. , fishman g. j. , 1994 , apj , 429 , l19 van straaten s. , ford e. c. , van der klis m. , m@xmath136ndez m. , kaaret p. 2000 , apj , 540 , 1049 veledina a. , poutanen j. , vurm i. , 2011 , apj , 737 , l17 weingartner j. c. , draine b. t. , 2001 , apj , 548 , 296 yuan f. , cui w. , 2005 , apj , 629 , 408 zurita c. , 2006 , apj , 644 , 432 | in x - ray binaries , compact jets are known to commonly radiate at radio to infrared frequencies , whereas at optical to @xmath0-ray energies , the contribution of the jet is debated .
the total luminosity , and hence power of the jet is critically dependent on the position of the break in its spectrum , between optically thick ( self - absorbed ) and optically thin synchrotron emission .
this break , or turnover , has been reported in just one black hole x - ray binary ( bhxb ) thus far , gx 3394 , and inferred via spectral fitting in two others , a062000 and cyg x1 . here , we collect a wealth of multiwavelength data from the outbursts of bhxbs during hard x - ray states , in order to search for jet breaks as yet unidentified in their spectral energy distributions . in particular , we report the direct detection of the jet break in the spectrum of v404 cyg during its 1989 outburst , at @xmath1 hz ( @xmath2 ) .
we increase the number of bhxbs with measured jet breaks from three to eight .
jet breaks are found at frequencies spanning more than two orders of magnitude , from @xmath3 hz for xte j1118 + 480 during its 2005 outburst , to @xmath4 hz for v4641 sgr in outburst . a positive correlation between jet break frequency and luminosity
is expected theoretically ; @xmath5 if other parameters are constant . with constraints on the jet break in a total of 12 bhxbs including two quiescent systems
, we find a large range of jet break frequencies at similar luminosities and no obvious global relation ( but such a relation can not be ruled out for individual sources ) .
we speculate that different magnetic field strengths and/or different radii of the acceleration zone in the inner regions of the jet are likely to be responsible for the observed scatter between sources .
there is evidence that the high energy cooling break in the jet spectrum shifts from uv energies at @xmath6 ( implying the jet may dominate the x - ray emission in quiescence ) to x - ray energies at @xmath7 . finally , we find that the jet break luminosity scales as @xmath8 ( very similar to the radio x - ray correlation ) , and radio - faint bhxbs have fainter jet breaks .
in quiescence the jet break luminosity exceeds the x - ray luminosity .
[ firstpage ] accretion , accretion discs , black hole physics , x - rays : binaries , ism : jets and outflows |
measurements of volume - averaged stellar mass , star formation rate and gas densities over cosmic time provide fundamental constraints on galaxy evolution by describing the integrated past history , present activity and future potential for galaxy growth . in the past decade
, substantial progress has been made in accurately measuring the first two the stellar mass function @xcite and star formation rate density @xcite .
on the other hand , quantifying gas content across redshift , and in particular identifying the gas supply and reservoir most closely linked to recent and future star formation has remained challenging .
recent progress has been made in the local universe through blind hi surveys such as hipass @xcite and alfalfa @xcite which are yielding hi mass functions of representative volumes @xcite and optical identifications of nearly all of hi detected sources ( e.g. , * ? ? ?
* ; * ? ? ?
this latter result , the fact that blind hi surveys do not reveal a population of hi - rich galaxies without optical counterparts , suggests that the study of the gas content of optically selected samples can provide a nearly complete census of hi in the local universe . in this paper , rather than perform a complete census of hi ,
we focus exclusively on a subsample of massive galaxies in the local universe that have been homogeneously observed by sdss , galex and now arecibo as part of the gass survey ( described in * ? ? ?
* hereafter paper i )
. briefly stated , the goals of the gass survey are to measure the hi properties of @xmath11 massive ( @xmath1 ) galaxies in the local universe ( z@xmath12 ) , in a mass range that includes the transition mass above which most galaxies have ceased forming stars . to maximize survey efficiency ,
galaxies are observed to a gas fraction limit of 2 - 5% , ensuring that the quantity of gas probed remains relevant for on - going and future star formation . in paper
i we found that the hi gas fraction decreased with stellar mass , stellar mass surface density and @xmath6 color .
observations and models suggest that star formation in massive galaxies may be quenched by internal ( agn or sf feedback ) or external means ( stripping or other environmental effects ) .
this has been used to explain both the decline of the global star formation rate ( sfr ) and the growth of stellar mass on the red sequence .
if quenching is due to the depletion of a gaseous reservoir , one might expect its signature to be evident in a decrease in the mean hi gas fraction .
however if the quenching is due to internal processes that inhibit star formation ( e.g. , * ? ? ?
* ) , then the signature of quenching might also be reflected in a decreased star formation efficiency ( e.g. sfe @xmath13 ) or the equivalent converse , an increased gas consumption timescale ( @xmath14 ) . while sfr efficiencies have been discussed extensively ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , this work marks the first time that it has been determined for a representative sample of galaxies selected exclusively by stellar mass .
in this paper we measure the distribution of sfr efficiencies across the gass sample and ask whether we find an excess of highly efficient or inefficient star - forming galaxies ?
we also investigate how the mean efficiency varies across the sample .
additionally , we take advantage of our simple selection criterion to produce a determination of several volume - averaged physical quantities .
we combine gass and galex measurements of @xmath0190 galaxies with recent determinations of the local stellar mass function to calculate the hi mass density and sfr density as a function of stellar mass .
this analysis allows us to compare the properties of our mass - selected sample to that of the full population .
we emphasize here that the sample , while restricted to relatively massive galaxies , includes many galaxies that are blue and/or not quiescent , in contrast with samples selected by color or early - type morphology , the latter of which are known to have low gas fractions and low star formation rates ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
in fact , as we discuss below , a significant fraction of the total star formation rate in the local universe is taking place within galaxies in this stellar mass range .
we use these measurements to address several fundamental questions : how much of the total hi in the local universe is associated with massive galaxies ?
how does this change across the so - called transition mass of @xmath15 ?
how does the hi density compare with the measured sfr density over the same mass range , and what does this suggest regarding the global sfr efficiency of massive galaxies ?
what fraction of this gas is in the process of efficiently forming stars , vs. the fraction that may be building up a reservoir for future star formation ?
we interpret our results in the context of quenching models , as well as scenarios that may lead to the return of a galaxy back onto the star - forming sequence . throughout this paper
, we make use of the flat @xmath16cdm cosmology with @xmath17 = 70 km s@xmath4 mpc@xmath4 and @xmath18 .
our hi data are taken from the gass first data release described in paper i and we refer the reader there for details on the observations and initial data analysis .
we briefly review the most pertinent information .
our large parent sample ( ps ) contains 12006 galaxies with @xmath1 and 0.025@[email protected] visible from arecibo and located within the footprint of the sdss primary spectroscopic survey , the projected galex medium imaging survey ( mis ) and alfalfa .
the first release of gass data contains @xmath20 galaxies of which 176 are new measurements , with 99 detections and 77 upper limits .
another @xmath21 previously detected gas - rich galaxies ( from alfalfa and other surveys ) are added to produce the statistically representative , volume - limited sample ( dr1 ) used in this paper .
dr1 contains 20% of the galaxies planned for the full gass survey .
hi masses are calculated for the detected galaxies , and upper limits determined for the non - detections . for this measurement our targets are considered to be unresolved by the arecibo 3.5@xmath22 beam .
additionally , no correction is made for self - absorption .
detected hi masses range from 4.6@xmath23 to 3.2@xmath24 .
upper limits ( 5@xmath25 ) from non - detections assume a 300 km / s velocity width .
upper limits are indicated on figures using arrows plotted at the location of the 5@xmath25 detection limit .
although velocity widths have been measured from the arecibo spectrum , we do not make use of those quantities in this paper .
we have investigated possible source confusion and contamination due to signal from galaxies close to or at the same redshift as our target galaxies . while such confusion does not have an influence on our main results , where appropriate we have indicated those galaxies for which confusion might cause an overestimate of the object s hi mass .
although the gass parent sample was defined using sdss dr6 , measurements and derived physical properties reported here were obtained using sdss dr7 , including the mpa / jhu value - added catalogs .
ultraviolet photometric measurements were calculated directly from pipeline - processed galex images based on galex data release gr6 .
these quantities have been tabulated in paper i. all photometric quantities used in this paper have been corrected for galactic extinction .
stellar masses have been calculated using sdss photometry only using the methodology described in @xcite .
following that work , we assume a @xcite imf for all derived quantities based on stellar masses and star formation rates , including gas mass fractions and star formation rate efficiencies .
stellar mass surface densities are calculated assuming that 50% of the stellar mass is contained within the r - band half - light radius . in this section
we describe the method we use to calculate star formation rates and star formation efficiencies .
star formation rates in this paper are derived from uv luminosities corrected for internal dust - attenuation .
a particular challenge is that the gass sample contains many galaxies with low level star - formation activity for which old and/or non - star - forming components ( evolved stars , agn ) can produce a uv luminosity comparable to those of a faint young stellar population .
this issue is not unique to the ultraviolet deriving sfrs for nearly passive galaxies is challenging using any star formation diagnostic ( e.g. uv , ir , h@xmath26 , radio continuum , etc . ) .
furthermore , measuring dust attenuation in individual galaxies is a complex problem , and nearly unfeasible with low s / n data .
a full treatment of these issues is beyond the scope of this paper , and we adopt a relatively simple uv - optical based approach following analyses in previous work .
interesting alternative multi - wavelength methods for deriving sfrs across the galaxy population have been developed @xcite and we will explore these in future work . a simple , one - component expression for calculating star formation rates using observed uv luminosities is given by : @xmath27 where @xmath28 is the ( star - formation history and metallicity - dependent ) conversion factor between uv luminosity and recent - past - averaged star formation rate , a@xmath29 is the ( geometrically - averaged ) ratio of intrinsic uv luminosity to measured uv luminosity , and f@xmath29(young ) is a measure of the fraction of uv light that originates in a young as opposed to a highly evolved stellar population . all of these quantities are highly simplified and can not be considered to be independent , but for our derivations and discussion below they will be treated as such . since the gass sample is drawn from sdss and therefore has 7 band photometry ( @xmath30 ) and spectral line indices , we use dust attenuations ( a@xmath31 ) derived using the methodology described in @xcite , slightly modified to produce more accurate sfrs for galaxies with evolved stellar populations .
variations on this methodology have been proposed ( e.g. * ? ? ?
* ) and we have checked this method against ours to ensure sure that applying different corrections has little impact on our conclusions .
we do not employ the `` hybrid '' attenuation correction used in @xcite because of the unavailability of the z - band dust attenuation measure ( @xmath32 ) , although we have calculated similar dust attenuation measures using @xmath33 in order to verify the robustness of the results reported here .
we also do not make use of h - alpha derived star formation rates because emission lines with sufficient s / n are only available for a subset of the sample and fiber aperture corrections can be large for our ( relatively ) low - redshift sample .
@xcite used the ir / uv flux ratio ( irx ) from a sample of 1000 sdss galaxies to derive a uv - based dust attenuation measure ( @xmath34 ) . by combining uv - optical colors with d@xmath35(4000 ) , which correlates with star formation history
, @xcite showed that : @xmath36 where @xmath37 and @xmath38 and coefficients have been taken from table 2 in @xcite .
@xcite show that the empirically derived @xmath34 shows a close correspondence to @xmath39 for galaxies with d@xmath35(4000)@xmath191.7 and using the @xcite attenuation curve one can show that @xmath40 . for galaxies with d@xmath35(4000)@xmath411.7
the ir / uv flux ratio is also sensitive to dust heating by light from evolved stars and is not a good measure of the dust attenuation of a young stellar population .
we attempt to bound the range of possible dust attenuations by considering two distinct cases . in the first ( dc , d4ncut ) , we assume that the ir luminosity in galaxies with d@xmath35(4000)@xmath411.7 is reprocessed light from an old population and we therefore set a@xmath42 for galaxies with d@xmath35(4000)@xmath411.7 .
in the second case ( dc ) , we apply our empirically derived dust attenuation corrections to the entire sample .
we make a further simplification by assuming that f@xmath29(young ) = 1 , or that all of the uv light in our measurements comes from a young stellar population .
measurements of the uv - optical colors of early - type galaxies ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) show nuv - r @xmath05 - 6 .
such a color , converted into a specific star formation rate , leads to a value close to @xmath43 yr@xmath4 .
this implies that specific star formation rates close to this value will very likely include a contribution from old stars and will most likely lead to an overestimate of sfr , since f@xmath29(young ) may be much less than 1 .
while our assumption leads to an overestimate of sfrs in weakly star - forming galaxies , it ends up having a small effect on average quantities discussed in the first part of this paper .
star formation rates were calculated assuming a constant @xmath44 and therefore : @xmath45 as derived by @xcite assuming a chabrier ( 2003 ) imf and a continuous recent ( 100 - 300 myr ) star formation history , which makes these star formation rates directly comparable to those in that work and most other recent determinations . for a standard salpeter imf ( between 0.1 and 100 @xmath46 ) , star formation rates would be a factor of @xmath47 higher .
in this paper we are primarily focused on quantities derived using the ratio of mean values . in several cases , but primarily for the star formation rates , these mean quantities can be dominated by a few very high star formation rate galaxies .
therefore , we took some care to ensure that our results were not affected by possible errors in measured sfrs for galaxies with the highest sfr .
many of these galaxies are ir - luminous and detected by iras .
we have compared our derived star - formation rate and the total infrared luminosity , obtained from the iifscz ( imperial iras fsc redshift catalog ; * ? ? ?
overall , the agreement is sufficiently good that differences are unlikely to affect our results .
it is possible , however , that we could be underestimating the sfr for the most ir - luminous galaxies , thereby underestimating the star formation efficiencies for a subset of our sample . in figure
[ fig : sfr_compare ] we compare sfrs obtained for both cases described above , to a subset of gass parent sample galaxies with sfrs calculated previously by @xcite .
we note that the dc , d4ncut sample compares quite favorably to @xcite and we adopt this as the primary sample for this paper .
finally , hi masses , stellar masses and star formation rates are used to derive hi - based gas fractions ( @xmath48 ) , star formation efficiencies ( sfe= ) and specific star formation rates ( ) .
we follow @xcite in adopting sfe , the sfr - to - gas ratio , as our primary derived quantity , as opposed to its reciprocal , often called the gas consumption timescale ( or the `` robert s time '' ) .
although our definition is straightforward , it is not necessarily identical to the star formation efficiency term sometimes used in star formation laws and related prescriptions . for all of our derived quantities
we have opted for the simplest possible definition .
we do not include a gas recycling term when calculating star formation efficiency .
although helium adds 26% to the cold gas mass we do not include this factor to calculate a cold gas mass , nor do we include a correction for the unmeasured molecular component .
this is in contrast , e.g. to the factor of 2.3 applied to hi masses in the singg survey @xcite .
we do note that the molecular phase in @xmath01/3 the gass sample is currently being studied in a corollary co survey being carried out at iram ( coldgass , kauffmann , kramer , saintonge 2010 ) .
the very simple selection criterion used for the gass ps allows us to combine our dr1 hi measurements with the local stellar mass function to determine the volume - averaged hi mass density for massive ( @xmath1 ) galaxies in the local universe . using sfr measurements we can also determine the analogous sfr density and volume - averaged sfe .
these volume - weighted quantities are derived using the local stellar mass function @xmath49 taken from @xcite , under the assumption that gass dr1 observations sample an unbiased , volume - limited distribution of galaxies in any particular stellar mass bin .
we first derive weights that allow us to scale the gass sample to match the stellar mass function of @xcite .
we also show that our weighting scheme allows us to ( trivially ) recover @xmath49 using the gass dr1 sample . to avoid effects that might arise from binning , we express the distribution of stellar masses in gass dr1 as a continuous ( gaussian - smoothed ) density function @xmath50 , shown in the bottom panel of figure [ fig : mf ] . for all @xmath51 such that @xmath52 : @xmath53 where @xmath54 is
a weight that corresponds to the inverse of the effective survey volume and @xmath55 is the bin width ( typically 0.25 ) .
we also define @xmath56 as the weighted - average stellar mass of any given bin . in the top panel of figure [ fig : mf ]
we plot the local stellar mass function from @xcite and compare it with @xmath57 based on gass dr1 , which we find to be consistent with the input stellar mass function .
we also plot the distribution of stellar masses in the ( volume - limited ) gass ps which shows a deficit at high stellar masses , likely due to sdss spectroscopic targeting criteria that avoid bright objects .
our method corrects for this incompleteness , which in any case is only significant for the highest stellar masses and is unlikely to impact our analysis .
we can use the same methodology to calculate @xmath58 and @xmath59 , the volume densities of hi and sfr as a function of m@xmath60 . again , because we are complete in any given @xmath61 bin we find that : @xmath62 and @xmath63
lastly , we apply a similar technique to determine the contribution of gass galaxies to the overall hi mass density function @xmath64 and sfr density function @xmath65 there are two caveats that should be noted with this calculation : 1
. the gass sample will not necessarily be complete in any given @xmath66 or sfr bin , and therefore the derived function should only be taken as a lower bound on the total function .
if we perform the calculation by setting the hi mass for non - detections to the hi upper limit , we may misrepresent the shape of the density function below the point where the gass sample is nearly complete . because this calculation can provide an approximate upper bound on the actual function
, we show these for illustrative purposes .
however , these upper - limit mass functions should be treated with caution .
if we consider a complete bivariate distribution function , e.g. @xmath67 we can evaluate a `` partial '' distribution function : @xmath68 where we evaluate this function for a given stellar mass limit ( e.g. @xmath69 for gass ) .
this partial distribution function can be simply calculated using our weight function derived above .
summing over all galaxies @xmath70 where @xmath71 and stellar mass @xmath72 , we find that @xmath73 as discussed above , we calculate this in two different ways to bracket our characterization of the function , either by omitting non - detections , or by setting their hi mass to our calculated upper limit .
the partial hi mass density function is then @xmath74 and integrating , we obtain the cumulative function @xmath75 @xmath76 the integral of the cumulative hi mass density function provides an estimate of the total contribution of gass galaxies to the hi density in the local universe .
a similar calculation applies for @xmath77 .
we show the partial hi mass density function calculated for the gass sample in figure [ fig : hi ] , and compare it to the total hi mass density distribution calculated using the @xcite hi mass function . in both plots
we show the hi mass density derived using only hi detections ( yellow curve ) , which can be taken to be a lower limit , and the hi mass density using upper limits as discussed above ( brown curve with points ) .
interestingly the gass - derived curve matches the total hi mass density down to @xmath78 , suggesting that nearly all galaxies with high hi - masses have stellar masses with @xmath79 .
the gass sample becomes increasingly less complete for hi masses below this limit .
we also find that gass galaxies with @xmath80 provide the bulk of the contribution to the total hi mass density , falling off significantly at lower gas masses .
this can also be seen in the sharp rise in the cumulative hi mass density @xmath81 over this hi mass range .
the cumulative mass density levels off at @xmath82 m@xmath83 mpc@xmath84 , @xmath9% of the total hi mass density derived using @xcite .
this integrated result shows that a significant fraction of the hi mass in the local universe is associated with massive galaxies .
there is little difference between the two curves , indicating that this result is robust with respect to the method that we use to account for non - detections .
we explore in figure [ fig : hisub ] how this distribution changes if we split our sample according to a galaxy s concentration , the ratio @xmath7 , which we use as a proxy for distinguishing disk - dominated ( @xmath7 @xmath192.6 ) and bulge - dominated ( @xmath7 @xmath412.6 ) galaxies ( as in e.g. * ? ? ?
* ) . this cut yields 54 ( 133 ) low ( high ) concentration galaxies .
we find that this split produces two nearly identical distributions , both contributing nearly equally to the total integrated hi mass density .
for the gass stellar mass range , disk - dominated galaxies do not account for the majority of the gas content of galaxies .
instead , hi appears to be evenly distributed across a range of galaxy types . from our derived sfr
density function we measure a partial sfr density of ( @xmath85 m@xmath83 mpc@xmath84 yr@xmath4 which is 47@xmath864% of the total sfr density @xcite and consistent with the results in those papers measured over the gass stellar mass range .
interestingly this suggests that gass galaxies account for nearly the same fraction of the total sfr density as they do in hi content .
we can also calculate an integrated volume - averaged sfe and implied gas consumption timescale which we determine to be @xmath87 yr@xmath4 ( timescale = [email protected] gyr ) for gass galaxies ( @xmath1 ) . because we have measurements of the total hi and sfr density
, we can calculate a sfe for lower mass galaxies ( @xmath88 ) , which we find to be @xmath89 yr@xmath4 ( timescale = [email protected] gyr ) .
the volume - averaged sfe appears to be relatively constant across the full galaxy population .
although our results suggest that the average sfe is similar above and below @xmath90 , it is possible that it only changes significantly above the transition mass at @xmath91 .
we show in figure [ fig : sfedist ] how the volume - averaged hi mass , sfr and sfe density functions vary with m@xmath60 .
quantities are plotted per m@xmath60 bin on the left , and as a cumulative quantity on the right . in the bottom panels
, we see that the trend in gas consumption remains quite flat across all stellar masses , remaining constant up to our highest stellar mass bin , though with increasing scatter . to explore this further , in fig [ fig : sfe_all ] and [ fig : sfe_all_obs ] we compare the global scaling relations of and sfe and find that they each paint a very different picture .
in the left panel of each pair of figures we plot the binned specific star formation rate ( @xmath92 ) as a function of stellar mass , mass surface density , concentration and color ( , @xmath93 , @xmath7 , and nuv - r ) . in all cases ,
is steadily declining , typically by a factor of 10 - 30 across the range of the gass sample .
in the right panels we show the average sfe ( @xmath2 , where hi non - detections are given zero hi mass ) .
the average sfe is nearly flat , straddling @xmath94 yr@xmath4 , corresponding to a gas depletion timescale of 3 gyr .
although in certain cases a small upwards or downward trend is suggested , these deviations are not statistically significant .
values for these quantities are given in tables [ tab : ssfr ] and [ tab : sfe ] .
we have chosen to plot bin - averaged quantities where the numerator and denominator are summed separately in order to simplify the treatment of non - detections .
we have checked that this trend is also apparent for mean or median @xmath95 and sfe measured for individual galaxies , the distribution of which is discussed in the next section .
this result is remarkable for two different reasons .
the first is that the timescale for gas consumption is nearly constant while the galaxy s specific star formation rate or `` building timescale '' is strongly dependent on stellar mass and correlated quantities .
the second remarkable aspect is that the global average sfe that we measure for the gass sample is very close to that observed locally for molecular gas in disks ( e.g. * ? ? ?
* using the things survey , walter et al .
this suggests that the entire hi reservoir is being converted into stars at the same rate as the molecular gas .
it would appear to argue against a `` bottleneck '' in the flow of gas onto galaxies occurring , on average , at the interface between the atomic and molecular phase .
instead it would appear that the gas - limiting step occurs prior to the hi phase ; that quenching of the detectable cold gas is not responsible for regulating the star formation history of galaxies .
we consider this result , and its relation to previous work , in the next section .
we first start out by asking whether a constant volume - averaged sfe is surprising , particularly for galaxies in the gass mass range ? while the global schmidt law suggests a star formation efficiency that rises with gas surface density or varies inversely with dynamical or free - fall timescale @xcite , more recent compilations of star formation laws ( e.g. , * ? ? ?
* ) consider fixed giant molecular cloud star - forming efficiencies ( with a varying atomic - to - molecular ratio ) and pressure regulation of the sfe and/or atomic - molecular ratio . because we measure neither the size of the gaseous and star - forming disk , nor the molecular phase , it is not easy to connect our global results to these theoretical predictions .
global galaxy - averaged quantities sample a range of gas surface densities , timescales and conditions within the ism .
additionally , our mean sfes combine measurements from galaxies with range of morphological types , environments and presumably dark halo masses and spin parameters .
it is a considerable theoretical challenge to interpret this result on its own .
observationally , low efficiencies have been measured for low - mass galaxies @xcite , lsb galaxies @xcite , and dlas @xcite . conversely high efficiencies have been measured in starburst galaxies ( e.g. * ? ?
* ; * ? ? ?
* ) and in galaxies undergoing interactions @xcite and/or some form of environmental disturbance ( e.g. stripping ) @xcite .
however , measurements of normal star - forming galaxies ( e.g. * ? ? ?
* ) show that galaxies follow the global schmidt law , implying a slowly varying sfe over the range of gas surface densities typically probed .
@xcite and @xcite noted a flat sfe across the hubble sequence using @xmath96 as a tracer .
a similar result has been obtained by @xcite and @xcite who obtain a nearly constant sfe in a sample of normal spirals .
more recently , @xcite suggested that the sfe is slowly increasing with galaxy luminosity ( implying that more luminous galaxies have shorter gas consumption timescales ) , but those data , which only include hi - detected galaxies , are nearly consistent with the constant values we have derived for the gass sample .
we return to the comparative question : why does the drop sharply with and @xmath93 while the sfe remains constant ? here we consider two different scenarios and leave a more detailed analysis for future work . _ internal regulation at the atomic - to - molecular transition : _ one possibility is that star formation is inhibited within the gas reservoir , at the sink point rather than the supply location . under the assumption that the efficiency of conversion of molecular gas into stars is nearly constant ( e.g. *
* ; * ? ? ?
* ) , regulation would then occur at the interface between the atomic and molecular phase .
processes that can stabilize a gaseous disk , such as those proposed by @xcite are possible examples . as discussed in the introduction , although such a process can explain a decreasing it may also result in a large reservoir of cold gas , leading to a decreasing sfe for galaxies that are actively being quenched .
therefore it appears hard to reconcile internal regulation with the constant sfe and decreasing vs. that we observe ( instead one might expect some correlation between sfe and ) . an alternative mechanism described in @xcite , with in - situ formation of molecular clouds in supershells ,
would also appear to be in conflict with our finding . _ quenching or throttling of the hi supply : _ a separate class of distinct processes are those that control the supply before ( or quasi - instantaneously as ) gas settles into the hi phase .
this includes both the `` ejective feedback '' and `` preventive feedback '' quenching mechanisms discussed in @xcite .
for example , star formation ( or agn ) may drive outflows , effectively ejecting gas from the system .
agn feedback may heat infalling gas , or prevent its cooling , and allow the build up of a reservoir of gas in a bound hidden phase ( e.g. ionized , warm - hot ) .
although such a hidden phase has been invoked to balance accretion and star formation rates in the milky way @xcite , more generally quenched gas in this phase may or may not provide an additional supply for star formation through halo accretion , in that it may be permanently quenched , or may subsequently cool on a longer timescale ( e.g. , * ? ? ?
* ) . additionally there are throttling scenarios that invoke varied gas accretion histories , essentially regulating the rate of infall onto the halo , with possible links to environmental conditions that lead to `` starvation '' or `` strangulation '' @xcite .
finally , this category also includes delayed or staged accretion histories @xcite where the effective accretion timescale varies with stellar mass . in principle , what all of these have in common is that they can produce significant variation in vs. stellar mass while allowing the sfe to remain constant . interestingly ,
if interpreted in light of recent models explaining the star formation law in atomic and molecular gas @xcite , our result suggests that on average , most of the hi in gass galaxies must reside at surface densities at or near the atomic - to - molecular transition , with little or no systematic trend across the sample .
our conclusion , that quenching / throttling mechanisms appear to better explain our data , has already been hinted at in previous work , most notably @xcite and @xcite both of which identified similar observational trends in disk galaxy samples .
we note here that our average result does not preclude internal regulation from taking place in some , but not all , of the massive galaxies in our sample .
it is possible that a `` conspiracy '' resulting from galaxies with low , internally quenched sfe being balanced by galaxies with high sfes could reproduce the mean scaling relation .
we can investigate this by studying the actual distribution of sfe vs. stellar mass and other properties across the sample , which we do below .
scatter around the mean relations discussed above may provide insight into the episodic nature of gas accretion and quenching in gass galaxies and we explore here the detailed distribution by plotting values and upper limits for individual galaxies in the dr1 sample .
we note here that while large uncertainties in some of the derived sfrs and sfes can obscure visible dependencies , such upper limits and sources of error for low - sfr or low gas mass galaxies are unlikely to affect the global trends discussed in previous sections . in general
, their contribution to summed values is small . on the other hand
, random and systematic errors will also shift high sfr galaxies in these plots . as discussed above
, we ve verified that our global results are robust with respect to such possible sources of error , but errors in sfrs , typically @xmath00.3 dex , may produce some of the outliers on the plots described here . results quoted in this section should be treated as suggestive , motivating further study and improvements in the accuracy of these measures . in figure
[ fig : ssfr_sfe_mass ] we plot the specific star formation rate and star formation efficiencies as a function of stellar mass . also shown is the distribution of for galaxies in the local universe derived using volume - corrected uv - optical data from galex and sdss @xcite .
as expected , in any given stellar mass bin , gass shows a similar distribution in as the volume - corrected galex+sdss sample , consistent with the fact that gass galaxies are selected purely based on stellar mass , with no other selection bias .
our sample of massive galaxies is not purely passive ; a significant fraction of gass galaxies are on or close to the star - forming sequence .
in fact , many of the galaxies are forming stars at rates higher than 1 m@xmath83 yr@xmath4 . among the population of massive galaxies are some objects that have some of the highest sfrs in the local universe . there is a locus of galaxies near @xmath43 yr@xmath4 , on the non - sf ( red or dead ) sequence .
as discussed above and in @xcite , the for many of these galaxies is likely to be an overestimate due to the fact that uv light from evolved populations has not been subtracted when calculating star formation rates .
although we do not label the red points as such , it is best to consider sfrs for these galaxies as an upper limit .
the right - hand side of figure [ fig : ssfr_sfe_mass ] shows the distribution of star formation efficiencies across the sample . there is a broad spread of efficiencies in the sample with implied gas consumption timescales ranging from less than 1 gyr to over 100 gyr .
in fact , this spread is indeed much larger than that typically seen for molecular gas within disk galaxies ( e.g. figure 15 of * ? ? ? * ) .
we highlight in this and future figures those galaxies with exceptionally high and low star formation efficiencies .
galaxies with sfe @xmath8 yr@xmath4 are forming stars at rates that are high compared to their present atomic gas mass ( though still longer than likely dynamical timescales , consistent with * ? ? ? * ) .
galaxies with sfe @xmath97 yr@xmath4 appear to be passive in comparison to their current gas content .
gass hi non - detections are shown with upward arrows , suggesting that their efficiencies could be higher .
the vast majority of the non - detections also have very low specific star formation rates which could suggest that estimates of their sfrs and sfes are too high . as a result ,
the red points that are hi non - detections may have higher _ or _ lower sfes , and are poorly constrained in this diagram .
overall , the broad scatter in star formation efficiencies vs. stellar mass does not point to a strong correlation between the two .
the high - sfe galaxies , indicated with large blue circles , are the galaxies with the shortest gas consumption timescales .
these galaxies are predominantly above the transition mass , @xmath98 .
although these galaxies have elevated sfes , they do not occupy an unusual location in the vs. @xmath61 diagram , with most situated on or near the star - forming sequence .
this suggests that the reason why their star formation efficiency is high is because they have low gas mass for their given present averaged star - formation rate .
stellar mass surface density is known to display a strong correlation with specific star formation rate @xcite and we show that the gass sample displays a similar correlation in figure [ fig : logmu ] . above a transition stellar mass surface density of @xmath99 kpc@xmath100 a much greater spread of is observed at a given @xmath93 .
this transition threshold has already been noted in paper i and in previous work @xcite . on the other hand ,
the sfe , unlike the trend vs. stellar mass , shows a rather curious `` t - shaped '' distribution . below the @xmath93 transition ,
the sfes are nearly constant , straddling the line around sfe @xmath101 yr@xmath4 . above the transition ,
the distribution in sfe becomes extremely broad , with very high and very low efficiencies present .
the highest sfe galaxies occupy the knee of the vs. @xmath93 diagram and might themselves represent a transition population ( see e.g. * ? ? ?
the low sfe galaxies , on the other hand , appear to reside with passive galaxies .
these galaxies host a sufficiently large gaseous reservoir that subsequent future star formation may lead to evolution off of the red sequence .
our results highlight the fact that the combination of gas , star formation and stellar mass measurements can be useful for obtaining information about the possible future evolution of massive galaxies .
we explore this further in figure [ fig : ssfr_gfrac_sfe ] by investigating three plots that link gas fraction to and sfe . in the first plot comparing sfe vs. hi gas mass fraction ( upper left )
, we see that high sfe galaxies have low gas fractions , with the converse being true for low sfe galaxies .
star - forming and green valley galaxies display a range of gas fractions from 2% to 100% .
we plot sfe vs. ( upper right ) , and find that the highest sfe galaxies do not have the highest specific star formation rates , but instead peak at slightly lower values .
this suggests that a high sfe is not driven by a high sfr .
not surprisingly , the gas fraction vs. relation ( lower right ) shows that star - forming and green valley galaxies reveal correlated gas fraction and specific star formation rates with the extreme sfe galaxies lying off of this relation .
all of these plots , taken together , suggest that it is low hi content , as opposed to an excess in star formation rate , that is responsible for the majority of high sfe galaxies .
figures [ fig : ssfr_sfe_mass]-[fig : ssfr_gfrac_sfe ] can be used to isolate galaxies that show gas excess or deficiency when compared to their current star formation rates , useful for determining what causes high or low sfe in galaxies .
figure [ fig : delta_ssfr_gfrac ] attempts to combine this diagnostic capability in one diagram , by plotting the offset of a galaxy s star formation rate and hi mass relative to the average value for a given stellar mass .
we use data from this paper and paper i and perform linear regression fits to the mean sfr and m@xmath102 vs . for the hi masses we use the mean values for which the masses of non - detections have been set to the upper limit ( paper i , table 4 , column 1 ) .
we find that @xmath103 and @xmath104 and define @xmath105 @xmath106 in this delta gas - fraction vs. delta specific star formation rate plot four quadrants are delineated .
the top right quadrant contains those galaxies with high and , in other words galaxies that are gas - rich and actively star - forming .
the majority of the gass detections fall here along a line where gas fraction excess equals the excess which corresponds approximately to a line of constant sfe .
this quadrant is also likely to include gas - rich mergers , starbursts and high surface brightness galaxies with extended uv - disks @xcite .
the upper left quadrant contains gas rich galaxies with lower than average star formation rates , including the low sfe galaxies highlighted above .
low surface brightness galaxies and galaxies that have recently accreted gas might be found here . the majority of galaxies identified as low sfe are forming stars below average rates , consistent with their being passive galaxies with large reservoirs of gas .
these galaxies do not appear to have extremely high gas masses and average star formation rates ( which may be more typical of galaxies with lower stellar mass ) .
gass3505 , an unusually gas - rich galaxy discussed in paper i , belongs to this class of object and is indicated with a red star in the figure .
the lower left quadrant contains gas - poor galaxies with below average sfr .
passive galaxies are found in this quadrant .
finally the lower right quadrant contains high sfe galaxies that are relatively gas poor but with above average sfr . on this plot
it becomes apparent that these galaxies are distributed across the quadrant and appear to have either high star formation rates when compared to their ( typical ) gas mass , or they have lower than average gas mass but typical star formation rates .
the most gas deficient in this category have possibly very recently experienced a process that disrupted gas flow and/or removed gas , such as starvation or stripping .
the latter scenario is likely to produce gas deficiencies even lower than we have observed .
later stages of such galaxies might also be found in the third quadrant below or near the line of constant sfe .
gass 7050 , identified in paper i as an unusual gas - poor galaxy , appears to belong to this category and is indicated with a green star in the figure .
lastly , we return to the question of whether or not the scatter around our mean scaling relations suggests a diversity in the processes that trigger and quench star formation in the gass sample . without question the gass sample does not show a tight distribution in sfes , and it is tempting to reconsider internal quenching mechanisms for at least some fraction of systems .
alternately , our results may rule out scenarios where the inflow of material is occurring as a steady flow ( or drizzle ) onto the galaxy . instead
, the scatter may be an indication that accretion of hi is episodic , with infalling gas arriving in larger discrete chunks .
it may suggest that large scale processes play a role in regulating the growth and evolution of gas and star formation in galaxies .
future work is being planned to investigate the large- and small - scale environment of gass galaxies , and in particular the outliers , which should reveal whether enviromental processes are driving this evolution .
some galaxies with high star formation rates may be undergoing a merger or interaction that is driving up the star formation rate in these galaxies and it will also be interesting to investigate the connection between signs of interaction , merging or other disturbance and location on this diagram .
we use measurements of the hi content , stellar mass and star formation rates in @xmath107 massive galaxies with @xmath108 , obtained from the galex arecibo sdss survey described in paper i @xcite to explore the global scaling relations associated with the ratio , which we call the hi based star formation efficiency ( sfe ) . we find that : 1
. we can measure the volume - averaged hi mass density and sfr density for the gass sample .
gass galaxies account for @xmath9 percent of the total hi mass density and @xmath10 percent of the sfr mass density .
2 . unlike the mean specific star formation rate , which decreases with stellar mass and stellar mass surface density , the star formation efficiency remains relatively constant across the sample with a value close to @xmath3 yr@xmath4 ( or an equivalent gas consumption timescale of @xmath5 yr ) .
we find little variation in sfe with stellar mass , stellar mass surface density , @xmath6 color and concentration .
we interpret these results as an indication that external processes or feedback mechanisms that control the gas supply are important for regulating star formation in massive galaxies .
approximately 5% of the sample shows high efficiencies with sfe @xmath8 yr@xmath4 , and we suggest that this is very likely due to a deficiency of cold gas rather than an excess star formation rate .
conversely , we also find a similar fraction of galaxies that appear to be gas - rich for their given specific star - formation rate , although such galaxies show both a higher than average gas fraction and lower than average specific star formation rate . both of these populations are plausible candidates for `` transition '' galaxies , with potential for a change ( either decrease or increase ) in their specific star formation rate in the near future .
we thank the arecibo staff , in particular phil perillat , ganesan rajagopalan and the telescope operators for their assistance , and hector hernandez for scheduling the observations . r.g . and
acknowledge support from nsf grant ast-0607007 and from the brinson foundation .
galex ( galaxy evolution explorer ) is a nasa small explorer , launched in april 2003 .
we gratefully acknowledge nasa s support for construction , operation , and science analysis for the galex mission , developed in cooperation with the centre national detudes spatiales ( cnes ) of france and the korean ministry of science and technology .
funding for the sdss and sdss - ii has been provided by the alfred p. sloan foundation , the participating institutions , the national science foundation , the u.s .
department of energy , the national aeronautics and space administration , the japanese monbukagakusho , the max planck society , and the higher education funding council for england .
the sdss web site is http://www.sdss.org/. the sdss is managed by the astrophysical research consortium for the participating institutions .
the participating institutions are the american museum of natural history , astrophysical institute potsdam , university of basel , university of cambridge , case western reserve university , university of chicago , drexel university , fermilab , the institute for advanced study , the japan participation group , johns hopkins university , the joint institute for nuclear astrophysics , the kavli institute for particle astrophysics and cosmology , the korean scientist group , the chinese academy of sciences ( lamost ) , los alamos national laboratory , the max - planck - institute for astronomy ( mpia ) , the max - planck - institute for astrophysics ( mpa ) , new mexico state university , ohio state university , university of pittsburgh , university of portsmouth , princeton university , the united states naval observatory , and the university of washington . | we use measurements of the hi content , stellar mass and star formation rates in @xmath0190 massive galaxies with @xmath1 , obtained from the galex arecibo sdss survey described in paper i @xcite to explore the global scaling relations associated with the bin - averaged ratio of the star formation rate over the hi mass ( i.e. , @xmath2 ) , which we call the hi - based star formation efficiency ( sfe ) . unlike the mean specific star formation rate , which decreases with stellar mass and stellar mass surface density , the star formation efficiency remains relatively constant across the sample with a value close to @xmath3 yr@xmath4 ( or an equivalent gas consumption timescale of @xmath5 yr ) .
specifically , we find little variation in sfe with stellar mass , stellar mass surface density , @xmath6 color and concentration ( @xmath7 ) .
we interpret these results as an indication that external processes or feedback mechanisms that control the gas supply are important for regulating star formation in massive galaxies .
an investigation into the detailed distribution of sfes reveals that approximately 5% of the sample shows high efficiencies with sfe @xmath8 yr@xmath4 , and we suggest that this is very likely due to a deficiency of cold gas rather than an excess star formation rate .
conversely , we also find a similar fraction of galaxies that appear to be gas - rich for their given specific star - formation rate , although these galaxies show both a higher than average gas fraction and lower than average specific star formation rate . both of these populations are plausible candidates for `` transition '' galaxies , showing potential for a change ( either decrease or increase ) in their specific star formation rate in the near future .
we also find that @xmath9 % of the total hi mass density and @xmath10 % of the total sfr density is found in galaxies with @xmath1 .
[ firstpage ] galaxies : evolution galaxies : fundamental parameters ultraviolet : galaxies radio lines : galaxies |
intermolecular energy transfer as a fundamental process in many biochemical and solid - state systems has been of increasing interest @xcite .
it is often distinguished between two cases , namely ( radiationless ) short - range transfer ( also called frster transfer @xcite ) and ( radiative ) long - range transfer . in the former the distance @xmath2 between donor and acceptor
is small compared with the electronic energy transfer wavelength @xmath3 , .
the free - space transfer rate behaves as @xmath0 , which can be explained by the instantaneous ( longitudinal ) coulomb interaction between the two molecules . in the latter
the intermolecular distance substantially exceeds the transition wavelength , .
the observed @xmath1 dependence of the transfer rate can be regarded as being the result of emission and reabsorption of real ( transverse ) photons .
it is worth noting that in a rigorous approach to the problem ( e.g. , within the framework of the multipolar formalism of qed @xcite ) the @xmath0 and @xmath1 distance dependences are limiting cases of a unified theory @xcite .
when the two molecules are near material bodies , then the electromagnetic field felt by them can be quite different from that in free space and the intermolecular energy transfer can change accordingly .
the effect has attracted attention , because it offers the possibility of controlling the energy transfer , with regard to potential applications , e.g. , in high - efficiency light - harvesting systems , optical networks , and quantum computing .
enhanced energy transfer between molecules randomly distributed within a single glycerol droplet ( of about 10@xmath4 m diameter ) @xcite and within a polymer fabry - prot microcavity @xcite has been observed .
using monomolecular layers of donor and acceptor molecules ( separated by distances of @xmath5 nm ) in planar microstructures , the dependence of short - range energy transfer on the local photon mode density has been demonstrated @xcite .
calculations of the energy transfer rate have been performed in order to include the effect of bulk material @xcite , microspheres @xcite , and planar microcavities @xcite . the quantum theory given in ref .
@xcite is based on a microscopic model that allows for both dispersing and absorbing bulk material . in refs .
@xcite the classical field generated by a donor dipole and felt by an acceptor dipole in the presence of a microsphere is substituted into the free - space fermi s golden rule expression .
a strictly quantum mechanical treatment that starts from a mode decomposition of the electromagnetic field according to the helmholtz equation of the macroscopic maxwell equations is given in refs .
unfortunately , the microscopic theory developed for bulk material @xcite becomes quite cumbersome when boundaries are present , and studies based on the standard mode expansion @xcite can not incorporate material absorption . in the present paper
we give a rigorous derivation of the rate of intermolecular energy transfer in the presence of arbitrarily shaped , dispersing , and absorbing material bodies , starting from the quantized version of the macroscopic electromagnetic field .
the quantization is based on the introduction of langevin noise current and charge densities into the classical maxwell equations , which can then be transferred to quantum theory , with the electromagnetic - field operators being expressed in terms of a continuous set of fundamental bosonic fields via the classical green tensor ( see @xcite and references therein ) . in particular , we show that the minimal - coupling scheme and the multipolar - coupling scheme yield exactly the same form of the rate formula .
it is worth noting that the formalism includes material absorption and dispersion in a consistent way , without restriction to a particular frequency domain , and applies to an arbitrary ( inhomogeneous ) medium configuration . here
, we apply the theory to bulk material , multi - slab planar structures , and microspheres , with special emphasis on media of drude lorentz type .
in particular , we show that the energy transfer can be strongly modified , if the two molecules are sufficiently near an interface and surface - guided waves at the energy transfer wavelength exist .
four- and five - layered planar structures are studied , and the results are compared with recent measurements @xcite . finally , the effect of surface - guided waves and whispering - gallery waves in the case of the molecules being near a microsphere is briefly discussed .
the paper is organized as follows . in section [ rate ] the basic - theoretical concept of electromagnetic - field quantization
is outlined and the energy transfer rate is derived .
section [ disc ] is devoted to applications , with special emphasis on multi - slab planar structures , and concluding remarks are made in section [ conl ] .
some deepening calculations are given in the appendix .
let us consider an ensemble of point charges , interacting with the quantized electromagnetic field in the presence of absorbing media .
the minimal - coupling hamiltonian in coulomb gauge reads @xcite @xmath6&&\hspace{4ex } + \sum_\alpha { 1\over 2m_\alpha } \left [ \hat{\bf p}_\alpha - q_\alpha \hat{\bf a}(\hat{\bf r}_\alpha ) \right]^2 % } \nonumber\\[.5ex]&&\hspace{4ex } + \,{\textstyle\frac{1}{2 } } \int\ ! { \rm d}^3{\bf r}\ , \hat{\rho}({\bf r } ) \hat{\phi}({\bf r } ) + \int\ !
{ \rm d}^3{\bf r}\ , \hat{\rho}({\bf r } ) \hat{\varphi}({\bf r } ) , \end{aligned}\ ] ] where @xmath7 is the position operator and @xmath8 is the canonical momentum operator of the @xmath9th ( nonrelativistic ) particle of charge @xmath10 and mass @xmath11 .
the first term of the hamiltonian is the energy of the medium - assisted electromagnetic field , expressed in terms of bosonic vector fields @xmath12 with commutation relations @xmath13
= \delta_{kk ' } \delta({\bf r}\!-\!{\bf r}')\delta(\omega\!-\!\omega'),\ ] ] @xmath14 = 0\,.\ ] ] the second term is the kinetic energy of the charged particles , and the third term is their coulomb energy , where the corresponding scalar potential @xmath15 is given by @xmath16 with @xmath17 being the charge density of the particles , and @xmath18 the vacuum dielectric permittivity .
the last term is the coulomb energy of interaction of the particles with the medium .
the scalar potential @xmath19 and the vector potential @xmath20 of the medium - assisted electromagnetic field are given by @xmath21 @xmath22 where @xmath23 with @xmath24 and @xmath25 being the transverse and longitudinal dyadic @xmath26-functions , respectively , and @xmath27 here , @xmath28 is the classical green tensor , which obeys the inhomogeneous , partial differential equation @xmath29 \bbox{g}({\bf r},{\bf r}',\omega ) = -\,\bbox{\delta}({\bf r}-{\bf r}')\ ] ] together with the boundary condition at infinity [ @xmath30 is the dyadic @xmath26-function ] , with @xmath31 @xmath32 @xmath33 @xmath34 @xmath35 being the complex , space- and frequency - dependent permittivity .
let us consider the case where the particles are constituents of neutral molecules ( at positions @xmath36 ) that are well separated from each other .
the hamiltonian ( [ n001 ] ) can then be decomposed into an unperturbed part @xmath37 and an interaction part @xmath38 as follows @xmath39 @xmath40 @xmath41 here , @xmath42 is the hamiltonian of the @xmath43th molecule , @xmath44 is the coulomb interaction energy between the @xmath43th and the @xmath45th molecule , and @xmath46&&\hspace{4ex } + \sum_{\alpha_m } \left(\frac{q_{\alpha_m}^2}{2m_{\alpha_m}}\right ) \hat{\bf a}^2(\hat{\bf r}_{\alpha_m } ) + \int\ ! { \rm d}^3{\bf r}\ , \hat{\rho}_m({\bf r } ) \hat{\varphi}({\bf r } ) \end{aligned}\ ] ] is the interaction energy between the @xmath43th molecule [ charge density @xmath47 and the medium - assisted electromagnetic field . in what follows
we shall restrict our attention to the ( electric-)dipole approximation , so that eq .
( [ n017 ] ) simplifies to @xmath48 where @xmath49 is the dipole operator of the @xmath43th molecule . disregarding the @xmath50 term in eq .
( [ n018 ] ) , which does not give rise to off - diagonal molecular matrix elements , making use of eqs .
( [ n009])([n011 ] ) , and applying the dipole approximation , @xmath51 takes the form of @xmath52 where @xmath53 \mbox{\boldmath $ \delta$}^\perp({\bf r}-{\bf r}_m ) } \nonumber\\&&\hspace{15ex } + \hat{\bf d}_m { } \mbox{\boldmath $ \delta$}^\parallel({\bf r}-{\bf r}_m ) .
\end{aligned}\ ] ] let us consider the resonant energy transfer between two molecules @xmath54 and @xmath55 at positions @xmath56 and @xmath57 .
the initial ( final ) state @xmath58 ( @xmath59 ) describes the excited molecule @xmath54 ( @xmath55 ) , the molecule @xmath55 ( @xmath54 ) being in the ground state , and the medium - assisted field in vacuum , @xmath60 @xmath61 ( cf .
@xcite ) . note that imposing this initial condition requires that the time of state preparation is sufficiently short compared with the time of energy transfer . using the born expansion @xcite up to the second order perturbation theory , the ( total ) rate of energy transfer can be given by @xmath62 where @xmath63 is the occupation probability of the state @xmath58 , and @xmath64 with @xmath65 applying the decomposition ( [ n015 ] ) ,
we may write @xmath66&&\hspace{4ex } = \langle a , b'| \hat{v}_{ab}|a',b\rangle + \langle a , b'|\hat{\cal t}|a',b\rangle , \end{aligned}\ ] ] where @xmath67 } \nonumber\\[.5ex]&&\hspace{4ex}\times \left[e_i-\hat{h}_0 + is\right]^{-1 } \left[\hat{h}_{a\,\rm int } + \hat{h}_{b\,\rm int}\right]|a',b\rangle .
\end{aligned}\ ] ] let us first consider the coulomb term @xmath68 , \end{aligned}\ ] ] where @xmath69 in order to calculate @xmath70 , we make use of eqs .
( [ n019 ] ) and ( [ n020 ] ) , perform the summation and integrations over the possible intermediate states @xmath71 and @xmath72 . after some calculation
we derive , on applying eq .
( [ n007 ] ) and the relationship @xcite , @xmath73&&\hspace{2ex } = \int { \rm d}^3{\bf s}\ , \frac{\omega^2}{c^2}\ , \varepsilon_{\rm i}({\bf s},\omega ) g_{km}({\bf r},{\bf s},\omega ) g^\ast_{lm}({\bf r'},{\bf s},\omega ) , \end{aligned}\ ] ] @xmath74&&\hspace{1ex } \times \left\ { \frac{\left[{\bf d}_{b'b } { } \bbox{\delta}_b({\bf r}',-\omega ) { } { \rm i m } \,\bbox{g } ( { \bf r}',{\bf r},\omega ) { } \bbox{\delta}_a({\bf r},-\omega ) { } { \bf d}_{aa'}\right ] } { -\hbar \omega_{a'a}-\hbar \omega + is}\right .
\nonumber\\[.5ex]&&\hspace{2ex } \left .
+ \,\frac{\left[{\bf d}_{b'b } { } \bbox{\delta}_b({\bf r}',\omega ) { } { \rm i m } \,\bbox{g } ( { \bf r}',{\bf r},\omega ) { } \bbox{\delta}_a({\bf r},\omega ) { } { \bf d}_{aa'}\right ] } { \hbar \omega_{a'a}-\hbar \omega + is}\right\ } , \end{aligned}\ ] ] where @xmath75 and @xmath76 [ note that ] .
recalling that @xmath77 @xmath32 @xmath78 , we may rewrite eq .
( [ n039 ] ) as @xmath79&&\hspace{2ex } \times \left\ { \frac{\left[{\bf d}_{b'b } { } \bbox{\delta}_b({\bf r}',\omega ) { } { \rm i m } \,\bbox{g } ( { \bf r}',{\bf r},\omega ) { } \bbox{\delta}_a({\bf r},\omega ) { } { \bf d}_{aa'}\right ] } { \hbar \omega_{a'a}-\hbar \omega + is\,\,\mbox{sign}\,(\omega ) } \right\ } \end{aligned}\ ] ] or , equivalently , @xmath80&&\hspace{1ex}\times\ , \frac{1}{2i}\,\left\ { \left [ \frac{\left[{\bf d}_{b'b } { } \bbox{\delta}_b({\bf r}',\omega ) { } \bbox{g } ( { \bf r}',{\bf r},\omega ) { } \bbox{\delta}_a({\bf r},\omega ) { } \bbox{\bf d}_{aa'}\right ] } { \hbar \omega_{a'a}-\hbar \omega + is\,\,\mbox{sign}\,(\omega)}\right]\right .
\nonumber\\&&\hspace{2ex } \left . - \left [ \frac{\left[{\bf d}_{bb ' } { } \bbox{\delta}_b({\bf r}',\omega ) { } \bbox{g } ( { \bf r}',{\bf r},\omega ) { } \bbox{\delta}_a({\bf r},\omega ) { } { \bf d}_{a'a}\right ] } { \hbar \omega_{a'a}-\hbar \omega -is\,\ , \mbox{sign}\,(\omega)}\right]^*\right\}. \end{aligned}\ ] ] the @xmath81-integral in eq .
( [ n041 ] ) may now be evaluated by means of contour - integral techniques , by taking into account that the green tensor is a holomorphic function of @xmath81 in the upper complex half - plane , which asymptotically behaves as @xcite @xmath82 we therefore close the path of integration by an infinitely large semicircle in the upper complex half - plane , , and subsequently subtract the semicircle integral .
it is easily seen that only the terms in @xmath83 and @xmath84 [ eq . ( [ n029 ] ) ] which are proportional to @xmath81 contribute to the integral over the semicircle , @xmath85.\ ] ] it is further seen that only the first term in the curly bracket contributes to the integral over the closed path .
we thus arrive at @xmath86 } \nonumber\\[.5ex]&&\hspace{6ex } -\,\frac{\omega_{a'a}^2}{\varepsilon_0 c^2 } \,\left[{\bf d}_{b'b } { } \bbox{g } ( { \bf r}_b,{\bf r}_a,\omega_{a'a } ) { } { \bf d}_{aa'}\right ] .
\end{aligned}\ ] ] substitution of the expressions ( [ n038 ] ) and ( [ n045 ] ) into eq .
( [ n037 ] ) yields the transition amplitude @xmath87 .
\end{aligned}\ ] ] note that the first term in eq .
( [ n045 ] ) and the coulomb term ( [ n038 ] ) exactly cancel out .
we eventually combine eqs .
( [ n023 ] ) and ( [ n046 ] ) and find that the rate of energy transfer between the chosen states @xmath88 and @xmath89 reads as ( ) @xmath90&&\hspace{4ex}\times\ , \left| { \bf d}_{b'b } { } \bbox{g } ( { \bf r}_b,{\bf r}_a,\omega_{a'a } ) { } { \bf d}_{aa'}\right|^2 \delta(\omega_{a'a}-\omega_{b'b } ) .
\end{aligned}\ ] ] it can be proved ( appendix [ multipol ] ) that the use of the multipolar hamiltonian @xcite instead of the minimal - coupling hamiltonian ( [ n001 ] ) exactly leads to the same expression of the energy transfer rate .
let us now consider the total energy transfer rate according to eq .
( [ n023a ] ) , by taking into account the vibronic structure of the molecular energy levels . restricting our attention to
oppenheimer approximation and neglecting the weak dependence of the electronic transition - dipole matrix element on the nuclear coordinates ( see , e.g. , @xcite ) , we may factorize the dipole transition matrix elements according to @xmath91 where @xmath92 is the purely electronic transition - dipole matrix element of the transition between the lower and the upper electronic state of the molecule @xmath93 , and @xmath94 are the overlap integrals between the vibrational quantum states in the two electronic states of the respective molecule .
note that the vibrational overlap integrals take account of both displaced and distorted energy surfaces .
combining eqs .
( [ n023a ] ) and ( [ n047 ] ) yields @xmath95&&\hspace{4ex}\times \left| { \bf d}^*_b { } \bbox{g } ( { \bf r}_b,{\bf r}_a,\omega_{a'a } ) { } { \bf d}_a\right|^2 \delta(\omega_{a'a}-\omega_{b'b } ) , \end{aligned}\ ] ] which can be rewritten as @xmath96 where @xmath97 and @xmath98 and @xmath99 respectively , are proportional to the ( single - photon ) emission spectrum of molecule @xmath54 and the ( single - photon ) absorption spectrum of molecule @xmath55 in _ free space _ each @xcite .
thus , the rate of energy transfer is proportional to the overlap of the two spectra weighted by the square of the absolute value of the actual green tensor .
it is worth mentioning that eqs .
( [ n047])([n047e ] ) apply to the resonant energy transfer between two molecules in the presence of an arbitrary configuration of dispersing and absorbing macroscopic bodies .
all the relevant parameters of the bodies are contained in the green tensor .
note that the emission ( absorption ) spectrum observed in this case is not proportional to @xmath100 $ ] in general , as it can be seen from a comparison of eq .
( [ n047d ] ) with eq .
( [ b8 ] ) . in particular
when the green tensor slowly varies with frequency on a scale given by the ( relevant ) vibrational frequencies of the molecules , then @xmath101 is also a slowly varying function of frequency and can ( approximately ) be taken at the electronic energy transfer frequency and put in front of the integral in eq .
( [ n048 ] ) , thus @xmath102 where @xmath103 in this case , the influence of matter environment on the ( total ) energy transfer rate is fully contained in @xmath104 .
clearly , when the two molecules are near a resonator - like equipment , so that the molecule can `` feel '' sharply - peaked field resonances , then @xmath101 can not be assumed to be a slowly varying function of frequency in general ( see section [ sphere ] ) .
it may be interesting to compare the rate of energy transfer with the donor decay rate .
straightforward generalization of the well - known formula for a two - level transition yields , on applying the born oppenheimer approximation , @xmath105 where @xmath106,\ ] ] and @xmath107 is given by eq .
( [ n047d ] ) .
whereas the decay rate is determined by the imaginary part of the green tensor ( taken at equal positions ) , the transfer rate is determined by the full green tensor ( taken at different positions ) .
thus , decay rate and transfer rate can quite differently respond to a change of the environment .
let us first consider the case when the two molecules are embedded in bulk material of arbitrary complex permittivity @xmath108 . using the well - known expression of the bulk - material green tensor @xmath109 ( see , e.g.,@xcite ) , application of eq .
( [ n047c ] ) yields @xmath110 where @xmath111 } \nonumber\\[.5ex]&&\hspace{1ex}\times \left [ - \left ( { \bf d}^*_b { } { \bf d}_a - 3 \frac{{\bf d}^*_b { } { \bf r}}{r}\ , \frac{{\bf d}_a { } { \bf r}}{r } \right ) \left ( { 1\over q^3(\omega)r^3}- { i\over q^2(\omega)r^2 } \right ) \right . \nonumber\\[.5ex]&&\hspace{7ex } \left
. + \left ( { \bf d}^*_b { } { \bf d}_a - \frac{{\bf d}^*_b { } { \bf r}}{r}\ , \frac{{\bf d}_a { } { \bf r}}{r } \right ) { 1\over q(\omega)r } \right ] \end{aligned}\ ] ] with @xmath112 the energy transfer rate is then obtained according to eq .
( [ n048 ] ) .
obviously , the green tensor of bulk material can be regarded as being a slowly varying function of frequency , so that the approximation ( [ n047f ] ) applies . from eqs .
( [ n057 ] ) and ( [ n051 ] ) it is seen that the energy transfer rate includes both the small - distance case ( frster transfer ) , with the rate being proportional to @xmath0 , and the large - distance ( radiative ) case , where the rate becomes proportional to @xmath1 .
note that the exponential @xmath113|^2 $ ] @xmath32 @xmath114 $ ] , which typically arises from material absorption , drastically diminishes the large - distance energy transfer [ @xmath115 @xmath32 @xmath116 @xmath32 @xmath117 . in eq .
( [ n057 ] ) local - field corrections are ignored .
they may be taken into account by applying , e.g. , the scheme used in ref .
@xcite for correcting the rate of spontaneous decay .
it is worth noting that the above given result , which is based on the quantization of the macroscopic maxwell field for given complex permittivity , exactly corresponds to the result obtained in ref .
@xcite within the framework of a fully microscopic approach on the basis of some model medium coupled to the radiation field and a heat bath .
already from the study of the spontaneous decay of an excited atom near an interface @xcite it is clear that in the case of inhomogeneous media ( of complicated atomic structure ) a microscopic approach would be rather involved and closed solutions would hardly be found .
let us consider a planar multi - slab structure and assume that the two molecules are in the same slab .
the relevant green tensor ( for the energy transfer between the two molecules relevant ) of an inhomogeneous system of this type can always be written in the form of @xmath118 where @xmath119 is the green tensor according to eq .
( [ n051 ] ) , with @xmath108 being the permittivity of the slab in which the two molecules are located , and the reflection term @xmath120 insures the correct boundary conditions at the surfaces of discontinuity .
clearly , a decomposition of the type of eq .
( [ n059 ] ) is also valid for other than planar systems , provided that the two molecules are located in a region of space - independent permittivity . to be more specific ,
let the @xmath121-direction be the direction of variation of the permittivity of the multi - slab system and assume that @xmath56 and @xmath57 are in the @xmath122th slab of thickness @xmath123 ( fig .
[ mlayer ] ) . the reflection term in eq .
( [ n059 ] ) can then be given by @xcite ( see also ref .
@xcite ) @xmath124&&\hspace{4ex } = { i \over 4\pi } \int_0^\infty { { \rm d } k_\| k_\| \over 2\beta_j } \,e^{i\beta_j d_j } \tilde{\bbox{g}}{^{\rm refl}}({\bf r}_b,{\bf r}_a,\omega , k_\|)\end{aligned}\ ] ] [ @xmath125 @xmath32 @xmath126 ; @xmath127 @xmath32 @xmath128 . choosing the coordinate system such that @xmath129 @xmath32 @xmath130 , the nonvanishing components of @xmath131 read @xmath132 \nonumber\\[.5ex]&&\hspace{11ex } + \,c^s_+\left[j_0(k_\|r_x ) + \!(-)\ ,
j_2(k_\|r_x)\right ] , \\[.5ex ] \label{ml3 } & & \tilde{g}^{\rm refl}_{xz(zx ) } = -(+)\ , 2i \ , { \beta_jk_\|\over k_j^2 } \ , s^p_{+(- ) } j_1(k_\|r_x ) , \\[.5ex ] \label{ml3a } & & \tilde{g}^{\rm refl}_{zz } = 2 \ , { k_\|^2\over k_j^2 } \,c^p_+ j_0(k_\|r_x)\end{aligned}\ ] ] [ @xmath133 - bessel function ] , where @xmath134&&\hspace{10ex } + \,(-)\ , 2 r^q_+ r^q_- \cos(\beta_j r_z ) e^{i\beta_j d_j } \bigr ] d_q^{-1 } , \\[.5ex ] \label{ml5 } & & s^q_{+(- ) } = \bigl [ r^q_-e^{i\beta_j(z_a+z_b - d_j ) } - r^q_+e^{-i\beta_j(z_a+z_b - d_j ) } \nonumber\\[.5ex]&&\hspace{10ex } + \,(-)\ , 2i r^q_+ r^q_- \sin(\beta_j r_z ) e^{i\beta_j d_j } \bigr ] d_q^{-1 } , \\[.5ex ] \label{ml6 } & & d_q = 1-r^q_+ r^q_- e^{2i\beta_j d_j}. % , \qquad q = p , s\ , \end{aligned}\ ] ] here , @xmath135
@xmath32 @xmath136 means tm(te ) polarized waves , and @xmath137 are the total reflection coefficients at the upper ( lower ) stack of slabs [ ] of the waves in the @xmath122th slab ( for details , see ref .
@xcite ) . note that when @xmath56 and @xmath57 are in the top ( bottom ) slab , then eqs . ( [ ml1])([ml6 ] ) ( formally ) apply provided that and are set .
if the frequencies of the vibronic transitions that are involved in the energy transfer are sufficiently far from a medium resonance , so that material absorption ( in the @xmath122th slab ) may be disregarded , then the permittivity may be considered as being real and positive . in this case
, it may be useful to decompose the integral in eq .
( [ ml1 ] ) into two parts , @xmath138 @xmath139&&\hspace{2ex } = { i \over 4\pi } \int_0^{\sqrt{\varepsilon_j}\omega / c } { { \rm d } k_\| k_\| \over 2\beta_j } \,e^{i|\beta_j| d_j } \tilde{\bbox{g}}^{\rm refl}({\bf r}_b,{\bf r}_a,\omega , k_\|),\end{aligned}\ ] ] @xmath140&&\hspace{2ex } = { i \over 4\pi } \int_{\sqrt{\varepsilon_j}\omega / c}^\infty { { \rm d } k_\| k_\| \over 2\beta_j } \,e^{-|\beta_j| d_j } \tilde{\bbox{g}}^{\rm refl}({\bf r}_b,{\bf r}_a,\omega , k_\|).\end{aligned}\ ] ] obviously , @xmath141 results from waves that have a propagating component in the @xmath121-direction , whereas the waves that contribute to @xmath142 are purely evanescent in the @xmath121-direction .
let the two molecules be embedded in a half - space medium ( medium @xmath143 ) and assume that in the relevant frequency interval the permittivity of the medium @xmath144 can be regarded as being real and positive . when the molecules are near the interface between the two half - space media such that , it can be proved that eqs .
( [ ml1])([ml3a ] ) reduce to ( ) @xmath145&&\hspace{4ex } \simeq \,{1 \over 4\pi k_1 ^ 2}\ , { \varepsilon_2\!-\!\varepsilon_1
\over \varepsilon_2 \!+\!\varepsilon_1}\ , { ( z_a\!+\!z_b)^2 -\!(+)\,2r_x^2 \over [ ( z_a\!+\!z_b)^2+r_x^2]^{5/2}}\,,\end{aligned}\ ] ] @xmath146&&\hspace{4ex } \simeq\ , + ( -)\,{1 \over 4\pi k_1 ^ 2}\ , { \varepsilon_2\!-\!\varepsilon_1 \over \varepsilon_2 \!+\!\varepsilon_1}\ , { 3(z_a\!+\!z_b)r_x \over [ ( z_a\!+\!z_b)^2+r_x^2]^{5/2}}\,,\end{aligned}\ ] ] @xmath147&&\hspace{15ex}\times \left\ { { 2(z_a\!+\!z_b)^2-r_x^2 \over k_1 ^ 2[(z_a\!+\!z_b)^2+r_x^2]^2 } \!+ \!1 \right\}\end{aligned}\ ] ] [ @xmath148 , complex permittivity of medium @xmath149 .
note that for @xmath56 @xmath32 @xmath150 , eqs .
( [ int6])([int5 ] ) just give the green tensor whose imaginary part determines the influence of the interface on the rate of spontaneous decay of a single molecule @xcite .
( for some special cases , see also ref .
@xcite . ) under the assumptions made , the main contribution to @xmath151 comes from @xmath142 .
hence surface - guided waves ( including decaying waves ) play an important role and can noticeably influence the resonant energy transfer . in particular when mediu is a metal or a dielectric with , ( and typically @xmath152 @xmath153 @xmath154 ) , then a strong effect is observed for , which is nothing but the condition for best excitation of surface - guided waves @xcite . in the numerical calculation of @xmath101 [ eq .
( [ n047c ] ) ] , which contains the relevant information about the influence of the interface on the rate of energy transfer [ see eqs .
( [ n048])([n047e ] ) ] , we have assumed that the two molecules are situated in vacuum [ ] above a half - space medium of drude lorentz type and restricted our attention to a single - resonance medium , @xmath155 here , @xmath156 corresponds to the coupling constant , and @xmath157 and @xmath158 are respectively the medium oscillation frequency and the linewidth . recall that the drude
lorentz model covers both metallic ( ) and dielectric ( ) matter and features a band gap between @xmath157 and .
we have performed the calculations using the exact green tensor [ eqs .
( [ n059])([ml6 ] ) ] . comparing the results with those obtained by using the approximately valid green tensor [ eq . ( [ n059 ] ) together with eqs .
( [ int6])([int5 ] ) ] , we have found good agreement . the behavior of @xmath101 is illustrated in fig . [ into ] .
it is seen that outside the band gap where the modification of @xmath101 due to the presence of the interface is small even for small distances of the molecules from the interface . since in this frequency domain @xmath101
may be regarded as being slowly varying on a frequency scale defined by the vibrational frequencies of the molecules , eq .
( [ n047f ] ) applies .
thus , the energy transfer rate is simply proportional to @xmath104 . inside the band gap ,
however , the interface can significantly affect @xmath101 if , according to eqs .
( [ int6])([int5 ] ) , ( in fig . [ into ] ) , that is to say , if the energy transfer transition under consideration is tuned to a surface - guided wave .
note that a negative real part of the medium permittivity can easily be realized by metals .
careful inspection of the contributions @xmath159 and @xmath151 to @xmath160 reveals that the enhancement of @xmath101 results from @xmath151 , whereas the reduction reflects some destructive interference of @xmath159 and @xmath151 .
another interesting feature is that the reduction of @xmath101 can go hand in hand with an enhancement of the corresponding quantity @xmath161 [ eq . ( [ n067a ] ) ] for the donor decay rate @xmath162 [ eq .
( [ n067 ] ) ] ( see the inset in fig . [ into ] ) .
further , fig .
[ into ] reveals that with increasing material absorption ( i.e. , with increasing value of @xmath158 ) @xmath101 varies less rapidly inside the band - gap region , and enhancement and reduction are thus less pronounced . clearly , the strong influence on @xmath101 of the interface which is observed for small material absorption must not necessarily lead to a correspondingly strong change of the energy transfer rate , because of the integration in eq .
( [ n048 ] ) .
nevertheless , the results show the possibility of controlling the resonant energy transfer by surface - guided waves .
figure [ intr ] illustrates the dependence of @xmath101 on the intermolecular distance for the case when @xmath81 corresponds to a surface - guided wave frequency and a noticeable change of @xmath101 is observed ( in the figure ) .
it is seen that the @xmath163 dependence , which is typical of the frster transfer in free space , is observed for much shorter intermolecular distances .
the relative minima of @xmath101 below the free - space level , which are observed for somewhat larger intermolecular distances , again result from destructive interference between @xmath159 and @xmath151 . eventually , the large - distance reduction of @xmath101 below the free - space level results from material absorption . as already mentioned , the behavior of @xmath101 in fig .
[ intr ] is dominated by surface - guided waves that decay exponentially along the @xmath164-directions . with increasing material absorption the penetration depths decrease , so that on average @xmath101 becomes closer to the free - space level .
the possibility of controlling the ultrashort - range energy transfer by varying the distance of the molecule from the surface is illustrated in the inset . in fig .
[ intz ] the dependence of @xmath101 ( again for ) on the molecule surface distance is plotted , and the contributions to @xmath101 from ordinary waves having a propagating component in @xmath121-direction ( @xmath141 ) and surface - guided waves ( @xmath142 ) are shown .
it is clearly seen that when the two molecules are very near the surface , then energy transfer between them is mediated by surface - guided waves , whereas for larger distances ordinary waves play the dominant role .
note that the oscillatory behavior is typical of the latter case .
clearly , for very large distances ( @xmath165 @xmath166 @xmath167 ) the free - space behavior is observed .
recently , experiments have been carried out to study the transfer of excitation energy between dye molecules confined within planar optical microcavities @xcite . in the experiments , donors ( eu@xmath168 complex ) and acceptors ( 1,1-dioctadecyl-3,3,3,3-tetramethylindodicarbocyanine ) embedded within a transparent material ( 22-tricosenoic acid ) bounded by no ( weak - cavity structure ) , one ( half - cavity structure ) , or two ( full - cavity structure ) silver mirrors
are considered . to compare the experimental results with the theoretical ones
, we have modeled the half - cavity structure by a planar four - layered system and the full - cavity structure by a five - layered system .
the former consists of vacuum , dielectric matter ( 22-tricosenoic acid , @xmath169 @xcite , thickness @xmath170 ) , metal ( silver , @xcite , thickness @xmath171 nm ) , and vacuum , and the latter consists of vacuum , metal ( silver , thickness @xmath172 nm ) , dielectric matter ( the same as above , thickness @xmath170 ) , metal ( silver , thickness @xmath171 nm ) , and vacuum . in each system , the donor is situated in the middle of the dielectric layer , while the position of the acceptor is shifted towards the silver mirror of @xmath171 nm thickness .
the green tensors of the two systems can be calculated according to eqs ( [ n059])([ml6 ] ) .
assigning to silver a drude lorentz - type permittivity @xcite , it can be proven that in the relevant frequency interval ( of overlapping donor emission and acceptor absorption spectra ) @xmath101 [ eq . ( [ n047c ] ) ] and @xmath161 [ eq . ( [ n067a ] ) ]
sufficiently slowly vary with @xmath81 , so that [ cf .
( [ n047f ] ) ] and , similarly , .
thus , @xmath104 and @xmath173 can be viewed as measures of the energy transfer rate and the donor decay rate , respectively .
figure [ barnes1 ] shows the dependence on @xmath170 of @xmath173 and @xmath104 ( averaged over the dipole orientations ) . from fig .
[ barnes1](a ) it is seen that at ( i.e. , for ) the ratio of the donor decay rates for the five- and four - layered systems is , which ( within the measurement accuracy ) is in sufficiently good agreement with experimental result ( see fig .
2d in ref .
note that in the vicinity of the ratio of the two rates sensitively responds to a change of @xmath174 .
comparing @xmath173 [ fig .
[ barnes1](a ) ] with @xmath104 [ fig .
[ barnes1](b ) ] , we see that for the four - layered system and @xmath174 @xmath175 @xmath176 ( i.e. , for ) both @xmath173 and @xmath104 decrease with increasing @xmath170 and an approximately valid linear relation between the energy transfer rate and the donor decay rate can be established in agreement with experimental results in ref . @xcite . from the data reported in ref .
@xcite it could be expected that the linear relation between the two rates is generally valid .
this is of course not the case . since the energy transfer rate is determined by the full ( two - point )
green tensor , whereas the donor decay rate is only determined by the imaginary part of the ( one - point ) green tensor , the two rates can behave quite differently , as it is demonstrated in fig .
[ barnes1 ] .
in particular , the increase of the donor decay rate at the cavity resonances can be accompanied with a decrease of the energy transfer rate , because of destructive interferences . in the experiments in ref .
@xcite , the measurements are performed on an ensemble of donors and acceptors whose distance is fixed in the @xmath121-direction but variable in the @xmath177-direction .
the question thus arises of whether the measured data refer to a single nearest - neighboring donor
acceptor pair or not . in fig .
[ barnes2 ] we have plotted the dependence on @xmath170 of @xmath104 ( averaged over the dipole orientations ) for the five - layered system and various values of @xmath178 , with @xmath179 being fixed . we see that the rates of energy transfer between molecules whose distances are larger than that of nearest - neighboring molecules can be quite comparable with those of the latter
. moreover there are also cases where the energy transfer rate increases with the donor acceptor distance .
the experimentally determined energy transfer rates are thus averaged rates , which not necessarily show the characteristic features of single - pair transfer rates . averaging in fig .
[ barnes2 ] @xmath104 over all values of @xmath178 , the resulting curve is expected to be substantially flatter than the solid - line curve , particularly when @xmath170 sweeps through @xmath3 .
an analysis of the contributions of @xmath141 [ eq . ( [ ml8 ] ) ] and @xmath142 [ eq .
( [ ml9 ] ) ] to @xmath151 [ eq . ( [ ml7 ] ) ] reveals that for cavity lengths of ( i.e. , for ) evanescent waves dominate the influence of the cavity system on both the rate of intermolecular energy transfer and the donor decay rate and lead to a strong increase of them .
whereas for cavities lengths of ( i.e. , for ) evanescent waves only weakly affect the donor decay rate , they can strongly affect the intermolecular energy transfer up to cavity lengths of a few micrometers .
note that the resonance lengths seen in fig .
[ barnes1 ] originate from propagating waves .
microspheres have been of increasing interest , because of the whispering - gallery ( wg ) and surface - guided ( sg ) waves , which may be employed , e.g. , for reducing the thresholds of nonlinear optical processes @xcite .
intermolecular energy transfer in the presence of microspheres has been considered for molecules near a small metallic spheroid ( spheroid s linear extension @xmath180 ) in the nonretardation limit , for molecules embedded within a dielectric microsphere @xcite , and for the case where one molecule is inside a dielectric microsphere and the other outside it @xcite . here
we restrict our attention to the influence of wg and sg waves on the energy transfer between two molecules outside a microsphere , taking fully into account retardation effects .
let @xmath144 and @xmath148 be respectively the permittivities outside and inside the sphere . if the transition dipole moments are parallel to each other and tangentially oriented with respect to the sphere , the relevant ( spherical - coordinate ) components of @xmath151 are ( , ) @xmath181&&\hspace{2ex } = { ik_1 \over 4\pi } \sum_{l=1}^\infty { ( 2l+1)\over l(l+1 ) } \biggl\ { { \cal b}_l^m h_l^{(1)}(k_1r_a ) h_l^{(1)}(k_1r_b ) \nonumber\\[.5ex]&&\hspace{4ex}\times\ , \bigl [ l(l+1)p_l(\cos\theta_b ) - \cos\theta_b p'_l(\cos\theta_b ) \bigr ] \nonumber\\[.5ex]&&\hspace{8ex } + \ , { \cal b}_l^n { [ k_1r_a\,h_l^{(1)}(k_1r_a)]'\over k_1r_a } \nonumber\\[.5ex]&&\hspace{10ex}\times\ , { [ k_1r_b\,h_l^{(1)}(k_1r_b)]'\over k_1r_b } p_l'(\cos\theta_b ) \biggr\ } \end{aligned}\ ] ] ( for the green tensor of a sphere , see , e.g. , @xcite ) , and for radially oriented dipoles the relevant components are ( , ) @xmath182&&\hspace{4ex}\times\ , { \cal b}_l^n h_l^{(1)}(k_1r_a ) h_l^{(1)}(k_1r_b ) p_l'(\cos\theta_b)\,,\end{aligned}\ ] ] where @xmath183&&\hspace{4ex } = - \frac { \bigl [ a_2j_l(a_2)\bigr ] ' j_l(a_1 ) - \bigl [ a_1j_l(a_1)\bigr ] ' j_l(a_2 ) } { \bigl [ a_2j_l(a_2)\bigr ] ' h_l^{(1)}(a_1 ) - j_l(a_2 ) \bigl[a_1 h_l^{(1)}(a_1)\bigr ] ' } \,,\end{aligned}\ ] ] @xmath184&&\hspace{0ex } = - \frac { \varepsilon_1(\omega ) j_l(a_2 ) \bigl[a_1 j_l(a_1)\bigr ] ' - \varepsilon_2(\omega)j_l(a_1 ) \bigl[a_2 j_l(a_2)\bigr ] ' } { \varepsilon_1(\omega ) j_l(a_2 ) \bigl [ a_1 h_l^{(1)}(a_1)\bigr ] ' -\varepsilon_2(\omega ) \bigl[a_2 j_l(a_2)\bigr ] ' h_l^{(1)}(a_1 ) } \nonumber\\ & & \end{aligned}\ ] ] [ ; @xmath185 , microsphere radius ; @xmath186 , spherical bessel function ; @xmath187 , spherical hankel function ; @xmath188 , associated legendre function ] . in fig .
[ sph ] the dependence on frequency of @xmath101 is illustrated for the case where vacuum is outside the sphere and the two molecules are placed at diametrically opposite positions , with the transition dipole moments being radially oriented .
it is clearly seen that the energy transfer can greatly be facilitated at the positions of the sphere - assisted field resonances , the enhancement of @xmath101 at the positions of sg resonances ( inside the band gap ) being larger than those at the positions of wg resonances ( outside the band gap ) .
maximum values of @xmath101 are observed where the sg resonances overlap .
the energy transfer rate for tangentially oriented dipoles ( not shown ) is in general smaller than that for radially oriented dipoles .
note that when @xmath101 is sharply peaked at the sphere - assisted field resonances , such that it is not slowly varying in the frequency interval where the ( free - space ) donor emission and acceptor absorption spectra overlap , then it can not be taken at the electronic energy transfer frequency and put in front of the integral in eq .
( [ n048 ] ) . in this case
, the change of the energy transfer rate will be less pronounced than it might be expected from the frequency response of the electronic part , because of the frequency integration .
we have given a rigorous , strictly quantum mechanical derivation of the rate of intermolecular energy transfer in the presence of dispersing and absorbing material bodies of arbitrary shapes , showing that both the minimal - coupling scheme and the multipolar coupling scheme lead to rate formulas of exactly the same form .
the dependence on the material bodies of the energy transfer rate is fully expressed in terms of the green tensor of the macroscopic maxwell equations for the medium - assisted electromagnetic field . in the macroscopic approach ,
the dispersing and absorbing material bodies are described , from the very beginning , in terms of a spatially varying permittivity , which is a complex function of frequency .
the macroscopic approach has
similar to classical optics the benefit of being universally valid , without the need of involved _ ab initio _ microscopic calculations .
in so far as such calculations for simple model systems have been performed , the results agree with those obtained from the microscopic approach .
clearly , macroscopic electrodynamics is valid only to some approximately fixed length scale which exceeds the average interatomic distance in the material bodies . whereas the donor spontaneous decay rate is determined by the imaginary part of the green tensor in the coincidence limit , the donor acceptor energy transfer rate depends on the full two - point green tensor .
hence , the decay rate and the energy transfer rate can be affected by the presence of material bodies quite differently .
our calculations for planar multilayer structures have shown that enhancement ( inhibition ) of spontaneous decay and inhibition ( enhancement ) of energy transfer can appear simultaneously .
they have further shown that surface - guided waves can strongly affect the energy transfer , thus being very suitable for controlling it . in free space
it is often distinguished between two limiting cases , namely the short - distance nonradiative ( frster ) energy transfer and the long - distance radiative energy transfer .
the former is characterized by the @xmath0 distance dependence of the transfer rate , and the latter by the @xmath1 dependence .
in particular , in the short - distance limit the energy transfer rate rapidly decreases with increasing distance between the molecules .
this must not necessarily be the case in the presence of material bodies , because of the possibly drastic change of the dependence on the distance of the energy transfer rate .
so , our calculations for planar multilayer structures have shown that the energy transfer rate can also increase with the distance .
we thank s. scheel and a. tip for discussions .
is grateful to the alexander von humboldt stiftung for financial support .
this work was supported by the deutsche forschungsgemeinschaft .
the multipolar - coupling hamiltonian can be obtained from the minimal - coupling hamiltonian by means of the power zienau transformation @xcite , @xmath189 where @xmath190 \,,\end{aligned}\ ] ] with @xmath191&&\hspace{10ex } \times \int_0 ^ 1 { \rm d}\lambda \ , \delta\!\left[{\bf r}\!-\!{\bf r}_m\!-\!\lambda \left(\hat{\bf r}_{\alpha_m } - { \bf r}_m\right)\right]\end{aligned}\ ] ] being the polarization associated with the @xmath43th molecule . using @xmath192 from eq . ( [ n001 ] )
, we derive ( see , for details , @xcite ) @xmath193&&\hspace{2ex } + \sum_m \sum_{\alpha_m}\frac{1}{2m_{\alpha_m } } \bigg\ { \hat{{\bf p}}_{\alpha_m } \nonumber\\[.5ex]&&\hspace{3ex } + q_{\alpha_m}\int_0 ^ 1\!{\rm d}\lambda\,\lambda \left ( \hat{\bf r}_{\alpha_m}\!-\!{\bf r}_m \right ) \times \hat{\bf b}\left [ { \bf r}_m\!+\!\lambda \left ( \hat{\bf r}_{\alpha_m } \!-\!{\bf r}_m \right ) \right ] \bigg\}^2 \nonumber\\[.5ex]&&\hspace{2ex } + \,\sum_{m } \int\ ! { \rm d}^3{\bf r } \ , \left[\frac{1}{2\varepsilon_0 } \hat{\bf p}_m({\bf r } ) \hat{\bf p}_{m}({\bf r } ) \right ] \nonumber\\[.5ex]&&\hspace{2ex } -\,\sum_m \int\ ! { \rm d}^3{\bf r}\ , \left[\hat{{\bf p}}_m({\bf r } )
\hat{\bf e}({\bf r } ) \right]\ , , \end{aligned}\ ] ] where [ with @xmath20 from eq .
( [ n010 ] ) ] , and @xmath194 and neutral molecules with non - overlapping charge distributions are again assumed .
note that in the multipolar - coupling scheme the operator of the electric field strength is defined according to @xmath195 - \bbox{\nabla } \hat{\varphi}({\bf r } ) - \bbox{\nabla } \hat{\phi}({\bf r})\ , , \end{aligned}\ ] ] which implies the following relation between @xmath196 and @xmath197 : @xmath198 hence , @xmath199 has the meaning of the displacement field with respect to the molecular polarization . from eq .
( [ n062 ] ) it is seen that the molecules now interact only via the medium - assisted electromagnetic field . in particular , in the ( electric-)dipole approximation eq .
( [ n062 ] ) simplifies to @xmath200 where @xmath201 with @xmath202 is the unperturbed hamiltonian of the medium - assisted electromagnetic field and the molecules , and @xmath203 is the interaction energy between them . comparing the multipolar - coupling energy given by eq .
( [ n066 ] ) with the minimal - coupling energy @xmath204 given by eq .
( [ n015 ] ) together with eqs .
( [ n017a])([n020 ] ) , we see that the two energies ( formally ) become equal to each other , if we remove in the latter the coulomb term and replace @xmath205/\hbar\omega$ ] with @xmath206 . having these changes in mind , we now follow step by step the derivation of eq .
( [ n047 ] ) in section [ two molecules ] . starting from the corresponding eigenstates of the unperturbed multipolar - coupling hamiltonian ( instead of those of the unperturbed minimal - coupling hamiltonian ) , it is not difficult to see that the result is again eq .
( [ n047 ] ) .
it should be pointed out that the above mentioned difference between @xmath196 and @xmath197 [ eq . ( [ n062b ] ) ] does not affect the energy transfer rate .
in the electric - dipole approximation and the rotating - wave approximation , the hamiltonian for a single molecule ( at position @xmath56 ) that ( with regard to the vibronic transitions @xmath207
@xmath208 @xmath209 ) resonantly interacts with the medium - assisted electromagnetic field reads , by appropriately specifying eqs .
( [ n014a])([n020 ] ) , @xcite @xmath210 @xmath211&&\hspace{4ex } + \sum_a \hbar\omega_a |a\rangle\langle a| + \sum_{a ' } \hbar\omega_{a ' } |a'\rangle\langle a'|,\end{aligned}\ ] ] @xmath212,\ ] ] where @xmath213 is the positive - frequency part of @xmath196 defined by eq .
( [ n006 ] ) , and the vibronic transition - dipole matrix elements @xmath214 of the vibronic transitions are given , in the born
oppenheimer approximation , by eq .
( [ n034 ] ) .
let us assume that the molecule is initially ( at time @xmath215 @xmath32 @xmath130 ) prepared in a statistical mixture of vibrational states in the upper electronic state and the medium - assisted electromagnetic field is in the vacuum state , i.e. , @xmath216 the time - dependent spectrum of light observed at position @xmath217 ( in free space ) by means of a spectral apparatus of sufficiently small passband width can be given by ( see , e.g. , @xcite ) @xmath218 , \end{aligned}\ ] ] where @xmath219 and @xmath220 are respectively the setting frequency and the operating time of the spectral apparatus . in order to calculate the electric - field correlation function associated with the light emitted by the molecule during the spontaneous decay of the upper electronic state , we may restrict our attention to the perturbative expansion of the time evolution operator up to the first order in @xmath204 @xcite , @xmath221 we make use of eqs .
( [ b1b ] ) , ( [ n006 ] ) [ together with eq .
( [ n007 ] ) ] , ( [ b6 ] ) , and ( [ b7 ] ) , apply eq .
( [ b5 ] ) , and derive after some calculation , on recalling the relation ( [ n030 ] ) , ( see also @xcite ) @xmath222&&\hspace{2ex } = 2\pi \sum_{a , a ' } p_{a ' } |v_{a'a}|^2 |{\bf f}({\bf
r},{\bf r}_a , \omega_{a'a})|^2 \delta(\omega_s - \omega_{a'a } ) , \end{aligned}\ ] ] where @xmath223&&\hspace{2ex } = { 1 \over \pi\varepsilon_0 } \int_0^\infty { \rm
d}\omega \ , { \omega^2 \over c^2}\ , { \rm im}\ , \bbox{g } ( { \bf r},{\bf r}_a,\omega ) { } { \bf d}_a \zeta(\omega_{a'a}\!-\!\omega ) \nonumber\\[.5ex]&&\hspace{15ex } \simeq\ , - { i \omega_{a'a}^2 \over \varepsilon_0 c^2}\ , \bbox{g}({\bf r},{\bf r}_a,\omega_{a'a}){\bf d}_a \end{aligned}\ ] ] [ @xmath224 @xmath32 @xmath225 @xmath34 @xmath226 ; @xmath227 , principal value ] . in the derivation of eq .
( [ b8 ] ) , retardation has been ignored and the relation @xmath228&&\hspace{4ex } = \lim _ { t\rightarrow\infty } { \sin^2(\omega t/2 ) \over t(\omega /2)^2 } = 2\pi\delta(\omega ) \end{aligned}\ ] ] has been used .
j. s. avery , proc .
london * 88 * , 1 ( 1966 ) ; l. gomberoff and e. a. power , _ ibid . _
* 88 * , 281 ( 1966 ) ; e. a. power and t. thirunamachandran , phys .
a * 28 * , 2671 ( 1983 ) ; d. l. andrews and b. s. sherborne , j. chem . phys . * 86 * , 4011 ( 1987 ) ; d. l. andrews , chem . phys . * 135 * , 195 ( 1989 ) ; d. l. andrews and g. juzelinas , j. chem . phys . * 96 * , 6606 ( 1992 ) ; d. p. craig and t. thirunamachandran , chem . phys . * 167 * , 229 ( 1992 ) .
ho trung dung , l. knll , and d .-
welsch , phys .
a * 57 * , 3931 ( 1998 ) ; s. scheel , l. knll , and d .-
welsch , _ ibid .
_ * 58 * , 700 ( 1998 ) ; a. tip , l. knll , s. scheel , and d .-
welsch , _ ibid .
_ * 63 * , 043806 ( 2001 ) . | by making use of the green function concept of quantization of the electromagnetic field in kramers
kronig consistent media , a rigorous quantum mechanical derivation of the rate of intermolecular energy transfer in the presence of arbitrarily shaped , dispersing , and absorbing material bodies is given .
applications to bulk material , multi - slab planar structures , and microspheres are studied .
it is shown that when the two molecules are near a planar interface , then surface - guided waves can strongly affect the energy transfer and essentially modify both the ( frster ) short - range @xmath0 dependence of the transfer rate and the long - range @xmath1 dependence , which are typically observed in free space .
in particular , enhancement ( inhibition ) of energy transfer can be accompanied by inhibition ( enhancement ) of donor decay .
results for four- and five - layered planar structures are given and compared with experimental results . finally , the energy transfer between two molecules located at diametrically opposite positions outside a microsphere is briefly discussed . |
in 1903 , first heusler alloys ( ha ) have been singled out as a special group of magnetic materials that exhibit ferromagnetism in compounds of non - magnetic elements @xcite . currently , the mn - based half - heusler alloys with the generic formula @xmath0mn@xmath1 ( with @xmath0 being a @xmath2 metal and @xmath1 belonging to the iii or iv group ) are well known magnetic systems that crystallizes in the heusler @xmath3 structure , which is closely related to the zincblende structure . especially , the half - metallic ferromagnet nimnsb @xcite has been triggered a variety of experimental and theoretical investigations due to the search for materials to be used as spin - injectors .
these materials are defined as magnetic materials with a band gap at the fermi level for electrons of one spin direction .
band structure calculations for bulk nimnsb show a gap of about 0.5 ev at t=0 for the minority electrons , which means 100% spin polarization at the fermi level @xmath4 . in fully relativistic calculations a reduced but not vanishing gap and 99% spin polarization at @xmath4 are predicted @xcite . as in the case of the group iii - v semiconductors ,
the crystal structure is the reason for the band gap .
it is , therefore , called a covalent band gap @xcite . as a consequence , the crystal structure and
the site occupation within the given structure are important for the appearance of the gap .
atomic disorder , especially interchange of atoms between the ni and mn sub - lattices , results in a strong reduction of the spin polarization at @xmath4 @xcite .
however , this interchange costs approximately as much energy as the evaporation of the metallic constituents @xcite .
therefore , half - metallic behavior is an idealization distorted by structural inhomogeneities @xcite and through the decrease of the spin polarization of spin carriers and vanishing of the semiconducting band gap for minority spin electrons in real crystals and interfaces caused by finite temperature effects .
a variety of experimental results for nimnsb is available in the literature .
while bulk measurements support the half - metallic behavior ( see , e.g. , @xcite ) , surface - sensitive techniques have failed so far to detect the energy gap for minority electrons or the 100% spin polarization at the fermi level @xmath4 .
most of the studies were performed on polycrystals or thin films , single and polycrystalline , prepared under various growth conditions on a variety of substrates .
spin - integrated photoemission ( pe ) spectra obtained with a photon energy of 45 ev showed clear signatures of the ni and mn @xmath2 contributions @xcite .
angle - integrated pe with variable photon energies identified ni and mn @xmath2 emission from the valence bands emphasized the need for further careful investigations both experimentally and theoretically @xcite .
a spin - polarized pe study found values of the spin polarization up to 50% near photothreshold @xcite .
the authors concluded that , if an energy gap for minority electrons should exist , it is smaller than 0.5 ev . a further spin - resolved pe work , yet with photon energies between 38 and 76 ev , reported spin polarization values of at most 40% close to the fermi level @xcite .
a reduced surface magnetization at remanence and/or a surface phase different from the bulk were proposed as possible explanations for the unexpected low polarization values .
furthermore , an angle - resolved photoemission study of the electronic structure single - crystalline nimnsb sample showed that the binding energies of the photoemission intensities are sensitive to the particular surface condition and/or preparation @xcite .
additionally , a possible surface state has been found in these @xmath5-resolved measurements . in another study , a temperature - dependent cross - over from half - metallic to normal ferromagnetic behavior at 80 k
was reported and discussed as a possible reason for the detected low polarization @xcite .
it was suggested that , at room temperature , the spin polarization at @xmath4 may be considerably lowered due to a populated minority band , although the magnetization is not much reduced compared to the value at t=0 .
( with a curie temperature of about 730 k , the magnetization at room temperature amounts to @xmath6 92% of the saturation value @xcite . )
however , even point contact andreev reflection measurements at 4.2 k at the free surface of nimnsb gave a maximal value of only 44% , independent of different surface preparations and magnetic domain structures @xcite .
this result is in line with former spin - polarized pe data , which did not find a higher polarization value for 20 k than for room temperature @xcite . as a complement to the reports above , a room - temperature study by angle and spin - resolved inverse photoemission reported close to 100% spin polarization at @xmath4 and @xmath7 under some conditions @xcite .
last but not least from a spin - resolved appearance potential spectroscopy ( aps ) study the measured spin asymmetry from the surface region was found to be significantly reduced compared to the theoretical prediction @xcite . in this contribution , we present spin - integrated and spin - resolved aps spectra from nimnsb(001 ) and discuss by comparison with experimental data the degree of spin polarization as a function of chemical disorder .
the paper is organized as follows .
section 2 is devoted to some experimental and computational details . in section 3
we discuss the theoretical results and compare with corresponding experimental data .
a summary is given in section 4 .
in this section the relevant details of the experimental investigations are given .
a more delayed report can be found in ref .
@xcite .
aps is a surface sensitive tool to study the unoccupied density of states ( dos ) with elemental resolution @xcite . for spin - resolved aps ,
the sample is bombarded with a spin - polarized electron beam of variable energy while the total yield of emitted x - rays or electrons is monitored @xcite . at energies high enough to excite a core electron into empty states above the fermi level
, the yield of emitted particles increases due to recombination of the created core hole via x - ray or auger electron emission . in our case
, we detect the emitted x - rays @xcite .
potential modulation together with lock - in techniques are used to separate the small signal from the dominating background . since both the exciting and the excited electron
are scattered into empty states , the rate of possible excitations and , thereby , of detected recombinations depends on the dos above @xmath4 . the spin - polarized electron beam used for excitation
is emitted from a gaas photocathode irradiated with circularly polarized 830 nm laser light .
this arrangement provides about 30% spin polarization of the emitted electrons @xcite .
the spectra shown have been renormalized to 100% hypothetical beam polarization .
the aps signal depends on the spin of the exciting electron in the case of ferromagnets because of the spin - dependent unoccupied dos .
the elemental resolution comes from the fact that core levels are involved whose energies are characteristic of the various elements . hence , the calculational procedure concerning aps is found in detail in @xcite we restrict ourselves on a short description of some computational details .
the fully relativistic dos calculations have been performed within the sprkkr computer program @xcite running in the cpa - mode @xcite .
the chemical disorder introduced in the spectroscopical analysis consists of an interchange between mn atoms on ni sites and vice versa with concentration values x of 0@xmath8 , 5@xmath8 , 10@xmath8 and 15@xmath8 .
this means we have calculated ( ni@xmath9mn@xmath10)(mn@xmath9ni@xmath10)sb in the l2@xmath11 crystal structure using the experimental lattice constant 5.91 @xmath12 . in figs .
2 and 3 we use the compact notation ni@xmath9mn@xmath10sb .
figure 1 shows the spin - resolved total dos for nimnsb .
clearly visible is the closing of the spin - dependent gap at the fermi energy caused by the chemical disorder . even at 5@xmath8 interchange of mn and ni atoms the gap completely vanishes because the spectral weight is shifted towards the fermi level .
a significant enhancement of the spectral weight around the fermi level is also observable in the majority dos and therefore affect the amount of spin polarization that is measured in an aps experiment .
the corresponding partial densities of states together with the spin - dependent potentials serve as input quantities for the spectroscopical calculations . in our case
we probed the local dos ( ldos ) at the mn atom in nimnsb .
the surface sensitivity is a result of the short mean free path of the electrons in the solid for energies that correspond to the mn@xmath13 excitation ( @xmath6 640 - 650 ev ) .
the elemental resolution results from the fact that a core level with an element - specific binding energy is involved . due to the excitation with spin - polarized electrons
the aps method becomes a magnetically sensitive technique .
the calculated spin - dependent aps - intensities result from the weighted self - convolution of the corresponding ldos .
the weighting factor itself is given by the spin - dependent effective cross section consisting of an appropriate combination of relativistic coulomb matrix elements .
for a quantitative comparison between aps measurements and theoretical spectra one has to take into account various lifetime effects and the experimental broadening .
electron lifetime effects have been included in our analysis in a phenomenological way introducing a parametrized complex inner potential with an imaginary part of v@xmath14 = 30 ev .
the core hole lifetime is set to zero .
lifetime effects in the valence - band states are accounted via a lorentzian with an energy - dependent width . for the explicit parametrization of the lorentzian the reader
is referred again to @xcite .
strong correlations manifest themselves in form of the simple scaling factor @xmath15 = 0.7(e - e@xmath16 ) . furthermore , due to experimental conditions
the first derivative of the aps signal has to be calculated and the apparatus broadening is considered by convoluting the calculated spectra by a gaussian of fwhm = 1.4 ev .
threshold as a function of chemical disorder .
experimental data are given by filled circles , height=453 ]
figure 2 presents calculated spin - integrated ap spectra for nimnsb(001 ) in comparison with aps data .
the spectra are normalized to equal maximum intensity .
the theoretical spectra have been calculated with respect to the fermi level . to be comparable with the experiment the calculated spectra were shifted along the energy axis until the peak maxima of the measured and calculated spectra coincidence . as it can be seen good agreement between experiment and theory
is achieved .
the almost quantitative agreement suggests that the dos as calculated resembles the mn local dos in nimnsb quite well in the spin - integrated case .
the spectra show a pronounced structure at 639.7 ev from the mn 2@xmath17 threshold .
these feature originate from the self - convoluted local density of unoccupied @xmath2 states .
the spectral feature at the high - energy side of the main line corresponds to the maximum of the @xmath18-like dos , whose appearance and energetic position is sensitive to the short - range crystallographic order around the atom where the local excitation occurs . as expected , the effect of chemical disorder is nearly negligible concerning the spin - integrated dos .
only slight deviations in the aps line shapes between the spectra calculated for the ordered and for the disordered structures are observable .
threshold as a function of chemical disorder .
experimental data are given by filled circles , height=453 ] by adding spin resolution ( see spectra in fig .
3 ) , we are able to obtain information about the spin - dependent dos .
the experimental data for majority ( @xmath19 ) and minority ( @xmath20 ) spins are presented by black and grey open circles , respectively .
first of all , the aps lines show a clear spin asymmetry , @xmath21 , between the spin - dependent intensities @xmath22 and @xmath23 [ @xmath24 .
the experimental spin - resolved spectra of fig .
3 represent an average from many different surface regions . by taking into account the reduced remanent magnetization @xcite
, we end up with an estimated spin asymmetry of @xmath25 for a nimnsb sample with saturated magnetization .
the negative asymmetry reflects the high density of unoccupied minority states in nimnsb , i.e. the unoccupied minority mn @xmath2 states .
the theoretical spectra calculated for the ordered structure , however , predict a higher negative spin asymmetry in the main aps peak of @xmath26 .
this discrepancy between experiment and theory by a factor of two and more is in line with the findings of earlier spin - resolved electron spectroscopic experiments . introducing chemical disorder the calculated value for the spin asymmetry decreases .
this has to expected because of the pronounced increase of spectral weight in the majority spin dos around the fermi energy that overcompensates the shift of spectral weight in the minority spin dos .
the concentration dependence of the spin asymmetry is demonstrated by the series of spin - dependent aps data shown in fig .
the spin asymmetry in the main peak is calculated to @xmath27 for 5@xmath8 interchange of ni and mn atoms and decreases from @xmath28 for 10@xmath8 disorder to @xmath29 for 15@xmath8 interchange of ni and mn atoms . as a guide for the eye we compare at the bottom of fig .
3 directly the spectra which belong to zero and 15@xmath8 disorder to clarify the decrease in the spin asymmetry .
therefore , our analysis supports the result found by orgassa et al . that chemical disorder significantly lowers the spin polarization at @xmath4 . on the other hand , introducing the effect of chemical disorder to our aps calculations , we end up with an improved agreement between experiment and theory but the predicted spin asymmetry is still too high by nearly a factor of two .
our analysis indicates that other mechanisms besides chemical disorder must be taken into account for a quantitative explanation of the strongly reduced spin polarization in nimnsb(001 ) .
we have presented spin - integrated and spin - resolved aps data for the unoccupied electronic states of nimnsb as a function of chemical disorder .
the calculated series of spin - resolved and concentration dependent aps spectra demonstrates that chemical disorder influences the spin asymmetry of the unoccupied states of nimnsb . concerning the spin polarization at the fermi level
, one must note that aps is not particularly sensitive in that region .
the nature of the aps signal does nt allow one to resolve small energy gaps of the order of 0.5 ev giving rise to a positive spin asymmetry at @xmath4 .
the aps signal is rather dominated by the minority @xmath30 states above @xmath4 leading to a negative spin asymmetry . on the other hand the significantly reduced spin polarization observed in the experiment clearly indicates a uncomplete spin polarization with a value well below 100% .
therefore , the expected 100% spin polarization at the fermi level is not verified , as in a number of spin - polarized electron spectroscopic experiments before .
furthermore , one has to take into account other mechanisms than chemical disorder to be able to explain quantitatively the unexpected low value for the measured spin asymmetry .
therefore , we conclude that the following issues must be examined in future studies on well - defined samples .
first the magnetization within the surface layers compared with the bulk should be investigated in more detail .
also surface phases with respect to composition and/or crystallographic order different from the bulk could be important and the position and width of the band gap should be analyzed in more detail in the calculations .
a first study on nimnsb(110 ) showed the formation of microstructures upon preparation , which has consequences for the magnetic properties of the surface @xcite .
even with a stoichiometric surface , the number of nearest neighbors is changed at the surface and surface / interface states may form and influence the gap @xcite . in the case that the free surface will not provide 100% spin polarization , there is still hope that specific interfaces may open the way for 100% spin - polarized charge - injection @xcite in spintronic devices .
other materials with half - metallic behavior must be considered as well in order to design functional materials for future applications @xcite .
99 donath m , rangelov g , braun j and grentz w 2005 magnetization , spin polarization and electronic structure of nimnsb surfaces _ local - moment ferromagnets : unique properties for modern applications ( lecture notes in physics vol 678 ) _ ed m donath and w nolting ( berlin : springer ) pp 261 - 272 ebert h 1999 fully relativistic band structure calculations for magnetic solids formalism and application _ electronic structure and physical properties of solids .
the use of the lmto method ( lecture notes in physics vol 635 ) _ ed h.dreyss ( berlin : springer ) pp 191 - 246 fonin m , dedkov yu s , rdiger u and gntherodt g 2005 growth and room temperature spin polarization of half - metallic epitaxial cro@xmath31 and fe@xmath32o@xmath33 thin films _ local - moment ferromagnets : unique properties for modern applications ( lecture notes in physics vol 678 ) _ ed m donath and w nolting ( berlin : springer ) pp 289 - 306 | the half - heusler alloy nimnsb is one of the local - moment ferromagnets with unique properties for future applications .
band structure calculations predict exclusively majority bands at the fermi level , thus indicating 100% spin polarization there . as one thinks about applications and the design of functional materials ,
the influence of chemical disorder in these materials must be considered .
the magnetization , spin polarization , and electronic structure are expected to be sensitive to structural and stoichiometric changes . in this contribution
, we report on an investigation of the spin - dependent electronic structure of nimnsb .
we studied the influence of chemical disorder on the unoccupied electronic density of states by use of the ab - initio coherent potential approximation method .
the theoretical analysis is discussed along with corresponding spin - resolved appearance potential spectroscopy measurements .
our theoretical approach describes the spectra as the fully - relativistic self - convolution of the matrix - element weighted , orbitally resolved density of states . |
dynamics on networks is a topic on which extensive works have been done in recent years .
phenomena which have been studied include evolution of spin systems , opinion dynamics , disease spreading dynamics , etc . @xcite .
the dynamical picture is quite different from that on regular lattices due to the topological features of network .
for example , one can define the dynamics in different ways for the voter model on a network while these rules become identical on lattices @xcite .
the study of ising model on networks has revealed a number of interesting features when static properties are considered . even in one dimension ,
when randomly new links are added ( or existing links rewired ) as in a watts strogatz ( ws ) network @xcite , one gets a phase transition @xcite which occurs with mean field criticality @xcite . on euclidean networks ,
indications of both mean field type and finite dimensional - like phase transitions have been shown to exist by varying the relevant parameter @xcite . on scale free networks
@xcite , the transition temperature shows a logarithmic increase with the system size which is perhaps the most surprising result @xcite . while considering ordering dynamics on regular lattices at zero temperature for the ising model using glauber dynamics , it is known that for any dimension greater than one , freezing occurs with a probability dependent on the dimension @xcite .
this happens when one considers a completely random initial condition . on random graphs and networks ,
one encounters similar freezing phenomena which depend on the density of added links @xcite . on random networks or graphs
, the evolution of the ising model from a completely random state shows that the system does not order . a freezing effect was observed and
although there could be an emergent majority of nodes with either spin up or spin down state , domains of nodes with opposing spins survive @xcite .
careful observations show that the disordered state is not an absorbing state @xcite .
it is instead a stationary active state , with some spins flipping , while keeping the energy constant .
the number of domains remaining in the system is just two .
the qualitative picture is then the same as on regular lattices for @xmath0 @xcite , the system wanders forever in an iso - energy set of states .
the distribution of the steady state residual energy ( which is identical to the number of bonds between oppositely oriented spins apart from a constant ) for the ising model on a random networks was investigated @xcite .
it was found that the distribution typically shows two peaks , one very close to the actual ground state where the residual energy is zero and one far away from it .
dynamics of the ising model on the ws network with restricted rewiring has also been considered .
here initially a spin is connected to its four nearest neighbours and then only the second nearest neighbour links are rewired with probability @xmath1 .
the system therefore always remains connected .
under the zero temperature glauber dynamics , freezing effect was observed for any @xmath2 @xcite . in this paper ,
we have considered in detail the variation of the relevant thermodynamic quantities as functions of time for the zero temperature dynamics of ising model on various networks .
our main emphasis is on the ising model on scale free networks at temperature @xmath3 as such studies have not been made so far to the best of our knowledge .
apart from the question whether the equilibrium state is reached or not , we have also explored the nature of the state in case it does not .
we are interested to see whether any active - absorbing phase transition occurs as the system parameters are varied .
both the random scale free network ( rsf ) and the barabsi - albert ( ba ) network have been considered for the study .
although many results are known for the ws network , we have explored specifically the possibility of an active - absorbing phase transition in this network .
the ba model is studied to make direct comparison with the random scale free network , where results can be quite different @xcite .
also , comparison with respect to issues like freezing and absorbing phase transition may be made for the rsf and the ws networks . in section
ii , we describe the network models and the dynamical evolution .
the quantities which have been calculated are defined in section iii .
the results are presented in the next section and in the last section we summarise and discuss the studies made .
we have considered three different types of network : ( a ) random scale free , ( b ) barabsi - albert and ( 3 ) watts strogatz ( addition type ) @xcite network .
we describe in brief how these networks are generated and the dynamical evolution process . in this section
we also include a brief discussion of how the numerical data had been analysed in previous studies . in the random scale free network
the degree distribution follows a power law but otherwise the network is random . to generate random scale free network @xcite , we assigned the degree of each node using the power law : @xmath4 where @xmath5 is the degree of node and @xmath6 is the characteristic degree exponent .
the minimum value of @xmath5 is @xmath7 and the maximum cut - off value is @xmath8 , where @xmath9 number of nodes .
this cut - off value ensures that there is no correlation @xcite .
starting from the node with the maximum degree , links have been established with randomly selected distinct nodes .
barabsi - albert network is a growing network where new nodes are joined to existing nodes with preferential attachment .
we start with the three fully connected nodes .
subsequently , a single node is added at a time to the network which is linked to one existing node .
the probability that the new node is connected to the existing @xmath10th node with degree @xmath11 is given by @xcite , @xmath12 degree distribution of the network is a power law with exponent @xmath13 ; @xmath14 @xcite .
addition type ws network is a one dimensional regular chain with two nearest neighbour links as well as with some extra randomly connected long range links . here
the long range links have been added with probability @xmath15 ( total long range links @xmath16 ) , where @xmath9 is the number of nodes and @xmath17 is a parameter , which denotes the number of extra long range links per node on an average .
so the average degree per node of this network is @xmath18 , which is a finite quantity as @xmath19 in the thermodynamic limit .
it is known @xcite that an order - disorder transition occurs at @xmath20 .
the hamiltonian of the ising system in these networks can be expressed as @xmath21 where @xmath22 and @xmath23 when sites @xmath10 and @xmath24 are connected and zero otherwise . starting with the random configuration , single spin flip energy minimizing glauber dynamics has been used to update the spin . in this dynamics a randomly selected spin is flipped if the energy of the updated configuration is lowered .
it is flipped with probability @xmath25 if the energy remains unchanged on flipping .
fifty different network configurations have been considered and for each network hundred different initial configurations have been taken .
the results are averaged over these configurations .
we have considered system sizes @xmath9 upto @xmath26 .
periodic boundary condition has been used for the ws model which is embedded in real space .
previous studies have shown that for the thermally driven phase transition , not only mean field critical behaviour exists for the ising model on small world networks but finite size scaling is also valid there @xcite .
the data collapse was obtained by rescaling the data and the scaling argument occurring in the scaling function was found to be of the familiar form @xmath27 ( with @xmath28 ) where @xmath29 denotes the deviation from the critical point .
@xmath30 is argued to be equal to @xmath31 where @xmath32 is the correlation length exponent and @xmath33 the effective dimension of the system . with mean field critical exponent @xmath34
, @xmath33 turns out to be equal to the upper critical dimension ; @xmath35 .
we have estimated the following quantities in the present work .
magnetisation : @xmath36 has been calculated by taking the average of the absolute values of the magnetisation , @xmath37 , as a function of time . since evolution to both up spin dominated and down spin dominated configurations are possible , the absolute value is taken to compute the configuration average .
distribution of saturation value of magnetisation has also been estimated for random scale free ( rsf ) network .
residual energy : @xmath38 where @xmath39 is the equilibrium energy of the ground state per spin and @xmath40 the energy per spin at time @xmath41 .
residual energy measurement indicates the closeness to the equilibrium ground state .
since we are employing a zero temperature quench , the equilibrium ground state configuration corresponds to either all spins up or down .
@xmath42 , where @xmath43 is the number of spin flips at time @xmath41 , has been studied as a function of time .
we count all the spin flips , i.e. , if the same spin flips more than one time , all these occurrences are taken into account .
freezing probability @xmath44 is defined as the probability that the system does not reach the true ground state .
it has been calculated as a function of the relevant parameters . for @xmath3 ,
the known ground state is the ferromagnetic state with @xmath45 .
a frozen configuration will have an absolute value of magnetisation less than @xmath7 . \5 . whether or not the dynamics continue
, the system always reaches an iso - energy state .
time @xmath46 to reach the iso - energy state has been calculated as a function of the relevant parameter for random scale free ( rsf ) network . the saturation values have been denoted using appropriate subscript , e.g. @xmath47 for magnetisation .
as a function of @xmath6 for different system size @xmath9 . inset shows the data collapse where @xmath48 has been plotted against @xmath49 . [ sat_frz_rsf ] , width=188 ] we first discuss the behaviour of the freezing probability @xmath44 on the rsf .
the variation of @xmath44 with the network parameter @xmath6 for different system sizes suggests that a freezing transition takes place here as the freezing probability is unity above a value of @xmath6 which we denote by @xmath50 ( fig .
[ sat_frz_rsf ] ) . of course
, if the initial state is fully ordered it is not counted as a frozen state . however , barring these two states ( all up/ all down ) , none of the other initial states reaches the actual equilibrium configuration .
hence the freezing probability is actually @xmath51 which becomes unity in the thermodynamic limit .
below @xmath50 , the freezing probability decreases with system size and shows a system size independent behaviour for @xmath52 . the freezing probability is non zero for all values of @xmath6 .
as the freezing probability shows finite size dependence close to @xmath50 and since it is dimensionless , we argue that it should show a finite size scaling behavior in the following manner @xmath53\ ] ] where @xmath54 is a scaling function .
we estimate @xmath55 and the exponent @xmath56 ( inset of fig.[sat_frz_rsf ] ) .
the estimates are obtained from the manifestly best collapse of the data after rescaling is done properly . as already discussed in section [ ising - net ] ,
finite size scaling is valid for the ising model on networks with @xmath57 where @xmath32 is the mean field value of the correlation length exponent and @xmath33 is equal to @xmath58 .
assuming the same to hold good here , we get @xmath59 from the fact that @xmath60 which is fairly close to the mean field value @xmath61 for ising model @xcite .
we next show that an order - disorder phase transition also occurs in this network .
the distribution of magnetisation @xmath62 has been studied ( fig . [ dist_rsf]a , b ) .
it is unimodal in nature for the values of @xmath6 far from the @xmath50 .
the peaks occur at @xmath63 for @xmath64 and @xmath65 for @xmath66 .
the distribution is bimodal close to @xmath50 with the peak values at @xmath65 and @xmath67 .
fluctuations of the saturation value of the magnetisation @xmath68 shows a peak which shifts toward @xmath50 and the the peak value increases as the system size is increased ( fig .
[ dist_rsf]c ) .
we have also estimated the time @xmath46 to reach the energetically stable state .
it shows a peak close to @xmath50 .
clearly @xmath46 diverges near @xmath50 , as the peak value increases with system size ( fig .
[ dist_rsf]d ) .
these behaviour suggest that there is an order - disorder transition also taking place as @xmath6 is varied . in principle
the order - disorder transition may take place at a value @xmath69 and we make further analysis to estimate @xmath70 more accurately .
we also check whether a mean field - like behaviour is present for the order - disorder transition . the variation with time of the magnetisation @xmath36 has been shown for different system sizes for two different values of the degree exponent @xmath6 ( fig .
[ dyn_mag_rsf]a , b ) .
the behaviour of the saturation values of the magnetisation @xmath47 for finite sizes shows the typical characteristics of a continuous phase transition with @xmath6 acting as the driving parameter ( fig .
[ sat_mag_rsf ] ) . using finite size scaling method
we indeed obtain a collapse of the data points for @xmath47 for different system sizes .
the following scaling form for @xmath47 has been used : @xmath71\\.\ ] ] we obtain @xmath72 , @xmath73 and @xmath74 ( inset of fig .
[ sat_mag_rsf ] ) .
the values of the exponents @xmath75 and @xmath32 are fairly close to the mean field values once again . for different system size @xmath9 .
inset shows the data collapse where @xmath76 has been plotted against @xmath77 .
[ sat_mag_rsf ] , width=188 ] next we discuss the behaviour of the residual energy and the spin flip probability which also attain a saturation value in time ( @xmath78 and @xmath79 ) . both the saturation values @xmath78 and @xmath79 show nonmonotonic variation with @xmath6 ( figs .
[ sat_flip_rsf ] , [ sat_enr_rsf ] ) , with a peak which shifts as the system size is increased .
@xmath78 in particular shows a very interesting behaviour with finite size . for @xmath80
, it decreases with system size indicating an absorbing phase which is actually the ordered phase as indicated by the behaviour of the magnetisation discussed above . however , there is a region between @xmath81 and @xmath82 , where @xmath78 increases with system size which indicates an active state . for @xmath83 ,
@xmath78 decreases with system size indicating an absorbing state once again .
hence we have two absorbing phases separated by an active phase and two transitions as the parameter @xmath6 is varied .
this is reminiscent of two distinct transitions observed in opinion dynamics models @xcite .
we found from the above studies that @xmath55 and @xmath84 are close but not exactly the same .
however , this could be due to finite size effects and these two might turn out to be identical , i.e. , @xmath85 in the thermodynamic limit .
this possibility is supported by the behaviour of both @xmath46 and @xmath79 .
while @xmath46 diverges at @xmath86 , @xmath79 decreases with system size below @xmath87 ( indicating an order phase ) and increases or remains constant above this value ( figs .
[ sat_enr_rsf ] ) . for different system size @xmath9 .
the vertical arrows separate different phases .
[ sat_flip_rsf ] , width=264 ] for different system size @xmath9 .
[ sat_enr_rsf ] , width=264 ] the ba network has no intrinsic parameter .
we calculate the relevant dynamic quantities for different system sizes .
the magnetisation @xmath36 slowly increases for the first few time steps and gets saturated at long times for all the system sizes .
both @xmath88 and @xmath89 decrease with time and then reach a saturation value .
saturation values of all these quantities as a function of system size have been plotted in fig .
[ sat_ba]a .
the saturation value of magnetisation @xmath90 decreases with system size and the variation shows a power law behaviour .
we fitted the variation with the form @xmath91 and the estimated exponents are @xmath92 and @xmath93 .
we have also studied the variation of saturation value of the fraction of spin flips @xmath94 and saturation value of residual energy @xmath95 .
both of them show a slowly increasing behaviour with system size and these increasing behaviour also follow power law ( fig .
[ sat_ba]a ) .
we fitted the variation of @xmath94 with the form @xmath96 and the exponents are @xmath97 and @xmath98 . for @xmath95 ,
the fitted form is @xmath99 and the exponents are @xmath100 and @xmath101 .
the important point to note here is this seems to be an active phase as the saturation value of the fraction of spin flips shows increase with system size .
in contrast , the same quantities plotted for the rsf with @xmath102 ( fig . [ sat_rsf ] ) shows that it is an absorbing phase ( as already noted in the previous subsection ) .
we will discuss more about this observation in section [ discussion ] .
the variation of @xmath95 with @xmath9 in rsf is however , similar , with @xmath103 .
@xmath90 for rsf decays as @xmath104 which signifies a faster decay compared to the ba network . .
in the ba model , the freezing probability @xmath105 increases with system size ( fig .
[ sat_ba]b ) .
freezing probability @xmath106 for @xmath107 .
thus the ba model is in an active disordered state where none of the configuration reaches the equilibrium ground state .
the variation of @xmath105 with system size fits well with the form @xmath108 and the calculated exponents are @xmath109 and @xmath110 .
the behaviour of the freezing probability as a function of @xmath9 is also quite different for the ba and the rsf networks ( with @xmath111 ) , in the latter we found a system size independent behaviour . :
variation of saturation value of magnetisation @xmath90 , fraction of spin flips @xmath94 and residual energy @xmath95 as a function system size .
[ sat_rsf ] , width=264 ] on the ws network , all the relevant quantities show a saturation behaviour as already noted previously on a slightly different version of the ws network @xcite . here in addition
, we have studied the spin flip probabilities .
the saturation value of the probability of spin flips @xmath78 has been plotted against the parameter @xmath17 for different system sizes and this quantity reveals interesting behaviour ( fig .
[ sat_ws]a ) .
it is clearly seen that above @xmath112 , the probability decreases as a function of system size @xmath9 while below @xmath113 it is almost size independent .
this indicates that there is an active absorbing phase transition taking place at this point .
the value of the freezing probability @xmath114 either increases with @xmath9 or remains constant which clearly shows that the entire phase is frozen for any @xmath115 .
we find that for @xmath116 the freezing probability reaches unity in the thermodynamic limit while it remains fairly constant beyond this value ( fig .
[ sat_ws]b ) .
this is consistent with the active - absorbing phase transition stipulated to take place at @xmath20 ; in the active state , one can never reach the equilibrium ground state configuration .
@xmath114 shows a peak at @xmath117 and the position of the peak is independent of system size which shows that the system is maximally disordered here .
the peak values of freezing probability as a function of system size @xmath9 is fitted with the form @xmath118 and the estimated values of the exponents are @xmath119 and @xmath120 ( inset of fig . [ sat_ws]b ) .
it may be noted that the same form is obeyed in the ba model .
we have studied zero temperature glauber dynamics of the ising model on three types of networks and compared the results .
frozen state is observed in all the three types of network models . for random scale free network ,
freezing probability is unity for @xmath121 , i.e. , the system never reaches the global equilibrium but for lower value of the parameter @xmath6 , it decreases with system size and shows a system size independent behaviour for @xmath122 .
this behaviour of freezing probability suggests a freezing transition point at @xmath123 .
we also find an order disorder transition point taking place very close to this point ; in fact we believe that they occur at the same point and the difference is only a finite size effect . also close to this point
, the first active - absorbing ( a - a ) phase transition takes place ; the disordered phase for @xmath124 is active while for @xmath125 , one gets an absorbing phase . a second
a - a transition takes place close to @xmath126 and the system evolves to an absorbing disordered state beyond this value . in all probability these two a - a transitions take place at @xmath127 and @xmath128 in absence of finite size effects ; these two points are significant as for @xmath129 , the average degree diverges while for @xmath130 , the degree variance diverges in the thermodynamic limit .
one can compare the results of rsf network and ba network for the same characteristic degree exponent @xmath131 .
the residual energy shows an increase with system size @xmath9 in both cases in a power law manner with an exponent which is fairly close .
also , the saturation value of magnetisation decreases with @xmath9 in both networks in a power law manner , corresponding exponents are however quite different .
the saturation energy and magnetisation behaviour are consistent with the fact that the state is disordered for both rsf and ba networks .
however freezing probability for ba model and rsf network show different behaviour with system size .
for rsf , the freezing probability shows a system size independent behaviour while for ba model , it has a nonlinear dependence .
as far as the spin flip probability is concerned rsf ( at @xmath102 ) again differs from the ba network .
the rsf network and ba model are intrinsically different , ba is a growing network generated with a particular strategy .
there is no loop in this network . in the case of rsf ,
the structure is completely different , there may be loops .
although in a numerical study , the value of @xmath6 may not be exactly 3 in either the rsf or ba networks due to finite size effects , we believe that this can not be the reason for the results being qualitatively different .
in fact previous studies on rsf and ba have shown that characteristic features may be quite different for the two networks @xcite .
the results obtained for the ws model can also be compared to those found for the rsf network and ba network . in the ws network
an a - a transition is observed at @xmath132 which is also the order - disorder transition point .
hence this is similar to the occurrence of an a - a and an order - disorder transition occurring simultaneously in the rsf .
however in the ws network , the entire disordered phase is active . for small average degree
the system is in an active and for large degree in an absorbing phase . for the rsf network ,
two a - a transitions exist where absorbing phase is observed for _ both _ large degree and small degree and an active state exists in between these two absorbing phases . in both rsf and ws
another interesting feature is present , the ordered state shows a finite freezing probability with negligible system size dependence . in the ws network maximum value of the freezing probability @xmath114
occurs at @xmath133 and shows a behaviour similar to the freezing probability in ba network as a function of system size . to summarise ,
a systematic study of ordering dynamics of the ising model on scale free networks has been made for the first time to the best of our knowledge .
it is observed that the system freezes to a non equilibrium steady state for all values of the relevant parameters in the random scale free networks ( rsf ) and the barabsi - albert model ( ba ) .
the presence of two active - absorbing phase transitions in the rsf makes it different from the ws network where only one such transition is observed .
it is also concluded that in rsf one of the active - absorbing phase transition takes place at the order - disorder transition point which is similar to what is observed in the ws network .
[ discussion ] acknowledgements : the authors thank soham biswas for his suggestions and encouragement .
ak acknowledges financial support from ugc sanction no .
f.7 - 48/2007(bsr ) .
ps acknowledges financial support from csir project .
computations made on hp cluster financed by dst ( fist scheme ) , india .
a. barrat and m. barthelemy and a. vespignani , _ dynamical processes on complex networks _ , cambridge university press , 2008 .
sen and b. k. chakrabarti , _ sociophysics : an introduction _ , oxford university press , 2013 . c. castellano , s. fortunato and v. loreto , rev .
phys . * 81 * 591 ( 2009 ) .
d. j. watts and s. h. strogatz , nature * 393 * 440 ( 1998 ) . a. barrat and m. weigt , eur .
j. b * 13 * , 547 ( 2000 ) .
m. gitterman , j. phys .
a : math . gen . * 33 * 8373 ( 2000 ) .
t. nikoletopoulos , a. c. c. coolen , i. prez castillo , n. s. skantzos , j. p. l. hatchett and b. wemmenhove , j. phys .
a * 37 * , 6455 ( 2004 ) .
b. j. kim , h. hong , p. holme , g. s. jeon , p. minnhagen and m. y. choi , phys .
e * 64 * 056135 ( 2001 ) . c. p. herrero , phys .
e * 65 * 066110 ( 2002 ) .
h. hong , b. j. kim and m. y choi , phys .
e * 66 * 011107 ( 2002 ) .
a. chatterjee and p. sen , phys .
e * 74 * 036109 ( 2006 ) .
m. leone , a. vzquez , a. vespignani , and r. zecchina , eur .
j. b * 28 * , 191 ( 2002 ) .
s. n. dorogovtsev , a. v. goltsev , and j. f. f. mendes , phys .
e * 66 * , 016104 ( 2002 ) .
f. igli and l. turban , phys .
e * 66 * , 36140 ( 2002 ) .
a. aleksiejuk , j. a. holyst and d. stauffer , physica a * 310 * 260 ( 2002 ) .
g. bianconi , phys .
lett . a * 303 * 166 ( 2002 ) . c. p. herrero , phys .
e . * 69 * 067109 ( 2004 ) .
j. viana lopes , yu .
g. pogorelov , j. m. b. lopes dos santos and r. toral , phys .
e * 70 * 026112 ( 2004 ) .
v. spirin , p. l. krapivsky and s. redner , phys .
e * 63 * 036118 ( 2001 ) .
p. svenson , phys .
e * 64 * 036122 ( 2001 ) .
o. haggstrom , physica a * 310 * 275 ( 2002 ) .
d. boyer and o. miramontes , phys .
e * 67 * r035102 ( 2003 ) .
p. k. das and p. sen , eur .
. j. b * 47 * 391 ( 2005 ) . c. castellano , v. loreto , a. barrat , f. cecconi and d. parisi , phys .
e * 71 * 066107 ( 2005 ) .
y. baek , m. ha and h. jeong , phys .
e * 85 * 031123 ( 2012 ) .
s. biswas and p. sen , phys .
e * 84 * 066107 ( 2011 ) .
r. albert and a. l. barabsi , rev .
phys . * 74 * 47 ( 2002 ) .
r. albert and a. l. barabsi , science , 286 , 509 ( 1999 ) . s. n. dorogovtsev and j. f. f. mendes , _ evolution of networks : from biological nets to the internet and www _ , oxford university press , oxford , 2003. m. bogu , r. pastor - satorras and a. vespignani , eur .
j. b * 38 * , 205 ( 2004 ) .
m. e. j. newman , _ networks : an introduction _
( oxford university press , 2010 ) .
sen , journal of statistical mechanics p04007 ( 2007 ) .
m. catanzaro , m. bogu and r. pastor - satorras , phys .
e * 71 * 027103 ( 2005 ) .
h. e. stanley , _ introduction to phase transitions and critical phenomena _
( oxford university press , jul 1987 ) .
m. droz , a. l. ferreira and a. lipowski , phys .
e * 67 * 056108 ( 2003 ) .
a. khaleque and p. sen , physica a * 413 * 599 ( 2014 ) . | a zero temperature quench of the ising model is known to lead to a frozen steady state on random and small world networks .
we study such quenches on random scale free networks ( rsf ) and compare the scenario with that in the barabsi - albert network ( ba ) and the watts strogatz ( ws ) addition type network .
while frozen states are present in all the cases , the rsf shows an order - disorder phase transition of mean field nature as in the ws model as well as the existence of two absorbing phases separated by an active phase .
the ws network also shows an active - absorbing ( a - a ) phase transition occurring at the known order - disorder transition point .
the comparison of the rsf and the ba network results show interesting difference in finite size dependence . |
wncyr10 at 12.0pt * 11 . * ... za nerazumnye pomyshleniya ih nepravdy , po kotorym oni sluzhili besslovesnym presmykayuwimsya i prezrennym chudoviwam , ty v nakazanie poslal na nih mnozhestvo besslovesnyh zhivotnyh , chtoby oni poznali , chto chem kto sogreshit , tem i nakazyvaetsya ... i bez `` 0btogo oni mogli pogibnut''7e ot odnogo dunoveniya , presleduemye pravosudiem i rasseivaemye duhom sily tvoe"1a ; no + ty vse raspolozhil meroyu , chislom i vesom .
wncyb10 at 12.0pt once upon a time the great russian physicist - theorist and mathematician bogoljubov said that the last line in the above written fragment from the book of proverbs which is a part of the bible ( non - canonical though ! ) ... ; but you did all arrange by measure , number and weight " represents the definition of the physics @xcite . probably this ... ; but you did all arrange by measure , number and weight " was an earliest evidence of the principle that all in the world have to be in harmony with each other . in fact , the word harmony ( syn : music : accord , concord , consonance ) has the greek origin from @xmath0 which means orderliness ( symmetry ) of the whole , commensurability ( proportionality ) of its parts .
the idea of harmony has intensively been elaborated by pythagoras , the greek philosopher and mathematician and founder of the pythagorean school @xcite .
originally from samos , pythagoras founded a society which was at once a religious community and a scientific school flourished at kroton in southern italy about the year 530 b.c .
pythagoras was the first genius of western culture .
he had a multifaceted magnetic personality
an intelligent mathematician and a religious thinker , both co - existed in him .
his main contributions are in geometry , numbers , music , cosmology , astronomy , philosophy and religion .
pythagoras must have been one of the world s greatest men , but he wrote nothing though numerous works are attributed to him , and it is hard to say how much of the doctrine one knows as pythagorean is due to the founder of the society and how much is later development .
it is also hard to say how much of what we are told about the life of pythagoras is trustworthy . for a mass of legend gathered around his name : sometimes he is represented as a man of science , and sometimes as a preacher of mystic doctrines , and we might be tempted to regard one or other of those characters as alone historical . certainly , it s true that there is no need to reject either of the traditional views . even though many wonderful things related to pythagoras , belong to legend , and seem to have no historical foundation , similarly the description of the learned works which he wrote is not attested by reliable historians and also belongs to the region of fable , nevertheless it is no doubt however , that he founded a school , or , rather , a religious philosophical society , which exerted great influence on the intellectual development of human civilization and had a fundamental importance all the time . of great influence were the pythagorean doctrines that numbers were the basis of all things and possessed a mystic significance , in particular the idea that the cosmos is a mathematically ordered whole .
aristotle wrote :
pythagorean having been brought up in the study of mathematics , thought that things could be represented by numbers ... and that the whole cosmos consists of a scale and a number "
. briefly stated , the doctrine of pythagoras was that all things are numbers .
pythagoras was led to this conception by his discovery that the notes sounded by stringed instrument are related to the length of the strings .
he conducted remarkable investigation in music " as he was a musician .
harmonies correspond to most beautiful mathematical ratio , he stated .
melodious musical tunes could be produced on a stringed instrument by plucking the string at particular points , which correspond to mathematical ratios .
such beautiful mathematical ratios are @xmath1 ( an octave ) , @xmath2 ( a fifth ) , and @xmath3 ( a fourth ) .
pythagoras recognized that first four numbers @xmath4 known as tetractys " , whose sum equals ten ( @xmath5 ) , contained all basic musical intervals : the octave , the fifth and the fourth .
in fact , all the major consonances , that is , the octave , the fifth and the fourth are produced by vibrating strings whose lengths stand to one another in the ratios of @xmath1 , @xmath2 and @xmath3 respectively .
recent major scale in according to pythagoras tune looks like @xmath6 where @xmath7 is major second ( fifth of fifth with octave lowering ) ; @xmath8 is major sixth ( fifth of major second ) ; @xmath9 is major third ( fifth of sixth with octave lowering ) ; @xmath10 is major seventh ( fifth of third ) .
the resemblance which pythagoras perceived between the orderliness of music , as expressed in the ratios which he had discovered and the idea that cosmos is an orderly whole , made up of parts harmoniously related to one another , led him to conceive of the cosmos too as mathematically ordered .
pythagoras compared the eight planets ( there were seven planets known the babylonians : moon , mercury , venus , sun , mars , jupiter and saturn ) , including the earth , with the musical octave and the seven planets , excluding the earth , as seven strings of the musical instrument lair .
the planets situated at different distances and moving at different speed correspond to different notes on musical octave .
the planets moving with higher speed produce higher notes and those with lower speed produce lower notes .
the celestial harmony of moving planets produces heavenly music ( the music of the spheres " ) analogous to different notes of musical octave . according to pythagoras ,
the sphere was the most beautiful solid and the circle the most beautiful shape .
thus , a spherical planet moving in circular orbit would form a harmonious constellation .
pythagoras worked out the distances of the planets from the earth .
he arranged the planets in order of increasing distances of the planets from the earth .
the order given by him was the moon , mercury , venus , the sun , mars , jupiter and saturn .
some pythagoreans believed that the earth moved round a central fire .
the earth did not always face the central fire .
this accounted for day and night on the earth .
they also believed that the moon as well as the sun shone because they reflected light from their surfaces received from the central fire .
perhaps the idea of central fire later on led to the heliocentric ( sun at the centre of the solar system ) configuration of solar system .
pythagoras observation of heavens suggested to him that the motion of the heavenly bodies was cyclic and that the heavenly bodies returned to the place from which they had started . from this
, pythagoras concluded that there must be a cycle of cycles , a greater year and on its completion the heavenly bodies returned to the original position and the same heavenly constellation would be observed again and again .
he called this the eternal recurrence .
pythagoras doctrine that mathematics contains the key to all philosophical knowledge , an idea , which was by his followers afterwards developed into an elegant number - theory .
the pythagorean philosophy in its later elaboration is dominated by the number - theory .
being the first , apparently , to observe that natural phenomena , especially the phenomena of the astronomical world , may be expressed in mathematical formulas , the pythagoreans held that numbers are not only the symbols of reality , but the very substance of real things .
pythagoras associated numbers with geometrical notions and numerical ratios with shapes .
he associated number one with a point , too with a line , three with a triangle ( the surface ) and four with a tetrahedron ( the solid ) .
thus , one point generates dimensions , two points generate a line of one dimension , three points generate a surface of two dimensions , and four points generate three - dimensional solid figures . in geometry , numbers represent lengths , their squares represent areas , their cubes represent volumes . starting from numbers , numerical ratios and their powers , one can construct geometrical figures of different shapes and geometrical solids of different sizes . using distance the arrangement of planets , their motion
, their orbital path , their distances from the center and their interrelations with each other can be worked out .
thus , according to pythagoras all relations could be reduced to number relations and hence , the whole cosmos is a scale and a number based phenomenon . according to pythagoras ,
ten is the perfect number , because it is the sum of one , two , three , and four the point , the line , the surface , and the solid .
there are the second type of perfect numbers : according to pythagoras the second type of perfect numbers are those were the numbers equal to sum of their factors .
for instance 28 has factors 1 , 2 , 4 , 7 , 14 and @xmath11 . from perfect numbers ,
pythagoras was led to amicable numbers like 220 and 284 .
amicable numbers form a pair of numbers where each number is equal to the sum of the factors of the other numbers .
for instance 220 has factors 1 , 2 , 4 , 5 , 10 , 11 , 20 , 22 , 44 , 55 , 110 .
the sum of these factors is @xmath12 .
moreover , 284 has factors 1 , 2 , 4 , 71 , 142 .
the sum of these factors is @xmath13 .
triangular numbers have been introduced by pythagoras : pythagoras called numbers 1 , 3 , 6 , 10 , 15 , 21 , 28 , 36 , 45 , 55 , 66 as triangular numbers because these numbers can be arranged so as to form triangles .
if @xmath14 are sides of a right - angled triangle and @xmath15 is the hypotenuse then according to pythagoras theorem @xmath16 .
the triad of positive integers @xmath17 satisfying the relation @xmath18 is called the pythagorean triad of numbers .
about fifteen such triads were previously known like ( 3,4,5 ) , ( 5,12,13 ) , ( 7,24,25 ) , ( 9,12,15 ) , ( 15,36,39 ) .
the pythagorean triads in which the numbers @xmath19 do not have a common factor are called primitive pythagorean triads . for example
( 3,4,5 ) , ( 5,12,13 ) , ( 7,24,25 ) etc .
are primitive pythagorean triads . but
( 9,12,15 ) , ( 15,36,39 ) are not primitive triads .
it is believed that pythagoras himself discovered the formula for determining triads of numbers satisfying the relation @xmath18 .
in fact , all pythagorean triads can be expressed via formulae a = m^2-n^2,b=2mn , c = m^2+n^2,[triad ] where @xmath20 are any positive integers @xmath21 . from his observations in music , mathematics and astronomy
, pythagoras generalized that everything could be expressed in terms of numbers and numerical ratios .
numbers are not only symbols of reality , but also substances of real things .
hence , he claimed - all is number .
the importance of this conception is very great , for example , it is the ultimate source of galileo s belief _ il libro della natura scritto in lingua matematica " _ that the book of nature is written in mathematical symbols and hence the ultimate source of modern physics in the form in which it came to us from galileo .
it may be taken as certain that the union of mathematical genius and mysticism is common enough .
pythagoras himself discovered the numerical ratios which determine the concordant intervals of the musical scale .
similar to musical intervals , in medicine there are opposites , such as the hot and the cold , the wet and the dry , and it is the business of the physician to produce a proper blend " of these in the human body .
the pythagoreans contended that the opposites are found everywhere in nature , and the union of them constitutes the harmony of the real world .
they also argued for the notion that virtue is a harmony , and may be cultivated not only by contemplation and meditation but also by the practice of gymnastics and music .
pythagoras held the theory that what gives form to the unlimited is the limit .
that is the great contribution of pythagoras to philosophy , and we must try to understand it .
it was natural for pythagoras to look for something of the same kind in the world at large .
musical tuning and health are alike means arising from the application of limit to the unlimited . in their psychology and their ethics
the pythagoreans used the idea of harmony and the notion of number as the explanation of the mind and its states , and also of virtue and its various kinds .
pythagoras argued that there are three kinds of men , just as there are three classes of strangers who come to the olympic games .
the lowest consists of those who come to buy and sell , and next above them are those who come to compete .
best of all are those who simply come to look on .
men may be classified accordingly as lovers of wisdom , lovers of honour , and lovers of gain .
that seems to imply the doctrine of the tripartite soul , which is also attributed to the early pythagoreans on good authority .
the pythagoreans were religiously and ethically inclined , and strove to bring philosophy into relation with life as well as with knowledge .
the pythagoreans believed also in reincarnation or transmigration ( doctrine of rebirth ) , that is , the soul , after death , passes into another living thing , which presupposes the ability of the soul to survive the death of the body , and hence some sort of belief in its immortality .
the above detailed introduction is made so as to show in the next sections that the great ancient pythagorean ideas have found themselves in the latest researches in high energy elementary particle and nuclear physics . in this respect
we will concern and discuss the mathematical , physical and geometrical aspects of the famous froissart theorem and in this way we will easily establish a link of this theorem to the mathematics and ideas elaborated in the pythagorean school . in other words , we would like to show a harmony of the froissart theorem just in the pythagoreans sense .
in the year 1961 french physicist marcel froissart discovered and proved a remarkable theorem , which stated that two - body reaction @xmath22 amplitude , satisfying mandelstam representation , is bounded by expressions of the form @xmath23 at the forward and backward angles , and @xmath24 at any fixed angle in the physical region , @xmath25 being a constant , @xmath26 being the total squared c.m .
energy ( one of the mandelstam invariant variables @xmath27 ) .
this corresponds to the total cross sections increasing at most like @xmath28 @xcite .
a little bit later it was shown that the analytical properties of two - particle scattering amplitude , which may be established strictly in the framework of axiomatic quantum field theory , bring us to the froissart statements as well . up - to - date derivation of the froissart theorem can be realized in a few steps , and we briefly sketch out it here . for simplicity we consider a reaction of elastic scattering @xmath29 for two scalar particles .
the scattering amplitude of the two - body reaction may be considered as a function of the invariant variable @xmath30 and two unit vectors @xmath31 and @xmath32 on two - dimensional sphere @xmath33 , which characterise the initial and final states of two - particle system : @xmath34 , @xmath35 is c.m .
momentum of particles in an initial state @xmath36 @xmath37 is lorentz boost , and the same with the primes in a final state . in the first step we wright the partial wave expansion @xmath38 =
_ lmy _ lm ( * n*)f_l(s)_lm(*n * ) = _ l(2l+1)f_l(s)p_l(*n* ) , [ 1 ] where @xmath39 , @xmath40 is two - particle phase space volume , @xmath33 is a surface of two - dimensional unit sphere , @xmath41 , and an addition theorem for the spherical harmonics in second line of eq .
( [ 1 ] ) has been used .
the second invariant mandelstam variable @xmath42 ( momentum transfer ) is related to @xmath43 by the following equation = 1 + .[2 ] a remarkable analytic properties of scattering amplitudes as functions of momentum transfer have been discovered in the year 1958 by harry lehmann @xcite using jost - lehman - dyson representation especially dyson s theorem for a representation of causal commutators in local quantum field theory @xcite .
lehmann proved that imaginary part of two - body interaction amplitude is analytic function of @xmath43 , regular inside an ellipse in @xmath43-plane with center at the origin and with semi - major axis z_0(s ) = 1 + _
l(s ) , _ l(s ) = , [ 3 ] where @xmath44 and @xmath45 define the support of spectral function in the jld representation by the requirements of spectral condition or spectrality .
actually , @xmath44 and @xmath45 are the lowest mass values of the physical states for which the following matrix elements are not equal to zero @xmath46 where @xmath47 and @xmath48 are local heisenberg s currents of particles @xmath49 and @xmath50 .
he also shown that two - body interaction amplitude , as itself , is analytic function of @xmath43 , regular inside an ellipse in @xmath43-plane with center at the origin and with semi - major axis @xmath51 which is related to @xmath52 by the equation x_0(s ) = .[4 ] afterwards the fundamental results of harry lehmann were improved by martin @xcite and sommer @xcite : it was shown that imaginary part of two - body interaction amplitude is analytic function of @xmath43 , regular inside an ellipse in @xmath43-plane with semi - major axis z_0(s ) = 1 + _
m(s ) , _ m(s ) = , t_0 = 4m_^2,[5 ] ^2 = = , [ lambda ] where @xmath53 is pion mass .
correspondingly two - body interaction amplitude , as itself , appears as analytic function of @xmath43 , regular inside an ellipse in @xmath43-plane with semi - major axis @xmath51 which is related to @xmath52 by eq .
( [ 4 ] ) .
the fundamental results derived by lehmann and improved by his followers are of great importance because it has been shown that the partial wave expansions ( [ 1 ] ) which define physical scattering amplitudes continue to converge for complex values of the scattering angle , and define uniquely the amplitudes appearing in the unphysical region of non - forward dispersion relations .
in fact , expansions converge for all values of momentum transfer for which dispersion relations have been proved .
the proved analyticity of two - body interaction amplitudes as functions of two complex mandelstam variables @xmath26 and @xmath42 in a topological product of cut @xmath26-plane with the cuts ( @xmath54 ) except for possible fixed poles and circle @xmath55 in @xmath42-plane allowed in a more general case to save the fundamental froissart results previously obtained at a more restricted mandelstam analyticity . really , let us wright cauchy representation for imaginary part of two - body interaction amplitude @xmath56 where contour @xmath25 is a boundary of an ellipse in @xmath43-plane with semi - major axis given by eq .
( [ 5 ] ) . using heine formula
@xmath57 we obtain i m f_l(s ) = _ c dz i m f_2(s;z)q_l(z).[6 ] from eq .
( [ 6 ] ) it follows i m f_l(s)_2(s)_zc|im f_2(s;z)|_zc|q_l(z)|(c),[7 ] where @xmath58 is a length of contour @xmath25 .
representation ( [ 6 ] ) where estimate ( [ 7 ] ) followed from is a good tool to study an asymptotic behaviour of partial waves at large orbital momentum . using asymptotic properties of the legendre functions @xmath59
@xcite @xmath60 @xmath61 is some polynomial in @xmath26 , we find i m f_l(s)_2(s)p_2(s)()^[z_0(s)+]^-l , |l| .
[ 8 ] if we put @xmath62 , @xmath63 then estimate ( [ 8 ] ) at large values of @xmath26 may be rewritten in the form i m f_l(s)(-l ) , s,[9 ] where _
2(s ) = ( ) ^_2(s ) p_2(s).[polinom ] thus we have obtained a very important result : analyticity of two - body interaction amplitudes as functions of @xmath43 , regular inside an ellipse in @xmath43-plane , results in exponential decrease of partial waves as functions of orbital momentum @xmath64 at large values of @xmath64 .
this means that the significant contribution to the partial wave expansion ( [ 1 ] ) is determined by partial waves for which the orbital momentum does not exceed the quantity l = .[10 ] the contribution of partial waves with @xmath65 to the partial wave expansion will be exponentially small .
let us decompose the partial wave expansion in two terms i m f_2(s;=1 ) = _ l=0^l-1(2l+1)im
f_l(s ) + i m f_2^l(s),[11 ] where the second term in eq .
( [ 11 ] ) contains the contribution of partial waves with @xmath66 .
now we would like to take advantage of unitarity condition which can be written for the partial waves as the following sequence of inequalities 0|f_l(s)|^2im f_l(s)|f_l(s)|1.[12 ] taking into account the unitarity condition we get for the first term in eq .
( [ 11 ] ) an estimate in the form _ l=0^l-1(2l+1)im f_l(s)_l=0^l-1(2l+1 ) = = , [ 13 ] where expression ( [ 10 ] ) for the quantity @xmath67 has been used .
froissart has shown that the second term in eq . (
11 ) is asymptotically small compared to the first one at large values of @xmath26 , so that we finally get i m f_2(s;=1 ) < .[14 ] the optical theorem relates a total cross section of two - body interaction with imaginary part of two - body forward elastic scattering amplitude @xmath68 @xmath69-function is defined by eq .
( [ lambda ] ) . hence from estimate (
[ 14 ] ) it follows an upper bound for the total cross section of two - body interaction _
ab^tot(s ) < .[15 ] where , as it was mentioned above , @xmath70 here is just the place to introduce the physical notion of the effective radius of two - body forces @xcite .
let us define the effective radius @xmath71 of two - body forces by the following equation r_2(s ) = , [ 16 ] where the definition ( [ 10 ] ) of the quantity @xmath67 and expression ( [ lambda ] ) for @xmath72 have been used . now upper bound ( [ 15 ] ) in terms of such defined quantity @xmath71 takes the form _
^ 2(s).[17 ] this form of the upper bound for experimentally measured quantity @xmath73 has a quite transparent physical and clear geometrical meanings : it means that the total cross section of two - body interaction is bounded by the area of a surface of two - dimensional sphere whose radius is defined by the effective radius of two - body forces .
a remarkable property of upper bound ( [ 17 ] ) consist in the fact that here all information about analytic properties of two - body interaction amplitudes is hidden in the physically tangible quantity ( [ 16 ] ) which is the effective radius of two - body forces . if we put @xmath74 equal to @xmath75 given by eq .
( [ 5 ] ) then from eqs .
( [ 13 ] ) and ( [ polinom ] ) it follows that @xmath76 in that case for the the effective radius of two - body forces we find from eq .
( [ 16 ] ) r_2(s ) = ~ ( s / s_0 ) = ( s / s_0),s.[18 ] in the article froissart gave an excellent semiclassical explanation corroborating his theorem .
we would like to present here a remarkable fragment from section ii of the froissart paper @xcite .
he wrote : _ to get intuitive idea why the amplitude is bounded in the physical region , let us consider a classical problem : two particles interact by means of absorptive yukawa potential @xmath77 .
if @xmath49 is the impact parameter , the total interaction seen by a particle for large @xmath49 is likely to be approximately @xmath78 .
if this is small compared to one , there will be practically no scattering .
if @xmath79 is large compared to one , there will be practically complete scattering , so that the cross section will be essentially determined by the value @xmath80 where @xmath81 .
it is @xmath82 .
if we now assume that @xmath83 is a function of the energy , and increases like a power of the energy , then @xmath84 will vary at most like the squared logarithm of the energy_. " in fact , froissart anticipated here a running coupling and quasi - potential character of strong forces . later on it
was shown @xcite that the hypothesis about validity of the dispersion relations in the momentum transfer leads , for any value of the energy @xmath26 , to a potential which is a superposition of yukawa potentials with energy dependent intensities .
this fact together with a theorem on single - time reduction in quantum field theory @xcite provides a strong basis for semiclassical consideration given by froissart .
however , it should be stressed that upper bound ( [ 17 ] ) has a quite different geometrical sense compared to semiclassical consideration given by froissart : eq .
( [ 17 ] ) shows that the total cross section of two - body interaction is bounded by the area of a surface of the sphere with the radius equal to the effective radius of two - body forces but not by the area of a disk with the same radius .
unitarity bound ( [ 17 ] ) states that the total probability ( per unit volume per unit time in fraction of particles density flux ) of all possible ( elastic and inelastic ) two - particle interactions , which take place in a limited volume @xmath85 during a limited interval of time @xmath86 , is limited by the area of a surface of the sphere which is , actually , a boundary of the volume @xmath85 .
this means that widely discussed in the recent literature concerning some physical problems at planck scale _ the holographic principle _
@xcite has been incorporated in the general scheme of axiomatic quantum field theory and resulted from the general principles of local quantum field theory .
in our works @xcite it was shown that there is a quite natural geometrical generalization of the froissart theorem to the case of multiparticle interaction . in this respect
it should be noted that the problem of finding such generalization is non - trivial because at least the known singularities of multiparticle scattering amplitudes related to disconnected parts by cluster structure of the amplitudes point to the fact that for the total amplitude of @xmath87-particle scattering ( @xmath88 ) there is no such generalization .
connected part of @xmath87-particle ( @xmath88 ) scattering amplitudes contains singular rescattering terms as well .
therefore , the first problem which arises in this case is to define a suitable object connected with the @xmath89 reaction amplitude which would permit a correct formulation of the problem .
it turns out there is a wide class of many - particle reaction amplitudes for which such a problem would be quite meaningful .
we have shown that these amplitudes should be understood as amplitudes of true @xmath87-particle interaction or @xmath87-body forces amplitudes ; see details in @xcite . here
we reproduce our results taking a line stated in previous section .
the scattering amplitude of the @xmath87-body reaction may be considered as a function of the invariant variable @xmath90 and two unit vectors @xmath91 and @xmath92 on @xmath93-dimensional sphere @xmath94 , which characterise the initial and final states of @xmath87-particle system : @xmath95 dimensionality @xmath96 of multidimensional space is related to the number of particles @xmath87 by the equation @xmath97 .
there are many ways to introduce the spherical coordinates in multidimensional space
. moreover , there are some peculiarities related to a parametrization of relativistic @xmath87-particle system .
however , we will not concern this subject here because it does not play any role for our main goal .
for the details we refer to @xcite and references therein . as above we may wright the partial wave expansion @xmath98 = _ lmy _ lm ( * e*)f_l(s)_lm(*e * ) = _ l(+1)f_l(s)c_l^ ( ) , [ 19 ] where @xmath99 , @xmath100 is @xmath87-particle phase space volume , @xmath101 is a surface of @xmath93-dimensional unit sphere , @xmath102 , and we have used in second line of eq .
( [ 19 ] ) an addition theorem for the ( hyper)spherical harmonics in multidimensional space @xmath103 @xmath104 where @xmath105 is gegenbauer polynomial .
here we contented ourself with a special class of @xmath87-body forces scattering amplitudes which are invariant under rotation in multidimensional space ( so called @xmath106-invariant amplitudes ) .
we will assume that for physical values of the variable @xmath26 imaginary part of @xmath87-body forces scattering amplitude is analytic function of @xmath107 , regular inside an ellipse @xmath108 in @xmath107-plane with center at the origin and with semi - major axis z_n(s ) = 1 + _
n(s ) , _ n(s ) = , [ 20 ] and for any @xmath109 is polynomially bounded in the variable @xmath26 , @xmath110 is some constant of mass dimensionality independent of @xmath26 , * q * is global momentum ( dependent of @xmath26 ) of @xmath87-particle system which will be defined later on .
such analyticity of @xmath87-body forces scattering amplitudes was called _ global _
if it is so , one can wright cauchy representation for imaginary part of @xmath87-body interaction amplitude @xmath111 where contour @xmath112 is a boundary of an ellipse @xmath108 in @xmath107-plane with semi - major axis given by eq .
( [ 20 ] ) .
there is a standard generalization of heine s expansion of the cauchy denominator @xcite = ( -i)2 ^ 2[()]^2(z^2 - 1)^-1/2 _ l=0^(l+)d_l^ ( z)c_l^(t),[21 ] which converges absolutely for @xmath113 is a second solution to gegenbauer s equation .
the restriction ( [ converge ] ) requires that the point @xmath42 lie within that ellipse in the complex @xmath42-plane with foci at @xmath114 which passes through the point @xmath115 .
in particular from eq .
( [ 21 ] ) it follows @xmath116 as a result we obtain i m f_l(s ) = _ c_n dz ( z^2 - 1)^-1/2d_l^(z)im f_n(s;z).[22 ] representation ( [ 22 ] ) is very useful to study an asymptotic behaviour of partial waves at large global orbital momentum .
taking into account asymptotic properties of the gegenbauer functions @xmath117
@xcite @xmath118 @xmath119 is some polynomial in @xmath26 , we find @xmath120 ( ) ^1- ( z_n(s)+)^-l , |l| . [ 23 ] finally if we put @xmath121 , @xmath122 then we get at large values of @xmath26 i m f_l(s)(-l ) , s,[24 ] where p_n(s , ) = 2^+1()[2_n(s)]^(-1)/2_n(s ) p_n(s).[25 ] estimate ( [ 25 ] ) shows that partial waves as functions of global orbital momentum @xmath64 exponentially decrease at large values of @xmath64 , i.e. the significant contribution to the partial wave expansion ( [ 19 ] ) is resulted from partial waves for which the global orbital momentum does not exceed the quantity = .[26 ] the contribution of partial waves with @xmath123 to the partial wave expansion will be exponentially small .
so , we decompose the partial wave expansion in two terms i m f_n(s;=1 ) = _ l=0^(+1)im
f_l(s)c_l^(1 ) + i m f_n^(s),[27 ] where the second term in eq .
( [ 27 ] ) contains the contribution of partial waves with @xmath123 . taking into account the unitarity condition ( [ 12 ] ) for the partial waves we get for the first term in eq .
( [ 27 ] ) an estimate @xmath124 = ( 1+o()),[28 ] where we inserted @xmath125 $ ] .
it can easily be seen that the second term in eq .
( 27 ) is asymptotically small compared to the first one at large values of @xmath26 , so that we finally get i m f_n(s;=1 ) < .[29 ] where we have used expression ( [ 26 ] ) for @xmath126 and relation @xmath127 .
by analogy with eq .
( [ 16 ] ) let us introduce the effective radius @xmath128 of @xmath87-body forces r_n(s)=p_n(s,),[30 ] where the definition ( [ 26 ] ) of the quantity @xmath126 and expression ( [ 20 ] ) for @xmath129 have been used . now upper bound ( [ 29 ] ) in terms of such defined quantity @xmath128 takes the form i m f_n(s;=1 )
< = j_n(s)s_d-1[r_n(s)]^d-1,[31 ] where j_n(s ) = = .[32 ] with account of the generalized optical theorem relating a total cross section of @xmath87-body interaction with imaginary part of @xmath87-body forces forward scattering amplitude @xcite @xmath130 from estimate ( [ 31 ] ) we obtain an upper bound for the total cross section of @xmath87-body interaction _
n^tot(s ) < s_d-1[r_n(s)]^d-1.[33 ] here again , as it should be , upper bound ( [ 33 ] ) has a quite clear geometrical meaning : the total cross section of @xmath87-body interaction is bounded by the area of a surface of @xmath93-dimensional sphere whose radius is defined by the effective radius of @xmath87-body forces .
again all information about global analyticity of @xmath87-body interaction amplitudes is hidden in the physical quantity ( [ 30 ] ) which is the effective radius of @xmath87-body forces . from eqs .
( [ 25 ] ) and ( [ 29 ] ) it follows that @xmath131 for the the effective radius of @xmath87-body forces we find from eq .
( [ 30 ] ) in that case r_n(s ) ~ ( s / s_0),r_n=,s.[34 ] upper bounds ( [ 31],[33 ] ) are a direct consequence of global analyticity of @xmath87-body forces scattering amplitudes which , in one s turn , is a direct geometrical generalization of analytic properties of two - body scattering amplitude strictly proved in axiomatic quantum field theory . at present
we do not know to what extend global analyticity of @xmath87-particle scattering amplitudes @xmath132 is a consequence of general principles of local quantum field theory .
the validity of such an assumption is obvious to us if we rely on the physical nature of @xmath87-body forces : our intuition tells us that true @xmath87-body interactions should manifest themselves only in the case when all the @xmath87 particles are in a sufficiently limited volume . on the other hand , from the beginning one
may , by definition , consider the @xmath87-body forces scattering amplitude to be a globally analytic part of the total @xmath133-matrix which may always be singled out from it @xcite . at last
, we have to give the definition of global momentum @xmath134 for the relativistic @xmath87-particle system . in this respect , first of all , note that momentum @xmath72 for two - particle system has been defined in a relativistic covariant way . under any lorentz transformation @xmath126 from the restricted lorentz group @xmath135 momentum
@xmath72 is transforming by wigner rotation : @xmath136 @xmath137 is lorentz boost .
this means that @xmath138 defined by eq .
( [ lambda ] ) is a lorentz invariant quantity .
moreover , we would like to emphasize the following asymptotic properties ^2s , s;^2 2_2(-m_2 ) , m_2 , m_2=m_a+m_b , _ 2=.[35 ] the expression of @xmath139 given by eq .
[ lambda ] can be rewritten in the form ^2 = 16s(_2(s)/s_2)^2 = 16sa_2 ^ 2(s).[36 ] the definition of global momentum for the relativistic @xmath87-particle system should be given such as to save the asymptotic properties shown by eqs .
( [ 35 ] ) .
such generalization for any number of particles looks like ^2=_n s^(n-1)/(3n-5)a_n^2/(3n-5 ) , [ 37 ] where @xmath140 is dimensionless constant _
n=2 ^ 2n/(3n-5)()^(2n-4)/(3n-5 ) , _ n=()^1/(n-1 ) , m_n=_i=1^n m_i .
[ 38 ] from the definition ( [ 37 ] ) we have the following asymptotic properties : ^2a_n^2 s , s,[39 ] where @xmath141 is dimensionless constant a_n^2 = ( ) ^2/(3 n-5)()^(2n-4)/(3n-5),[40 ] for example @xmath142 and ^22_n(-m_n),m_n.[41 ]
let us come back to eq .
( [ 13 ] ) and remind an ancient pythagoras theorem stated that the sum of first @xmath143 odd numbers beginning from unity is equal exactly to the square of @xmath143 i.e. _ n = n^2.[pyth ] this pythagoras theorem can easily be proved with the help of the formula for an arithmetical progression .
however , pythagoras theorem can be proved without a knowledge of the formula for an arithmetical progression but using only some remarkable observations in a game with the numbers .
we will not touch here the simplest proof , we would only like to stress a deep link between the froissart bound and this pythagoras theorem . of course , to take advantage of this link we have to learn apart from differential calculus and integral calculus that : * symmetry properties of the space - time continuum are described by inhomogeneous lorentz group or poincar group .
we had also to know how to construct the unitary representations of this group as well , as it was made in the fundamental paper of wigner @xcite .
* there is a very deep connexion between general physical principles such as causality , spectrality , unitarity and analytic properties of physical scattering amplitudes .
the very essence of this connexion is expressed by brilliant jost - lehmann - dyson representation which provided the fundamental results of lehmann . *
it takes many other attainments and the knowledge acquisitions as well .
there is a generalization of pythagoras theorem ( [ pyth ] ) .
really , let us consider any polynomial @xmath144 degree of @xmath87 @xmath145 let @xmath146 be a sum of the polynomial values when the argument @xmath147 takes an integer value @xmath148 then it can be proved that @xmath146 is also a polynomial @xmath149 in @xmath143 degree of @xmath150 s(n)=q_n+1(n),q_n+1(x)=c_0(q)+c_1(q)x+c_2(q)x^2++c_n+1(q)x^n+1,[pythag ] and there is correspondence between @xmath151 and @xmath152 : @xmath153 for example , if @xmath154 is polynomial of fourth degree then we have c_5(q)=c_4(p)/5 , + c_4(q)=c_3(p)/4+c_4(p)/2 , + c_3(q)=c_2(p)/3+c_3(p)/2+c_4(p)/3 , + c_2(q)=c_1(p)/2+c_2(p)/2+c_3(p)/4 , + c_1(q)=c_0(p)+c_1(p)/2+c_2(p)/6-c_4(p)/30
. + c_0(q)=c_0(p).[p4 ] we will call that statement as a generalized pythagoras theorem . it can easily be seen that usual pythagoras theorem ( [ pyth ] ) corresponds to @xmath155 . from eq .
( [ 28 ] ) it s clear that the generalized froissart theorem is related to the generalized pythagoras theorem where @xmath156 is being used . in according with the theory held by pythagoras the unitarity bounds ( [ 17 ] ) and ( [ 33 ] )
give form to the unlimited and therefore they are limit ; see introduction .
recently @xcite a simple theoretical formula describing the global structure of @xmath157 and @xmath158 total cross - sections in the whole range of energies available up today has been derived by an application of single - time formalism in qft and general theorems a l froissart .
the fit to the experimental data with the formula was made , and it was shown that there is a very good correspondence of the theoretical formula to the existing experimental data obtained at the accelerators .
moreover , it turned out there is a very good correspondence of the theory to all existing cosmic ray experimental data as well @xcite .
the predicted values for @xmath159 obtained from theoretical description of all existing accelerators data are completely compatible with the values obtained from cosmic ray experiments . the global structure of ( anti)proton - proton total cross section is shown in figs . 1 - 2 extracted from papers @xcite . ( 188,200 ) ( -35,0 ) ( 92,-15)@xmath160 ( -55,65 ) ( 288,200 ) ( 15,10 ) ( 144,0)@xmath160 ( -5,95 ) the theoretical formula describing the global structure of ( anti)proton - proton total cross section has the following structure _ ( |p)pp^tot(s ) = ^tot_asmpt(s ) , [ 42 ] where ^tot_asmpt(s ) = 2= ( mb),[43 ] @xmath161(gev^{-2}),\ ] ] @xmath162(gev^{-2 } ) = \ ] ] = ( gev^-2),[44 ] @xmath163 @xmath164 is the slope of nucleon - nucleon differential elastic scattering cross section , @xmath71 is the effective radius of two - nucleon forces , @xmath165 is the effective radius of three - nucleon forces , @xmath166 characterizes the internucleon distance in a deuteron , the functions @xmath167 describe low - energy parts of ( anti)proton - proton total cross sections and asymptotically tend to zero at @xmath168 ( see details in the original paper @xcite ) .
the mathematical structure of the formula ( [ 42 ] ) is very simple and physically transparent : the total cross section is represented in a factorized form .
one factor describes high energy asymptotics of total cross section and it has the universal energy dependence predicted by the general froissart theorem in local quantum field theory .
the other factor is responsible for the behaviour of total cross section at low energies and it has a complicated resonance structure .
however this factor has also the universal asymptotics at elastic threshold .
it is a remarkable fact that the low energy part of total cross section has been derived by application of the generalized froissart theorem for a three - body forces scattering amplitude .
( [ 43 ] ) shows that geometrical scaling in a naive form @xmath169 is not valid .
however , from eq .
( [ 43 ] ) it follows the generalized geometrical scaling which looks like ^tot_asmpt(s ) = 2b_el(s)[1 + 2(1-)],[geom - scale ] where @xmath170 is defined above and @xmath171 here , we would like to point out some remarkable features of the global structure in the ( anti)proton - proton total cross sections .
first of all , the ( anti)proton - proton total cross sections have a minimum at @xmath172 , and the question is what this minimum corresponds to .
it turns out that the effective radius of three - nucleon forces at the point @xmath172 satisfys the following harmonic ratio , [ 45 ] where @xmath173 is the proton charge radius . in other words , at the minimum @xmath172 it takes place the _ octave consonance " _ of the three - nucleon forces with the proton charge distribution .
going further on , we have applied our approach to study a shadow dynamics in scattering from deuteron in some details . in this way a new simple formula for the shadow corrections to the total cross - section in scattering from deuteron has been derived and new scaling characteristics with a clear physical interpretation have been established .
we shall briefly sketch the basic results of our analysis of high - energy particle scattering from deuteron .
as has been shown in @xcite , the total cross - section in the scattering from deuteron can be expressed by the formula @xmath174 where @xmath175 are the total cross - sections in scattering from deuteron , proton and neutron , ( s ) = ^el(s ) + ^inel(s)=2^el(s)a^el(x_el ) + 2_sd^ex(s)a^inel(x_inel ) , [ 46 ] @xmath176 @xmath177 the total single diffractive dissociation cross - section @xmath178 is defined by the following equation @xcite _ sd^(s ) = _ m_min^2^s_t_-(m_x^2)^t_+ ( m_x^2 ) dt , [ 47 ] where = ^ex = /2m_n r_d,[ex ] and we supposed that @xmath179 and @xmath180 at high energies . the first term in the r.h.s .
( [ 46 ] ) generalizes the known glauber correction @xmath181 but the second term in the r.h.s . of eq .
( [ 46 ] ) is totally new and comes from the contribution of the three - body forces to the hadron - deuteron total cross section .
the expressions for the shadow corrections have quite a transparent physical meaning , both the elastic @xmath182 and inelastic @xmath183 scaling functions have a clear physical interpretation @xcite .
the function @xmath182 measures out a portion of elastic rescattering events among of all the events during the interaction of an incident particle with a deuteron as a whole , and this function attached to the total probability of elastic interaction of an incident particle with a separate nucleon in a deuteron .
correspondingly , the function @xmath183 measures out a portion of inelastic events of inclusive type among of all the events during the interaction of an incident particle with a deuteron as a whole , and this function attached to the total probability of single diffraction dissociation of an incident particle on a separate nucleon in a deuteron . the scaling variables @xmath184 and @xmath185 have quite a clear physical meaning too .
the dimensionless quantity @xmath184 characterizes the effective distances measured in the units of fundamental length " , which the deuteron size is , in elastic interactions , but the similar quantity @xmath185 characterizes the effective distances measured in the units of the same fundamental length " during inelastic interactions . the functions @xmath182 and @xmath183 have a different behaviour : @xmath182 is a monotonic function while @xmath183 has the maximum at the point @xmath186 where @xmath187 .
the existence of the maximum in the function @xmath183 results an interesting physical effect of weakening the inelastic eclipsing ( screening ) at superhigh energies .
the energy @xmath188 at the maximum of @xmath183 can easily be calculated from the equation @xmath189 and here we faced with the harmonic ratio ( in square ) .[48 ] using the above mentioned global structure for the ( anti)proton - proton total cross sections , we have made a preliminary comparison of the new structure for the shadow corrections in elastic scattering from deuteron with the existing experimental data on proton - deuteron and antiproton - deuteron total cross sections .
the results of this comparison are shown in figs . 3 - 4 ( 298,184 ) ( 20,10 ) ( 155,0)@xmath160 ( 0,90 ) ( 288,184 ) ( 15,10 ) ( 155,0)@xmath160 ( 0,88 ) we would like to emphasize that in the fit to the data on antiproton - deuteron total cross sections @xmath190 was considered as a single free fit parameter .
after that a comparison with the data on proton - deuteron total cross sections has been made without any free parameters : @xmath190 was fixed by the previous fit to the data on antiproton - deuteron total cross sections , and our fit yielded @xmath191 . if we take into account the latest experimental value for the deuteron matter radius @xmath192 @xcite then we can find that the fitted value for the @xmath190 satisfies with a good accuracy the equality r_d^2 = r^2_d , m,(r^2_d , m = 3.853fm^2 = 98.96gev^-2).[rd ] so , we have established a harmonic _ consonance " _ between the internucleon distance in a deuteron and the deuteron matter distribution .[50 ] now , let us come back to eq .
( [ 43 ] ) . taking into account that @xmath193 , from the froissart bound ( [ 17 ] ) and eq .
( [ 43 ] ) we have the following bound .[51 ] on the other hand for the effective radii of @xmath87-body forces we have obtained an asymptotic behaviour given by eq .
( [ 34 ] ) where it follows from = , s.[52 ] bound ( [ 51 ] ) with account of eq .
( [ 52 ] ) gives m_3>m_2=,(m_2=2m_).[54 ] however , if we conjecture that @xmath194 which is fulfilled for @xmath195 then = , s.[55 ] the ratio given by eq .
( [ 55 ] ) corresponds to the harmonic ratio for the major second in the major scale in according to pythagoras tune ; see introduction .
we would like especially to emphasize that the ratio ( [ 55 ] ) is compatible with the global structure of ( anti)proton - proton(deuteron ) total cross sections described above .
in this minireview we have tried in the spirit of the pythagorean school to show the mathematical , physical and geometrical beauty of the froissart theorem .
no doubt , we were enchanted with the aesthetic aspects of the froissart theorem : there were heard _ the new notes of the music of the spheres _ produced by the froissart theorem in the fundamental dynamics of particles and nuclei . starting from abstract mathematical structures of axiomatic quantum field theory by applying the general theorems , a physically transparent intuitively clear and visual picture of particles and nuclei interactions was arisen before our eyes .
we found a very simple relations between physically tangible quantities which looked like pythagoras harmonic ratios mentioned above and hence might be considered as a hadronic symphony " in the fundamental dynamics .
in fact , we came back to the great pythagorean ideas reformulated in terms of the objects living in the microcosmos .
it appears that the study of fundamental processes in high energy elementary particle physics makes it possible to establish a missing link between cosmos and microcosmos , between the great ancient ideas and recent investigations in particle and nuclear physics and to confirm the unity of physical picture of the world .
anyway , we believe in it . at last , in our previous papers we repeatedly criticized the so called supercritical pomeron phenomenology in hadronic physics . in our opinion this phenomenology might be compared with a cacophony " in particle physics .
certainly , someone likes cacophony in the music .
however , we prefer a symphony in the music and a harmony in the fundamental dynamics as well .
it is a great pleasure to thank professor o.a .
khrustalev who initiated , encouraged and supported my scientific flame and research in the youth and professor v.i .
savrin for friendly and successfully collaboration in the middle of seventieth . * * v.v .
belokurov , o.a .
khrustalev , o.d .
timofeevskaya , quantum teleportation
usual wonder , in russian , izhevsk , 2000 .
lebedev , fragments of ancient greek philosophers , part i , in russian , moscow , nauka " , 1989 .
m. froissart , phys . rev . * 123 * , 1053 ( 1961 ) . h. lehmann , nuovo cim .
* 10 * , 579 ( 1958 ) .
r. jost , h. lehmann , nuovo cim . * 5 * , 1598 ( 1957 ) .
dyson , phys .
rev . * 110 * , 1460 ( 1958 ) .
h. lehmann , nuovo cimento suppl .
* 14 * , 153 ( 1959 ) .
a. martin , nuovo cim . * 42a * , 930 ( 1966 ) ; ibid * 44a * , 1219 ( 1966 ) . g. sommer , nuovo cim . *
48a * , 92 ( 1967 ) ; * 52a * , 373 ( 1967 ) ; * 52a * , 850 ( 1967 ) ; * 52a * , 866 ( 1967 ) .
l. durand , p.m. fishbane , l.m .
simmons , jr . , journ . of math
* 17 * , 1933 ( 1976 ) .
wigner , _ on unitary representations of the inhomogeneous lorentz group _ , ann . of math . *
40 * , 149 ( 1939 ) .
khrustalev , v.i .
savrin , preprint ihep 68 - 19k , serpukhov , 1968 .
khrustalev , a.a .
logunov , nguen van hieu , in : _ problems of theoretical physics _ , essays dedicated to n.n .
bogolyubov on the occasion of his sixtieth birthday , publishing house nauka " , moscow , 1969 , p. 90 .
khrustalev , a.a .
logunov , a.n .
tavkhelidze , i.t .
todorov , nuovo cim . * 30 * , 134 ( 1963 ) .
arkhipov , sov .
* 74 * , 69 ( 1988 ) ; ibid * 83 * , 247 ( 1990 ) .
g.t hooft , _ the holographic principle _ , e - print hep - th/0003004 .
arkhipov , v.i .
savrin , sov .
phys . * 49 * , 3 ( 1981 ) .
arkhipov , rep . on math
. phys . * 20 * , 303 ( 1984 ) .
arkhipov , _ what can we learn from the study of single diffractive dissociation at high energies ? _ in proceedings of viiith blois workshop on elastic and diffractive scattering , protvino , russia , june 28july 2 , 1999 , world scientific , singapore , 2000 , pp
. 109 - 118 ; report ihep 99 - 43 , protvino , 1999 ; e - print hep - ph/9909531 .
a.a . arkhipov , _ on global structure of hadronic total cross sections _ , preprint ihep 99 - 45 , protvino , 1999 ; e - print hep - ph/9911533 .
arkhipov , _ proton - proton total cross sections from the window of cosmic ray experiments _ , preprint ihep 2001 - 23 , protvino , 2001 ; e - print hep - ph/0108118 ; in proceedings of ixth blois workshop on elastic and diffractive scattering , pruhonice near prague , czech republic , june 9 - 15 , 2001 , eds . v. kundrat , p. zavada , institute of physics , prague , czech republic , 2002 , pp . 293 - 304 . a.a .
arkhipov , _ three - body forces , single diffraction dissociation and shadow corrections in hadron - deuteron total cross sections _ , preprint ihep 2000 - 59 , protvino , 2000 ; e - print hep - ph/0012349 ; in proceedings of xvth workshop on high energy physics and quantum field theory , tver , russia ,
september 7 - 13 , 2000 , eds .
m. dubinin , v. savrin , institute of nuclear physics , moscow state university , russia , 2001 , pp .
241 - 257 .
arkhipov,_diffraction 2000 : new scaling laws in shadow dynamics _ , nucl .
b ( proc . suppl . )
* 99a * , 72 ( 2001 ) .
f. schmidt - kaler et al . , phys .
lett . * 70 * , 2261 ( 1993 ) . | it has been shown that the great ancient pythagorean ideas have found themselves in the latest researches in high energy elementary particles and nuclear physics . in this respect
we concern and discuss the mathematical , physical and geometrical aspects of the famous froissart theorem and in this way one establishes a link of this theorem to the mathematics and ideas elaborated in the pythagorean school .
a harmony of the froissart theorem in fundamental dynamics of particles and nuclei has been displayed .
we argue that a harmony of the froissart theorem allow us to hear the new notes of _ the music of the spheres " _ just in the pythagoreans sense .
* harmony of the froissart theorem + in fundamental dynamics of particles and nuclei * + a.a .
arkhipov + _ state research center
institute for high energy physics " + 142280 protvino , moscow region , russia
_ + pythagorean watchword |
the properties of a superconducting film or thin wire s are modified when they are in contact with a normal metal n. this phenomenon was first observed in the pioneering experiments by meissner @xcite who explored the properties of superconducting wires covered with normal metals .
it is generally called the `` superconducting proximity effect '' ( spe ) .
it was intensively studied in the 1960 s @xcite , @xcite , @xcite , @xcite , @xcite , @xcite . during the last decade
it has experienced a renewed interest theoretically @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite as well as experimentally @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite .
recently the spe has been extended to sn - multi layers @xcite , @xcite .
a few years ago , our group @xcite investigated the proximity effect between pb and several alkali metals . for a better analysis of these measurements we developed a quantitative numerical method for the calculation of the transition temperature of an sn double layer @xcite .
our numerical results show that , when a superconductor s is covered with a normal metal n , that the initial slope @xmath3 is independent of the mean free paths of the two metals and the transparency of the interface ( if the transmission is not dramatically changed ) . if one defines a normalized initial slope @xmath4 then @xmath5 is independent of the thickness @xmath6 of the superconductor up to relatively large values of @xmath6 .
if the superconductor is very weak coupling ( @xmath7 , @xmath8=debye temperature ) then our result for the initial slope converges towards the results for the thin film cooper limit @xcite ( see below ) .
when we compared our experimental initial slope @xmath9 with our numerical calculation we observed that the experimental results were considerably smaller than the theoretical predictions .
surprised by the discrepancy we searched the literature and found early experiments from the 1960s , particularly by hilsch @xcite , @xcite and minigerode @xcite , from which the normalized initial slope can be derived .
these measurements showed a similar disagreement in the initial slope with the theory ( see ref .
@xcite ) . since we were rather amazed by the discrepancy between our experiments and theory in the spe and also by the fact that this discrepancy had not been detected previously , we decided to re - investigate the spe . in this paper
we investigate the spe in the range , where a minimum of experimental parameters is needed to perform a quantitative comparison with the theory .
we focus on the normalized initial slope @xmath5 of sn sandwiches and the transition temperature of very thin ns sandwiches in the thin film limit .
we use thermal evaporation to condense the thin films onto a substrate at liquid helium temperature . to obtain clean films all the evaporation sources are surrounded with liquid n@xmath10 and the vacuum in our system is better than @xmath11torr . we first evaporate 10 atomic layers of insulating sb on a helium cold quartz substrate .
the sb film acts as a fresh substrate and insures that the following quench condensed films are flat and homogeneous . in a series of experiments a film of the superconductor pb
is first condensed onto the sb substrate .
afterwards the pb is covered in several step with an increasing thickness of the normal metal ag .
the thickness of the films is measured with a quartz oscillator .
the accuracy of the thickness measurement is about 15% .
after each evaporation the superconducting transition curve , the magnetoresistance and the hall effect of the double layer are measured .
fig.1 shows a plot of @xmath0 versus the ag thickness @xmath12 on top of a @xmath13 thick pb film .
this plot yields graphically the initial slope @xmath14 and the normalized initial slope @xmath5 @xmath15 where @xmath16 is the thickness of the pb films and @xmath17 is the transition temperature for pb alone .
@xmath18{f_154_1.eps}}\\ & \begin{tabular } [ c]{l}fig.1 : $ t_{c}$ versus $ d_{ag}$ for an pbag double layer . \end{tabular}\end{aligned}\ ] ]
this experiment is repeated for different thicknesses of the superconductor pb . in fig.2 @xmath5
is plotted versus the pb thickness @xmath19 .
it is essentially independent of the pb thickness .
this was the prediction of our numerical results .
the value of the normalized initial slope is @xmath20 .
@xmath21{c}{\includegraphics [ height=3.0369 in , width=4.0174 in ] { f_154_2.eps}}\\\begin{array } [ c]{c}\begin{array } [ c]{c}\text{fig.2 : the normalized initial slope } s_{sn}\text { for the pb / ag } \\ \text{sandwiches as a function of the pb thickness}\end{array } \end{array } \end{array}\ ] ] @xmath22 in the next experimental series we investigate ns double layers in the thin film limit and express the results in terms of the normalized initial slope @xmath5 . as an example a thin ag film ( @xmath23a ) is condensed onto the insulating sb substrate and covered in several steps with increasing thickness of pb .
fig.3a shows the results .
the inverse @xmath0-reduction @xmath24 is plotted as a function of the pb thickness . from these plots
we extract the value of the normalized initial slope . for the agpb double layer
the value is @xmath25 this is in excellent agreement with the results from the first experimental series .
@xmath26{cc}{\includegraphics [ height=2.7231 in , width=3.1415 in ] { f_154_3a.eps } } & { \includegraphics [ height=2.714 in , width=3.1996 in ] { f_154_3b.eps}}\\ & \\{\includegraphics [ height=2.8269 in , width=3.32 in ] { f_154_3c.eps } } & { \includegraphics [ height=2.8244 in , width=3.2553 in ] { f_154_3d.eps}}\end{array } $ ] @xmath27{l}$]t_c1/ ( t_s - t_c ) d_pb@xmath28@xmath29 @xmath22 we use this much more efficient procedure to investigate sandwiches of pb with the normal metals ag , au , cu , mg .
furthermore we included also as `` normal metals '' superconductors with a transition temperatures @xmath30 which lie below the value of @xmath0 for pb . these metals are zn , cd , al , in and sn .
fig.3b - d shows some of the results for mg , zn and sn .
in table i the experimental data are collected .
the first column gives the experimental code ; which includes the symbols for the normal metal and the superconductor .
the following columns give the thickness of the normal metal ( or the superconductor with the lower @xmath0 ) , the transition temperature @xmath30 of n if n is a superconductor , the ratio of the experimental ( i.e. , phonon enhanced ) density of states and the experimental normalized initial slope @xmath31 the estimated error of @xmath32 is about 15% . the density of states are taken from kittel s book @xcite .
the theoretical prediction of @xmath33the normalized initial slope by the werthamer theory @xmath34 and our numerical result @xmath35will be discussed below .
the transition temperatures for quench condensed cd and zn are taken from ref .
@xcite and @xcite , the transition temperatures of quench condensed al , in and sn are taken from @xcite and @xcite .
@xmath36{|l|l|l|l|l|l|l|l|l|}\hline \textbf{exp . } & $ \mathbf{d}_{n}$(a ) & $ \mathbf{t}_{n}$(k ) & $ \mathbf{n}_{n}^{\ast}\mathbf{/n}_{s}^{\ast}$ & $ \mathbf{s}_{\exp}$ & $ \mathbf{s}_{\text{wh.}}$ & $ \mathbf{s}_{\text{sim.}}$ & $ \mathbf{s}_{\exp}/\mathbf{s}_{\text{sim.}}$ & $ \mathbf{s}_{\text{sim.}}/\left ( \mathbf{n}_{n}^{\ast}\mathbf{/n}_{s}^{\ast}\right )
$ \\\hline agpbjb & 41.2 & 0 & 0.387 & 0.66 & 0.95 & 1.74 & 0.38 & 4.\thinspace5\\\hline cupbje & 32.9 & 0 & 0.603 & 0.98 & 1.49 & 2.67 & 0.37 & 4.\thinspace \allowbreak43\\\hline aupbjd & 29.5 & 0 & 0.442 & 0.62 & 1.09 & 1.98 & 0.31 & 4.\thinspace \allowbreak48\\\hline mgpbjg & 28.5 & 0 & 0.572 & 0.48 & 1.41 & 2.54 & 0.19 & 4.\thinspace \allowbreak44\\\hline cdpbjj & 31.4 & 0.80 & 0.329 & 0.39 & 0.40 & 0.528 & 0.74 & 1.\thinspace \allowbreak60\\\hline znpbuu & 26.0 & 1.39 & 0.430 & 0.33 & 0.44 & 0.54 & 0.61 & 1.\thinspace \allowbreak26\\\hline alpbut & 21.5 & 2.28 & 0.833 & 0.44 & 0.68 & 0.76 & 0.58 & 0.91\\\hline inpbup & 32.1 & 4.1 & 0.663 & 0.30 & 0.31 & 0.354 & 0.85 & 0.53\\\hline snpbun & 33.0 & 4.7 & 0.664 & 0.29 & 0.25 & 0.283 & 1.02 & 0.43\\\hline & & & & & & & & \\\hline \end{tabular } \\ & \begin{tabular } [ c]{l}$\text{table i : the experimental normalized initial slope } s_{\text{exp}}\text { } $ are compared with\\ the theoretical predictions by werthamer 's theory $ s_{\text{wh}}$ and the author's\\ numerical calculations $ s_{\text{sim}}$. the first four columns give the experimental\\ code ( containing the symbols of the normal conductor n and superconductor s),\\ the thickness of n , the transition temperature of n ( if superconducting ) and\\ the ratio of the experimental density of states $ \mathbf{n}_{n}^{\ast } \mathbf{/n}_{s}^{\ast}$ ( which includes\\ electron - phonon enhancement ) .
the second to last column gives the ratio\\ between the experimental $ s_{\text{exp}}$ and the numerical results $ s_{\text{sim}}$. \end{tabular}\end{aligned}\ ] ]
let us consider a double layer with the superconducting transition temperature @xmath0 .
this transition temperature defines a characteristic time @xmath37 . the superconducting coherence length is the distance that an electron propagates during this time @xmath38 . in a clean metal
this is equal to @xmath39 and in a dirty metal one has @xmath40 where @xmath41 is the diffusion constant of the dirty metal .
when , during the time @xmath42 an arbitrary conduction electron propagates through the whole thickness range of the double layer with roughly equal probability then the superconducting properties of the system are averaged over both films . in this case the mean free paths of the individual metal have no bearing on the superconducting transition temperature @xmath43 ( however , the mean free paths determine whether the system is in this limit ) .
a well - known example is the thin film or cooper limit of a double layer when both films are much thinner than their coherence lengths and the transmission time through the interface is much shorter than @xmath44 in the limit that @xmath45is much smaller than the debye temperature , @xmath46 the transition temperature of such a double layer is given by the bcs - cooper formula @xmath47 where the effective bcs interaction @xmath48 of the double layer is @xmath49 and @xmath50 are the bcs interactions in the super- and normal conductor .
using the bcs - cooper formula for an sn double layer yields for the normalized initial slope @xmath51 the predictions of equation ( [ s_cp ] ) for the normalized initial slope ( which is derived from equation ( [ tc_cp ] ) ) has two problems , ( i ) pb is not a superconductor with @xmath52 and therefore the equation ( [ tc_cp ] ) does not yield a good representation of pb , ( ii ) equation ( [ tc_cp ] ) uses the same debye temperature for both metals and this is generally not fulfilled in the experiment .
one can replace the bcs - cooper formula for @xmath0 by an ( implicit ) expression @xmath53 where @xmath54 .
we do not evaluate equation ( [ tc_g ] ) here because its results are in included in our numerical procedure as a limiting case .
werthamer @xcite derived a set of equations for the transition temperature of a double layer of a super- and a normal conductor in the dirty limit , using de gennes interface boundary condition @xcite . in this theory
the gap parameter is approximated as a cosine and hyperbolic cosine function in the superconductor and normal conductor respectively .
( this is sometimes called the single mode expansion ) .
this theory agrees quite well with the experimental results for double layers of two superconductors @xcite .
the normalized initial slope can be expressed as @xmath55 here @xmath56 is the inverse function of @xmath57 and @xmath58 is the digamma function .
if we assume that the transition temperature @xmath30 for the normal metals cu , ag , au and mg is infinitely small , then @xmath59 takes the value one .
the values of @xmath34 according to werthamer s theory are included in table i. the values of @xmath34 do nt show a good agreement for the normal metals cu , ag , au and mg .
finally we compare the experimental values @xmath32 with our numerical calculation .
this numerical calculation derives the transition temperature of a double or multi layer of a superconductor and a normal conductor .
the equivalence in the propagation of the superconducting pair amplitude and a single electron in gorkov s linear gap equation is used .
the single electrons act as messengers which carry the information about the superconducting gap ( @xmath60 ) from one position - time @xmath61 to another position - time @xmath62 .
this message which decays thermally with time as @xmath63 , is integrated at @xmath64 over all starting position - times @xmath65 and , after multiplication with the bcs interaction @xmath66 yields the new gap function @xmath67 . at the transition temperature
the procedure has to be self - consistent , i.e. the initial and final gap function have to be identical .
the propagation of the single electrons is then quasi - classically simulated .
the frame work of the calculation is the weak coupling theory of superconductivity .
this numerical procedure to calculate the transition temperature of double or multi - layers consisting of thin films of superconductors and normal conductors is very flexible .
it uses the following input parameters of the individual metal films ( i ) thickness , ( ii ) density of states and fermi velocity , ( iii ) transition temperature , ( iv ) debye temperature , ( v ) mean free path and ( vi ) transmission through the interface between the films .
an important outcome of the numerical simulation is the result that the normalized initial slope of an sn double layer as a function of @xmath68 at @xmath69 does not depend on ( i ) the mean free paths of the two metals , ( ii ) the thickness of the superconductor and ( iii ) a finite ( but not too large ) barrier between the two metals .
we include the numerical results @xmath70 in table i. for the double layers with the normal conducting metals cu , ag , au and mg the discrepancy between experiment and theory is very large , of the order of 3 . it is remarkable that the deviation is considerably smaller when the normal metal is really a superconductor with a smaller transition temperature although the deviation is still factor of about 1.5 . for in and sn
, however , there is a good agreement between theory and experiment .
it is surprising that this disagreement between experiment and theory has not been noticed in the past .
the main reason is that the majority of the experimental and the theoretical work focused on ns sandwiches with thick normal metal films .
then superconductivity is only obtained for a finite thickness of the superconductor . in this case a comparison between experiment and theory requires many fit parameters such as the transparency of the interface and the mean free paths of the superconductor and the normal conductor .
therefore it is quite possible to fit the experimental data by using the wrong parameters that ca nt be checked otherwise .
the physical origin of this disagreement between experiment and theory is not understood .
our theoretical simulation of the spe uses the frame work of weak coupling superconductivity .
quench condensed pb , in and sn are not weak coupling .
the ratios of @xmath71 for quench condensed films are 4.6 for pb , 3.9 for in and 4.0 for sn @xcite , @xcite
. an obvious proposal would be to solve the superconducting proximity effect for strong coupling superconductors .
this means to develop and solve a series of equations for the energy and position dependent gap function @xmath72
. this would be an extremely demanding job . as a start
, we considered the thin film limit for the spe of strong superconductors .
this consideration , which will be published elsewhere , does not remove the discrepancy between experiment and theory .
the fact that the werthamer result @xmath34 disagrees less with the experimental data @xmath32 for cu , ag , au and mg than our numerical result might be accidental .
the single mode expansion of @xmath67 in the werthamer theory is surely less appropriate than a self - consistant gap function .
the experimental normalized initial slope is proportional to the density of states ratio .
although the density of states can be modified in quench condensed films it is inconceivable that this explains a factor of three in the initial slope .
let us return to fig.1 where the reduction of @xmath0 of pb by a thin layer of ag is plotted .
in this paper , the superconducting proximity effect is investigated for sn double layers in a regime where the resulting @xmath0 does not depend on the mean free path of the films and , within limits , not on the transparency of the interface .
this regime includes the thin film limit and the normalized initial slope @xmath73 .
while the layer s is always pb the layer n is either a non - superconducting metal such as cu , ag , au and mg or a superconductor with a @xmath0 below the transition temperature of pb . the experimental results for the transition temperature @xmath0 are compared with a numerical calculation which was recently developed in our group .
the results for the sn double layers can be divided into three groups : * when n represents a non - superconducting metal film ( n = cu , ag , au and mg ) we observe grave deviations between experiment and theory by a factor of the order of three .
* when n represents a superconductor with a low @xmath0 ( n = cd , zn , al ) the deviation between experiment and theory is still there but reduced by a factor of two .
* when n represents a superconductor with a @xmath0 which is about half the @xmath0 of pb ( n = in , sn ) then we observe a reasonably good agreement between experiment and theory .
prior to our recent experiments we believed that the proximity effect between a superconductor and a normal conductor represented an intensively studied phenomenon with a good theoretical understanding .
we are deeply puzzled by the large observed discrepancy between experiment and theory .
it would be very desirable if other theoretical approaches would give quantitative predictions for the normalized initial slope in sn double layers .
there have been a number of theoretical papers published which extended the proximity effect to more complex systems , for example between a superconductor and a ferromagnet but which include implicitly the simpler case of an sn double layer .
these authors would be able to calculate quantitatively the normalized initial slope from their theory .
experimentally it would be desirable to extend the measurements to sn layers where s is a weak coupling superconductor .
this requires lower temperatures but permits the use of thicker films because the coherence lengths are larger at lower temperatures .
aluminum would be a good candidate for the superconductor if evaporated in ultra - high vacuum so that the al is not granular . | the superconducting proximity effect is investigated for sn double layers in a regime where the resulting transition temperature @xmath0 does not depend on the mean free paths of the films and , within limits , not on the transparency of the interface .
this regime includes the thin film limit and the normalized initial slope @xmath1
@xmath2 the experimental results for @xmath0 are compared with a numerical simulation which was recently developed in our group .
the results for the sn double layers can be devided into three groups : ( i ) when n = cu , ag , au , mg a disagreement between experiment and theory by a factor of the order of three is observed , ( ii ) when n = cd , zn , al the disagreement between experiment and theory is reduced to a factor of about 1.5 , ( iii ) when n = in , sn a reasonably good agreement between experiment and theory is observed . |
high - mass stars are influential in the evolution of galaxies by injecting a large amount of energy of ultraviolet photons , stellar winds and supernova explosions .
it is of fundamental importance to understand the formation mechanism of high - mass stars in order to elucidate structure and evolution of galaxies , and considerable efforts have been devoted to understand the formation mechanism of high - mass stars .
it has been discussed that self - gravitational massive aggregation of molecular gas or protostars are probable sites of high - mass star formation .
the theories of high - mass star formation include the monolithic collapse and the competitive accretion ( for recent reviews see * ? ? ?
* ; * ? ? ?
* ) , and massive and dense clouds including the hot cores and the infrared dark clouds are studied as possible sites of such high - mass star formation ( e.g. , * ? ? ?
it is however puzzling that these massive objects are not commonly associated with hii regions , the most obvious signature of o / early b star formation , and we do not yet have convincing confrontation between the theories and observations @xcite . recently , supersonic collision between the interstellar molecular clouds is discussed as a possible triggering mechanism of high - mass star formation both observationally and theoretically .
up to now we have 8 regions of cloud - cloud collision toward the super star clusters and the hii regions in the milky way disk , and toward the young o stars in the lmc @xcite .
these collisions take place supersonically at a velocity of 1030kms@xmath3 , and the high velocity indicates that the cloud motion is not governed by the cloud - cluster self - gravity .
it is shown by theoretical works that cloud - cloud collision increases the cloud density and the effective sound speed in the shocked interface layer @xcite . according to the recent magnetohydrodynamical numerical simulations for a head - on collision at 20kms@xmath3 by @xcite ,
the mass accretion rate is enhanced by two orders of magnitude to @xmath7@xmath8@xmath9yr@xmath3 due to the collision as compared with the pre - collision state . the mass accretion rate @xmath7@xmath8@xmath9yr@xmath3 satisfies the theoretical requirement to overcome the radiation pressure feedback of the forming o star @xcite .
the typical collision time scale is estimated to be 0.1myr from a ratio of the o star cluster size and velocity , 1pc/10kms@xmath3 , and stellar mass of 10100@xmath9 can be accumulated at the enhanced mass accretion rate in the timescale . currently , a key question is if cloud - cloud collision is one of the major mechanisms of high - mass star formation .
it is premature to conclude that cloud - cloud collision is common because the number of samples is yet too small .
one of the difficulties in studying high - mass star formation is that the number of high - mass star forming regions is small as compared with low - mass star formation sites .
the nearest young o star cluster is the orion nebula cluster ( onc ) including m42 and m43 located at 410pc from the sun ( see for reviews * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the onc has been best studied among the o star clusters at various wavelengths from x rays to radio emissions and is the primary object where theories of high - mass star formation are tested .
the other known regions of high - mass star formation are located at a distance around 2kpc or larger .
the orion a molecular cloud which harbors the onc is an active star forming region , and includes omc(orion molecular cloud)-1 , omc-2 , and omc-3 in the m42m43 region .
a number of molecular line observations were made toward the orion a cloud in the last few decades mainly in the millimeter co emission lines @xcite .
@xmath6100 ob stars and 2000 low mass young stars are distributed in and around the molecular cloud having @xmath10@xmath9 . in particular , the hii region m42 is the most active site of o star formation which includes about ten o / early b stars . on a large scale including the orion b cloud , four ob associations named ori ob1a , ori ob1b , ori ob1c , and ori ob1d are identified @xcite and they may be distributed spatially in age sequence . a model of the sequential formation of the ob associations
was therefore proposed , where ob stars are formed due to triggering by the ionization - shock fronts @xcite .
the orion a cloud appears to be converging toward m42 showing a v shape pointing to the north and a model consisting of self - gravitationally - bound converging motion was presented to explain the shape and formation of the onc @xcite , whereas details of the star formation remain elusive ( see also for low - mass star formation , * ? ? ?
in this paper we present a new analysis of the archival data of @xmath0co(@xmath1=10 ) transition obtained with large single - dish telescopes aiming to test the scenario of cloud - cloud collision as a possible mechanism of high - mass star formation .
the present paper focuses on the distribution and kinematics of the cloud . in section 2
we give a summary of observational and theoretical papers on cloud - cloud collision and present common signatures characteristic to the collision scenario .
section 3 explains an analysis of the co data , and section 4 presents a model of the colliding clouds along with discussion .
section 5 gives discussion from a broader perspective and section 6 concludes the paper .
evidence for triggered formation of o / early b star(s ) by cloud - cloud collision is found in the four super star clusters , westerlund2 , ngc3603 , rcw38 and [ dbs2003]179 ( * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* kuwahara et al .
2016 , in prep . ) , in the two hii regions with single o stars , rcw120 and m20 ( @xcite , 2016 ) , and in the single high - mass stars in the lmc , n159w - south and the papillon - nebula yso of n159e @xcite .
the physical parameters of the 8 objects along with the onc are summarized in table 1 .
the four super star clusters include 1020 o stars with age of 0.14.0myrs , while the others with a single o star have age less than 1.0myr .
theoretical studies of cloud - cloud collision are made by using hydrodynamical numerical simulations @xcite .
these studies allow us to gain an insight into the observational / theoretical signatures of colliding clouds .
it is also notable that synthetic observations present observable characteristics of cloud - cloud collision including the complementary distribution at the two velocities and the bridging feature between the two colliding clouds @xcite .
we shall summarize the major properties obtained by these previous studies in the followings .
the observed cloud - cloud collision indicates the relative velocity between the two clouds is supersonic in a range of 1030kms@xmath3 , except for a few cases where the relative velocity is apparently small probably due to projection ( n159w - south and n159e - papillon ) .
such high velocity is gravitationally unbound by the typical cloud mass within a 10pc radius of the formed o star(s ) , @xmath11@xmath12@xmath9 .
observational verification of heating by o stars / hii regions is used to establish the physical association of the clouds with the formed o stars in spite of their large velocity difference ( e.g. , * ? ? ? * ) which is otherwise interpreted as due to different kinematic distance .
the number density of the molecular clouds in the galactic spiral arms is high enough to cause frequent cloud - cloud collision every 8myr for a giant molecular cloud as demonstrated by numerical simulations @xcite and co observations by nanten ( fukui et al . 2016b , in prep .
the magnetohydrodynamical numerical simulations demonstrate that in the shock - compressed interface layer between the two clouds the density increases by a factor of 10 and the turbulent velocity and alfvn velocity increase by a factor of 5 @xcite .
this is because the shock front is deformed into disordered turbulence by density inhomogeneities of the initial clouds and the turbulence amplifies the field strength .
the mass accretion rate is proportional to the third power of the effective sound speed including the turbulent velocity and alfvn velocity , and the supersonic motion produces a high - mass accretion rate in the order of @xmath7@xmath8@xmath9yr@xmath3 , which is large enough to overcome the stellar radiation pressure and allow mass growth of the star(s ) .
we note that the turbulent velocity field in the interface layer is isotropic , providing a signature of the collision even when the velocity between the colliding clouds becomes small by the projection effect as shown in the interface layer of n159w - south and the papillon nebula in n159e @xcite .
we see in table 1 that the number of forming o stars has a threshold value of molecular column density @xcite .
1020 o star formation takes place for high column density arround @xmath13@xmath14 derived from @xmath0co(@xmath1=10 ) line intensity by using the x@xmath15 factor , whereas formation of a single o star happens for lower column density around @xmath16@xmath14 .
o star(s ) is formed if one of the two clouds , not always both of them , satisfies these conditions .
we suggest that in the high column density case the o star formation takes place in a small path length of the collision , e.g. , @xmath60.1pc ( see rcw38 , * ? ? ?
* ) , and that in the low column density case the path length prior to the o star formation is longer , @xmath61pc ( see rcw120 , * ? ? ? * ) molecular clouds within 110pc of o star formation will be rapidly dispersed by the ionization / winds and may not be observable at a timescale larger than @xmath62myr @xcite , while the details will depend on the number of o stars , i.e. , the ultraviolet radiation field , and the gas density distribution .
collisional interaction also converts the natal gas into the shocked interface layer and stars , and dissipates the natal clouds as discussed further in section 2.3 .
figure 1 shows the radial distribution of the associated molecular emission in the nine regions of o star formation in table 1 .
the super star clusters including 1020 o stars in a volume smaller than 1pc show often a central cavity of 10pc radius as seen in westerlund2 , ngc3603 , and [ dbs2003]179 , whose ages are 24myr .
similar cavities are observed in the other super star clusters including arches , quintuplet , and r136 ( e.g. , * ? ? ?
* ; * ? ? ?
the only exception is rcw38 which shows rich molecular gas at apc scale
. a close look at rcw38 indicates a cavity of 0.5pc radius in the molecular gas created by the ionization , and the small ionization cavity likely reflects the small age 0.1myr @xcite .
these super star clusters suggest that the ionization propagates at a typical velocity of @xmath65kms@xmath3 ( a ratio 0.5pc/0.1myr , or 10pc/2myr ) .
this velocity is consistent with theoretical values ( e.g. , * ? ? ?
we expect that the clouds within 1pc will soon be ionized and dispersed in another 0.1myr , leading to termination of star formation in a few times 0.1myr .
the age spread of the cluster members formed by triggering is therefore short in the order of 0.1myr in super star clusters .
such a small age spread is consistent with that of 0.10.3myr of the young cluster members estimated by careful vlt and hst observations of the two super star clusters ngc3603 and westerlund1 @xcite . among the regions of single o star formation , the three regions m20 , n159w - south and n159e - papillon show rich molecular gas within 1pc and rcw120 shows a molecular cavity of 2pc radius
the size of the molecular cavity in single o star formation seems to be smaller than in the superstar clusters perhaps due to the less ionizing photons . m20
whose age is 0.3myr is extended over 3pc , whereas it also shows a clear cavity of neutral matter with a 0.1pc radius centered on the o star in the hst image ( esa / hubble & nasa , bruno conti ) .
the papillon - nebula yso in n159e shows a molecular cavity of 0.5pc radius , whose inside is filled by the hii region and the age of the papillon - nebula yso is estimated to be 0.2myr @xcite .
n159w - south has no hii region and no molecular cavity , and is driving protostellar outflow with the associated accretion disk ( fukui et al .
n159w - south has a small age of 0.1myr , and is younger than the papillon - nebula yso .
so , the molecular cavities in m20 and the papillon - nebula are likely created by the ionization .
it is suggested that the molecular cavity in rcw120 was created by the cloud - cloud collision as first modeled by @xcite , and the effect of the ionization is not dominant in forming the cavity ( @xcite ; see also for a different scenario @xcite ) . the single o star forming regions in the current sample are fairly young and the effect of the ionization does not seem to be prevailing , whereas the cloud dispersal by ionization will become dominant later .
it is often the case that colliding clouds are not of the same size as simulated by @xcite , @xcite , and @xcite .
if one of the colliding clouds is smaller than the other , the small cloud can create a cavity in the large cloud .
this produces complementary distribution of the two clouds at different velocities , which is used to identify candidates for colliding clouds unless the cloud dispersal by the ionization is significant . in order to visualize the collision
, we adopt the numerical simulations by @xcite , and table 2 lists the simulation parameters .
a schematic is shown in figure 2(a ) . in the model where the two clouds are spherical ,
the cross section of the cavity is determined by the size of the small cloud and the length of the cavity by the travel distance since the initiation of the collision .
the cavity is identified observationally as an intensity depression in the molecular distribution at the velocity of the large cloud .
the projected distributions of the small cloud and the intensity depression are complementary , and they are in general displaced with each other because the angle of the relative motion to the line of sight , @xmath17 , is not 0deg .
( figure 5 ) . while the real cloud shape can be more complicated , it is still worthwhile to use the simple two - sphere model to gain an insight into the cloud distribution created by the collision .
the relative motion of the colliding clouds generally has an inclination angle to the line of sight , and three cases of the inclination angle @xmath17 , 0deg .
, 45deg . , and 90deg .
, are shown in the velocity channel distributions in figures 35 . for @xmath17=0deg .
( figure 3 ) we see two velocity features whose distribution varies with velocity .
we identify the initial two clouds at @xmath18@xmath19kms@xmath3 ( panels ( e)(g ) in figure 3 ) and @xmath20@xmath21kms@xmath3 ( panels ( i)(k ) in figure 3 ) and the intermediate velocity feature mainly at @xmath19@xmath20kms@xmath3 ( panels ( g)-(i ) in figure 3 ) .
the small cloud fits well with the cavity created in the large cloud by collision , showing the complementary distribution between them ( figure 3(l ) ) . for @xmath17=45deg .
( figure 4 ) we see the two clouds at @xmath22@xmath23kms@xmath3 ( panels ( f)(g ) in figure 4 ) and @xmath20@xmath21kms@xmath3 ( panels ( i)(j ) in figure 4 ) and the intermediate velocity feature mainly at @xmath23@xmath20kms@xmath3 ( panels ( h ) in figure 4 ) . in figure 4
we see a displacement of the complementary distribution between the cavity and the small cloud , which is a natural result of projection for @xmath17=45deg . if the relative motion is vertical to the line of sight @xmath17=90deg .
( figure 4 ) , we do not see the two velocity components .
all these simulations show only single - peaked line profiles with skew in velocity ( figures 3(l ) and 4(l ) ) , where the two clouds seem to be continuous due to the @xmath24bridging@xmath25 feature ( section 2.4 ) .
figures 35 illustrate that , even if we see only a singly - peaked velocity component in observations , a possibility of cloud - cloud collision is not excluded . in regions of single o star formation
, the most significant complementary distribution toward an o star at a 0.1pc scale is found in m20 ( torii et al .
2016 , in prep . ) .
in rcw120 the dispersal of the small cloud is significant and only the large cloud with the intensity depression , one of the complementary clouds , is seen along with the remnant of the small cloud outside the `` bubble '' @xcite . in n159 it is possible that the angular resolution of alma , [email protected] , is not high enough to resolve the complementary distribution which is likely to be in a scale of 0.1pc . in the super star clusters where the ionization is more significant , it is probable that the complementary distribution is dispersed more quickly . even in the youngest cluster rcw38
we see the ionization is dominant in shaping the clouds @xcite .
we note that the oldest cluster [ dbs2003]179 among the four shows a large - scale complementary distribution at a scale more than 10pc , which is not affected much by the ionization ( kuwahara et al .
2016 , in prep . ) . to summarize , the colliding clouds with small age and relatively low ionization are the most probable site to find the complementary distribution , whereas it is possible that the complementary distribution is often dispersed by the stellar feedback and the collisional interaction , depending on the cloud distribution .
another possible signature of cloud - cloud collision is seen in the optical / near - infrared extinction toward the hii region . in cloud - cloud collision
the blue - shifted cloud is always on the far - side of the red - shifted cloud prior to the collision but moves to the near - side after the collision . as a result
, one may expect that only the blue - shifted cloud , not the red - shifted cloud , is located in front of the hii region and can create a dark heavily obscured feature .
such dark lanes are in fact observed in m20 where the collision took place @xmath60.3myr ago to form a single o star ; the dark lanes toward m20 show a good correspondence with the blue - shifted cloud @xcite .
a caveat is that this may not always be the case if the collision happened very recently where the line - of - sight separation between the two clouds is still small .
the super star cluster rcw38 shows a narrow dark lane in the near infrared as irs1 , whereas irs1 is identified as part of the red - shifted cloud @xcite .
the blue - shifted cloud in rcw38 also corresponds to the near - infrared dark features surrounding the cluster .
these authors interpret that irs1 remains neutral without full ionization due to the small age of the system ; the o star formation occurred only 0.1myr ago and part of the red - shifted cloud within 1pc of the cluster center is still surviving in front of or inside the hii region .
so , the dark lanes can arise from the both clouds depending on the age and shape of the clouds , and we need to be cautious in examining the extinction features when the collision is a very recent event .
numerical simulations show that the two colliding clouds become connected in velocity by a bridging feature in synthetic observations if the projected velocity separation is larger than the linewidths of the clouds @xcite .
the shocked interface layer appears as the bridging feature connecting the clouds in the position - velocity plots as shown in figures 3 and 4 .
the bridging features are observable in the spots of collision , although all of them are not necessarily related to o star formation as shown later . in table 1
we see that only the youngest two cases , rcw38 and m20 , show the bridging feature toward the o star formation .
rcw38 shows the bridging feature within 0.5pc of the @xmath620 o star candidates , where the small colliding cloud is localized @xcite .
the connection among the three features [ the bridging feature , @xmath620 o stars , and the small colliding cloud ] is tight ( see figure 3 in * ? ? ?
m20 shows three bridging features and one of them is found toward the o star ( torii et al .
2016 , in prep . ) .
the three super star clusters , ngc3603 , westerlund2 and [ dbs2003]179 , show the bridging features in the outer regions of more than 10pc radius of the cluster , where molecular gas still remains .
rcw120 also shows the bridging features outside the hii region in co and hi @xcite .
n159w - south and n159e - papillon have a small projected velocity separation of a fewkms@xmath3 and the bridging features are not seen .
to summarize , the main observational signatures of o star formation by cloud - cloud collision are listed as follows ; 1 . the supersonic velocity separation of the two clouds , 2 . the complementary distribution of the two velocity components , and 3 .
the bridging feature connecting the two clouds .
we note that the cloud dispersal by the ionization and the collisional merging of the clouds , in addition to the projection effect , can weaken or extinguish these signatures , and one is required to carefully examine these effects which can smear out the collision signatures .
even if a cloud - cloud collision happens , it is possible that one sees none of the signatures . in order to catch the signatures observationally
it is required that the system is young , without significant dispersal of the natal gas , and that the projection effect is small . in more than a fewmyrs after the o star formation by collision , it probably becomes impossible to find any of the observable signatures , for instance , as in case of the relatively young cluster ngc6231 with an age of 35myr ( baume et al .
1999 ) which has no molecular gas within 30pc of the center ( fukui et al . 2016c , in prep .
the @xmath0co(@xmath1=10 ) distributions in the m42m43 region obtained with the nro 45-m telescope are presented and discussed by @xcite .
the observational parameters are listed in table 3 .
the paper shows that the orion a molecular cloud is singly peaked in velocity , which is consistent with the previous co results summarized in section 1 .
the cloud properties created in cloud - cloud collision may however show no obvious double component ( section 2 ) , and it is worthwhile to make a thorough investigation of the kinematic properties in the m42m43 region from a view point of the cloud - cloud collision scenario .
the numerical simulations of the cloud - cloud collision in section 2 assumed physical parameters which do not exactly coincide with the observed properties of the present cloud .
nonetheless , it is conceivable that the basic characteristics of the colliding clouds hold for different physical parameters , and we here consider the qualitative properties presented in section 2 . in the followings , we apply an x@xmath15 factor of 2@xmath2610@xmath27 ( kkms@xmath3)@xmath3@xmath14 @xcite to convert the @xmath0co(@xmath1=10 ) intensity into the molecular column density .
figure 6 shows the dec .- velocity diagram covering @xmath61deg in dec .
integrated in a r.a .
range from @xmath28 to @xmath29 , and figure 7 shows velocity - channel distributions in a region of @xmath60.5deg ( r.a . ) by @xmath61deg ( dec . )
centered on @xmath4ori .
although most ( 75% ) of the co emission is apparently seen in a velocity range of 613kms@xmath3 as a single component , in figure 7 we find the distribution of the @xmath0co emission is significantly different between the two velocity ranges 5.39.8kms@xmath3 ( panels ( b ) , ( c ) , and ( d ) ) and 11.414.4kms@xmath3 ( panels ( f ) and ( g ) ) .
the strongest peak of the @xmath0co emission is seen at 9kms@xmath3 toward @xmath4ori where the @xmath0co linewidth is enhanced from 5kms@xmath3 to 14kms@xmath3 at a 1 kkms@xmath3 level toward the onc .
this main peak having a size of @xmath61pc@xmath261pc is called the omc-1 clump in the followings .
we see that the blue - shifted emission at @xmath68kms@xmath3 , which constitutes the integral shape of the cloud ( mainly in panel ( d ) of figure 7 , * ? ? ?
* ) , is extended on the northern and southern sides of the omc-1 clump , whereas the red - shifted component at @xmath612kms@xmath3 ( mainly in panel ( f ) of figure 7 ) has a sharp intensity decrease toward the south at dec.=@xmath30 .
we see the trend also in figure 6 .
these different spatial distributions at the two velocity ranges suggest a possibility that the two clouds are overlapping as modeled later in section 4 .
the intermediate velocity range 1011kms@xmath3 ( panel ( e ) of figure 7 ) shows mixed morphology of the two velocity components .
based on the empirical signatures of cloud - cloud collision ( section 2 ) , we searched the @xmath0co data for complementary distribution and found three possible pairs .
figure 8 shows the three complementary distributions at the two velocity ranges in panels ( a ) , ( b ) and ( c ) .
figure 8(a ) shows the distribution toward m42 , and figure 8(b ) toward m43 and omc-2 .
figure 8(c ) is the distribution toward omc-3 , where no hii region is associated ( e.g. , * ? ? ?
* ) . in figure 8(a
) , we define the omc-1 clump in the blue - shifted cloud at the contour level of 15 kkms@xmath3 .
we find the u - shaped cloud in the red - shifted cloud appears to surround the omc-1 clump , and define the cloud at the contour level of 2 kkms@xmath3 .
the defined levels are used for calculating the cloud mass . in figure 8(b ) and 8(c ) we find that the red - shifted component shows complementary distributions to the blue - shifted component as detailed below . in order to show the complementary distributions clearly figure 9 overlays the two velocity components in the three regions with appropriate displacements as indicated by arrows in figures 9(b ) and ( c ) , whereas no displacement is applied in figure 9(a ) . in figure 9(a ) , the complementary distribution is clear in the two components .
the omc-1 clump is extended by more than 1pc only toward the north - west at an intensity level of @xmath615 kkms@xmath3 ( figure 8) .
it is possible that this extension forms a complementary pair with the intensity depression in the red - shifted component toward ( r.a . ,
dec.)=(@xmath31 , @xmath32 ) ( panel ( f ) of figure 7 ) , although the diffuse nature of the extension makes the distribution less evident than toward omc-1 .
we note that the intensity depression in the south of the omc-1 clump in the blue - shifted component , which is extended from r.a.=@xmath33 to @xmath29 at dec.=@xmath34 , shows also a possible complementary distribution to the red - shifted component in the same direction .
further , in the south of the omc-1 clump we see an intensity depression elongated in the east to west ( r.a.=@xmath35@xmath36 , dec.=@xmath37@xmath38 ) in the blue - shifted component ( figure 10(a ) ) .
this depression seems similar in shape to the southern part of the complementary distribution of the red - shifted component , and possibly represents another sign of complementary distribution .
the depression appears to be shifted to the south as compared with the red - shifted component .
a displacement similar to figure 9(b ) may be applicable , while the large size of the omc-1 clump @xmath61pc tends to complicate the complementary correspondence more than in figure 9(b ) . in order to explain the estimate of the displacement in figure 9(b ) , we show in figure 10 , the upper panels , the inner part of the two @xmath0co components toward the central part of m42 and m43 in the red- and blue - shifted velocity ranges .
figure 10 also shows enlarged @xmath0co images toward m43 in the lower two panels .
the blue - shifted cloud shows an intensity depression toward the exciting star nuori , which is elongated in the north - south by @xmath60.5pc with a width of @xmath60.2pc as enclosed by a box in figure 8(c ) .
the eastern side of the depression is bound by a filamentary feature that corresponds to the dark lane in m43 ( see also figure 12 ) . a corresponding emission feature to the intensity depression
is seen in the red - shifted component which has a similar shape to the depression .
we call the two features as the @xmath24orion keyhole@xmath25 and the @xmath24orion key@xmath25 in order to avoid confusion with the keyhole nebula in carina
. we applied the method to estimate the displacement in the complementary distribution which is presented in the appendix .
in fitting the displacement , we restricted the area of the analysis to the region of the orion keyhole and the orion key as shown in figure 11 ; in the north of the orion keyhole , the possible intensity depression is not closed , making it difficult to apply the method . by inspection
we find that the orion key has a discontinuous @xmath24bent@xmath25 in the middle at @xmath60.25pc from the southern edge , which seems to morphologically correspond to the edge of the eastern filament of the orion keyhole ( figure 8) .
we then chose the direction of the displacement at a position angle of 108deg . so that the two features coincide spatially after the displacement .
figure 11 shows the overlapping function @xmath39 in pc@xmath40 defined in the appendix and indicates that @xmath41=0.3pc gives the optimum fit .
we confirmed that the present position angle adopted gives the best fit , i.e. , the maximum value of @xmath39 , by changing the angle in a range 103113deg . a displacement of 0.3pc shown by the arrow in figure 9(b ) gives the fit between the orion keyhole and key . by the displacement
, we also see that the western edge of the red - shifted component shows good correspondence with the eastern edge of the blue - shifted component over @xmath61pc in the north - south , which lends support for the displacement . in figure 9(c ) , while not so clear as in the other two cases , the two peaks of the blue - shifted component at ( r.a .
, dec)=(@xmath42 , @xmath43 ) and ( @xmath44 , @xmath45 ) are located toward the intensity depression of the red - shifted component , forming a possible pair of complementary distribution , after a displacement of 0.1pc shown by an arrow .
this displacement was chosen based on the eye inspection because the distribution is relatively simple .
figures 12 and 13 show overlays of the two velocity components on the optical image .
the most notable correspondence is seen toward m43 where the thin dark lane in the center of the nebula coincides with the co filament in the east of the orion keyhole ( figure 12(a ) ) .
the other blue - shifted features of @xmath60.5deg in length show coincidence with the dark areas in the northeast of m42 .
this correspondence is consistent with that the blue - shifted component lies on the nearside of the optical nebula .
the red - shifted component in figure 13 shows no correspondence with the optical features , suggesting that the red - shifted component lies on the far side or the inside of the nebula .
we remark that the orion key has no corresponding dark lane in m43 .
we present a model of the two molecular clouds which are colliding to trigger the formation of the o / b stars in m42 and m43 .
table 5 lists the main physical parameters of the model clouds and figures 14 gives a schematic of the clouds before and after the collision .
the model consists of two clouds of projected velocities peaked at @xmath68kms@xmath3 and @xmath613kms@xmath3 , while the actual relative velocity between the two clouds is in the order of 10kms@xmath3 .
both of the clouds are elongated along the galactic plane .
the mass of the blue - shifted component is dominant ( @xmath680% ) toward omc-1 , explaining the apparent single peak , while the border velocity 11kms@xmath3 between the two model clouds may have some uncertainty ; i.e. , a border velocity between 10kms@xmath3 and 11kms@xmath3 may not be excluded .
the total mass of the two clouds @xmath610@xmath46@xmath9 is too small by an order of magnitude to gravitationally bind the estimated relative velocity @xmath67kms@xmath3 .
the model clouds have an overlap toward the omc-1 clump at dec.=@xmath47@xmath48 which corresponds to the region of the enhanced velocity span in figure 6 .
the enhancement is interpreted as due to the turbulence by the collisional interaction @xcite , while the effect of the protostellar outflow may also be in part responsible for the enhancement .
the model offers an explanation of the complementary distribution which is found the present cloud .
the collisional interaction is seen at least in the three regions as indicated by the complementary spatial distribution ; one is toward m42 and triggered formation of @xmath4ori and @xmath5ori ( figure 9(c ) ) , and another toward m43 , which triggered formation of nuori ( figure 9(b ) ) .
the other is toward the north of m43 without an hii region ( figure 9(a ) ) .
two more regions associated with possible complementary distribution are noted in section 3 , while no clear star formation is seen except for m42 and m43 . in the present case ,
the half - power linewidths of the two clouds are 23kms@xmath3 and the projected velocity separation @xmath65kms@xmath3 is too small for observing the bridging feature . in the model , @xmath4ori and @xmath5ori ,
including nearly ten o stars , were formed on the near side of the blue - shifted cloud which was collided and shock - compressed by the more extended red - shifted cloud .
this collision created the cavity in the red - shifted cloud following the cloud - cloud collision model @xcite .
the interface layer between the two clouds became strongly - compressed and highly turbulent to form o stars as shown by @xcite .
the o stars formed on the nearside of the dense blue - shifted cloud ionized the nearside of the cloud to form the m42 nebula as exposed to us .
nuori was formed in the collision between the orion key and the red - shifted component having the orion keyhole .
nuori is a b3 star and ionized m43 ( see section 4.3 for more details ) .
the ionization possibly has widened the size of the orion keyhole , whereas the smaller uv photon flux of nuori than @xmath4ori and @xmath5ori is not powerful enough to dissipate the orion keyhole at present .
the timescale of the collision is approximately estimated from a ratio of the cloud size and the relative velocity between the two clouds . if we assume
tentatively the relative motion between the two clouds has an angle of 45deg to the line of sight , the relative velocity corrected for the projection is estimated to be @xmath67kms@xmath3 , which is consistent with the enhanced @xmath0co broadening toward the onc ( figure 6 ) .
the apparent spatial extent of the colliding clouds forming o / b stars ranges from @xmath60.3pc to @xmath60.8pc ( figure 8) .
the ratio between the timescale and the cloud size yields roughly 0.050.11myr as the collision timescale and poses a constraint on the upper limit for the age of the stars formed by triggering .
the complementary distribution is characterized by the projected displacement between the colliding clouds ( figure 3 , see also the appendix ) .
the displacement is small for a small collision timescale .
also , the displacement @xmath49 to the size of the small cloud becomes smaller , if the size of the small cloud in the collision model is large .
figure 10 shows different displacements between the two velocity components among the three regions .
we apply a detailed analysis of the timescale to the orion key and keyhole , where information is obtained on their relative location by the analysis in section 3 thanks to the small size , @xmath60.2pc , of the small cloud .
the separation between the two features are estimated to be 0.3pc at a relative velocity is 4kms@xmath3 .
if we assume 45deg as the angle of the relative motion to the line of sight , the time scale of the collision is estimated to be 0.4pc/5.7kms@xmath3=0.07myr if no deceleration of the relative motion is assumed .
the northern region in figure 9(c ) shows a displacement of @xmath60.1pc , while it may be less accurate than in figure 9(b ) .
the omc-1 clump in figure 9(a ) shows no appreciable displacement , whereas the larger clump size , @xmath61pc , may tend to smear out a possible displacement .
the separation between the two velocity components may vary from region to region reflecting the initial distribution of the two components , leading to different collision parameters . in summary , we conclude that the typical collision timescale is @xmath60.1myr with some regional minor variation . at the present epoch ,
the blue - shifted component has high peak column density toward the peak of the omc-1 clump , @xmath62@xmath2610@xmath50@xmath14 , while the red - shifted cloud significantly smaller column density , @xmath610@xmath51@xmath14 . in m43 ,
the typical column density in the orion key and keyhole is @xmath610@xmath51@xmath14 .
we assume that the initial conditions of the two clouds are similar to those of the present clouds , whereas it is probable that the present column densities have become somewhat lower than the initial values due to the collisional interaction and ionization by the formed stars . by comparing o star formation in several regions ,
@xcite argue that the number of o stars formed by collisional triggering depends on the initial column density ; more than ten o stars are formed for a threshold column density @xmath610@xmath50@xmath14 and a single o star for a threshold column density @xmath610@xmath51@xmath14 .
the difference in the o / b star formation in m42 and m43 is consistent with the suggestion .
the mhd numerical simulations by @xcite have shown that the interface layer between the colliding clouds become denser and more turbulent with amplified magnetic field , realizing the high mass accretion rate @xmath610@xmath5210@xmath53@xmath9yr@xmath3 which satisfies the formation of the high - mass stars greater than 20@xmath9 . in the present typical collision timescale 0.1myr , the stellar mass which is attained by mass accretion is 10100@xmath9 .
so , the present short timescale is consistent with the o star formation . more details of the star formation may be explored . in m42 ,
the ionized gas has higher density in @xmath4ori than in @xmath5ori .
there are even younger heavily - embedded high - mass protostars showing outflows , the bn / kl object and the orion - s region , in the west of @xmath4ori .
we see an age sequence from the east to the west , where @xmath5ori is the oldest and the bn / kl object and the orion - s region the youngest .
a possible scenario is that the age sequence is caused by the time sequence of the collision ; i.e. , the colliding two components has a spatial gradient in the sense that the western part collided first and then the collision is propagating to the west , and the most recent triggering formed the bn / kl object and the orion - s region .
we speculate that this apparent time sequence from the east to the west is due to a gradient in the separation between the colliding surfaces of the two clouds in the sense that the western part collided later than the eastern side .
a similar age gradient is seen in rcw38 ; the eastern part is already ionized , whereas the western part is more heavily embedded @xcite . in m43 ,
the position of nuori is separated to the west by @xmath60.15pc from the eastern filament of the orion keyhole .
the velocity of the interface layer generally becomes smaller by the momentum conservation ( e.g. , * ? ? ?
* ) , which places the formed star behind the small cloud .
a possible scenario is that the b star formation took place by the collision in the eastern edge of the orion key .
this explains the location of the nuori and the situation is similar to m20 where a single o star is formed by triggering in the eastern edges of the clouds ( torii et al .
2016 , in prep . ) .
the age of the onc members ranges from less than 0.1myr to more than 1myr while the majority is in the range 0.11myr as estimated in the hr diagram by comparing with the theoretical stellar evolutionary tracks @xcite .
the collision timescale less than 0.1myr corresponds to the youngest end of the age distribution .
we infer that the blue - shifted cloud which dominates the molecular column density in the orion a cloud was forming low - mass stars of the onc prior to the collision in the last 1myr .
the distribution of low - mass young stars of the onc , 2mass variables stars in figures 15 and 16 @xcite , indicates that these stars toward the omc-1 clump are correlated well with the column density distribution of the blue - shifted cloud except for the @xmath0co peak including the bn / kl object where the extinction may be large at k band @xcite .
these low - mass members are spatially more extended than the o stars @xcite , beyond the region of the collisional interaction .
we see little molecular gas in the two regions in the north of m42 , ( r.a .
, dec.)=(@xmath54@xmath55 , @xmath32@xmath43 ) , and the south of the m42 clump ( r.a . , dec.)=(@xmath56@xmath57 , @xmath58@xmath34 ) ,
whereas 2mass stars are distributed there .
the two correspond to the regions of the suggested collisional interaction and the blue - shifted gas in the regions has possibly been removed by the collision after the low - mass star formation .
the red - shifted cloud had lower column density , not actively forming stars before the collision , while the northern part of the red - shifted component may have been forming part of the stars as indicated by the spatial correspondence with the stars ( figure 15 ) .
it is possible that not only o / b stars but also low mass - stars were formed by the collisional triggering ( cf . * ? ? ?
the young low - mass stars in the hr diagram by @xcite suggest very young low - mass stars of an age less than 0.1myr .
part of them might represent protostars whose accretion disk was destructed by the uv photons of @xmath4ori and @xmath5ori before they reach @xmath5910@xmath9 .
the monolithic collapse and the competitive accretion are the two scenarios for massive cluster formation and have been discussed for more than a decade ( e.g. , * ? ? ?
* ) . in either case
the massive gas / star system is self - gravitationally bound and multiple sources evolve into stars of various masses by mass accretion .
it is also discussed that the bound system may evolve in time and becomes unbound by gas dispersal ( e.g. , * ? ? ?
these two gravitationally - bound scenarios have no explicit physical mechanism of realizing the high - mass accretion rate like the shock - induced turbulence in the cloud - cloud collision scenario , and it is assumed that the gas achieves the high - mass accretion rate by some gravitationally driven mechanism .
the colliding clouds in the present hypothesis are not gravitationally bound by the supersonic velocity and the collision increases the mass accretion rate to be @xmath7@xmath8@xmath9yr@xmath3 , which is higher than in the bound scenarios , favoring the high - mass star formation @xcite . the evolution of the monolithic mass aggregation has been tested theoretically . in the simulations of proto - clusters , a single gravitationally
bound n - body system is adopted as the initial condition of a cluster whose number of stars are assumed to be 800 2000 in the same order with that of the onc ( e.g. , * ? ? ?
* ; * ? ? ?
the effect of the stellar feedback is also incorporated in recent simulations ( e.g. , * ? ? ?
the typical timescale of the evolution of the n - body system is myrs , which is significantly larger than that of the cloud - cloud collisionless than 0.1myr .
it is probable that these simulations are applicable to the pre - collision cluster of low - mass stars in the orion blue - shifted cloud of the present hypothesis .
some n - body simulations assume @xmath24rapid@xmath25 gas dispersal by ionization due to the high - mass stars , and it is argued that such rapid dispersal is consistent with the observations @xcite .
the rapid dispersal is however yet of a large time scale , close to myrs .
there is increasing observational evidence for a short time scale less than 0.1myr in o star formation ; observations of the onc shows that the youngest stars have age less than 0.1myr , and in the two super star clusters ngc3603 and westerlund1 the age spread of young clusters is short in the order of 0.1myr @xcite .
it is also becoming established that the age and mass segregation is outstanding in young massive clusters ; the o stars are concentrated in the inner most part of ngc3603 , the onc , and r136 .
while the stellar mass function depends on the area for averaging the stellar properties , the innermost part tends to have a top - heavy mass function in these clusters .
we need to be cautious that the onc may consist of at least two different populations .
the onc has a usual mass function similar to the field initial mass function imf having a salpeter - like slope @xcite . in the present model the stars toward m42
is a superposition of two different populations ; the older low - mass members formed prior to the collision are likely dominant in the blue - shifted cloud and the other youngest stars including o / early b stars were formed in triggering by the collision , which are not dominant in mass .
the pre - existent low - mass stars dominate the stars of the onc , making the steep slope of the mass function , while about a half of them may still be deeply embedded in the blue - shifted cloud @xcite .
one may expect that the imf has discontinuous slopes if the star formation by the two different modes are working .
we suggest that the total mass of the o stars in the center of the onc is @xmath6150@xmath9 @xcite . on the other hand ,
the molecular mass of the omc-1 clump is 3.3@xmath2610@xmath60@xmath9 . if we assume tentatively that the depth where of the shock front propagated in the omc-1 clump is 0.3pc , one third of the clump size , the molecular mass already shocked is estimated to be @xmath61000@xmath9 , large enough to form the ten o stars of 2030@xmath9 .
the present collision model suggests that the forming stars by triggering are located on the front side of the m42 clump , producing a non - spherically symmetric distribution of the cluster .
this is a natural consequence of the collision that has a directivity at least in the early phase where part of the molecular gas still remains without ionization .
we note that the n - body simulations result in spherically symmetric system after myrs do not explain such asymmetry .
the total mass of the orion key and keyhole region is estimated to be @xmath6150@xmath9 for a size of @xmath60.5pc .
this is large enough to form nuori with 17@xmath9 @xcite .
the initial conditions of the present cloud - cloud collision assume a molecular cloud having high column density 10@xmath50@xmath14 , which has been continuously forming low - mass stars .
these low - mass stars are not able to ionize the cloud and the blue - shifted cloud continues to form low - mass stars until it forms high - mass stars which are capable of ionizing the cloud .
the o star formation is suddenly triggered by the impact of the supersonic collision with the red - shifted cloud which has lower column density .
the epoch of o star formation may range in a certain time span , whereas the cloud - cloud collision poses a limit on the age spread of the youngest members including o stars by the ratio of the cluster size divided by the collision velocity to be @xmath60.1myr .
it is unlikely that the two gravitationally bound scenarios can explain the very short discontinuous history of star formation less than 0.1myr .
we do not see any gap in the age distribution of the cluster member .
this is a natural consequence of the cloud - cloud collision following myr low - mass star formation .
we have made a new analysis of the molecular gas toward m42 and m43 , and test a hypothesis that two clouds collided about 0.1myr ago to form the northern part of the orion a cloud .
we have shown the collision model offers a reasonable explanation of the observed cloud and cluster properties in the m42 and m 43 region . although , currently , we do not have an extensive observational basis that the cloud - cloud collision is a dominant mechanism of o star formation in the galaxy .
the main conclusions of the present study are summarized as follows . 1 . the present analysis of the high resolution co data shows a possibility that the orion a cloud toward m42 and m43 consists of two velocity components , one at 4.011.1kms@xmath3 and the other at 11.114.9kms@xmath3 .
the mass of the blue - shifted component is 1.5@xmath2610@xmath46@xmath9 with its velocity range 4.011.1kms@xmath3 and the red - shifted component 3.4@xmath2610@xmath60@xmath9 with its velocity range 11.114.9kms@xmath3 having a size of 4pc by 7pc in r.a . and
we find at least three pairs of complementary distributions between the two clouds having scale sizes of 0.11pc .
these complementary distributions of molecular gas are an observational signature characteristic to cloud - cloud collision according to the hydrodynamical numerical simulations at the early phase of o star formation .
the major complementary distribution is found toward @xmath4ori and @xmath5ori where the blue - shifted component of @xmath60.5pc radius , the omc-1 clump , is surrounded by the red - shifted component , the u - shaped cloud , which is extended with a radius of @xmath61pc .
the secondary complementary distribution is seen toward m43 as named the orion key and keyhole , where a spatial displacement of 0.3pc is seen between the two complementary distributions .
we frame a hypothesis that the two velocity components collided with each other to trigger formation of the nearly ten o stars in m42 and the b star in m43 .
the displacement between the two complementary features in m43 suggests that the relative motion makes a large angle to the line - of - sight . by assuming the angle to be 45deg , the relative velocity and separation between the clouds is estimated to be @xmath66kms@xmath3 and @xmath60.5pc , respectively , after correction for the projection , and the typical collision time scale is calculated to be @xmath60.1myr .
although the collision is extended over an area of 4pc@xmath267pc in r.a . and
the triggering of the o / b star formation is seen only toward the localized regions with ob stars .
we suggest that this trend is consistent with the threshold for o star formation in cloud - cloud collision suggested by @xcite ; multiple o star formation is possible for column density @xmath610@xmath50@xmath14 while the single o / b star formation for @xmath610@xmath51@xmath14 .
the high segregation of the high - mass cluster members to the cluster center is explained as a reflection of the initial column density distribution prior to the collision .
the stellar contents of the onc has been best studied among the rich clusters thanks to its unrivaled small distance to the sun although the cluster is not overwhelmingly rich in high - mass stars .
the ob stars in the onc are minor members in terms of stellar mass , and the initial mass function is not significantly influenced by the formation of the ob stars .
the youngest members including the o / early b stars are formed by the collision within 0.1myr , and the large age spread of the cluster , from a few 0.01myr to a few times 1.0myr , is understood by a combination of low - mass star formation in the blue - shifted cloud prior to the collision and the recent o / b star formation by the collisional triggering .
formation of o stars may still be continuing toward the western edge of the m42 clump where the bn object and omc-1-s are located along the north - south direction . in the present study
we have shown that cloud - cloud collision offers a possible consistent interpretation of the formation of the o / b stars in m42 and m43 . since the immediate vicinity of @xmath4ori and @xmath5ori is significantly ionized , we are not allowed to directly witness the triggering as in case of rcw38 .
we note that the bn / kl object can be such a candidate as discussed in section 5 , and suggest that a future more careful scrutiny of the region and its surroundings at higher resolutions with alma etc . will offer a further insight into the possible role of the collisional interaction .
we are grateful to thomas haworth for valuable discussion on this paper .
we also thanks to robert gendler for optical image of the onc , and kazuki tokuda for alma archive data .
this publication makes use of data products from the two micron all sky survey , which is a joint project of the university of massachusetts and the infrared processing and analysis center / california institute of technology , funded by the national aeronautics and space administration and the national science foundation .
nobeyama radio observatory is a branch of the national astronomical observatory of japan , national institutes of natural sciences .
the aste telescope is operated by national astronomical observatory of japan ( naoj ) .
this paper makes use of the following alma data : ads / jao.alma#2012.1.00554.s .
alma is a partnership of eso ( representing its member states ) , nsf ( usa ) and nins ( japan ) , together with nrc ( canada ) , nsc and asiaa ( taiwan ) , and kasi ( republic of korea ) , in cooperation with the republic of chile .
the joint alma observatory is operated by eso , aui / nrao and naoj .
this work was supported by jsps kakenhi grant numbers 15h05694 , 15k17607 , 15k05014 , 16k17664 , and 26247026 ; by jsps and by the mitsubishi foundation .
natexlab#1#1 , s. v. 2010 , , 405 , 1431 , j. , langer , w. d. , stark , a. a. , & wilson , r. w. 1987 , , 312 , l45 , s. , & kroupa , p. 2015 , arxiv e - prints , arxiv:1512.03074 , a. 1991 , in nato advanced science institutes ( asi ) series c , vol .
342 , nato advanced science institutes ( asi ) series c , ed . c. j. lada & n. d. kylafis , 125 , j. m. , hillenbrand , l. a. , & skrutskie , m. f. 2001 , , 121 , 3160 , a. , duvert , g. , dutrey , a. , et al .
1990 , , 234 , 469 , j. e. , ercolano , b. , & bonnell , i. a. 2013 , , 431 , 1062 , g. , paresce , f. , panagia , n. , et al .
2011 , , 739 , 27 , c. l. , pringle , j. e. , & duarte - cabral , a. 2015 , , 446 , 3608 , m. p. , shipman , r. f. , price , s. d. , et al .
1998 , , 494 , l199 , b. g. , & lada , c. j. 1977 , , 214 , 725 , e. d. , getman , k. , townsley , l. , et al .
2005 , , 160 , 379 , y. , ohama , a. , hanaoka , n. , et al .
2014 , , 780 , 36 , y. , harada , r. , tokuda , k. , et al .
2015 , , 807 , l4 , y. , torii , k. , ohama , a. , et al .
2016 , , 820 , 26 , n. , dawson , j. r. , ohama , a. , et al .
2009 , , 696 , l115 , a. , & ohta , k. 1992 , , 44 , 203 , l. , & burkert , a. 2007 , , 654 , 988 , t. j. , shima , k. , tasker , e. j. , et al . 2015 , arxiv e - prints , arxiv:1509.00859 , t. j. , tasker , e. j. , fukui , y. , et al .
2015 , , 450 , 10 , m. h. , morgan , j. , schloerb , f. p. , snell , r. l. , & goldsmith , p. f. 1992 , , 395 , l99 , l. a. 1997 , , 113 , 1733 , l. a. , & carpenter , j. m. 2000 , , 540 , 236 , t. , & fukui , y. 2013 , , 774 , l31 , p. 2001
, , 322 , 231 , p. , aarseth , s. , & hurley , j. 2001 , , 321 , 699 , n. , brandner , w. , gennaro , m. , et al .
2012 , , 750 , l44 , m. l. , tucker , k. d. , chin , g. , & thaddeus , p. 1977
, , 215 , 521 , r. j. , morris , m. , moscowitz , j. , & thaddeus , p. 1986
, , 303 , 375 , f. , pomars , m. , deharveng , l. , zavagno , a. , & bouret , j. c. 2010 , , 510 , a32 , c. f. , & tan , j. c. 2003 , , 585 , 850 , a. , getman , k. , hillenbrand , l. , & preibisch , t. 2008 , star formation in the orion nebula i : stellar content , ed .
b. reipurth , 483 , t. , mizuno , a. , ogawa , h. , & fukui , y. 1998 , , 116 , 336 , f. , miura , t. , kitamura , y. , et al .
2012 , , 746 , 25 , a. , tokuda , k. , kimura , k. , et al .
2015 , , 216 , 18 , c. r. , muench , a. , smith , n. , & zapata , l. 2008 , star formation in the orion nebula ii : gas , dust , proplyds and outflows , ed . b. reipurth , 544 , a. , dawson , j. r. , furukawa , n. , et al . 2010 , , 709 , 975 , d. e. , & megeath , s. t. 2008 , the orion molecular cloud 2/3 and ngc 1977 regions , ed . b. reipurth , 590 , r. , bensch , f. , howe , j. e. , et al . 2000 , , 539 , l133 , f. , heyer , m. h. , gutermuth , r. , snell , r. l. , & brunt , c. m. 2013 , , 431 , 1296 , k. , onishi , t. , nayak , o. , et al .
2016 , arxiv e - prints , arxiv:1604.06010 , s. , hayashi , m. , hasegawa , t. , handa , t. , & oka , t. 1994 , , 425 , 641 , y. , kawabe , r. , takakuwa , s. , et al .
2011 , , 63 , 105 , a. , humann , b. , olczak , c. , et al .
2015 , , 578 , a4 , a. w. , bloemen , j. b. g. m. , dame , t. m. , et al .
1988 , , 207 , 1 , a. m. , & gould , a. 2016 , , 590 , a2 , k. , fukui , y. , ogawa , h. , & kawabata , k. 1986 , , 303 , 667 , k. , yamaguchi , c. , nakai , n. , et al .
2000 , in , vol . 4015 , radio telescopes , ed . h. r. butcher , 237246 , h. , fukui , y. , fujimoto , y. , et al .
1986 , , 166 , 276 , k. , tasker , e. j. , & habe , a. 2014 , , 792 , 63 , j. c. , beltrn , m. t. , caselli , p. , et al .
2014 , protostars and planets vi , 149 , k. , enokiya , r. , sano , h. , et al .
2011 , , 738 , 46 , k. , hasegawa , k. , hattori , y. , et al .
2015 , , 806 , 7 , g. j. , & sandell , g. 1995 , , 299 , 179 , a. p. , & zinnecker , h. 2004 , , 427 , 299 , b. a. , dame , t. m. , masheder , m. r. w. , & thaddeus , p. 2005
, , 430 , 523 , m. g. , & cassinelli , j. p. 1986
, , 310 , 207 , c. , sunada , k. , iizuka , y. , iwashita , h. , & noguchi , t. 2000 , in , vol . 4015 , radio telescopes , ed . h. r. butcher , 614623 , a. , russeil , d. , motte , f. , et al .
2010 , , 518 , l81 , h. , & yorke , h. w. 2007 , , 45 , 481 [ table:1 ] [ table:2 ] .velocity range of the molecular emission in the 8 cases ( table 1 ) . [ cols="^,^,^,^,^,^,^,^,^,^,^,^ " , ]
we describe the method for a simple case of the complementary distribution in the x - y plane ( figure 17 ) where the small cloud and the intensity depression are assumed to be circles of the same radius a with an initial displacement of @xmath61 along the x - axis in the sky . by shifting the small cloud by @xmath41 in the direction of the intensity depression along the x - axis ,
we calculate the overlapping product between the two circles represented as @xmath62 and @xmath63 by assuming that they have uniform intensity 1.0 in an arbitrary unit inside the circle and 0.0 outside the circle . by taking a product of the circle and the intensity depression at @xmath41 ,
the overlapping between the two circles is given by @xmath64 , the product of @xmath65 and @xmath66 .
@xmath67 is integrated in the x - y plane and is expressed by @xmath39 .
the displacement of the two circles is then estimated to be @xmath68 at the peak of @xmath39 ( figure 2 ) .
the method is applied to the synthetic observations in figures 2(b ) , ( c ) , and ( d ) . in figures 2(b ) and
( c ) , the distributions of the small cloud and the intensity depression are taken from figure 4 .
grid spacing is given as [email protected] in the x - y plane according to the numerical simulations @xcite , and we assign a numerical value to be 1.0 , for each pixel inside the both components , and to be zero outside of them as above ( figures 17(f ) and ( g ) ) .
@xmath39 is calculated in the same manner as above with an incremental step of 0.1pc in @xmath41 and is shown in figure 2(d ) .
the peak position of @xmath39 gives the displacement to be @xmath62pc . by correcting for the projection effect
, the displacement corresponds to the length of the cavity @xmath63pc , where the size of the small cloud is taken into account .
we find the value is roughly consistent with the length of the cavity in the large cloud , @xmath64pc , at 1.6myr in the numerical simulations . in section 3 ,
we applied the method to estimate the displacement between the orion key and keyhole . | the orion nebula cluster toward the hii region m42 is the most outstanding young cluster at the smallest distance 410pc among the rich high - mass stellar clusters . by newly analyzing the archival molecular data of the @xmath0co(@xmath1=10 ) emission at 21@xmath2 resolution , we identified at least three pairs of complementary distributions between two velocity components at 8kms@xmath3 and 13kms@xmath3 .
we present a hypothesis that the two clouds collided with each other and triggered formation of the high - mass stars , mainly toward two regions including the nearly ten o stars , @xmath4ori and @xmath5ori , in m42 and the b star , nuori , in m43 .
the timescale of the collision is estimated to be @xmath60.1myr by a ratio of the cloud size and velocity corrected for projection , which is consistent with the age of the youngest cluster members less than 0.1myr .
the majority of the low - mass cluster members were formed prior to the collision in the last one myr .
we discuss implications of the present hypothesis and the scenario of high - mass star formation by comparing with the other eight cases of triggered o star formation via cloud - cloud collision . |
a natural standpoint about quantum mechanics , which is however not the one usually considered in textbooks written for physics students , is to look at it as a new probability theory , different and reacher than the classical one @xcite .
this point of view becomes mandatory or at least very fruitful if one is faced with more advanced research topics , such as the description of open quantum systems or quantum information and communication theory ( for a general reference see @xcite ) . in these fields tools and concepts obtained relying on a probabilistic approach , also working in direct analogy with classical probability theory ,
have become of paramount importance .
an example in this direction is given by quantum dynamical semigroups , which provide the quantum generalization of classical markov semigroups .
the subject has been the object of active research in the mathematics , physics and chemistry community over decades by now , but it is still of great interest . in particular covariance properties of such mappings under translations have been considered in detail only recently . besides a mathematical characterization @xcite also the actual physical relevance @xcite of such covariant quantum dynamical semigroups has been considered . in the present contribution we will focus mainly on the application of such translation - covariant quantum dynamical semigroups to the study and the description of the phenomenon called decoherence in the physics literature @xcite . by such a term a whole variety of situations
is meant , all having in common a loss of typical quantum interference capability , arising as a dynamical consequence of interaction of the system of interest with some other , typically much bigger , system .
the phenomenology of decoherence is ubiquitous when considering open quantum systems , but its actual quantitative study requires very special experimental conditions , which can be realized e.g. in interferometric setups for massive particles , observing loss of interference fringes as a consequence of external disturbance , arising because the approximation of isolation of the system is no more realistic . for a quantitative study of the phenomenon it is in fact crucial that such decoherence effects can actually be engineered , so that their strength is under the control of the experimenter .
the paper is organized as follows . in sect .
[ sec : tv ] we briefly sketch the formal expression of the generator of a translation - covariant quantum dynamical semigroup . in sect .
[ sec : lp ] we show how such a general structure in a suitable limit can account for decoherence behaviors quantitatively described by means of the characteristic function of a classical lvy process . in sect .
[ sec : mp ] we further explore how a particular physical example of realization of such generators applies to the description of decoherence in both position and momentum space , finally mentioning possible extension of the formalism in sect .
[ sec : co ] .
provided memory effects can be neglected , quantum dynamical semigroups @xcite give a general setting for the description of the dynamics of an open quantum system @xcite . in the physical literature
major efforts have been devoted to the derivation or phenomenological assessment of possible generators of such quantum dynamical semigroups , so called master equations .
the typical benchmark is the lindblad structure of such generators , which goes back to the work of gorini , kossakowski and sudarshan @xcite and of lindblad @xcite , holding true for a generator given by a bounded mapping .
attention was later devoted to possible constraints on the structure of such generators arising as a consequence of symmetry ( see e.g. @xcite for references ) . in this respect
the results of holevo for symmetry under translations are of particular importance because of the many possible physical applications , especially in connection with typical quantum phenomena such as decoherence .
we first consider the general expression of formal generators of translation - covariant quantum dynamical semigroups as obtained by holevo @xcite .
the covariance of the mapping corresponds to the requirement that its action has to commute with the unitary representation of translations on the hilbert space of interest .
the physical system we are going to consider is the centre of mass of a particle in free space , so that @xmath0 .
let @xmath1 be the mapping describing the dynamics in heisenberg picture , thus acting on an observable @xmath2 .
in order to be covariant @xmath1 has to satisfy the requirement @xmath3 & = & { \mathrm{e}}^{i \mathsf{{\ensuremath{\boldsymbol{a } } } \cdot \mathsf{p } } / \hbar } \mathcal{l } ' \left [ \mathsf{a } \right ] { \mathrm{e}}^{- i \mathsf{{\ensuremath{\boldsymbol{a } } } \cdot \mathsf{p } } / \hbar } \hspace{2em } \forall { \ensuremath{\boldsymbol{a } } } \in \mathbb{r}^3 , \label{eq : cov}\end{aligned}\ ] ] where @xmath4 denotes the momentum operator of the massive particle .
the general structure of generator complying with this requirement is given by the formal operator expression @xmath5 \right .
& = & \frac{i}{\hbar } [ \mathsf{h } \left ( \mathsf{p } \right ) , \mathsf{a } ] + \mathcal{l}_g [ \mathsf{a } ] + \mathcal{l}_p [ \mathsf{a } ] , \label{eq : g+p}\end{aligned}\ ] ] where the symbols @xmath6 and @xmath7 denote a gaussian and a poisson component , the names arising from the connection with the classical lvy - khintchine formula .
one has in particular for the gaussian component @xmath8 & = & \frac{i}{\hbar } \left [ \mathsf{y}_0 + \frac{1}{2 i } \sum_{k = 1}^3 \left ( \mathsf{y}_k l_k \left ( \mathsf{p } \right ) - l^{^{\dag}}_k \left ( \mathsf{p }
\right ) \mathsf{y}_k \right ) , \mathsf{a } \right ] \nonumber\\ & & + \frac{1}{\hbar } \sum_{k = 1}^3 \left [ \left ( \mathsf{y}_k + l_k \left ( \mathsf{p } \right ) \right)^{^{\dag } } \mathsf{a } \left ( \mathsf{y}_k + l_k \left ( \mathsf{p } \right ) \right ) \right . \nonumber\\ & & \left .
- \frac{1}{2 } \left\ { \left ( \mathsf{y}_k + l_k \left ( \mathsf{p } \right ) \right)^{^{\dag } } \mathsf { } \left ( \mathsf{y}_k + l_k \left ( \mathsf{p } \right ) \right ) , \mathsf{a } \right\ } \right ] \nonumber\end{aligned}\ ] ] where @xmath9 with @xmath10 for @xmath11 , that is to say it is a linear combination of the three position operators of the test particle , appearing at most quadratically , while for the poisson component @xmath12 & = & \int { \mathrm{d}}\mu \left ( { \ensuremath{\boldsymbol{q } } } \right ) \sum_j \left [ l_j^{^{\dag } } \left ( \mathsf{p } ; { \ensuremath{\boldsymbol{q } } } \right ) { \mathrm{e}}^{- i{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } / \hbar } \mathsf{a } { \mathrm{e}}^{i{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } / \hbar } l_j \left ( \mathsf{p } ; { \ensuremath{\boldsymbol{q } } } \right ) \right .
\nonumber\\ & & \left .
- \frac{1}{2 } \left\ { l_j^{^{\dag } } \left ( \mathsf{p } ; { \ensuremath{\boldsymbol{q } } } \right ) l_j \left ( \mathsf{p } ; { \ensuremath{\boldsymbol{q } } } \right ) , \mathsf{a } \right\ } \right ] \nonumber\\ & & + \int { \mathrm{d}}\mu \left ( { \ensuremath{\boldsymbol{q } } } \right ) \sum_j \left [ \omega_j \left ( { \ensuremath{\boldsymbol{q } } } \right ) l_j^{^{\dag } } \left ( \mathsf{p } ; { \ensuremath{\boldsymbol{q } } } \right ) \left ( { \mathrm{e}}^{- i{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } / \hbar } \mathsf{a } { \mathrm{e}}^{i{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } / \hbar } - \mathsf{a } \right ) \right . \nonumber\\ & & \left .
+ \left ( { \mathrm{e}}^{- i{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } / \hbar } \mathsf{a } { \mathrm{e}}^{i{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } / \hbar } - \mathsf{\mathsf{a } } \right ) l_j \left ( \mathsf{p } ; { \ensuremath{\boldsymbol{q } } } \right ) \omega^{\star}_j \left ( { \ensuremath{\boldsymbol{q } } } \right ) \right ] \nonumber\\ & & \nonumber\\ & & + \int { \mathrm{d}}\mu \left ( { \ensuremath{\boldsymbol{q } } } \right ) | \omega_j \left ( { \ensuremath{\boldsymbol{q } } } \right ) |^2 \sum_j \left [ { \mathrm{e}}^{- i{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } / \hbar } \mathsf{a } { \mathrm{e}}^{i{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } / \hbar } - \mathsf{a } - \frac{i}{\hbar } \frac{\left [ \mathsf{a } , { \ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } ] \right.}{1 + q^2 / q^2_0 } \right ] . \nonumber\end{aligned}\ ] ] such expressions can cover a huge variety of physical situations , accounting for both dissipative and decoherence effects . some rough insight can be gained considering the dummy integration label @xmath13 as a momentum .
the dynamics of the open system , in our case the centre of mass of a tracer particle , is thus described by an interaction only characterized by the momentum transfers between system and environment , taking place e.g. as a consequence of collisions , thus complying with translational invariance .
the unitary operators @xmath14 appearing in the poisson part describe in fact a momentum kick , with rates which are not only given by functions of the momentum transfer @xmath13 itself , but also depend on the momentum operator @xmath4 , thus becoming dynamic quantities .
this is in particular necessary in order to correctly describe phenomena like energy transfer and approach to equilibrium .
the gaussian part corresponds to a dynamics arising as a consequence of a big number of small momentum transfers , leading to a diffusive behavior .
an interesting limiting situation appears if we neglect dissipative effects and therefore the dynamics of the momentum operator , apart from its appearance in the free kinetic term , so that , also switching to the preadjoint mapping in schrdinger picture , and assuming @xmath15 to be absolutely continuous with respect to the lebesgue measure , the two contributions can be written @xmath16 & = & - \frac{i}{\hbar } \sum_{i = 1}^3 \mathbf{b}_i [ \mathsf{x}_i , \rho ] - \sum_{i , j = 1}^3 \frac{1}{2 } { \ensuremath{\boldsymbol{d}}}_{{\ensuremath{\text{ij } } } } [ \mathsf{x}_i , [ \mathsf{x}_j , \rho ] ] \label{eq : decoh1}\\ \mathcal{l}_p [ \rho ] & = & \int d{\ensuremath{\boldsymbol{q}}}| \lambda ( { \ensuremath{\boldsymbol{q } } } ) |^2 \left [ e^{i{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } / \hbar } \rho e^{- i{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } / \hbar } - \rho - \frac{i}{\hbar } \frac{[{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } , \rho]}{1 + q^2 / q^2_0 } \right ] \label{eq : decoh2}\end{aligned}\ ] ] where @xmath17 , and the integration measure satisfies the lvy condition @xmath18 it is very convenient to write the contributions given by eq .
( [ eq : decoh1 ] ) and eq .
( [ eq : decoh2 ] ) in the position representation , leading to the simple expression @xmath19 + \mathcal{l}_p [ \rho ] |{\ensuremath{\boldsymbol{y } } } \rangle & = & - \psi \left ( { \ensuremath{\boldsymbol{x}}}-{\ensuremath{\boldsymbol{y } } } \right ) \langle { \ensuremath{\boldsymbol{x}}}| \rho |{\ensuremath{\boldsymbol{y } } } \rangle , \label{eq : xrepr}\end{aligned}\ ] ] where according to eq .
( [ eq : decoh1 ] ) and eq . ( [ eq : decoh2 ] ) we have introduced the function @xmath20 , \nonumber\end{aligned}\ ] ] only depending on the difference @xmath21 due to translational invariance . the action of the contributions given by eq .
( [ eq : decoh1 ] ) and eq .
( [ eq : decoh2 ] ) in the position representation is therefore very simple , it only amounts to multiplying the matrix elements of the statistical operator by a function of the particular form ( [ eq : psi ] ) , whose general properties as we shall see naturally account for a description of decoherence .
the master equation corresponding to eq .
( [ eq : decoh1 ] ) and eq .
( [ eq : decoh2 ] ) can be easily solved in the position representation , giving a dynamics which only changes the initial statistical operator by a multiplicative time dependent factor @xmath22 a key point is now the observation that eq .
( [ eq : psi ] ) actually gives the general expression of the characteristic exponent appearing in the characteristic function of a lvy process , corresponding to the celebrated lvy - khintchine formula @xcite . as a consequence
the function @xmath23 gives the general possible expressions for the characteristic function of a classical lvy process , different processes , e.g. gaussian , poisson , compound poisson or lvy stable processes arising corresponding to the different possible values of @xmath24 and of the positive weight @xmath25 in the measure .
these different lvy processes intuitively correspond to the different ways according to which momentum is transferred to the test particle as a consequence of interaction with the environment .
thus for example a poisson process corresponds to a situation in which the different possible interaction events are characterized by a fixed momentum transfer , given by the height of the jumps in the poisson process .
more generally a physically realistic situation involves a compound poisson process , characterized by the fact that the momentum transfer in the single interaction events is not a deterministic quantity , but it is itself described by a probability density , depending on the detail of the microscopic interaction mechanism , according to which the poisson process is composed .
the function @xmath26 is a characteristic function , so that it has the following interesting properties , explaining why eq .
( [ eq : sol ] ) generally gives a well defined master equation describing loss of coherence in the position representation : @xmath27 @xmath28 @xmath26 is positive definite @xmath29 for @xmath30 @xmath29 for @xmath31 , provided there exists a probability density .
these properties typical of characteristic functions @xcite automatically entail that the diagonal matrix elements in the position representation are not affected with elapsing time , thus preserving normalization of the statistical operator , while the off - diagonal matrix elements are generally suppressed as expected due to decoherence .
furthermore for a fixed spatial distance @xmath21 the off - diagonal matrix elements in the position representation are fully suppressed for long enough interaction times , while for a fixed interaction time these off - diagonal matrix elements only go to zero if the associated process admits a proper probability density , which is not the case e.g. for a compound poisson process . depending on the particular process describing the random momentum transfers in each scattering event different characteristic functions appear , corresponding to different behaviors in the suppression of the off - diagonal matrix elements for large spatial separations .
the function @xmath32 , which is responsible for the loss of visibility in interferometric experiments testing decoherence , for a fixed interaction time @xmath33 might monotonically decrease to zero for growing values of @xmath21 , or also oscillate and reach asymptotically a finite value corresponding to a residual coherence .
these quite different behaviors , corresponding to a more or less effective decoherence effect , are all encoded in the possible expressions of the characteristic function @xmath34 .
application of this formalism to actually realized experiments has been considered in @xcite .
typical experiments testing decoherence in a quantitative way involve an interferometer for massive particles ( such as fullerenes @xcite or atoms @xcite ) , in which the interfering particle is exposed to some environment during the time of flight , such as a background gas , a laser field or even the internal degrees of freedom of the interfering particle itself .
the general structure of translation - covariant quantum dynamical semigroups allows for the description of decoherence effects provided one considers the behavior in time of the so called coherences , that is to say the off - diagonal matrix elements of the statistical operator in a given basis , selected by the dynamics itself or by the observation which can be performed on the open system . for the considered massive particle interacting with some environment the natural basis
are given by momentum or position . in order to describe both phenomena
we obviously can not neglect the momentum dynamics as implicitly done going over from eq .
( [ eq : g+p ] ) to eq .
( [ eq : decoh1 ] ) and eq .
( [ eq : decoh2 ] ) .
we therefore need a physical example of realization of the general structure eq .
( [ eq : g+p ] ) , as given by the quantum version of the classical linear boltzmann equation @xcite .
such a master equation describes the dynamics of a quantum test particle interacting through collisions with a homogeneous gas , thus providing a quantum counterpart of the classical linear boltzmann equation . for the case of a scattering cross section
@xmath35 only depending on the momentum transfer the equation can be written @xmath36 & = & \frac{n_{{\ensuremath{\text{gas } } } } } { m^2_{\ast } } { \int}{\mathrm{d}}{\ensuremath{\boldsymbol{q } } } \sigma \left ( { \ensuremath{\boldsymbol{q } } } \right ) \left [ { \mathrm{e}}^{i{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } / \hbar } \sqrt{s \left ( { \ensuremath{\boldsymbol{q } } } , \mathsf{p } \right ) } \rho \sqrt{s \left ( { \ensuremath{\boldsymbol{q } } } , \mathsf{p } \right ) } { \mathrm{e}}^{- i{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } / \hbar } \label{eq : qdsf } \right.\\ & & \left .
- \frac{1}{2 } \left\ { s \left ( { \ensuremath{\boldsymbol{q } } } , \mathsf{p } \right ) , \rho \right\ } \right ] , \nonumber\end{aligned}\ ] ] with @xmath37 the density of gas particles with mass @xmath38 , @xmath39 the mass of the test particle , @xmath40 the reduced mass , @xmath41 a two - point correlation function of the gas known as dynamic structure factor and explicitly given by @xmath42 with @xmath43 the energy transfer in the single collision and @xmath44 .
we are not going to delve on details of the structure of such an equation .
we only point out that it actually provides an example of translation - covariant master equation complying with the general mathematical result .
we are however interested to show that such a structure actually describes decoherence phenomena in both momentum and position .
in fact while the classical linear boltzmann equation only describes dissipative effects , corresponding to the behavior of populations in momentum space , that is the diagonal matrix elements in the momentum representation of eq .
( [ eq : qdsf ] ) , the quantum master equation also describes coherences and therefore possibly interference phenomena and suppression thereof as a consequence of the dynamics , provided suitable quantum states given by linear superpositions states are considered .
looking at coherence in momentum space implies considering coherent superpositions of momentum eigenstates .
such highly non classical motional states can show interference effects which are expected to be suppressed as a consequence of the interaction with the environment . as a consequence matrix elements of the form @xmath45 , so that for long enough times the dynamics only affects the behavior of the probability density @xmath46 , and the master equation eq .
( [ eq : qdsf ] ) goes effectively over to a classical rate equation for such a probability density . due to the complexity of eq .
( [ eq : qdsf ] ) obtaining an analytical solution is hardly feasible , so that the natural strategy is to numerically solve the master equation , relying on a so called unraveling of the master equation itself @xcite , to be solved by means of monte carlo methods . in this case
setting @xmath47 one can consider the following stochastic differential equation for the stochastic wave vector @xmath48 @xmath49 | \psi \left ( t \right ) \rangle { \ensuremath{\text{dt } } } \nonumber\\ & & + { \int}{\mathrm{d}}{\ensuremath{\boldsymbol{q } } } \left [ \frac{v \left ( { \ensuremath{\boldsymbol{q } } } \right ) | \psi \left ( t \right ) \rangle}{\|v \left ( { \ensuremath{\boldsymbol{q } } } \right ) | \psi \left ( t \right ) \rangle \| } - | \psi \left ( t \right ) \rangle \right ] { \ensuremath{\text{dn}}}_{{\ensuremath{\boldsymbol{q } } } } \left ( t \right ) , \label{eq : d}\end{aligned}\ ] ] where the field of increments satisfies @xmath50 & = & \|v \left ( { \ensuremath{\boldsymbol{q } } } \right ) | \psi \left ( t \right ) \rangle \|^2 { \ensuremath{\text{dt } } } , \nonumber\end{aligned}\ ] ] so that indeed the solutions of the stochastic differential equation ( [ eq : d ] ) provide unravelings of the master equation eq . (
[ eq : qdsf ] ) , in the sense that @xmath51 .
\nonumber\end{aligned}\ ] ] despite the formal complexity of eq .
( [ eq : d ] ) , for initial states given by momentum eigenvectors one can develop a simple algorithm to study the dynamics of such states , essentially corresponding to the gillespie algorithm @xcite , leading to a pure jump process in momentum space . on similar grounds one can also study the dynamics of coherent superpositions of the form @xmath52 with @xmath53 , which evolve in time according to @xmath54 where again @xmath55 .
an estimate of loss of coherence can be obtained studying the quantity @xmath56 .
\nonumber\end{aligned}\ ] ] as it turns out this measure for the coherence of the state in the momentum basis behaves for a constant scattering cross section approximately as @xcite @xmath57 , \label{eq : lossp}\end{aligned}\ ] ] where the argument of the exponential is given by @xmath58 with @xmath59 @xmath60 denotes the error function , and @xmath61 is a reference scattering rate given by @xmath62 , with @xmath63 the most probable velocity for the gas particles .
( [ eq : lossp ] ) clearly predicts an exponential loss of coherence in the momentum basis , depending on the relative distance in momentum space of the states making up the coherent superposition . for the study of decoherence in position space we can follow a different strategy . neglecting in eq .
( [ eq : qdsf ] ) the dynamics of the momentum , we can replace the corresponding operator by a classical label @xmath64 giving the mean value of the momentum of the incoming particle .
the master equation then reads @xmath36 & = & \frac{n_{{\ensuremath{\text{gas } } } } } { m^2_{\ast } } { \int}{\mathrm{d}}{\ensuremath{\boldsymbol{q } } } \sigma \left ( { \ensuremath{\boldsymbol{q } } } \right ) s \left ( { \ensuremath{\boldsymbol{q } } } , \mathsf{{\ensuremath{\boldsymbol{p}}}_0 } \right ) \left [ { \mathrm{e}}^{i{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } / \hbar } \rho { \mathrm{e}}^{- i{\ensuremath{\boldsymbol{q } } } \cdot \mathsf{x } / \hbar } - \rho \right ] , \label{eq : qdsfp}\end{aligned}\ ] ] corresponding to a particular realization of eq .
( [ eq : decoh2 ] ) . considering a constant scattering cross section and defining the rate @xmath65 according to ( [ eq : rate ] )
one can introduce the following characteristic function @xmath66 so that the master equation ( [ eq : qdsfp ] ) can be solved in the position representation as in ( [ eq : sol ] ) , leading to @xmath67 t \right ) \langle { \ensuremath{\boldsymbol{x}}}| \rho_0 |{\ensuremath{\boldsymbol{y } } } \rangle , \label{eq : aa}\end{aligned}\ ] ] where according to the general framework presented in sect .
[ sec : lp ] the characteristic function of a compound poisson process appears . a suitable measure of decoherence
is given in this case by @xmath68 for a test particle slower than the gas particles , so that @xmath69 , one has @xmath70 with @xmath71 the thermal de broglie wavelength of the gas particles given by @xmath72 , and @xmath73 the confluent hypergeometric function , so that @xmath74 t \right ) , \nonumber\end{aligned}\ ] ] which for spatial distances above the thermal de broglie wavelength @xmath75 is well approximated by a fixed decoherence rate @xmath76exp@xmath77 , expressing the fact that for large enough distances off - diagonal matrix elements in the position representation are uniformly suppressed .
we have given a brief presentation of how quantum dynamical semigroups can be useful for the description of decoherence in quantum mechanics , as also pursued in @xcite , coping in a quantitative way with experimentally realizable situations .
this has been obtained relying on a characterization of translation - covariant quantum dynamical semigroups , leading to a quantum non - commutative generalization of the lvy - khintchine formula . when applied to the study of decoherence , neglecting dissipative phenomena ,
such a structure leads to a description of loss of coherence with a wide variety of possible behaviors , each corresponding to the characteristic function of a classical lvy process . despite pursued within the framework of the markov assumption , thus supposing that the dynamics does not entail memory effects , the approach to the description of decoherence building on covariance properties , recently also followed in @xcite ,
can be of more general validity , as it appears from recent results pointing to a generalization of the lindblad structure for the description of a class of non - markovian evolutions @xcite .
the author is grateful to the organizers for the kind invitation and hospitality at cimat .
the work was partially supported by the italian mur under prin2005 .
r. alicki , invitation to quantum dynamical semigroups , in _ dynamical semigroups : dissipation , chaos , quanta _ , eds .
p. garbaczewski and r. olkiewicz , lecture notes in physics , vol . 597 ( springer - verlag , berlin , 2002 ) pp . | translation - covariant markovian master equations used in the description of decoherence and dissipation are considered in the general framework of holevo s results on the characterization of generators of covariant quantum dynamical semigroups .
a general connection between the characteristic function of classical lvy processes and loss of coherence of the statistical operator describing the center of mass degrees of freedom of a quantum system interacting through momentum transfer events with an environment is established .
the relationship with both microphysical models and experimental realizations is considered , focusing in particular on recent interferometric experiments exploring the boundaries between classical and quantum world . |
over past decades , cavity quantum electrodynamics ( qed ) has become an important platform to demonstrate quantum characteristics of atoms and photons @xcite . some remarkable experiments , such as strong coupling of the trapped atoms with cavity @xcite , conditional logic gating @xcite , efficient generation of single photons single , singlephoton and
so on have been performed successfully . with atoms strongly interacting with the local cavity mode as quantum nodes and the photons flying between different nodes as quantum bus
, we may set up a quantum network , which has been considered as a promising way to scaling of the qubit system for large - scale quantum information processing .
many theoretical @xcite and experimental works interf , beu have been done in this direction and some advances have been achieved over recent years . among the theoretical work mentioned above , the single photon experiencing input - output process of a cavity duan04,feng is of particular interests , in which the atomic qubits could be entangled by a single moving photon and the successful implementation is monitored by a click of the detector .
the present paper will focus on the input - output process regarding optical cavities .
motivated by a very recent experiment with microtoroidal resonator ( mtr ) @xcite , we intend to carry out some quantum information processing ( qip ) tasks by means of the input - output process relevant to optical cavities with low - q factors .
we have noticed that the proposals duan04,feng required the cavities to be with high quality and with strong coupling to the confined atoms , otherwise the schemes would not work well or would be pretty inefficient .
some simulation has been done to check how well the cavity input - output process works @xcite .
it was shown that the gating time should be much shorter than the decay time of the cavity if we expect to have the gating with high success probability @xcite .
evidently , it is not an easy experimental task to meet those requirements because the efficient output of photons , to some extent , implies the larger cavity decay rate , i.e. , the cavity with a relatively lower q factor . in this sense , the recent achievement of the mtr gives us hopes to solve the problem .
although it is still of large decay rate ( i.e. , called ` bad ' cavity in @xcite ) and moderate coupling to the atom , the mtr , with individual photons input and output through a microresonator , has explicitly shown the effect of the photon blockade .
so this mrt seems a promising candidate setup to be meeting the requirements in those qip schemes @xcite .
our present work will , however , show the possibility of accomplishing some interesting qip tasks with the currently achieved mtr .
the key step is to design a scheme for entangling two atoms confined respectively in two spatially separate cavities with low - q factors .
so our work is actually not only relevant to the mtr , but also related to any single - sided optical cavities with one wall perfectly reflective but the other partially reflective @xcite . as a result
, the mtr and the single - sided optical cavity will be mentioned alternately in what follows .
we will first present an analytical expression for the reflection rate of the input - output process , which works for a wide range of parametric variation , from weak to strong coupling regimes , and in the presence or absence of the confined atom .
then we will try to use a single photon to entangle two atoms confined respectively in two spatially separate cavities , based on which further qip tasks could be carried out .
we argue that qip with single photons is an efficient way in the low - q cavity situation , which is considerably different from the high - q cavity and strong - coupling cases .
different from the previous schemes @xcite with photonic polarization unchanged or changed by a phase @xmath0 , the large cavity decay and moderate coupling in our case lead to a certain angle rotation of the photonic polarization after the input - output process , which is called faraday rotation .
the faraday rotation was originally studied in a resonant medium with zeeman level splitting under the radiation of linearly polarized light @xcite and has been recently observed experimentally in cold atom and quantum dot systems @xcite . the key point for the faraday rotation is the bi - refringent propagation of the light through the medium . by using the values from @xcite , we will show the desired faraday rotation for our purpose can be obtained in a two - mode cavity by using single photons with suitable frequencies .
the paper is organized as follows . in sec .
ii , we introduce a general input - output relation for a single photon with respect to a low - q optical cavity confining a single atom under jaynes - cummings model .
we show in sec .
iii the faraday rotation in a two - mode cavity holding a three - level atom by @xmath1-type configuration .
based on the faraday rotation , we may entangle remote atoms by a single photon , as shown in sec .
v is devoted to the application of the generated entanglement .
finally , we end with some discussion and a summary in sec .
in this section , we present the basic input - output relation for a cavity coherently interacting with a trapped two - level atom . under the jaynes - cummings model , we have the following hamiltonian , @xmath2where @xmath3 and @xmath4 are the annihilation and creation operators of the cavity field with frequency @xmath5 , respectively ; @xmath6 , @xmath7 , and @xmath8 are , respectively , inversion , raising , and lowering operators of the two - level atom with frequency difference @xmath9 between the two levels .
consider a single - photon pulse with frequency @xmath10 input in an optical cavity .
the pulse can be expressed by @xmath11 @xcite , where @xmath12 is the normalized pulse shape as a function of @xmath13 , @xmath14 is the pulse duration , @xmath15 , a one - dimensional field operator , is the cavity input operator satisfying the commutation relation @xmath16=\delta ( t - t^{\prime } ) $ ] , and we denote the vacuum of all the optical modes by @xmath17 . in the rotating frame with respect to the frequency of the input pulse , the quantum langevin equation of the cavity mode @xmath3 driven by the corresponding cavity input operator
@xmath18 is @xcite@xmath19a(t)-g\sigma _ { -}(t)-\sqrt{\kappa }
a_{\mathrm{in}}(t ) , \label{eom}\]]where @xmath20 is the cavity damping rate .
moreover , the atomic lowering operator also obeys a similar equation and , in the rotating frame of the frequency @xmath10 , we have , @xmath21\sigma _ { -}(t)-g\sigma _ { z}(t)a(t)+\sqrt{\gamma } \sigma _ { z}(t)b_{\mathrm{in% } } ( t ) , \label{eom2}\]]where @xmath22 , with the commutation relation @xmath23=\delta ( t - t^{\prime } ) $ ] , is the vacuum input field felt by the two - level atom , @xmath24 is the decay rate of the two - level atom .
the input and output fields of the cavity are related by the intracavity field as @xcite @xmath25now assuming a large enough @xmath26 to make sure that we have a weak excitation by the single - photon pulse on the atom initially prepared in the ground state , i.e. , keeping @xmath27 throughout our operation , we can adiabatically eliminate the cavity mode and arrive at the input - output relation of the cavity field,@xmath28[i(\omega _ { 0}-\omega _ { \mathrm{p}})+\frac{\gamma } { 2}]+g^{2}}{% [ i(\omega _ { c}-\omega _ { \mathrm{p}})+\frac{\kappa } { 2}][i(\omega _ { 0}-\omega _ { \mathrm{p}})+\frac{\gamma } { 2}]+g^{2 } } , \label{rela}\]]where @xmath29 is the reflection coefficient for the atom - cavity system , and we have assumed in eq .
( 5 ) that the input field of the two - level atom , @xmath30 , as a vacuum field , gives negligible contribution to the output cavity field @xmath31 .
( [ rela ] ) is a general expression for various cases .
for example , in the case of @xmath32 , eq . ( 5 )
can recover the previous result for an empty cavity @xcite , @xmath33 eq .
( [ rela ] ) also fits very well the results in @xcite : if @xmath34 is dominant with respect to other parameters , @xmath35 would be 1 , implying that the input photon remains unchanged when it is output . using the models in @xcite
, we could explain the dominant @xmath34 case as that , due to the strong resonant coupling between the cavity field and the atom , the energy levels of the cavity will be shifted by the large vacuum rabi splitting , yielding a large detuning between the dressed - cavity - mode and the single photon which is of the same frequency as that of the original cavity . equivalently the total system can be seen as a photonic pulse interacting with a far - detuned bare cavity . as a result of eq .
( [ r0 ] ) we have the reflective coefficient being 1 .
this implies that the photon enters and then leaks out of the cavity without being absorbed by the cavity mode .
in contrast , in the case that the cavity mode is far - detuned with respect to the confined atom , no level shift would occur in the cavity .
since the single photon is of the same frequency as that of the cavity , we can equivalently see the model as the photon interacting resonantly with a bare cavity , which yields @xmath36 from eq .
( [ r0 ] ) . from eq .
( 5 ) the large detuning between the cavity mode and the atom means @xmath32 . as a result
, we have @xmath37 . in this sense
, we may argue that the controlled phase flip designed in @xcite seems a special case of eq .
( [ rela ] ) .
in fact , the key condition for eq .
( 5 ) is @xmath27 .
so as long as @xmath26 is large enough , which makes @xmath38 always satisfied , eq .
( 5 ) should work well even if @xmath34 is bigger than @xmath26 . as this condition was met in @xcite , it is not strange that we could apply eq .
( 5 ) to some physics in @xcite . for more general cases , we plot fig .
[ r ] by eq .
( 5 ) with the values from @xcite for the absolute value and the phase shift of the reflection coefficient .
in the case of an empty cavity , the pulse will experience a perfect reflection in the whole frequency regime , i.e. @xmath39 . the phase shift is @xmath40 at @xmath41 , but reduces to zero rapidly with the frequency of the input pulse deviating from the resonant point . in the case of an atom presented in the cavity
, the coupling between the cavity field and the atom will shift the cavity mode , which leads to the vacuum splitting .
such a splitting also makes the reflection coefficient a corresponding splitting , as shown in the absolute value and the phase shift of the reflective coefficient in fig .
so it is the interaction between the cavity field and the atom that induces weak absorption of the photon with detuned frequency by the cavity mode and also decreases the reflection rate @xmath42 .
however , as shown in fig .
[ r ] , even for a bad cavity with strong damping , the decrease of the reflection rate is very small so that we can still have @xmath43 .
in the preceding section , we did not consider the polarization degrees of freedom of the single photon . in this section ,
we show that , with the general input - output relation , we can obtain a rotation regarding the polarization of the single photon pulse after the input - output process , known as the faraday rotation @xcite . to explain the mechanism of faraday rotation in cavity - qed , we may consider the atom with the level structure as in fig .
[ level1 ] .
the states @xmath44 is the excited state .
we assume that the transitions of @xmath45 and @xmath46 are due to the coupling to two degenerate cavity modes @xmath47 and @xmath48 with left ( ) and right ( ) circular polarization , respectively .
if the atom is initially prepared in @xmath49 , the only possible transition is @xmath50 will take action .
so from eq .
( [ rela ] ) we have the output pulse related to the input one as @xmath51 with @xmath52 the corresponding phase shift determined by the parameter values .
it also means that an input circularly polarized single - photon pulse @xmath53 would only sense the empty cavity . as a result , the corresponding output governed by eq .
( [ r0 ] ) is @xmath54 with @xmath55 a phase shift different from @xmath56 .
therefore , for an input linearly polarized photon pulse @xmath57 , the output pulse is @xmath58the polarization degrees of freedom of a linearly polarized optical field can be characterized by the stokes vector @xmath59 with @xcite,@xmath60 where @xmath61 ( @xmath62 ) with @xmath63 or @xmath64 is the annihilation ( creation ) operator regarding different polarization .
it is easily verified that @xmath65 corresponds to @xmath66 and eq .
( 7 ) could be rewritten as @xmath67 , for which we define @xmath68 to be faraday rotation . similarly , if the atom is initially prepared in @xmath69 , then only the @xmath70 circularly polarized photon could sense the atom , whereas the @xmath71 circularly polarized photon only interacts with the empty cavity .
so we have , @xmath72 where the faraday rotation is @xmath73 .
using the faraday rotation introduced in last section , we could generate entanglement between the atoms in spatially separate cavities with low - q factors , as plotted in fig .
we consider the level structure of each trapped atom as in fig .
[ level](b ) , where @xmath74 and @xmath75 ( @xmath76 ) are the degenerate zeeman sublevels of a typical alkali atom with @xmath77 . as the atom is in resonance with the cavity modes , the possible cavity - mode - induced transitions are @xmath78 ( or @xmath79 ) by absorbing or emitting a ( or ) circularly polarized photon .
the transitions @xmath80 ( @xmath76 ) can be realized by a classical laser pulse . to entangle two atoms confined in spatially separate locations , we encode the qubits in @xmath81 and @xmath82 . as shown in fig .
level(a ) , we input a single photon pulse in superposition of horizontal and vertical polarizations , i.e. , @xmath83 ( which changes to @xmath84 after going through the quarter - wave plate ( qwp ) ) into the first cavity , and then we direct the output pulse by a fiber to the second cavity .
the detection of the output photon from the second cavity , assisted by a qwp and a hwp , would yield the atoms to be entangled . specifically if the two atoms are prepared in @xmath85 with @xmath86 for different cavities , the photonic input and output regarding the first cavity yield , @xmath87which means an entanglement between the photonic and atomic qubits due to different faraday rotations .
the photon going in and then reflected out of the second cavity corresponds to , @xmath88|\psi _ { 2}\rangle \notag \\ & \rightarrow & \alpha _ { 1}\alpha _ { 2}\frac{1}{\sqrt{2}}(e^{i(\phi + \phi ^{\prime } ) } |\mathrm{h}\rangle + e^{i(\phi _ { 0}+\phi _ { 0}^{\prime } ) } |\mathrm{% v}\rangle ) |0\rangle _ { 1}|0\rangle _ { 2 } \notag \\ & & + \beta _ { 1}\beta _ { 2}\frac{1}{\sqrt{2}}(e^{i(\phi _ { 0}+\phi _ { 0}^{\prime } ) } |\mathrm{h}\rangle + e^{i(\phi + \phi ^{\prime } ) } |\mathrm{v}\rangle ) |1\rangle _ { 1}|1\rangle _ { 2 } \notag \\ & & + \alpha _ { 1}\beta _ { 2}\frac{1}{\sqrt{2}}(e^{i(\phi + \phi _ { 0}^{\prime } ) } |%
\mathrm{h}\rangle + e^{i(\phi _ { 0}+\phi ^{\prime } ) } |\mathrm{v}\rangle ) |0\rangle _ { 1}|1\rangle _ { 2 } \notag \\ & & + \beta _ { 1}\alpha _ { 2}\frac{1}{\sqrt{2}}(e^{i(\phi _ { 0}+\phi ^{\prime } ) } |% \mathrm{h}\rangle + e^{i(\phi + \phi _ { 0}^{\prime } ) } |\mathrm{v}\rangle ) |1\rangle _ { 1}|0\rangle _ { 2 } , \notag \\ \label{ewe}\end{aligned}\]]where the actions of the qwps have been included . as we may adjust the frequency of the input pulse to @xmath89 , we actually have @xmath90 from eq .
( [ rela ] ) with @xmath91 and @xmath92 , and we have @xmath93 from eq .
( [ r0 ] ) with @xmath94 .
so the output state becomes@xmath95after the output photon goes through a hwp , which makes @xmath96 and @xmath97 state , yielding a projection onto the atomic state as @xmath98 , \label{e1}\]]where @xmath99 is the normalization constant .
alternatively , we may also detect the photon in the @xmath100 state , yielding @xmath101 , \label{e2}\]]with @xmath102 the normalization constant .
the atomic states we obtained in eqs .
( [ e1 ] ) and ( [ e2 ] ) are entangled with arbitrary amount of entanglement determined by @xmath103 , @xmath104 , @xmath105 and @xmath106 .
it is straightforward to extend above operations to the cases involving three atoms . for example , if the output photon pulse from the second cavity is directed to get in and then reflected out of the third cavity , we can obtain the atomic state , @xmath107,\end{aligned}\]]corresponding to the output state of the photon with polarization @xmath108 , and @xmath109,\end{aligned}\]]with the output state as @xmath110 . by using another hwp ( with a different tilted angle from the previously used one ) to achieve @xmath111 and @xmath112
, we distinguish the states of the output pulse by the single - photon detector .
therefore , we can generate the three - qubit entangled states in eqs .
( 15 ) and ( 16 ) at our will .
evidently , the scheme can be generalized to the situation involving more atoms .
in this section , we carry out some quantum information processing tasks using the generated atomic entanglement .
the key integrant is the efficient conversion between the quantum nodes and the flying qubits .
we will first show a conversion from the atomic entanglement to the photonic entanglement by some single qubit rotations on the atoms .
then we transfer any unknown photonic state to the atom trapped in a distant cavity and transfer an unknown state from one atom to another . supposing that the atoms in two separate cavities have been entangled as in eq .
( [ e1 ] ) with @xmath113 and @xmath114 , we apply two @xmath115-polarized classical laser pulses simultaneously on the two atoms , and the states of the atoms change as , @xmath116 produce a @xmath70- ( or @xmath71- ) circularly polarized photon .
( [ efp ] ) becomes @xmath117 ( or @xmath118 ) is the generated photonic state with @xmath71 ( or @xmath64 ) polarization from the @xmath119th cavity , and the photon will change the polarization to @xmath120 ( or @xmath121 ) after going through the qwp . in eqs .
( 17 ) and ( 18 ) , we have omitted the photonic states in vacuum for simplicity .
the two equations above clearly show that the entanglement is converted from the atomic states to the states of the emitting photons , which is actually a source of entangled photons and will be used for further quantum information processing mission .
our another scheme is to use the atomic entanglement generated above for state transfer from a photon to an atom in distance .
suppose that the atoms have been prepared in a maximally entangled state @xmath122 . as we want to transfer an unknown photonic state @xmath123 ( @xmath124 ) via the atomic entanglement from one side to another , the photon only needs to go through one of the cavities .
first we input the single photon to the cavity in its local side ( for convenience , we call it as first cavity and the other one the second cavity ) , the state of the total system changes as , latexmath:[\]]by detecting the photon polarization and the states of the first atom , the second atom would be projected onto four corresponding states with equal probability . after the local operations conditioned on the measurement results , the unknown state , i.e. ( @xmath133 ) , on the second atom is reconstructed .
eq . ( 5 ) is the key result of the present paper , based on which we have made the schemes for entanglement generation and state transfer .
it is easy to check that eq .
( 5 ) fits very well the numerical results in @xcite . as an analytical expression working for a wide range of parametric values , eq .
( 5 ) should be very useful in the study of cavity qed . the entanglement generation and state transfer in sections iv and v are sketched for the low - q cavities , such as the achieved mtr @xcite or the single - sided cavity @xcite .
one point we have to mention is that we have not yet figured out the controlled logic gate between the two distant atomic qubits based on the faraday rotation , although we could achieve entanglement in between .
so we could not achieve teleportation between the two atoms from the conventional viewpoint @xcite .
anyway we have shown the possibility to transfer an atomic or a photonic qubit state to a distant atomic qubit .
different from the standard teleportation steps , we need local operations on the qubits on both sides as well as classical information to achieve the transfer of a quantum state .
the entanglement of two distant atomic qubits by a single photon in our scheme is more efficient than that by interference of two photons emitted from two respective atoms @xcite . in the latter case ,
the entanglement between the two atomic qubits relies on the two leaking photons reaching at the beam splitter simultaneously , as well as the high efficiency of the two detectors , which strongly restricts the success rate in real implementation .
in contrast , as only a single photon is involved in our scheme , there is no requirement for simultaneous detection .
so our scheme should be more efficient than those using two detectors in the case of the large inefficiency of current single - photon detector .
the imperfection in our scheme is the photon loss , which is also a problem in previously published schemes with photon interference .
the photon loss occurs due to the cavity mirror absorption and scattering , the fiber absorption , and the inefficiency of the detector .
as the successful detection of the photon ensures the accomplishment of our implementation , the photon loss actually only affects the efficiency of the scheme , but not the fidelity of the entanglement generation . moreover , even if we implement the scheme with low success rate , due to the highly efficient single - photon source , such as 10,000 single photons per second @xcite , we are able to accomplish our schemes within a short time
. we may simply assess the implementation of our schemes , where the failure rate due to atomic decay is about @xmath134 .
moreover , the current dark count rate of the single - photon detector is about 100 hz , which can reduce the efficiency of our scheme by a factor of @xmath135
. other imperfection rate , regarding the photon absorption by the fiber and the scattering of the cavity mirror , can be assumed to be 6@xmath136 .
thus the success rate of our implementation should be @xmath137 , where the square is due to two atoms involved . by using the generation rate @xmath138 @xmath139 of single photons ,
the two atoms can be entangled within 2 sec , provided that the two atoms are not very distant , i.e. , without considering the time of the photon traveling in between .
evidently , our scheme is in principle scalable with the single photon going through the spatially separate cavities one by one .
however , with more atoms ( confined in cavities ) involved , the photon loss would be more serious and thereby it will take a longer time for a successful event . before ending the paper , we would like to reiterate the difference of our present work from the relevant work published previously @xcite .
the previous schemes prefer to work in strong coupling condition , i.e. , @xmath140 .
the fidelity of the operations decreases significantly with the decrease of @xmath141 .
in contrast , our study focuses on the input - output process of the low - q cavity .
we have not only analytically presented an important expression for the photonic reflection , which straightforwardly leads to our understanding of faraday rotation , but also demonstrated a way to entanglement generation and state transfer in the case of low - q cavity .
we argue the impossibility of controlled -not gating in the case under our consideration , which is different from the strong - coupling or other cases considered in @xcite .
meanwhile , we have shown the preference with a single photon to carry out qip tasks . in summary
, the general reflection rate of input - output process we have analytically presented can be applied to different cases for cavity qed : with and without atoms confined , and with strong or weak coupling , and particularly useful for weak coupling case .
based on this general reflection rate and a recently achieved mtr , we have proposed a scheme using cavities with low - q factors to entangle distant atoms by a single photon , to generate entangled photons , and to transfer quantum state to a distant qubit .
we argue that our work would be useful for qip in cavity qed with current technology .
the work is supported by nus research grant no .
r-144 - 000 - 189 - 305 .
j.h.a . also thanks the financial supports of the nnsf of china under grant no .
10604025 and the fundamental research fund for physics and mathematics of lanzhou university under grant no .
lzu05 - 02 .
is partially supported by nnsf of china under grant no .
10774163 .
note added : after completing the manuscript , we become aware of a recent work @xcite with some similarities for quantum dots in micropillar cavities .
but our study on the cavity - atom system , strongly relevant to the latest progress of cavity qed experiments , is feasible with current technology .
a. rauschenbeutel , g. nogues , s. osnaghi , p. bertet , m. brune , j. m. raimond , and s. haroche , phys . rev .
lett . * 83 * , 5166 ( 1999 ) ; s. osnaghi , p. bertet , a. auffeves , p. maioli , m. brune , j. m. raimond , and s. haroche , phys . rev . lett . * 87 * , 037902 ( 2001 ) .
duan and h. j. kimble , phys .
* 90 * , 253601 ( 2003 ) ; x .-
feng , z .- m .
zhang , x .- d .
li , s .- q .
gong , and z .- z .
xu , ibid . * 90 * , 217902 ( 2003 ) ; x. b. zou , k. pahlke , and w. mathis , phys .
a * 69 * , 052314 ( 2004 ) ; s. d. barrett and p. kok , ibid . * 71 * , 060310(r ) ( 2005 ) ; y. l. lim , s. d. barrett , a. beige , p. kok , and l. c. kwek , ibid . * 73 * , 012304 ( 2006 ) ; l. jiang , j. m. taylor , a. s. srensen , and m. d. lukin , phys . rev . a * 76 * , 062323 ( 2007 ) .
z. zhao , y .- a .
chen , a .-
zhang , t. yang , h. j. briegel , and j .- w .
pan , nature ( london ) * 430 * , 54 ( 2004 ) ; p. walther , k. j. resch , t. rudolph , e. schenck , h. weinfurter , v. vedral , m. aspelmeyer , and a. zeilinger , _
ibid_. * 434 * , 169 ( 2005 ) ; d. leibfried , e. knill , s. seidelin , j. britton , r. b. blakestad , j. chiaverini , d. b. hume , w. m. itano , j. d. jost , c. langer , r. ozeri , r. reichle , and d. j. wineland , ibid . * 438 * , 639 ( 2005 ) ; h. hffner , w. hnsel , c. f. roos , j. benhelm , d. chek - al - kar , m. chwalla , t. krber , u. d. rapol , m. riebe , p. o. schmidt , c. becher , o. ghne , w. dr , and r. blatt , ibid . * 438 * , 643 ( 2005 ) ; c. w. chou , h. de riedmatten , d. felinto , s. v. polyakov , s. j. van enk , and h. j. kimble , ibid . * 438 * , 828 ( 2005 ) .
duan and h. j. kimble , phys .
* 92 * , 127902 ( 2004 ) ; y .- f .
xiao , x .-
lin , j. gao , y. yang , z .- f .
han , and g .- c .
guo , phys .
a * 70 * , 042314 ( 2004 ) ; j. cho and h .- w .
lee , phys .
* 95 * , 160501 ( 2005 ) ; l .- m .
duan , b. wang , and h. j. kimble , phys .
a * 72 * , 032333 ( 2005 ) .
lin , z .- w .
zhou , m .- y .
ye , y .- f .
xiao , and g .- c .
guo , phys rev a * 73 * , 012323 ( 2006 ) ; z.j .
deng , m. feng , and k.l .
gao , ibid . * 75 * , 024302 ( 2007 ) ; z. j. deng , x. l. zhang , h. wei , k. l. gao , and m. feng , ibid . * 76 * , 044305 ( 2007 ) ; h. wei , z. j. deng , x. l. zhang , and m. feng , ibid . * 76 * , 054304 ( 2007 ) .
j. beugnon , m. p. a. jones , j. dingjan , b. darquia , g. messin , a. browaeys , and p. grangier , nature * 440 * , 779 ( 2006 ) ; y .- a .
chen , s. chen , z .- s .
yuan , b. zhao , c .- s .
chuu , j. schmiedmayer , and j .- w .
pan , nat .
phys . , * 4 * , 103 ( 2008 ) .
this unitary rotation can be achieved by a half - wave plate tilted by 22.5@xmath142 with respect to the horizontal direction . with different tilted angles ,
different unitary operations are realized on the photonic polarization .
we have noticed a previous publication ( phys .
lett * 94 * , 107401 ( 2005 ) ) to teleport quantum state of electron spin between two quantum dots by means of faraday rotation with similar operations to ours in sec .
v ( c ) . but strictly speaking , the teleportation accomplished there without bell - state measurement is different from the conventionally defined teleportation .
so we prefer to call our operation state tranfer , instead of teleportation . on the other hand ,
as bell - state measurement can be replaced by a controlled - not gate plus some single - qubit rotations and measurements , any conditional gate between any two qubits , e.g. , between a photonic qubit and an atomic qubit , would achieve a teleportation .
but it seems unrealizable in our present treatment unless we assume @xmath143 as done in duan04,feng .
we have also noticed the conditional phase gate in kimble95 with two input photons by using both faraday rotation and bi - photonic nonlinearity , which is beyond our model . | both cavity qed and photons are promising candidates for quantum information processing .
we consider a combination of both candidates with a single photon going through spatially separate cavities to entangle the atomic qubits , based on the input - output process of the cavities .
we present a general expression for the input - output process regarding the low - q cavity confining a single atom , which works in a wide range of parameters .
focusing on low - q cavity case , we propose some schemes for quantum information processing with faraday rotation using single photons , which is much different from the high - q cavity and strong coupling cases . |
the braid group @xmath1 is a group defined by the presentation @xmath6 the monoid generated by the positive generators @xmath7 is called _ the positive braid monoid _ and is denoted by @xmath8 .
an element of @xmath8 is called a _
positive braid_. the braid group @xmath1 , first introduced by artin , appears in various fields of mathematics and have been studied from many points of view . in 1990
s a new feature of the braid groups , the left - invariant total orderings called _ the dehornoy ordering _ was discovered by dehornoy @xcite .
after the discovery of the orderings , numerous studies have been done from both algebraic and geometrical prospectives and the subject of braid orderings and related topics have been developed rapidly @xcite,@xcite . in @xcite ,
infinite families of left - invariant total orderings of the braid groups , called _ thurston type orderings _ are constructed via the hyperbolic geometry .
this family of orderings contains the dehornoy ordering , hence it is an extension of the dehornoy ordering .
it is known that thurston type orderings have a property called the property @xmath9 , and consequently they define the well - ordering when they are restricted to the positive braid monoid @xmath8 .
one of the problems of braid orderings is to determine the order type of the well - ordered set @xmath10 for a thurston type ordering @xmath11 , and to construct a method to compute ordinals for each element in @xmath8 . for the dehornoy ordering case
, burckel solved these problems by using a correspondence between positive braid words and rooted labeled trees @xcite,@xcite .
he introduced a normal form of a positive braid called _
burckel s normal form _ , which makes it possible to compare the dehornoy ordering and to compute ordinals .
in @xcite , dehornoy gives alternative description of burckel s normal form by introducing the @xmath5-normal forms ( alternate normal form ) of positive braids .
he utilized the alternate decomposition , which uses the garside structure of braid groups .
he showed that for the positive braid monoid , the @xmath5-normal form coincides with burckel s normal form .
however , the connection between the @xmath5-normal forms and the dehornoy ordering is indirect .
it depends on burckel s result , so there still lie gaps to understand the dehornoy ordering on positive braid monoid directly . in this paper , we give a new combinatorial description for special kinds of thurston type orderings , called _ finite type orderings_. the dehornoy ordering is a typical example of finite type orderings so our results extend the results of dehornoy and burckel . for a positive braid word @xmath12 with additional structure , which we call a _ tower of subword decompositions _ , we define the code , which is a sequence of parenthesized non negative integers .
the set of codes has two lexicographical orderings @xmath13 and @xmath14 . the @xmath4-normal form of a positive braid @xmath15 is defined as a positive word representative of @xmath15 whose code is maximal with respect to the ordering @xmath14 .
this maximal code is denoted by @xmath16 . using these notions ,
our main theorem is stated as follows .
[ thm : main ] let @xmath11 be a finite thurston type ordering of @xmath1 .
then for @xmath17 , @xmath18 holds if and only if @xmath19 holds .
thus on the positive braid monoid @xmath8 , the structure of finite thurston type orderings is simple .
it can be regarded as a composition of two lexicographical orderings .
this @xmath4-normal form description of finite thurston type orderings makes it possible to compare the ordering of braids by purely algebraic and combinatorial ways . using the @xmath4-normal form description ,
we determine the order type of the well - ordered set @xmath10 for all finite thurston type orderings @xmath11 .
this result is a partial answer to the question listed in @xcite and @xcite and provides an alternative proof of burckel s results .
[ thm : ordertype ] let @xmath11 be a finite thurston type ordering of @xmath1 .
then the order type of the well - ordered set @xmath10 is @xmath3 .
the @xmath4-normal form can be seen as an extension of burckel s normal form and dehornoy s @xmath5-normal form .
we define the tail - twisted @xmath5-normal form of positive braids , which is an extension of the @xmath5-normal form .
we show the tail - twisted @xmath5-normal forms coincide with the @xmath4-normal forms for special kinds of finite type orderings , called _
normal _ finite type orderings . using this algebraic formulation of the @xmath4-normal forms , we give a computational complexity of @xmath4-normal forms and finite thurston type orderings .
[ thm : complexity ] let @xmath11 be an arbitrary finite thurston type ordering of @xmath1 . 1 .
the @xmath4-normal form @xmath20 of a positive @xmath21-braid @xmath15 is computed by the time @xmath22 , where @xmath23 is the word length of @xmath15 .
2 . for an @xmath21-braid ( not necessarily positive ) @xmath15 , whether @xmath24 holds or not can be decided by the time @xmath25 where @xmath23 is the word length of @xmath15 .
thus , the @xmath4-normal form argument provides an efficient algorithm to compare finite thurston type orderings in terms of artin generators .
the plan of the paper is as follows . in section 2 , we summarize the definitions and basic facts about thurston - type orderings of the braid groups . in section 3
we introduce the code of braid words and define the @xmath4-normal form .
the proof of theorem [ thm : main ] and theorem [ thm : ordertype ] will be given in section 4 .
section 5 is devoted to the study of the relationships between the @xmath4-normal form and dehornoy s @xmath5-normal form .
we define the tail - twisted @xmath5-normal form and show that it coincides with the @xmath4-normal form , and prove theorem [ thm : complexity ] .
finally we mention other aspects of our code constructions .
there is another submonoid @xmath26 of @xmath1 called _ the dual braid monoid _ , which has similar properties of the positive braid monoid @xmath8 .
indeed , the dual braid monoid also defines the garside structure on the braid group @xmath1 , and two monoids @xmath8 and @xmath26 are closely related @xcite,@xcite .
fromentin studied the dehornoy orderings via the dual braid monoids using the rotating normal forms , which corresponds to dehornoy s @xmath5-normal form in dual braid monoids settings @xcite,@xcite,@xcite .
our code constructions and the @xmath4-normal forms equally work for the dual braid monoids with an appropriate modification . in particular , all results in this paper holds for dual braid monoids as well .
however , in the dual braid monoid case we need a new method to prove the counterpart of theorem [ thm : main ] , the @xmath4-normal form description of the ordering .
we will describe the @xmath4-normal form construction for the dual braid monoid in the subsequent paper @xcite .
+ * acknowledgments . *
the author would like to express his gratitude to professor toshitake kohno for his encouragement and helpful suggestions .
he also wishes to thank to bert wiest and matthieu calvez for careful reading of an early version of the paper , and for pointing out mistakes .
this research was supported by jsps research fellowships for young scientists .
first we briefly review a construction of thurston type orderings . for details , see @xcite .
let @xmath27 be the @xmath21-punctured disc .
the braid group @xmath1 is naturally identified with the relative mapping class group @xmath28 , which is the group of isotopy classes of homeomorphisms of @xmath27 whose restrictions on @xmath29 are identity maps . the generator @xmath30 corresponds to the isotopy class of the positive half dehn - twist along the straight line connecting the @xmath31-th and the @xmath32-st puncture points .
let us choose a complete hyperbolic metric on @xmath27 .
then the universal covering @xmath33 is isometrically embedded in the hyperbolic plane @xmath34 . by attaching the points at infinity to @xmath35
, we obtain a topological disc . by abuse of notation
, we still denote this disc by the same notation @xmath35 .
take a lift of the base point @xmath36 and let @xmath37 be the connected component of @xmath38 which contains @xmath36 .
then , @xmath39 is identified with the real line @xmath40 . for a point @xmath41
, we define the relation @xmath42 on the braid group @xmath1 as follows . for @xmath43,[g ] \in b_{n}$ ] , let us choose their representative homeomorphisms @xmath44 and take their lifts @xmath45 so that they fix the point @xmath36 .
we define @xmath43 < _ { x } [ g]$ ] if @xmath46 holds under an orientation - preserving identification of @xmath39 with the real line @xmath40 .
although there are many choices of representative homeomorphisms @xmath47 and @xmath48 , the restrictions of their lifts are independent of these choices , so @xmath42 is well - defined .
the relation @xmath42 defines a left - invariant partial ordering of @xmath1 .
if we choose a point @xmath49 in a nice way , @xmath42 becomes a total ordering .
we call left - invariant total orderings constructed in this way _ thurston type orderings_. although there are many choices of a hyperbolic metric and a base point in the above construction , constructed families of orderings are independent of these choices .
it is known that thurston type orderings fall into two types , _ finite type _ and _ infinite type_. in next section , we formulate a finite type ordering as an ordering which can be described by curve diagrams . in this section
we give an alternative description of a special kind of thurston type orderings using a curve diagram .
the curve diagram formulation gives more accessible meanings and is closely related to our combinatorial description .
let @xmath50 be a diagram in @xmath27 , which consists of a union of closed oriented , embedded arcs @xmath51 satisfying the following properties .
* interiors of @xmath52 are disjoint to each other . *
the initial point of @xmath52 lies at @xmath53 and the end point of @xmath52 lies at puncture points or @xmath54 . *
each component of @xmath55 is a disc or an one - punctured disc .
we call such a diagram _ curve diagram_. see figure [ fig : curvediagram ] .
we say two curve diagrams @xmath50 and @xmath56 are tight if they form no bigon .
a bigon is an embedded disc whose boundary consists of two subarcs @xmath57 , @xmath58 .
if we equip a complete hyperbolic metric on @xmath27 and realize both @xmath50 and @xmath56 as a union of geodesics , then @xmath50 and @xmath56 become tight .
thus , we can always isotope given two curve diagrams so that they are tight . for distinct braids
@xmath59 , if we put the image of the curve diagrams @xmath60 and @xmath61 tight , then they must diverge at some point .
we define @xmath62 if the image @xmath61 moves the left side of @xmath60 at the first divergence point .
the relation @xmath63 defines a left - invariant total ordering of @xmath1 .
we say a thurston - type ordering @xmath11 is of _ finite type _ if it coincides with the ordering @xmath63 for some curve diagram @xmath50 .
although this definition of finite type ordering is different from that in @xcite,@xcite , @xcite , this definition agrees with the usual definition .
conversely , every ordering constructed by a curve diagram is always a thurston - type ordering .
we remark that the notion of curve diagrams and its defining orderings are generalized to other mapping class groups for arbitrary surface with non - empty boundaries .
see @xcite for such a construction of left - invariant total orderings on mapping class groups .
we say a curve diagram @xmath50 is _ normal _ if all arcs @xmath64 are vertical and oriented upwards . here
we say an arc @xmath52 is vertical if the arc @xmath52 is written by the equation @xmath65 by using the @xmath49-@xmath66 coordinate , regarding @xmath27 as the unit disc removed puncture points lying on the @xmath49-axis .
see figure [ fig : curvediagram ] . for a normal curve diagram @xmath50 of @xmath27
, we define the integer @xmath67 so that @xmath52 lies between the @xmath67-th and the @xmath68-st punctures .
then the normal curve diagram @xmath50 is represented as the permutation of @xmath69 integers @xmath70 .
a finite thurston type ordering @xmath11 is called _ normal _ if @xmath71 for some normal curve diagram @xmath50 . in this paper
, we mainly consider a normal finite thurston - type ordering .
this is justified by the following facts .
two left invariant total orderings @xmath11 and @xmath72 of a group @xmath73 are called _ conjugate _ if there exists an element @xmath74 such that @xmath75 is equivalent to @xmath76 for all @xmath77 .
we call such @xmath48 a _ conjugating element _ between @xmath11 and @xmath72 . in @xcite , it is shown that the number of conjugacy classes of finite thurston type orderings on @xmath1 is finite and each conjugacy class of orderings contains at least one normal finite type ordering .
for the braid group @xmath1 and the positive braid monoid @xmath8 , there is a good property which makes it easier to handle conjugate orderings . [
lem : conjpos ] let @xmath11 and @xmath72 be two left - invariant total orderings of the braid group @xmath1 . if @xmath11 and @xmath72 are conjugate , then we can choose a conjugating element @xmath78 as a positive braid .
let @xmath79 be a conjugating element between @xmath11 and @xmath72 , and let @xmath80 be the garside fundamental braid , defined by @xmath81 it is known that @xmath82 is central in @xmath1 and @xmath83 is a positive braid for sufficiently large @xmath84 @xcite .
thus @xmath83 is a positive conjugating element if @xmath85 is sufficiently large .
we closed this section by giving examples of normal finite thurston type orderings .
[ exam : dtype ] let @xmath86 be a curve diagram as shown in the left diagram of figure [ fig : curvediagram ] , which corresponds to the trivial permutation @xmath87 .
the ordering @xmath88 defined by @xmath86 is called the _
dehornoy ordering_. algebraically , the dehornoy ordering @xmath88 is defined as follows . for @xmath59
, we define @xmath89 if @xmath90 admits a word representative which contains no @xmath91 but contains at least one @xmath30 for some @xmath31 @xcite .
next we consider the curve diagram @xmath92 in the middle diagram of figure [ fig : curvediagram ] , which corresponds to the permutation @xmath93 .
the ordering @xmath94 defined by @xmath92 is called the _ reverse of the dehornoy ordering_. in some papers ( for example , @xcite ) the dehornoy ordering is used to represent the reverse of the dehornoy ordering .
this is not a serious difference because these two orderings are conjugate by the garside fundamental braid @xmath80 .
that is , @xmath95 is equivalent to @xmath96 . [
exam : ttype ] let us present the most simple finite thurston type ordering which is not conjugate to the dehornoy ordering .
let @xmath50 be the curve diagram of @xmath97 , as shown in the right diagram of figure [ fig : curvediagram ] , corresponding to the permutation @xmath98 .
it is known that the ordering defined by @xmath50 is not conjugate to the dehornoy ordering @xmath88 @xcite,@xcite .
thurston type orderings are generalization of the dehornoy ordering .
many properties of the dehornoy orderings hold for thurston type orderings as well .
the most important property we use in the paper is the property @xmath9 .
let @xmath11 be a thurston type ordering of @xmath1 .
then @xmath99 holds for all @xmath59 and @xmath100 . roughly speaking
, the property @xmath9 means we can freely insert positive generators @xmath101 at arbitrary points without decreasing the ordering .
the property @xmath9 will be essentially used to prove our description of orderings .
using property @xmath9 , we can prove the following corollary by the same argument as for the dehornoy ordering case .
( see ( * ? ? ?
* chapter1 ) ) for every thurston type ordering @xmath11 of the braid group @xmath1 , @xmath10 is a well - ordered set .
in this section , we define the codes and the @xmath4-normal forms and provide various examples .
first we consider a normal finite thurston - type ordering and after we extend the results about normal orderings to general one .
let @xmath11 be a normal finite thurston type ordering on @xmath1 , define by the curve diagram @xmath50 which corresponding to the permutation @xmath102 . throughout this section we use the following notation .
we always denote @xmath103 by @xmath104 .
let @xmath105 be a flip homomorphism defined by the conjugate of the garside fundamental braid @xmath80 .
that is , the map @xmath106 is defined by @xmath107 or equivalently , @xmath108 .
similarly , we define the shift map @xmath109 by @xmath110 for @xmath111 and @xmath112 . for positive braids
@xmath113 , we denote the submonoid of @xmath8 generated by @xmath114 by @xmath115 .
first we define the code of a positive braid word with additional information which we call the _ tower of subword decomposition_. our starting point is the positive @xmath116-braid monoid @xmath117 . there is only one normal finite thurston type ordering , namely the dehornoy ordering @xmath88 of @xmath117 . for a positive 2-braid word @xmath118
, we define the code of @xmath12 with respect to the ordering @xmath88 by the positive integer @xmath119 .
that is , we define @xmath120 .
assume that we have already defined the code of positive @xmath31-braid words with respect to arbitrary normal finite thurston type orderings of @xmath121 for all @xmath122 .
let @xmath11 be a normal finite thurston type ordering of @xmath1 .
now we define the code of positive @xmath21-braid word @xmath12 as follows .
first of all , for @xmath123 , let us define two integers @xmath124 and @xmath125 by @xmath126 and let @xmath127 .
the set @xmath128 is nothing but the set of integers of the connected component of @xmath129 - \{k(1)+\frac{1}{2 } , \ldots , k(i)+\frac{1}{2}\}$ ] which contains @xmath130 .
for example , @xmath131 , @xmath132 . now take a decomposition of the word @xmath12 into the products of subwords as @xmath133 where @xmath134 and take a subword decomposition of @xmath135 as @xmath136 where @xmath137 .
let @xmath88 be the dehornoy ordering of @xmath138 .
observe that @xmath139 for @xmath140 .
thus from our inductive hypothesis , the code @xmath141 is defined if we assign the tower of subword decompositions of @xmath142 .
let us denote by @xmath143 the code @xmath144 .
let @xmath145 be the restriction of the ordering @xmath11 on @xmath146 . under the identification of the shift map @xmath147
, the ordering @xmath145 can be seen as a normal finite thurston type ordering of @xmath148 .
now , by giving the tower of subword decompositions of @xmath149 , the code @xmath150 are defined .
we denote by @xmath151 the code @xmath152 .
finally , let @xmath153 be the restriction of the ordering @xmath11 to @xmath154 . by identifying @xmath154 with @xmath155 by the @xmath156-fold shift map @xmath157 , we regard @xmath153 as a normal finite thurston type ordering of @xmath155 . now by giving a tower of subword decompositions , the codes @xmath158 are defined .
we denote this code by @xmath159 .
now assume that we have assigned the subword decompositions of each subword @xmath142 and @xmath160 so that all of the codes @xmath161 are defined .
we call this iteration of subword decompositions a _ tower of subword decompositions_. then we define the code @xmath162 of the positive word @xmath12 with the tower of subword decompositions as the sequence of codes of subwords @xmath163 this completes the definition of codes .
+ we remark that to define the codes , it is not sufficient to choose a positive braid word .
we need to indicate the tower of subword decompositions . to indicate a subword decomposition of @xmath12
, we use the symbol @xmath164 .
for example , to express the decomposition of @xmath165 with @xmath166 and @xmath167 , we use a notation @xmath168 .
let @xmath169 be the set of all codes of positive @xmath21-braid words with respect to a normal finite thurston type ordering @xmath11 .
next we define two lexicographical orderings @xmath13 and @xmath14 of the set @xmath169 . to this end , we regard the code of a positive @xmath21-braid words as a left - infinite sequence of the code of subwords @xmath170 , with only finitely many codes @xmath161 are non - trivial .
this means , for example , we regard a code @xmath171 of a 2-braid word @xmath172 as a left infinite sequence of integers @xmath173 , not as a single integer @xmath171 .
let @xmath11 be a normal finite thurston type ordering .
the _ lexicographical ordering from left _ @xmath13 and the _ lexicographical ordering from right _
@xmath13 are total orderings of @xmath169 defined by the following way . as in the definition of codes , we begin with the case @xmath174 . for two codes @xmath119 and @xmath175 in @xmath176 , we define @xmath177 if and only if @xmath178 , and @xmath179 if and
only if @xmath178 . assume that we have already defined @xmath13 and @xmath14 for all normal finite thurston type orderings @xmath11 of @xmath121 for all @xmath122 .
let @xmath180 , @xmath181 be two codes in @xmath182 .
we define @xmath183 if @xmath184 similarly , we define @xmath185 if @xmath186 we are now ready to define the _
@xmath4-normal form _ of a positive braids .
first of all let @xmath11 be a normal finite thurston - type ordering .
for a positive braid @xmath187 , the _
@xmath4-normal form _ of @xmath15 with respect to the ordering @xmath11 is the positive word representative of @xmath15 which has the tower of subword decompositions whose code is maximal among the set of all codes of positive word representatives of @xmath15 , with respect to the ordering @xmath14 .
since the number of positive braid word representatives of a given positive braid is finite , the @xmath4-normal forms is always uniquely determined .
we denote the @xmath4-normal form of a positive @xmath21-braid @xmath15 with respect to a normal finite thurston type ordering @xmath11 by @xmath20 or simply @xmath188 , and its code by @xmath16 or @xmath189 . recall that from lemma [ lem : conjpos ] , for conjugate left invariant total orderings @xmath11 and @xmath72 of @xmath1 , we can always choose a positive conjugating element @xmath78 .
using this fact , we define the @xmath4-normal form for arbitrary finite thurston type orderings .
let @xmath11 be a finite thurston type ordering on @xmath1 which is conjugate to a normal finite thurston type ordering @xmath190 .
let @xmath191 be a positive word representative of a positive conjugating element between @xmath11 and @xmath190 .
fix such @xmath190 and @xmath191 . for a positive @xmath21-braid @xmath15 ,
the @xmath4-normal form of @xmath15 with respect to the ordering @xmath11 is a braid word @xmath192 .
the code of @xmath15 , denoted by @xmath16 , is defined by @xmath193 . by definition , the set of codes @xmath169 is a subset of @xmath194 .
we define the lexicographical orderings @xmath13 and @xmath14 of @xmath169 as the restriction of the orderings @xmath13 and @xmath14 of @xmath194 to the subset @xmath194 . for a general finite thurston type ordering ,
the @xmath4-normal form of a positive braid is no longer a positive braid word .
moreover , we need to fix @xmath190 and @xmath191 , hence for a general orderings , the @xmath4-normal form is not canonically defined . before proceeding to the proof of theorem [ thm : main ] , we give some examples of codes and the @xmath4-normal forms , which give a better understanding of the @xmath4-normal form description .
[ exam : dehornoy3 ] first of all , we consider the first non - trivial case , the dehornoy ordering of 3-braids . for a positive 3-braid word @xmath12
, first we take a subword decomposition @xmath133 where @xmath195 for even @xmath31 and @xmath196 for odd @xmath31 .
the above decomposition can be explicitly written by @xmath197 and the code of @xmath12 is given by a sequence of integers @xmath198 now we give a concrete example .
let @xmath199 .
then the code of @xmath12 with respect to this subword decomposition ( @xmath200 , @xmath201 , @xmath202 , @xmath203 ) is given by @xmath204 .
this is not the @xmath4-normal form .
the @xmath4-normal form of the braid @xmath12 is @xmath205 and its code is @xmath206 .
[ exam : dehornoy4 ] next we proceed to more complicated examples , the dehornoy ordering of @xmath207 . as in example
[ exam : dehornoy3 ] , first we take a subword decomposition of a positive @xmath208-braid word @xmath12 as @xmath209 where @xmath210 for even @xmath31 and @xmath211 for odd @xmath31 . by definition ,
@xmath212 , @xmath213 and @xmath214 , @xmath215 .
thus , the subword decomposition of @xmath135 is written as @xmath216 where @xmath217 and @xmath218 .
it is easy to see the orderings @xmath219 and @xmath220 are also the dehornoy orderings .
since we have already described a code of the dehornoy ordering of @xmath221 , we can explicitly write each code @xmath161 . now we give a concrete example .
let @xmath12 be a positive @xmath208-braid word with a subword decomposition @xmath222 that is , @xmath203 , @xmath223 , @xmath224 , @xmath225 , and @xmath226 .
we take a subword decomposition of @xmath135 as @xmath227 . for a sake of simplicity , we choose the subword decompositions of each @xmath142 so that they have the maximal code with respect to @xmath14 among all subword decompositions of @xmath142 .
then , codes @xmath161 are given as follows .
@xmath228 consequently , the code of @xmath12 is given by @xmath229 the word @xmath12 is not the @xmath4-normal form .
in fact , the @xmath4-normal form of @xmath12 is given by @xmath230 , and the code is given by @xmath231 [ exam : thurston4 ] finally , we study the thurston type ordering @xmath11 defined in example [ exam : ttype ] , which corresponds to the permutation @xmath98 .
thus , @xmath232 , @xmath233 and @xmath214 , @xmath215 .
for a positive 4-braid word @xmath12 , we take a subword decomposition of the form @xmath234 where @xmath235 if @xmath31 is odd and @xmath236 if @xmath31 is even . since @xmath237 , @xmath238 .
first observe that the restriction of the ordering @xmath0 to @xmath239 is the dehornoy ordering of @xmath221 , under the identification by the shift map @xmath240 .
thus , the code of @xmath149 is given as a code with respect to the dehornoy ordering of @xmath221 .
next we consider a subword decomposition of @xmath135 . since @xmath232 , @xmath233 and @xmath214 , @xmath215 , the subword decomposition of @xmath135
is given by @xmath241 where @xmath242 and @xmath218 .
thus , for the word @xmath243 , @xmath244 and @xmath245 .
now we give a concrete example .
let @xmath12 be the positive 4-braid word which appeared in example [ exam : dehornoy4 ] .
take a subword decomposition @xmath246 that is , the decomposition is defined by @xmath247 , @xmath248 , and @xmath201 . from the above observations , each @xmath249
is given as follows .
@xmath250 summarizing , the code of @xmath12 is given by @xmath251 this is not the @xmath4-normal form
. the @xmath4-normal form of the braid @xmath12 with respect to the ordering @xmath0 is given by @xmath252 , and its code is given by @xmath253 as example [ exam : dehornoy4 ] and [ exam : thurston4 ] suggest , different orderings give completely different @xmath4-normal forms .
in this section we proof theorem [ thm : main ] and theorem [ thm : ordertype ] . throughout this section , we use the same notation as in the section 3 .
unless otherwise stated , we always consider a normal finite thurston type ordering @xmath11 which is defined by a normal curve diagram @xmath50 represented by the permutation @xmath254 .
we always put @xmath255 , and all the @xmath4-normal forms are considered as the @xmath4-normal forms with respect to the normal ordering @xmath11 .
first of all we introduce a partial ordering @xmath256 , which is bi - invariant under the multiplication of @xmath1 from both sides .
let @xmath257 be positive @xmath21-braid words .
we define @xmath258 if @xmath12 is obtained from @xmath259 by inserting positive generators @xmath260 . from property @xmath9 , if @xmath258 then @xmath261 holds for all thurston type orderings @xmath262 . more strongly , if @xmath258 then @xmath263 holds for all @xmath264braids @xmath265 and for all thurston type orderings @xmath262 .
thus , we can regard the relation @xmath256 as a bi - invariant part of thurston type orderings .
now we begin with studying properties of the @xmath4-normal form .
[ lem : observation0 ] let @xmath266 be a @xmath4-normal form . 1 . if @xmath267 and @xmath268 , then @xmath149 is non - empty and the last letter of @xmath149 is @xmath269 .
2 . for all @xmath270 ,
the last letter of @xmath142 is @xmath271 if @xmath31 is odd and @xmath272 if @xmath31 is even .
we only prove ( 2 ) for odd @xmath31 .
the proofs of ( 1 ) and ( 2 ) for even @xmath31 are similar .
first assume that @xmath142 is an empty word .
let @xmath273 be the last letter of @xmath274 .
if @xmath275 , then a subword decomposition of @xmath12 defined by @xmath276 defines the bigger code with respect to @xmath14 . if @xmath277 , then a decomposition @xmath278 also defines the bigger code
. thus @xmath142 must be non - empty .
now assume that the last letter of @xmath142 is not @xmath271 .
then the last letter of @xmath142 is @xmath273 @xmath279 , so the subword decomposition @xmath280 defines the bigger code , which is a contradiction .
we remark that in the case @xmath281 , @xmath149 must be an empty word .
thus , from now on , we neglect the term @xmath149 if @xmath281 . by lemma [ lem : observation0 ] ,
we can write the @xmath4-normal form in the following form . @xmath282 thus , the code of the @xmath4-normal form must have the following form .
@xmath283 where @xmath284 are non - zero codes .
we call zeros @xmath285 in the code of the @xmath4-normal form _
trivial zeros_. to consider the code of @xmath4-normal forms , we do not need to consider trivial zeros , so we always neglect trivial zeros .
next we study each word @xmath286 more precisely .
[ lem:1stobservation ] let @xmath12 be the @xmath4-normal form and put @xmath287 as the above . 1 .
if @xmath288 and @xmath268 then @xmath289 .
2 . for all @xmath290 , @xmath291
we only prove ( 2 ) for odd @xmath31 .
the proof of ( 1 ) and ( 2 ) for even @xmath31 are similar .
suppose @xmath292 .
then @xmath293 so @xmath286 commutes with @xmath272 . by lemma
[ lem : observation0 ] , the last letter of @xmath294 is @xmath272 .
thus , we obtain a new word decomposition @xmath295 which defines the bigger code .
thus , @xmath296 .
now let us write @xmath297 , where @xmath298 . by the similar arguments , we obtain @xmath299 .
iterating this argument , we conclude that @xmath300 . in some special cases ,
the above subword estimation results can be improved .
[ lem : subword ] let @xmath301 be a @xmath4-normal form and @xmath140 . 1 . if @xmath302 , @xmath303 and the last letter of @xmath304 is not @xmath271 , then @xmath305 2 .
if @xmath306 is odd and the last letter of @xmath307 is not @xmath272 , then @xmath308 3 .
if @xmath306 is even and the last letter of @xmath307 is not @xmath271 , then @xmath309 we only prove ( 3 ) .
other cases are similar . from lemma [ lem:1stobservation ] , we already know that @xmath310 .
thus , the word @xmath286 is of the form @xmath311 where @xmath312 , possibly empty .
put @xmath313 .
first we observe @xmath314 . because otherwise , the last letter of @xmath307 commutes with @xmath271 ,
so we obtain a subword decomposition @xmath315 which defines the bigger code .
thus , the word @xmath12 is written as @xmath316 if @xmath317 , then @xmath318 contains neither @xmath271 nor @xmath319 , so @xmath318 commutes with @xmath320 .
since @xmath271 commutes with each @xmath321 for @xmath111 , we can change @xmath12 as @xmath322 by using the braid relations . here
the third equality is obtained by applying the relations @xmath323 and @xmath324 @xmath325 repeatedly .
then the last subword decomposition defines the bigger code , which is contradiction . thus @xmath326 . if @xmath327 , we are done .
otherwise , by iterating the similar arguments for each @xmath321 , we obtain the desired result . for a @xmath4-normal form @xmath328
, we define the integer @xmath329 and @xmath284 as follows . for odd @xmath290
, we define @xmath330 if the last letter of @xmath286 is @xmath271 and @xmath331 otherwise .
similarly , for even @xmath31 , we define @xmath330 if the last letter of @xmath286 is @xmath272 and @xmath331 otherwise . by definition , @xmath332 if and only if the assumption of lemma [ lem : subword ] is satisfied for @xmath31 . therefore if @xmath332 , then @xmath333 holds for some @xmath334 . for
even @xmath31 , let us define an integer @xmath335 @xmath336 as the maximal integer satisfying @xmath337 for odd @xmath31 , we define @xmath335 @xmath338 as the minimal integer satisfying @xmath339 we will consider , for example when @xmath340 , the above formula simply means @xmath341 .
we define @xmath342 if @xmath332 .
now we show that if @xmath332 , then the subword estimation given in lemma [ lem:1stobservation ] for @xmath294 can also be improved .
[ lem : subword2 ] if @xmath332 , then there exists an integer @xmath284 such that @xmath343 holds .
we prove the lemma for even @xmath31 .
the proof for odd @xmath31 is similar . as in the proof of lemma [ lem : subword ] ,
first let us denote @xmath286 as @xmath344 where @xmath345 . from the definition of @xmath335 , we can write the word @xmath318 as @xmath346 where @xmath347 .
assume that @xmath348 for all @xmath349 .
then from the assumption that @xmath12 is the @xmath4-normal form , the word @xmath294 is written as @xmath350 because otherwise , we can produce a new word representative and subword decomposition which defines a bigger code .
we prove the lemma by induction of @xmath335 and @xmath351 , the length of the word @xmath352 .
first we consider the case @xmath353 and @xmath354 . in this case
, we can change the word @xmath355 as @xmath356 hence , by the same argument in the proof of lemma [ lem : subword ] , from the existence of the subword @xmath357 we can construct a new word representative and subword decomposition which defines the bigger code , by pushing out @xmath271 .
hence it is contradiction .
next we consider a general case .
if @xmath358 , then we can change the word @xmath355 as @xmath359 if@xmath360 , let us put @xmath361 .
then we can change the word @xmath355 as @xmath362 hence in both cases , by inductive hypothesis , we obtain a new word representative and subword decomposition which defines a bigger code .
this is a contradiction .
finally , we define @xmath363 if @xmath364 .
this completes the definition of @xmath329 and @xmath284 .
now we construct two positive braid words @xmath365 and @xmath366 for each @xmath4-normal form @xmath12 .
these two words play an important role in the proof of theorem [ thm : main ] .
first of all , we define the word @xmath367 .
if @xmath332 , then we define @xmath368 we remark that , for example , if @xmath369 for even @xmath31 , the above definition simply implies @xmath370 if @xmath364 , then we define @xmath371 by lemma [ lem:1stobservation ] , [ lem : subword ] , and [ lem : subword2 ] , in either case @xmath372 holds for all @xmath31 . using the words @xmath367 , we define the braid word @xmath365 by @xmath373 let @xmath374 and @xmath375 be @xmath21-braids defined by @xmath376 we define the braid word @xmath366 by @xmath377}^{p } d_{[m-2]}^{p } \cdots d_{[0]}^{p } a_{-1}\ ] ] where @xmath85 is a sufficiently large integer , and @xmath378=0 $ ] if @xmath31 is even and @xmath378=1 $ ] if @xmath31 is odd . by taking a sufficiently large integer @xmath85 ,
@xmath366 is obtained by inserting positive generators @xmath101 into the word @xmath12 .
thus , from property @xmath9 , @xmath379 holds .
our idea to prove theorem [ thm : main ] is that , for two @xmath4-normal forms @xmath12 and
@xmath259 , we compare @xmath365 and @xmath380 instead of comparing @xmath259 and @xmath12 directly . an idea to insert or to delete the generators for a given braid word so that the action of the braid become simpler and is easier to compare had already appeared in the author s previous paper @xcite , to estimate the dehornoy floor .
the arguments given belows are straightforward generalization and refinements of the arguments given @xcite . by definition ,
the most parts of the words @xmath366 and @xmath381 are explicitly given , we can draw " the image of the curve diagram @xmath50 under their braid actions for initial parts , which is enough to compare @xmath366 and @xmath381 .
the action of the braids @xmath374 and @xmath375 on the punctured disc @xmath27 are easy to describe . the braid @xmath374 corresponds to the full dehn - twist along the circle enclosing the punctures @xmath382 and the braid @xmath375 corresponds to the full dehn - twist along the circle enclosing the punctures @xmath383 .
similarly , we can see actions of @xmath367 , as we will explain the next section . before stating the actions of the braids @xmath367 and @xmath365
, we introduce a cutting sequence presentation of embedded arcs .
this is a method of encoding an embedded curve into a sequence of signed integers , introduced in @xcite .
let @xmath384 be a curve diagram consists of the horizontal arcs which connect the puncture points ( see figure [ fig : cutting ] ) . for a properly embedded oriented arc @xmath37 in @xmath27 which transversely intersects @xmath384 , we associate the sequence of signed integers as the following manner .
let @xmath385 be the @xmath31-th intersection point of @xmath37 with @xmath384 .
we define @xmath386 ( resp .
@xmath387 ) if @xmath385 lies on @xmath388 and the sign of the intersection at @xmath385 is positive ( resp .
negative ) .
since the complement of the curve diagram @xmath384 is a disc , the isotopy class of @xmath37 can be uniquely determined by the finite sequence of signed integers @xmath389 .
we call this sequence the _ cutting sequence presentation _ of the arc @xmath37 . in a cutting sequence presentation
, an existence of a bigon between @xmath37 and @xmath384 corresponds to an existence of the subsequence of the form @xmath390 .
therefore @xmath37 is tight to @xmath384 if and only if the cutting sequence presentation of @xmath37 contains no subsequences of form @xmath390 . removing a bigon corresponds to removing the subsequence @xmath390 .
we call the cutting sequence presentation is _ tight _ if the corresponding arc is tight to @xmath384 . in the right figure in figure
[ fig : cutting ] , the cutting sequence presentation of the arc @xmath391 is @xmath392 .
as is easily observed , this is the tight cutting sequence presentation .
the actions of the braid groups on cutting sequence presentations are easy to describe .
since the braid @xmath393 corresponds to the half dehn - twist along the arc @xmath394 , the cutting sequence presentation of the arc @xmath395 is obtained from the original cutting sequence of @xmath37 by replacing a subsequence @xmath396 ( resp .
@xmath397 ) with @xmath398 ( resp .
@xmath399 ) .
the obtained cutting sequence presentation might be non - tight and depends on a particular choice of word representative of a braid , even if the original cutting sequence presentation is tight .
therefore , sometimes the tight cutting sequence presentation of an arc might change dramatically .
however , in some cases we can obtain an initial segment of the tight cutting sequence presentation of an arc as the following lemma shows .
[ lem : cutcalc ] let @xmath15 be a positive @xmath21-braid and @xmath50 be a properly embedded arc .
let us denote the tight cutting sequence presentation of @xmath61 by @xmath400 and assume that @xmath401 for all @xmath402 . 1 . if @xmath403 and @xmath404 , then the tight cutting sequence presentation of the braid @xmath405 is @xmath406 2 . if @xmath407 and @xmath408 , then the tight cutting sequence presentation of the braid @xmath405 is @xmath409 we prove ( 1 ) .
the proof of ( 2 ) is similar . by a direct computation ,
the cutting sequence presentation of @xmath410 is given by @xmath411 since the original cutting sequence is tight , this sequence does not contain a subsequence of form @xmath412 .
this means , under the process of making the sequence tight , @xmath413 disappear if and only if @xmath414 . from the assumption of @xmath415 ,
the term @xmath416 does not disappear , so the tight cutting sequence presentation of @xmath410 is given by @xmath406 by using lemma [ lem : cutcalc ] , we can calculate an initial segment of the tight cutting sequence presentation of the following special braids which are related to @xmath366 and @xmath365 . [
prop : cutcalc ] let @xmath417 be the @xmath4-normal form of a positive @xmath21-braid @xmath15 and @xmath418}^{p } d_{[m-2]}^{p } \cdots d_{[0]}^{p } a_{-1},\ ; \underline{w}= a_{m } \underline{a_{m-1}}\ , \underline{a_{m-2}}\cdots \underline{a_{0}}a_{-1}.\ ] ] be positive braid words defined in the previous section .
( @xmath419 ) : : the tight cutting sequence presentation of the arc + @xmath420}^{p } \cdots d_{[0]}^{p } a_{-1 } ] ( \gamma_{1 } ) $ ] for odd @xmath421 is @xmath422 ( @xmath423 ) : : the tight cutting sequence presentation of the arc + @xmath424}^{p } d_{[m-2]}^{p } \cdots d_{[0]}^{p } a_{-1 } ] ( \gamma_{1 } ) $ ] for even @xmath421 is @xmath425 ( @xmath426 ) : : the tight cutting sequence presentation of the arc + @xmath427(\gamma_{1 } ) $ ] for odd @xmath421 is @xmath428 ( @xmath429 ) : : the tight cutting sequence presentation of the arc + @xmath430(\gamma_{1 } ) $ ] for even @xmath421 is @xmath431 the above calculations are confirmed by a combination of the following sub - calculations , which are easily observed from lemma [ lem : cutcalc ] . *
if @xmath50 has the tight cutting sequence @xmath432 where @xmath433 is a subsequence which contains no @xmath434 , then the tight cutting sequence of @xmath435 is @xmath436 . *
if @xmath50 has the tight cutting sequence @xmath437 where @xmath433 is a subsequence which contains no @xmath438 , then the tight cutting sequence of @xmath439 is @xmath440 . *
if @xmath50 has the tight cutting sequence @xmath441 , where @xmath433 is a subsequence which contains no @xmath442 , then the tight cutting sequence of @xmath443 is @xmath444 . *
if @xmath50 has the tight cutting sequence @xmath440 where @xmath433 is a subsequence which contains no @xmath445 , then the tight cutting sequence of @xmath446 is @xmath447 . *
if @xmath50 has the tight cutting sequence @xmath448 where @xmath449 and @xmath433 is a subsequence which contains no @xmath442 , then the tight cutting sequence of @xmath450 is @xmath451 . *
if @xmath50 has the tight cutting sequence @xmath452 where @xmath453 and @xmath433 is a subsequence which contains no @xmath454 , then the tight cutting sequence of @xmath446 is @xmath455 .
now we are ready to prove our main theorems .
first we remark that from the definition of the @xmath4-normal forms and codes , we only need to prove the theorem for normal finite thurston type orderings .
we prove the theorem by induction of @xmath21 .
the case @xmath174 is trivial .
assume that the assertion is proved for all normal finite thurston type orderings of @xmath121 for all @xmath122 .
let @xmath11 be a normal finite thurston type ordering on @xmath1 define by the normal curve diagram @xmath50 . for positive @xmath21-braids
@xmath78 and @xmath15 , assume that @xmath19 holds .
we denote the @xmath4-normal form of @xmath15 by @xmath417 and the @xmath4-normal form of @xmath78 by @xmath456 .
we prove @xmath18 by showing @xmath457 .
since the ordering @xmath11 is left - invariant , we can always assume that the initial letters of @xmath458 and @xmath20 are different , by annihilating the common prefixes .
let us put @xmath459 and @xmath460}^{p } d_{[l-2]}^{p } \cdots d_{[0]}^{p } c_{-1}.\ ] ] since @xmath19 , the inequality @xmath461 holds .
if @xmath462 , then by proposition [ prop : cutcalc ] , we conclude the arc @xmath463 moves more left than @xmath464 , so @xmath465 .
therefore we only need to consider the case @xmath466 .
first we consider the case @xmath467 or @xmath468 .
assume that @xmath469 , thus @xmath470 and @xmath471 .
let us denote the decompositions of @xmath135 and @xmath472 as @xmath473 and @xmath474 respectively .
since we have already assumed that the initial letters of @xmath475 and @xmath188 are different , so our assumption @xmath476 implies @xmath477 for @xmath478 and @xmath479 .
let us write the decomposition of @xmath160 as @xmath480 the assumption that @xmath135 is the @xmath4-normal form implies @xmath481 must be the empty word . because otherwise , by regarding @xmath482 as a new @xmath483 , we obtain a new subword decomposition which defines the bigger code .
similarly , for the decomposition @xmath484 of @xmath485 , the last subword @xmath486 is the empty word .
now , the fact both @xmath481 and @xmath486 are empty words implies that both the braids @xmath135 and @xmath472 move the arc @xmath52 .
therefore the comparison of @xmath135 and @xmath472 can be done by seeing the image of @xmath52 . recall that we have defined the code of @xmath160 and @xmath485 so that they coincide with the code with respect to the restriction of the ordering to @xmath487 .
now by inductive hypothesis , we conclude that @xmath488 moves left side of@xmath489 , hence @xmath18 .
the case @xmath467 is similar .
next we consider the case @xmath490 .
we only show @xmath421 is odd case .
the even @xmath421 case is similar . from proposition [ prop :
cutcalc ] , the tight cutting sequence presentation of the arc @xmath491(\gamma_{1})\ ] ] have the initial segment @xmath492 if @xmath493 and @xmath494 if @xmath495
. from the definition of @xmath4-normal form , both @xmath496 and @xmath497 are also the @xmath4-normal form with respect to the dehornoy ordering of @xmath138 .
let @xmath498 be the normal curve diagram in example [ exam : dtype ] , which defines the dehornoy ordering . since @xmath499 , the inequality @xmath500 holds .
therefore by the inductive hypothesis , the arc @xmath501 moves left side of @xmath502 .
in particular , @xmath503 moves more left than @xmath504 .
now let us observe the tight cutting sequence presentation of the arc @xmath505 is @xmath506 , which is identical with the subsequence @xmath507 , and the tight cutting sequence of @xmath508 is @xmath509 , which is identical with the subsequence @xmath510 .
thus , we can apply the results in the proposition [ prop : cutcalc ] for the actions of @xmath511 and @xmath512 for the subsequence @xmath285 .
therefore , the arc @xmath513 has the initial segment @xmath514 if @xmath493 and @xmath515 if @xmath495 . thus in either case , the initial segment of the tight cutting sequence of the arc @xmath513 is given by @xmath516 where @xmath503 is an initial segment of the tight cutting sequence presentation of arc @xmath503 .
similarly , the tight cutting sequence of the arc @xmath517}^{p } d_{[m-2]}^{p } \cdots d_{[0]}^{p } a_{-1})(\gamma_{1})\ ] ] has the initial segment @xmath518 where @xmath504 is a initial segment of the tight cutting sequence presentation of arc @xmath504 . by inductive hypothesis ,
the arc @xmath503 left side of the arc @xmath504 , so we conclude that @xmath519}^{p}\cdots < \underline{a_{m}}\,\underline{a_{m-1}}\cdots < \underline{w } < \beta .\ ] ] next we proceed to a computation of the order - type of the well - ordered set @xmath10 for arbitrary finite thurston type orderings .
first we observe the following lemma .
[ lem : conjugate ] let @xmath11 , @xmath72 be conjugate left - invariant total orderings of @xmath1 , having the property @xmath9 .
then the order type of @xmath10 and @xmath520 are the same .
first we remark that the assumption that the ordering has the property @xmath9 implies its restriction to @xmath8 defines a well - ordering .
let @xmath78 be a conjugating element between @xmath11 and @xmath72 . from lemma [ lem : conjpos ] , we can choose @xmath78 as a positive braid . from property @xmath9 ,
the map @xmath521 defined by @xmath522 is an order - preserving injection , so the order - type of @xmath523 is smaller than that of @xmath524 . by interchanging the role of @xmath11 and @xmath72
, we also obtain that the order type of @xmath524 is smaller than that of @xmath523 .
therefore they have the same order - type .
thus , it is sufficient to compute the order - type of @xmath10 for a normal finite thurston type ordering @xmath11 to determine the order - type of @xmath520 for general normal finite thurston type ordering @xmath72 .
we prove the theorem by induction on @xmath21 .
the case @xmath174 is trivial .
let @xmath11 be a normal finite thurston type ordering on @xmath1 and let @xmath525 be the set of all codes of the @xmath4-normal forms with respect to the ordering @xmath11 .
theorem [ thm : main ] asserts that @xmath526 is order - isomorphic to @xmath10 .
a direct computation shows that the order - type of @xmath527 is @xmath3 .
since @xmath526 is a subset of @xmath527 , the order type of @xmath526 is at most @xmath3 . to show the converse inequality
, we consider the set of even codes @xmath528 that is , a code @xmath4 belongs to the set @xmath529 if and only if each entry of @xmath4 is strictly positive even integer except trivial zeros . by inductive hypothesis
, a direct computation shows that the order type of @xmath530 is @xmath3 . on the other hand , tits conjecture , proved in @xcite ,
asserts the subgroup of @xmath1 generated by the square of generators @xmath531 has the only trivial commutative relations @xmath532 .
this implies the set of even codes @xmath533 is a subset of @xmath534 .
so the order type of @xmath526 is at least @xmath3 .
therefore the order type of @xmath526 is @xmath3 .
in this section we study the relationships between the @xmath4-normal form and dehornoy s @xmath5-normal form ( alternate normal form ) or burckel s normal form .
this gives an algebraic description , the efficient method of computations , and the computational complexities of the @xmath4-normal forms .
the @xmath5-normal form is a normal form of positive braids defined by dehornoy in @xcite .
first we review the definition of the @xmath5-normal form of the positive braid monoid @xmath8 .
the construction of such a normal form is valid for much wider class of monoids , called a locally garside monoid .
since we are interested in the braid groups , we only describes the positive braid monoid case .
as is well known , the positive braid monoid @xmath8 defines a garside structure of @xmath1 .
this implies the positive braid monoid @xmath8 has many good properties .
here we do not describe what a garside structure is .
only we need is existence and uniqueness of the maximal right divisor . for a subset @xmath535 of @xmath536 , let @xmath537 be the submonoid of @xmath8 generated by @xmath538
. for each @xmath187
, there exists the unique maximal right divisor of @xmath15 which belongs to @xmath537 .
that is , there exists the unique element @xmath539 which satisfies the following two properties : let @xmath545 and @xmath546 . for a positive @xmath21-braid @xmath15 , the @xmath547-decomposition ( the alternate decomposition ) of @xmath15 is a factorization of @xmath15 given by @xmath548 where @xmath549 is defined by the inductive formula @xmath550 for even @xmath140 , we identify @xmath537 with @xmath148 by the flip map @xmath106 .
similarly , we identify @xmath551 with @xmath148 by the identity map for odd @xmath140 . for @xmath552 ,
we use another identification .
we identify @xmath537 with @xmath148 by the shift map @xmath240 .
then by using these identifications , we can perform the above alternate decompositions for each @xmath549 . by iterating the alternate decomposition until each @xmath537 is generated by only one generator ,
we finally obtain the unique positive word representative of @xmath15 .
we call this unique positive word representative the @xmath5-_normal form _ ( alternate normal form ) .
now we iterate the alternate decomposition for each @xmath549 . using an identification @xmath555 ,
the alternate decomposition of @xmath556 is given by @xmath557 .
similarly , the alternate decomposition of @xmath558 is given by @xmath559 . finally , by using an identification @xmath560 , the alternate decomposition of @xmath561 is given by @xmath562 .
summarizing , we obtain the @xmath5-normal form of @xmath15 , @xmath563 [ prop : m - normal ] let @xmath88 be the dehornoy ordering of the braid group @xmath1 . then for each element @xmath15 in @xmath8 , the @xmath4-normal form @xmath564 of @xmath15 with respect to the ordering @xmath88 coincide with dehornoy s @xmath5-normal form .
let @xmath15 in @xmath8 and @xmath266 be the @xmath4-normal form of @xmath15 .
let us denote the alternate decomposition of @xmath15 by @xmath565 . from the definition of the @xmath4-normal form
, @xmath149 is always an empty word . since @xmath566
, we obtain @xmath567 for some @xmath568 .
if @xmath569 is not a trivial braid , then the subword decomposition @xmath570 defines a strictly bigger code , so it leads a contradiction .
so we conclude @xmath571 by the similar argument , we obtain @xmath572 for @xmath140 .
an iteration of this argument for each @xmath142 shows that the @xmath4-normal form and the @xmath5-normal form coincide .
[ [ tail - twisted - phi - normal - form - and - mathcalc - normal - form ] ] tail - twisted @xmath5-normal form and @xmath4-normal form ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ now we generalize proposition [ prop : m - normal ] for general normal finite thurston type orderings .
to describe the @xmath4-normal form for general normal finite thurston type orderings in view of dehornor s @xmath5-normal form , we introduce a generalized notion of the @xmath5-normal form , the _ tail - twisted @xmath5-normal form_. let @xmath573 be a permutation of @xmath69 integers @xmath87 . for @xmath187 , the @xmath574-tail twisted @xmath5-normal form is a positive word representative of @xmath15 defined by the following inductive procedure .
assume that we have already constructed the @xmath574-tail twisted normal forms for all @xmath576 and for all permutations in @xmath577 @xmath578 .
here @xmath577 is the permutation group of degree @xmath31 .
then we define the @xmath574-tail twisted normal form of positive @xmath21-braids @xmath15 for @xmath579 as follows .
let @xmath124 and @xmath125 be integers and @xmath128 be the set of integers appeared in the definition of the code ( see section 3.1 again ) .
let @xmath585 be the permutation of @xmath586 integers , defined by @xmath587 we denote the @xmath5-normal form of the braid @xmath592 by @xmath593 . by some powers of the shift map , each @xmath594
is naturally identified with @xmath595 . using this identification , we denote by @xmath596 the @xmath585-tail twisted @xmath5-normal form of @xmath597 . by the inductive assumption
, we have already defined these tail - twisted @xmath5-normal forms .
now , the _ @xmath574-tail twisted @xmath5-normal form _ of a positive @xmath21-braid @xmath15 is a positive word representative @xmath598 defined by @xmath599 we remark that for the trivial permutation @xmath600 , the @xmath574-tail twisted @xmath5-normal form is nothing but the usual dehornoy s @xmath5-normal form .
using the tail - twisted @xmath5-normal form , proposition [ prop : m - normal ] is generalized as follows .
let @xmath11 be a normal finite thurston type ordering of @xmath1 which corresponds to the permutation @xmath601 .
then for each @xmath187 , the @xmath574-tail twisted @xmath5-normal form @xmath598 coincide with the @xmath4-normal form @xmath20 .
the proof is almost the same as the proof of proposition [ prop : m - normal ] , so we only describe the correspondence between these two normal forms . for
a positive @xmath21-braid @xmath15 , let @xmath602 be the @xmath4-normal form of @xmath15 , and denote the subword decomposition of @xmath135 by @xmath603 .
similarly , let @xmath604 be a twisted version of the alternate decomposition of @xmath15 , and @xmath605 be the alternate decomposition of @xmath592 .
as is easily observed , the braid @xmath561 is always trivial , so we omit it . by the same arguments of the proof of proposition [ prop : m - normal ] ,
we obtain a correspondence @xmath606 , @xmath607 , @xmath608 , and @xmath609 .
thus , an inductive argument shows these two normal forms are indeed identical .
the definition of the tail - twisted @xmath5-normal form is a bit complex , so we give a concrete example of the tail - twisted normal form in a simple case .
let @xmath610 .
the normal finite thurston type ordering corresponding to the permutation @xmath574 is the ordering given in example [ exam : thurston4 ] . since for this permutation @xmath574 , all permutations @xmath582 , @xmath617 and @xmath618 are the trivial permutations , so all of the tail - twisted @xmath5-normal forms with respect to these permutations are merely the ordinary @xmath5-normal forms .
thus , we obtain @xmath619 combing the above results , we conclude that the @xmath574-tail twisted normal form of @xmath15 is given by @xmath620 which is of course identical with the @xmath4-normal form of @xmath15 given in example [ exam : thurston4 ] . finally , using these results and dehornoy s results about the @xmath5-normal forms , we prove theorem [ thm : complexity ] .
theorem [ thm : complexity ] is interesting in the following sense .
although in @xcite a computational complexity of finite thurston - type orderings is given , but it relies on mosher s automatic structure of mapping class groups @xcite .
so the relationships between a usual word representative and a computational complexity is indirect .
the tail - twisted @xmath5-normal form formulation allows us to determine the computational complexity in usual word representatives , and provide more convenient and practical method to compute finite thurston type orderings . for dehornoy s @xmath5-normal form
, these results are proved in @xcite .
first we consider normal case . in this case , the @xmath4-normal form is identical with the tail - twisted @xmath5-normal form .
a computation of twisted tail parts , that is , a computation of @xmath621 and tail - twisted normal forms @xmath622 , is merely a computation of maximal right divisors and the tail - twisted @xmath5-normal forms for smaller @xmath21 .
so the computation of these parts does not affect the order of the total computational complexity .
therefore the theorem also holds for normal thurston type orderings . for a general finite thurston type ordering , computation of @xmath4-normal form for word length @xmath23 braid is a computation of @xmath4-normal form with respect to normal finite type of a braid with the word length @xmath623 , where @xmath85 is the word length of a positive conjugating element .
since @xmath85 is a constant , therefore the conclusion holds .
the @xmath4-normal form constructions and arguments in this paper can be seen as a geometric background of the @xmath5-normal form and burckel s normal form .
our @xmath4-normal form brings new information about the questions raised by dehornoy in @xcite .
our construction gives an answer to the question @xmath624 in @xcite , which is in short , give a direct proof of the fact that the @xmath5-normal form gives a combinatorial description of the dehornoy ordering . " here the word direct " means without using burckel s result of the dehornoy orderings .
our proof requires neither burckel s normal form nor its tree construction .
we use a geometric definition of the dehornoy ordering given in @xcite , @xcite and the property @xmath9 , which is confirmed by the geometric definition .
thus our proof is independent of burckel s results . of course , the above question is still open when we restrict to use the usual algebraic definition of the dehornoy ordering and algebraic method
. 1 d.bessis , _ the dual braid monoid , _ ann .
36 * , no.5 ( 2003 ) , 647 - 683 .
j.birman , _ braids , links , and mapping class groups , _ ann . of math .
studies * 82 * , princeton univ . press ( 1975 ) .
j.birman , k.ko , and s.lee , _ a new approach to the word problem in the braid groups , _ adv . math . * 139 * ( 1998 ) , 322 - 353 .
s.burckel , _ the well - ordering on positive braids , _ j. pure appl .
alg , * 120 * , ( 1997),1 - 17 .
s.burckel , _ computation of the ordinal of braids , _ order , * 16*,(1999),291 - 304 .
j.crisp , l.paris , _ the solution to a conjecture of tits on the sub group generated by the squares of the generators of an artin group , _ invent .
math , * 145 * , ( 2001),19 - 36 .
p.dehornoy , _ braids and left distributive operations , _ trans . of ams , * 345 * , ( 1994 ) , 115 - 151 .
p.dehornoy , _ alternating normal forms in braid monoids and locally garside monoids , _ j. pure appl .
algebra , * 212 * , ( 2008 ) , 2416 - 2439 .
p.dehornoy , i.dynnkov , d.rolfsen and b.wiest , _ why are the braids orderable ? , _ panoramas et synthses * 14 * , soc .
p.dehornoy , i.dynnkov , d.rolfsen and b.wiest , _ ordering braids , _ math . surveys and monographs * 148 * , amer .
j.fromentin , _ the cycling normal form on dual braid monoids , _ arxiv : math.gr/0712.3836 .
j.fromentin , _ a well - ordering of dual braid monoids , _ comptes rendus mathematique , * 346*,(2008 ) , 729 - 734 .
j.fromentin , _ every braid admits a short sigma - definite representative ,
_ arxiv:0811.3902v1 .
r.fenn , m.greene , d.rolfsen , c.rourke and b.wiest , _ ordering the braid groups , _ pacific j. of math , * 191 * , ( 1999 ) , 49 - 73 .
t.ito , _ braid ordering and the geometry of closed braid , _
e - print , arxiv:0805.1447v2 t.ito , _ finite thurston type orderings on the dual braid monoids , _ e - print , arxiv:0902.0833v2 l. mosher , _ mapping class groups are automatic , _ ann . of math . * 142 * , ( 1995 ) , 303 - 384 .
c.rourke , b.wiest , _ order automatic mapping class groups , _ pacific j. math , * 194 * , no.1 ( 2000 ) , 209 - 227 .
h.short , b.wiest , _ ordering of mapping class groups after thurston , _ enseign .
math , * 46*,(2000 ) , 279 - 312 . | we prove that for any finite thurston - type ordering @xmath0 on the braid group @xmath1 , the restriction to the positive braid monoid @xmath2 is a well - ordered set of order type @xmath3 .
the proof uses a combinatorial description of the ordering @xmath0 .
our combinatorial description is based on a new normal form for positive braids which we call the @xmath4-normal form .
it can be seen as a generalization of burckel s normal form and dehornoy s @xmath5-normal form ( alternating normal form ) . |
let @xmath3 be an i.i.d .
sequence of non negative random variables with common cumulative distribution @xmath4 .
the arrival times @xmath5 are defined by @xmath6 and , for @xmath7 , @xmath8 .
we consider the counting process @xmath9 defined by @xmath10 and the associated renewal function @xmath11=\sum_{n=0}^\infty f^{\ast n}(x),\quad x\ge 0.\ ] ] the renewal theorem states that if the inter - arrival distribution @xmath4 has a finite first moment @xmath12 , then @xmath13 recall that the distribution @xmath4 is called lattice if @xmath4 is supported by @xmath14 for some mesh @xmath15 .
it is well known ( see for example asmussen ( * ? ? ?
* proposition 6.1 ) ) that if @xmath4 has a finite second moment @xmath16 , then @xmath17 when @xmath4 has infinite first or second moment , teugels @xcite provides asymptotics for the renewal function @xmath18 under some regular variation conditions . in this paper , we focus on the case when @xmath4 is light - tailed and we assume that @xmath19 has some finite exponential moment so that @xmath20 then the moment generating function @xmath21 is well defined and holomorphic on @xmath22 using complex analysis , stone @xcite proved that under the strong non lattice condition @xmath23 there exists some @xmath24 such that @xmath25 up to now , relatively few results concern either expansions or closed form expressions for @xmath26 .
it appears that the only case where closed form expansions are available are when @xmath4 is matrix exponential distributed , in which case an expression of @xmath26 is given in asmussen and bladt @xcite .
mitov and omey @xcite provide heuristics on many terms asymptotics of @xmath26 that are verified on the already known cases .
however , as the authors point out , those interesting expansions are only given formally and are not proved .
other expansions are available in ( * ? ? ?
* theorem 4 ) in the context of potential densities of lvy processes in terms of the associated lvy jump distribution .
the approach by stone @xcite for obtaining expansion ( [ expansion_stone ] ) is mainly based on complex analysis and proved fruitful for obtaining expansions involving survival functions of random sums , see blanchet and glynn @xcite .
the approach in @xcite was later generalized for spread out distributions in @xcite ; note that the generalization of our results to spread out distribution is not available in the present paper because the main technical assumption that enable us to obtain higher expansions ( namely , assumption ( [ conda ] ) thereafter ) , which is stronger than the non lattice condition ( [ eq : stone_strong_lattice ] ) , is not verified for spread out distributions .
this is not really surprising , as it is on the other hand known that a spreadout distribution is strongly non lattice ( see e.g. proposition 1.6 p.189 of @xcite ) , which is why many results in renewal theory that hold for strongly non lattice distributions also hold for spreadout distributions .
we consider in this paper both the lattice and the non lattice cases and , in view of equation , we introduce the function @xmath27 following stone s approach but with more detailed computations , we obtain higher order expansions for the function @xmath28 both in the lattice and non - lattice cases ( theorems [ thm_expansion_v_lattice ] and [ thm_expansion_v ] respectively ) .
when @xmath0 has a meromorphic extension to the whole complex plane and under some technical conditions , we are even able to provide exact expansions for @xmath28 ( corollary [ cor_expansion_exact ] ) .
the paper is structured as follows .
section 2 presents our results on high order expansions for the renewal function as well as some examples .
section 3 is devoted to applications to ruin theory : we provide asymptotics of the ruin probability in the setting of continuous or discrete time risk processes and also consider a two dimensional risk process .
proofs are gathered in sections 4 and 5 .
in the sequel , the solutions of the equation @xmath29 play a major role . note that @xmath1 is the unique real solution in @xmath30 and that other solutions satisfy @xmath31 and come in pairs , i.e. if @xmath32 is a solution then so is @xmath33 .
we consider first the lattice case and we assume without loss of generality that the mesh of the distribution is equal to @xmath34 , i.e. @xmath4 is supported by @xmath35 . in this case , the moment generating function @xmath36 is @xmath37-periodic on @xmath38 and we introduce the fundamental domain @xmath39 .
[ thm_expansion_v_lattice ] let @xmath40 be such that equation @xmath29 has no root verifying @xmath41 .
let us denote by @xmath42 the solutions of this equation in @xmath43 .
then , @xmath44 has the asymptotic expansion @xmath45 where the notation @xmath46 denotes the residue of the meromorphic function @xmath47 at pole @xmath48 .
+ if @xmath48 is a simple zero of @xmath49 , i.e. @xmath50 , then the @xmath51-th term in has the simple form @xmath52 next we consider the case when @xmath4 is non - lattice .
it is well known that a distribution is non - lattice if and only if @xmath53 is the unique real solution of the equation @xmath54 .
we will need here the following stronger technical assumption : for all @xmath55 , @xmath56 assumption is stronger than the strong lattice condition ( [ eq : stone_strong_lattice ] ) from stone @xcite .
it is however not too restrictive and satified by a large class of distributions as shown by the following propositon .
[ prop_a ] suppose that the distribution @xmath4 is absolutely continuous with respect to the lebesgue measure .
then assumption is satisfied for all @xmath57 .
our main result in the non - lattice case is the following theorem .
[ thm_expansion_v ] consider a strictly non - lattice distribution @xmath4 satisfying assumption .
then for all @xmath58 , the equation @xmath29 has a finite number of solutions in @xmath59 that we denote by @xmath60 . then , supposing that @xmath61 is such that there is no solution to that equation verifying @xmath41 , @xmath62 has the asymptotic expansion @xmath63 if @xmath48 is simple zero of @xmath49 , i.e. @xmath50 , then the @xmath51-th term in has the simple form @xmath64 it is worth noting that theorems [ thm_expansion_v_lattice ] and [ thm_expansion_v ] can be extended to obtain asymptotics of higher order , i.e. of order @xmath65 with @xmath66 , if we assume that the moment generating function @xmath0 has a meromorphic extension to @xmath67 for some @xmath68 .
theorems [ thm_expansion_v_lattice ] and [ thm_expansion_v ] and their proofs extend in a straightforward way , but not proposition [ prop_a ] . in the case
when the moment generating function @xmath0 has a meromorphic extension to the whole complex plane , i.e. @xmath69 , it is even possible under some technical assumption to get an exact expansion for the renewal function .
[ cor_expansion_exact ] assume that @xmath70 has a meromorphic extension to the whole complex plane .
* in the lattice case , we assume furthermore that @xmath71 we denote by @xmath72 ( possibly @xmath73 ) the solution of @xmath29 in the fundamental domain @xmath74 with @xmath75 . then , we have the exact expansion @xmath76 * in the non - lattice case , we suppose that assumption holds for all @xmath77 .
let @xmath78 .
let us furthermore suppose that * * one has an infinite number of roots @xmath79 of equation @xmath29 , all of which are simple , real , and such that the following series converges : @xmath80 * * the following holds : @xmath81 + then , we have the exact expansion @xmath82 [ comment_referee ] one of the assumptions of corollary [ cor_expansion_exact ] in the non - lattice case is that there is an _
infinite _ number of roots of equation @xmath29 .
one may wonder what happens in the case when those roots are in finite number @xmath9 .
in fact , one shows in this case , and thanks to condition ( [ eq : cond_nonlattice ] ) , that the moment generating function @xmath70 is rational , in which case a finite expansion for @xmath62 can be deduced almost straightforwardly . in the previous results ,
a kind of dichotomy arises between the lattice and non - lattice cases .
interestingly , a unified statement can be deduced for the renewal mass function or the renewal density function in the non - lattice and lattice case respectively .
[ cor_expansion_exact_2 ] under the same assumptions as corollary [ cor_expansion_exact ] : * in the lattice case , the renewal measure has mass function @xmath83 as a particular case , if all the poles are simple , @xmath84 * in the non - lattice case , the renewal measure has density function @xmath85 as a particular case , if all the poles are simple , @xmath86 to conclude this subsection , we present an informal argument leading to ( and motivating ) expansion ( [ expansion_v ] ) and that may lead to some better comprehension of proof of theorem [ thm_expansion_v ] given in section [ sub_section_proof_theo_expansion ] .
one verifies , using fubini , that the laplace transform of @xmath87 is @xmath88 dx=\sum_{k=0}^\infty { \mathbbm{e}}\left [ \int_{s_k}^\infty e^{-xz } dx\right]\\ = \sum_{k=0}^\infty \frac{g(-z)^k}{z}= \frac{1}{z(1-g(-z))},\end{gathered}\ ] ] so that , the inversion formula for the laplace transform leads formally to @xmath89 where @xmath90 is such that all singularities of @xmath91 are on the left of vertical line @xmath92 .
the right hand side integral of ( [ mellin_transf ] ) does not necessarily converge , however let us suppose that this is the case .
the poles of @xmath91 satisfying @xmath93 are @xmath94 , ... , @xmath95 , and one can verify that the residue of @xmath96 at @xmath94 is @xmath97 .
thus , using a rectangular contour and the theorem of residues leads to the following @xmath98 the last integral is an @xmath99 , so that one would obtain expansion ( [ expansion_v ] ) .
however , the main failing points in this reasoning are first that the integral in ( [ mellin_transf ] ) is not convergent , and that the contour argument leading to ( [ mellin_transf2 ] ) is more delicate than it seems .
the convergence issue will be solved by introducing a gaussian kernel ( an idea already introduced by stone @xcite ) which , by inversion , will make the corresponding integral converge , see step 2 in the proof in forthcoming section [ sub_section_proof_theo_expansion ] .
the contour argument will involve assumption ( [ conda ] ) , which will enable two of the pieces of the contour to vanish in the proof , see again step 2 in section [ sub_section_proof_theo_expansion ] as well as the corresponding figure [ figure_contour ] .
we provide some examples that illustrate the results above . in the lattice case
, we consider the negative binomial distribution with parameters @xmath100 and @xmath7 defined by @xmath101 its moment generating function is given by @xmath102 so that @xmath103 defines a meromorphic function on @xmath104 .
the poles are the solutions of @xmath105 for @xmath106 the only pole in the fundamental domain @xmath107 is @xmath108 so that theorem [ thm_expansion_v_lattice ] implies @xmath109 for all @xmath24 . in the general case @xmath7 , there are exactly @xmath110 poles in the fundamental domain @xmath111 given by @xmath112 with @xmath113 furthermore , assumption is easily satisfied ( comparison between an exponential and a power growth ) so that corollaries [ cor_expansion_exact ] and [ cor_expansion_exact_2 ] apply . we obtain that the renewal measure has mass function @xmath114 in the non - lattice case let us consider _ matrix exponential distributions _ with parameters
@xmath115 , where @xmath116 is an @xmath117 probability row vector and @xmath118 is an @xmath119 subintensity matrix .
the definition and principal properties of phase type distributions can be found in asmussen and albrecher ( * ? ? ?
* chapter ix ) .
the moment generating function @xmath70 is a rational function given by @xmath120 where @xmath121 and @xmath122 ( see ( * ? ? ?
* theorem 1.5 ) ) .
the equation @xmath29 is a polynomial equation with @xmath123 solutions @xmath124 and condition is satisfied with @xmath125 and assumption is easily satisfied ( comparison between an exponential and a power growth ) . from corollaries [ cor_expansion_exact ] and [ cor_expansion_exact_2 ] ,
we obtain a closed formula for the renewal distribution @xmath18 and the renewal density @xmath126 . on the other hand , asmussen and bladt ( * ? ? ?
* theorem 3.1 ) provide the simple expression @xmath127 let us check that this formula agrees with corollary [ cor_expansion_exact_2 ] in the case when the roots @xmath128 are simple .
it is easy to check that the @xmath129 s are exactly the eigenvalues of the matrix @xmath130 .
denoting by @xmath131 the @xmath132 matrix made up with the eigenvectors @xmath133 corresponding to eigenvalues @xmath128 , we have @xmath134 denoting by @xmath135 the @xmath119 matrix with @xmath136 at the @xmath137th position and @xmath1 elsewhere , we compute @xmath138 on the other hand , we deduce from ( [ mgf_ph ] ) @xmath139 hence , in order to prove that ( [ ph_1 ] ) and ( [ renewal_density_polynome ] ) agree , we need to prove @xmath140 or equivalently @xmath141 this can be easily verified with elementary algebra ( the relation @xmath142 is useful ) .
[ example_uniform ] let us consider the simple case of the _ uniform distribution _ on @xmath143 $ ] , i.e. @xmath144 $ ] .
in that case @xmath145 and the equation @xmath29 is equivalent to @xmath146 .
the solutions are @xmath147 , @xmath148 , where @xmath149 is the @xmath51th generalized lambert function .
proposition [ prop_a ] together with theorem [ thm_expansion_v ] provide an asymptotic expansion of @xmath28 .
using the relations @xmath150 and @xmath151 , we obtain , for all @xmath152 , @xmath153 with @xmath147 and @xmath154 . as @xmath155 , @xmath156 so that the expansion has arbitrary high order .
at this point , there still lacks an example of distribution @xmath157 such that a meromorphic extension @xmath70 exists , equation @xmath29 admits an infinite number of solution and infinite expansion ( [ exact_expansion_v_non_lattice ] ) holds .
an example of such an infinite expansion of @xmath62 will be given in upcoming example [ example_meromorphic ] , in the context of meromorphic lvy processes .
as an application of theorem [ thm_expansion_v ] , we provide asymptotic expansions for the ruin probability in risk theory .
we consider both a continuous setting ( compound poisson risk process ) and a discrete setting ( binomial risk process ) .
estimation of the ruin probability in a two - dimensional model is also investigated .
we consider the following classical continuous time risk process @xmath158 with @xmath159 a poisson process with intensity @xmath160 and @xmath161 an i.i.d .
sequence of non - negative random variables with common distribution @xmath162 and finite expectation @xmath163 , independent from @xmath159 .
such a process models the capital of an insurance company with premium rate @xmath90 , initial reserve @xmath164 , and incoming claims @xmath161 , see e.g. asmussen and albrecher @xcite .
we define @xmath165 the tail function and assume the following on the moment generating function @xmath166 we are interested in the ruin probability @xmath167 and its asymptotic when the initial reserve @xmath164 is large .
it is well known that @xmath168 if and only if the safety loading is positive , i.e. @xmath169 in the asymptotic analysis , a key role is played by the lundberg equation @xmath170 under assumption ( [ safety_loading ] ) , this equation restricted to real numbers admits a unique real solution denoted by @xmath171 .
the lundberg inequality states that @xmath172 while the cramr - lundberg approximation provides the asymptotic behavior as @xmath173 @xmath174 we provide high order asymptotic expansions for the ruin probability @xmath175 .
similar considerations as well as exact expansions have been proved with different methods by kuznetsov and morales @xcite for a so called meromorphic risk process and by roynette et al .
@xcite . using the fact that @xmath171 solves the lundberg equation
, one can define the probability measure @xmath4 on @xmath176 by @xmath177 the moment generating function @xmath178 is well defined and holomorphic on the complex plane . for future reference , note that the lundberg equation is equivalent to @xmath179 using a renewal equation solved by the ruin probability function @xmath180 and the asymptotic behavior of the renewal function provided by theorem [ thm_expansion_v ] , we can deduce an asymptotic expansion for @xmath175 as @xmath173 .
[ prop_exp_ruin ] assume conditions and are satisfied .
let @xmath24 be fixed and @xmath60 the solutions of @xmath29 in @xmath181 .
then , the ruin probability @xmath175 has the asymptotic expansion @xmath182 if @xmath48 is a simple zero of @xmath49 , i.e. @xmath50 , then the @xmath51-th term in has the simple form @xmath183 the term @xmath184 of the asymptotic expansion is exactly the cramer - lundberg approximation .
[ ex_stop_loss ] let us consider the case where the claims are of the form @xmath185 where @xmath186 and @xmath187 has an exponential distribution with parameter @xmath188 .
this models a reinsurance scenario where a reinsurance company covers the excess of claim above @xmath189 only , i.e. according to a stop loss contract with priority @xmath186 . in that case ,
lundberg equation ( [ lundberg ] ) reads @xmath190 and the lundberg exponent is @xmath191 .
solutions @xmath48 satisfy @xmath192 where @xmath149 is the @xmath51th generalized lambert function . it is easy to check that @xmath193 so that theorem [ prop_exp_ruin ] entails the asymptotic expansion ( [ exp_ruin_new_continuous ] ) with @xmath51th term given by ( [ g_prime_not_zero_continuous_risk ] ) . since claims here are bounded , the expression of the ruin probability can in fact be made more precise .
indeed @xmath194w(x)$ ] where @xmath195 is the so - called scale function associated to lvy process @xmath196 ( see expression ( 8.7 ) p.215 of @xcite ) .
the expression of @xmath195 is available in theorem 3 of @xcite as an infinite series , it yields that ( [ exp_ruin_new_continuous ] ) can in fact be written as an infinite series , i.e. @xmath197 [ rem_risk_meromorph]it is worth comparing theorem [ prop_exp_ruin ] with the results of kuznetsov and morales @xcite .
they consider a so called meromorphic risk process @xmath198 , which amounts to assume that the claims @xmath199 s have density @xmath200}{dx}=\sum_{m=1}^\infty b_m e^{-\rho_m x},\quad x\ge 0,\ ] ] for some positive coefficients @xmath201 and increasing sequence @xmath202 satisfying @xmath203 .
corollary 1 of @xcite states that the laplace exponent @xmath204 of the lvy process @xmath198 admits a meromorphic extension on @xmath205 and that all the solutions of the ( extended ) lundberg equation are real , negative and simple .
furthermore , denoting these solutions by @xmath206 , the ruin probability has expansion @xmath207 } { \lambda'(-\zeta_n ) } e^{-\zeta_n x } \label{expansion_km}\ ] ] in this framework , condition is not satisfied but we check below that these results are still consistent with theorem [ prop_exp_ruin ] .
the laplace exponent satisfies @xmath208
\quad \mbox{and}\quad \lambda'(z)=c-\alpha\ee \left ( z_1e^{-z z_1}\right ) . \ ] ] elementary computations reveal that @xmath209 one can check that the solutions of @xmath29 satisfy @xmath210 and that @xmath211,\ ] ] so that expansion ( [ exp_ruin_new_continuous ] ) , with corresponding terms given by ( [ g_prime_not_zero_continuous_risk ] ) , corresponds to the @xmath123 first terms of ( [ expansion_km ] ) . with some more effort
, theorem [ prop_exp_ruin ] can be extended to more general gerber - shiu functions , e.g. of the form @xmath212 where @xmath213 , @xmath214 , @xmath215 are non negative , @xmath216 is the ruin time of the risk process and @xmath217 is the running minimum at time @xmath218 , see theorem 2.8 of @xcite as well as theorem 1 of @xcite for example of such expansions .
for ease of presentation , we stick in this paper to @xmath175 as defined by ( [ def_ruin_proba ] ) .
[ example_meromorphic ] we give an example of an infinite expansion in the non lattice case of @xmath62 as in corollary [ cor_expansion_exact ] . conditions ( [ exact_expansion_abs_convergence ] ) and ( [ eq : cond_nonlattice ] ) may look hard to verify in practice .
to exhibit such an @xmath157 , we again use the theory of meromorphic lvy processes . as in remark
[ rem_risk_meromorph ] , we pick spectrally negative process @xmath219 , @xmath220 where @xmath221 is a poisson process with intensity @xmath160 , such that laplace exponent is of the form @xmath222 for some @xmath223 , where sequences of positive real numbers @xmath224 and ( strictly ) increasing @xmath225 are such that series @xmath226 converges so that lvy process @xmath219 is indeed a compound poisson process , see ( 3.2 ) in @xcite .
we will additionally suppose that sequence @xmath225 grows like a polynomial ( in addition to being increasing ) , i.e. there exists @xmath227 such that @xmath228 for some @xmath229 .
remembering that @xmath162 is the cdf of the @xmath199 s , we then consider r.v .
@xmath157 with descending ladder height distribution of lvy process @xmath230 , with corresponding moment generating function @xmath231 where @xmath171 is solution to lundberg equation ( [ lundberg ] ) , see relation ( 5.7 ) p. 87 of @xcite .
we proceed to show that an infinite expansion for the corresponding function @xmath62 is available .
the sole condition for this expansion is ( [ cond_convergence_rho_m ] ) , which is not too stringent and covers a wide range of processes .
the relation between @xmath70 and @xmath232 is given by ( [ compute_rem_km ] ) .
this has two important consequences .
the first one is that @xmath36 is meromorphic , as @xmath232 is .
the second one is that @xmath32 is a solution to @xmath29 iff @xmath233 . by properties ( v ) and ( vi ) of @xcite ( see also theorem 1 ( 7 ) of @xcite ) , one deduces that roots @xmath79 are _ real non negative _ and verify @xmath234 we now turn back to conditions ( [ exact_expansion_abs_convergence ] ) and ( [ eq : cond_nonlattice ] ) .
we start by ( [ exact_expansion_abs_convergence ] ) .
we compute from ( [ compute_rem_km ] ) @xmath235 hence , for all @xmath236 , @xmath237 , which happens to be positive .
we now write @xmath238}.\ ] ] using ( [ roots_ordered ] ) and the dominated convergence theorem , one easily shows that @xmath239}\longrightarrow \sum_{m=1}^\infty \frac{b_m}{c\rho_m } $ ] as @xmath240 .
one then deduces from the above inequality that @xmath241 now , ( [ cond_convergence_rho_m ] ) and ( [ roots_ordered ] ) implies that @xmath242 is a convergent series for all @xmath78 which , combined with ( [ lim_inf_example_infinite_expansion ] ) , implies the convergence ( [ exact_expansion_abs_convergence ] ) . +
we now prove ( [ eq : cond_nonlattice ] ) , by establishing that @xmath243 with @xmath244 . using ( [ expression_g_infinite_expansion ] ) and
@xmath245 implies for all @xmath246 @xmath247 \\ & = & \sum_{m=1}^\infty \frac{b_m}{c\rho_m^2 } \left[\frac{\kappa}{\rho_m-\kappa } + 1 -\rho_m \frac{\rho_m-\rho_n+i\theta}{(\rho_m-\rho_n)^2+\theta^2}\right]\\ & : = & r_n(\theta)+ii_n(\theta).\end{aligned}\ ] ] let us set @xmath248 + \sum_{m=1}^n \frac{b_m}{c\rho_m^2}\rho_m \frac{\rho_n-\rho_m}{(\rho_m-\rho_n)^2+\theta^2}$ ] and @xmath249 .
as @xmath250 is increasing one gets the following inequalities @xmath251:=\xi>0,\\
0\ge r_{n,2}(\theta ) & \ge & \sum_{m = n+1}^\infty \frac{b_m}{c\rho_m^2 } \rho_m \frac{\rho_n-\rho_m}{(\rho_m-\rho_n)^2}=\sum_{m = n+1}^\infty \frac{b_m}{c\rho_m}\frac{1}{\rho_n-\rho_m}:=\chi_n,\end{aligned}\ ] ] so that the real part of @xmath252 verifies the inequality @xmath253 ) entails that @xmath254 as @xmath255 .
one then deduces that @xmath256 for @xmath110 large enough , and @xmath257 proving ( [ eq : cond_nonlattice ] ) . hence infinite expansion ( [ exact_expansion_v_non_lattice ] ) holds .
quite unlike its continuous time counterpart , risk theory in discrete time seems to have been less studied .
we refer to @xcite for an overview of such processes , as well as ( * ? ? ?
* chapter xvi ) .
this type of process is but a skip free random walk , i.e. a random walk with at most unit upward movement , and is in fact studied in many fields of applied probability .
we consider here the so - called binomial discrete time risk model defined by @xmath259 where @xmath260 is the initial reserve , the premium rate is assumed w.l.o.g . to be equal to @xmath136 , the claims @xmath261 form an i.i.d .
sequence taking values in @xmath262 .
we let @xmath263 $ ] and assume that @xmath264 .
the discrete ruin probability is defined by @xmath265 this corresponds to the probability that a @xmath258-valued random walk starting from @xmath266 eventually becomes nonpositive .
the condition @xmath267 ensures that the random walk has a positive drift so that @xmath168 . a closed form expression for @xmath175
may be found in @xcite , however this expression requires computing an infinite number of convolutions of distribution of @xmath268 .
we are here interested in finding a simple expansion of @xmath175 as @xmath269 . similarly to condition in the continuous case , we assume here that @xmath270\mbox { is finite for all } s>0.\ ] ] in this discrete setting , the lundberg equation writes @xmath271=1,\quad z\in{\mathbbm{c}}. \label{eq_lundberg_discrete}\ ] ] restricted to the real numbers , this equation has , thanks to convexity of mean generating function of @xmath272 , a unique solution @xmath171 .
we define the probability mass function @xmath47 defined by @xmath273,\quad k\in\mathbb{n},$ ] with moment generating function @xmath274 the following theorem provides an asymptotic expansion of @xmath175 as @xmath269 , @xmath260 and is the discrete analog of theorem [ prop_exp_ruin ] .
[ thm : ruin_discret ] assume conditions holds .
let @xmath24 be fixed and @xmath60 the solutions to equation @xmath29 in @xmath275 .
then , the ruin probability @xmath175 has the asymptotic expansion @xmath276 + o(e^{-(r+\kappa)x } ) \quad \mbox{as } x\to \infty.\label{exp_ruin_new}\ ] ] if @xmath48 is a simple zero of @xmath49 , i.e. @xmath50 , then the @xmath51-th term in has the simple form @xmath277 we consider a two dimensional ruin problem motivated by applications in reinsurance .
the capitals of two insurance companies are modeled by the risk processes @xmath278 where , @xmath279 , @xmath280 are the respective initial reserves and premium rates , @xmath281 are poisson processes with intensities @xmath282 , and @xmath283 are the corresponding claims with mean @xmath284 . for each @xmath285 , independence between @xmath281 and @xmath283 is assumed .
however no independence is required between processes @xmath286 and @xmath287 .
we suppose that the mean drifts @xmath288 , @xmath285 , are positive , and then define the eventual ruin probabilities for each company @xmath289 we also consider the probability that ( at least ) one of the companies is eventually ruined @xmath290 i.e. the probability that the two dimensional process @xmath291 exits the first quadrants @xmath292 .
we are interested here in the asymptotics of @xmath293 as @xmath294 tend to infinity along a fixed direction @xmath295 .
we refer to @xcite for related results that concern light tailed claims , or @xcite for models featuring heavy tailed claims . for @xmath285 , letting @xmath296 , we suppose that equation @xmath297 has solutions @xmath298 , @xmath299 , @xmath300 in @xmath301 for some @xmath24 , and that those @xmath299 , @xmath300 are simple zeros of @xmath49 .
thus @xmath302 has the following @xmath303 terms expansion from theorem [ prop_exp_ruin ] @xmath304 + \varepsilon_j(x_j)e^{-(r+\kappa_j)x_j},\quad x_j\to + \infty,\quad
j=1,2 , \label{two_dim_expansions}\ ] ] where @xmath305 as @xmath306 , and @xmath307 the main result of this subsection is the following theorem .
[ prop_2dim ] a two term asymptotic for @xmath293 as @xmath294 tend to infinity along the fixed direction @xmath295 , is given by @xmath308+\re\left[d_1 e^{-d_1(q)x}\right ] + \eta_q(x)e^{-\re(d_1(q))x},\quad x\to \infty , \label{two_terms_two dim}\ ] ] where @xmath309 , @xmath310 , @xmath311 and where @xmath312 $ ] and @xmath313 $ ] are the two first dominant functions among @xmath314 , @xmath315 , @xmath316 $ ] and @xmath317 $ ] .
four different cases occur in the asymptotic described in proposition [ prop_2dim ] , depending on the asymptotic direction @xmath318 : @xmath319 to each case corresponds a different two terms expansion for @xmath320 as summarized in figure [ fig_2dim1 ] .
proposition [ prop_2dim ] generalizes the one term expansion given in theorem 3 of @xcite .
the last term in ( [ two_terms_two dim ] ) is only @xmath321 but the condition @xmath322 provides information on how fast this term tends to @xmath1 .
let us consider the stop loss contract scenario with priority @xmath186 .
we assume that @xmath323 is the capital of an insurance company with claims @xmath324 distributed as @xmath325 where @xmath187 is exponentially distributed with parameter @xmath326 .
the second risk process @xmath327 corresponds to the capital of a reinsurance company which covers the excess of claims with priority @xmath189 , i.e. claims @xmath324 are distributed as @xmath328 , as described in example [ ex_stop_loss ] . in that case , the two risk processes @xmath329 and @xmath330 are dependent . because of the memoryless property of the exponential distribution , the @xmath331 s are exponentially distributed with parameter @xmath326 and the intensity of the poisson process @xmath332
is given by @xmath333 as the @xmath331 s are exponentially distributed , @xmath334 see corollary 3.2 p.78 of @xcite .
the two terms expansion for @xmath335 is given by @xmath336 + o(e^{-(r+\kappa_1)x_1})\ ] ] where @xmath337 is the solution to equation ( [ eq_stop_loss ] ) with smallest real part @xmath338 and @xmath339 .
note that even though @xmath323 and @xmath327 are correlated , proposition [ prop_2dim ] may be applied and , since @xmath340 , only three cases occur : @xmath341 this is summarized in figure [ fig_2dim2 ] .
the above lemma is akin to the preliminary result of stone @xcite , see also breiman @xcite and feller and orey @xcite . in these references ,
only the case of a real valued function @xmath47 is considered . in the complex case
, we give a short proof inspired by lemma 10.11 p.221 of @xcite . inspecting the proof of @xcite
, one can see that it is sufficient to check that @xmath348 is integrable at @xmath53 .
the rest of the proof may be applied similarly ( with minor modification ) in order to prove . since @xmath349 and @xmath350
it is sufficient to prove local integrability of @xmath351 at @xmath53 .
we compute further @xmath352 the first term is integrable at @xmath1 since @xmath353 } |f(\theta)| \times \int_{-\varepsilon}^\varepsilon \frac{1-\ee(\cos(\theta x_1))}{\theta^2}d\theta < + \infty.\ ] ] for the integrability of the second term , we need the assumption @xmath344 which implies the existence of a constant @xmath229 such that @xmath354 for @xmath355 .
we use also the inequality @xmath356 , @xmath357 . using this ,
we have @xmath358 let us define @xmath359 with the convention @xmath360 .
we use the basic fact that @xmath361 has fourier transform @xmath362 and that the probabilities @xmath363 s are linked to the fourier transform by @xmath364 which can be verified by writing @xmath365 and using fubini . hence , by lebesgue s dominated convergence and fubini s theorems , @xmath366 note that @xmath367 .
we deduce , thanks to lemma [ lemma_weak_conv ] , @xmath368 we apply the same argument to the i.i.d .
@xmath369 with distribution @xmath370 .
this yields , for all @xmath371 , @xmath372 whence we deduce @xmath373 d\theta \\
= \frac{1}{2\pi } \int_{-\pi}^\pi e^{-ik\theta }
\left(\frac{1}{1- g(i\theta)}- \frac{1}{\mu}\frac{1}{1- e^{i\theta}}\right ) d\theta , \quad k\in{\mathbbm{n } } , \label{proof_theo_lattice_0}\end{gathered}\ ] ] the last line justified by the fact that the integral is convergent .
the integrand function @xmath374 is meromorphic on the domain @xmath375 , @xmath376 .
for @xmath377 small enough , the poles inside this domain are exactly @xmath378 ( the pole at @xmath108 has been removed ) .
cauchy s residue theorem with contour given in the left panel of figure [ figure_contour ] implies @xmath379 using this , we obtain finally @xmath380= \sum_{m=0}^\infty [ 1/\mu - u(k+m+1 ) ] \nonumber\\ & = & \sum_{m=0}^\infty \sum_{j=1}^n\mathrm{res}\left(\frac{e^{-(k+m+1)z}}{1- g(z ) } ; z_j\right)\nonumber\\ & & - \frac{1}{2\pi}\sum_{m=0}^\infty \int_{-\pi}^\pi e^{-(k+m+1)(r+i\theta ) } \left(\frac{1}{1- g(r+i\theta ) } - \frac{1}{\mu}\frac{1}{1- e^{r+i\theta}}\right)d\theta\nonumber \\ & = & \sum_{j=1}^n\mathrm{res}\left(\frac{e^{-kz}}{(e^z-1)(1- g(z ) ) } ; z_j\right)\nonumber\\ & & - \frac{e^{-kr}}{2\pi } \int_{-\pi}^\pi \frac{e^{-ik\theta}}{e^{(r+i\theta)}-1 } \left(\frac{1}{1- g(r+i\theta ) } - \frac{1}{\mu}\frac{1}{1- e^{r+i\theta}}\right)d\theta.\label{expansion_v_lattice_proof_c1}\end{aligned}\ ] ] here we have used that all residues @xmath381 are obtained by integrating @xmath382 on a compact contour around @xmath48 , so that exchanging @xmath383 and @xmath384 is indeed justified by fubini s theorem . by the lebesgue lemma , the last term in ( [ expansion_v_lattice_proof_c1 ] ) is @xmath385 and this proves equation .
we prove that if @xmath4 is absolutely continuous , then for all @xmath57 , @xmath386 clearly , equation implies proposition [ prop_a ] .
it is worth noting that this is a uniform version of the riemann - lebesgue lemma .
let us set for all @xmath246 , @xmath387 , @xmath205 .
since @xmath4 admits a density , the riemann - lebesgue lemma implies that @xmath388 converges pointwise to @xmath1 as @xmath389 when @xmath390 .
let us now note that we have the uniform bound @xmath391\right|\le { \mathbbm{e}}[e^{r_0 x}]<+\infty,\quad \forall z\in\omega.\ ] ] by theorem 1.6.4 p.26 of @xcite , @xmath388 converges uniformly towards @xmath1 as @xmath389 when @xmath32 lies in any compact subset @xmath392 . picking in particular @xmath393\mbox { and } \im(z)=0 \}$ ] yields @xmath394}|g(r+i\theta)|\longrightarrow 0,\quad \theta\to\infty,\ ] ] which we were set to prove.@xmath395 * step 1
: * we prove that condition implies that the equation @xmath29 has only a finite number of solutions in @xmath396 .
condition entails the existence of @xmath397 such that the equation @xmath398 has no solution with @xmath399 and @xmath400 . since furthermore the obvious
bound @xmath401 if @xmath402 exclude solutions in the half - plane @xmath402 , the only possible solutions of @xmath29 in @xmath396 belong to the compact set @xmath403 .
the function @xmath404 being holomorphic , its zeros are isolated .
hence the equation @xmath29 has finitely many solution in the compact set @xmath405 .
* step 2 : * following stone @xcite , we introduce for positive @xmath215 and @xmath406 , @xmath407=\mathbb{e}[u(x - az , h)]\ ] ] with @xmath408 and @xmath272 a random variable with standard gaussian distribution . using fourier analysis
, one can show ( similarly to equation ( 7 ) in @xcite ) @xmath409 setting @xmath410 and @xmath411 , we then split the integral @xmath412 into @xmath413 the analysis of the first term @xmath414 relies on cauchy s residue theorem .
in the domain @xmath396 , the function @xmath415 is meromorphic with poles @xmath378 ( note that the pole at @xmath108 has been removed ) .
we apply cauchy s residue theorem with the contour given in the right panel of figure [ figure_contour ] and we let the parameter @xmath416 tend to @xmath417 . thanks to condition , the contribution of the horizontal parts of the contour vanishes as @xmath418 and we get , for all @xmath419 , @xmath420 with @xmath421 note that the sum of residues is a real number because for conjugate poles @xmath32 and @xmath33 , the residues at @xmath32 and @xmath33 are also conjugate so that the sum is real . similarly for the second term @xmath422 , cauchy s residue theorem yields @xmath423 with @xmath424 equations , , and together yield @xmath425 note that the rieman - lebesgue lemma implies that the second term @xmath426 satisfies , for all @xmath24 , @xmath427 * step 3 : * we analyze here the term @xmath428 and , similarly as equation ( 8) from @xcite , we show that @xmath429 the argument for this statement is given almost without proof in @xcite but can be adapted from a similar argument in @xcite in the following way .
one notices that @xmath430 so that we have , for @xmath431 , @xmath432 with @xmath433 since @xmath434 is analytic on @xmath30 ( the pole at @xmath435 has been removed ) , cauchy s residue theorem and lebesgue s lemma yield @xmath436 for the other terms , we use the fact that the function @xmath437 is bounded for @xmath438 and satisfies @xmath439 and @xmath440 this justifies the following estimates @xmath441 ) , we obtain ( [ proof_thm_expansion_v_8 ] ) .
the term @xmath442 can be replaced by @xmath443 because @xmath444 is arbitrary .
equations , and together yield @xmath445 we prove the following inequality : @xmath446 recall the definition @xmath447 $ ] with @xmath448 and @xmath272 a standard gaussian random variable . using the fact that @xmath87 is non decreasing , we check , for @xmath449 , @xmath450 let @xmath451 be such that , for all @xmath452 , @xmath453 .
besides , the fact that renewal function @xmath454 is sub - additive implies that there exists some constant @xmath455 independent from @xmath406 and @xmath164 such that @xmath456 by the definition @xmath447 $ ] , both sides of equation are obtained by splitting @xmath457+{\mathbbm{e}}[u(x\pm e^{-rx}-az , h){\mathbbm{1}}_{\ { |z| > x\}}]\ ] ] with the values @xmath458 and @xmath459 .
let us now prove the following estimate @xmath460 this is roughly obtained from by setting @xmath461 this can be made rigorous thanks to equation .
we have , uniformly in the neighborhood of @xmath48 , @xmath462 , @xmath463\\ & & .\left[t(1,z)+ o(e^{-rx})\right][1+o(e^{-rx})]\\ & = & e^{-xz } t(1,z)+ o(e^{-rx})\end{aligned}\ ] ] so that @xmath464 besides , left hand side of equation and equation entail @xmath465 now from , we have @xmath466 together with and since @xmath467 , this yields @xmath468 a similar upper bound is proved in the same way , using the right hand side of equation .
we deduce @xmath469 we can replace @xmath470 by @xmath471 because @xmath472 is arbitrary in @xmath473 .
this proves eq . .
[ [ proof - of - corollaries - cor_expansion_exact - and - cor_expansion_exact_2 ] ] proof of corollaries [ cor_expansion_exact ] and [ cor_expansion_exact_2 ] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ we consider first the lattice case . according to condition
, one can consider @xmath478 an increasing sequence such that @xmath479 and @xmath480 according to equation , @xmath481 with @xmath482 the number of solutions of the equation @xmath29 in @xmath483 .
lebesgue s dominated convergence theorem implies , for @xmath484 , @xmath485 this yields the result ( [ exact_expansion_v_lattice ] ) .
we now consider the non - lattice case .
thanks to assumption ( [ eq : cond_nonlattice ] ) , let @xmath478 be such that @xmath479 and @xmath486 we will prove below that setting @xmath34 , @xmath487 and @xmath488 in equation and letting @xmath489 , we obtain @xmath490 remembering that roots @xmath491 are simple .
this is justified as follows : * condition together with lebesgue s dominated convergence theorem implies @xmath492 * recalling that @xmath493 and @xmath494 are respectively given by ( [ def_i_tilde_1 ] ) and ( [ def_i_tilde_2 ] ) , the integral term @xmath495 with @xmath34 , @xmath487 and @xmath496 converge to @xmath1 as @xmath497 . indeed , condition and the upper bound @xmath498
imply @xmath499 this last quantity goes to @xmath1 as @xmath479 .
* let us recall inequality @xmath500 for all @xmath501 .
as roots @xmath79 are simple and real , one has for all @xmath502 , and for @xmath110 large enough,@xmath503 e^{-xz}t(1,z ) \varphi(z);z_j\right)\right|\\ = \left| \left [ e^{\pm e^{-r_nx } z_j}e^{a_n(x)^2z_j^2/2}-1\right]\right|.\left|\mathrm{res}\left ( e^{-xz}t(1,z ) \varphi(z);z_j\right)\right|\\ \le e\left| \pm e^{-r_nx}z_j + a_n(x)^2z_j^2/2 \right|.\left|\mathrm{res}\left ( e^{-xz}t(1,z ) \varphi(z);z_j\right)\right|\\ \le m e^{-2 r_nx}r_n^2 .\left|\mathrm{res}\left ( e^{-xz}t(1,z ) \varphi(z);z_j\right)\right|\end{gathered}\ ] ] where @xmath504 is a constant independent from @xmath110 and @xmath51 , but which may depend on @xmath78 .
one thus deduces @xmath505 which tends to zero as @xmath255 because of assumption ( [ exact_expansion_abs_convergence ] ) . *
the fact that @xmath506 tends to @xmath1 as @xmath255 and convergence ( [ conv_v_h_a ] ) ( with @xmath34 ) as well as inequality ( [ inequality_infinite_sum ] ) implies from ( [ proof_thm_expansion_v_2_0 ] ) . in the lattice case ,
we deduce from corollary [ cor_expansion_exact ] that @xmath507 finally , note that @xmath108 , @xmath245 , @xmath508 so that @xmath509 in the non - lattice case , we deduce from corollary [ cor_expansion_exact ] that @xmath510 the result follows since @xmath511 .
assumption that the moment generating function of @xmath162 is finite everywhere is equivalent to the fact that the tail function @xmath165 decreases super - exponentially fast , i.e. @xmath512 is in @xmath513 with @xmath514 the following lemma can be verified easily .
it is well known that @xmath175 satisfies the following defective renewal equation @xmath519 where @xmath520 can be seen as the cdf of a defective distribution ( because @xmath521 ) , see e.g. corollary 3.3 p.79 of asmussen and albrecher @xcite or eq .
( 7.2 ) p.377 of feller @xcite .
we introduce the probability distribution @xmath522 and use the notation @xmath523 for the moment generating function and the first and second moments of @xmath4 .
elementary calculation yields @xmath211 .
\label{g(zj)_prime_ruin}\ ] ] thanks to the assumption @xmath524 and lemma [ lemme_d ] , the tail function of @xmath4 belongs also to @xmath513 whence the moment generating function @xmath70 is defined for all @xmath205 . setting @xmath525 ,
entails the non - defective renewal equation @xmath526 see feller @xcite eq .
( 6.12 ) p.376 .
the distribution @xmath4 is non - lattice so that smith s renewal theorem entails the classical cramer - lundberg asymptotics : @xmath527 the last equality being obtained by direct computation , see e.g. theorem 5.3 p.86 of @xcite .
we now wish to provide an expansion of @xmath528 and provide extra terms in ( [ constant_c ] ) .
let us introduce @xmath529 the solution to ( [ renewal_eq_dieze ] ) is given by @xmath530 using the fact that @xmath531 , an integration by parts and the definition of @xmath532 , we obtain @xmath533 we recall that @xmath534 belongs to @xmath535 defined in ( [ def_cal_d ] ) , so that one easily sees that @xmath536 defined in ( [ renewal_eq_dieze ] ) belongs to @xmath535 , and that @xmath537 and @xmath538 are @xmath539 from lemma [ lemme_d ] .
thus we only need to study asymptotics of the terms @xmath540 and @xmath541 as @xmath269 in ( [ expansion_ruin1 ] ) .
we start with @xmath541 .
let us note that , as @xmath70 is defined for all @xmath205 , @xmath542 defined in ( [ def_r ] ) is equal to @xmath417 , as well as @xmath61 in theorem [ thm_expansion_v ] .
expansion ( [ expansion_v ] ) in theorem [ thm_expansion_v ] yields @xmath543 now turning to @xmath540 . by a change of variable
one has that @xmath544 , so that , using expansion ( [ expansion_v ] ) in theorem [ thm_expansion_v ] for @xmath545 yields @xmath546 for some function @xmath547 vanishing at @xmath417 . furthermore , it is easy to check that @xmath548 .
one then verifies that @xmath549 is an @xmath471 , by writing the residue as an integral over a fixed contour including @xmath48 then using lemma [ lemme_d ] .
one also checks that the last term on the right hand side of ( [ exp_i2_1 ] ) is @xmath471 by a dominated convergence argument .
one may thus write in ( [ exp_i2_1 ] ) @xmath550 deriving @xmath551 with @xmath552 given in ( [ expression_l(x ) ] ) , expressing the residue as an integral over a fixed contour and using fubini , thus yields the following expression for the terms in the summation on the right hand side of ( [ exp_i2_1_bis ] ) : @xmath553;z_j\right)\nonumber\\ & & = \mathrm{res}\left(\frac{e^{-x z}}{z(1-g(z ) ) } \left [ \frac{\alpha}{c } \ee\left ( \int_0^\infty \int_0^\infty \kappa e^{(z+\kappa)y } { \mathbbm{1}}_{[z\ge v\ge y ] } dy dv \right)\right . \right.\nonumber\\ & & - \left . \left.\frac{\alpha}{c } \ee\left ( \int_0^\infty e^{(z+\kappa)y } { \mathbbm{1}}_{[z\ge y]}dy\right)\right];z_j\right)\nonumber\\ & & = \mathrm{res}\left(\frac{e^{-x z}}{z(1-g(z ) ) } \left [ \frac{\alpha}{c } \frac{\kappa}{z+\kappa}\ee\left ( \frac{e^{(z+\kappa)z}-1}{z+\kappa}-z\right)\right.\right.\nonumber\\ & & - \left .
\left.\frac{\alpha}{c}\frac{1}{z+\kappa}\ee\left ( e^{(z+\kappa)z}-1\right)\right];z_j\right),\end{aligned}\ ] ] which , since @xmath48 satisfies @xmath554 with @xmath70 given by ( [ def_g_ruin ] ) , yields @xmath555;z_j\right)\nonumber\\ & & = \mathrm{res}\left(\frac{e^{-x z}}{z(1-g(z ) ) } \frac{[-z - \frac{\alpha}{c } \kappa m]}{z+\kappa};z_j\right).\label{exp_i2_2}\end{aligned}\ ] ] plugging ( [ exp_i2_2 ] ) in ( [ exp_i2_1_bis ] ) and summing @xmath540 and expression ( [ exp_i4_1 ] ) of @xmath541 then yields from ( [ expansion_ruin1 ] ) @xmath556}{z+\kappa};z_j\right)+ o(e^{-rx}).\ ] ] let us now note that in the case where @xmath50 , the residue in the @xmath51th term in the above can be explicitly given and , thanks to ( [ g(zj)_prime_ruin ] ) , is equal to ( [ g_prime_not_zero_continuous_risk ] ) .
let us also note that expression of @xmath532 provided in ( [ constant_c ] ) is but @xmath557 with @xmath108 and that @xmath558 , so that the above yields expansion ( [ exp_ruin_new ] ) .
we let @xmath559 , @xmath371 , the probability mass distribution of @xmath268 and denote its survival function by @xmath560=\sum_{j = k+1}^\infty b(k)$ ] .
let us in particular note that @xmath561 can be positive .
in other words , there is a possibility that there is no claim ( or equivalently , a claim of size @xmath1 ) at time @xmath110 .
we define @xmath562 and suppose here that the @xmath199 s verify that @xmath563 .
we let @xmath171 its unique real positive solution .
similarly to ( [ expression_l(x ) ] ) , ( [ lundberg ] ) and ( [ def_g_ruin ] ) we define the following discrete measures @xmath568 , associated probability mass function @xmath569 and complex valued function @xmath570 which we recall here : @xmath571 @xmath70 is thus defined for all @xmath572 and is equal to @xmath573 for some integer valued r.v .
@xmath157 with probability mass function @xmath569 , and with expectation and second moment @xmath574 and @xmath575 .
we proceed along the lines of proof of proposition [ prop_exp_ruin ] .
setting @xmath576 , @xmath260 , the discrete time analog of ( [ renewal_eq ] ) is @xmath577 see e.g. proposition 1.2 p.488 in @xcite . letting @xmath525 and @xmath578 , and by the definition of @xmath579 for all @xmath580 in ( [ def_f_ruin_discrete ] )
, we obtain from ( [ renewal_eq_discrete ] ) the renewal equation @xmath581 since @xmath157 is lattice , the corresponding smith s renewal theorem implies that @xmath582 the last equality can be verified by direct computation .
let us note in passing that ( [ constant_c_discrete ] ) provides the cramer lundberg asymptotics @xmath583 for @xmath269 , @xmath260 , in the discrete case .
defining now @xmath584 with @xmath360 , solution to ( [ renewal_eq_dieze_discrete ] ) is given by @xmath585,\\ & = & \frac{1}{\mu } \sum_{k=0}^x
z(x - k ) + \sum_{k=0}^x z(x - k ) v(k ) - \sum_{k=-1}^{x-1 } z(x - k-1)v(k),\\ & = & \frac{1}{\mu } \sum_{k=0}^x z(k ) + \sum_{k=0}^{x-1 } [ z(x - k)-z(x - k-1)]v(k ) + z(0)v(x
) - z(x ) v(-1).\end{aligned}\ ] ] substracting in the above constant @xmath532 given by ( [ constant_c_discrete ] ) yields @xmath586v(k ) + z(0)v(x ) - z(x ) v(-1 ) .
\label{expansion_ruin1_discrete}\ ] ] the first and last term on the right hand side of ( [ expansion_ruin1_discrete ] ) are @xmath385 by lemma [ lemme_d_d ] , hence we are interested in the following quantities @xmath587v(k)=\sum_{k=1}^{x } [ z(k)-z(k-1)]v(x - k),\\ i_2(x)&:= & z(0)v(x)=\bar{l}(0 ) v(x)=m v(x ) .
\end{aligned}\ ] ] thanks to theorem [ thm_expansion_v_lattice ] , one has that @xmath588 and we then turn to @xmath537 . writing @xmath589e^{\kappa k}\bar{l}(k )
+ e^{\kappa ( k-1 ) } [ \bar{l}(k ) - \bar{l}(k-1)]\nonumber\\ & = & [ 1-e^{-\kappa}]e^{\kappa k}\bar{l}(k ) - e^{\kappa ( k-1 ) } l(k-1)\nonumber\\ & = & [ 1-e^{-\kappa } ] z(k)- e^{\kappa ( k-1 ) } l(k-1 ) , \label{i1_disctete1}\end{aligned}\ ] ] then writing , in view of expansion ( [ expansion_v_lattice ] ) , @xmath590 for some @xmath591 vanishing at @xmath592 , one obtains ,
@xmath593 e^{\kappa k}\bar{l}(k ) e^{-(x - k)z};z_j \right]\label{i1_disctete2}\\ & - & \sum_{j=1}^n \mathrm{res}\left [ \frac{1}{(e^{z}-1)(1-g(z)}\sum_{k=1}^{x } e^{\kappa ( k-1 ) } l(k-1 ) e^{-(x - k)z};z_j \right]\label{i1_disctete3}\\ & + & \sum_{k=1}^x [ z(k)-z(k-1 ) ] \varepsilon_{x - k } e^{-r(x - k)}. \label{i1_disctete4}\end{aligned}\ ] ] using lemma [ lemme_d_d ] as well as a dominated convergence theorem , it is not hard to see that ( [ i1_disctete4 ] ) is an @xmath471 as @xmath269 , @xmath260 .
we then study ( [ i1_disctete2 ] ) and ( [ i1_disctete3 ] ) .
one verifies that @xmath594\in { \cal d}_d$ ] by writing the residue as an integral over a fixed contour including @xmath48 then using lemma [ lemme_d_d ] , so that in ( [ i1_disctete3 ] ) we compute @xmath595\nonumber\\ & = & \mathrm{res}\left[\frac{1}{(e^{z}-1)(1-g(z))}e^{-x z } e^{z}\sum_{k=0}^{\infty}e^{(z+\kappa ) k}\pp(z\ge k+1 ) ; z_j\right]\nonumber\\ & -&\mathrm{res}\left[\frac{1}{(e^{z}-1)(1-g(z ) ) } e^{-x z } \sum_{k = x+1}^{\infty } e^{\kappa ( k-1 ) } l(k-1 ) e^{k z};z_j\right]\nonumber\\ & = & \mathrm{res}\left[\frac{1}{(e^{z}-1)(1-g(z ) ) } e^{-x z}e^{z}g(z);z_j\right]+ o(e^{-rx}),\quad j=1, ... ,n . \label{i1_disctete5}\end{aligned}\ ] ] next , we compute , for all @xmath32 , @xmath596 l(n)\nonumber\\ & = & \sum_{n=1}^\infty \left [ e^{\kappa + z } \frac{1- e^{(\kappa + z ) n}}{1-e^{\kappa + z } } \right ] l(n)\nonumber\\ & = & \frac{e^{\kappa + z } } { 1-e^{\kappa + z } } \sum_{n=1}^\infty l(n)- \frac{1}{1-e^{\kappa + z } } e^{\kappa+z } \sum_{n=1}^\infty e^{(\kappa+ z ) n } l(n)\nonumber\\ & = & \frac{e^{\kappa + z } } { 1-e^{\kappa + z } } [ m - l(0 ) ] - \frac{1}{1-e^{\kappa + z } } e^{\kappa+z } [ g(z)-l(0)]\nonumber\\ & = & \frac{e^{\kappa + z}}{1-e^{\kappa + z}}[m - g(z ) ] , \label{i1_disctete6}\end{aligned}\ ] ] one verifies thanks to lemma [ lemme_d_d ] and writing the residue as a contour around @xmath48 that @xmath597\in { \cal d}_d$ ] , so that inserting ( [ i1_disctete6 ] ) in ( [ i1_disctete2 ] ) yields @xmath598 e^{\kappa k}\bar{l}(k ) e^{-(x - k)z};z_j\right]\\ = \mathrm{res}\left[\frac{1}{(e^{z}-1)(1-g(z ) ) } e^{-x z}\frac{e^{\kappa + z}}{1-e^{\kappa + z}}[m - g(z)];z_j\right]+ o(e^{-rx } ) .
\label{i1_discrete30}\end{gathered}\ ] ] respectively inserting ( [ i1_disctete5 ] ) and ( [ i1_discrete30 ] ) in ( [ i1_disctete3 ] ) and ( [ i1_disctete2 ] ) in @xmath537 , and adding @xmath540 in ( [ i2_discrete ] ) , yields @xmath599 e^{-x z};z_j\right]\nonumber\\ & - & \sum_{j=1}^n \mathrm{res}\left [ \frac{g(z)}{(e^{z}-1)(1-g(z ) ) } e^{z } e^{-x z};z_j\right]\nonumber\\ & + & \sum_{j=0}^n \mathrm{res } \left [ \frac{m}{(e^{z}-1)(1-g(z))}e^{-x z};z_j\right]+ o(e^{-rx } ) , \nonumber\\ & = & -\sum_{j=1}^n \mathrm{res } \left[\frac{1}{1-g(z ) } \frac{m - g(z)e^{\kappa+z}}{1-e^{\kappa+z } } e^{-x z};z_j\right]+o(e^{-rx } ) . \label{i1_i2_discrete}\end{aligned}\ ] ] plugging ( [ i1_i2_discrete ] ) into ( [ expansion_ruin1_discrete ] ) yields ( [ exp_ruin_new ] ) , provided that we prove that residue of @xmath600 at @xmath601 equals @xmath602 , @xmath532 given by ( [ constant_c_discrete ] ) .
this computation is in fact included in the case where the @xmath603 s are non zero , in which case @xmath51th term of ( [ i1_i2_discrete ] ) is equal to ( [ expression_term_ruin_discrete ] ) , which we proceed to consider now .
direct computation from ( [ def_g_ruin_discrete ] ) coupled to the fact that @xmath554 gives @xmath604+[1-\ee(e^{(\kappa+z_j)z})]e^{\kappa+z_j}}{(1- e^{\kappa+z_j})^2}\\ = -\frac{\ee\left(z e^{(\kappa+z_j)z}\right)}{1- e^{\kappa+z_j}}+ \frac{e^{\kappa+z_j}}{1- e^{\kappa+z_j}}\end{gathered}\ ] ] so that it is easy to verify that @xmath605 $ ] is ( [ expression_term_ruin_discrete ] ) times @xmath606 .
we recall that two functions @xmath47 and @xmath406 satisfy @xmath607 iff @xmath608 tends to @xmath1 as @xmath609 . in the following we consider cases where @xmath610 is such that functions respectively satisfy @xmath611 \gg \left\{c_0 ^ 2 e^{-q\kappa_2 x } , \
\re\left[c_1 ^ 2 e^{-q(\kappa_2+z^2_1 ) x}\right]\right\},\label{case1_2dim}\\ & & c_0 ^ 1 e^{-\kappa_1 x } \gg c_0 ^ 2 e^{-q\kappa_2 x } \gg \left\{\re\left[c_1 ^ 1 e^{-(\kappa_1+z^1_1 ) x}\right],\
\re\left[c_1 ^ 2 e^{-q(\kappa_2+z^2_1 ) x}\right]\right\},\label{case2_2dim}\end{aligned}\ ] ] all other cases being treated similarly .
we will use the following inequalities ( an easy and direct consequence of lemma 2 ( i ) of @xcite ) @xmath612 let us first consider the case ( [ case1_2dim ] ) . using expansions ( [ two_dim_expansions ] ) with @xmath613 , @xmath614 , as well as ( [ ineq_app ] )
we get @xmath615 \\\le c_0 ^ 2 e^{-q\kappa_2 x } + \re\left[c_1 ^ 2 e^{-q(\kappa_2+z^2_1 ) x}\right ] + \varepsilon_1(x)e^{-(r+\kappa_1)x}+ \varepsilon_2(qx)e^{-(r+\kappa_1)qx}.\label{expansion_2dim_1}\end{gathered}\ ] ] we now note that the terms on the far left and right hand side of ( [ expansion_2dim_1 ] ) divided by @xmath616 tend to @xmath1 because of assumption ( [ case1_2dim ] ) , which proves ( [ two_terms_two dim ] ) .
+ we now consider case ( [ case2_2dim ] ) . using again ( [ ineq_app ] ) we get , similarly to ( [ expansion_2dim_1 ] ) , @xmath617 - c_0 ^ 2 e^{-q\kappa_2 x } \le \psi_{\small \mbox{or}}(x , qx)- c_0 ^ 1 e^{-\kappa_1 x } - c_0 ^ 2 e^{-q\kappa_2 x } \\\le \re\left[c_1 ^ 1 e^{-(\kappa_1+z^1_1 ) x}\right ] + \re\left[c_1 ^ 2 e^{-q(\kappa_2+z^2_1 ) x}\right ] + \varepsilon_1(x)e^{-(r+\kappa_1)x}+ \varepsilon_2(qx)e^{-(r+\kappa_1)qx}.\label{expansion_2dim_2}\end{gathered}\ ] ] because of assumption ( [ case2_2dim ] ) , we get from ( [ expansion_2dim_2 ] ) that @xmath618 - c_0 ^ 2 e^{-q\kappa_2 x}}{c_0 ^ 2 e^{-q\kappa_2 x}}\\ \le \liminf_{x\to\infty}\frac{\psi_{\small \mbox{or}}(x , qx)- c_0 ^ 1 e^{-\kappa_1 x } - c_0 ^ 2 e^{-q\kappa_2 x}}{c_0 ^ 2 e^{-q\kappa_2 x}}\\ \le \limsup_{x\to\infty}\frac{\psi_{\small \mbox{or}}(x , qx)- c_0 ^ 1 e^{-\kappa_1 x } - c_0 ^ 2 e^{-q\kappa_2 x}}{c_0 ^ 2 e^{-q\kappa_2
x}}\\ \le \lim_{x\to\infty}\frac{\re\left[c_1 ^ 1 e^{-(\kappa_1+z^1_1 ) x}\right ] + \re\left[c_1 ^ 2 e^{-q(\kappa_2+z^2_1 ) x}\right ] + \varepsilon_1(x)e^{-(r+\kappa_1)x}+ \varepsilon_2(qx)e^{-(r+\kappa_1)qx}}{c_0 ^ 2 e^{-q\kappa_2 x } } \\=0,\end{gathered}\ ] ] which proves ( [ two_terms_two dim ] )
. * acknowledgment . *
the authors wish to thank an anonymous referee for pointing out the informal proof given at the end of section [ sec : main_results ] , comments leading to remark [ comment_referee ] , as well as reference @xcite leading to the infinite expansion for @xmath175 in example [ ex_stop_loss ] .
this work was partially supported by the french national agency of research anr grant , project ammsi , number anr-2011-bs01 - 021 .
asmussen s. and bladt m. ( 1996 ) .
renewal theory and queueing algorithms for matrix exponential distributions .
matrix - analytic methods in stochastic models ( a.s .
alfa & s. chakravarty , eds . ) , pp.313-341 .
marcel dekker , new york .
biard , r. ( 2013 ) .
asymptotic multivariate finite - time ruin probabilities with heavy - tailed claim amounts : impact of dependence and optimal reserve allocation . _
bulletin franais dactuariat _ , 13(26 ) , pp.7992 . | a high order expansion of the renewal function is provided under the assumption that the inter - renewal time distribution is light tailed with finite moment generating function @xmath0 on a neighborhood of @xmath1 .
this expansion relies on complex analysis and is expressed in terms of the residues of the function @xmath2 . under the assumption that @xmath0 can be extended into a meromorphic function on the complex plane and some technical conditions ,
we obtain even an exact expansion of the renewal function .
an application to risk theory is given where we consider high order expansion of the ruin probability for the standard compound poisson risk model .
this precises the well known crmer - lundberg approximation of the ruin probability when the initial reserve is large . |
the one - way model for quantum computation @xcite , or measurement - based quantum computation ( mbqc ) , is an alternative to the standard circuit model @xcite , providing another possible route towards practical quantum information processing . in this scheme ,
quantum computation proceeds by adaptive single - qubit measurement on an entangled universal resource state .
particular physical implementations , however , may necessitate modifying the original scheme , which used a cluster state @xcite on the @xmath0-dimensional square lattice as its universal resource .
other resource states that have been identified as universal under this measurement - based scheme include the cluster state corresponding to a @xmath0-dimensional hexagonal lattice @xcite , and , in addition , other extended measurement - based schemes have been proposed @xcite . in this paper , rather than searching for classes of universal states , we consider the impact of possible experimental imperfections on the resource states to see if universal mbqc can be achieved . we also show that our model exhibits a phase transition in the computational power of the resource state , in the sense that its potential for universal quantum computation is lost suddenly at a particular critical degree of imperfection .
the notion of universality we use throughout is that of _ efficient _ universality ( in the sense of ref .
@xcite ) i.e. that any quantum computation can be implemented with only a polynomial overhead in spatial resources and time,)-sized quantum circuits on @xmath1 qubits can be generated using a resource state of size poly(@xmath1 ) , and in time poly(@xmath1 ) ( both in terms of quantum measurements and classical side - processing ) .
] as efficiency is a crucial aspect of computational power .
typically , one imagines a cluster state as being prepared by placing qubits in an @xmath2 lattice and then entangling the qubits with an ising type interaction between nearest neighbours .
the model we consider in this paper is one in which there is only a finite probability @xmath3 that there was initially an atom in any given lattice site .
the value of @xmath3 is independent and identical for each lattice site .
we could imagine this situation arising when we fill an optical lattice with atoms from a mott - insulator with some sub - unital filling factor @xcite .
we assume there is only at most one atom per site , and furthermore that the atom positions are heralded .
the resulting state looks like a regular lattice graph state , but with some vertices and their incident edges removed .
percolation is the widely - known phenomenon that in lattices with imperfect filling there exists a unique giant connected cluster of size @xmath4 ( where @xmath5 ) , when the filling factor lies above a certain percolation threshold ( in the so - called supercritical phase ) .
one might naively think it is obvious that above this threshold the corresponding state in our model would be universal as a resource of mbqc .
however , a connected graph state of size @xmath4 is generally not sufficient to be a universal resource .
for example , a 1d chain ( i.e. , 1d cluster state ) of @xmath4 is not universal , nor is any @xmath4 tree structure nor even the complete graph on @xmath6 vertices @xcite . in this paper , we show that the faulty lattice is indeed universal by showing explicitly how to convert the supercritical resource state into the regular 2d cluster state with constant overhead ( dependent only on @xmath3 ) and polynomial - time classical processing . finally , we interpret the transition to universality at the percolation threshold as a phase transition , with the entanglement of the faulty lattice as an order parameter .
specifically , we use an entanglement measure called the _ entanglement width _
@xcite which is well suited as an entanglement measure for states being used in mbqc .
we show that the scaling exponent of the entanglement width of the faulty lattice goes through a discontinuity at the critical probability @xmath7 in the large @xmath6 limit .
it should be noted that in this paper all statistical statements , such as discussion of phase transitions , should be interpreted in the following way .
we consider properties that hold true almost surely ( more precisely , with probability exponentially close to one in the thermodynamic limit ) , for all the possible realizations of the configuration of holes with a given @xmath3 .
this paper does not directly consider mixed states , although clearly the mixed state corresponding to a mixture of states weighted according to their probability for a given @xmath3 , would reduce to the same ensemble of pure states under a non - demolition site - occupation number measurement .
we briefly mention the literature which utilized the idea of percolation theory for the quantum information processing .
a simple modification of the situation above is to consider the effect when all lattice sites are filled with an atom but the entangling gate fails with some probability , leading to a graph state with missing edges instead of vertices .
this is the model considered in ref .
@xcite , motivated by linear optical implementations of quantum computation . in that paper the authors consider an initial faulty resource of size @xmath8 , a @xmath9-dimensional resource , and show that above the edge percolation threshold for a @xmath9d lattice , it is possible to concentrate an @xmath10 hexagonal cluster state .
the approach used in refs .
@xcite is derived from renormalization theory and is applicable to a wide variety of lattice geometries , and may also be adapted to work with vertex defects , rather than edge defects .
we note that in the context of quantum networks , on the other hand , classical percolation methods were not found to be optimal @xcite .
because this paper relies on results from rather distinct disciplines , this section provides the necessary background for readers who may be unfamiliar with one or more topics .
see ref .
a _ graph _
@xmath11 is a set of points , called _ vertices _ , along with a set of pairs of vertices which are called _
edges_. vertices are usually labeled with numbers , such as vertex @xmath12 , and edges are usually labeled by the vertices they connect , such as edge @xmath13 that connects vertices @xmath12 and @xmath14 .
the vertices @xmath12 and @xmath14 are sometimes said to be the _
endpoints _ of the edge , or be _ connected _ by the edge , and also sometimes the edge is said to be _ incident _ on vertices @xmath12 and @xmath14 .
a graph is represented pictorially with dots representing vertices and lines connecting the dots representing edges .
two vertices connected by an edge are said to be _ neighbors _ , and the number of neighbors of a vertex is dubbed its _
degree_. the _ neighborhood _ of vertex @xmath12 , denoted as @xmath15 , is the set of vertices that are neighbors of vertex @xmath12 .
the neighborhood of a set of vertices is the union of the neighborhoods of the vertices in the set ( excluding the vertices themselves ) .
a _ path _ in @xmath11 is an ordered set of vertices and their connecting edges , @xmath16 and @xmath17 , and each vertex and edge appear exactly once .
thus , paths have no `` dangling '' edges and can not self - intersect .
the first and last vertices in the path are the _ starting _ and _ ending _ vertices , respectively .
there are two more important concepts we need to define : graph minors and topological minors .
these are illustrated in figure [ f : minor ] .
graph minor _ , or simply minor , is obtained by first partitioning the vertices into disjoint connected sets , and then considering the graph between these sets , while disregarding any self - loops or parallel edges ; see figure [ f : minor](a ) .
topological minor _ is a special case of a graph minor where the resulting minor has the same topology as the original graph .
more precisely , if we view the original graph as a topological space , a topological minor is homeomorphic to the original graph .
this is illustrated in figure [ f : minor](b ) . for a rigorous definition of these concepts ,
see @xcite .
the universal resource states we address are called _ graph states _
@xcite and are defined in terms of a graph @xmath11 as follows .
let @xmath19 , @xmath20 , and @xmath21 denote the pauli operators on the @xmath12-th qubit . for each vertex @xmath12 of the graph @xmath11
, prepare a qubit initially in the @xmath22 eigenstate of @xmath19 .
then apply the controlled - z operator between each pair of qubits who are neighbors in @xmath11 . for a connected graph , what results is an entangled pure quantum state that is well represented by the underlying graph @xmath11 .
the class of graph states resulting from @xmath0d regular lattices are called _
cluster states_. the two lattices that we use are the square lattice and the hexagonal lattice .
both have been shown to be universal for one - way quantum computation @xcite .
there are three types of errors : _ closeness _ errors , _ degree _ errors and _ lattice _ errors .
a closeness error occurs when a path gets too close to itself or another path .
a degree error is where a vertex is not degree three , as is proper for a hexagonal lattice . a lattice error happens when the global topology is not that of a regular hexagonal lattice .
additionally , the first two error classes are subdivided according to what type of paths or degrees are involved , as explained below .
* _ self - h _ errors , where a single horizontal path gets too close to itself .
similarly , there are _ self - v _ errors . *
_ h - h _ errors , where a horizontal path gets too close to another disjoint horizontal path . similarly
, there are _ v - v _ errors . * _ h - v _ errors , where a horizontal path gets too close to a vertical path , except for a single intersection .
degree errors are also subdivided . at the end of the entire algorithm
, we want every vertex to be degree 3 ( except on the boundary of the lattice , which is dealt with separately ) .
thus , we need to eliminate vertices of degrees 1,2 and 4 . _
degree 1 _ errors correspond to `` dangling '' qubits while _ degree 2 _ errors correspond to `` wires '' , both of which can be contracted by measuring @xmath23 .
while these are handled in the quantum stage of the algorithm , _ degree 4 _ errors are handled in the classical stage . as discussed in section [ s : overview ] , the classical part of our algorithm proceeds in three main stages : path identification , error localization , and correction of local errors .
after this is complete , we will have identified a subset of the qubits whose topology is that of a hexagonal lattice , and the quantum part of the algorithm ( discussed in the next subsection ) will contract this lattice into a hexagonal lattice cluster state .
as discussed in subsection [ s : vertexdisjoint ] , we know that there exists @xmath24 vertex - disjoint h- and v - paths . the wall follower algorithm defined above will necessarily find all these paths since it finds boundary paths @xcite . for the h - paths , we use the 2-local version of rhwf and for the v - paths , we use rhwf , but keep only every third path . clearly then we do not have h - h or v - v errors . to eliminate self errors
we simply find the shortest path amongst the original vertices and their accompanying internal edge sets , which can only contain degree @xmath25 and @xmath0 vertices and hence no self errors .
taking the shortest path also can not introduce any closeness errors because we do not add any vertices , we only delete them .
thus we have shown that after stage c.1 of the algorithm we have identified @xmath24 valid h- and v - paths . , while the blue lines are the bridges incident on @xmath26 for two vertical paths , @xmath27 and @xmath28 .
the solid blue line is associated with the index @xmath14 , while the dashed line is associated with @xmath29 .
closures of upper and lower abutments are denoted by black circles . in ( a ) and ( b )
, we assume the closures of the upper and lower abutments are disjoint , but that the total abutments overlap . in case ( c ) ,
we treat the possibility that the upper or lower abutment closures themselves overlap .
note that , while the case of complete containment is not illustrated as a sub - case of ( c ) , it is still treated in the text.,width=302 ] once this is done , we can find the complete bridge decomposition as defined above .
since we do this by removing vertices while preserving connectivity , this process clearly does not introduce any new closeness errors .
this step also localizes h - v errors to the region near the abutments as in fig .
[ f : abutments - ordered ] .
the next step is to take an _ alternating bridge decomposition _ , which is defined as follows .
in a @xmath12-stripe we can delete the bridges @xmath30 such that @xmath31 has odd parity , which is alternating in the sense that we take alternating bridges as indexed by @xmath12 and @xmath14 .
we want to show that the alternating bridge decomposition is also alternating in a stronger , topological sense .
we do this by showing below that the natural partial ordering on the abutments is actually a total ordering ; that is , each pair of abutments along an h - path can be compared . intuitively , then
, the global topology of the lattice is that of a square lattice except for possible errors localized to the abutments .
taking alternating bridges will then clearly give a hexagonal lattice .
this intuition is illustated in fig .
[ f : abutments - ordered ] . showing that abutments are totally ordered ( as sets ) follows from showing that closures of abutments are disjoint .
we show that they are disjoint by considering the various ways they might overlap and showing that each can not occur .
this leaves several configurations , as illustrated in fig .
[ f : total - order ] . .
horizontal paths are colored red , and their 1- and 2-neighborhoods are colored different shades of yellow .
( we only highlight neighborhoods that are used in the proof . )
vertical paths are denoted by black and/or blue .
stage c.1 of the algorithm keeps only the vertical path with the blue bridges , and stage c.2 keeps only the bridges from this path .
there is no room for the dashed line to create a bridge in the upper stripe that would result in an error of the form depicted in figure [ f : total - order](b.2).,width=302 ] the only possible configuration for fig .
[ f : total - order](b.2 ) is more carefully illustrated in fig .
[ f : bridges - alternate ] .
the fact that this is the only possible configuration follows from first considering figure [ f : total - order](b.2 ) and drawing in the two v - paths that were deleted in stage c.1 . because the configuration in figure [ f : total - order](a.2 ) ca nt happen
, we must draw figure [ f : bridges - alternate ] so that the bridges connect as shown .
the v - path furthest to the right can not go any lower without creating a new , further right , bridge .
the next furthest v - path must lie along the middle h - path , because if it went any lower , then rhwf would have found that lower path when identifying the middle h - path .
( this is nt an issue for the furthest right v - path because it enters the 2-neighborhood of the lowest h - path . )
the last existing v - path can not intersect the others , and so must remain at least as high as shown in figure [ f : bridges - alternate ] .
therefore this configuration is exhaustive .
we now show that the configuration in figure [ f : bridges - alternate ] leads to a contradiction , and so case ( b.2 ) can not occur . since in stage c.1 of the algorithm , we take every third path , the two black vertical paths necessarily exist and do nt contribute bridges . to obtain a diagram like that in fig .
[ f : total - order](b.2 ) , we would need a fourth vertical path , the assumed dashed blue line , that has the following property : its bridge in the lower stripe must lie to left of the lower blue bridge , while its bridge in the upper stripe must lie to the right of that same lower blue bridge . to form a bridge in the appropriate place in the upper stripe , this assumed fourth path would need to pass through the neighborhood of the middle h - path .
but since v - paths are nt allowed to cross , this neighborhood is blocked by one of the deleted ( black ) v - paths .
therefore , case ( b.2 ) can not occur .
figure [ f : total - order](c.1 ) breaks into the three non - trivial cases in figure [ f : three - cases ] .
they are enumerated by the length @xmath32 of the closure of the lower abutment for the solid blue bridge . for case
[ f : three - cases](a ) , where @xmath33 , there is a clear contradiction with the requirement of keeping every third v - path .
likewise for case [ f : three - cases](b ) , where @xmath34 . for case
[ f : three - cases](c ) , which cover all cases with @xmath35 , it becomes slightly more complicated .
the argument that this ca nt happen is very similar to that discussed above for figure [ f : total - order](b.2 ) .
the essential point is that keeping every third v - path does nt leave enough space for this kind of error .
thus , the case depicted in figure [ f : total - order](c.1 ) does nt happen either .
( c.1 ) . in configuration ( a ) ,
where @xmath33 , there is a clear violation of the requirement that we have kept every third path . in ( b ) , where @xmath34 , regardless of which dashed path is chosen there is again not enough room for two more v - paths . finally ,
in case ( c ) , which covers all @xmath35 , we have the dreaded octopus configuration .
, width=302 ] finally , we have case ( c.2 ) . since we have kept only every third vertical path , neighborhoods of identified v - paths can not overlap .
for case ( c.2 ) to occur , the horizontal path would have to enter the neighborhood of the dashed blue bridge , _ then leave it _ before entering the neighborhood of the solid blue bridge .
but in order to create an error like ( c.2 ) , it would have to _ re - enter _ the neighborhood of the dashed blue bridge .
this is a contradiction , since then the h - path would either have to cross itself ( eliminated in stage c.1 ) or a bridge ( not possible by the definition of a bridge ) .
thus we conclude that the closures of the abutments are indeed disjoint and taking alternating bridges gives a hexagonal lattice structure , up to errors localized to the abutments . in other words , the graph minor achieved by collapsing the region near the abutment
has a topological minor which is exactly a regular hexagonal lattice .
thus , if we can transform these regions to have exactly the topology of a single local vertex in a hexagonal lattice , then the quantum part of the algorithm will succeed .
now the only step left for the classical stage is to fix the local h - v errors in the region of the abutments . to do this , we can actually enumerate the possibilities .
first observe that all vertices in an abutment are degree @xmath9 , other vertices in an h - path are degree @xmath0 , and the endpoint ( or starting point ) of a bridge , @xmath36 ( or @xmath37 ) is either degree @xmath0 , @xmath9 , or @xmath38 . there is no lattice error if there is one vertex in the region near an abutment of degree @xmath9 and the rest are degree @xmath0 .
therefore if @xmath36 has degree @xmath0 there is no problem ; the only degree 3 vertex in this local region is the single vertex belonging to the abutment . in the other cases , where @xmath36 has degree 3 or 4
, we can eliminate the errors in the following manner .
first delete everything in the h - path that lies in the closure of the abutment _ except _ for the endpoints of the abutment .
then , by adding the endpoint of the bridge to the remaining vertices , we once again have a valid h - path , which additionally has no h - v errors , since the only degree 3 vertex is now the endpoint of the bridge , which is now considered as part of the h - path .
although we have done this procedure for the endpoints , we can also do it for the starting points .
both of these points can be done separately , since bridges always have distinct starting and ending vertices as a consequence of using the 2-local rhwf .
one final subtle point is that , when the reconfiguration is done for both the start and end of the bridge , this might result in an h - h error .
because of the restricted nature of these h - h errors , these errors do not affect the global topology ; in fact , an example of this subtlety is present in fig .
[ f : algorithm ] .
the quantum part of the algorithm is straightforward compared to the classical part . as stated before , we simply measure @xmath18 on vertices that were not identified in the classical part , thus disconnecting them
. then we measure @xmath23 on the remaining degree 1 or 2 vertices , which will topologically contract the lattice .
the only delicate part is the boundary , where we should leave one additional vertex present between each bridge in the h - paths , so that we have a true hexagonal lattice .
a valid vertex is always available because we take every third v - path .
[ [ the - existance - of - mathcal - ol - vertex - disjoint - crossingssvertexdisjoint ] ] the existance of @xmath39 vertex - disjoint crossings[s : vertexdisjoint ] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ we argue that we can expect @xmath39 vertex - disjoint crossings in both the horizontal and vertical directions .
let us consider the event ( increasing with respect to @xmath3 ) that there exists a path between the one side and the opposite side in the 2d square grid .
denote by @xmath40 the maximal number of vertex - disjoint paths crossing the 2d square grid of size @xmath41 from one side to the other . according to a basic result in independent percolation , in the supercritical phase @xmath42 , the event such that at least a single crossing path exists happens with probability @xmath43 , for some @xmath44 , that is , almost surely in the thermodynamic limit .
the existence of a crossing is stable with respect to changes in any @xmath45 or fewer sites if there exists at least @xmath46 vertex - disjoint crossings . in order to show that the stability is @xmath47 with probability close to one in the thermodynamic limit
, we utilize an argument for the stability of an event from section 2.6 of ref .
@xcite , in which the following inequality is proved in the bond percolation setting , but is easily adapted to site percolation , as discussed in section 5 of ref .
we first write @xmath48 with some spatial overhead @xmath49
. then the existence of @xmath46 vertex - disjoint crossings is still expected with probability close to one given by @xmath50 as long as the reduced exponent @xmath51 is positive and @xmath52 .
namely , for a given @xmath3 in the supercritical phase , we can almost surely expect @xmath40 is @xmath39 with a spatial overhead depending only on @xmath3 , since @xmath53 can be chosen optimally over the range of @xmath54 . in this section
we show that each step in the algorithm runs in time at most @xmath4 .
first , rhwf runs @xmath24 times to find @xmath24 paths .
each time it finds a path , it will visit any given vertex at most @xmath38 times , since this is the maximum degree of a vertex .
furthermore , each time rhwf is called , it can only visit vertices that are distinct from those visited by any previous call .
since the most number of vertices possible is @xmath6 , all of the rhwf calls will visit no more than @xmath55 vertices , which is @xmath4 .
the bridge and alternating bridge decompositions again visit at most @xmath4 vertices , this time only once , since there is no backtracking .
reconfiguring the regions near the abutments requires only @xmath4 steps as well , since it only checks the degrees of the bridge start and end points .
thus , the entire classical part of the algorithm has run time @xmath4 .
the quantum part only requires @xmath4 measurements , so the @xmath4 scaling holds for the entire algorithm . | we study how heralded qubit losses during the preparation of a two - dimensional cluster state , a universal resource state for one - way quantum computation , affect its computational power . above the percolation threshold we present a polynomial - time algorithm that concentrates a universal cluster state , using resources that scale optimally in the size of the original lattice . on the other hand , below the percolation threshold
, we show that single qubit measurements on the faulty lattice can be efficiently simulated classically .
we observe a phase transition at the threshold when the amount of entanglement in the faulty lattice directly relevant to the computational power changes exponentially . |
the perturbation to the luminosity distance @xmath0 caused by cosmological structure is of interest for two reasons ; on one hand it can degrade the precision of cosmological parameters inferred from the hubble diagram obtained from standard candles such as sn1a ( e.g. riess et al . 1998 ; perlmutter et al .
1999 ) while on the other it can provide a useful probe of structure . at very low redshift
the distance is , in the absence of structure , simply linearly proportional to the redshift and the dominant effect of density perturbations comes from the peculiar velocities associated with the growth of structure .
these velocities cause a change to the redshift through the first order non - relativistic doppler effect with the result that that if a source has a positive radial peculiar velocity @xmath4 then its distance is lower than for an object of the same redshift in an unperturbed universe ; i , e . it has a peculiar displacement @xmath5 and a corresponding fractional apparent luminosity enhancement of @xmath6 . exploiting these measurable perturbations to the luminosity as a function of redshift as a probe of structure has a long and rich history . in the past
these have been limited to very low redshift and in such studies it is legitimate to adopt a newtonian analysis and consider only the effect of the motions the effects here being the same for motions associated with the growth of structure or for hypothetical motions of sources relative to the cosmic frame that are not associated with density perturbations ( we shall refer to such motions as `` unsupported '' ) .
but to understand the effect on supernova cosmology , or to exploit motions as a probe of structure using data from deeper future surveys , a more detailed treatment is needed that allows for the non - negligible redshift of the sources . for one thing
, one can not simply apply the non - relativistic velocity addition formula to relate velocity and distance perturbations . and for finite redshifts other effects come into play .
these include the gravitational redshift ( sachs & wolfe 1967 ) ; the effect of aberration which modifies both the angular size and also the surface brightness of sources ; and the effect of weak gravitational lensing , an effect distinct from aberration , which modifies sizes of galaxies leaving their surface brightness unaffected . the purpose of this paper is to elucidate some features of the observable effects of peculiar velocities at non - negligible redshifts .
the structure of the paper is as follows : in the following section ( [ sec : review ] ) we provide a review of previous work .
we briefly describe the history of low - redshift peculiar velocity observational studies ( [ sec : review - standard - candles ] ) and we also contrast these with other probes of cosmic motions such as the kinematic sz effect ( [ sec : review - ksz ] ) ; the observer motion induced dipole anisotropy of isotropic background radiation fields and the dipole in the source counts of very distant galaxies ( [ sec : review - dipoles ] ) ; the effect of motions at low - redshift on structures in redshift space ( [ sec : review - zspacedistortion ] ) sometimes called the ` rocket - effect ' ; and the weak lensing effect ( [ sec : review - gravlensing ] ) . in
[ sec : review - theory ] we review recent theoretical developments .
these are mostly based on the pioneering work of sasaki and co - workers who , starting in 1987 , provided a framework for describing the effects of linear density perturbations on the luminosity distance that includes the effect of peculiar velocities , gravitational lensing and the gravitational redshift ( including the integrated sachs - wolfe effect ) .
many of the more recent studies isolated the terms that depend only on the peculiar motions and have applied the resulting simplified formulae to supernovae cosmology and to make forecasts for how future surveys can provide constraints on large - scale structure ( coining the terms ` doppler - lensing ' and ` anti - lensing ' to describe these effects , though , as we note , these are not particularly novel probes ) .
we show , however , that the modifications to the normal newtonian analysis that become non - negligible at finite redshift have a puzzling feature in that they appear to depend on the absolute motion of the sources relative to the cosmic frame .
in [ sec : deltamofz ] we address the apparent violation of the equivalence principle in the above studies .
we show how these apparent violations are banished once one allows for the decay of the peculiar velocity in the case of unsupported motions and , in the case of motions associated with structure , once one includes the sachs - wolfe effect .
in [ sec : deltazofm ] we consider a novel proposal to measure peculiar velocities which is to try to measure the perturbation to the mean redshift as a function of flux density ( rather than the perturbation to the flux - density as a function of redshift ) .
we show that there is a non - trivial observable influence from the modulation of the surface brightness in this method .
but we argue that there is a problem with this method in that , for most scales of interest , the observable is swamped by the effect of galaxy clustering .
we then conclude with a discussion ( [ sec : discussion ] ) including a critical analysis of the prospects for extending peculiar velocity studies to larger scales with future surveys , and in appendix [ sec : appendix ] we show how measurements of large - scale motions are biased by smaller scale structure .
exploiting @xmath7 as a probe of structure so called ` peculiar velocity ' or ` cosmic - flow ' studies has a long and rich history .
rubin et al ( 1976 ) found a dipole moment of flux densities of galaxies with @xmath8 in the range 3,500 - 6,500 km / s , indicating a motion of the sun with respect to this shell at a speed of @xmath9km / s and also a substantial motion of galaxies in this shell with respect to the cosmic microwave background frame .
this was a most surprising and unexpected result since , outside of the local group , the hubble flow appeared to be remarkably cold and close to linear . in a similar study , but with a more careful treatment of biases from selection effects , tammann , yahil & sandage ( 1979 ) compared magnitudes of galaxies in the virgo cluster with those of galaxies at the same redshift scattered around the sky to provide an impressively precise estimate of peculiar motion associated with the growth of the local super - cluster . about the same time
it was realised that using hi velocity widths for spirals ( tully & fisher , 1977 ; hereafter tf ) gave distances with @xmath10 20% fractional distance uncertainty .
central velocity dispersions for ellipticals ( faber & jackson , 1976 ) provided somewhat less precise distances but the discovery of the ` fundamental plane ' in surface brightness , velocity dispersion and magnitude ( djorgovski and davis , 1987 ; dressler et al 1987 ) allowed the use of the @xmath11 relation which uses velocity dispersion and surface brightness to predict the proper size of the galaxy and which provides distance precision similar to the tf relation .
further peculiar motion data have been obtained from sn1a ( e.g. turnbull et al .
2012 ) , which have much higher precision of @xmath12% fractional distance error per object , but which are , as yet , a sparser sample of the velocity field than tf or fp samples .
other methods include using the tip of the red giant branch ( trgb ) as a standard candle ( lee , freedman & madore , 1993 ) and surface - brightness fluctuations for early type galaxies which were pioneered by tonry et al .
( 2000 ) and show generally good agreement with fp distances ( blakeslee et al . 2002 ) .
all of these techniques are , one way or another , directly measuring the perturbation to the luminosity distance caused by peculiar motions .
major reviews of the status of peculiar velocity studies in the mid 90s were provided by dekel ( 1994 ) and by strauss and willick ( 1995 ) and a good snapshot of the state of the subject at the turn of the century is given by hudson ( 1999 ) and by courteau and dekel ( 2001 ) , highlighting the results presented at the famous conference in victoria , b.c . in 1999 .
after something of a lull , the subject is now undergoing a revival , with major advances in the data available , including the ` cosmicflows-2 ' compilation of @xmath138,000 mostly tf distances ( tully et al 2013 ; incorporating the sfi++ tf catalog of springob et al .
2007 as well as snia distances ) and with significant results coming from the 6dfgs survey with an impressive @xmath1310,000 fundamental plane distances ( springob et al 2013 ) .
in addition , there has been something of a revival in using the individually less precise galaxy flux densities as distance estimators ( nusser , branchini and davis 2011 ; branchini , davis and nusser 2012 ) exploiting the larger sample size and excellent photometric accuracy of the 2mass redshift survey ( huchra , 2012 ) .
nusser , branchini and feix ( 2013 ) have discussed how this might be extended to massive photometric surveys ( @xmath14 square degrees with 10-@xmath15 multi - passband optical through near - ir photometry for galaxies to @xmath16 ) that will be provided by the euclid satellite along with the anticipated ground - based support photometry ( laureijs et al .
2011 ) .
there are other potential probes of velocity that are related to the above but are not exactly equivalent to measurement of the luminosity distance .
one is the kinematic sz effect on the cosmic microwave background ( cmb ) which has been attempted with clusters of galaxies with wmap by kashlinsky et al .
2010 and kashlinsky , atrio - barandela , & ebeling 2011 and with planck ( planck collaboration et al .
2014 ) and with galaxies ( lavaux , afshordi , & hudson , 2013 ) .
this method has the distinguishing characteristic that the observable is largely independent of distance unlike standard candle techniques that have poor sensitivity at large distance and should in principle provide the most powerful constraints on peculiar velocities at large distances .
this method is also quite different from the other probes considered here in that it provides a remote measurement of the cmb dipole as would have been seen by observers in the clusters , providing essentially unique constraints on departures from homogeneity as opposed to more easily measured departures from isotropy of our universe .
another technique is to exploit the dipole induced in otherwise isotropic radiation backgrounds or distant source counts by our motion .
this was originally applied to cosmic rays by compton & getting ( 1935 ) but is best known , and most accurately measured , from the dipole of the cosmic microwave background ( e.g. hinshaw et al .
2009 ) which indicates a solar motion of 369 km / s .
the dipole moment of the x - ray background has been measured by plionis and georgantopoulos ( 1999 ) at 1.5kev and by scharf et al .
( 2000 ) in the 2 - 10kev energy range , both of whom find results broadly compatible with the cmb motion .
lahav , piran and treyer ( 1997 ) have shown that in these bands the effects of clustering of sources and that induced by our motion are expected to be comparable .
a closely analogous effect is the dipole of counts of discrete sources .
ellis & baldwin ( 1984 ) showed that the perturbation to the flux density of a distant source with spectral index @xmath17 caused by our motion is @xmath18 with @xmath19 the direction to the source .
so , for unresolved sources with @xmath20 , and allowing for aberration , the amplitude of the dipole of the counts is @xmath21 .
baleisis et al . (
1998 ) have attempted to measure this with the green bank ( 87 gb ) ( gregory & condon 1991 ) and parkes - mit - nrao ( pmn ) catalogues ( griffith & wright 1993 ) , but with limited success ; they attribute the very large apparent signal to calibration errors .
the dipole has been determined in the nvss survey ( condon et al .
1998 ) by blake & wall ( 2002 ) and more recently by singal ( 2011 ) and rubart & schwarz ( 2013 ) , the last including sources from the westerbork wenss survey ( rengelink et al .
these studies have found generally quite good agreement of the dipole direction with that of the cmb dipole , but , puzzlingly , finding something like 2 - 4 times the expected amplitude .
gibelyou & huterer ( 2012 ) have measured dipoles of various galaxy redshift surveys and also gamma - ray bursts as well as the nvss radio sources ; they find only the latter to be discrepant with theoretical expectations and suggest the nvss dipole may be corrupted by systematic errors .
mertens et al .
2013 have suggested that the milky way motion would induce a dipole signal in the magnification measured in large - area weak lensing surveys such as lsst or des .
but the quantity that they consider is the perturbation to the angular size of objects at a constant distance , which is not directly comparable to what is actually observed .
the kinematic sz and source count dipole effects are physically distinct from methods that measure the perturbation to the luminosity distance using flux densities of sources or features in the distribution of luminosities like the ` knee ' in the galaxy luminosity function that can be considered to be ` standard candles ' .
these different methods are nonetheless supposed to be measuring the same physical quantity .
another manifestation of the perturbation to the distance is the distortion of the appearance of cosmological structure in redshift space ( kaiser 1987 ) .
there are two aspects to this .
the most well known , and most highly exploited , is the distortion of the correlation function ; a combination of smearing along the line of sight from small scale motions and a coherent squashing associated with large scale motions ( see e.g. hamilton 1997 ; percival et al .
2011 for reviews ) .
a separate effect , less widely considered , but more closely related to the subject here , is the distortion of the density contrast @xmath22 inferred from the density of galaxies as this is estimated by dividing the counts in a region by the selection function @xmath23 ; this is a function of the distance but is usually evaluated in redshift space , resulting is a biased picture of local structure
. sometimes referred to as the ` rocket - effect ' , this effect , like peculiar velocity measurements , is most important at low redshifts . in one sense one can consider this to be another probe of peculiar motions , but the difficulty is that the velocity induced density perturbation is superposed on the real - space clustering pattern , and decoupling these is not easy .
the distance measurement methods as described above , in contrast , provide distances , and hence peculiar velocities , that are not biased or otherwise affected by perturbations to the galaxy density .
such methods give a clean determination of the velocity , though subject to the assumption that the the luminosity function of the sources being used is universal ( i.e. unperturbed by local density or other environmental influences ) .
this decoupling of density and velocity is achieved at the cost of requiring that the estimator , in the case of galaxy - based measurements , may only use information contained in the _ shape _ of the flux density ( or angular size ) distribution of those objects that survive the selection criteria ; they may not use the actual number of detected objects . in the local universe the effect of velocities on the galaxy density field is not so much a useful probe of structure as an impediment to attempts to determine the local peculiar gravity field for comparison with the velocity field with the aim of determining the growth rate of structure ( nusser , davis & branchini 2014 ) .
the fractional perturbation to the luminosity distance from peculiar motions falls rapidly with redshift , but at larger redshift another effect comes into play which is the modulation of sizes of objects by gravitational lensing .
this was first considered theoretically by zeldovich ( 1964 ) and by gunn ( 1967a ; citing unpublished work of richard feynman presented at a caltech colloquium in 1964 ) .
the essential question is : given a space - time with inhomogeneity and metric fluctuations , how is the angular size of distant objects modified by the structures along the line of sight .
this can be obtained by propagating a narrow bundle of rays back from the observer to a distant source plane at some specified distance , which provides the 2x2 matrix giving the mapping from source plane positions to sky - plane angles whose determinant is the amplification ( the off - diagonal and asymmetric parts of the diagonal components being the shear and the mean of the diagonal terms being conventionally written as @xmath24 with @xmath25 being the convergence ) . at linear order this may be calculated in the born approximation ; the first order relative deviation of a pair of null rays is calculated using tides evaluated along the unperturbed paths . the above papers focused on the fluctuations in the convergence caused by structures along the line of sight .
also of interest is the mean amplification which will be non - zero if the paths to observed objects tend to avoid foreground objects .
this was considered by gunn ( 1967b ) and by dyer and roeder ( 1972 , 1974 ) .
the lensing effect is a cumulative one ; the amplification of a background object depends on an integral of the transverse tidal field along the line of sight .
consequently the convergence changes quite slowly with distance .
liouville s theorem tells us that the surface brightness depends only on the redshift , so the fractional perturbation to the luminosity is , at linear order , just the fractional change in the solid angle , and therefore the luminosity distance perturbation is related to the convergence by @xmath26 .
the early theoretical papers cited above considered universes containing point masses or used ` swiss - cheese ' models . a significant advance in understanding of the magnitude - redshift relation in more realistic perturbed flrw models with metric fluctuations and motions given according to the prediction for the spectrum of fluctuations in @xmath27cdm or similar models was provided by sasaki ( 1987 ) and elucidated by futamase and sasaki ( 1989 ) and by sugiura , sugiyama and sasaki ( 1999 ; hereafter s@xmath2899 ) .
these papers provided a unified treatment valid at both low and high redshift and revealed clearly how the lensing and peculiar velocity effects come to dominate at high and low redshifts respectively ( they also included other , typically subdominant , effects such as the integrated sachs - wolfe ( 1967 ) effect ) . ignoring the effects of the peculiar gravitational potential , one finds from s@xmath2899 s equation 3.15 that the effect of peculiar velocities of sources ` s ' and the observer ` o ' induce a fractional change in luminosity distance ( which is to say minus half the fractional change in flux density at constant observed redshift ) @xmath29 here @xmath30 is the scale factor , @xmath31 is conformal distance , prime denotes derivative with respect to conformal time and @xmath19 is a unit vector from the observer along the line of sight back to the source .
velocities here and below are in units such that the speed of light @xmath32 . at low redshift ,
@xmath33 so the first term in ( [ eq : deltadl ] ) is dominant and is simply the peculiar displacement @xmath5 .
the last term can be safely ignored for all of the observations described above , but becomes significant at finite redshift .
interestingly , and indeed rather puzzlingly , this term depends only on the motion of the source , unlike the leading order low - redshift effect that depends on the relative peculiar motion of the source with respect to that of the observer . thus ( [ eq : deltadl ] ) might appear to say that in a situation where the observer and the sources share the same common motion , perhaps because they have been accelerated by some very distant mass excess , there would be an observable dipole perturbation to the luminosity distance at constant @xmath34 from which we can infer our motion relative to the ` cosmic frame ' .
but that would violate the equivalence principle ( ep ) which dictates that for sources and observer in free fall the only locally observable effect of distant masses is the tidal distortion , which would show up as a quadrupole
. the velocity of the observed region as a whole with respect to the state of motion of more distant matter can not be determined by local measurements . yet for @xmath35 equation ( [ eq : deltadl ] ) seems to provide an absolute cosmic speedometer .
subsequently , many papers have repeated this kind of analysis .
pyne & birkinshaw ( 2004 ) also computed @xmath7 in perturbed flrw models .
they focussed mostly on gravitational lensing effects at high @xmath34 but the velocity dependent terms in e.g. their equations 32 & 23 are equivalent to ( [ eq : deltadl ] ) .
their analysis makes clear that ( [ eq : deltadl ] ) does not rely on the assumption that the velocities are purely growing mode linear perturbations .
hui & greene ( 2006 ) performed a similar perturbation analysis , and they also showed how ( [ eq : deltadl ] ) can be understood heuristically in the case of an unperturbed frw model in which sources and the observer have peculiar velocities ( i.e. what we are calling ` unsupported ' motions ) .
they argue that on all relevant scales equation ( [ eq : deltadl ] ) , augmented by the usual gravitational lensing effect , provides an adequate description .
hui & greene ( 2006 ) , cooray & caldwell ( 2006 ) and davis et al .
( 2011 ) have all used ( [ eq : deltadl ] ) to calculate the impact of peculiar velocities on the precision of sn1a cosmology ; these papers calculate the extra contribution to the error covariance matrix over and above the diagonal contribution from uncorrelated measurement errors and the corresponding increase in the uncertainty in the derived dark energy properties .
hui & greene make the important point that even small velocities can have a big impact because the errors they induce are strongly spatially correlated .
since the primary goal of sn1a is to determine the dark energy ` equation of state ' , what is relevant is the monopole of the distance perturbation . for an isotropic uniformly sampled full sky survey
the observer motion would be irrelevant as it generates only a dipole .
but for realistic current or near term future surveys the sky coverage is strongly non - uniform and so there is coupling of the dipole and other low order moments into the monopole .
so for real surveys , the motion of the observer which couples to all of the errors at some level is the most critical .
yet all of the papers cited above only compute the extra covariance arising from source motions .
the solution , as suggested by hui and greene , is to calculate the correlated motions of sn hosts , conditional on the motion of the local group motion with respect to the cmb .
they find that this makes little difference for most current sn1a surveys , but may be important for measurements that use lots of low-@xmath34 sn to pin down the current expansion rate more accurately .
bonvin , durrer and gasparini ( 2006 ) also calculated @xmath7 in linearly perturbed frw models , again reproducing the dependence on velocities of s@xmath2899 above ( see their equation 59 , and noting that in their notation @xmath36 is the same as @xmath37 in ( [ eq : deltadl ] ) ) .
they suggest that the power spectrum of @xmath7 provides a new observational tool which can be used to determine cosmological parameters .
second order perturbations to the distance have been considered by umeh , clarkson & maartens ( 2012 , 2014 ) and by marozzi ( 2015 ) .
bonvin ( 2008 ) calculated the effect of peculiar velocities on weak gravitational lensing convergence in perturbed frw models .
the version of the paper in the journal has an expression for the velocity induced lensing convergence that , unlike ( [ eq : deltadl ] ) , appears to depend only on the relative source and observer velocities , but this has been emended in a revised version on the arxiv . following this , boleko et al .
2013 have noted that the standard lensing convergence effect ( that produced by the transverse deflection of rays ) `` is swamped at low redshifts by a relativistic doppler term that is typically neglected '' . they have dubbed this ` anti - lensing ' .
the same group ( bacon et al .
2014 ) have followed this with a more detailed analysis estimating how well this should be measurable with deeper future redshift surveys and have coined the phrase ` doppler - lensing ' to describe the effect of peculiar velocities .
there is clearly a bit of a disconnect between two disparate communities here .
the cosmological perturbation theorists are evidently unaware that the probes of cosmology that they are proposing have a long and productive history of practical application going back about 4 decades as described in [ sec : review - standard - candles ] above .
the use of power - spectrum or auto - correlation analysis was also well developed early on ( gorski , 1988 ; kaiser , 1988 ; gorski et al .
1989 ; groth , juszkiewicz & ostriker 1989 ) .
somehow the equivalence of the physical quantities being discussed by the different communities here has been lost somewhere along the line .
the confusion may have been exacerbated by the description of the velocity induced effects as ` weak lensing convergence ' .
there is no lensing here , at least in the commonly used sense of deflection of light rays by cosmological structure ( there is a deflection of rays caused by aberration associated with motion of the observer but that is largely ignored in the papers above ) , and therefore no convergence of rays ; all that is happening is that velocities perturb the relation between observed redshift and angular diameter distance .
as already mentioned , the asymptotic @xmath38 behaviour of equation ( [ eq : deltadl ] ) can be easily understood without recourse to relativity or cosmological perturbation theory ; if a source has a peculiar velocity ( relative to any peculiar velocity of the observer ) then it has a peculiar displacement , in redshift space , that is simply @xmath39 . for most of the measurements described above
, this simple notion provides a reasonable model for interpreting the observations .
what is interesting about the formalism of sasaki and colleagues ( and the more recent theoretical papers cited above ) is not just that it provides a nice unified treatment of peculiar motions and gravitational lensing , but that it should be valid to describe the effect of peculiar velocities at all redshifts , not just in the limit @xmath40 .
future redshift surveys will probe to larger redshifts and at even at @xmath41 the next order correction to the dominant @xmath42 term a contribution to @xmath43 which is independent of redshift is a 10% effect .
similarly , it would seem prudent to get this right when computing the impact of flows on scales of a few hundred mpc on sn1a cosmology , given how much is riding on the result .
but , as we have noted , the formalism used exhibits seemingly unphysical behaviour . in the following section
we show that in the case of ` unsupported ' motions that is to say motions of observers or sources considered as massless test particles in an unperturbed cosmology the resolution of the puzzle of the apparent violation of the ep is that the velocities are decaying , so a spatially uniform flow at a given instant of cosmic time will appear non - uniform on the past light cone of the observer , and it is light - cone variables that appear in the formula ( [ eq : deltadl ] ) .
the finite redshift term can be thought of as the correction so that the total effect is determined by the relative velocities at constant cosmic time .
allowing for this restores the symmetry between source and observer motions . for finite matter density and velocities associated with growth of structure
the situation is a little more complicated , and the resolution is that the ep violating dipole from the velocity field alone is cancelled in a consistent treatment by the gravitational redshift effect . in the process
we obtain a somewhat improved expression for the perturbation to the luminosity distance that properly accounts for the motion of the observer .
this can be used to better predict the effect of structure on the precision of the sn1a cosmological parameter estimation .
we then turn attention to the prospects for measuring peculiar velocities from the perturbation to the redshift as a function of flux density , as proposed by nusser , branchini & feix ( 2013 ) .
we show that this has the expected asymptotic behaviour at very low redshift that @xmath44 but that , as with @xmath7 , there are finite - redshift modifications ; here caused in part by relativistic beaming effect modifying the surface brightness .
but the challenge with this method is that the proposed measurable is swamped for most scales of interest by fluctuations in the mean redshift caused by large - scale structure .
in the discussion section ( [ sec : discussion ] ) we consider the extension of peculiar velocity studies to larger scales with future redshift surveys . in appendix
[ sec : appendix ] we present a linearised analysis of the bias in the velocity field measurement caused by small scale motions or measurement errors and we show that this accounts nicely for the damping found empirically in the numerical studies of koda et al .
first let s recall the origin of ( [ eq : deltadl ] ) .
hui and greene ( 2006 ) have given a very nice heuristic derivation of this .
they compute the effect of source and observer velocities in a homogeneous frw model ; i.e. they consider the effect of what we are calling ` unsupported ' motions .
they calculate the perturbation to the object flux density and to the redshift at fixed distance , and then combine these to obtain the perturbation to @xmath45 at fixed observed redshift .
this is legitimate , but is somewhat roundabout as the perturbation to the flux densities involves consideration of relativistic beaming in which the redshift perturbation modulates the surface brightness of the sources ( see e.g. the discussion in davis et al .
a more direct approach is simply to compute the perturbation to the angular size and redshift to obtain the ` convergence ' or equivalently minus @xmath7 .
consider , for simplicity , a spatially flat background cosmology with metric @xmath46 where the scale factor is dimensionless and is equal to unity at present : @xmath47 and with an observer at the origin of the spatial coordinate system @xmath48 . in the absence of source or observer
motions the redshift of a source at conformal distance @xmath49 is @xmath50 and the distance is a function of redshift : @xmath51 .
the angular diameter of a small spherical standard source of physical diameter @xmath52 at this distance is @xmath53 , which varies with redshift as @xmath54 as is well known , this vanishes ( at about @xmath55 in the standard @xmath27cdm model ) where @xmath56 is stationary owing to the focussing of rays by the smoothly distributed matter . an observer moving with peculiar velocity @xmath57 relative to a local fundamental observer would perceive a standard source at distance @xmath49 to have angular size @xmath58 because of special relativistic aberration .
this is easily shown by applying a lorentz boost and linearising in the assumed small velocity @xmath57 .
there will also be a perturbation to the redshift @xmath59 where , in the present circumstances , @xmath60 is the fractional perturbation to the wavelength caused by the 1st order doppler effect of peculiar velocities of the source and observer .
the quantity of interest here is how the perturbed angular size ( [ eq : theta ] ) at this perturbed redshift differs from the angular size for a source at redshift @xmath34 in the fictitious case that there is no source or observer motion .
working to 1st order precision this is @xmath61 so , subtracting ( [ eq : theta0 ] ) from ( [ eq : theta ] ) , and dividing by @xmath62 gives the first - order fractional perturbation to the angle at this redshift @xmath34 from peculiar motions as @xmath63 this is ( minus ) the perturbation to the angular diameter distance , but as this is at constant redshift this is the same as minus the perturbation to the luminosity distance , as the surface brightness depends only on the redshift . in the situation described here of unsupported peculiar motions in a homogeneous background cosmology the perturbation to the redshift
is @xmath64 which in ( [ eq : deltatheta0 ] ) is equivalent to ( [ eq : deltadl ] ) .
more generally , the fractional wavelength perturbation @xmath60 in ( [ eq : deltatheta0 ] ) will also include contributions from the perturbation to the potential ; the peculiar gravity .
but note that for perturbations on scales much less than the hubble scale , which is what is relevant here as the perturbations to the velocity on the horizon scale is on the order of the primary cmb anisotropy or about @xmath65 , which is only a few km / s , the perturbation to the redshift is well described by newtonian theory , so the only significant ` relativistic ' effect here is the aberration caused by the observer motion . and it is this aberration or ` beaming ' that is the source of the asymmetry between observer and source velocities in ( [ eq : deltadl ] ) .
returning to the effect of unsupported peculiar motions , equation ( [ eq : deltatheta0 ] ) can be written as @xmath66 .
\label{eq : deltatheta1}\ ] ] this might appear , at face value , to have a zeroth order ( in redshift ) non - zero value in the case that the observer and source share the same motion .
but that is illusory .
the velocities appearing in ( [ eq : deltatheta1 ] ) are at different times , and in the situation described here the velocities are decaying as ( e.g. peebles 1980 ) @xmath67 , so @xmath68 , and @xmath69 \label{eq : deltatheta2}\ ] ] where the velocities are both at the time of observation @xmath70 . but
@xmath71 and consequently at low redshift the perturbation to the angular ( or luminosity ) distance depends only on the relative motion of the source and observer ( with corrections that are quadratic in the redshift ) : @xmath72 where , by ` @xmath10 ' , we mean equal up to fractional corrections of order @xmath73 .
thus the apparent ep violation in ( [ eq : deltadl ] ) appears because , in that equation , the source and observer velocities are not evaluated at the same time .
when they are , then to lowest order the ep violation disappears and the symmetry between observer and source motions is restored .
we are more interested in the case that the motions are associated with the growth of structure , in which case the velocities do not simply decay as @xmath74 , rather the equation of motion for the peculiar velocity is @xmath75 where @xmath76 is the gravity sourced in poisson s equation by the matter density perturbation @xmath77 ( peebles 1980 ) . consequently @xmath78 which
, in ( [ eq : deltatheta1 ] ) would seem to give a zeroth order ( in redshift ) contribution to @xmath79 of @xmath80 .
this is a dipole which , if observable , would allow one to determine the acceleration from distance structures from local observations , again in violation of the ep .
but if one is going to incorporate the peculiar gravity then , to be consistent , one should use , following s@xmath2899 , for @xmath60 in ( [ eq : deltatheta1 ] ) not just the doppler shift , but one should also include the gravitational redshift @xmath81 which , when multiplied by the large pre - factor @xmath82 gives an effect of the same order for structures on the scale of the source - observer separation .
we then have @xmath83 for the case of a distant attractor the gravity @xmath76 will be nearly spatially constant within the observed region and since @xmath84 the last two terms cancel and the gravity from distant matter is therefore unobservable from local measurements , again in accord with the ep .
in addition to being blind to super - survey size sources of gravity , expression ( [ eq : deltatheta4 ] ) has the desirable characteristic is that it is independent of the choice of ` background ' that we consider the actual universe to be a perturbation about .
for example , were one to consider the observable region to be a perturbation to a background with the `` wrong '' density and therefore the wrong value for the hubble parameter , say , this would change the peculiar velocities .
but the corresponding spatially constant density perturbation would produce a spatially constant tidal field which would compensate so that the physical observable @xmath7 is invariant . finally , in linear theory the gravity and velocity are proportional to one another , with @xmath85 , where the perturbation growth rate factor @xmath86 , and we can express the perturbation of the luminosity distance entirely in terms of velocities as @xmath87 where the velocities are all to be understood to be at the time of observation .
this formula ( [ eq : deltadd ] ) provides , at least in the context of linear theory , a physically consistent way to allow for the effect of peculiar motions on supernova cosmology .
given an assumed power spectrum for the density , and hence velocity , perturbations , it can be used as in the studies referenced above to compute the extra contribution to the covariance function .
also , given some estimate of the actual structure in the nearby universe , it can be used to correct the distances of sn for the associated motion , with the additional covariance from larger scale motions being calculated from the conditional probability distribution for the residual motions .
nusser , branchini and feix ( 2013 ) have suggested that useful measurements of large scale motions could be obtained using the photometric redshifts from euclid ( laureijs et al .
2011 ) which will be obtained from a combination of ground - based surveys ( providing at least three of the @xmath88 , @xmath89 , @xmath90 and @xmath34 bands ) and the euclid photometry in the broad vis band and in @xmath91 , @xmath92 & @xmath93 .
this is expected to provide @xmath94 photometric redshifts with better than @xmath13 5% precision in @xmath95 to mean redshift @xmath96 and covering 15,000 square degrees .
the ground based surveys will perform photometry with @xmath97-sigma detections limits deeper than 24th magnitude and may by themselves provide useful photometric redshifts in advance of euclid .
indeed some of these data in the southern sky are already being collected by the des dark energy survey which is planned to cover @xmath13 5,000 square degrees .
the nusser , branchini and feix ( 2013 ) proposal differs from conventional peculiar velocity measurements in that rather than the considering the perturbation to the distance , and therefore the flux densities of objects , at a given redshift they are proposing to measure the perturbation to the redshift , as a function of direction , relative to the mean redshift for sources of a given flux density . to this end
they define a redshift perturbation per galaxy @xmath98 where @xmath99 is the redshift of the @xmath90th galaxy and @xmath100 is the mean redshift for galaxies of apparent magnitude @xmath101 .
this , they argue is dominated , for sub - horizon scale structures , by the doppler shift , so that the average of @xmath102 over galaxies lying in some solid angle on the sky provides an estimate of an appropriately weighted average of @xmath103 along the line of sight .
what they propose is to use the angular power spectrum of this quantity to constrain the power spectrum of matter fluctuations .
the motivation for this is that they claim that this would provide measurements of large - scale motions with a precision corresponding to a fractional distance error of approximately 30% per galaxy from photometry and relatively low resolution spectroscopy .
this is not very much worse than the precision obtained from the considerably more expensive tf or @xmath11 techniques .
however , this technique differs from the normal method in several important respects : one is that the finite - redshift behaviour is somewhat different from that for the luminosity distance .
a second is that by using only the angular variation of the line - of - sight averaged peculiar velocity this method discards useful information about the variation of motions along the line of sight ( this is significant since it is the angular variations that are most susceptible to systematic errors in the photometric zero - point ) .
but the most important distinction is that the estimator will tend to be heavily influenced by large - scale structure in the galaxy spatial distribution .
in fact , as we will show , the signal from peculiar motions will be swamped by fluctuations from galaxy clustering .
the assumption underlying the idea that an average of ( [ eq : nbf ] ) provides an estimate of the average peculiar velocity is that galaxies in some direction selected by flux density will have a distribution of distances @xmath104 that is unbiased by peculiar motions .
but that is not the case .
large - scale motion will affect @xmath105 , the surface brightnesses of the galaxies , and therefore also their flux densities at a given distance .
for bolometric observations the effect is the usual @xmath106 dimming , so @xmath107 . for filter band - pass limited observations
there is also a linear relation , but the coefficient depends on the spectral energy distribution ( sed ) of the sources and on the filter transmission function . here , for simplicity , we will consider bolometric observations . if one considers a population of standard objects of identical proper size @xmath52 and rest - frame surface brightness @xmath108 then , in the absence of motions , the flux density from a source at distance @xmath49 is @xmath109,\ ] ]
where ` @xmath13 ' means ignoring geometric factors of order unity and where we have used the solid angle @xmath110 as appropriate for a spatially flat universe . in the presence of unsupported motions , the flux density for a standard source at the same distance
@xmath49 is @xmath111 with the first factor being the fractional change in solid angle from observer motion induced aberration and the second being the fractional change in the surface brightness . in general @xmath112 , rather requiring @xmath113 to be equal to the ` background ' flux density for some perturbed distance @xmath114 gives @xmath115
such a displacement in the background cosmology would correspond to a change in redshift @xmath116 . adding this to the first order doppler shift @xmath117 at
fixed @xmath49 gives the fractional perturbation to the observed wavelength for objects at fixed flux density as @xmath118 for @xmath38 the rhs becomes simply the sum of the source and observer doppler shifts ( recall that @xmath119 for low @xmath34 ) .
but for finite @xmath34 there are corrections coming from the relativistic beaming effects .
these reduce the observable effect relative to the naive prediction .
the numerator here is identical in its dependence on velocity to the fractional perturbation to the angular size of standard objects at constant redshift ( [ eq : deltatheta1 ] ) , so it can also be shown to be only dependent on relative velocities at constant proper time ( plus corrections of order @xmath73 ) for the case of unsupported motions and can be cast into a physically consistent form in the case of motions associated with growing structure with the inclusion of the sachs - wolfe term .
this is an idealised model .
the assumption of standard candles means that observations at some @xmath101 probe only a thin shell at the corresponding redshift . allowing a distribution function for the absolute magnitude and surface brightness results in a distribution @xmath120 but with the same general dependence on redshift . as mentioned
, the details also depend on galaxy seds and filter transmission functions .
so far this technique seems quite promising , particularly if the scatter in redshift is as small as the @xmath1330% claimed .
unfortunately , however , the estimator ( [ eq : nbf ] ) for motions also differs from conventional luminosity distance estimators in that it is not unbiased by the density of galaxies .
this has negative consequences . in a universe with no structure , but with moving sources and observers , the average of ( [ eq : nbf ] ) would provide a probe of motions with the relatively good precision indicated from the width of @xmath121 .
but in reality , even in the absence of peculiar velocities , there will be additional fluctuations in the mean redshift along each line of sight caused by density perturbations .
if one were to compute the average of @xmath102 for all of the galaxies in a cone of small solid angle @xmath122 and in some modest range of magnitudes @xmath123 around @xmath101 as @xmath124 then the cosmological signal would be essentially just the mean of the line of sight source velocity weighted by the density of galaxies ( and the finite redshift terms ) .
but with large scale structure there will be an additional contribution @xmath125 the integral here will , on average , be zero , but there will be fluctuations arising from the fluctuations in the number density of galaxies @xmath126 . in a simple model of ` blobs ' with some characteristic scale @xmath127 and rms density contrast @xmath128 then the rms contribution to @xmath129 from structure will be equal to @xmath130 times some factor of order unity , where @xmath131 is the number of random structures along the line of sight .
the ` signal ' , on the other hand , is on the order @xmath132 .
so for structures much smaller than the depth of the survey which for deep surveys means much less than the hubble scale the signal will be swamped by the noise from large - scale structure .
of course the density of galaxies in redshift space is measurable , so one can try to correct for this bias . but
this leads one back to an estimator for the luminosity distance perturbation as a function of redshift where one has to infer the distance from the shape of the distribution of fluxes of objects above the flux limit ( perhaps sub - divided by distance independent properties ) , and the statistical precision will not be the same as that of ( [ eq : nbf ] ) .
to recapitulate the major results so far , in section [ sec : deltamofz ] we showed how the somewhat puzzling dependency of the finite - redshift corrections to the luminosity distance perturbations on the absolute motion of the source and observer is removed once allowance is made for the time evolution of the velocity field and for the gravitational redshift .
we note that sasaki and colleagues never separated the different contributions , so this does not cast any doubt on the validity of their results .
the only question is regarding the legitimacy of using the peculiar velocity sourced effects alone .
we obtained an expression for @xmath133 in terms of the peculiar velocity that is manifestly respectful of the equivalence principle and which provides a consistent way to allow for the motion of the source .
this provides an improved way to estimate of the impact of motions on sn1a cosmology precision .
but this is not a huge effect ; for local perturbations on scale of a few hundred mpc this is a 10% correction , and in the standard @xmath27cdm model there is not much effect from larger scales . nor is this a very big effect for peculiar velocity measurements ( or ` doppler - lensing ' ; we hope we have dispelled any notion that these are different ) if one is measuring the effect of perturbations on scales much smaller than the typical observer - source separation , but again ( [ eq : deltadd ] ) should provide a more accurate result and so should be used in forecasting the performance of future surveys . in section [ sec : deltazofm ] we considered the perturbation to the redshift for sources of a given flux density .
we showed that one needs here to consider the effect of the motions on the surface brightness which results in a non - trivial dependence of the effect on the depth of the survey .
but we also argued that the signal would , except for very large - scale modes , be swamped by galaxy clustering ( as this measure of velocity is not unbiased with respect to galaxy number density ) .
we turn now to the prospects for extending peculiar velocity studies to greater depth , with particular focus on the possibility of exploiting future photometric surveys , possibly augmented with low resolution spectroscopy to provide redshifts , but without the more expensive velocity width information required for tf or @xmath134 measurements . extending peculiar velocity studies to large distances
generally has a rather poor return on telescope time because the distance errors grow linearly with depth @xmath135 .
and surveys without velocity dispersions for early type galaxies or hi velocity widths for spirals have generally higher distance error per galaxy .
but there is still potentially useful information .
the number of regions of a given size of interest @xmath136 grows as @xmath137 so the net error on the variance of flows on scale @xmath136 does decrease ( but only as @xmath138 so the figure of merit ( fom ) scales linearly with depth or only as the cube - root of the number of galaxies surveyed ) .
the measurement of tammann , yahil and sandage ( 1979 ) using luminosities of galaxies as a distance indicator gave a measurement of the coherent peculiar motion on the @xmath139 scale of the local supercluster with @xmath13 100 km / s precision .
so scaling up to a survey volume with huge numbers of such structures tells us that obtaining significant measurements of the variance of flows on similar scales should be do - able . plus
, of course , in addition to making more precise estimates of the velocity power spectrum at fixed scale , deeper surveys are invaluable for measuring bulk motion on the scale of the entire survey volume
. it might be hard to justify mounting a survey on these grounds , but if they are carried out for other reasons then there is no question that they should yield information on peculiar velocities .
it is reasonable to ask whether the information thus gained will be competitive with or complementary to that garnered from redshift space distortion studies ( rsd ) , for which the fom scales as the volume . at the present , these two techniques are roughly equally statistically powerful for measurement of the growth rate parameter @xmath140 , despite there being many , many more galaxies in the surveys used for rsd ( @xmath141 ) as compared to the samples of currently @xmath13 a thousand distance measurements ( hudson & turnbull 2012 ; koda et al . 2013 ) .
the reason is that rsd measurements are an auto - correlation measurement and therefore suffer from cosmic variance whereas velocity based determinations exploit cross - correlation and are therefore free of cosmic variance .
future surveys such as taipan ( koda et al .
2013 ) and wallaby ( duffy et al .
2012 ) with tens of thousands of distances will yield significant improvement in peculiar velocity measurements ( see e.g. koda et al .
but ultimately in the future the balance will shift in favour of rsd because distance - based velocities scale poorly with depth as noted above ( kinematic sz measurements , however , do not suffer this penalty ) . a further argument for measuring peculiar velocities as well as rsd
is that both are affected by small scale motions , but in somewhat different ways . modelling rsd involves the pairwise velocity distribution as a function of projected separation which is difficult to predict .
it is dangerous to assume a dispersion that is independent of projected separation as in e.g. the landy , szalay & broadhurst ( 1998 ) method since it is expected that , qualitatively speaking , the dispersion at small projected separation is dominated by the ` 1-halo ' contribution which is heavily weighted to massive clusters while at larger separation one is dealing with pairs that are in separate virialised systems where the appropriate average is dominated by small systems .
one can try to estimate the pairwise velocity dispersion from group and cluster catalogs , but the result from e.g. the crook et al .
( 2007 ) group and cluster catalogue from 2mrs is small and seems to be at odds with what is needed to match the data .
this , of course , may be simply because this simple modelling does not include the relative motions of groups and clusters .
these small - scale peculiar motions also cause a bias in measurements of larger - scale flows .
peculiar velocity measurements suffer are well known to be susceptible to biases of various kinds ( see e.g. faber et al .
1994 ; strauss & willick 1995 ) that are often described as ` malmquist ' bias .
these biases depend on how the data are selected and analysed ( e.g. whether one regresses magnitude onto line - width or _ vice versa _ ) and on what quantity one tries to measure ( e.g. peculiar velocity as a function of distance _ vs. _ peculiar displacement as a function of redshift ; or in strauss and willick s terminology method i _ vs _ method ii ) .
peculiar velocity measurements may be also be biased by environmental ( i.e. density ) dependence of the luminosities of galaxies .
but in the context of measurements of large - scale linear structure these can in principle be measured by cross - correlating apparent peculiar velocity and density in fourier space .
the true peculiar velocity is 90 degrees out of phase with the density and so appears as an imaginary contribution to the cross - spectrum while linear environmental bias shows up in the real part .
see mcdonald 2009 for further discussion .
the most serious ` malmquist'-type effects can be avoided if one uses the flux densities , together possibly with auxiliary distance independent quantities such as velocity widths or dispersions , to estimate the distance to a collection of galaxies in a localised region of redshift space ( i.e. method ii ) .
for example , if there is no selection on velocity width , as is often a reasonable approximation , and if residuals in log velocity width are approximately gaussian , as is again often assumed , then an unbiased estimate of the peculiar displacement is obtained from the deviation of the mean log velocity width from that expected if the galaxies are at the distance indicated by their redshift .
this is known , for historical reasons , as the ` inverse ' method . more generally ,
if the joint distribution of absolute magnitude and velocity width , or other distance independent attributes , is known then one can obtain the likelihood of the distance under the assumption that all of the galaxies in that region lie at the same distance , and thereby obtain an unbiased maximum likelihood estimate for the distance .
these results rely critically on the assumption that the velocity field is locally ` cold ' , so the galaxies in question can be assumed to have a common distance .
motions associated with small and intermediate scale structure violate this assumption and this introduces a bias not dissimilar to that which effects rsd .
the effect is easy to understand ; if one is measuring the flow in a region of redshift space on the back of a supercluster where the density of galaxies will be falling with radius and if there is any dispersion in distance at a given redshift then in a region of redshift space there will tend to be more galaxies that have scattered away from us than have scattered towards us from behind . in the appendix 1 .
we show that in linear theory , one would expect this to cause the measured velocity in redshift space to be biased downwards by a factor @xmath142 . in principle
( though subject to having a viable model for the biasing of the space distribution of galaxies ) this velocity bias , and hence @xmath143 also , is measurable .
so peculiar velocity measurements provide independent constraints on the distribution of relative motions of galaxies that is the major nuisance factor in the interpretation of rsd .
another potentially useful difference between these probes is that rsd measures only that component of the large - scale velocity field that is associated with the large - scale density contrast . in conventional models for structure formation that is
all there is , but in principle one could imagine that the actual cause of peculiar motions is not given by the gravity predicted by the density contrast either because of some kind of exotic ` biasing ' ( e.g. large - scale astrophysical modulation of galaxy creation ) or some exotic source of gravity
. it would be nice to be able to test this aspect of the prediction of gravitational instability directly . turning to the precision that might be obtained from future surveys without velocity - width information , bacon et al .
( 2014 ) made forecasts for ` doppler - lensing ' experiments with des , euclid & ska .
the most conservative example they considered was the des photometric survey with spectroscopic follow - up to obtain redshift for those galaxies at @xmath144 .
they assumed a surface density of 0.7 objects per square arc - minute , and that distances could be obtained with fractional error per galaxy of 30% .
these numbers both seem a little optimistic .
this redshift range and surface density implies a space density of @xmath145 but to reach that density requires observations of galaxies about 4 magnitudes fainter than @xmath146 , or about 24th magnitude ( i.e. about as deep as the 10-@xmath15 photometric limits though only a sub - set selected by photometric redshift to lie at this low redshift would need to be targeted ) . to obtain spectra
this faint requires @xmath147 hour integrations on a 10 m telescope .
the surface density is well matched to e.g. deimos on keck - ii but the problem is that the field of view of this instrument is less than 100 square arc - minutes so to cover the an area the size of the des footprint would take @xmath148 hours , which is impractical .
the 30% fractional distance error estimate was based on the small scatter in sizes of objects in the hst cosmos survey found by schmidt et al .
( 2012 ) who devised a weak lensing convergence estimator based on the size and magnitude of these objects . but
scatter in observed size does not simply equate to the distance precision as there are efficiency factors arising from incompleteness ( schmidt et al .
2012 attempted to estimate these ) .
if one were simply to use the flux densities of all galaxies in a flux limited survey taken willy - nilly without any auxiliary information the distance error is very much higher ( and the figure of merit scales inversely as the square of this ) .
an interesting question is whether one can do significantly better using information such as size ( or surface - brightness ) , morphology and colour . in this vein tammann , yahil &
sandage divided their galaxies by morphologically determined ` luminosity classes ' and similarly nusser , branchini & davis ( 2011 ) sub - divided the 2mrs galaxies according to their classification as spirals or ellipticals .
these two sub - samples have similar shaped lfs but @xmath146 differs by about 0.4 magnitudes so combining velocity estimates obtained from these separately has slightly better performance than taking them together and ignoring the difference in their lfs . for elliptical galaxies it may be possible to improve the precision of photometric distances using other distance independent information .
kormendy ( 1977 ) showed that the size of elliptical galaxies is strongly correlated with their surface brightness , and applied this to measure distances to the virgo cluster .
the small scatter in this ` kormendy relation ' would indicate something like 30% fractional distance uncertainty .
adding colour information , de carvalho & djorgovski ( 1998 ) claimed distance precision of 25% and argued that it might be possible to get 15% distance uncertainty with more homogeneous data , which would be better even than the @xmath149 relation .
this is all very encouraging , but subsequent studies suggest that the promised performance from the ` photometric fundamental plane ' may be hard to achieve . graham ( 2002 ) showed that for early type galaxies the sersic index correlates with velocity dispersion giving another photometric method for distance estimation .
he fit for log effective radius as a function of surface brightness and sersic index .
however , he found fractional rms distance errors of @xmath10 38% ( e - galaxies only ) or 48% ( e + s0 ) .
this was using hst photometry at relatively low redshift , so that may be optimistic .
la barbera et al .
( 2005 ) performed the same kind of analysis for a pair of clusters at @xmath150 and obtained a tighter relation with @xmath151% fractional distance error .
huff and graves ( 2014 ) fit for a photometric fundamental plane for early type galaxies ( those with best - fit photo-@xmath34 template being passive ) in sdss using surface brightness and concentration index ( sdss did not measure sersic index ) .
they used this to measure weak lensing convergence but if used as a distance indicator would have a precision of approximately 40% .
these studies therefore find relations with larger scatter and therefore poorer distance estimate performance
than some of the earlier studies indicated might be achievable . and even these results are probably over - optimistic as most of them are affected by selection effects .
for example , the kormendy relation for ellipticals is that the surface brightness is a decreasing function of size .
there is no doubt that the absence of large galaxies with high surface brightness is real , but the absence of small galaxies of low surface brightness is in part at least a selection effect .
this selection bias acts to artificially tighten up the relations , inflating the apparent distance precision .
a more realistic estimate of the precision that may be obtainable requires analysis of surveys with well defined selection criteria .
smith , loveday and cross ( 2009 ) have measured the bivariate luminosity - surface brightness distribution in the ukids las .
they show that with a optical @xmath152 colour split both blue and red subsamples are quite well described by choloniewski models ( choloniewski , 1985 ) .
the red sub - sample has nearly luminosity independent surface brightness and has a faint end slope @xmath17 close to zero and much larger than the value of @xmath153 characteristic of the galaxy population as a whole .
the bigger @xmath17 the better the precision per galaxy , and @xmath154 gives approximately 50% error in distance per galaxy , so this kind of precision should be obtainable for the red , primarily early - type , sub - sample , and the above studies suggest that including morphological information should improve this . but
the downside is that a large @xmath17 means that one can not exploit the large numbers of sub-@xmath155 galaxies that are found for later types .
for @xmath154 the space density rapidly converges to @xmath156 , which for these objects is @xmath157 and one reaches a density of @xmath13 half of this by going about half a magnitude below @xmath146 , or to about @xmath158 , and this would be a more reasonable goal for spectroscopic follow - up with the 400 fibre 2df / aaomega spectrograph on the aat perhaps augmented by the 1000 fibre weave spectrograph on the wht if this goes ahead , but it would still be a massive effort .
if this were done , it would provide then @xmath13 50% distance errors ( rather than 30% ) for objects within @xmath159 but these would have a space density of only @xmath160 or a little over a factor 10 smaller than assumed by bacon et al .
so overall this results in a big decrease in the fom . to obtain high space density at @xmath150 as anticipated by bacon et al .
2014 would require advanced spectroscopes such as pfs or the maunakea spectroscopic explorer ( mse ) that combine very wide field with the reach of a 8 - 10 m scale telescope , but of course there would be competing demands for these , and the currently proposed surveys for e.g. pfs would not be suitable . in the nearer term it will be possible to try to do peculiar velocity measurements with photometric redshifts using des . but the challenge there is that with 5% errors in @xmath161 the effect of bias described in the appendix is expected to be strong even for very large - scale perturbations , and the signal on smaller scales will be corrupted .
in addition , there is the concern regarding the impact of catastrophic photometric redshift errors .
an alternative to des , though likely delivering data on a slightly longer time - scale , is the j - pas survey with 56 narrow band filters that , it is claimed , will give photometric redshifts with @xmath162% precision in @xmath161 to @xmath163 over 8,000 square degrees . in summary ,
the prospects for extension of peculiar velocity studies to larger scales using photometric surveys , perhaps augmented with low resolution spectroscopy , are interesting but will not deliver the kind of @xmath164% fractional distance precision that has been suggested in some previous studies
. the spatial sampling density will also be an issue , and it will be interesting to see what information can be gained from ` photometric - plane ' studies of the more numerous late - type galaxies . and without spectroscopy , the measurements will be limited to very large spatial scales . nonetheless , one can expect to obtain useful information as a spin - off from these surveys that are being carried out primarily for weak - lensing and for rsd / bao measurements .
but , more conventionally , we can expect to see good progress at lower redshift ( where the signal per volume element is much stronger ) with the taipan spectrograph that is being developed for the ukst schmidt telescope and which represents a significant advance on the current 6df instrument and which will provide fp distances as well as redshifts , and from the askap wallaby survey .
these will both extend the range of spatial scales that peculiar velocities probe and will also improve the signal to noise at lower redshift by increasing the density of sampling .
we are grateful to the referee for helpful suggestions .
mh acknowledges the support of nserc .
nk is grateful for the support of cifar and for the hospitality of the institute for advanced studies at durham .
bacon d. j. , andrianomena s. , clarkson c. , bolejko k. , maartens r. , 2014 , mnras , 443 , 1900 baleisis a. , lahav o. , loan a. j. , wall j. v. , 1998 , mnras , 297 , 545 blake c. , wall j. , 2002 , nature , 416 , 150 blakeslee j. p. , lucey j. r. , tonry j. l. , hudson m. j. , narayanan v. k. , barris b. j. , 2002 , mnras , 330 , 443 bolejko k. , clarkson c. , maartens r. , bacon d. , meures n. , beynon e. , 2013 , phrvl , 110 , 021302 bonvin c. , durrer r. , gasparini m. a. , 2006 , phrvd , 73 , 023523 bonvin c. , 2008 , phrvd , 78 , 123530 bonvin c. , hui l. , gaztaaga e. , 2014 , phrvd , 89 , 083535 branchini e. , davis m. , nusser a. , 2012 , mnras , 424 , 472 broadhurst t. j. , taylor a. n. , peacock j. a. , 1995 , apj , 438 , 49 choloniewski j. , 1985 , mnras , 214 , 197 compton a. h. , getting i. a. , 1935 , phrv , 47 , 817 condon j. j. , cotton w. d. , greisen e. w. , yin q. f. , perley r. a. , taylor g. b. , broderick j. j. , 1998 , aj , 115 , 1693 cooray a. , caldwell r. r. , 2006 , phrvd , 73 , 103002 courteau s. , dekel a. , 2001 , aspc , 245 , 584 crook a. c. , huchra j. p. , martimbeau n. , masters k. l. , jarrett t. , macri l. m. , 2007 , apj , 655 , 790 davis t. m. , et al . , 2011 , apj , 741 , 67 de carvalho r. r. , djorgovski s. , 1989 , apj , 341 , l37 dekel a. , 1994 , ara&a , 32 , 371 di dio e. , durrer r. , 2012 , phrvd , 86 , 023510 djorgovski s. , davis m. , 1987 , apj , 313 , 59 dressler a. , faber s. m. , burstein d. , davies r. l. , lynden - bell d. , terlevich r. j. , wegner g. , 1987 , apj , 313 , l37 duffy , a. r. , meyer , m. j. , staveley - smith , l. , bernyk , m. , croton , d. j. , koribalski , b. s. , gerstmann , d. , westerlund , s. , 2012 , mnras , 426 , 3385 dyer c. c. , roeder r. c. , 1972 , apj , 174 , l115 dyer c. c. , roeder r. c. , 1974 , apj , 189 , 167 ellis g. f. r. , baldwin j. e. , 1984 , mnras , 206 , 377 faber s. m. , jackson r. e. , 1976 , apj , 204 , 668 faber s. m. , courteau s. , dekel a. , dressler a. , kollatt t. , willick j. a. , yahil a. , 1994 , jrasc , 88 , 92 futamase t. , sasaki m. , 1989 , phrvd , 40 , 2502 gibelyou c. , huterer d. , 2012 , mnras , 427 , 1994 gorski k. , 1988 , apj , 332 , l7 gorski k. m. , davis m. , strauss m. a. , white s. d. m. , yahil a. , 1989 , apj , 344 , 1 graham a. w. , 2002 , mnras , 334 , 859 gregory p. c. , condon j. j. , 1991 , apjs , 75 , 1011 griffith m. r. , wright a. e. , 1993 , aj , 105 , 1666 groth e. j. , juszkiewicz r. , ostriker j. p. , 1989 ,
apj , 346 , 558 gunn j. e. , 1967a , apj , 147 , 61 gunn j. e. , 1967b , apj , 150 , 737 hamilton a. j. s. , 1998 , assl , 231 , 185 hildebrandt h. , et al . , 2011 , apj , 733 , l30 hinshaw g. , et al . , 2009 , apjs , 180 , 225 huchra j. p. , et al .
, 2012 , apjs , 199 , 26 hudson , m. j. 1994 , mnras , 266 , 468 hudson m. j. , turnbull s. j. , 2012 , apj , 751 , l30 huff e. m. , graves g. j. , 2014 , apj , 780 , l16 hui l. , greene p. b. , 2006 , phrvd , 73 , 123526 jenkins a. , et al . , 1998 ,
apj , 499 , 20 kaiser n. , 1987 , mnras , 227 , 1 kaiser n. , 1988 , in _ large scale structure and motions in the universe , _ ed .
giuricin , g. , mardirossian , f. , mezzetti , m. & ramella , m. ( dordrecht : kluwer ) , p. 197 kaiser n. , hudson m. j. , 2015 , mnras , submitted kashlinsky a. , atrio - barandela f. , ebeling h. , edge a. , kocevski d. , 2010 , apj , 712 , l81 kashlinsky a. , atrio - barandela f. , ebeling h. , 2011 , apj , 732 , 1 koda j. , et al . , 2014 , mnras , 445 , 4267 kormendy j. , 1977 , apj , 218 , 333 la barbera f. , covone g. , busarello g. , capaccioli m. , haines c. p. , mercurio a. , merluzzi p. , 2005 ,
mnras , 358 , 1116 lahav o. , piran t. , treyer m. a. , 1997 , mnras , 284 , 499 landy s. d. , szalay a. s. , broadhurst t. j. , 1998 , apj , 494 , l133 laureijs r. , et al . , 2011 , arxiv , arxiv:1110.3193 lavaux g. , afshordi n. , hudson m. j. , 2013 , mnras , 430 , 1617 lee m. g. , freedman w. l. , madore b. f. , 1993 , apj , 417 , 553 lynden - bell d. , 1992 , in statistical challenges in modern astronomy , eds .
feigelson & babul , p201 marozzi g. , 2015 , cqgra , 32 , 045004 mcdonald p. , 2009 ,
jcap , 11 , 26 mertens j. b. , yoho a. , starkman g. d. , 2013 , mnras , 432 , 1315 nusser a. , branchini e. , davis m. , 2011 , apj , 735 , 77 nusser a. , branchini e. , feix m. , 2013 , jcap , 1 , 18 nusser a. , davis m. , branchini e. , 2014 , apj , 788 , 157 peebles p. j. e. , 1980 , the large scale structure of the universe , princeton , princeton university press percival w. j. , samushia l. , ross a. j. , shapiro c. , raccanelli a. , 2011 , rspta , 369 , 5058 perlmutter s. , et al . , 1999 ,
apj , 517 , 565 planck collaboration , et al . , 2014 , a&a , 561 , a97 plionis m. , georgantopoulos i. , 1999 , mnras , 306 , 112 pyne t. , birkinshaw m. , 2004 , mnras , 348 , 581 rengelink r. b. , tang y. , de bruyn a. g. , miley g. k. , bremer m. n. , roettgering h. j. a. , bremer m. a. r. , 1997 , a&as , 124 , 259 riess a. g. , et al . , 1998 , aj , 116 , 1009 rubart m. , schwarz d. j. , 2013 , a&a , 555 , a117 rubin v. c. , thonnard n. , ford w. k. , jr .
, roberts m. s. , 1976 , aj , 81 , 719 sachs r. k. , wolfe a. m. , 1967 , apj , 147 , 73 sasaki m. , 1987 , mnras , 228 , 653 sasaki m. , 1993 , pthph , 90 , 753 scharf c. a. , jahoda k. , treyer m. , lahav o. , boldt e. , piran t. , 2000 , apj , 544 , 49 schmidt f. , leauthaud a. , massey r. , rhodes j. , george m. r. , koekemoer a. m. , finoguenov a. , tanaka m. , 2012 , apj , 744 , l22 scodeggio m. , giovanelli r. , haynes m. p. , 1997
, aj , 113 , 2087 scranton r. , et al . , 2005 , apj , 633 , 589 singal a. k. , 2011 , apj , 742 , l23 smith a. j. , loveday j. , cross n. j. g. , 2009 , mnras , 397 , 868 springob c. m. , masters k. l. , haynes m. p. , giovanelli r. , marinoni c. , 2007 , apjs , 172 , 599 springob c. m. , magoulas c. , colless m. , jones d. h. , campbell l. , lucey j. , mould j. , erdodu p. , 2013 ,
iaus , 289 , 269 strauss m. a. , willick j. a. , 1995 , phr , 261 , 271 sugiura n. , sugiyama n. , sasaki m. , 1999 , pthph , 101 , 903 tammann g. a. , yahil a. , sandage a. , 1979 , apj , 234 , 775 tonry j. l. , blakeslee j. p. , ajhar e. a. , dressler a. , 2000 , apj , 530 , 625 tully r. b. , fisher j. r. , 1977 , a&a , 54 , 661 tully r. b. , et al . , 2013 , aj , 146 , 86 turnbull s. j. , hudson m. j. , feldman h. a. , hicken m. , kirshner r. p. , watkins r. , 2012 , mnras , 420 , 447 umeh o. , clarkson c. , maartens r. , 2012 , arxiv , arxiv:1207.2109 watkins r. , feldman h. a. , 2007 , mnras , 379 , 343 zeldovich y. b. , 1964 , sva , 8 , 13
one way to analyse cosmic - flow data would be to consider each galaxy to provide a measure of the peculiar velocity at a position in space which is the estimated distance .
if one lived in a rich cluster then mapping velocity vs. estimated - distance space coordinates would make considerable sense .
but for large - scale peculiar velocity studies this would be subject to biases of various kinds . at cosmological distances redshift
gives a more faithful representation of real space than estimated distance space ( where the errors increase linearly with distance ) . if one considers the offsets of the flux densities or magnitudes from their expected values as giving an estimate of the peculiar displacement as a function of redshift space position ( method ii in the terminology of strauss & willick 1995 ) then a lot of the bias problems are ameliorated .
but it is not completely free of bias .
if you imagine galaxies as having random motions , much like molecules in a gas , ( or large random measurement errors , as is the case for photometric redshifts ) then , if you look in a localised region in redshift space ( perhaps a thin shell within a cone of some solid angle ) then , if there is , for instance , a radially outwards increasing density of galaxy density then there will be more galaxies in the selected region that are scattered in from greater distances than from the nearer side .
that results in a positive bias in the estimated luminosity distance at the chosen redshift that is proportional to the radial gradient of the density of galaxies .
this bias is present even in a homogeneous universe at low redshift because the number density of galaxies increases outwards @xmath165 .
the bias arising from this effect was first considered by lynden - bell ( 1992 ) and is explored more recently by us in a separate paper ( kaiser and hudson 2015 ) .
there is a related effect arising due to the presence of density inhomogeneities , where , for example , on the near side of a supercluster the density is rising more quickly than in a homogeneous universe and so there will be more galaxies scattered down from greater distances than scattered up from nearer distances .
this effect is opposite in sign to that produced by peculiar motions , which are proportional to ( minus ) the radial gradient of the potential .
the peculiar displacement is @xmath166 .
if this has a probability distribution , independent of position , @xmath167 then the mean peculiar displacement of galaxies is @xmath168 where @xmath169 is the real - space galaxy density contrast .
if there is only a large - scale density perturbation then one can locally expand @xmath170 as a taylor series and , since the mean of the distribution of @xmath89 is taken to be zero , we have @xmath171 corresponding to a peculiar velocity @xmath172 .
for a plane wave , @xmath173 the inferred peculiar velocity is @xmath174 so if the velocity dispersion is known , one can estimate , and correct for , the bias on any scale of interest ( noting , of course , that wavelength and wave number are related by @xmath175 ) . this bias or ` damping ' of the measured peculiar velocity will tend to bias estimates of @xmath140 from comparison of density and velocity fields . in the analysis of willick & strauss ( 1998 ) the effect of small - scale motions was included in the model .
it is worth noting that that a similar bias also affects peculiar velocities measured at a given estimated distance ( method i ) . in this case
the bias is known as `` inhomogeneous malmquist bias '' ( hudson 1994 , strauss & willick 1995 ) .
it has the opposite sign as the bias considered above since uncertainties in the distances cause the estimated real - space positions of galaxies to scatter out of overdensities , leading to an apparent peculiar velocity signal that has the same sign as the infall signal .
usually this bias is eliminated by using the galaxy density field to correct individual distance estimates .
however it is also possible in principle to make a statistical correction on a mode - by - mode basis . in this case
, the bias is very similar to the one described above , but with opposite sign : @xmath176 note that for most realistic surveys @xmath177 , hence this bias is larger than the method ii bias .
returning to the case of peculiar displacement in redshift space , it is clear that the model for random motions discussed above is rather idealised .
it may be a reasonable model for redshift measurement errors , but real galaxy motions are a complicated mix of virialised velocities within bound systems and streaming motions on larger scales .
we would argue nonetheless that this model should be applicable provided one is dealing with large - scale waves in the linear regime to describe the effect of small @xmath178 scale motions in virialised groups and clusters . whether one should include the effect of motions on intermediate scales depends rather on how the data are analysed . a very large scale density enhancement ( the scale of interest ) by itself causes a smooth deformation of the physical distance shell corresponding to the chosen constant @xmath34 .
perturbations on some smaller scale ( which we ll assume here is either linear or quasi - linear so position and redshifts are single - valued functions of one - another ) will further corrugate the constant-@xmath34 distance surface .
if one makes an estimate of the very large scale velocity by averaging the flux densities of the galaxies then the intermediate scale perturbations will bias the result in the manner of ( [ eq : velbias ] ) .
this is because , again in the case that the large - scale mode has a positive radial density gradient , wherever the intermediate scale motions result in a positive peculiar displacement there will , on average , be a slightly higher density of galaxies and these being more distant will have a slightly higher hubble velocity .
but if one were to bin the measurements on the intermediate scale and create peculiar velocity estimates normalising by the local density of galaxies , then the intermediate scale motions would no longer couple to the larger - scale density perturbations . in redshift space distortions , large - scale peculiar motions cause a ` squishing ' of the clustering pattern , enhancing the apparent amplitude of modes with wave vector out of the plane of the sky .
but the peculiar velocity from smaller scale motions tends to counteract that . in ( [ eq : velbias ] ) we see that large - scale motions are also biased .
the velocity dispersion here is a ` single - galaxy ' or ` 1-point ' statistic . in rsd
we deal with the pairwise velocity dispersion . at small
projected separation , where both pairs most probably reside in the same virialised halo the pairwise averaging gives a large variance as it tends to be dominated by large clusters .
but for pairs with substantial projected separation the distribution function is the convolution of the distribution for two places which , as far as small - scale motions are concerned provide a velocity dispersion very similar to that in ( [ eq : velbias ] ) .
thus it would seem that measurements of the velocity bias given by ( [ eq : velbias ] ) which depends to some extend on assumptions about the galaxy density bias since @xmath179 could provide independent constraints on the distribution of small - scale motions that is the nuisance factor in rsd measurements . in @xmath27cdm
the pairwise line - of - sight velocity dispersion falls from about 700 km / s at @xmath180 scales to about 500 km / s at a few mpc/@xmath181 and above ( jenkins et al .
this is consistent with the idea that at small separation one is seeing pairs in the same halo , so one is measuring the velocity dispersion of clusters weighted by the square of the number of galaxies in those clusters , rather than weighting galaxies equally .
the estimate from lcrs ( jing , mo & borner 1997 ) is flat at @xmath10 500 km / s out to a few mpc ( the uncertainty becoming large for larger separation ) .
we interpret this to mean that at these separations the pairwise velocity difference distribution is the convolution of the single - particle distribution with itself so that the appropriate value for the 1-particle rms velocity is @xmath182 km / s . with @xmath183 ( i.e. @xmath184 )
this suggests that the bias in ( [ eq : velbias ] ) would be a @xmath13 25% effect for @xmath185 ( or wavelength @xmath186 ) .
koda et al .
2013 have calculated the power spectrum of peculiar velocities in numerical cosmological simulations .
they observed a damping at high frequencies of the velocity power as measured in redshift space that they found could be described by a ` sinc ' function : @xmath187 with @xmath188 km / s . expanding this to lowest non trivial order would give @xmath189 or equivalently @xmath190 with @xmath191 km / s ( with @xmath192 ) , in excellent agreement with our prediction .
however , we would emphasise that this 1-particle velocity dispersion from simulations is considerably larger than the value seemingly preferred by the data of willick & strauss ( 1998 ) who found a ml value of @xmath193 km / s . equation ( [ eq : velbias ] ) can also be used to estimate the bias of velocities caused by errors in photometric redshifts based on broad - band filter photometry .
but the results , for conventional broad - band imaging surveys delivering @xmath194% fractional errors in @xmath95 are quite depressing
. a wider range of scales can be probed using j - pas with its 56 narrow band filters , which is designed to deliver much better precision of @xmath195 . | we consider some aspects of the perturbation to the luminosity distance @xmath0 that are of relevance for sn1a cosmology and for future peculiar velocity surveys at non - negligible redshifts .
\1 ) previous work has shown that the correction to the lowest order perturbation @xmath1 has the peculiar characteristic that it appears to depend on the absolute state of motion of sources , rather than on their motion relative to that of the observer .
the resolution of this apparent violation of the equivalence principle is that it is necessary to allow for evolution of the velocities with time , and also , when considering perturbations on the scale of the observer - source separation , to include the gravitational redshift effect .
we provide an expression for @xmath2 that provides a physically consistent way to measure peculiar velocities and determine their impact for sn1a cosmology .
\2 ) we then calculate the perturbation to the redshift as a function of source flux density , which has been proposed as an alternative probe of large - scale motions .
we show how the inclusion of surface brightness modulation modifies the relation between @xmath3 and the peculiar velocity , and that , while the noise properties of this method might appear promising , the velocity signal is swamped by the effect of galaxy clustering for most scales of interest .
\3 ) we show how , in linear theory , peculiar velocity measurements are biased downwards by the effect of smaller scale motions or by measurement errors ( such as in photometric redshifts ) .
our results nicely explain the effects seen in simulations by koda et al . 2013 .
we critically examine the prospects for extending peculiar velocity studies to larger scales with near - term future surveys .
cosmology : theory , observations , distance scale , large - scale structure ; galaxies : distances and redshifts |
in the past few years there has been increasing interest in the investigation of financial markets as complex systems in statistical mechanics .
the empirical analysis of the high frequency financial data reveals nonstationary statistics of market fluctuations and several mathematical models of markets based on the concept of the nonequilibrium phenomena have been proposed @xcite .
recently mizuno _ et al .
_ investigate high frequency data of the usd / jpy exchange market and conclude that dealers perception and decision are mainly based on the latest 2 minutes data @xcite .
this result means that there are feedback loops of information in the foreign currency market . as microscopic models of financial markets
some agent models are proposed @xcite . specifically ising - like models are familiar to statistical physicists and
have been examined in the context of econophysics .
the analogy to the paramagnetic - ferromagnetic transition is used to explain crashes and bubbles .
_ consider the effect of a weak external force acting on the agents in the ising model of the financial market and conclude that apparently weak stimuli from outside can have potentially effect on financial market due to stochastic resonance @xcite .
this conclusion indicates that it is possible to observe the effect of the external stimuli from the market fluctuations . motivated by their conclusion we investigate high - frequency financial data and find a potential evidence that stochastic resonance occurs in financial markets . in this article the results of data analysis
are reported and the agent - based model is proposed in order to explain this phenomenon .
we analyze tick quotes on three foreign currency markets ( usd / jpy , eur / usd , and jpy / eur ) for periods from january 1999 to december 2000 @xcite .
this database contains time stamps , prices , and identifiers of either ask or bid . since generally market participants ( dealers ) must indicate both ask and bid prices in foreign currency markets the nearly same number of ask and bid offering are recorded in the database . here
we focus on the ask offering and regard the number of ask quotes per unit time ( one minute ) @xmath0 as the market activity .
the reason why we define the number of ask quotes as the market activity is because this quantity represents amount of dealers responses to the market . in order to elucidate temporal structure of the market activity power spectrum densities of @xmath0 , estimated by @xmath1 where @xmath2 represents frequency , and @xmath3 a maximum period of the power spectrum density , are calculated .
fig : power - spectrum - usdjpy ] , [ fig : power - spectrum - eurusd ] , and [ fig : power - spectrum - eurjpy ] show the power spectrum densities for three foreign currency markets ( usd / jpy , uer / usd , and eur / jpy ) from january 1999 to december 2000 .
it is found that they have some peaks at the high frequency region .
there is a peak at 2.5 minutes on the usd / jpy market , at 3 minutes on the eur / usd market , and there are some peaks on the jpy / eur .
we confirm that these peaks appear and disappear depending on observation periods . on the usd / jpy market
there is the peak for periods of january 1999july 1999 , march 2000april 2000 , and august 2000november 2000 ; on the eur / usd market july 1999september 1999 ; and on the eur / jpy market january 1999march 1999 , april
1999june 1999 , november 1999 , and july 2000december 2000 .
these peaks mean that market participants offer quotes periodically and in synchronization .
the possible reasons for these peaks to appear in the power spectrum densities of the market activity are follows : 1 .
the market participants are affected by common periodic information .
2 . the market participants are spontaneously synchronized . in the next section the double - threshold agent model
is introduced and explain this phenomenon on the basis of the reason ( 1 ) .
here we consider a microscopic model for financial markets in order to explain the dependency of the peak height on observation periods . we develop the double - threshold agent model based on the threshold dynamics . in foreign exchange markets the market participants attend the markets with utilizing electrical communication devices , for example , telephones , telegrams , and computer networks .
they are driven by both exogenous and endogenous information and determine their investment attitudes .
since the information to motivate buying and one of selling are opposite to each other we assume that the information is a scaler variable .
moreover the market participants perceive the information and determine their investment attitude based on the information .
the simplest model of the market participant is an element with threshold dynamics .
we consider a financial market consisting of @xmath4 market participants having three kinds of investment attitudes : buying , selling , and doing nothing .
recently we developed an array of double - threshold noisy devices with a global feedback @xcite . applying this model to the financial market
we construct three decision model with double thresholds and investigate the dependency of market behavior on an exogenous stimuli .
the investment attitude of the @xmath5th dealer @xmath6 at time @xmath7 is determined by his / her recognition for the circumstances @xmath8 , where @xmath9 represents the investment environment , and @xmath10 the @xmath5th dealer s prediction from historical market data .
@xmath6 is given by @xmath11 where @xmath12 and @xmath13 represent threshold values to determine buying attitude and selling attitude at time @xmath7 , respectively .
@xmath14 is the uncertainty of the @xmath5th dealer s decision - making . for simplicity
it is assumed to be sampled from an identical and independent gaussian distribution , @xmath15 where @xmath16 is a standard deviation of @xmath14 . of course this assumption can be weakened .
namely we can extend the uncertainty in the case of non - gaussian noises and even correlated noises .
the excess demand is given by the sum of investment attitudes over the market participants , @xmath17 which can be an order parameter .
furthermore the market price @xmath18 moves to the direction to the excess demand @xmath19 where @xmath20 represents a liquidity constant and @xmath21 is a sampling period .
@xmath22 may be regarded as an order parameter .
the dealers determine their investment attitude based on exogenous factors ( fundamentals ) and endogenous factors ( market price changes ) . generally speaking ,
the prediction of the @xmath5th dealer @xmath10 is determined by a complicated strategy described as a function with respect to historical market prices , @xmath23 . following the takayasu
s first order approximation @xcite we assume that @xmath10 is given by @xmath24 where @xmath25 is the @xmath5th dealer s response to the market price changes . it is assumed that the dealers response can be separated by common and individual factors , @xmath26 where @xmath27 denotes the common factor , and @xmath28 the individual factor . generally these factors are time - dependent and seem to be complicated functions of both exogenous and endogenous variables . for simplicity it is assumed that these factors vary rapidly in the limit manner . then this model becomes well - defined in the stochastic manner .
we assume that @xmath27 and @xmath28 are sampled from the following identical and independent gaussian distributions , respectively : @xmath29 where @xmath30 represents a mean of @xmath27 , @xmath31 a standard deviation of @xmath27 , and @xmath32 a standard deviation of @xmath33 .
since we regard the market activity as the number of tick quotes per unit time it should be defined as the sum of dealers actions : @xmath34 the market activity @xmath35 may be regarded as an order parameter .
this agent model has nine model parameters .
we fix @xmath36 , @xmath37 , @xmath38 , @xmath39 , @xmath40 , and @xmath41 throughout all numerical simulations . it is assumed that an exogenous periodic information to the market is subject to @xmath42 at @xmath43 , @xmath44 and @xmath45 .
we calculate the signal - to - noise ratio ( snr ) of the market activity as a function of @xmath46 .
the snr is defined as @xmath47 where @xmath48 represents a peak height of the power spectrum density , and @xmath4 noise level . from the numerical simulation we find non - monotonic dependency of the snr of @xmath35 on @xmath46 .
[ fig : snr ] shows a relation between the snr and the noise strength @xmath46 .
it has an extremal value around @xmath49 .
namely the uncertainty of decision - making plays a constructive role to enhance information transmission . if there are exogenous periodic information and the uncertainty of decision - making we can find the peak on power spectrum densities at appropriate uncertainty of decision - making due to stochastic resonance @xcite . for the double - threshold agent model
is plotted against the uncertainty of decision - making of agents at @xmath36 , @xmath37 , @xmath38 , @xmath39 , @xmath50 , @xmath41 , @xmath51 , @xmath43 , @xmath44 , and @xmath45 . ]
we analyzed time series of the number of tick quotes ( market activity ) and found there are short - time periodicities in the time series .
the existence and positions of these peaks of the power spectrum densities depend on foreign currency markets and observation periods .
the power spectrum densities have a peak at 2.5 minutes on the usd / jpy market , 3 minutes on the eur / usd .
there are some peaks at a few minutes on the jpy / eur .
we developed the double - threshold agent model for financial markets where the agents choose three kinds of states and have feedback strategies to determine their decision affected by last price changes .
from the numerical simulation we confirmed that the information transmission is enhanced due to stochastic resonance related to the uncertainty of decision - making of the market participants .
we propose a hypothesis that the periodicities of the market activity can be observed due to stochastic resonance .
appearance and disappearance of these peaks may be related to the efficiency of the markets .
the efficiency market hypothesis @xcite says that prices reflect information .
because quotes make prices tick frequency can reflect information .
if the peaks of the power spectrum densities come from exogenous information then snr is related to the efficiency of the market .
namely the market may be efficient when the peaks appear .
the author thanks prof .
t. munakata for stimulative discussions and useful comments .
this work is partially supported by the japan society for the promotion of science , grant - in - aid for scientific research # 17760067 .
99 r.n . mantegna and h.e .
stanley , an introduction to econophysics : correlations and complexity in finance , cambridge university press , cambridge ( 2000 ) . m.m .
dacorogna , r. genay , u. mller , r.b . olsen and o.v .
pictet , an introduction to high - frequency finance , academic press , san diego ( 2000 ) .
t. mizuno , t. nakano , m. takayasu , and h. takayasu , physica a , * 344 * ( 2004 ) 330 . a .- h .
sato and h. takayasu , physica a * 250 * ( 1998 ) 231 . t. lux and m. marchesi , nature , * 397 * ( 1999 ) 498500 .
d. challet , m. marsili and yi - cheng zhang , physica a , * 294 * ( 2001 ) 514524 .
p. jefferies , m.l .
hart , p.m. hui and n.f .
johnson , the european physical journal b , * 20 * ( 2001 ) 493501 .
a. krawiecki and j.a .
hoyst , physica a , * 317 * ( 2003 ) 597 .
the data is provided by cqg international ltd .
sato , k. takeuchi , and t. hada , physics letters a , * 346 * ( 2005 ) 27 .
h. takayasu , and m. takayasu , physica a , * 269 * ( 1999 ) 24 .
l. gammaitoni , p. hnggi , p. jung , and f. marchesoni , review of modern physics , * 70 * ( 1998 ) 223 .
e. fama , journal of finance , * 46 * ( 1991 ) 1575 . | power spectrum densities for the number of tick quotes per minute ( market activity ) on three currency markets ( usd / jpy , eur / usd , and jpy / eur ) for periods from january 1999 to december 2000 are analyzed .
we find some peaks on the power spectrum densities at a few minutes .
we develop the double - threshold agent model and confirm that stochastic resonance occurs for the market activity of this model .
we propose a hypothesis that the periodicities found on the power spectrum densities can be observed due to stochastic resonance .
tick quotes , foreign currency market , power spectrum density , double - threshold agent model , stochastic resonance + * pacs * 89.65.gh , 87.15.ya , 02.50.-r |
important in both fundamental science and numerous applications , optimization problems of various degrees of complexity are challenging ( see @xcite for an excellent introduction ) .
optimization conditioned by constraints that may vary from event to event is of especial theoretical and practical importance . as a first example , when dealing with a system under a random potential , each realization of the random potential demands a separate optimization resulting in a different ground state .
the thermodynamic behavior of such a system in a quenched random potential crucially depends on the random potential realized .
a similar but practical problem may arise in routing passengers at various cities to reach their destinations . in the latter case ,
the optimal routing depends on the number of passengers at various locations , the costs from one location to the others , which likely to vary from time to time .
this type of conditional optimization also occurs in modern proteomics problem , that is , in the mass spectrometry ( ms ) based peptide sequencing . in this case , each tandem ms ( ms@xmath0 ) spectrum constitute a different condition for optimization which aims to find a database peptide or a _ de novo _ peptide to best explain the given ms@xmath0 spectrum . when the cost function of an optimization problem can be expressed as a sum of independent local contributions , the problem usually can be solved using the transfer matrix method that is commonly employed in statistical physics .
a well - studied example of this sort in statistical physics is the directed polymer / path in a random medium ( dprm ) @xcite .
even when a small non - local energetics is involved , the transfer matrix approach still proves useful @xcite . as an example , the close relationship between the dprm problem and ms - based peptide sequencing , where a small nonlocal energetics is necessary to enhance the peptide identifications , was sketched in an earlier publication @xcite and the cost value distribution from many possible solutions other than the optimal one is explored .
indeed , obtaining the cost value distribution from _ all _ possible solutions in many cases is harder than finding the optimal solution alone . in this paper
, we will provide the solution to a generic problem that enables a full characterization of the peptide sequencing score statistics , instead of just the optimal peptide .
the 1d problem considered is essentially a hopping model in the presence of a random potential .
the solution to this problem may also be useful in other applications such as in routing of passengers and even internet traffic . in what follows
, we will first introduce the generic 1d hopping model in a random potential , followed by its transfer matrix ( or dynamic programming ) solution .
we then discuss the utility of this solution in the context of ms - based peptide sequencing , and demonstrate with real example from mass spectrum in real ms - based proteomics experiments . in the discussion section , we will sketch the utility of the transfer matrix solution in other context and then conclude with a few relevant remarks .
along the @xmath1-axis , let us consider a particle that can hop with a set of prescribed distances @xmath2 towards the positive @xmath3 direction .
that is , if the particle is currently at location @xmath4 , it can move to location @xmath5 , @xmath6 ,
@xmath7 in the next time step . at each hopping step
, the particle will accumulate an energy @xmath8 from location @xmath1 that it just visited .
the score @xmath9 ( negative of the on - site potential energy ) is assumed positive and may only exist at a limited number of locations . for
locations that @xmath9 do not exist , we simply set @xmath10 there .
the energy of a path starting from the origin specified by the sequential hopping events @xmath11 would have visited locations @xmath12 with @xmath13 and has energy @xmath14 in general , there can be more than one path terminated at the same point . treating each path as a state with energy given by @xmath15 ,
one ends up having the following recursion relation for the partition function @xmath16 [ pf_global.0 ] z(x ) = _
i=1^k e^s(x - m_i ) z(x - m_i ) , where @xmath17 plays the role of inverse temperature ( with @xmath18 chosen ) .
if one were only interested in the best score terminated at point @xmath1 , it will be given by the zero temperature limit @xmath19 and the recursion relation may be obtained by taking the logarithm on both sides of ( [ pf_global.0 ] ) and divided by @xmath20 then taking @xmath19 limit to reach [ pf_zt_best ] s_best(x ) = _ 1i k \ { s(x - m_i)+s_best(x - m_i ) } , where @xmath21 records the best path score among all paths reaching position @xmath1 .
this update method , also termed dynamic programming , records the lowest energy and lowest energy path reaching a given point @xmath1 .
the lowest energy among all possible at position @xmath1 is simply @xmath22 and the associated path can be obtained by tracing backwards the incoming steps .
it is interesting to observe that one can also obtain the worst score at each position via dynamical programming [ pf_zt_worst ] s_worst(x ) = _ 1i k \ { s(x - m_i)+s_worst(x - m_i ) } .
the full thermodynamic characterization demands more information than the ground state energy . in principle , one may obtain the full partition function using eq .
( [ pf_global.0 ] ) evaluated at various temperatures .
this procedure , however , hinders analytical property such as determination of the average energy @xmath23 a better starting point may be achieved if one can obtain the density of states @xmath24 . in this case , we have z & & de e^-e d(e ) + e & = & .note that if the ground energy @xmath25 of the system is bounded from below , the partition function is simply a laplace transform of a modified density of states given by @xmath26 where @xmath27 and @xmath28 this implies that the density of states @xmath24 together with the ground state energy @xmath25 determine all the thermodynamic behavior of the system . in the next section
, we will explain how to obtain the density of states using the dynamical programming technique as well as how to extend this approach to more complicated situations that will be useful in characterizing the score statistics in ms - based peptide sequencing .
the density of states is related to the energy histogram in a simple way .
the number of states between energies @xmath29 and @xmath30 ( with @xmath31 ) is given by @xmath32 . if we happen to use @xmath33 as the energy bin size for energy histogram , the count @xmath34 in the bin with energy @xmath29
is simply @xmath35 and the density of states @xmath36 . for simplicity
, we will assume that the all the on - site energies @xmath8 are integral multiple of @xmath37 .
this implies that each path energy / score is also an integral multiple of @xmath33 . in the following subsections
, we will use score density of states instead of energy density of states .
we denote by @xmath38 the number of paths reaching position @xmath1 with score @xmath39 . with this notation ,
we can easily write down the recursion relation for @xmath38 as follows [ s.dos ] c(x , n ) = _ i=1^k c(x - m_i , n- ) .
this recursion relation allows us to compute the density of states in the same manner as computing the partition function ( [ pf_global.0 ] ) except that we need to have an additional dimension for score at each position @xmath1 . as an even simpler application of this recursion relation ,
suppose that one is only interested in the number of paths reaching position @xmath1 , one may sum over the energy part on both side of ( [ s.dos ] ) and arrives at [ p_count ] c(x ) = _ i c(x - m_i ) , which enables a very speedy way to compute the total number of paths reaching position @xmath1 .
in the context of _ de novo _ peptide sequencing @xcite , this number corresponds to the total number of all possible _ de novo _ peptides within a given small mass range .
although simply obtained , this number may be useful for providing rough statistical assessment in _ de novo _ peptide sequencing . in general , one may wish to associate with each hop an energy @xmath40 or one may wish to introduce some kind of score normalization based on the number of hopping steps .
this is indeed the case when applying this framework to ms - based peptide sequencing where a peptide length factor adding or multiplying to the overall raw score is a common practice . in this case
, it becomes important to keep track the number of hops made in each path .
we may further categorize the counter @xmath38 into @xmath41 .
that is , we may separate the paths with different number of steps from one another and arrive at a finer counter @xmath42 which records the number of paths reaching position @xmath1 with score @xmath39 and with @xmath43 hopping steps .
it is rather easy to write down the recursion relation obeyed by this fine counter [ sl.dos ] c(x , n , l ) = _
i=1^k c(x - m_i , n-,l-1 ) .
this recursion relation allows us to renormalize the raw score based on the number of steps taken .
for example , for raid_dbs @xcite , a database search method we developed , we divide the raw score obtained by @xmath44 for any peptide ( path ) of @xmath43 amino acids ( hopping steps ) to get better sensitivity in peptide identification . in principle , the recursion relations given by ( [ s.dos]-[sl.dos ] ) are all one - dimensional updates .
the only difference is the internal structure of counters at each position @xmath1 . for ( [ p_count ] ) ,
the counter is just an integer and has no further structure . for ( [ s.dos ] ) ,
the counter at each position has a 1d structure indexed by the score . for ( [ sl.dos ] ) ,
the counter at each position @xmath1 has a 2d structure indexed by both the score and the number of hopping steps .
this means that in terms of solving the problem using dynamical programming , it is always a 1d dynamical programming with different degrees of internal structure that may lengthen the execution time when shifting from the simplest case ( [ p_count ] ) to the more complicated case ( [ sl.dos ] ) . obviously at each position
@xmath1 , there is an upper bound and a lower bound for score and also for the number of hopping steps accumulated .
we shall call them @xmath21 , @xmath45 , @xmath46 and @xmath47 respectively .
the first two quantities may be obtained by eqs .
( [ pf_zt_best ] ) and ( [ pf_zt_worst ] ) respectively .
we provide the recursions for the two latter quantities below l_max(x ) & = & _ 1i k \ { l_max(x - m_i ) } + 1 [ l_max ] , + l_min(x ) & = & _ 1i k \ { l_min(x - m_i ) } + 1 [ l_min ] .
( [ pf_zt_best]-[pf_zt_worst ] ) and ( [ l_max]-[l_min ] ) provides the ranges for both the scores and the number of cumulative hopping steps at each position @xmath1 via simple dynamic programming . as we will discuss later
, this information enables a memory - efficient computations of score histograms .
in this section , we focus on an important subject in modern biology using ms data to identify the numerous peptides / proteins involved in any given biological process . because of the peptide mass degeneracies and the limited measurement accuracy for the peptide mass - to - charge ratio , using ms@xmath0 spectra is more effective in peptide identifications . in a ms@xmath0 setup ,
a selected peptide with its mass identified by the first spectrometer is fragmented by noble gas , and the resulting fragments are analyzed by a second mass spectrometer .
although such ms@xmath0-based proteomics approaches promise high throughput analysis , the confidence level assignment for any peptide / protein identified is challenging .
the majority of peptide identification methods are so - called database search approaches .
the main idea is to _ theoretically _ fragment each peptide in a database to obtain the corresponding _ theoretical _ spectra .
one then decides the degree of similarity between each theoretical spectrum and the input query spectrum using a scoring function .
the candidate peptides from the database are ranked / chosen according to their similarity scores to the query spectrum .
although one may assign relative confidence levels among the candidate peptides via various ( empirical ) means , an objective , standardized calibration exists only recently @xcite . in our earlier publications
@xcite , we proposed to tackle this difficulty by using a _ de novo _ sequencing method to provide an objective confidence measure that is both database - independent and takes into account spectrum - specific noise . in this paper , we will provide concrete algorithms for such purpose . to begin , consider a spectrum @xmath48 with parent ion mass range @xmath49 $ ] , we denote by @xmath50 the set of all `` possible '' peptides with masses in this range .
given a peptide @xmath51 from @xmath50 , the associated quality score @xmath52 is defined by a _ prescribed _ scoring system .
the score distribution of @xmath52 within @xmath50 provides naturally a likelihood measure for any given peptide @xmath51 to the the correct one . however , as described earlier @xcite , this seemingly straightforward idea faces two difficulties in terms of implementations .
first , unlike the dprm problem for which the function to be optimized is defined without ambiguity , the choice of the scoring function is somewhat empirical because the parameters used in the scoring must be trained using a training data set .
further , because of different instruments and experimental setups , it seems impossible to design a scoring system such that the correct peptide for each spectrum has the highest score among _ all possible _ peptides ; the application of a given scoring function to general cases may require a leap of faith .
second , even after the scoring function is chosen , it is not known how to find the peptide @xmath53 that maximizes @xmath54 as well as the score distribution @xmath55 within @xmath50 other than by the generally impractical procedure of examining all members of @xmath50 .
the first difficulty can be alleviated by validating high scoring _
de novo _ peptides via database searches @xcite and is not the main focus of the current paper .
note that a partial solution to the second problem via iterative mapping when nonlocal score contributions exist is provided earlier @xcite . here
we tackle the second problem head on when the scoring function used does not contain nonlocal contribution other than a final renormalization with respect to the peptide length .
our algorithms contains two parts : computer memory allocations and dynamical programming update .
prior to discussing these two parts , however , we first address the important issue of choosing a good mass unit .
the goal here is to choose a mass unit @xmath56 and expresses the molecular mass of each amino acid as an integral multiple of this unit .
for example , one may choose @xmath56 to be @xmath57 dalton ( da ) , and round the molecular mass of each amino acid to be an integral multiple of @xmath57 da .
once a mass unit is chosen , all the masses under consideration are integral multiples of this unit .
it turns out that different choices of the mass unit leads to different maximum cumulative mass error . as a specific example , consider using @xmath58 da as the mass unit .
the mass of alanine , with true mass @xmath59 da , is now represented as @xmath60 .
this molecular mass expression is @xmath61 da smaller than the true molecular mass of alanine . when this happens , the integral mass representation has a mass smaller than the true mass , and we call such type of mass error a _ down_-error .
now the amino acid tryptophan with molecular mass @xmath62 da will be assigned an integral mass of @xmath63 , which has an extra of @xmath64 da compared to the true mass .
we call this type of mass error the _ up_-error . the ratio of the mass error to the real molecular mass when multiplied by @xmath65 da provides the cumulative maximum error that can be induced by a single amino acid at @xmath66 da mass . for a fixed mass unit , we went over this mass error analysis for each of the twenty amino acids and documented the largest up - error and down - error .
the larger one between the maximum up - error and the maximum down - error is called the max - error . to search for best mass units that minimize the max - error at @xmath66 da
, we went over all possible mass unit ranging from @xmath67 da to @xmath68 da in step of @xmath69 da .
interestingly enough , we found a discrete list of mass units that have smaller max - error compared to their nearby mass units .
these numerically found _ magic _ mass units are summarized in table [ tab.magic ] . [ cols="^,^,^,^,^,^",options="header " , ] once a mass unit is chosen , all the amino acid masses are effectively integers . to obtain the score histogram of _ all _ _ de novo
_ peptides when queried by a spectrum @xmath48 with parent molecular mass @xmath70 ( with n- and c- terminal groups of the peptide stripped away ) , we first construct a mass array where index @xmath71 corresponds a molecular mass @xmath72 . to encode all possible peptides with molecular mass up to @xmath70 , we need to have an array of size @xmath73 .
apparently , when a larger mass unit is used , the size of the mass array is smaller and thus reduces computation time .
however , as one may see from table [ tab.magic ] , the larger mass unit is also accompanied by a larger max - error and might not be preferred when high mass accuracy is the first priority .
the basic idea of our algorithm is to encode all possible peptides in the mass array by linking pointers , analogous to the consecutive hopping steps in the 1d hopping model . for an amino acid @xmath74 ,
let @xmath75 represents its corresponding integer mass in unit of @xmath56 .
for a peptide made of @xmath76 $ ] , it will have a hopping trajectory in the molecular array given by @xmath77 $ ] with @xmath78 .
let us also denote @xmath79 by @xmath80 to indicate that it is the terminating point of the path .
apparently , all possible peptides with molecular masses equal to @xmath81 will all have corresponding hopping paths starting at the origin and terminating at @xmath80 . through appropriate pointer linking , one may therefore encode _ all _ possible peptides with molecular mass @xmath81 in a one - dimensional mass array . for a given spectrum @xmath48 , depending on the score function used ,
one may calculate local score contributions at each mass index .
this step is done once only for the whole mass array , and need not be repeated for each candidate peptide . in a typical ms@xmath0 experimental spectrum , there always exists some level of parent ion mass uncertainty .
once the size of the mass uncertainty is specified , we only need to examine _ de novo _ peptides whose corresponding hopping paths terminating at a few consecutive mass indices .
this indicates that some of the mass indices of the aforementioned mass array may not even be used in this context .
below we describe how to efficiently obtain _ relevant mass indices _ and only allocate computer memories for those masses .
assume that the possible terminating points are @xmath82 with @xmath83 .
the update rules described in eqs .
( [ pf_zt_best]-[pf_zt_worst ] ) , ( [ p_count ] ) , and ( [ l_max]-[l_min ] ) will also be used at this stage .
the following pseudocode describes our algorithm .
initialize the mass_index = 0 entry + @xmath84 ; @xmath85=1 ; + remark : max_aa is the maximum number of amino acids considered + for ( aa_index = 0 ; aa_index @xmath86 max_aa ; aa_index + + ) \ { + label occupancy of n(aa_index ) ; + at n(aa_index ) attach a pointer back to 0 ; + update @xmath87 , @xmath88 , @xmath89 , @xmath90 , @xmath85 at n(aa_index ) ; + } + for ( mass_index = 1 ; mass_index @xmath91 ; mass_index + + ) \ { + if ( mass_index occupied ? ) \ { + for ( aa_index = 0 ; aa_index @xmath86 max_aa ; aa_index + + ) \ { + label occupancy of ( mass_index + n(aa_index ) ) ; + at mass_index+n(aa_index ) attach a pointer to mass_index ; + update @xmath87 , @xmath88 , @xmath89 , @xmath90 , @xmath85 at ( mass_index + n(aa_index ) ) ; + } + } + } + for ( mass_index = @xmath92 ; mass_index @xmath93 ; mass_index )\ { + backtrack all possible paths @xmath94 final occupied entries ; + } + the last step in the algorithm above identifies _ relevant mass indices _ , mass_indices that will be traversed by the hopping paths of all peptides with molecular masses in the range @xmath95 $ ] . we only need to allocate computer memory associated with those sites .
for each of these relevant sites , we also know the values of @xmath87 , @xmath88 , @xmath89 , @xmath96 , and the total number of peptides reaching that site through the algorithm above . one may therefore allocate a 2d array of size @xmath97 for each relevant mass_index @xmath98 for later use . once memory allocation for relevant mass_indices
is done , we can efficiently go through those relevant sites to obtain the 2d score histogram that we mentioned . in the pseudocode
below , update is performed using eq .
( [ sl.dos ] ) .
we now demonstrate the very simple main algorithm initialize all the fine counters @xmath99 + except @xmath100=1 ; + for ( aa_index = 0 ; aa_index @xmath86 max_aa ; aa_index + + ) \ { + update @xmath42 at @xmath101n(aa_index ) ; + } + for ( mass_index in ascendingly ordered _
relevant mass_indices_)\ { + for ( aa_index = 0 ; aa_index @xmath86 max_aa ; aa_index + + ) \ { + update @xmath42 at @xmath102(mass_index + n(aa_index ) ) ; + } + } + we now define the final 2d counter [ 2dhist_final ] y(n , l ) _ i=1^k c(f_i , n , l ) . apparently , in the 1d hopping model when allowing @xmath71 consecutive terminating points , the resulting density of states @xmath24 can now be expressed as @xmath103 . if one were interested in normalizing the final score in a path - length dependent manner , one will has the following generic transformation h(e ) = _ lde ( e - f(e,l ) ) where @xmath104 is a generic length - normalized energetic function that takes the raw energy @xmath105 with @xmath43 hopping steps and turn them into a new energy @xmath104 , and @xmath106 is understood . using a real experimental ms@xmath0 spectrum of parent ion mass @xmath107 da and a raw scoring function ( raid_dbs @xcite raw score without divided by @xmath44 with @xmath43 being the peptide length ) , we obtained a 2d score histogram . from this 2d score histogram
, we can compute the average peptide length @xmath108 as well .
we then transform the 2d score histogram using two different @xmath109 functions . in the first case , @xmath110 , meaning that one just divides the score by a constant given by @xmath111 . in the second case
, we use the raid_dbs scoring function where @xmath112 . in figure
[ fig1 ] , we show the two resulting score histograms along with the fits to theoretical distribution function @xcite . as one may see from the figure , both histograms are well fitted by the theoretical distribution function over at least 15 order of magnitudes .
there is difference , however , in the histograms obtained .
in the first case , where the score is merely divided by the average length , we have a wider score distribution than that of the second case .
this implies that a high scoring hit out of the first type of scoring function will have a larger @xmath113-value than that of the second type .
this is perfectly reasonable because when using the first type of raw scoring , very long peptides which by random chances are more likely to hit on fragment peaks in the mass spectrum are less penalized than the shorter peptides . as a consequence ,
one anticipates more false long peptides out of the first type of scoring method than that of the second scoring method .
therefore , one should assign a larger @xmath113-value to the former case and a smaller @xmath113-value to the latter case .
it is apparently important to be able to obtain score histograms of the second scoring method .
however , this can only be achieved if one keeps the length information in the dynamical programming update , see eq .
( [ sl.dos ] ) .
our method may also be extended to other applications . in the case of passenger routings ,
the @xmath1-axis actually represents time .
the local score may be viewed as the additional cost that may vary for different stops . once the problem is laid out
, the 2d histogram obtained from our solution indicates the number of equivalent routes in terms of additional costs and the total number of stops .
this problem should be interesting in its own right . in this paper , we developed a new approach to obtain the density of states of a 1d hopping problem in random potential .
we have extended the simplest case scenario and have shown that we can apply this method to provide a _
complete _ score histogram for ms - based peptide sequencing problem .
this important information may be used for a more objective statistical significance assignment in peptide identification .
our algorithm may also serve as a speedy _ de novo _ algorithm .
if one is only interested in getting the best scoring peptide with length normalized score , one only needs to keep track of @xmath114 .
furthermore , it is straightforward to include in our _ de novo _ algorithm post - translationally modified amino acids .
the effect is simply an enlargement of the alphabet .
that is , instead of having 20 amino acids , we will simply have more allowed masses but without needing to change any part of the algorithm .
in the near future , we would like to build a web application that allows the users to obtain information of interest .
for example , a user might be interested in knowing : given a parent ion molecular mass and a mass error tolerance , how many _ de novo _ peptides can there be ?
furthermore , we plan to provide users with the full score histogram when a query spectrum is provided and a scoring method is chosen .
our approach , founded on statistical physics , can easily address this type of questions to provide useful information for biological researches .
this work was supported by the intramural research program of the national library of medicine at the national institutes of health . | we provide a complete thermodynamic solution of a 1d hopping model in the presence of a random potential by obtaining the density of states .
since the partition function is related to the density of states by a laplace transform , the density of states determines completely the thermodynamic behavior of the system .
we have also shown that the transfer matrix technique , or the so - called dynamic programming , used to obtain the density of states in the 1d hopping model may be generalized to tackle a long - standing problem in statistical significance assessment for one of the most important _ proteomic _ tasks peptide sequencing using tandem mass spectrometry data . statistical significance , dynamic programming , mass spectrometry , directed paths in random media , peptide identification |
for the nearest - neighbor ( nn ) ising antiferromagnet on the square lattice in a uniform magnetic field , the low temperature ordered phase is separated from the paramagnetic phase by a simple , 2nd order phase boundary .
( within the context of the lattice gas model this system could be described as having repulsive nn - coupling and forming a @xmath2 ordered state . ) with the addition of repulsive ( antiferromagnetic ) next - nearest - neighbor ( nnn ) interactions the situation becomes more complicated .
early monte carlo simulations suggested that a single , super - antiferromagnetic , or @xmath3 , phase existed , separated from the paramagnetic phase by a single phase boundary@xcite .
a degenerate , row - shifted @xmath4 state was also predicted at zero temperature .
( see fig .
[ f1 ] for a schematic representation of these states . ) on the other hand , symmetry arguments based on landau theory@xcite predict the order - disorder transitions of @xmath0 and @xmath1 structures belong to xy model with cubic anisotropy .
the original motivation of this study was to investigate the possibility of xy - like behavior of the ising spins on the square lattice , since there is numerical evidence@xcite that ising antiferromaget with attractive nnn interaction on the triangular lattice has an xy - like intermediate state between the low temperature ordered state and high temperature disordered state .
in fact , the present model has already been studied by many authors using different approaches .
an early monte carlo study@xcite comprehensively showed the phase diagrams for several different interaction ratios ( r ) of nnn to nn interaction .
but due to the deficiencies of computer resources at that time , for the @xmath5 case , a disordered region was missed between the two ordered phases which was pointed out in a later interfacial free energy study@xcite
. meanwhile , transfer matrix studies@xcite found reentrant behavior for the @xmath0 transition lines . while a study using the cluster variation method@xcite concluded that for a range of r ( [email protected] ) the system undergoes a first order transition , a recent monte carlo study@xcite using a variant of the wang - landau method@xcite focused on the @xmath5 case without external field and found the phase transition is of second order .
for external field @xmath7 ( in the unit of nn interaction constant ) the two ordered phases , namely @xmath0 and row - shifted @xmath1 , are degenerate at zero temperature so it is tempting to think that the cubic anisotropy would be zero for this field and that there could be a kosterlitz - thouless transition . in this paper , we carefully study the location of phase boundaries and the critical behavior for the case @xmath5 . in sec .
[ s1 ] the model and relevant methods and analysis techniques are reviewed .
our results are presented in sec . [ s2 ] , along with finite size scaling analyses , and we summarize and conclude in sec.[s3 ] .
the ising model with nnn interaction is described by the hamiltonian @xmath8 where @xmath9 , @xmath10 and @xmath11 are nn and nnn interaction constants , respectively , h is an external magnetic field , and the sums in the first two terms run over indicated pairs of neighbors on a square lattice with periodic boundary conditions .
both @xmath10 and @xmath11 are positive ( antiferromagnetic ) and the ratio @xmath12 . , @xmath3 and row - shifted @xmath4 ordered structures within the context of the lattice gas model .
( in magnetic language , filled circles correspond to up spins and empty circles correspond to down spins.),width=312 ] for the @xmath5 case , the ground states would be the @xmath0 state , also known as super - antiferromagnetic state , in small magnetic fields ; and at higher fields it would be a row - shifted @xmath1 state , which differs from the @xmath1 state in the sense that the antiferromagnetic chains in the former state can slide freely without energy cost .
see fig .
locally , the structure may appear to be @xmath1 , but for large enough lattices the equilibrium structure always shows row shifting . as a result ,
such a row - shifted @xmath1 state is highly degenerate , and the antiferromagnetic sublattice exhibits only one dimensional long range order . in terms of the sublattice magnetizations @xmath13 we can define two components of the order parameter for the @xmath0 state @xmath14/4,\ ] ] @xmath15/4,\ ] ] with a computationally convenient root - mean - square order parameter @xmath16 since @xmath17 would have a limiting value of @xmath18 for the row - shifted @xmath1 state and be zero for the disordered state , it can also be used as an order parameter for the row - shifted @xmath1 state .
other observables , such as the finite lattice ordering susceptibility @xmath19 and fourth - order cumulant @xmath20 , are defined in terms of the order parameter @xmath17 as @xmath21\ ] ] @xmath22 where n is the total number of spins and t is the simulation temperature . in some cases ,
the true ordering susceptibility @xmath23 , which is @xmath24 , is used to eliminate simulation errors resulting from @xmath25 , where the order parameter is known to be zero for the infinite lattice . for small lattice sizes ,
wang - landau sampling @xcite was used to obtain a quick overview of the thermodynamic behavior of our model . a two - dimensional random walk in energy and magnetization space
was performed so that the density of states @xmath26 could be used to determine all thermodynamic quantities ( derived from the partition function ) for any value of temperatures and external field . consequently , `` freezing '' problems are avoided at extremely low temperatures . this allowed us to determine the `` interesting '' regions of field - temperature space ; however , it quickly became apparent that , because of subtle finite size effects , quite large lattices would be needed .
unfortunately , as l increases , the number of entries of histogram explodes as @xmath27 and it proved to be more efficient to use parallel tempering instead .
since a large portion of interesting phase boundary is at relatively low temperatures and many local energy minima exist which makes the relaxation time rather long , the parallel tempering method@xcite is a good choice for simulating our model .
the basic idea is to expand the low temperature phase space by introducing configurations from the high temperatures .
so , many replicas at different temperatures are simulated simultaneously , and after every fixed number of monte carlo steps , a swap trial is performed with a metropolis - like probability which satisfies the detailed balance condition . the transition probability from a configuration @xmath28 simulated at temperature @xmath29 to a configuration @xmath30 simulated at temperature
@xmath31 would be @xmath32,\ ] ] @xmath33 we chose the temperatures for the replicas to be in a geometric progression@xcite , which would make acceptance rates relatively constant among neighboring temperature pairs , and the total number of temperatures was chosen to make the average acceptance rate above @xmath34 .
the multiplicative , congruential random number generator ranecu was used @xcite , and some results were also obtained using the mersenne twister @xcite for comparison .
no difference was observed to within the error bars .
typically , data from @xmath35 to @xmath36 mcs were kept for each run and 3 to 6 independent runs are taken to calculate standard statistical error bars . for parallel termpering
, the swap trial was attempted after every mcs . in all the plots of data and
analysis shown in following sections , if error bars are not shown they are always smaller than the size of the symbols
. in general , such replica exchange not only applies to the temperature set , but also can apply to any other sets of fields , such as the external magnetic field . following the same argument , one can get the transition probability from @xmath37 to @xmath38 @xmath39,\ ] ] @xmath40 where @xmath41 are the uniform magnetizations of replica m and n , respectively . to extract critical exponents from the data , we performed finite - size scaling analyses along the transition lines . since the maximum slope of the fourth - order cumulant @xmath20 follows@xcite @xmath42 where @xmath43 , the correlation length exponent @xmath44 can be estimated directly . with the exponent @xmath44 and critical temperature @xmath45 at hand
, the critical exponent @xmath46 and @xmath47 can be extracted from the data collapsing of the finite - size scaling forms , @xmath48 @xmath49 where @xmath50 , and @xmath51 and @xmath52 are universal functions whose analytical forms are not known .
one can also estimate the exponent @xmath53 from the relation of the peak position with lattice size for the specific heat @xmath54 where @xmath55 is the `` background '' contribution . in some cases
when the appropriate paths , i.e. which are perpendicular to the phase boundary , are ones of constant temperatures , then the critical behavior would be expressed in terms of reduced field @xmath56 , and all the above analysis still applies .
general purpose computing on graphics processing units(gpu ) attracts steadily increasing interest in simulational physics@xcite , since the computational power of recent gpu exceeds that of a central processing unit(cpu ) by orders of magnitude . the advantage continues to grow as the performance of gpu s doubles every year .
recently , a gpu accelerated monte carlo simulation of ising models@xcite was performed . compared to traditional cpu calculations , the speedup was about 60 times .
the idea behind the implementation in ref @xcite can be easily extended to our model , and the parallel tempering algorithm is naturally realized .
initially , all the replica are loaded to the global memory of the gpu . for each replica ,
the entire lattice is divided into four sublattices , then spins in the same sublattice can be updated simultaneously by the gpu using a metropolis scheme , and the swap of configurations of replica pairs can also be achieved in parallel . on the geforce gtx285 graphics unit ,
our code runs about 10 times faster than it does on the 32 cpus of ibm p655 cluster using message passing interface(mpi ) for parallel computation . .
open circles and pluses denote simulation results .
the solid lines are second order transition lines , while the dashed line indicates the short range ordering line . ]
from the ground state analysis , with zero or low field the ordered state would be the superantiferromagnetic , or @xmath0 , structure . as the external field increases to @xmath57 , more spins align in the opposite field direction , and
the ordered structure would be the row - shifted @xmath1 . with even stronger fields ,
the system becomes paramagnetic . in the region near @xmath58
a mixture of @xmath0 and row - shifted @xmath1 is visible . for finite temperatures
, we found that the fourth - order cumulant is always a good quantity to use to locate the transition points , while the data for other quantities , such as the specific heat or susceptibility , may look `` strange '' due to the effect of neighboring critical points . to help the reader understand the observed thermodynamic properties , the final phase diagram for @xmath5
is plotted in fig .
[ f2 ] : the solid lines are the phase boundaries , all of which are continuous .
the dashed line inside the @xmath0 ordered phase indicates a `` short range ordering '' line , which was located from the peak position of the specific heat . as shown in fig .
[ f3 ] , for paths of constant temperature @xmath59 and @xmath60 , there are two peaks , and the one that increases with lattice size corresponds to the critical point . versus field @xmath61 with lattice sizes @xmath62 for paths of constant : ( a)@xmath63 and ( b)@xmath59.,title="fig : " ] + versus field @xmath61 with lattice sizes @xmath62 for paths of constant : ( a)@xmath63 and ( b)@xmath59.,title="fig : " ] an indication of the complexity of the finite size behavior is clearly seen in the bottom portion of fig .
[ f3 ] in which the small lattices actually have minima in the specific heat for field values that eventually show phase transitions for sufficiently large systems .
the round - shaped size - independent peak is due to the short range ordering of the @xmath0 `` clusters '' of different orientation from the ordered background .
no corresponding behavior was observed from the susceptibility or the fourth - order cumulant . in order to confirm the above argument
, we also calculated the nn and nnn pair correlation function , that is @xmath64 , for paths of different fields crossing this line .
the correlation function data are plotted in fig .
the nn pair correlation decreases from zero to a minimum and then increases to positive values , while the nnn pair correlation increases monotonically from @xmath65 . and 0.6 across the short range ordering line.,title="fig : " ] + and 0.6 across the short range ordering line.,title="fig : " ] the data for the specific heat and susceptibility for three different values of @xmath61 are plotted in fig . [ f5 ] . ; l=120 , @xmath66 ; l=160 , @xmath67 ; l=200 @xmath68 ; l=300 @xmath69 ; l=400 @xmath70.,title="fig : " ] ; l=120 , @xmath66 ; l=160 , @xmath67 ; l=200 @xmath68 ; l=300 @xmath69 ; l=400 @xmath70.,title="fig : " ] without the field , they both show very sharp peaks , and from the magnitudes of the peak values of the specific heat , as shown in fig.[f6](b ) , we can obtain a rather accurate estimate of the exponent ratio @xmath71 , which is obviously not zero . in fig.[f6](a ) , we also show the curve - fit for the maximum slope of @xmath72 for @xmath73 , and extract the exponent @xmath74 directly . and [ eq2 ] for the maximum slopes of @xmath72 ( a ) and peak values of specific heat ( b ) , respectively.,title="fig : " ] + and [ eq2 ] for the maximum slopes of @xmath72 ( a ) and peak values of specific heat ( b ) , respectively.,title="fig : " ]
both values of @xmath53 and @xmath44 are quite consistent with the early estimates in ref @xcite , and the value of @xmath75 is different from ref @xcite , in which the @xmath76 correction term was assumed up to lattice size @xmath77 .
the same procedure was repeated for @xmath78 and @xmath79 , however , as shown in fig .
[ f5 ] , the peaks of the specific heat become increasingly rounded as the field increases , which makes it rather difficult to get a direct estimate of the exponent @xmath53 . because of this it was necessary to obtain data for much larger lattice sizes , a task that was only possible with the use of gpu computing . in fact , as the value of @xmath44 increases with the field , for @xmath80 , according the hyper - scaling law @xmath81 , where @xmath82 is the dimension of the system , @xmath53 should be negative .
although the curve - fit is not stable , given the value of @xmath53 we can get a fit within error bars .
such continuous increasing of the exponent @xmath44 up to values much greater than one is actually consistent with the findings of an early transfer - matrix study@xcite . to estimate the critical exponents @xmath46 and @xmath47 , we performed data collapsing with a large range of lattice sizes for the order parameter and its susceptibility .
as shown in fig .
[ f7 ] , the data in both finite size scaling plots collapse very nicely onto single curves , and the ratio @xmath83 and @xmath84 agree with values of the 2d ising universality class within error bars . and @xmath85 for root - mean - square order parameter and its ordering susceptibility , respectively . @xmath86 and @xmath87 .
data are for : l=80 , @xmath88 ; l=100 , @xmath89 ; l=120 , @xmath66 ; l=160 , @xmath67 ; l=200 @xmath68 ; l=300 , @xmath69 ; l=400 , @xmath70.,title="fig : " ] + and @xmath85 for root - mean - square order parameter and its ordering susceptibility , respectively . @xmath86 and @xmath87 .
data are for : l=80 , @xmath88 ; l=100 , @xmath89 ; l=120 , @xmath66 ; l=160 , @xmath67 ; l=200 @xmath68 ; l=300 , @xmath69 ; l=400 , @xmath70.,title="fig : " ] hence , although the individual exponents are non - universal , suzuki s weak universality holds quite well . another data collapsing along the path of constant @xmath90 across the phase boundary of the row - shifted @xmath91 state
is shown in fig .
the quality of the data collapsing is also excellent , and again , the exponents are non - universal .
the estimate for @xmath83 is a bit low but @xmath84 agrees well with prediction of weak universality . for root - mean - square order parameter and its ordering susceptibility , respectively .
data are for : l=80 , @xmath88 ; l=100 , @xmath89 ; l=120 , @xmath66 ; l=160 , @xmath67 ; l=200 @xmath68 ; l=300 , @xmath69 ; l=400 , @xmath70.,title="fig : " ] + for root - mean - square order parameter and its ordering susceptibility , respectively .
data are for : l=80 , @xmath88 ; l=100 , @xmath89 ; l=120 , @xmath66 ; l=160 , @xmath67 ; l=200 @xmath68 ; l=300 , @xmath69 ; l=400 , @xmath70.,title="fig : " ] in table .
[ t1 ] , the critical points and exponents @xmath92 and @xmath47 for several typical paths of constant h or t across the phase boundary are listed . .
values of critical point temperatures or magnetic fields and corresponding critical exponents for several paths of constant temperature or field across the phase boundary of the @xmath0 and * row - shifted @xmath1 ordered phases . [ cols="^,^ , < , < ,
< , < , < " , ] close to the region between the two ordered phases the correlation length exponent @xmath44 turns out to be quite large , and correspondingly the location of the critical points becomes very difficult to determine .
in addition , the specific heat curves look `` strange '' , see fig .
[ f3 ] , because the exponent @xmath75 would have a negative value with large magnitude , much larger lattice size is needed to approach to the limiting peak value .
since suzuki s weak universality seems to hold along the transition line , we fixed the values of @xmath93 and @xmath94 for the data collapsing analysis to get a better estimate of the critical point and the exponent @xmath44 . versus field @xmath61 along the path of constant @xmath95 for lattice size @xmath96 . ] for root - mean - square order parameter and its susceptibility , respectively . @xmath97 and @xmath98 .
data are for : l=32 , @xmath99 ; l=64 , @xmath100 ; l=128 , @xmath101 ; l=256 , @xmath102.,title="fig : " ] + for root - mean - square order parameter and its susceptibility , respectively . @xmath97 and @xmath98 .
data are for : l=32 , @xmath99 ; l=64 , @xmath100 ; l=128 , @xmath101 ; l=256 , @xmath102.,title="fig : " ] as shown in fig
. [ f9 ] , the crossing point of the fourth order cumulant curves for a path of constant @xmath95 is slightly above @xmath58 , and from the data collapsing analysis , see fig . [ f10 ] , we obtained an estimate of the critical point to be @xmath103 .
hence , we confirm the reentrant behavior of the @xmath0 transition line . for the paths of constant @xmath104 ,
the curves of the fourth order cumulant shows two crossing points and the finite size effect is quite obvious .
see fig .
[ f11 ] . versus field @xmath61 along the path of constant @xmath104 for lattice size @xmath105 .
] for the larger lattice size , the two crossing points move towards lower fields but they do not approach each other .
thus , a region of disorder remains between the two different ordered states , even down to quite low temperatures .
( if however , small lattices are used with insufficient data precision , it looks as though the curves for different lattice sizes coincide .
such behavior would indicate , erroneously , the presence of an xy - like region . ) in fig .
[ f12 ] , we show data collapsing fits along the path of constant @xmath106 ( and excellent data collapsing is also found along the path of constant @xmath107 ) , which confirms that there is a disordered region in between the two ordered states . for root - mean - square order parameter and its susceptibility , respectively .
data are for : l=80 , @xmath88 ; l=100 , @xmath89 ; l=120 , @xmath66 ; l=160 , @xmath67 ; l=200 @xmath68.,title="fig : " ] + for root - mean - square order parameter and its susceptibility , respectively .
data are for : l=80 , @xmath88 ; l=100 , @xmath89 ; l=120 , @xmath66 ; l=160 , @xmath67 ; l=200 @xmath68.,title="fig : " ] we thus conclude that there is no xy - like region and that the two phase boundaries probably only come together at a bicritical point at @xmath108 , although we can not exclude the possibility of a bicritical point at very low , but non - zero , temperature .
however , the data do not yield any hint of such a bicritical point ; but the lack of data points for @xmath109 in fig.2 precludes us from making a definitive statement about this issue .
( moreover , the reentrant behavior of the @xmath0 phase makes it very difficult to study the approach to @xmath108 . )
the variation of the critical exponents is consistent with the changing magnetic field producing different effective anisotropies which , in turn , is expected to yield non - universal behavior@xcite . due to the large values of @xmath44 near the bicritical point ( and correspondingly strongly negative values of @xmath53 ) , we consider it also extremely unlikely that tricritical points could be found along these transition lines as @xmath110 becomes small .
we have carried out large - scale monte carlo simulations for the square - lattice ising model with repulsive ( antiferromagnetic ) nearest- and next - nearest - neighbor interactions . from the finite size scaling analysis , both transitions from @xmath0 and row - shifted @xmath1 ordered states to disordered states turn out to be continuous and non - universal .
the reentrance behavior of the @xmath0 transition line is confirmed , and the proximity to the transition line to the @xmath1 state makes it difficult to untangle the low temperature behavior unless quite large lattices are used .
it was only possible to obtain the precise , large lattice data needed though the use of gpu computing .
no evidence for xy - like behavior is found , and we conclude that there is probably a zero temperature bicritical point . although the exponent @xmath44 varies along the transition line , the exponent ratio @xmath83 and @xmath84 seem to agree with that of the 2d ising universality class . + * acknowledgements : * we would like to thank k. binder and s .- h .
tsai for illuminating discussions and comments and also t. preis for introducing us to the use of gpu s for our calculations .
numerical computations have been performed partly at the research computing center of the university of georgia .
this work was supported by nsf grant dmr-0810223 . | using the parallel tempering algorithm and gpu accelerated techniques , we have performed large - scale monte carlo simulations of the ising model on a square lattice with antiferromagnetic ( repulsive ) nearest - neighbor(nn ) and next - nearest - neighbor(nnn ) interactions of the same strength and subject to a uniform magnetic field . both transitions from the @xmath0 and row - shifted
@xmath1 ordered phases to the paramagnetic phase are continuous . from our data analysis ,
reentrance behavior of the @xmath0 critical line and a bicritical point which separates the two ordered phases at t=0 are confirmed . based on the critical exponents we obtained along the phase boundary
, suzuki s weak universality seems to hold . |
the study of random graphs , initiated more than four decades ago , has been since an issue of major interest in probability theory and in statistical mechanics .
examples of random graphs abound @xcite , from the original erds - renyi model @xcite , where edges are chosen independently of each other between pairs of a set of @xmath0 vertices ( with a fixed probability of @xmath1 ) , to the scale - free graphs with power law degree distribution @xcite , only introduced in recent times .
an interesting and useful distribution is the one that generates random @xmath2-regular graphs @xcite .
these are uniformly drawn from the set of all graphs over @xmath0 vertices , each restricted to have degree @xmath2 .
random regular graphs can be easily generated when @xmath0 is large ( and @xmath2 finite ) @xcite , and an instance of @xmath3-regular graph is symbolized in figure [ figtree ] . around a randomly picked up vertex called _ root _ the graph looks like a regular tree .
the probability that there exists a _ circuit _ ( a self - avoiding closed path ) of length @xmath4 going through the root vanishes when @xmath0 is sent to infinity and @xmath4 is kept finite .
finite - length loops may be present in the graph , but not in extensive number . ]
@xcite .
the purpose of this article is to reach some quantitative understanding of the presence of long circuits in random graphs .
our motivation is at least two - fold .
first , improving our knowledge on circuits in random graphs would certainly have positive fall - out on the understanding of equilibrium properties of models of interacting variables living on these graphs .
for instance , frustration in spin systems emerge from the presence of circuits with odd length .
the number of these circuits is therefore directly related to the amount of glassiness present in the system at low temperature . from the dynamical point of view too , the properties of models of interacting or communicating agents e.g. routers , ... strongly depend on the topological properties of the underlying graph , and particularly upon feedbacks resulting from the presence of circuits .
these `` practical '' incentives , together with purely academic motivations have led to recent studies on the distributions of unicyclic components @xcite and cycles @xcite in random graphs .
secondly , the absence of short circuits in random graphs or , equivalently , the observation that random graphs essentially exhibit a local tree - like structure is crucial to the analytical treatment of spin models on random graphs with some common techniques , e.g. the replica or cavity methods .
these models are characterized by ( realistic ) strong interactions between spins , in contrast to usual mean - field systems , e.g. curie - weiss or sherrington kirkpatrick models . yet , the absence of short - ranged loops in random graphs make them , around any vertex ( root ) , locally similar to a tree surrounded by a remote bulk . because of that one conjectures that the free - energy of a spin model on a random graph is equal to the one on the tree with self - consistent boundary conditions requiring that appropriate properties ( for example the magnetization for the ising model ) are identical on the leaves and at the root vertex @xcite .
one of the goals of this note is to point out that this procedure _ la _ bethe amounts to make assumptions on the large - scale structure of random graphs , _
i.e. _ on the distribution of long loops . we will first discuss _ loops _ , and compute their number in the thermodynamic limit by analyzing the partition function of the ising model .
next we will compute the number of circuits by discussing the @xmath5 limit of the @xmath6 ferromagnetic model .
for gathering information about the finite @xmath0 behavior of the number of loops ( or of the loop entropy ) we will use an effective counting algorithm .
we will end this note with a discussion of our findings and of the ( many ) open issues .
in this section , we enumerate long loops in random graphs from the study of associated statistical mechanics models .
we will give our _ ad hoc _ definition of a _ loop _ in the following section .
let us consider for example ising spins @xmath7 on vertices @xmath8 of the graph of figure [ figtree ] , and ferromagnetic couplings @xmath9 on edges @xmath10 . to calculate the partition function @xmath11 of the ising model at inverse temperature @xmath12 on the graph of figure [ figtree ]
, we consider the leaves of the uncovered local tree _
i.e. _ vertices at distance @xmath13 from the root ( in figure [ figtree ] the maximum drawn distance is @xmath14 ) @xcite . at sufficiently large @xmath12 ,
a spontaneous , say , positive magnetization @xmath15 is expected to be present in the bulk . as a result
, the spins attached to the leaves will feel an external field @xmath16 .
integrating these spins out will in turn produce an external field @xmath17 acting on spins at distance @xmath18 from the root , with @xcite @xmath19\ .\ ] ] after repeated iterations of this procedure , the field at the root reaches a stationary value , @xmath20 , from which we can derive the magnetization @xmath21 $ ] and the free - energy density @xmath22 the critical inverse temperature is the smallest value of @xmath12 for which @xmath20 is non zero _ i.e. _ @xmath23 . as is well known , the high temperature expansion of the partition function @xmath24 of the ising model can be written as a sum over closed paths going through each vertex an even number of times and each edge at most once @xcite .
each path is given a weight @xmath25 depending upon its length @xmath4 . in the following
we will refer to such paths as _
loops_. gathering all loops with the same weight _
i.e. _ same length , we obtain @xmath26 where @xmath27 is the number of loops of length @xmath4 that can be drawn on the graph .
it is a sensible assumption that this multiplicity exhibits an exponential growth with the graph size , @xmath28\ , \ ] ] where @xmath29 is the intensive length of the loops , and @xmath30 is the entropy of loops having length @xmath29 .
we stress here that @xmath31 corresponds to @xmath32 .
knowledge of the entropy @xmath30 will provide us with information about large - scale loops _
i.e. _ with lengths of the order of @xmath0
. insertion of the above scaling hypothesis for @xmath27 in the partition function @xmath24 , and use of the laplace method yield @xmath33 \label{g - sig}\ ] ] in the infinite @xmath0 limit is self - averaging ( as is known and true ) but that also @xmath30 enjoys the same property . ] . in other words ,
the free - energy density @xmath34 is essentially the legendre transform of the loop entropy @xmath35 .
.4 in a2.6 in b + .2 in .2 in we show in figure [ figpaire ] our results for the loop entropy .
the entropy departs from @xmath36 with a slope equal to @xmath37 from ( [ g - sig ] ) .
its maximum is reached in @xmath38 , that is , for loops going through half of the edges .
the left part of the entropy curve ( @xmath39 ) is parametrized by inverse temperatures @xmath12 going from @xmath40 ( @xmath41
the para / ferro transition takes place when extensive loops start contributing to partition function @xmath24 ) to @xmath42 ( top of the curve at zero temperature , @xmath24 is dominated by the most numerous paths ) .
the right part of the entropy curve ( @xmath43 ) is obtained when @xmath44 , that is , for inverse temperatures with an imaginary part equal to @xmath45 allowing to transform the hyperbolic tangent of @xmath12 into the cotangent of its real part .
this is done by considering a real @xmath46 , the stationary field @xmath20 for this temperature , and the transformation @xmath47 , @xmath48 . _
the case of even degree . _ for even @xmath2 @xmath30 is left unchanged under the transformation @xmath49 .
the right part of the entropy curve is thus the mirror symmetric of the left part with respect to @xmath50 ( figure [ figpaire]a ) .
this results from a duality between long and short extensive loops .
let @xmath51 be the full loop ( all edges occupied , which is allowed for even @xmath2 ) .
then , for any loop @xmath52 with length @xmath53 , @xmath54 ( that is @xmath55 without the edges belonging to @xmath52 ) is a loop with length @xmath56 .
notice that the largest loop is @xmath57 , with length @xmath58 .
_ the case of odd degree .
_ here duality does not hold and there is no simple transformation rule for @xmath29 and @xmath30 .
an explicit calculation gives the entropy curve shown in figure [ figpaire]b .
the maximal length @xmath59 is reached with an infinite slope , and corresponds to a finite entropy @xmath60 . for odd degrees @xmath2 ,
indeed , loops can not occupy all edges since the number of incoming edges onto each vertex is even .
the longest loops have one free edge per vertex : these act as defects , the positions of which can be chosen with some freedom , giving rise to a finite entropy .
the larger is @xmath2 , the smaller is the ratio @xmath61 : intuitively , the frustration coming from the parity of @xmath2 is less and less important as @xmath2 increases indicate that the replica symmetric solution is correct in this range of ( complex - valued ) temperature @xcite . ] .
the iteration procedure can be applied to calculate the free - energy of other short - range models and then derive the entropies corresponding to large - scale diagrams generated through high temperature expansions .
for instance the ferromagnetic @xmath6 model with @xmath62 gives information on circuits _
i.e _ loops with vertices having degree 2 at most @xcite .
a spin @xmath63 is submitted to two fields @xmath64 conjugated to the magnetization @xmath65 and its squared value limit . ] .
the iteration equations for these fields read @xmath66 inserting the fixed point values for the fields in the expression of the free - energy per spin component , @xmath67 - \frac{c}{2\beta } \ln \big [ ( 1+h_2)^2 + \beta h_1 ^ 2 \big ] \ , \ ] ] yields ( for an alternative derivation , see @xcite ) , @xmath68 + \frac{c}{2\beta } \ln \big [ ( c-1)\beta\big ] \ .\ ] ] the entropy of circuits , @xmath69 is obtained through @xmath70 $ ] .
results are shown in figure [ figpaire ] .
the maximal entropy , @xmath71 , is reached in @xmath72 .
the rightmost point , with coordinates @xmath73 corresponds to hamiltonian cycles .
expression ( [ mc ] ) coincides with the output of rigorous calculations @xcite , which shows the exactness of the replica symmetric hypothesis for the @xmath74 model @xcite .
note the similarity of the entropy curves for loops on odd @xmath2-regular graphs and circuits ( figure [ figpaire ] ) .
in the two cases the slopes at minimal and maximal lengths are respectively finite and infinite , see the discussion above .
let @xmath75 and @xmath27 be in all the rest of this note the number of circuits of length @xmath4 in a random , @xmath2-regular graph . for finite @xmath4 , @xmath27 is asymptotically poisson - distributed when @xmath76 @xcite , @xmath77 = \frac1{m ! } \left [ \frac{\left(c-1\right)^l}{2l } \right]^{m } \exp\left\{- \frac{\left(c-1\right)^l}{2l } \right\}\ .\ ] ] the above identity holds indeed for circuit - length @xmath78 .
the expected number of circuits of intensive length @xmath79 is therefore , for @xmath80 , @xmath81 with @xmath82 and @xmath83 .
note that this value of @xmath84 coincides with expression ( [ mc ] ) at the first order in @xmath29 .
to study what happens at finite @xmath0 we have implemented a fast algorithm for finding all circuits in a given graph @xcite .
it is important that we explicitly _ find _ , not only count , all the circuits so that by our method we can in principle give all interesting characterizations . in our enumeration
we first generate a random graph and then count the circuits .
we average this outcome over a number of samples .
we describe the algorithm for enumeration in next section . to generate a fixed connectivity random graph we start by assuming that each site has @xmath2 connections that connect it to @xmath2 different sites : we allow no self - connections and no double edges . at the beginning , all connections are free .
two of the connections are extracted and matched together .
we continue filling them up ( we use a table that we resize at each step to keep the process effective ) till all connections are set or we are stuck because there are two left connections that both belong to the same site or that belong to two sites that are already connected . to be sure not to introduce any systematic bias in this case we just discard the full graph and restart the procedure from scratch . for enumerating circuits we have implemented and used an algorithm introduced by johnson @xcite .
the algorithm finds all elementary circuits of a graph .
the computer time needed for this task is bounded by @xmath85 , where @xmath0 is the number of vertices of the graph , @xmath86 the number of edges and @xmath87 the total number of circuits in the graph : indeed one proves that the time used between the output of two consecutive circuits is bound by @xmath88 ( and that this is true also for time elapse before the output of the first circuit and after the output of the last one ) .
the memory needed by the algorithm is bounded by @xmath88 .
let us just sketch the crucial steps of the procedure .
one first orders the vertices in some lexicographic sequence , and labels them with integers ranging from @xmath89 to @xmath90 .
the search is started from a _ root _
vertex @xmath91 , in the subgraph induced by @xmath91 and by vertices after @xmath91 : one of the crucial performance issues is that all vertices that become roots are the smallest vertex in at least one elementary circuit .
the input to the procedure is the adjacency structure @xmath92 for each vertex @xmath93 , that represents the graph : it contains @xmath94 if and only if @xmath95 , where @xmath96 is the set of edges of the graph .
we _ block _ a vertex @xmath93 when it is added to a path beginning in @xmath91 .
we build elementary paths starting from @xmath91 .
the vertices of the current trial paths are loaded on a stack .
a procedure adds the vertex to the path , if appropriate , and appends the vertex to the stack : the vertex is deleted from the stack when returning from this procedure .
the ingenious part of the algorithm is in keeping a vertex _ blocked _ as long as possible , to minimize the execution time .
this has to be done while keeping the procedure correct : the basic rule that has to be satisfied to guarantee that all circuits are found ( only once ) is that if it exists a path from the vertex @xmath93 to @xmath91 that does not intersect the path loaded on the stack , then @xmath93 has to be free ( i.e. it can not be in a _ blocked _ state ) .
one uses a list to keep the information needed to satisfy this constraint while staying time effective .
some details about performances are as follows : on an intel xeon @xmath97 @xmath98 processor our implementation takes of the order of @xmath99 seconds for finding all circuits of a @xmath100 graph ( they are @xmath101 ) , @xmath102 for @xmath103 ( @xmath104 circuits ) and @xmath105 for @xmath106 ( with @xmath107 circuits ) . in order to test the results of our code
we have also implemented a `` quick and dirty '' backtracking procedure to count hamiltonian circuits .
since our procedure crucially depends on the quality of the random number generator we have also checked that different ( high quality ) random number generators lead to statistically compatible answers . thanks to our algorithm and implementation
we have been able to enumerate of the order of @xmath108 circuits ( a large number ) . for small @xmath2 values
we can study larger graphs ( we have analyzed graphs with up to @xmath109 vertices in the @xmath110 case and up to @xmath111 vertices for @xmath112 , and averaged our results over samples ranging from @xmath113 to @xmath114 random graphs ) .
typically we find for example of the order of @xmath115 million circuits on a @xmath116 , @xmath110 graph , one billion circuits on a @xmath117 , @xmath118 graph and @xmath119 billion circuits on a @xmath120 , @xmath112 graph .
for each value of @xmath0 , we average over of the order of @xmath114 samples for all the @xmath110 enumerations , and @xmath113 graphs for @xmath121 . in figure [ figallc3 ]
we plot the theoretical results obtained in the ( replica symmetric ) calculation of section 2.2 for @xmath110 , together with results obtained by finite enumeration on finite lattices with @xmath0 vertices .
numerical data from finite graphs are very slowly approaching the theoretical , infinite volume data .
we try in the following to analyze the quality of this asymptotic agreement of the two sets of data . at first we know that @xmath122 as @xmath123 where @xmath124 is the average number of circuits of length @xmath125 . with our numerical data
we can not work really close to @xmath41 , since for finite @xmath0 we have a value @xmath126 that goes to zero as @xmath127 : finite size effects appear as a flattening of @xmath128 when @xmath29 becomes very close to the minimum allowed value ( see figure [ figallc3 ] ) .
one can easily see by eye that the slope is very similar to the asymptotic slope in the small @xmath29 region where we are relatively safe from finite size effects .
we have fitted a linear behavior ( that is indeed clear in the data ) for example for @xmath29 in the range @xmath129 for @xmath110 . using this approach we find for @xmath2 from 3 to 6 slopes about @xmath130 smaller than the theoretical prediction ( on the larger graphs we can study ) : for @xmath110 we find @xmath131 versus a theoretical @xmath132 ; for @xmath133 we find @xmath134 versus @xmath135 ; for @xmath118 we find @xmath136 versus @xmath137 ; for @xmath112 we find @xmath138 versus @xmath139 .
finite size effect can be drastically reduced if we compare directly different @xmath2 value ( since we are using graphs of a size that is limited by the same criterion for each @xmath2 value ) .
for example the ratio of the slope of @xmath2 and @xmath140 is @xmath141 for @xmath110 versus a theoretical @xmath142 , @xmath143 versus @xmath144 for @xmath133 and @xmath145 versus @xmath146 for @xmath118 .
this really remarkable agreement gives us confidence that we have a good control over finite size effects .
equation ( [ form ] ) tells us that the difference between the measured circuit entropy @xmath147 and the asymptotic value given by formula ( [ mc ] ) , once multiplied by @xmath0 , behaves as @xmath148 for vanishing small values of @xmath29 , with @xmath149 .
it is therefore independent of @xmath2 , with a logarithmic dependence upon the graph size @xmath0 . to check how equation ( [ mc1 ] ) applies to finite values of @xmath29
, we look first in figure [ figallcl05 ] at the number of circuits with @xmath150 , for different values of @xmath0 and @xmath2 ( we have harvested our most precise data at @xmath110 ) .
data indeed show only a very weak dependence upon @xmath2 , and this dependence becomes weaker with increasing @xmath2 .
data for @xmath118 are already indistinguishable from the ones for @xmath112 . in figure [ figallcl1 ]
, we show the same quantity for @xmath151 _ i.e. _ for hamiltonian circuits , that pass through all vertices of the random graph .
as far as the scaling with @xmath2 is concerned figure [ figallcl1 ] shows that the scaling of hamiltonian circuits is excellent already at @xmath110 .
we will come back later about the fact that scaling property of hamiltonian circuits turn out to be very different from the ones of all other finite @xmath29 , less dense circuits . inspired by the result obtained for small circuits ( [ form]),([mc1 ] )
we look for the following fit for the circuit entropy for finite values of @xmath29 , @xmath152 in figure [ figfit ] we show our results for @xmath110 , @xmath150 .
the quality of the best fit to data with sizes @xmath153 only is excellent , and is very good agreement with all data with @xmath154 .
the two parameter fit is clearly superior to power law fits .
we find that with very good accuracy ( surely better than one percent ) @xmath155 i.e. that even at finite @xmath29 the relation ( [ mc1 ] ) gives the correct leading corrections .
this result tells us that for all @xmath29 values ( maybe excluding @xmath151 , see later ) we have that the average number of circuits of reduced length @xmath29 equals @xmath156 where @xmath157 is a bounded function of @xmath29 . for @xmath158 , we find values close to @xmath159 e.g. @xmath160 in the case of @xmath150 . here
precision is not as good since this is a sub - leading correction .
what is clear from our data is that sub - leading corrections to the circuit entropy are of the order of @xmath161 as encoded in formula ( [ mc2 ] ) . as we have said above the case of hamiltonian circuits ( @xmath151 )
is exceptional .
finite size transient effects are very strong ; this is intuitively expected since these circuits fill the graph and are deeply affected by its finite size .
we show in figure [ fignofit ] the analogous of figure [ figfit ] but for @xmath151 .
it is clear that the structure of finite size effects is completely different . on the contrary
we have already explained that we find exactly the same behavior for all intermediate @xmath29 values : the case @xmath151 appears as isolated .
to sum up , the absence of small loops in random graphs allows one to argue that the free - energy of a spin model defined on the graph is equal to the one on a regular tree with a self - consistent external field at boundary ( leaves ) . in turn , this free - energy fully determines the distribution of large - scale loops in the random graph .
we stress that the whole complexity of this cavity procedure is actually hidden in the assumption of self - consistency for the field(s ) @xcite . assumptions corresponding to broken symmetry in the replica framework
@xcite may be necessary for models with random interactions .
anyhow , the salient feature of the above standard approach is that it predicts global thermodynamical quantities from local considerations about the graph only .
we have added to our exact computation , valid in the @xmath162 limit , results from exact enumeration at finite @xmath0 .
thanks to them we have been able to determine precisely the behavior of the leading corrections to the thermodynamical behavior ( at least for circuits with @xmath163 : we have found that hamiltonian circuits have stronger finite size corrections and a peculiar finite @xmath0 behavior
. it would be very interesting to characterize properties or quantities that can adequately be described by a local procedure @xcite .
the question is however well beyond the scope of the present paper , which only shows that large loops are ( remarkably ) among those ones .
we are grateful to g. parisi for insights about the validity of replica symmetry for the ising model with complex temperature .
we also thank s. cocco , a. montanari for interesting discussions .
thanks the les houches summer school and the smc - infm center in rome for the hospitality in august and september 2003 respectively .
thanks the `` laboratoire de physique thorique '' of ens in paris for a visiting professorship during may - june 2004 .
wormald _ models of random regular graphs _ in _ surveys in combinatorics _ , edited by j.d .
lamb and d.a .
preece , london mathematical society lecture note series * 276 * pp .
239298 ( cambridge university press , cambridge , uk , 1999 ) | we compute the number of circuits and of loops with multiple crossings in random regular graphs .
we discuss the importance of this issue for the validity of the cavity approach . on the one side we obtain analytic results for the infinite volume limit in agreement with existing exact results . on the other side
we implement a counting algorithm , enumerate circuits at finite @xmath0 and draw some general conclusions about the finite @xmath0 behavior of the circuits . |
we address here the following question : what is the asymptotic size and the limit laws of the smallest and largest gaps or spacings in the spectra of random matrices ?
typical spacings between eigenvalues of random matrices have been well understood for invariant ensembles for quite some time .
more recently , the behavior of these typical spacings has even been proved to be universal for much larger classes of random matrices ( @xcite , @xcite , @xcite ) .
much less is known for atypically large or small spacings .
this question was first considered for the smallest spacings in the unpublished phd thesis of j.vinson ( @xcite ) and raised by persi diaconis in @xcite for the largest ones .
it was also discussed in an interesting debate during a conference at the courant institute in 2006 , in honor of percy deift .
we solve here completely the question of the smallest spacings for the simplest invariant ensembles , i.e the cue and the gue .
we give the scaling and the limit laws for the joint distribution of the smallest spacings .
the answer is simple to state : it is given by the trivial poissonian ansatz where the spacings would be treated as i.i.d random variables .
the answer we find for the largest spacings is less complete , since we can only obtain at this point a first - order approximation which gives the asymptotic size of the largest spacings and not their limit laws .
we believe that the same poissonian ansatz should work as well for the largest gaps , but this question is left open .
the question of the universality of the behavior of the extremal spacings is also left open .
dyson @xcite showed that the repulsion between eigenvalues of the gaussian unitary ensemble ( the gue ) could be described asymptotically in terms of the _ determinantal _ point process associated with the sine kernel @xmath6 for this limiting determinantal point process , the probability of having no eigenvalue in an interval of size @xmath7 is known to be the fredholm determinant @xmath8 where @xmath9 is the convolution operator acting on @xmath10 with kernel @xmath11 .
the density of the spacing between two successive points is then given by ( see @xcite ) @xmath12 this spacing distribution was shown to appear in a number - theoretic context : some statistics produced by odlyzko @xcite presented a correspondence between the histogram of normalized gaps between zeros of the riemann zeta function and @xmath13 .
this gave further evidence for the analogy discovered by dyson and montgomery : they realized that the local dependence of the zeros of @xmath14 , previously calculated by montgomery , involved the sine kernel ( see @xcite , @xcite for an historical account and other steps of this fruitful analogy ) . at the same mean or typical gap scale , a precise analysis of the joint distribution of the gaps between eigenvalues was performed by katz - sarnak @xcite and soshnikov @xcite , for the circular unitary ensemble ( the cue ) .
less attention was paid to eigenvalues statistics at smaller and larger scales
. this paper concerns the extreme gaps .
this study was initiated by j. vinson @xcite : he showed that the smallest gap between elements of the @xmath15 , multiplied by @xmath1 , converges in law to a random variable with distribution function @xmath16 , as the size @xmath17 of the unitary matrix increases . in his thesis ,
similar results for the smallest gap between eigenvalues of a generalization of the @xmath18 were obtained .
vinson also gives interesting heuristics suggesting that the largest gap between @xmath15 eigenvalues should be of order @xmath19 , with poissonian fluctuations around this limit . using a different technique ,
soshnikov @xcite investigated the smallest gaps for determinantal point processes on the real line , with a translation invariant kernel : amongst points included in @xmath20 $ ] , this extreme spacing multiplied by @xmath21 converges weakly to the distribution with distribution function @xmath16 , as @xmath22 .
heuristically , the above extreme gaps asymptotics can be obtained using the known asymptotics @xcite of the spacing distribution @xmath23 and treating the gaps as independent random variables .
the difficulty to obtain rigorous results lies in showing that this poissonian ansatz is asymptotically correct for the extreme gaps , and in making the above estimates uniform in the dimension @xmath17 .
we first consider the _ joint _ law of the smallest gaps ( theorem [ thm : unitary ] , corollary [ cor : smallestunitary ] ) between eigenvalues of unitary matrices .
this relies both on soshnikov s method and a convergence of the process of small gaps to a poisson point process .
the same reasoning equally applies to the small gaps between eigenvalues from the gaussian unitary ensemble ( theorem [ thm : g ] , corollary [ cor : smallestgue ] ) .
the proofs of the small spacings asymptotics are in section 2 .
the first order asymptotics of the largest gaps is then proved . concerning unitary random matrices ( theorem [ thm : largestgaps ] )
, this makes use of two important tools .
a key ingredient , by deift & al @xcite , is the uniform asymptotics about the probability for a given arc of the circle to be free of eigenvalues .
the proof also requires the negative correlation property for the event that two disjoint arcs are free of eigenangles . on account of the @xmath18 ( theorem [ thm : largestgapsgue ] )
, we also make a essential use of the negative correlation , and the large gap probability is evaluated by comparing the @xmath18 fredholm determinant with the unitary one .
these large gaps asymptotics are proved in section 3 .
the extreme spacings between random eigenvalues are important quantities for statistical physics , computational mathematics and number theory .
for this reason , diaconis @xcite mentions the open question of maximal spacings , answered in theorem [ thm : largestgaps ] . after making our results explicit , successively for unitary matrices and the gue ensemble , we give applications of our extremal spacings statistics at the end of this introduction .
let @xmath24 , a haar - distributed ( measure @xmath25 ) unitary matrix over @xmath26 .
suppose @xmath24 has eigenvalues @xmath27 s , with ordered eigenangles @xmath28 consider the point process on @xmath29 @xmath30 our first result is about the convergence of @xmath31 to a poisson point process , thanks to this normalization by @xmath32 .
[ thm : unitary ] suppose @xmath33 . as @xmath34 , the process @xmath31 converges to a poisson point process @xmath35 with intensity @xmath36 for any bounded borel sets @xmath37 and @xmath38 .
the intensity is proportional to @xmath39 because of the rotational invariance of the haar measure .
the corresponding factor will be less trivial in the case of the @xmath18 ensemble .
our method to prove theorem [ thm : unitary ] relies on the _ s - modified _ random point field technique initiated by soshnikov @xcite : one can calculate the correlation functions of the process obtained by keeping only the @xmath40 s for which @xmath41 contains exactly one other eigenvalue .
contrary to @xcite , we do not use the notion of cluster functions , because we characterize the convergence to poisson random variables thanks to the convergence of the factorial moments ; this allows also to consider easily non translation invariant kernels , like in the @xmath18 case . moreover , theorem [ thm : unitary ] gives information about the joint distribution of the number of gaps taking values in disjoint intervals ( convergence in terms of point processes ) .
in particular we can compute the limiting joint law of the smallest gaps .
let @xmath42 be the @xmath0 smallest eigenangles gaps ( i.e. of the form @xmath43 , where the indexes are modulo @xmath17 and gaps are defined as small as possible , i.e. @xmath44 $ ] ) .
for the sake of brevity , write @xmath45 the limiting joint law of the @xmath46 s is a corollary of theorem [ thm : unitary ] .
[ cor : smallestunitary ] for any @xmath47 , under the haar measure on @xmath48 , @xmath49 in particular , the @xmath50 smallest normalized space @xmath51 converges in law to @xmath46 , with density @xmath52 note that this result , for @xmath53 , is proved in vinson s thesis @xcite by a different method : he characterizes the number of small gaps as a symmetric function of the eigenvalues , and computes its moments .
it is not clear how his method can be extended to provide the joint law of the @xmath0 smallest gaps .
+ we now turn to our next question about extreme gaps , i.e the asymptotic behavior of the largest gaps , which were guessed by vinson , based on the supposed asymptotic independence of distant gaps .
we obtain thanks to the precise asymptotics of one gap probability , obtained by the steepest descent method for riemann hilbert problems in @xcite , and the negative association property of determinantal point processes ( see e.g. @xcite ) .
note that both results are posterior to vinson s thesis .
consider @xmath54 the largest gaps between successive eigenangles of @xmath55 , i.e. of the form @xmath56 , where the indexes are modulo @xmath17 and @xmath57 .
then , as @xmath17 goes to infinity , the largest gap converges in @xmath58 to a constant , for any @xmath5 : @xmath59 actually the above limit holds for all the @xmath60 largest gaps if @xmath61 is subpolynomial .
[ thm : largestgaps ] let @xmath62 be positive integers .
then for any @xmath5 , @xmath63 as @xmath34 , @xmath64 the problem of the exact fluctuations will be addressed in a future work . ]
. note that , for independent uniform eigenangles on the unit circle , the largest gap is of order @xmath65 , more than in the above theorem , as expected from the repulsion of the eigenvalues in the determinantal case .
similar results hold for the @xmath18 . for this ensemble ,
the distribution of the eigenvalues has density @xmath66 with respect to the lebesgue product measure , on the simplex @xmath67 .
the empirical spectral distribution @xmath68 converges in probability to the semicircle law ( see e.g. @xcite ) @xmath69 like for the unitary group , we first consider the smallest gaps , studying the point process @xmath70 for any arbitrarily small fixed @xmath71 ( this is a technical restriction allowing the use of the plancherel - rotach asymptotics of the hermite polynomials ) .
[ thm : g ] as @xmath34 , the process @xmath72 converges to a poisson point @xmath73 process with intensity @xmath74 for any bounded borel sets @xmath75 and @xmath76 .
the following corollary about the smallest gaps is an easy consequence of the previous theorem . as
for the unitary group , introduce @xmath77 the @xmath0 nearest spacings in @xmath78 , i.e. of the form @xmath79 , @xmath80 , with @xmath81 , @xmath82 $ ] , @xmath83 .
let @xmath84 [ cor : smallestgue ] for any @xmath47 , with the above notations and for the @xmath18 ensemble measure ( [ eqn : gue ] ) , @xmath85 in particular , the @xmath50 smallest normalized space @xmath86 converges in law to @xmath46 , with density @xmath52 the above result gives for @xmath53 the probability density of the normalized smallest gap in a given interval @xmath78 : @xmath87 .
the following precises the distribution of this smallest spacing location in the bulk of the spectrum .
[ cor : smallestlocation ] let @xmath88 be the index of the smallest gap between eigenvalues of the @xmath18 in a compact subset @xmath89 with nonempty interior : @xmath90 as @xmath34 , @xmath91 converges weakly to the probability measure with density proportional to @xmath92 we now turn to the largest gaps for the @xmath18 ensemble .
the result is completely different inside the bulk and on the edge . indeed , for eigenvalues strictly inside the support of the limiting measure , the maximal spacings have order @xmath19 ( see the following theorem [ thm : largestgapsgue ] ) ,
while the eigenvalues on the border have an average distance of higher order , @xmath93 : for any @xmath0 , @xmath94 converges weakly as @xmath34 to a multivariate tracy - widom distribution ( see e.g. @xcite ) .
strictly inside the bulk , the result is analogous to the circular case , the only difference being the normalization , due to the average density of eigenvalues .
let @xmath95 be the largest gaps of type @xmath96 with @xmath97 , a compact subset of @xmath98 with nonempty interior .
[ thm : largestgapsgue ] let @xmath99 be positive integers .
then for any @xmath5 @xmath100 when seen in a window of size proportional to the average gap , the spacings between the zeros of dirichlet l - functions are distributed like particles of a determinantal point process with sine kernel , this is the montgomery - odlyzko law . here
we want to discuss the accuracy of this analogy when looking at rare events , the extreme gaps between the zeta zeros , relying on theorems [ thm : unitary ] and [ thm : largestgaps ] . due to the availability of many numerical data , we focus on the riemann zeta function @xmath101 ( @xmath102 ) , which admits an analytic continuation to @xmath103 .
let @xmath104 be the non trivial zeta zeros , @xmath105 , with @xmath106 then @xmath107 has an average value 1 .
the quantity @xmath108 has been widely studied .
conditionally to the generalized riemann hypothesis , the best known result is @xmath109 @xcite . from the @xmath18 hypothesis for the zeta zeros , it is expected that @xmath110 . however , to the best of our knowledge , more precise conjectures about the growth speed of large gaps between zeta zeros were not proposed . from theorem [ thm : largestgaps ] , amongst @xmath17 successive gaps with fermionic repulsion , the maximal gap has size about @xmath111 times the average gap , suggesting @xmath112 in particular @xmath113 odlyzko s numerical data @xcite give 3.303 for the maximal value of @xmath114 , @xmath115 , while @xmath116 , giving a difference of @xmath117 with the observed gaps .
r0.5 [ fig : maximal ] further tests can be performed at distinct heights along the critical axis , thanks to numerical data of gourdon @xcite : he computed @xmath118 successive zeta zeros at height @xmath119 along the critical axis for each @xmath120 .
the extreme normalized gaps are given on the joint graph , where the expectation from our random matrices result is the straight line @xmath121 .
for example , amongst the @xmath122 gaps following the height @xmath123 , @xmath124 , i.e. a difference of @xmath125 with the expected value .
concerning the smallest gaps , does the poisson intensity @xmath126 from theorem [ thm : unitary ] appear in the context of the zeta zeros ?
note @xmath127 the normalized gaps . we know that , as @xmath34 , the set of gaps @xmath128 converges weakly to a poisson point process with intensity @xmath126 .
we therefore expect that , as @xmath34 , @xmath129 converges to a poisson point process with the same intensity .
r0.5 the joint graph gives the histogram of the @xmath130 smallest gaps , normalized as previously , amongst the @xmath131 first zeta zeros , based again on numerical data from @xcite .
more precisely , the histogram gives the 3000 smallest values of @xmath132 .
the straight line is the function @xmath133 ( the step of the histogram is 5 ) .
this presents a good relevance of the @xmath18 hypothesis for the riemann zeta function , even at the scale of rare events , here the extreme spacings .
the most classical method to diagonalize a matrix is the well - known qr algorithm . in the case of hermitian matrices ,
an alternative approach was proposed by deift & al @xcite , based on the isospectral property of the toda flow .
more precisely , given a @xmath134 hermitian matrix @xmath135 , the first step is to reduce it in a tridiagonal form @xmath136 ( which is a robust and fast operation ) , conjugating with successive householder reflections , @xmath137,\ ] ] keeping the same spectrum as @xmath135 .
after such a tridiagonalization , the @xmath138 and @xmath139 s are real .
the above matrix is important in the analysis of the toda lattice : flaschka , hnon , and manakov proved independently in the 70 s that the following evolution of @xmath17 particles on a line ( @xmath140 , @xmath141 , @xmath142 ) , @xmath143 is an integrable system .
more precisely , after the change of variables @xmath144 the differential equation takes the lax pair form @xmath145 where @xmath146.\ ] ] in particular and importantly , the spectrum of @xmath147 does not depend on time .
moser proved that @xmath148 , @xmath149 as @xmath150 , with @xmath67 .
this implies that @xmath151 converges to @xmath152 , hence @xmath136 converges to a diagonal matrix , whose entries give the eigenvalues of @xmath153 .
deift @xcite asked about the speed of convergence of the toda flow till its equilibrium .
more precisely , for a given @xmath154 , what is the necessary time @xmath155 such that @xmath156 for all @xmath157 ? as @xmath158 the speed of convergence to the spectrum is governed by the minimal gap between eigenvalues . a good choice for a _ typical _ hermitian matrix is a matrix from the @xmath18 , with independent ( up to symmetry ) complex gaussian entries ( of variance 2 on the diagonal , 1 elsewhere ) . from corollary [ cor : smallestgue ] , the minimal gap for such matrices is of order @xmath159 for a given required precision @xmath160 , the toda flow time necessary to evaluate the eigenvalues is expected to grow as @xmath161 with the dimension as a function of @xmath17 ] .
we first review some results about convergence of processes , here with values in @xmath29 and a finite number of atoms .
a point process @xmath163 is said to converge in distribution to @xmath164 is for any bounded continuous function @xmath165 @xmath166 to show the convergence of @xmath31 to a poisson point process @xmath35 , we only need to show the convergence in law of @xmath167 to @xmath168 for all bounded intervals @xmath169 and @xmath78 , the independence for disjoint @xmath170 s being an automatic consequence : this is a very practical property of poisson point processes , detailed in proposition [ prop : facmom ] below .
moreover , the convergence of @xmath167 will be shown thanks to the convergence of the factorial moments to those of a poisson random variable .
this is particularly adapted to our situation because as we will see the correlation functions of point processes are defined through factorial moments , and are explicit in the case of determinantal point processes .
note that this is the same technique employed in @xcite , where the following result is given , in the case of point processes with values in @xmath171 .
[ prop : facmom ] let @xmath163 be a sequence of point processes on @xmath29 , and @xmath35 a poisson point process on @xmath29 with intensity @xmath172 having no atoms ( and @xmath173-finite ) .
assume that for any bounded intervals @xmath169 and @xmath78 and all positive integers @xmath174 @xmath175 then the sequence of point processes @xmath31 converges in distribution @xmath35 .
we need to check the three conditions of the following theorem [ thm : kallenberg ] by kallenberg , which is written here in the specific case @xmath176 , @xmath177 and @xmath178 the set of compact rectangles @xmath170 ( @xmath169 and @xmath78 intervals ) , with the notations from @xcite .
conditions ( 1 ) , ( 2 ) and ( 3 ) will be verified if @xmath179 indeed , if ( [ eqn : convlaw ] ) holds , for any @xmath155 @xmath180 and as @xmath181 is almost surely finite this goes to @xmath152 as @xmath150 ( is is a consequence of the dominated convergence theorem ) ; this proves part ( 1 ) .
equations ( 2 ) and ( 3 ) are also consequences of the above convergence in law .
to prove ( [ eqn : convlaw ] ) , as @xmath181 is a poisson random variable , the convergence of all the moments is sufficient .
a moment is a finite linear combination of factorial moments , which concludes the proof .
[ thm : kallenberg ] let @xmath35 be a point process on @xmath29 and assume @xmath35 is almost surely simple ( i.e. the atoms of the measure @xmath35 all have weight 1 almost surely ) .
then @xmath31 converges weakly to @xmath35 if and only if the three following conditions are satisfied for any compact intervals @xmath169 and @xmath78 in @xmath171 : 1 . @xmath182 ; 2 .
@xmath183 ; 3 .
@xmath184 .
references for the main properties of correlation functions of determinantal point processes are @xcite and @xcite .
we follow this last survey to present the notions used in the following .
if @xmath185 is a simple point process on a complete separate metric space @xmath186 , consider the point process @xmath187 on @xmath188 .
one can define this way a measure @xmath189 on @xmath188 by @xmath190 for any borel set @xmath191 in @xmath188 .
most of the time , there is a natural measure @xmath192 on @xmath186 , in our cases @xmath193 or @xmath194 and @xmath192 is the lebesgue measure .
if @xmath195 is absolutely continuous with respect to @xmath196 , there exists a function @xmath197 on @xmath188 such that for any borel sets @xmath198 in @xmath186 @xmath199 hence one can think about @xmath200 as the asymptotic ( normalized ) probability of having exactly one particle in neighborhoods of the @xmath201 s .
more precisely under suitable smoothness assumptions , and for distinct points @xmath202 in @xmath193 , @xmath203 note that @xmath197 is not a probability density
. moreover specifically , if @xmath204 are disjoint in @xmath186 and @xmath205 , @xmath206 see @xcite for a proof .
in particular , @xmath207 note the analogy with formula ( [ eqn : facmompoisson ] ) we want to prove . for unitary matrices or the @xmath18 ensemble
, our method to prove convergence of small spacings counting measures is the same : * for given compact intervals @xmath169 and @xmath78 consider the modified process obtained from @xmath208 by keeping only the points @xmath209 in @xmath78 such that @xmath210 . *
show that the correlation function @xmath211 of this new process uniformly converges to @xmath212 .
this is possible thanks to the determinantal aspect of @xmath213 and the hadamard - fischer inequality , lemma [ lem : fischer ] . *
conclude that the factorial moments converge to those of the expected poisson random variables thanks to ( [ eqn : facmompoisson ] ) and ( [ eqn : facmomcorrelation ] ) for the smallest gaps asymptotics , the following inequality will be repeatedly used .
a concise proof can be found in @xcite .
[ lem : fischer ] let @xmath135 be a positive - definite @xmath134 ( hermitian ) matrix . for any @xmath214 , let @xmath215 ( resp .
@xmath216 ) be the submatrix of @xmath135 using rows and columns numbered in @xmath217 ( resp @xmath218 ) .
then @xmath219 we begin with the proof of theorem [ thm : unitary ] . we know
that for a unitary matrix @xmath33 , the density of the eigenangles @xmath220 , with respect to the lebesgue measure on the corresponding simplex is @xmath221 moreover , a remarkable fact about the point process @xmath222 is that it is determinantal : all its correlation functions @xmath223 , @xmath142 are determinants based on the same kernel : @xmath224 this classical property relies on gaudin s lemma , see @xcite . in the following for any bounded interval @xmath37 we write @xmath225 .
we want to show that for an interval @xmath38 @xmath226 with @xmath227 .
we consider the point process @xmath228 and its thinning @xmath229 obtained from @xmath213 by only keeping the eigenangles @xmath40 for which @xmath230 .
the following lemma means that @xmath231 is properly estimated by @xmath232 .
it is analogous to lemma 3 in @xcite .
[ lem : nosucsmall ] for any interval @xmath38 , as @xmath34 , @xmath233 .
let @xmath234 be such that @xmath235 , and @xmath236 .
if @xmath237 , then @xmath238 .
hence @xmath239 where the last inequality comes from the definition ( [ eqn : multdimpcess ] ) , where @xmath191 is the set of points @xmath240 with @xmath241 and @xmath242 . to prove that this positive random variable converges in law to @xmath152 , we consider its expectation @xmath243 and show it goes to 0 .
thanks to the multilinearity of the determinant , @xmath244 as @xmath245 and @xmath246 , the first column of this determinant is @xmath247 and the two others are @xmath248 . thus @xmath249 the integration domain is @xmath250 , concluding the proof
let @xmath251 , @xmath252 , be the correlation functions of the point process @xmath229 .
if , for any @xmath174 , the convergence of the factorial moment @xmath253 can be shown , then theorem [ thm : unitary ] will be proved , thanks to the above lemma [ lem : nosucsmall ] .
the way to show ( [ eqn : convfacmom ] ) relies on three steps , to apply a simple dominated convergence argument : * if all @xmath40 s are distinct @xmath251 converges to @xmath254 as @xmath34 ( lemma [ lem : simpleconvergence ] ) ; * in the set @xmath255 @xmath251 is uniformly bounded ( lemma [ lem : uniformbounded ] ) ; * even if @xmath251 is not uniformly bounded in the complement of @xmath256 in @xmath257 ( @xmath258 ) , the contribution to the integral is negligible because the volume of @xmath258 decreases sufficiently fast ( lemma [ lem : neglectset ] ) .
[ lem : simpleconvergence ] let @xmath259 be distinct elements in @xmath257 .
then @xmath260 first note that , as all the @xmath40 s are distinct , for sufficiently large @xmath17 the point @xmath261 is in @xmath256 ( see ( [ eqn : defomega ] ) ) .
this means that if @xmath259 are points of @xmath229 , the point in each of the @xmath262 is not another one of the @xmath263 s .
this makes the combinatorics easy : the correlation functions of @xmath229 can be explicitly given in terms of those of @xmath213 , as noted in @xcite , by an inclusion - exclusion argument : for sufficiently large @xmath17 @xmath264 note that there is no convergence issue here as @xmath265 if @xmath266 .
we first show that the term corresponding to @xmath267 in the above sum gives the expected asymptotics .
the determinantal aspect of the process makes things easy : @xmath268 is a @xmath269 determinant , and only the terms in the @xmath270 diagonal blocks make a significant contribution ( which leads to the idea of asymptotic independence ) .
more precisely , we write formally @xmath271 as @xmath272 and @xmath273 , if @xmath274 all terms of the corresponding @xmath270 above matrix are @xmath275 .
moreover , the above determinant is unchanged by subtracting an odd column to the following even column , and then by subtracting an odd line to the following even line . in this way ,
the diagonal @xmath270 matrices becomes @xmath276 where the last equality relies on @xmath245 , @xmath277 and @xmath278 . as a consequence , in the expansion of the determinant over all permutations of @xmath279 , the terms corresponding to entries only in the diagonal @xmath270 block matrices have order at most @xmath280 , while all other terms have a strictly lower order ( at most @xmath281 ) .
as the integration domain of @xmath282 is @xmath283 , the only permutations hopefully giving a non - zero limit need to come from the block diagonal @xmath270 matrices . indeed they give a non - trivial limit : their contribution is exactly @xmath284 a simple change of variable @xmath285 allows to conclude , thanks to the easy limit , uniform on compacts , @xmath286 our last task is to show that in the limit ( [ eqn : inclusionexclusion ] ) is equivalent to its @xmath267 term . by iterations of the hadamard - fisher inequality , lemma [ lem : fischer ] , @xmath287 the contribution of all terms with @xmath288 in ( [ eqn : inclusionexclusion ] ) is therefore bounded by @xmath289 where the integration domain @xmath290 has size @xmath291 and @xmath292 .
the first term of the above product converges , as previously proved ( it corresponds to @xmath267 ) , so the whole term goes to 0 as @xmath34 , concluding the proof .
[ lem : uniformbounded ] there is a constant @xmath234 depending only on @xmath169 such that , for any @xmath293 and @xmath294 ( see ( [ eqn : defomega ] ) ) , @xmath295 as previously mentioned , the formula ( [ eqn : inclusionexclusion ] ) is true whenever , for all distinct @xmath296 and @xmath297 , @xmath298 , i.e. @xmath299 is in @xmath256 . using the hadamard - fisher inequality as in the proof of the previous lemma ,
@xmath251 is therefore bounded by @xmath300 with @xmath290 .
once again , the hadamard - fisher inequality gives @xmath301 this gives the upper bound , uniform in @xmath302 , @xmath303 converging to @xmath304 as previously seen .
a better upper bound in the previous proof can be obtained as follows . by a direct ensembles argument ,
@xmath251 is bounded by @xmath305 this also comes from the fact that the inclusion - exclusion series ( [ eqn : inclusionexclusion ] ) is alternate .
we know by fisher - hadamard that this upper bound is lower than @xmath306 , which is interpreted as follows : @xmath307 converges to its limit from below , which is a sign of repulsion before the asymptotic independence .
[ lem : neglectset ] let @xmath258 be the complement of @xmath308 in @xmath257 ( see ( [ eqn : defomega ] ) ) . then @xmath309 _ proof .
_ let @xmath310 , and note @xmath311 the set of these points .
r.5 for notational convenience , one can suppose @xmath312 .
a set of points @xmath313 is said to be a cluster of @xmath314 if , for all @xmath315 , @xmath316 .
let @xmath317 be the set of points which can not be included in a maximal cluster : @xmath318 this way we get a partition @xmath319 where there are @xmath320 maximal clusters @xmath321 .
suppose that @xmath322 where @xmath323 .
then the following obvious bound holds : @xmath324 where we used the hadamard - fisher inequality .
this last product of @xmath325 elements is bounded , uniformly in @xmath261 , as it is equal to the @xmath325th power of @xmath326 which converges .
concerning the first product , we analyze each term in the following way . for any @xmath327 ,
the correlation @xmath328 is a @xmath329 determinant of @xmath330 .
suppose that all arguments of @xmath328 are included in an interval of size @xmath331 . by subtracting the first row from the @xmath332 others one obtains that the first row is @xmath247 and the others are @xmath333 , because @xmath277 .
hence @xmath334 . as a consequence ,
as all points of a cluster @xmath335 are within distance @xmath291 , @xmath336 this leads to the upper bound @xmath337 but the size of the set of points @xmath310 with such a clusters configuration is @xmath338 , because there needs to be @xmath339 small gaps ( i.e. of size @xmath291 ) between successive @xmath40 s .
hence the total contribution of such clusters to the integral over @xmath258 is @xmath340 which goes top 0 because @xmath341 ( with equality if all clusters have only one point ) and @xmath342 ( @xmath343 is not possible : this would mean that @xmath344 ) . +
as previously explained , the three above lemmas conclude the proof of theorem [ thm : unitary ] .
note that the events @xmath345 and @xmath346 are almost surely the same ( @xmath347 ) .
the independence property of the limit poisson point process @xmath35 in disjoint subsets therefore yields @xmath348 where we noted that , for any interval @xmath349 , @xmath350 is a poisson random variable with parameter @xmath351 .
a straightforward simplification of the above products gives the expected result .
concerning the limiting density of @xmath352 , we proceed in two steps .
first , from formula ( [ eqn : jointlarge ] ) , the joint density of @xmath353 is , in the limit and on the simplex @xmath354 , proportional to @xmath355 consequently , @xmath356 so the density of @xmath46 is proportional to @xmath2 .
the small gaps asymptotics for the @xmath18 are obtained exactly in the same way as for the unitary group .
the only difference is that the determinantal kernel is not translation invariant anymore , leading to some complications .
more precisely , let @xmath357 be the hermite polynomials , more precisely the successive monic orthogonal polynomials with respect to the gaussian weight @xmath358 . following @xcite , where the following results on the determinantal aspect of the @xmath18 can be found , we introduce the functions @xmath359 then the set of point @xmath360 with law ( [ eqn : gue ] )
is a determinantal point process with kernel ( with respect to the lebesgue measure on @xmath171 ) given by @xmath361 we will now discuss why lemma [ lem : nosucsmall ] till lemma [ lem : neglectset ] still hold in the case of the @xmath18 ensemble , restricted to @xmath362 for a given @xmath71 .
note that the hadamard - fischer inequality , an important tool for the proof of all the following lemmas , still holds for the gaussian unitary ensemble because the set of its eigenvalues is a determinantal process .
* we now note @xmath363 and its thinning @xmath229 obtained from @xmath213 by only keeping the eigenvalues @xmath364 for which @xmath365 .
the absence of successive small gaps , lemma [ lem : nosucsmall ] , only requires in its proof that @xmath366 , @xmath367 .
this is proved in lemma [ lem : infinitenorms ] . *
the analogue of the simple convergence of the correlation function associated to the thinned point process , lemma [ lem : simpleconvergence ] , is now , for any distinct points @xmath368 in @xmath362 , @xmath369 the proof , in the same way as the unitary case , requires uniform bounds on the partial derivatives of the kernel , given in lemma [ lem : infinitenorms ] , and the asymptotics @xmath370 which is a direct consequence of lemma [ lem : taylorexpansion ] . *
the uniform boundness result , lemma [ lem : uniformbounded ] , only requires the hadamard - fischer inequality , lemma [ lem : fischer ] , which applies to any determinantal point process . * finally , we give the analogue of lemma [ lem : neglectset ] in the following way .
let @xmath371 and @xmath372 be its complement in @xmath257 , where @xmath78 is any borel set in @xmath362 .
@xmath373 the only estimate necessary to this result is that the first order partial derivative of the kernel is uniformly @xmath374 , which is one of the estimates of the following lemma [ lem : infinitenorms ] .
[ lem : infinitenorms ] let @xmath71 . uniformly for @xmath375 , @xmath376 is @xmath374 , the first order partial derivatives of @xmath377 are @xmath378 and the second order ones are @xmath379 .
moreover , under the additional condition @xmath380 , @xmath376 is uniformly bounded , by a constant depending only on @xmath381 and @xmath382 . from the plancherel - rotach asymptotics for the hermite polynomials ( theorem 8.22.9 in @xcite ) , for any nonnegative integer @xmath0 , @xmath383 is @xmath384 , uniformly in @xmath385 .
consequently , if @xmath386 , from the first line of ( [ eqn : kernelhermite ] ) , @xmath377 is uniformly @xmath387 . to prove the first assertion of the lemma
, we use the stability property of the functions @xmath388 by derivation ( see @xcite lemma 3.2.7 ) : @xmath389 injecting this expression of @xmath390 in the last two lines of ( [ eqn : kernelhermite ] ) and using @xmath391 yields @xmath392 uniformly for @xmath375 . iterating this procedure , any partial derivative of @xmath377 of order @xmath0 are @xmath393 times a linear combination of factors of type @xmath394 , which shows that any partial derivative of order @xmath0 is @xmath395 .
the plancherel - rotach asymptotics for hermite polynomials also yield a precise evaluation of the correlation functions evaluated at close points , in particular the following result .
[ lem : taylorexpansion ] let @xmath71 , @xmath396 .
then as @xmath34 , uniformly in @xmath385 , @xmath397 , @xmath398 the intuition for this result is as follows . from the well - known convergence to the sine kernel , @xmath399 if this convergence is sufficiently uniform , one expects for small @xmath400 a taylor expansion @xmath401 injecting this expansion in @xmath402 gives the expected result . to justify the accuracy of the sine kernel approximation for @xmath400 close to 0 , we rely on corollary 1 in @xcite ( more general asymptotics in the full complex plane for , in particular ,
hermite polynomials were obtained in @xcite ) : using the plancherel - rotach asymptotics of hermite polynomials the authors show that . ] @xmath403 uniformly for @xmath404 in @xmath405 where @xmath406 , @xmath407 , @xmath408 , @xmath409 , @xmath410 . as @xmath411 and @xmath412 , a simple expansion yields @xmath413 for @xmath414 , this implies @xmath415 and the expected result for @xmath416 .
the above discussion concludes the proof of theorem [ thm : g ] .
its corollary [ cor : smallestgue ] follows exactly in the same way as corollary [ cor : smallestunitary ] from theorem [ thm : unitary ] .
corollary [ cor : smallestlocation ] is a straightforward consequence of the scission of the limiting lvy measure : if @xmath417 is a poisson point process with measure @xmath418 , conditionally to the @xmath419 s the @xmath420 s are distributed independently of each other , independently of the @xmath419 s , as a poisson point process with intensity proportional to @xmath421 . in particular , in our situation ,
the abscissa associated to the minimal ordinate is distributed with density proportional to @xmath422
to evaluate the extreme gaps , we first investigate the queuing distribution of one given spacing . a large part of the literature concerns the probability of having no eigenvalues in a given interval ( e.g. @xcite )
. a rigorous derivation of the queuing distribution for the large gaps was only given recently thanks to the steepest descent method for riemann hilbert problems @xcite or by operator theory tools for toeplitz determinants @xcite .
all the asymptotics we need in the following are direct consequences of the precise analysis of @xmath428 .
such estimates were first given by widom @xcite , for the following equation ( [ eqn : est1 ] ) without the error term , ( [ eqn : est2 ] ) first by krasovsky @xcite and ( [ eqn : est1 ] ) together with ( [ eqn : est2 ] ) by deift & al @xcite : they prove that for some sufficiently large @xmath429 and any @xmath154 , uniformly in @xmath430 , @xmath431 for an explicit constant @xmath432 , which remained conjectural for 20 years . from this
, we can give upper bounds on @xmath433 , the expectation of the number of gaps greater than @xmath400 .
from lemma [ lem : meh ] , @xmath437 evaluated for @xmath438 the first term is evaluated thanks to ( [ eqn : est2 ] ) : it is @xmath439 uniformly in @xmath440 . concerning the second term , ( [ eqn : est1 ] ) and
the inequality @xmath441 on @xmath442 imply that it is @xmath443 , concluding the proof .
note that , substituting @xmath444 in the asymptotics ( [ eqn : est1 ] ) and ( [ eqn : est2 ] ) yields @xmath445 where the @xmath446 terms are uniform in @xmath447 $ ] , for any given positive @xmath448 and @xmath449 .
hence , if @xmath450 , the expected number of gaps greater than @xmath451 goes to @xmath152 .
if , @xmath452 , this goes to @xmath453 .
the transition in the number of large gaps at @xmath454 is a strong clue for theorem [ thm : largestgaps ] , but it is not sufficient : concluding the proof requires additional knowledge about the correlations between gaps , lemma [ lem : negcor ] , assumed for the moment . take @xmath154 and @xmath5 .
we note @xmath455 where @xmath99 .
then @xmath456 and we aim to show that the last three terms converge to 0 ( the value @xmath457 is arbitrary , any angle strictly smaller than @xmath458 would be appropriate for this proof ) . first , decompose the unit circle into 8 fixed angles of size @xmath459 . if @xmath460 , on of these arcs is free of eigenvalues . from ( [ eqn : est1 ] )
, the probability of this event decreases exponentially , so the last term in ( [ eqn:3terms ] ) converges to 0 .
moreover , integrating by parts , @xmath461 because @xmath462 needs to be shorter than @xmath463 . the probability that @xmath462 is greater that @xmath400 is obviously shorter than @xmath464 , the expectation of the number of gaps greater than @xmath465 .
hence the above quantity goes to @xmath152 thanks to the uniform estimate of lemma [ lem : boundexpectation ] . finally , showing that @xmath466 requires an additional argument , the negative correlation property for empty sets events , lemma [ lem : negcor ] .
in particular , this negative correlation implies the following result .
[ lem : varianceexpectation ] consider a set of disjoint arcs on the unit circle .
let @xmath467 be the number of such intervals free of eigenangles , i.e. those @xmath468 s such that @xmath469
. then @xmath470 .
this is a straightforward consequence of lemma [ lem : negcor ] : as @xmath213 is a determinantal point process , for disjoints @xmath78 and @xmath471 , @xmath472 .
hence , noting @xmath473 the number of initial arcs , @xmath474 as expected .
consider now a number @xmath473 of disjoint intervals @xmath475 of length @xmath476 in @xmath194 .
we can find @xmath477 of them .
let @xmath467 be the number of such intervals free of eigenangles , i.e. those @xmath468 s such that @xmath469 .
if there are less than @xmath60 gaps larger than @xmath476 , either there are less than @xmath60 intervals @xmath468 s free of eigenangles , or there is a gap between successive eigenangles containing two intervals : @xmath478 this last term is bounded by the expectation of the number of gaps greater than @xmath479 ; from ( [ eqn : transition ] ) this goes to 0 if @xmath480 ( true for @xmath160 sufficiently small ) . concerning @xmath481 , first note that by the estimate ( [ eqn : transition ] ) , @xmath482 , so @xmath483 .
this allows to use chebyshev s inequality , for sufficiently large @xmath17 : @xmath484 where we used lemma [ lem : varianceexpectation ] in the last inequality .
this last term is equivalent to @xmath485 , thus going to @xmath152 , which concludes the proof . for the proof of theorem [ thm : largestgapsgue ] , precise asymptotics like ( [ eqn : est1 ] ) and ( [ eqn : est2 ] ) related to unitary groups are not available in the @xmath18 context .
this difficulty can be overcome , our main observation being lemma [ lem : compfredholm ] : the probability that an interval is free of eigenvalues is equivalent in the @xmath486 and @xmath48 cases , up to a normalization , if the interval size is shorter than the expected extreme gap size .
the proof relies on a comparison of the fredholm determinants associated to @xmath377 and @xmath330 .
the rest of the proof is similar to the one concerning @xmath48 .
indeed , consider an @xmath154 , @xmath5 , @xmath487 the @xmath60th largest gap in @xmath78 , and @xmath488 we then decompose @xmath489 this last expectation is , by integration by parts , @xmath490 the probability @xmath491 is lower than the expectation of the number of gaps greater than @xmath492 .
lemma [ lem : inthetailgue ] therefore yields @xmath493 as @xmath34 .
concerning @xmath494 , we proceed as for the unitary group : for @xmath495 with @xmath496 , consider @xmath497 disjoint intervals of length @xmath498 included in @xmath78 .
let @xmath499 be the number of these intervals free of eigenvalues .
then @xmath500 from lemma [ lem : inthetailgue ] this last probability goes to @xmath152 as @xmath34 if @xmath480 , @xmath501 .
moreover , from lemma [ lem : expectgue ] , @xmath502 for some @xmath503 depending only on @xmath160 and @xmath78 .
moreover , from the negative correlation property lemma [ lem : negcor ] and the same reasoning as lemma [ lem : varianceexpectation ] , @xmath504 as @xmath62 , from ( [ eqn : expboundgue ] ) @xmath505 for sufficiently large @xmath17 , which allows to use chebyshev s inequality : @xmath506 the last inequality being ( [ eqn : varianceexpectationgue ] ) . from ( [ eqn : expboundgue ] ) this last term is equivalent to @xmath507 and going to 0 , as @xmath34 , concluding the proof . concerning the gaussian unitary ensemble , the same type of asymptotics hold , being a direct consequence of formula ( [ eqn : gueexpansion ] ) .
note that when taking @xmath516 the speed of convergence to the sine kernel is @xmath517 for the unitary group , much better that @xmath518 for the @xmath18 , whose correlation kernel is not translation invariant . by inclusion - exclusion ,
the probability that a determinantal point process with kernel @xmath11 has no points in a measurable subset @xmath169 is the fredholm determinant ( see e.g. lemma 3.2.4 in @xcite ) @xmath522 to compare empty sets probabilities , we therefore need to compare fredholm determinants .
a classical inequality ( see e.g. @xcite chap iv , ( 5.14 ) ) is @xmath523 is the trace norm of a nuclear operator @xmath136 .
however , for positive ( like @xmath169 or @xmath524 ) operators @xmath525 the trace norm is @xmath526 , but for non - positive operators , like @xmath527 , the trace norm is difficult to evaluate , even for a compactly supported continuous kernel @xmath11 .
however , in such a case , the hilbert - schmidt norm is computable : @xmath528 ) , we will use the modified carleman - fredholm determinant @xmath529 and the inequality @xmath530 ) , after a simple change of variables , the probability that there are no eigenvalues in @xmath531 $ ] is equal to @xmath532 in the same way , the probability that there are no eigenvalues in @xmath533 $ ] is @xmath534 hence the inequality ( [ eqn : ineqhs ] ) will be applied with @xmath169 and @xmath524 integral operators with respective kernel @xmath535 and @xmath536 from lemma [ lem : asymptsine ] , the infinite norm between the two kernels above is @xmath537 , so by ( [ eqn : hsnorm ] ) , integrating on a domain of area @xmath538 , @xmath539 moreover , consider a parameter @xmath540 and decompose @xmath541 from lemma [ lem : asymptsine ] , if @xmath542 then @xmath543 is smaller than @xmath544 , and when @xmath545 it is bounded by @xmath546 .
consequently , @xmath547 by choosing @xmath548 . in the same way ,
hence @xmath550 , here on the diagonal , @xmath551 and the same for @xmath552 .
finally , write @xmath553 and @xmath554 show that the first term is bounded by @xmath555 .
moreover , @xmath556 is a probability , so bounded by 1 , and the estimates of @xmath557 and @xmath552 , formula ( [ eqn : estimatetrace ] ) , show that @xmath558 is @xmath559 , concluding the proof .
the first step allows reasoning on fixed intervals instead of gaps .
more precisely , note @xmath82 $ ] with @xmath496 , and consider the intervals of length @xmath565 by successive slips of size @xmath566 : @xmath567,\ 0\leq k\leq p_n=\lfloor ( b - a)/s_n\rfloor.\ ] ] there an injective map associating to any eigenvalues gap of size at least @xmath568 an interval @xmath569 included in this gap , for example the one with lower index .
consequently , @xmath570 where we use the abbreviation @xmath571 .
the second step consists in obtaining uniform upper bounds for these empty intervals probabilities . for this purpose , the negative correlation property , lemma [ lem : negcor ] ,
is used by partitioning the interval @xmath569 : @xmath572 where * @xmath573 has length @xmath574 , with @xmath575 to be chosen positive but sufficiently small compared to @xmath160 ; * @xmath576 has length @xmath577 ; * the intervals @xmath578 all have length @xmath579 and their number is @xmath580 ; * @xmath189 is a residual interval .
the negative correlation property yields @xmath581 this now can be upper - bounded using lemma [ lem : compfredholm ] because all intervals are shorter than @xmath582 : for example , noting @xmath520 one extremity of @xmath573 , @xmath583=0\right)+n^{{\mathrm{o}}(1)-1}\\ \leq \operatorname{\mathbb{p}}^{\operatorname{u}(n)}\left([0,(1-{\varepsilon}')\alpha_n]=0\right)+n^{{\mathrm{o}}(1)-1}=n^{-(1-{\varepsilon}')^2+{\mathrm{o}}(1)}\end{gathered}\ ] ] with @xmath446 not depending on the index @xmath0 , and where the last estimate relies on ( [ eqn : est1 ] ) . in the same way , @xmath584 gathering all these results , @xmath585 we can chose @xmath575 sufficiently small such that @xmath586 ( e.g. @xmath587 ) .
for such a choice , for sufficiently large @xmath17 not depending on @xmath588 the above exponent is smaller than @xmath589 for some @xmath396 .
such a function is smaller than @xmath590 on @xmath563 , for some @xmath591 .
[ lem : expectgue ] let @xmath82 $ ] with @xmath496 , @xmath560 , @xmath592 and @xmath561 . consider a maximal number @xmath593 of disjoint intervals @xmath475 of length @xmath594 included in @xmath78
let @xmath595 be the number of those intervals containing no eigenvalues .
then @xmath596 for some @xmath503 depending only on @xmath160 and @xmath78 we first take a restricted interval to avoid the fluctuations in the spectral measure : there is @xmath597 with @xmath598 such that @xmath599 then , for a given interval @xmath600 $ ] included in @xmath601 , from lemma [ lem : compfredholm ] , @xmath602,1\leq i\leq n\right)+n^{{\mathrm{o}}(1)-1}\ ] ] with @xmath446 uniform in @xmath520 . from ( [ eqn : restrictedi ] ) this is greater than @xmath603,1\leq i\leq n\right)+n^{{\mathrm{o}}(1)-1}=n^{-\left(1-\frac{{\varepsilon}}{2}\right)^2+{\mathrm{o}}(1)}.\ ] ] from ( [ eqn : est1 ] ) .
there are @xmath604 intervals @xmath468 s included in @xmath601 , so as @xmath34 @xmath605 with @xmath606 for example . in the context of a point process on a finite set @xmath607 ,
let @xmath186 and @xmath608 be distinct disjoint subsets of @xmath607 , and write @xmath609 for the event that the elements of @xmath186 are free of particles .
shirai and takahashi @xcite showed that for determinantal point processes , the empty sets events are negatively correlated : @xmath610 this negative association property has received a considerable attention in the past few years , in the context of asep for example .
still for discrete determinantal point processes , formula @xmath611 was generalized to all increasing events @xcite , and general criteria for the negative association property were given in @xcite .
the following continuous analogue of ( [ eqn : negdiscr ] ) holds .
it can be shown by a simple discretization , relying on results from @xcite .
we give another justification , which relies on a work of georgii and yoo @xcite .
[ lem : negcor ] let @xmath213 be the point process associated to the eigenvalues of haar distributed unitary matrix ( resp .
an element of the @xmath18 ) .
let @xmath612 and @xmath613 be compact disjoint subsets of @xmath614 ( resp .
@xmath171 ) . then @xmath615 the general negative correlation result , corollary 3.3 in @xcite , requires a locally trace class operator @xmath11 with a restriction on its spectrum .
more precisely , consider the unitary case , the @xmath18 proof , being similar .
for an operator @xmath11 acting on @xmath616 , if @xmath617 $ ] , there exist a unique determinantal point process @xmath618 with kernel @xmath11 ( see @xcite ) ; under the additional hypothesis @xmath619 , georgii and yoo proved that for any compact and disjoint borel sets @xmath620 , @xmath621 in the case of haar distributed unitary matrices , @xmath622 is a nuclear operator with kernel @xmath623 , and defines a projection : @xmath624 is in its spectrum , and the general statement does not directly apply . to care of this minor problem ,
look at the restriction @xmath625 of @xmath330 to a compact subset @xmath186 of @xmath194 ( @xmath626 , @xmath627 being the projection on @xmath186 ) .
suppose that the set @xmath628 has a nonempty interior . as a projection of @xmath330 ,
@xmath625 is still non - negative and trace class . as for any determinantal point process , @xmath629 in the sense of fredholm determinants of a trace class operator .
suppose that @xmath630 has an eigenvalue @xmath631 .
then by choosing @xmath632 , ( [ eqn : freddet ] ) yields @xmath633 in each case , this is absurd because the joint law of the eigenvalues is absolutely continuous with respect to the lebesgue measure on @xmath634 , and @xmath635 has a nonempty interior : both quantities need to be strictly positive . hence @xmath636 is a trace class operator with spectrum in @xmath637 , and the result from @xcite applies .
another way to prove lemma [ lem : negcor ] is as follows : the inequality can be stated for the determinantal point process with kernel @xmath638 with @xmath639 , and then the inequality remains true for @xmath330 by continuity of the application @xmath640 in the set of trace class operators , by ( [ eqn : ineqtrace ] ) .
anderson , a. guionnet , o. zeitouni , an introduction to random matrices , cambridge university press , 2009 .
g. ben arous , a. guionnet , `` wigner matrices '' , chapter 21 , handbook on random matrix theory , editors : g.akemann , j.baik , p.di francesco , to appear oxford university press , 2010 . g. ben arous , a. kuptsov , rem universality for random hamiltonians , preprint .
j. borcea , p. brndn .
t. liggett , negative dependence and the geometry of polynomials , j. amer .
( 2009 ) , 521 - 567 .
h. bui , large gaps between consecutive zeros of the riemann zeta - function , preprint .
d. bump , lie groups , springer , 2004 .
p. deift , some open problems in random matrix theory and the theory of integrable systems .
contemporary mathematics ( 2008 ) 458:419 - 30 .
p. deift , t. kriecherbauer , k.t - r mclaughlin , s. venakides , x. zhou : uniform asymptotics for polynomials orthogonal with respect to varying exponential weights and applications to universality questions in random matrix theory .
pure appl .
52 , 1335 - 1425 ( 1999 ) p. deift , a. its , i. krasovsky , x. zhou , the widom dyson constant for the gap probability in random matrix theory , journal of computational and applied mathematics volume 202 , issue 1 , 1 may 2007 , pages 26 - 47 .
p. deift , t. nanda , c. tomei , ordinary differential equations and the symmetric eigenvalue problem , siam j. numer .
( 1983 ) , pp . 1 - 22
. b. delyon , j. yao , on the spectral distribution of gaussian random matrices , acta mathematicae sinica , english series vol .
22 , no . 2 ( 2006 ) 297 - 312 .
p. diaconis .
patterns in eigenvalues : the 70th josiah williard gibbs lecture . bull .
40 , 155178 ( 2003 ) .
f. dyson , statistical theory of the energy levels of complex systems , i , ii , iii , j. math .
phys . 3 140 ( 1962 )
. t. ehrhardt , dyson s constant in the asymptotics of the fredholm determinant of the sine kernel , comm .
262 , 317 - 341 ( 2006 ) .
l. erds , b. schlein , h .-
universality of random matrices and local relaxation flow , preprint .
georgii , h.j .
yoo , conditional intensity and gibbsianness of determinantal point processes , journal of statistical physics , vol .
118 , nos 1/2 , january 2005 .
i. goldberg , s. goldberg , n. krupnik , traces and determinants of linear operators , operator theory : advances and applications , vol .
116 , birkhuser , 2000 .
x. gourdon , computation of zeros of the zeta function , 2004 , unpublished .
r. horn , c. johnson , matrix analysis , cambridge university press , 1985 .
k. johansson , random matrices and determinantal processes , lectures given at the summer school on mathematical statistical mechanics in july 05 at ecole de physique , les houches .
o. kallenberg , random measures , 3rd , rev . and enl . ed .
new york : academic press , 1983 .
katz , p. sarnak , random matrices , frobenius eigenvalues and monodromy , american mathematical society colloquium publications , 45 .
american mathematical society , providence , rhode island , 1999 .
j.p . keating , n.c .
snaith , random matrix theory and @xmath641 , comm .
214 , p 57 - 89 , 2000 .
i. v. krasovsky , gap probability in the spectrum of random matrices and asymptotics of poly orthogonal on an arc of the unit circle , international mathematics research notices 2004 ( 25 ) : 1249 - 1272 .
r. lyons , determinantal probability measures , publ .
hautes tudes sci .
98 ( 2003 ) , 167212 .
mehta , random matrices , third edition .
pure and applied mathematics ( amsterdam ) , elsevier / academic press , amsterdam , 2004 .
montgomery , the pair correlation of zeros of the zeta function , analytic number theory ( st . louis , 1972 ) , proc . sympos .
pure math .
24 , amer .
( providence ) , 181 - 193 ( 1973 ) .
a.m. odlyzko , on the distribution of spacings between zeros of the zeta function , math .
, 48 ( 1987 ) , pp .
273 - 308 .
t. shirai , y. takahashi , random point fields associated with certain fredholm determinants ii : fermion shifts and their ergodic and gibbs properties , ann .
volume 31 , number 3 ( 2003 ) , 1533 - 1564 .
a. soshnikov , level spacings distribution for large random matrices : gaussian fluctuations annals of mathematics , vol .
148 , pp.573 - 617 , 1998 .
a. soshnikov , determinantal random point fields russian mathematical surveys , vol.55 , no.5,923 - 975 , 2000 a. soshnikov , statistics of extreme spacings in determinantal random point processes moscow math .
j. , vol . 5 , no.3 , 705 - 719 , 2005 . g. szego , orthogonal polynomials , 4th ed .
, american mathematical society , colloquium publications , vol .
23 , american mathematical society , rhode island , 1975 .
t. tao , v. vu , random matrices : universality of local eigenvalue statistics , to appear in acta mathematica . j. vinson , closest spacing of eigenvalues , phd thesis , princeton university , 2001 .
h. widom , the strong szeg limit theorem for circular arcs , indiana univ .
j. 21 , 277 - 283 ( 1971 ) . | this paper studies the extreme gaps between eigenvalues of random matrices .
we give the joint limiting law of the smallest gaps for haar - distributed unitary matrices and matrices from the gaussian unitary ensemble .
in particular , the @xmath0th smallest gap , normalized by a factor @xmath1 , has a limiting density proportional to @xmath2 . concerning the largest gaps , normalized by @xmath3 , they converge in @xmath4 to a constant for all @xmath5 .
these results are compared with the extreme gaps between zeros of the riemann zeta function . |
large - amplitude oscillations ( laos ) of prominences are recently attracting more attention in the solar physics community .
the number of reported events is increasing as the current telescope capabilities allow for continuous full sun monitoring .
laos are motions involving speeds larger than @xmath1 according to the classification of @xcite .
these speeds are larger than the local sound speed in the cool prominence plasma , typically @xmath2 .
thus , laos involve supersonic motions and non - linear effects are expected .
laos are typically divided in two kinds of motions : longitudinal and transverse with respect to the spine of the filament ( see a review by * ? ? ?
* ) . in this work we focus exclusively on the large - amplitude longitudinal oscillations ( lalos ) case .
these motions were first reported by @xcite followed by two more publications by @xcite and @xcite . in these works
the authors reported five lalos events where the filament plasma move almost parallel to their spines .
more recently @xcite reported an oscillation at the limb showing that the motion is along the prominence dipped magnetic field lines .
more lalos observations were reported and characterized in later work @xcite . in all the lalos
reported so far the periods range from 40 to 100 minutes and the amplitudes are between @xmath1 to @xmath3 . in lalos the prominence threads move parallel to themselves indicating that the motion is along the magnetic field .
@xcite demonstrated that in the considered event the motion formed @xmath4 with respect to the filament spine .
that orientation coincided with the typical observed orientations of the magnetic field of filaments ( see e.g. , * ? ? ?
* ; * ? ? ?
thus , all above evidences suggest that lalos are actually oscillations along the magnetic field .
the nature of lalos is currently a matter of discussion since , in principle , several forces could act as the restoring force .
based on 1d numerical simulations , @xcite proposed the pendulum model to explain lalos , where the restoring force is the gravity projected along the magnetic field lines .
these authors discarded the magnetic origin of the restoring force because the lorentz force is always perpendicular to the magnetic field and does not affect the longitudinal motion .
similarly , the numerical simulations presented by @xcite show that the gas pressure gradient force is not important and that the nature of lalos is not magnetosonic .
these results were confirmed by other 1d simulations by @xcite .
@xcite performed an analytical study of the influence of the magnetic field curvature on linear longitudinal oscillations .
it was found that the longitudinal oscillations are in general a combination of the pendulum and the slow modes .
however , for typical prominence parameters the contribution of the slow modes is very small confirming the findings of previous studies .
all these results support the pendulum model for the lalos .
the main drawback of the 1d modeling is that the magnetic field is assumed rigid and the motion of the plasma perfectly follows the field lines . in this regime
the magnetic field does not react to the motion of the plasma .
however , the heavy mass of the prominence and the non - zero value of the plasma-@xmath5 could make the role of the magnetic field more important and influence the motion of threads @xcite .
for example , the mass could change dynamically the magnetic structure producing non - linear coupling of the longitudinal and transverse motions and the pendulum model might be no longer valid .
recently , @xcite studied the different modes of oscillations of a prominence embedded in a 2d structure in the linear regime .
they used a realistic scenario where the magnetic field reacts to the plasma motion .
the authors found evidences that the longitudinal motion is affected by the gravity but also by the gas pressure .
however , the authors placed the prominence in a very high position where the field lines are essentially flat and convex - downwards . at such position
, the longitudinal motion is strongly influenced by the slow modes as was found in @xcite .
due to the potential applicability for prominence seismology , it is necessary to validate or to discard the pendulum model . in this study
we demonstrate that the pendulum model is robust , with non - linear 2d time - dependent numerical simulations .
we prove that the period of the oscillations depends exclusively on the radius of curvature of the dipped field lines and that the longitudinal and transverse motions ( relative to the magnetic field ) are not strongly coupled .
the back reaction of the magnetic field is small when lalos are present and the restoring force is the projection of the gravity along the magnetic field lines .
the aim of the numerical experiment is to reproduce the lalos observed in filaments . in those events nearby energetic disturbances perturb an existing filament from the side , probably along the filament - channel magnetic structure .
the hot plasma produced at the energetic event flows along the magnetic field lines pushing an already formed prominence . in order to have a prominence we instantaneously load a force free magnetic structure with a prominence mass .
initially , such a configuration is not in force equilibrium .
we then let the system evolve in order to have a relatively relaxed configuration .
the amplitude of the motions of the system decreases with time but the velocity field does not disappear completely .
once we reach a relatively relaxed situation we instantaneously apply a velocity perturbation along the magnetic field of the prominence in order to mimic a typical observed excitation event such as a jet or a small flare . the velocity perturbation is placed over the almost relaxed prominence with a speed of @xmath1 , typical for lalos . in the relaxation phase ,
the initial configuration consists of a force - free magnetic field described in @xcite .
it represents a symmetric double arcade given by equations ( [ eq : fff - conf - x ] ) and ( [ eq : fff - conf - z ] ) and plotted in figure [ fig : initial - conf ] .
@xmath6 this structure has dips closer to the surface , where we initially place the prominence mass .
the depth of the dips decreases monotonically from bottom to top positions and the curvature at the central position of the field lines changes from concave upwards to convex downwards forming a overlying arcade of loops .
the magnetic field has a null point at @xmath7 that we have placed outside the numerical domain by taking @xmath8 .
in addition , placing the null point bellow the numerical domain we avoid partially the non realistic region of very high plasma-@xmath5 ( ratio of gas pressure to magnetic pressure ) surrounding the null
. both pressures change dynamically with time , but the plasma-@xmath5 is always constrained to @xmath9 in all the numerical domain .
we have considered the parameters @xmath10 and @xmath11 . at the central position of the structure
the magnetic field strength is @xmath12 at @xmath13 and reaches a maximum of @xmath14 at @xmath15 . beyond this point
the field strength decreases with height .
this point coincides with the change of curvature of the field lines .
thus , below @xmath16 the magnetic structure has dips whereas above this point the structure is an arcade of loops .
the numerical domain consists of a box of @xmath17 points and @xmath18 , with a spatial resolution of 125 km .
we assume open boundary conditions for the top and the two side boundaries . at the bottom boundary , line - tying
is considered by imposing current free conditions and zero velocities as in @xcite . ) and ( [ eq : fff - conf - z ] ) [ fig : initial - conf ] and an initial density and temperature prominence perturbation . for the force - free magnetic configuration we have imposed @xmath19 , @xmath11 , and @xmath8 .
a selected set of field lines have been plotted as white lines .
the color field represents the density at the prominence and corona .
the density of the internal part of the prominence is @xmath20 times the density of the corona.,scaledwidth=50.0% ] the coronal plasma is assumed initially stratified with a uniform million degree temperature , @xmath21 .
the prominence is placed on the structure by perturbing the coronal atmosphere with a temperature of the form @xmath22 where @xmath23 is the position of the center of the pulse and @xmath24 are the widths of the perturbation in @xmath25- and @xmath26-directions respectively .
the temperature at the central position of the prominence is @xmath27 and @xmath28 is selected accordingly to @xmath29 .
the initial temperature perturbation has a quartic dependence in the exponent in order to have a large region of cool and dense plasma .
the pressure is unperturbed and equal to that of the ambient corona .
the density of the perturbation is computed accordingly to the ideal gas law ( see fig .
[ fig : initial - conf ] ) .
in this situation , the gas pressure along the horizontal direction is uniform , @xmath30 .
we numerically solve the ideal mhd equations in the above scenario by using our recently developed mancha code @xcite .
mancha solves the time dependent nonlinear problem given initial conditions .
the relaxation phase is not relevant for this study , but it is sufficiently interesting to be briefly described . the initial configuration is not in equilibrium and the weight of the prominence mass is not compensated by the magnetic structure or gas pressure gradients . the prominence plasma rapidly starts to fall and collapses , increasing its internal pressure .
after some time the collapse ceases producing a shock wave that propagates along the magnetic field lines at both sides of the prominence .
these shocks bounce at the field footpoints moving back again to the prominence .
the shocks travel between the two footpoints , and the prominence oscillates vertically producing again more shocks by the same process .
after @xmath31 the prominence is more or less relaxed into a dynamical equilibrium . at this point
vertical oscillations and waves along the magnetic field remain but the speeds involved are below @xmath32 .
it is important to note that we include the remnant velocities in the next phase where we add the longitudinal perturbation to the already existing velocity field .
the resulting dynamics will not be strongly influenced by these remnant motions .
once the prominence is sufficiently relaxed we perturb the resulting configuration with a velocity field along the magnetic structure , namely @xmath33 in order to excite longitudinal motions .
we have considered this time as @xmath34 .
the relaxation phase has @xmath35 , while @xmath36 is taken after the longitudinal velocity perturbation .
@xmath37 is the magnetic field at the end of the relation phase or at @xmath34 .
the perturbation has @xmath38 and @xmath39 , much larger than the initial temperature perturbation in order to excite oscillations of the cool plasma and in their surrounding .
the perturbed initial velocity is set to @xmath40 which is in the range of large - amplitude oscillations . due to the dimensions of the considered structure , larger initial velocities produce drainage of cool plasma on the sides of the structure . in figure
[ fig : time - sequence - lalo ] we have plotted a four panel time sequence of the density and the magnetic field . in figure
[ fig : time - sequence - lalo]a the almost relaxed prominence is plotted .
in contrast with the non - equilibrium initial situation ( fig . [
fig : initial - conf ] ) the density is now more vertically elongated because the cool plasma has deformed the initial magnetic field .
we perturb the system with the velocity of equation ( [ eq : initial_vel ] ) and the whole prominence material starts to move towards the increasing values of @xmath25-coordinate following clearly the magnetic field lines .
the velocity also decreases with time and after approximately @xmath41 most of the prominence reaches the maximum displacement ( fig .
[ fig : time - sequence - lalo]b ) and the motion reverses .
however , the bottom parts of the prominence reverse the direction of motion slightly before the rest of the structure .
the motion along the vertical structure is different and the phase difference becomes important , as it is apparent in figure [ fig : time - sequence - lalo]c . at this time the bottom part moves rightward and the central part leftwards with the top part considerably delayed .
the increase in the delay between the different parts of the structure seems gradual .
the last panel ( fig .
[ fig : time - sequence - lalo]d ) shows very well that the motion between different parts of the structure becomes out of phase with opposite directions of motion .
this indicates that the plasma moves along the magnetic field with different periods of oscillation .
the zig - zag structure resembles the motion of the apparent tornado structure observed with the sdo / aia , as reported by @xcite . from the figure it is clear that the density changes with time in reaction to both longitudinal and transverse movements .
the magnetic field structure also changes during the temporal evolution . in the figure ,
the white solid lines are a set of selected field lines at different times and the white dashed lines represent the magnetic field at @xmath34 ( fig .
[ fig : time - sequence - lalo]a ) . in the first period
the magnetic field changes slightly ( figs . [ fig : time - sequence - lalo]b and [ fig : time - sequence - lalo]c ) but rapidly recovers the initial configuration and remains more or less unchanged for the rest of the temporal evolution ( see fig .
[ fig : time - sequence - lalo]d ) . in the time - dependent evolution
we do not find a swaying motion of the magnetic structure .
in contrast , the magnetic structure remains more or less unchanged except for some small - amplitude vertical oscillations .
we conclude that the motion of the plasma mainly follows the magnetic field lines .
we see that different parts of the structure oscillate with different frequencies forming a continuous spectrum @xcite .
this behavior makes the spectral study of the motion complicated .
in addition , these oscillations are essentially non - linear and the plasma is subject to important advection . in our fluid , the frozen - in condition applies because we consider it to be perfectly conducting .
thus , it is necessary to advect every field line considered in order to catch the longitudinal and transverse motions of the plasma .
the bottom line - tying boundary conditions imply that the magnetic field is unperturbed during the temporal evolution .
thus , any field line starting at the bottom boundary follows the plasma motion simplifying the study of the oscillations .
we select a large set of equally spaced field lines starting at the bottom .
these field lines cross the central axis ( @xmath42 ) with approximate separations of @xmath43 .
the lines plotted in figure [ fig : time - sequence - lalo ] are only a small subset of the field lines considered in the following analysis .
we have computed the longitudinal and transverse motions of the center of mass as @xmath44 the @xmath45-index corresponds to the field line considered and @xmath46 is the coordinate along the field line @xmath45 that depends implicitly on the time , @xmath47 .
the integrands @xmath48 , @xmath49 and @xmath50 are the longitudinal velocity , transverse velocity and density along the field line @xmath45 respectively .
these velocities are plotted in figure [ fig : lalo - velocities ] for the different field lines .
the left vertical axis shows the velocity scale . for comparison ,
the right vertical axis shows the @xmath51-coordinate at the central position @xmath42 of the field lines at @xmath34 . in a linear situation
this is equivalent to studying the velocities along the vertical axis at @xmath42 .
we have plotted the temporal evolution of the velocities from @xmath52 to @xmath53 . before @xmath34
we clearly see the velocities of the relaxation phase at the right panel . at the relaxation
phase the central positions of the prominence oscillate vertically because we place the mass centered at the structure .
we see that the longitudinal velocities are zero ( left panel ) but there are clear oscillations along the vertical direction ( right panel ) . at @xmath34
we impose the instantaneous velocity perturbation that produces a sudden increase of the longitudinal velocity to @xmath1 , but no significant changes to the transverse motion .
the different nature of the longitudinal and transverse oscillations is also clear in their very different periods of oscillation .
we define the density at the center of mass of each field line as @xmath54 being @xmath55 the position of the center of mass along the field line @xmath45 .
the color coding is computed as the temporal average of @xmath56 . with this color coding
we can identify the oscillation corresponding to coronal or prominence plasma .
the longitudinal period increases monotonically from the bottom positions up to @xmath57 .
this range of heights includes almost all the prominence . above @xmath57
the period of the oscillations decreases . from the figure
we see that the longitudinal velocity at denser plasma shows a triangular shape with peaked maximums and minimums instead of a sinusoidal shape .
this could be related to the non - linearity of the oscillations .
the longitudinal oscillations show a significant damping ( fig .
[ fig : lalo - velocities]a ) in contrast to the weak damping of the transverse motions ( fig .
[ fig : lalo - velocities]b ) .
lalos show different damping times for different parts of the structure .
the plasma oscillations at the bottom parts of the structure are damped quicker than at the top parts . in order to study the convergence of the numerical solutions
, we have performed a check repeating our simulations with a decreased resolution of 500 and 250 km grid sizes .
we found that at all resolutions the simulations show a similar damping scale , with slightly larger damping for larger grid sizes and identical periods .
therefore , these tests demonstrate that our simulations have converged .
we may speculate that a kind of numerical phase - mixing is probably occurring for lal oscillations , mimicking a real process that may occur on the sun .
lalos at different field lines have different frequencies because they have different field line curvature
. then two adjacent field lines at the numerical grid lose their relative coherence and due to the numerical viscosity the motion tends to be canceled by stress .
the numerical viscosity depends on the grid size .
however , the convergence test has shown that we are very far from reaching realistic values of dissipation because the models with different resolution exhibit similar damping .
thus , significantly better grids are required to resolve the transition region surrounding the oscillating thread and to avoid the enhanced numerical viscosity phase - mixing .
this will be addressed in a future work using a new module of mancha with adaptive mesh refinement .
the study of the damping mechanism is out of the scope of this work .
however , understanding the nature of this damping is also important for lalos and it will be considered in a future publication .
the transverse velocities in figure [ fig : lalo - velocities ] ( right panel ) show an amazingly regular and coherent motion .
after the perturbation ( @xmath58 ) , there is an alteration of the regular pattern , but the system quickly recovers its very regular motion .
there are no important phase differences , and the oscillation periods are very similar ( @xmath59 in all field lines analyzed ) .
this behavior suggests that the transverse motions is a global normal mode of the structure .
however , one of the most important results is that there are no transverse motions associated to the longitudinal motion.the back reaction of the magnetic field should have produced displacements perpendicular to the magnetic field with the same or half period than the longitudinal motion .
however , this is not observed in figure [ fig : lalo - velocities ] or in the spectral analysis of the curves . we have analyzed the periods of oscillations at every field line considered in the two polarization directions .
the results are plotted in figure [ fig : lalo - periods ] .
-position of the field lines dips , as in figure [ fig : lalo - velocities ] ( top panel ) , and radius of curvature ( bottom panel ) of the field lines .
the radius of curvature is computed numerically along the field lines and averaged in space and time .
the radius of curvature for every field line is spatially averaged in @xmath60 along the line centered at the dip for every snapshot .
this spatially averaged radii are also temporally averaged in all the simulation time @xmath36 .
the color code is identical to fig .
[ fig : lalo - velocities].,scaledwidth=50.0% ] the longitudinal period increases with height ( top panel ) as we have discussed previously .
the increase is monotonic but above @xmath61 the behavior changes , in agreement with the visual impression from figure [ fig : lalo - velocities ] . in the figure
we have plotted the theoretical period , @xmath62 , of the mass loaded into the deformed structure ( solid line ) using the pendulum model expression from @xcite , @xmath63 where @xmath64 is the radius of curvature of the dipped portion of the field lines and @xmath65 is the solar gravity acceleration . we have computed the radius of the curvature at the center of the dipped portion of the field lines .
these radii slightly change in reaction to the mass motion .
we therefore have considered the temporal average of the radii .
we can see an excellent agreement between the time - dependent simulations and the pendulum model in the range @xmath66 that includes most of the prominence mass .
above @xmath61 there is no agreement with the model . above this point
the gas pressure gradient contributes largely to the restoring force .
the reasons are that the field lines supporting the cool plasma are flatter with small projection of the gravity along the field and the density contrast is relatively small . in this situation
the gas pressure gradient becomes important ( see * ? ? ?
for this reason , above @xmath67 the frequencies do not fit the pendulum curve . in this work we are considering field lines of the order of @xmath68 .
this length is probably in the range of short prominence field lines . in typical prominences
the field lines supporting the cool plasma are larger . in @xcite
we found that for longer magnetic field lines the pendulum works even better .
similarly , below @xmath69 the medium is essentially coronal plasma and the oscillation modes are slow modes . in the same figure
we have plotted the period of the transverse oscillation as diamonds ( top panel ) .
the periods are uniform with a common value of @xmath70 minutes .
the uniformity of the period suggests that the transverse motion is related to the fast wave global normal mode .
figure [ fig : lalo - periods ] ( bottom ) shows a very important result for prominence seismology with profound observational implications . in this figure
we have plotted the period as a function of the radius of curvature of the field lines .
this clearly demonstrates that the pendulum model of @xcite works exceedingly well for lalos in prominences .
thus , the period of oscillation of the lalos is only dependent on the radius of curvature of the dipped field lines that support the plasma .
this very good match between the numerical simulations and analytic expressions shows that the plasma contained at different heights is oscillating along the magnetic field and gravity is the restoring force .
the zig - zag shape of the oscillating prominence is entirely associated to the different periods of oscillation along the vertical structure that produces phase differences between the motion of the different flux - tubes .
in this work we have performed non - linear time - dependent numerical simulations of lalos considering a dipped magnetic field line structure .
the motion is very complex but we have not found evidences of strong coupling between longitudinal and transverse motions of the plasma .
this indicates that the longitudinal motion of the plasma along the magnetic structure does not produce significant vertical motions of the field due to back reaction .
we have also demonstrated that lalos are very well described by the pendulum model .
the periods of the numerical simulations agree very well with the results of the analytical model .
this demonstrates that there is one to one relation between the period of the oscillations and the radius of curvature of the field lines .
this result has important observational implications because it is possible to easily infer the geometry of the filaments when lalos are present in such structures .
m. luna , e. khomenko , m. collados and a. de vicente acknowledge the support by the spanish ministry of economy and competitiveness through projects aya2011 - 24808 , aya2010 - 18029 and aya2014 - 55078-p .
this work contributes to the deliverables identified in fp7 european research council grant agreement 277829 , `` magnetic connectivity through the solar partially ionized atmosphere '' ( pi : e. khomenko ) .
j. t. acknowledges support from the spanish `` ministerio de educacin y ciencia '' through a ramn y cajal grant and support from mineco and feder funds through project aya2014 - 54485-p . m. l. and j. t. acknowledge support from the international space science institute ( issi ) to the team 314 on `` large - amplitude oscillation in prominences '' led by m. luna .
the authors acknowledge the partial contribution of teide high - performance computing facilities to the results of this research .
teidehpc facilities are provided by the `` instituto tecnolgico y de energas renovables '' ( iter , sa ) .
url : http://teidehpc.iter.es .
resources partially supporting this work were provided by the nasa high - end computing ( hec ) program through the nasa center for climate simulation ( nccs ) at goddard space flight center . | large - amplitude longitudinal oscillations ( lalos ) in prominences are spectacular manifestations of the solar activity . in such events
nearby energetic disturbances induce periodic motions on filaments with displacements comparable to the size of the filaments themselves and with velocities larger than @xmath0 .
the pendulum model , in which the gravity projected along a rigid magnetic field is the restoring force , was proposed to explain these events .
however , it can be objected that in a realistic situation where the magnetic field reacts to the mass motion of the heavy prominence , the simplified pendulum model could be no longer valid .
we have performed non - linear time - dependent numerical simulations of lalos considering a dipped magnetic field line structure . in this work
we demonstrate that for even relatively weak magnetic fields the pendulum model works very well .
we therefore validate the pendulum model and show its robustness , with important implications for prominence seismology purposes . with this model
it is possible to infer the geometry of the dipped field lines that support the prominence . |
the orbital decay of massive black hole pairs from galactic scale separations to distances at which gravitational wave ( gw ) emission takes over is governed by a number of different physical mechanisms .
in particular the decay has a different nature in a predominantly gaseous background as opposed to a predominantly stellar background .
these two regimes are usually studied separately , although galactic nuclei in a galaxy host might transition from one regime to the other even multiple times in their evolution as a result of galaxy mergers , starvation of cosmological gas accretion , disk instabilities , and any process that can significantly affect the gas supply and star formation rate in galactic nuclei .
the last decade has seen significant improvements in the realism of the numerical models dedicated to tackle one or the other regime , and even the first attempts to model both regimes in the same calculation ( khan et al .
2013a ) .
two important general results can be drawn from the studies of the last few years .
first , it is now largely agreed that the last parsec problem that once plagued the stellar background calculations does not exist in a realistic galaxy host , namely one with a triaxial , rotating stellar distribution ( berczik et al .
2006;khan et al . 2011;2012;2013b
; vasiliev , antonini & merritt 2015 ) .
one of the most convincing demonstrations comes from a recent analysis of orbital diffusion in the loss cone using different types of potentials , including triaxial ones , with increasing number of particles in collisionless simulations ( gualandris et al .
loss cone refilling thus appears to be effective enough in realistic galaxy hosts , allowing a successful hardening via 3-body encounters down to separations at which gravitational waves take over .
recently cosmological hydrodynamical simulations have confirmed that realistic merger remnants , in which gas dissipation and star formation takes place , do possess the required degree of triaxiality to keep the loss cone nearly full and drive coalescence ( khan et al .
2016 ) .
the second result is that the decay in a gaseous background , once believed to be much faster than in a stellar background ( eg escala et al .
2005 ; dotti , colpi & haardt 2006 ; mayer et al .
2007 ) , turned out to be a much more complex problem than initially suspected , with the consequence that there is no simple general scenario at the moment . at variance with the stellar background case ,
in which dynamical fiction and 3-body encounters are well defined and well understood sources of drag , in gaseous backgrounds different types of torques can dominate at different scales and in different stages of the orbital evolution , such as dynamical friction , linear and nonlinear torques from a background circumnuclear rotating disk ( cnd ) , stochastic torques from massive clouds / clumps and large - amplitude spiral density waves , and tidal torques by circumbinary disks at the smallest separations ( see eg mayer 2013 , roedig et al . 2012 ; farris et al 2015 ) . the stochastic torque regime due to perturbations in an inhomogeneous interstellar medium ( ism ) , in particular , was pointed out only recently , and can delay or shorten orbital decay by orders of magnitude relative to the smooth disk case ( fiacconi et al . 2013 ; mayer 2013 ) . finally , even in the simple case of a smooth cnd there are indications that the orbital decay may stall , or at least slow down significantly , once the separations of the two holes drops below a tenth of a parsec ( mayer et al . 2008 ; chapon et al .
2013 ) . as a result of the multiple regimes and physical parameters of galactic nuclei that can affect orbital decay , it is fair to say that black hole merger timescales are really uncertain overall , with values ranging from as small as 10 myr ( eg khan et al .
2016 ) to as large as a few billion years ( khan et al .
2011;2012 ) depending on the specific regime and physical conditions considered .
indeed in this article we focus on this pivotal issue of timescales as it is highly relevant to define the science that will be possible with lisa .
in the last few years a number of works , both numerical and semi - numerical , have provided evidence that there is no last - parsec problem in realistic galactic potentials , where non - axisymmetry and some degree of rotation are always present ( berczik et al .
2006 ; vasiliev , antonini & merritt 2015 ; khan et al . 2013b
; holley - bockelmann & khan 2015 ) .
although convergence of the hardening rates is not yet demonstrated in collisional n - body simulations ( but see holley - bockelmann & khan 2015 on highly rotating systems ) , alternative approaches such as the numerical convergence study of orbital diffusion in the loss cone using collisionless n - body simulations ( gualandris et al .
2016 ) strongly support the notion that the loss cone should remain nearly full in realistic galaxy merger remnants .
this in turn yields confidence in the results of collisional n - body simulations of black hole binary hardening in galaxy mergers at the currently affordable resolutions ( less than @xmath5 particles , see eg khan et al .
2012 ; holley - bockelmann & khan 2015 ) . on the left
we show a grey - scale luminosity map ( in the uvj filter ) from the argo cosmological simulation showing the galaxy merger that was selected and re - simulated at much higher resolution ( green circles ) .
the white circle marks the virial radius of the group - sized halo , while the green circles mark the merging galaxies . from the spatial scale
it is clear how the galaxies at study , which have stellar masses almost as large as those of our present - day milky way ( @xmath6 ) have sizes below 1 kpc , hence 4 - 5 times smaller than our milky way .
this is expected based on the scaling arguments described in the text . on the right
we show the complete orbital separation curve for the two massive black holes at the center of the re - simulated merger with pc scale resolution , ending with the in - spiral driven by gw emission ( green dashed line ar 32 myr ) .
we plot also the results of simulations without the direct n - body integration , that captures the hardening phase , as well as with direct n - body but without the post - newtonian corrections ( blanchet 2006 ) , which are relevant in the very last phase .
we refer to khan et al .
( 2016 ) for details . ]
the latter simulations find that massive black hole binaries can merge on timescales shorter than an hubble time .
a particularly successful case is that of ultramassive black holes in stellar dominated hosts , for which dynamical friction is so effective to bring them all the way to the gravitational wave emission phase without the need of a long hardening stage ( khan & holley - bockelmann 2015 ) .
however , with this last exception , all published timescales are really long , ranging from slightly below 1 to a few gyr , namely they encompass a significant fraction of the hubble time .
note that the latter result is consistent with the simple analytical predictions of gould & rix ( 2000 ) , which were assuming efficient loss cone refilling in a potential with the properties of present - day spheroids .
n - body simulations mostly consider black holes at the high mass end , @xmath7 , hence the coalescence timescale for lisa black holes , which are much lighter , would be even longer , assuming the same properties of the ( stellar ) background hold . if merging timescales of a few gyr are the norm at any redshift , the consequence would be that black hole mergers would be exceedingly rare at @xmath8 since the lookback time is a less than 3 gyr then
. this would be unfortunate since this is also the epoch for which the discovery potential of lisa is very high .
the inefficient coalescence in stellar backgrounds has been and still is the motivation behind exploring the alternative regime in which orbital decay is driven by gasdynamical processes .
however , as we will illustrate in section 3 , coalescence does not become necessarily faster in gaseous backgrounds .
therefore it is fair to say we are facing a _ timescale problem _ rather than a _
last parsec problem_. we ought to make progress in order to obtain robust forecasts black hole coalescence rates in the lisa band .
uncertainties in the black hole seeding models would also generate important variations in the expected event rates ( sesana , volonteri & haardt 2007 ) . yet
black hole seeding models tend to yield different black hole population properties only at at very high redshift , while afterwards the physics of black hole growth via gas accretion and mergers with other black holes become the dominant driver of the evolution of the black hole mass function ( @xmath9 , see eg volonteri & natarajan 2009 ; bonoli et al .
2014 ) .
sesana & khan ( 2015 ) have pointed out that the hardening timescale is determined by the longest decay phase at sub - pc separations .
this is the phase associated with the transition between dominance of 3-body encounters and dominance of energy loss by gravitational wave emission .
the two processes have indeed rates proportional to @xmath10 and @xmath11 , respectively , where @xmath12 is the semi - major axis ( hereafter separation ) of the binary . by defining a characteristic separation @xmath13
at which the hardening rate by 3-body encounters and the hardening rate by gravitational wave emission are equal one obtains the lifetime of the binary at sub - pc scales as that corresponds to the orbital decay time in the longest phase of the decay .
following sesana & khan ( 2015 ) .
for the latter characteristic decay timescale one finds : @xmath14 where @xmath15 and @xmath16 are , respectively , the characteristic stellar density and mean ( 1d ) stellar velocity dispersion in the galactic nucleus , @xmath17 is an adimensional parameter expressing how close the loss cone is to full ( @xmath18 for a full loss cone , @xmath19 for triaxial systems employed in numerical simulations such as in khan et al .
the dependence on density and stellar velocity dispersion is linear .
however this equation has no apparent connection with cosmology , although the parameters that appear in it need to be informed by cosmology in modern galaxy formation theory .
elucidating such connection is the main purpose of this section .
we start by recalling that in the galaxy formation scenario arising in a universe dominated by cold dark matter ( cdm ) cosmic structures form bottom - up via collapse of density fluctuations combined with hierarchical merging .
the characteristic density of a dark matter halo is determined by the virial overdensity @xmath20 . inside the region whose mean overdensity is the virial overdensity a halo is virialized , namely there is a well defined equlibrium structure for which the binding energy is set . in the simple spherical collapse model
the virial overdensity can be calculated exactly once the cosmological parameters are known ( eg bryan & norman 1998 ) .
often for simplicity the redshift - independent overdensity defined to be 200 times the critical density of the universe , @xmath21 , is adopted . while such overdensity is only strictly valid in a universe with matter density parameter @xmath22 , quantitative statements concerning redshift evolution of structures are very weakly dependent on the presence or not of a cosmological constant ( with corresponding density parameter @xmath23 ) until it begins to be dominant at @xmath24 .
this is because the virial ovedensity varies only by factors of 2 as a function of redshift relative to @xmath21 ( bryan & norman 1998 ) . in the remainder of this section
we will thus drop @xmath23 since equations are simplified considerably , and afterwards we will comment on the magnitude of the eventual corrections . inside a virialized
halo baryons would collapse and form a luminous galaxy .
hence , as it is customary , one can assume that scaling relations between key galaxy structural parameters , such as mass , radius , circular velocity ( or velocity dispersion ) , and thus also characteristic density , directly reflect those of the virialized host halo because this comprises most of the mass , and therefore most of the binding energy ( white & rees 1978 ; mo , mao & white 1998 ) .
galaxy size will depend also on the amount of angular momentum , but this is also assumed to be proportional to the angular momentum of the halo and to be approximately conserved , introducing a simple scaling factor between galaxy size and halo virial radius .
the virial overdensity @xmath20 is , by definition , a multiple of the mean density of the universe at the corresponding epoch , thereby it has to increase as @xmath25 towards high redshift . with our assumption that @xmath26 we are simply enforcing that the multiplication factor is constant over time .
as a consequence , _ if the characteristic density of galaxies follows the halo overdensity then it should also scale as @xmath25 .
_ for instance at @xmath0 galaxies should be approximately @xmath27 times denser relative to their @xmath28 counterparts .
while this is essentially textbook structure formation one should appreciate the very important implications in the context of massive black hole binary coalescence .
_ indeed based in eq .
( 1 ) this means that coalescence timescales can be shorter by two orders of magnitude at @xmath0 relative to @xmath28 .
the latter is a significant effect .
_ this simple approach neglects two aspects .
the first one is that , based on the same scaling relations between halos and galaxies , also the stellar velocity dispersion , which enters in equation ( 1 ) , depends on redshift , as we will illustrate in a moment .
the second one is that baryonic physics can decouple to some extent the dynamics of the baryons from that of the dark matter .
hence the scaling of halo and galaxy properties is not necessarily homologous ( although cosmological hydro simulations show it is so to some degree , see eg sokolowska et al .
2016 ) . indeed modern theory of galaxy formation is essentially based on the effort to try achieve the necessary decoupling suggested by observational constraints on galaxy properties via what we collectively indicate as `` feedback mechanisms '' . within the same framework ,
radiative cooling is also an obvious process that tends to decouple baryons from the collisionless dark matter fluid , with a tendency to increase density beyond the the inference of the scalings just described because important coolants such as recombination and atomic radiative transitions increase their rate with increasing density .
these processes will shape the mass distribution in galaxies , hence they will determine the density profile as a function of radius .
we will return to this second point in section 3 but for now we can continue our reasoning as if galaxies were sufficiently described by a characteristic density , i.e. neglecting their actual mass profiles . assuming an isothermal sphere for the velocity distribution we can move on easily with our calculations .
this is an oversimplification given that ( 1 ) the nfw model , that is conventionally used to describe cdm halos , is close to the isothermal law only in the intermediate region , and that ( 2 ) both halos and galaxies are not sppherical so they will possess significant orbital anisotropy . yet here we want to capture the essential notions and scalings rather than exact numbers . hence continuing with the isothermal model one can write that the 1d velocity dispersion is @xmath29 , where we impose that the ( constant ) circular velocity @xmath30 is equal to the virial velocity @xmath31 of the dark matter halo .
the latter is automatically determined by the halo virial mass in dissipationless spherical collapse theory once the virial overdensity and the reference cosmological parameters are set .
the virial radius is also automatically determined in the spherical collapse model as the radius containing a region with the virial overdensity , eg the radius at which the density equals @xmath21 .
the relation between halo virial mass @xmath32 and virial velocity @xmath31 is ( see mo , mao & white 1998 ) : @xmath33 the redshift dependence is contained in the hubble parameter @xmath34 which , in a generic cosmology with non - zero cosmological constant , is defined as : @xmath35}^{1/2}}\ ] ] where @xmath36 and @xmath37 are , respectively , the total density parameter and the cosmological constant density parameter at @xmath28 , and @xmath38 is the hubble constant . assuming @xmath39 and @xmath40 in equation ( 3 ) and substituting the result into equation ( 2 ) one obtains the following scaling of the velocity dispersion with redshift for a fixed halo virial mass @xmath32 : @xmath41 assuming
the mass of the galaxy is simply proportional to halo mass , as strongly suggested by abundance matching ( eg behroozi et al .
2013 ) , this simple scaling relation can also be used for the galaxy stellar velocity dispersion as long as the galactic potential does not modify greatly the relation between velocity and mass ( which it can not if the galaxy and halo are in equilibrium ) . at @xmath0
the relation implies an increase in velocity dispersion of a factor of @xmath42 relative to @xmath28 as opposed to an increase in density of more than a factor of a 100 ( see above ) .
bringing the cosmological constant back in inside equation ( 3 ) would only change the result by about @xmath43 . hence ,
if we assume that the stellar velocity dispersion @xmath16 is of order of the velocity dispersion of the halo , @xmath44 , we can safely state that the main effect as redshift increases is that the increasing density should shorten the hardening timescale almost as fast as the density increases , namely as @xmath45 .
the previous discussion is instrumental to understand the results of recent simulation work .
khan et al . (
2016 ) have made the first attempt to use a a lcdm galaxy formation simulation as the initial condition to follow the evolution of a massive black hole binary into the gravitational wave emission regime .
they selected the most massive pair of galaxies within the high resolution sub - volume of the cosmological zoom - in hydrodynamical simulation argo ( fiacconi & feldmann 2015 ; fiacconi , feldmann & mayer 2015 ) , having stellar masses of a few times @xmath46 in halos with @xmath47 .
then they increased the resolution further in two steps using particle splitting .
finally they converted into stars the very little residual gas after the major merger in order to continue the simulation as an n - body collisional calculation with post - newtonian corrections ( blanchet 2006 ) .
uncertainties in baryonic processes , especially feedback , are of course present .
nevertheless the galaxy merger was chosen after verifying that it produces a galaxy whose key structural parameters match the available observational constraints , such as the relation between stellar mass and halo mass and the scaling relations between galaxy effective radius and its stellar velocity dispersion ( feldmann & mayer 2015 ) . in particular , the characteristic densities and stellar velocity dispersions , which enter eq ( 1 ) , match reasonably those inferred for observed massive galaxies at high redshift ( szomoru et al . 2012 ; bezanson et al . 2013 ) .
the selected merger starts at @xmath48 and is over by @xmath49 ( figure 1 , left panel ) .
wave - form for the massive black hole merger simulated in khan et al .
( 2016 ) , assuming a luminosity distance ( 32 gpc ) corresponding to @xmath48 in the concordance lcdm cosmology .
the lisa sensitivity curve is also shown in red , according to klein et al .
( 2016).the merger should be detected by lisa in the last phase of the in - spiral . ]
the timescale that we obtain for the coalescence after the galaxy merger is completed is surprisingly small , about 10 myr , and corresponds to the duration of the hardening phase , as the previous phase in which the two galactic cores merge by dynamical friction inside the common halo lasts only about a myr ( figure 1 , right panel ) .
can we explain such a short timescale based on the scaling arguments just outlined ?
the answer is yes . indeed we attempted the rescaling procedure suggested by the above equations in the non - cosmological models of khan et al .
note that khan et al ( 2012 ) used dehnen models with with three different inner slopes ( with exponents -1.5 , -1 and 0.5 , respectively ) .
we found that the merger remnants of models with the steeper inner slope provide a better fit to the cosmological merger remnant of khan et al .
( 2016 ) , which reflects the effect of gas dissipation increasing the central mass density prior to the completion of the merger .
hence we first rescaled those remnants to the same mass of the khan et al .
( 2016 ) cosmological merger remnant , and then rescaled the effective radius according to the expected redshift evolution so that the density was increased by a factor @xmath45 ( the original models were constructed to represent at @xmath28 galaxies ) .
using the characteristic density of the rescaled models in equation ( 1 ) we obtained an hardening timescale of about 30 myr , as opposed to almost a gyr in the mergers using the original models adopted in khan et al .
( 2012 ) , just slightly higher than the timescale we find in the cosmological simulation .
this supports the notion that density scaling with redshift is the main reason behind our short timescales as a similar result can be obtained starting with simple idealized models once these are appropriately rescaled to account for the redshift - dependent effects ( but note that we did not scale velocity dispersion ) .
the very central density and stellar velocity dispersion are the product of the balance between gas cooling and heating / stirring by feedback processes occurring during a nuclear starburst driven by the merger ( see eg capelo et al .
2015 for a recent large simulation survey of gas - rich galaxy mergers ) .
the absence of black hole accretion and agn feedback in the simulations of khan et al .
( 2016 ) should not be a worry because we verified the local relation between black hole mass and central stellar velocity dispersion is reproduced .
furthermore , as already mentioned , central velocity dispersions are reasonable when compared to observed massive galaxies at @xmath8 with similar luminosities . as there is mounting evidence that , at high z , black hole masses are shifted upwards relative to the relation , probably suggesting a lesser role of agn feedback in regulating the concurrent growth of host galaxy and black hole masses ( trakhtebroth et al .
2015 ) , the black hole masses chosen in these simulation are probably on the conservative side . in lower mass galaxies ,
in which feedback has a stronger effect due to their shallower potential well , it is conceivable that the characteristic nuclear density might be reduced as a result of the starburst , an effect that has indeed been studied extensively for dwarf galaxies in cosmological hydrodynamical simulations . in this scale
the evolution of the characteristic baryonic density with redshift might depart from the simple scaling arguments based on the dark halo evolution . in progenitors of milky way - sized galaxies ,
whose mass would be an order of magnitude lower than the current mass of the milky way at @xmath28 at @xmath50 ( see eg guedes et al .
2011;agertz & kravtsov 2015 ) , the conditions in the host might be just intermediate .
the study of black hole growth of a milky way - like galaxy carried out by bonoli et al .
( 2016 ) using the erisbh simulation does indeed reveal that nuclear gas inflows are rather weak even at @xmath51 , when the last significant mergers occur .
this reflects the fact that feedback processes are very effective at counteracting radiative cooling in this case . since such milky - way progenitors would likely host black holes in the range @xmath52 , the typical target of lisa ( amaro - seoane et al .
2013 ) , massive black hole binary evolution in this type of galaxies should be studied thoroughly in the future .
the previous section focused on the sinking and coalescence of supermassive black holes ( @xmath53 ) in massive galaxies , a category of objects that will be only marginally in the detection window of lisa . in the early phase of their in - spiral they would
, however , provide a very strong signal , with strong signal - to - noise ratio , at relatively low redshifts ( @xmath54 ) as in the case of the system studied by khan et al .
( 2016 ) ( see fig.2 ) .
furthermore , the results presented in the last section are mostly relevant to gas - poor galaxies . the galaxy host merger remnant in khan et al .
( 2016 ) indeed belongs to the category of galaxies that leave the star forming main sequence and become `` red and dead '' relatively early , at @xmath8 , such as the red nuggets identified in the candels survey ( barro et al .
2013 ) . while it is generally believed that more massive galaxies quench their star formation earlier , galaxy evolution is a nonlinear process which occurs along multiple tracks , hence even among massive galaxies star formation may continue or even peak at an epoch later than that of the merger remnant of khan et al .
indeed at @xmath50 we witness the existence of massive star forming galaxies that are very clumpy and disky ( tacconi et al . 2013 ; forster - schreiber et al .
2011 , wisnioski et al .
this has often interpreted as evidence for massive , self - gravitating gas disks that fragment due to a toomre - like instability ( bournaud et al . 2010 ; ceverino et al . 2010 ) . while the quantitative aspects of gravitational instability of massive gas disks at high - z are currently debated and
may strongly depend on feedback processes ( tamburello et al .
2015 ; mayer et al . 2016 ) , it is important to ask how would massive black hole pairs evolve and sink in such clumpy disks . in particular , what would be the typical timescales for massive black hole coalescence compared to those in relatively smooth galactic disks at low redshift , or to those in gas - poor galaxies dominated by the stellar background physics ? on the left we an edge - on color - coded density map of one of the clumpy hi - z galaxies from tamburello et al .
( 2017 ) , showing the ejection of the secondary black hole out of the disk plane ( black dot on top left corner of panel ) . on the right
we show a few examples of how the orbital separation of the secondary bh relative to the primary evolves with time .
all runs shown have bh accretion and feedback except the one used in the top panel , which only includes bh accretion .
parameters related to the details of the accretion and feedback sub - grid models differ in the various runs ( see labels in figure and tamburello et al . 2017 ) .
as it is seen there is a high degree of stochasticity in the orbital separation evolution , but in most cases the secondary keeps wandering at large distances even after a billion years from the beginning of the simulation ( but see tamburello et al .
( 2017 ) on the effect of a massive extended stellar bulge).,title="fig : " ] on the left we an edge - on color - coded density map of one of the clumpy hi - z galaxies from tamburello et al .
( 2017 ) , showing the ejection of the secondary black hole out of the disk plane ( black dot on top left corner of panel ) . on the right
we show a few examples of how the orbital separation of the secondary bh relative to the primary evolves with time .
all runs shown have bh accretion and feedback except the one used in the top panel , which only includes bh accretion .
parameters related to the details of the accretion and feedback sub - grid models differ in the various runs ( see labels in figure and tamburello et al .
as it is seen there is a high degree of stochasticity in the orbital separation evolution , but in most cases the secondary keeps wandering at large distances even after a billion years from the beginning of the simulation ( but see tamburello et al .
( 2017 ) on the effect of a massive extended stellar bulge).,title="fig : " ] in tamburello et al . (
2017 ) we considered the decay of two massive black holes that are already embedded in a gas - rich massive galactic disks subject to fragmentation into massive star forming clumps .
we considered also the effect of black hole accretion and feedback , as well as star formation and feedback from supernovae explosions .
we explored a large suite of simulations with galactic hosts having different masses and gas fractions , and choosing different eccentricities for the orbit of the secondary black hole ( the mass ratio is fixed of the two black holes to 10:1 , which should be statistically representative as the same mass ratio is most typical for halo / galaxy mergers ) .
the main result is that the lighter black hole , with mass in the range @xmath55 ( @xmath56 of the mass of the primary ) experiences repeated gravitational scattering by the most massive clumps as well as by strong spiral density waves often this often results in its ejection from the disk plane , slowing down its orbital decay by at least an order of magnitude as dynamical friction is suppressed in the low density envelope around disk midplane ( fig .
3 ) . in a companion paper ( tamburello et al .
2016 ) we have shown how such clumps do reproduce the properties of observed star forming clumps in high - z massive galaxies once appropriate corrections for observational biases ( limited sensitivity and resolution ) are taken into account .
the suppression of orbital decay is exacerbated when agn feedback is included as both dynamical friction and disk torques are weaker even when the secondary is still in the disk plane due to local heating of the gaseous background .
since it sinks slower , the secondary can be ejected when it is further away from the primary in this case .
the secondary , following an ejection , can still wander at distances greater than 1 kpc more after 1 gyr .
decay has started .
taken at face value this would imply a possible complete abortion of the coalescence process .
however , since tamburello et al used galaxy models that lack the an extended bulge / spheroid component , dynamical friction was underestimated away from the galactic disk plane . by computing analytically the extra drag resulting from an hernquist spheroid with realistic structural parameters
, it was found that , if no more ejections occur , , wandering secondaries should coalesce on timescales of order 1 billion year .
this is still a very long timescale , comparable , if not longer , than the typical merging timescales of galaxies at @xmath8 .
the clumpiness of the ism is an important factor even for the late stage of black hole pair decay , at 10 pc scales and below , near the phase in which a bound binary forms .
this applies to gas - rich galactic hosts at any redshift as circumuclear disks ( cnds ) of gas and stars are ubiquitous at low z in merger remnants as well as in the central regions of seyfert galaxies ( medling et al .
2014 ; izumi et al .
2016 ) , which structurally typical late - type spiral galaxies .
fiacconi et al ( 2013 ) carried out the first study of black hole pairs in cnds , documenting the dramatic effect of gravitational scattering by clumps and spiral density waves that we just illustrated .
they considered black holes right in the lisa window ( @xmath57 ) .
they also found that in some cases the decay can be accelarated due to the capture of a massive gas cloud ( @xmath58 ) by the secondary .
they proposed the general notion that orbital decay is _ stochastic _ in a clumpy cnd , with orbital decay timescales to @xmath59 pc separations ranging from a few myr to as much as 100 myr .
the shortest timescales agree with previous results for smooth cnds ( see escala et al . 2005 , dotti et al .
2006;2007 , mayer 2013 ) .
statistically , though , a largest fraction of the simulations resulted in long timescales , in line with the same findings of tamburello et al .
( 2017 ) , especially for black holes on eccentric orbits .
these simulations were quite idealized in the physical treatment , lacking feedback mechanisms in particular , which should play a major role in cnd , regions that are often subject to starbursts .
del valle et al .
( 2015 ) and souza lima et al .
( 2016 ) recently carried out a similar studies but included a much richer inventory of physical processes .
del valle et al .
( 2015 ) included star formation and a weak form of sn feedback , while souza lima et al .
( 2016 ) employed a popular effective sub - grid model of sn feedback , the blastwave feedback ( stinson et al .
2006 ) , which is known to produce realistic stellar masses and disk sizes in galaxy formation simulations ( guedes et al 2011 ) , and even added black hole accretion and agn feedback in the same way as tamburello et al .
they confirmed the stochastic orbital decay picture and the prevalence of suppressed orbital decay with timescales in the range 50 - 100 myr .
the same long orbital decay timescales were found in a complimentary study modeling the entire massive black pair evolution before and after a major galaxy merger .
this study also addresses the formation of the circumnuclear disk after the merger , but was restricted to only one initial condition for the galaxy merger due to the high computational burden introduced by deep the multi - scale nature of the calculation ( roskar et al .
souza lima et al . (
2016 ) is the only work , however , that includes also the effect of bh accretion and agn feedback at cnd - scales .
thanks to that they uncovered a new process , dubbed `` wake evacuation effect '' , by which the secondary black hole carves a hole in the outer , lower density region of the cnd , as it launches a hot pressurized bubble resulting from agn feedback ( figure 5 ) .
the result is that dynamical friction and disk torques are stifled as the cnd - black hole dynamical coupling is temporarily stifled by the presence of a large cavity . as a result ,
the secondary decays slower even before a strong interaction with a clump or spiral wave occurs , often resulting in an ejection when it is still far from the center . in the runs with agn feedback
the outcome is thus an even longer delay , of order a gyr , which however should be regarded as an upper limit in absence of an extended massive stellar spheroid , for the same argument made in tamburello et al .
2017 in the case of high - z galactic disks .
when dynamical friction by an extended spheroid is added , for realistic parameters we obtained a decay timescale ( to fraction of parsec separation , namely close to the resolution limit ) of order a a few @xmath60 yr .
this is still almost two orders of magnitude longer than the decay timescale in a smooth cnd and no feedback processes included ( see eg mayer 2013 ) .
these effects are only some of those that are possible in a complex multi - phase cnd .
for example , recently park & bogdanovic ( 2017 ) have shown that radiation pressure can also produce an extra drag pulling opposite to the dynamical friction wake , which again slows down the orbital decay . in summary ,
the presence of a clumpy ism as well as the concurrent effect of feedback processes complicate significantly the picture in gas - rich systems , suggesting that timescales in the range @xmath61 yr are likely for the decay to sub - pc separations from galactic scales , with the longer timescales more likely in the high - z clumpy galaxy phase . despite uncertainties
it is clear that , at high z , such timescales are much longer than those in gas - poor galaxies in which the decay is governed by the physics of the stellar dynamical background . at low z , instead
, they turn out to be comparable or shorter than those in purely stellar backgrounds , as the simulations of khan et al . ( 2012 ) found a typical timescale of order a gyr for the hardening phase by 3-body encounters .
the latest developments in modeling the orbital decay of massive black hole binaries that we just illustrated have turned upside - down the conventional notion that orbital decay of massive black hole binaries should be faster and more efficient in gas - rich galaxies relative to gas - poor galaxies . in particular at high reshift , while simulations finally show fast coalescence for massive black holes in gas - poor merger remnants of massive galaxies , with timescales at least an order of magnitude shorter than the merging time of galaxies themselves ( @xmath5 yr as opposed to @xmath61 yr ) , the opposite seems to be the case for similarly massive gas - rich disks due to the perturbing action of the their inhomogeneous , clumpy interstellar medium .
the existence of such clumpy medium is an inevitable consequence of the self - gravity of massive gas disks , although detailed properties such as the mass spectrum of cold clouds depend on the complicated interplay between gas physics and feedback processes . in these systems
orbital decay timescales acquire a stochastic character , with values ranging from @xmath60 yr to @xmath62 gyr , namely longer than the galaxy merging timescales at @xmath63 . these estimates still do no account for the final phase of decay below @xmath64 pc due to the resolution limitations of current multi - scale hydrodynamical simulations .
a word of caution must be spent in that there is still significant uncertainty in how to model feedback processes in galaxy formation and evolution , which in turn has a major impact on the nature of the ism .
the latest generation of strong feedback models , which are favoured by some constraints available on the rate of assembly of stellar masses in galaxies , do reduce significantly the clumpiness of the ism even in the most massive high redshift disks ( mayer et al .
2016 ; oklopcic et al .
while massive star forming clumps are observed in massive disks at high - z , their actual nature is debated , especially whether they are bound coherent structures or the collection of much smaller , weakly bound structures ( behrendt et al .
2016 ; tamburello et al .
this of course will have an impact on how strong are the dynamical perturbations that disk substructure can induce on the orbits of massive black holes . at low z
it is established and inevitable that the ism is clumpy at the scale of giant molecular clouds ( gmcs ) , namely at the scale of @xmath65 ( the largest cloud masses are observed in galactic nuclei , eg in our galactic center , see oka et al .
these cloud masses appear to be large enough to cause significant perturbations to decaying massive black holes with masses best accessible to lisa , namely @xmath66 , and their scattering action delays decay , in most cases leading to timescales @xmath67 yr to reach sub - pc scale separations ( souza lima et al .
more massive black holes would be unaffected as shown by simple analytical arguments ( fiacconi et al .
new effects caused by agn feedback , such as wake evacuation or anisotropic radiation pressure , can stifle the orbital decay further ( souza lima et al .
2016 ; park & bogdanovic 2017 ) .
the picture that is emerging is that the orbital decay of massive black holes is tightly linked to the properties of their galactic hosts , and to how they evolve in redshift .
it has never been so evident as it is now that understanding the evolution of massive black holes requires control of the background galaxy formation theory , which currently is developed primarily with supercomputer simulations of galaxy formation in a cosmological context .
this has two important implications , both crucial in preparation for lisa the first is that the potential of gravitational wave detection from massive black holes as a new probe of cosmic structure formation is much higher than we could ever suspect .
the second is that to fully exploit such potential we need to devote significant resources to supercomputer simulations addressing all possible regimes of orbital decay , from large to small scales , and ranging across a wide variety of possible galaxy hosts , especially at @xmath63 and for massive black holes below @xmath68 , the regime where the discovery potential of lisa is highest .
in essence the latest simulations reveal a high complexity of massive black hole orbital decay , with multiple regimes and multiple physical effects involved .
it is now clear that semi - analytical models describing evolving black hole populations in hierarchical galaxy formation , the tool that has been used so far to make predictions for lisa detections ( eg barausse 2012 ) need the input of such simulations to make reliable forecasts . without that black hole merger rates
could be off by orders of magnitude .
the same applies to recent models coupling recipes for black hole orbital decay with large scale cosmological hydro simulation that can not resolve better than kpc scales ( salcido et al .
2016).a joint effort of the different communities connected to the lisa science goals seems necessary to fill this gap and make future progress .
the author thanks his many collaborators , in particular fazeel khan , pedro capelo , rok roskar , andreas just , peter berczik , silvia bonoli , romain teyssier & jillian bellovary .
a special thank goes to current and former phd students , who have carried out a large part of the research described in this article , in particular simone callegari , davide fiacconi , valentina tamburello & rafael souza lima .
he acknowledges stimulating discussions over last years with monica colpi , piero madau , tamara bogdanovic , alberto sesana , scott tremaine , zoltan haiman , massimo dotti , andres escala & alessia gualandris .
amaro - seoane , p. , aoudia , s. , babak , s. , et al .
2013 , _ gw notes _ , * 6 * , 4 | we discuss the latest results of numerical simulations following the orbital decay of massive black hole pairs in galaxy mergers .
we highlight important differences between gas - poor and gas - rich hosts , and between orbital evolution taking place at high redshift as opposed to low redshift .
two effects have a huge impact and are rather novel in the context of massive black hole binaries .
the first is the increase in characteristic density of galactic nuclei of merger remnants as galaxies are more compact at high redshift due to the way dark halo collapse depends on redshift .
this leads naturally to hardening timescales due to 3-body encounters that should decrease by two orders of magnitude up to @xmath0 .
it explains naturally the short binary coalescence timescale , @xmath1 myr , found in novel cosmological simulations that follow binary evolution from galactic to milliparsec scales .
the second one is the inhomogeneity of the interstellar medium in massive gas - rich disks at high redshift . in the latter star forming clumps
1 - 2 orders of magnitude more massive than local giant molecular clouds ( gmcs ) can scatter massive black holes out of the disk plane via gravitational perturbations and direct encounters .
this renders the character of orbital decay inherently stochastic , often increasing orbital decay timescales by as much as a gyr . at low redshift
a similar regime is present at scales of @xmath2 pc inside circumnuclear gas disks ( cnds ) . in cnds
only massive black holes with masses below @xmath3 can be significantly perturbed .
they decay to sub - pc separations in up to @xmath4 yr rather than the in just a few million years as in a smooth cnd .
finally implications for building robust forecasts of lisa event rates are discussed . |
in the quest to understand hadron structure within qcd , small violations of fundamental symmetries play a vital role .
the sigma commutator , @xmath0 : @xmath1 \right ] \left| n\right\rangle = \bar{m}\left\langle n\right| \bar{u}u+\bar{d}d \left| n\right\rangle = \bar{m}\frac{\partial m_{n}}{\partial \bar{m } } \label{eqn : fh - commutator}\ ] ] ( with @xmath2 the two - flavour ( @xmath3=1 , 2 , 3 ) axial charge ) is an extremely important example of such a symmetry .
@xmath0 can not be accessed directly by experimental measurements . however , one can infer from world data a value of @xmath4 mev @xcite .
this result has been under some scrutiny recently due to the progress in new determinations of the pion - nucleon scattering lengths @xcite and new phase shift analyses @xcite .
the full lattice qcd calculations upon which our work is based involve only two active flavours , the heavier third flavour is approximated by a renormalisation of the strong coupling constant . as a guide , recent work suggests that the best value of @xmath0 may be 8 to 26 mev larger than the value quoted above @xcite . one can notionally use qcd to directly calculate the value of @xmath0 , but in practice the calculation has proven to be difficult .
early attempts @xcite to extract @xmath5 from the quark mass dependence of the nucleon mass ( using eq.([eqn : fh - commutator ] ) ) in quenched qcd with naive extrapolations produced values in the range 15 to 25 mev .
attention subsequently turned to determining @xmath5 by calculating the scalar matrix element of the nucleon @xmath6 .
there it was discovered that the sea - quark loops make a dominant contribution to @xmath5 @xcite .
these works , based on quenched qcd simulations found values in the 40 to 60 mev range , which are more compatible with the experimental values quoted earlier .
on the other hand , the most recent estimate of @xmath5 , and the only one based on a two - flavour , dynamical - fermion lattice qcd calculation , comes from the sesam collaboration .
they obtain a value of @xmath7 mev @xcite , through a direct calculation of the scalar matrix element @xmath8 and the quark mass @xmath9 .
the fact that neither @xmath8 , nor @xmath10 is renormalisation group invariant introduces a major difficulty in calculating the sigma commutator in this approach .
one must reconstruct the scale invariant result from the product of the scale dependent matrix element and the scale dependent quark masses .
the latter are extremely difficult to determine precisely and are the chief sources of uncertainty .
furthermore , since lattice calculations are made at quite large pion masses , typically above 500 or 600 mev , one needs to extrapolate , in the pion mass down to the physical value at 140 mev . an important innovation adopted by dong _
_ was to extrapolate @xmath11 using a form motivated by chiral symmetry , namely @xmath12 .
regrettably , the value of @xmath13 used was _ not _ constrained by chiral symmetry and higher order terms of the chiral expansion were not considered .
furthermore , since the work was based on a quenched calculation , the chiral behaviour implicit in the lattice results involves incorrect chiral physics @xcite .
our recent work @xcite was motivated by the improvements in computing power , together with the development of improved actions @xcite , which have led to accurate calculations of the mass of the nucleon within _ full qcd _
( for two flavours ) as a function of @xmath9 down to @xmath14 mev .
( since @xmath15 is proportional to @xmath9 over the range studied we choose to display all results as a function of @xmath15 . )
we showed that provided that one has control over the extrapolation of this lattice data to the physical pion mass , one can calculate @xmath0 from @xmath16 ( which is equivalent to eq .
( [ eqn : fh - commutator ] ) where @xmath9 was used ) at @xmath17 mev .
this approach has the important advantage that one _ only _ needs to work with renormalization group invariant quantities .
chiral perturbation theory ( @xmath18pt ) predicts that the leading non - analytic ( lna ) correction to the self energy contribution to the nucleon mass is proportional to @xmath19 ( or @xmath20 ) .
it can be seen in fig .
[ fig : fits to data ] that the preliminary point from cp - pacs @xcite at @xmath21 gev@xmath22 does indeed suggest some curvature in this low mass region .
these observations led the cp - pacs group to extrapolate their data with the simple , phenomenological form : @xmath23 rather than a naive linear form ( @xmath24 ) , as shown in fig .
[ fig : linear fits to data ] .
the corresponding fit to the combined data set , using eq .
( [ eqn : phen_form_cp - pacs ] ) , is shown as the short - dashed curve in figs .
[ fig : linear fits to data ] and [ fig : fits to data ] .
we found that this fit gives @xmath25 mev .
the difficulty with this purely phenomenological analysis was discussed in ref .
the problem is that a derivative is required when evaluating @xmath0 and the value of @xmath26 found in the fit ( -0.761 gev@xmath27 ) is almost an order of magnitude smaller than the model independent lna coefficient , @xmath28 gev@xmath27 , indicated by @xmath18pt .
recently , an alternative approach was suggested in ref .
there it was realised that the pion loop diagrams shown in fig .
[ fig : se - n?n ] yield not only the most important non - analytic structure , but also give rise to the most significant variation in the nucleon mass as @xmath29 .
this leads to the following extrapolation function for @xmath30 : @xmath31 where @xmath32 and @xmath33 are the self - energy contributions of figs .
[ fig : se - n?n](a ) and [ fig : se - n?n](b ) , respectively , using a cut - off in momentum controlled by @xmath34 .
the full analytic expressions for @xmath32 and @xmath33 are given in ref .
@xcite . for our purposes
it suffices that they have precisely the correct lna and next - to - leading non - analytic behaviour required by chiral perturbation theory as @xmath35 .
in addition , @xmath33 contains the correct , square root branch point ( @xmath36^{\frac{3}{2}}$ ] ) at the @xmath37 mass difference , which is essential for extrapolations from above the @xmath38 threshold .
fitting eq .
( [ eqn : full - eqn ] ) to the data , including the point near 0.1 gev@xmath22 , gives the dot - dash curve in fig .
[ fig : fits to data ] .
the corresponding value of @xmath0 is 54.6 mev and the physical nucleon mass is 870 mev .
omitting the lowest data point from the fit yields the long - dash curve in fig .
[ fig : fits to data ] with @xmath39 mev , demonstrating the need for lattice simulations of qcd at light quark masses .
the importance of the inclusion of the correct chiral behaviour is clearly seen by the fact that it increases the value of the sigma commutator from the 30 mev of the unconstrained cubic fit to around 50 mev .
nevertheless , it is a remarkable result that the present lattice data for two - flavour dynamical - fermion qcd , yields a stable @xcite and accurate answer for the sigma commutator , an answer which is already within the range of the experimental values .
99 j. gasser , h. leutwyler and m. e. sainio , _ phys .
lett . _ * b253 * , 252 ( 1991 ) .
d. sigg _ et al .
_ , _ nucl .
phys . _ * a609 * ( 1996 ) 269 ; + d. chatellard _ et al .
_ , _ nucl .
phys . _ * a625 * , 855 ( 1997 ) .
t. e. ericson , b. loiseau and a. w. thomas , nucl .
phys . * a663 - 664 * , 541 ( 2000 ) , hep - ph/9907433 .
m. m. pavan , r. a. arndt , i. i. strakovsky and r. l. workman , nucl - th/9912034 .
d. v. bugg , pin newsletter * 15 * ( 1999 ) 319 .
m. knecht , hep - ph/9912443 .
see for example , s. cabasino _
_ , _ nucl .
( proc . suppl . )
* 20 * , 399 ( 1991 ) .
s. j. dong , j. f. lagae and k. f. liu , _ phys .
rev . _ * d54 * , 5496 ( 1996 ) .
m. fukugita , y. kuramashi , m. okawa and a. ukawa , _ phys .
rev . _ * d51 * , 5319 ( 1995 ) .
collaboration , s. gusken _ et al .
_ , _ phys .
rev . _ * d59 * , 054504 ( 1999 ) . s.r .
sharpe , _ phys .
_ * d56 * , 7052 ( 1997 ) .
labrenz and s.r .
sharpe , _ phys .
rev . _ * d54 * , 4595 ( 1996 ) . d. b. leinweber , a. w. thomas and s. v. wright , hep - lat/0001007 . to appear in _ phys .
* b*. b. sheikholeslami and r. wohlert , _ nucl .
phys . _ * b259 * ( 1985 ) 609 ; + m. alford , w. dimm , p. lepage , _ phys .
lett . _ * b361 * , 87 ( 1995 ) ; + p. hasenfratz , _ nucl .
* b63 * , 53 ( 1998 ) .
collaboration , s. aoki _ et .
_ , _ phys .
rev . _ * d60 * , 114508 ( 1999 ) .
collaboration , c. r. allton _ et .
rev . * d60 * , 034507 ( 1999 ) .
d. b. leinweber , a. w. thomas , k. tsushima , and s. v. wright , phys .
rev . * d61 * , 074502 ( 2000 ) , http://xxx.lanl.gov/abs/hep-lat/9906027[hep-lat/9906027 ] . | as a direct source of information on chiral symmetry breaking within qcd , the sigma commutator is of considerable importance . since hadron structure is a non - perturbative problem , numerical calculations on a space - time lattice are currently the only rigorous approach . with recent advances in the calculation of hadron masses within full qcd ,
it is of interest to see whether the sigma commutator can be calculated directly from the dependence of the nucleon mass on the input quark mass .
we show that , provided the correct chiral behaviour of qcd is respected in the extrapolation to realistic quark masses , one can indeed obtain a fairly reliable determination of the sigma commutator using present lattice data . for two - flavour dynamical fermion
qcd the sigma commutator lies between 45 and 55 mev based on recent data from cp - pacs and ukqcd .
adp-00 - 24/t407 |
the standard model ( sm ) of elementary particle physics describes the structure of fundamental particles and how they interact via gauge bosons . to date , almost all experimental results have agreed with the prediction by the standard model .
however , many questions can be raised , which indicate that the standard model is not complete .
for example , `` why is there a hierarchy between the electroweak scale ( 1 tev ) and the gravitational scale ( @xmath3 tev ) ? '' , `` what are the origins of mass ? '' , `` why is there a spectrum of fermion masses ? are there only three generations ? '' , _
etc_. although the most popular extension of the standard model is supersymmetry ( susy ) , there are other equally well - motivated models , such as extra dimension , compositeness , 4@xmath4 generation , technicolor ,
_ etc_. in this document , we present results of signature - based searches and searches inspired by non - susy models , using 1.02.5 @xmath0 of data collected with the cdf and d0 detectors . in signature - based searches ,
we apply generic selection criteria in order to be sensitive to a wide range of new physics .
we report on the event counts and various kinematic distributions of data and predicted backgrounds . in model - inspired searches ,
we optimize selection criteria to obtain the best sensitivity for selected models .
if no significant excess is found , we report limits on model parameters .
the cdf collaboration has performed a signature - based search in the inclusive @xmath6 final state using 2.0 @xmath0 of data .
the @xmath6 signature raised great interest for two main reasons .
first , this final state has been predicted by several susy models event observed in run i @xcite.]@xcite , _
e.g. _ , the production of a chargino and a neutralino , when @xmath7 is photino - like and the lsp @xmath8 is higgsino - like , via the decay chain : @xmath9 .
second , the dominant backgrounds are mis - identifications of either the photon or the @xmath10-quark candidates and mismeasurements of the jet energy which induce @xmath11 not associated with unobserved neutral particles ( fake @xmath11 ) .
the sm processes which produce real @xmath6 are expected to contribute at most 2% .
therefore , a significant excess in data will be an indication of new physics .
events are required to have a central . ]
photon with transverse energy @xmath12 , at least two jets with @xmath13 and @xmath14 , at least one of the jets must be identified as originating from a @xmath10 quark ( `` @xmath10-tagged '' ) using the tight secvtx algorithm @xcite , and missing transverse energy @xmath15 .
figure [ fig : gbjmet ] shows the @xmath11 and dijet mass @xmath16 distributions from data and predicted background .
other kinematic distributions , such as jet multiplicity , @xmath17 of photon , @xmath17 of @xmath10-tagged jet , _ etc .
_ , have also been examined and no significant excess has been found .
the observed number of events in data is 617 , which is consistent with the expected number of background events , @xmath18 . [
cols="^,^ " , ]
the cdf and d0 collaborations have performed extensive signature - based searches and searches inspired by non - susy models .
we have not yet found significant excess in 1.02.5 @xmath0 of data .
however , the result of the cdf search for high - mass @xmath19 resonances is exciting : a 3.8 @xmath20 excess is observed in the region @xmath21 with a @xmath22-value of 0.6% .
the same analysis will be updated with more data .
in addition , similar searches in the @xmath23 channel by both cdf and d0 are expected in the near future and will help understanding whether the excess is an indication of new physics or a statistical fluctuation .
moreover , several novel detectors and techniques have been developed , such as the cdf em timing system , @xmath11 significance model , and the d0 em pointing algorithm .
these allow us to explore signatures which were considered difficult before . as more
data data are being collected , we expect many new and interesting results from both cdf and d0 .
the author wishes to thank the cdf and d0 exotic group conveners , b. brau , c. hays , t. adams , and p. verdier , for their suggestions of the presentation in the conference .
the author also would like to thank the following people for answering author s almost non - stop questions : r. culbertson , h.j .
frisch , d. krop , c. pilcher , s. wilbur , a. loginov , i. shreyber , j. adelman , a. pronko , m. goncharov , c. henderson , g. choudalakis , v. krutelyov , y. gershtein , y. maravin , b.r .
ko , a. das , k. hatakeyama , c. magass , p.h .
beauchemin , e. james , y. hu , y. nagai , w .- m .
yao , and s. bar - shalom .
this work has been supported by u.s .
department of energy .
99 f. abe , _ et al . _
( cdf collaboration ) , phys .
d * 59 * , 092002 ( 1999 ) ; f. abe , _ et al .
_ ( cdf collaboration ) , phys . rev .
lett . *
81 * , 1791 ( 1998 ) ; d. toback , ph.d .
thesis , university of chicago , 1997 . g. l. kane and s. mrenna , phys .
lett . * 77 * , 3502 ( 1996 ) .
s. ambrosanio , g. l. kane , g. d. kribs , s. p. martin , and s. mrenna , phys .
lett . * 76 * , 3498 ( 1996 ) ; s. ambrosanio , g. l. kane , g. d. kribs , s. p. martin , and s. mrenna , phys .
d * 55 * , 1372 ( 1997 ) . the leading - order ( lo ) production cross - section of @xmath24 has been estimated using the madgraph generator to be 0.076 pb . a @xmath25-factor to scale the lo to the nlo cross - section , @xmath26 , is used after a discussion with f. petriello and u. baur .
s. dimopoulos , s. d. thomas , and j. d. wells , nucl .
b * 488 * , 39 ( 1997 ) ; s. ambrosanio , g. d. kribs , and s. p. martin , phys .
d * 56 * , 1761 ( 1997 ) ; g. f. giudice and r. rattazzi , phys .
* 322 * , 419 ( 1999 ) ; s. ambrosanio , g. l. kane , g. d. kribs , s. p. martin , and s. mrenna , phys .
d * 55 * , 1372 ( 1997 ) .
w. m. yao , _ et al .
_ ( particle data group ) , j. phys .
g * 33 * , 1 ( 2006 ) .
l. randall and r. sundrum , phys .
lett . * 83 * , 3370 ( 1999 ) .
j. l. hewett and t. g. rizzo , phys .
* 183 * , 193 ( 1989 ) ; d. london and j. l. rosner , phys .
d * 34 * , 1530 ( 1986 ) .
v. m. abazov , _ et al . _
( d0 collaboration ) , phys .
lett . *
100 * , 091802 ( 2008 ) .
j. bagger , c. schmidt , and s. king , phys .
d * 37 * , 1188 ( 1988 ) ; p. h. frampton and s. l. glashow , phys .
b * 190 * , 157 ( 1987 ) .
k. d. lane and m. v. ramana , phys .
d * 44 * , 2678 ( 1991 ) ; e. eichten and k. d. lane , phys .
b * 327 * , 129 ( 1994 ) ; k. lane and s. mrenna , phys .
d * 67 * , 115011 ( 2003 ) .
j. bijnens , p. eerola , m. maul , a. mansson , and t. sjostrand , phys .
b * 503 * , 341 ( 2001 ) .
s. rolli and m. tanabashi , `` leptoquarks , '' j. phys .
g * 33 * , 1 ( 2006 ) and 2007 partial update for edition 2008 .
h. c. cheng and i. low , jhep * 0309 * , 051 ( 2003 )
. s. dimopoulos and l. susskind , nucl .
b * 155 * , 237 ( 1979 ) ; s. dimopoulos , nucl .
b * 168 * , 69 ( 1980 ) ; e. eichten and k. d. lane , phys .
b * 90 * , 125 ( 1980 ) .
j. c. pati and a. salam , phys . rev .
d * 10 * , 275 ( 1974 ) ; h. georgi and s. l. glashow , phys . rev .
lett . * 32 * , 438 ( 1974 ) .
b. schrempp and f. schrempp , phys .
b * 153 * , 101 ( 1985 ) ; w. buchmuller , r. ruckl , and d. wyler , phys .
b * 191 * , 442 ( 1987 ) , erratum - ibid .
b * 448 * , 320 ( 1999 ) .
t. scanlon , ph.d .
thesis , imperial colledge london , 2006 .
s. bar - shalom and a. rajaraman , phys .
d * 77 * , 095011 ( 2008 ) .
n. cabibbo , phys .
* 10 * , 531 ( 1963 ) ; m. kobayashi and t. maskawa , prog .
* 49 * , 652 ( 1973 ) .
s. bar - shalom , a. rajaraman , d. whiteson , and f. yu , arxiv:0803.3795 [ hep - ph ] .
a. a. affolder , _ et al . _
( cdf collaboration ) , phys .
d * 64 * , 032002 ( 2001 ) , erratum - ibid .
d * 67 * , 119901 ( 2003 ) ; a. abulencia , _ et al .
_ ( cdf collaboration ) , phys . rev .
d * 74 * , 072006 ( 2006 ) .
t. aaltonen , _ et al .
_ ( cdf collaboration ) , phys . rev .
lett . * 100 * , 041801 ( 2008 ) .
v. m. abazov , _ et al . _
( d0 collaboration ) , arxiv:0806.2223 [ hep - ex ] .
p. h. frampton , p. q. hung , and m. sher , phys .
rept . * 330 * , 263 ( 2000 ) .
t. han , z. si , k. m. zurek , and m. j. strassler , jhep * 0807 * , 008 ( 2008 ) ; m. j. strassler and k. m.
zurek , phys .
b * 651 * , 374 ( 2007 ) .
a. scott and d. stuart , _
http://www - cdf.fnal.gov / physics / exotic / r2a/20040826.mumulxy_longlifez/. _ | we present the results of searches for non - standard model phenomena , with focus on signature - based searches and searches driven by non - supersymmetry ( non - susy ) models .
the analyses use 1.02.5 @xmath0 of data from @xmath1 collisions at @xmath2 tev collected with the cdf and d0 detectors at the fermilab tevatron .
no significant excess in data has been observed .
we report on the event counts , kinematic distributions , and limits on selected model parameters . |
chronometry is the science of the measurement of time .
as the time flow of clocks depends on the surrounding gravity field through the relativistic gravitational redshift predicted by einstein @xcite , chronometric geodesy considers the use of clocks to directly determine earth s gravitational potential differences .
instead of using state - of - the - art earth s gravitational field models to predict frequency shifts between distant clocks ( @xcite , itoc project ) , the principle is to reverse the problem and ask ourselves whether the of frequency shifts between distant clocks can improve our knowledge of earth s gravity and geoid @xcite .
for example , two clocks with an accuracy of @xmath1 in terms of relative frequency shift would detect a 1-cm geoid height variation between them , corresponding to a geopotential variation @xmath2 of about @xmath3 ( for more details , see e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
until recently , the performances of optical clocks had not been sufficient for the determination of earth s gravity potential . in 2010
, @xcite demonstrated the ability of the new generation of atomic clocks , based on optical transitions , to sense geoid height differences with a 30-cm level of accuracy . to date , the best of these instruments reach a stability of @xmath4 ( nist , riken + univ .
tokyo , @xcite ) after 7 hours of integration time .
more recently , an accuracy of @xmath5 ( jila , @xcite ) has been obtained , equivalent to geopotential differences of 0.2 , or 2 cm on the geoid . .
such results the question of what can we learn about earth s gravity and mass sources using clocks , that we can not easily derive from existing gravimetric data .
recent studies address this question ; for example , @xcite discuss the value and future applicability of chronometric geodesy for direct geoid mapping on continents and joint gravity potential surveying to determine subsurface density anomalies .
they find that a geoid perturbation caused by a 1.5 km radius sphere with 20 per cent density anomaly buried at 2 km depth in the earth s crust is already detectable by atomic clocks with present - day accuracy .
they also investigate other applications , for earthquake prediction and volcanic eruptions @xcite , or to monitor vertical surface motion changes due to magmatic , post - seismic , or tidal deformations @xcite . here
we will consider the `` static '' or `` long - term '' component of earth s gravity .
our knowledge of earth s gravitational field is usually expressed through geopotential grids and models that integrate all available observations , globally or over an area of interest .
these models are , however , not based on direct observations with the potential itself , which has to be reconstructed or extrapolated by integrating measurements of its derivatives .
the potential is reconstructed with a centimetric accuracy at resolutions of the order of 100 km from grace and goce satellite data @xcite , and integrated from near - surface gravimetry for the shorter spatial scales . as a result ,
the standard deviation ( rms ) of differences between geoid heights obtained from a global high - resolution model as egm2008 , and from a combination of gps / leveling data , reaches up to 10 cm in areas well covered in surface data @xcite .
the uneven distribution of surface gravity data , especially in transitional zones ( coasts , borders between different countries ) and with important gaps in areas difficult to access , indeed limits the accuracy of the reconstruction when aiming at a centimetric level of precision .
this is an important issue , as large gravity and geoid variations over a range of spatial scales are found in mountainous regions , and because a high accuracy on altitudes determination is crucial in coastal zones .
airborne gravity surveys are thus realized in such regions @xcite ; local clock - based geopotential determination could be another way to overcome these limitations . in this context , here
, we investigate to what extent clocks could contribute to fill the gap between the satellite and near - surface gravity spectral and spatial coverages in order to improve our knowledge of the geopotential and gravity field at all wavelengths . by nature , potential data are smoother and more sensitive to mass sources at than gravity data , which are strongly influenced by local effects .
thus , they could naturally complement existing networks in sparsely covered places and even also contribute to point out possible systematic patterns of errors in the less recent gravity datasets .
we address the question through test case examples of high - resolution geopotential reconstructions in areas with different characteristics , leading to different variabilities of the gravity field .
we consider the massif central in france , marked by smooth , moderate altitude mountains and volcanic plateaus , and an alps - mediterranean zone , comprising high reliefs and a land / sea transition .
the paper is organized as follows . in section [ sec:2 ] , we briefly summarize the method schematically . in section [ sec:3 ] , we describe the regions of interest and the construction of the high - resolution synthetic datasets used in our tests . in section [ sec:4 ] , we present the methodology to assess the contribution of new clock data in the potential recovery , in addition to ground gravity measurements .
numerical results are shown in section [ sec:5 ] .
we finally discuss in section [ sec:6 ] the influence of different parameters like the data noise level and coverage .
the rapid progress of optical clocks performances opens new perspectives for their use in geodesy and geophysics . while they were
until recently built only as stationary laboratory devices , several transportable optical clocks are currently under construction or test ( see e.g. @xcite . the technological step toward state - of - the art transportable optical clocks is likely to take place within the next decade . in parallel , in order to assess the capabilities of this upcoming technology , we chose an approach based on numerical simulation in order to investigate whether atomic clocks can improve the determination of the geopotential .
, our method is adapted to the determination of the geopotential at regional scales . in figure
[ fig : methodo ] a scheme of the method used in this paper is shown : 1 . in the first step ,
we generate a high spatial resolution grid of the gravity disturbance @xmath6 and the disturbing potential @xmath7 , considered as our reference solutions .
this is done using a state of the art geopotential model ( eigen-6c4 ) , and by removing low and high frequencies .
it is described in details in section [ sec:3 ] ; 2 . in the second step ,
we generate synthetic measurements @xmath8 and @xmath9 from a realistic spatial distribution , then we add generated random noise representative of the measurement noise .
this is described in details in section [ sec:4 ] ; 3 . in a third step
, we estimate the disturbing potential @xmath10 from the synthetic measurements @xmath8 and/or @xmath9 on a regular grid thanks to least - square collocation ( lsc ) method .
is realized by making an assumption on the a priori gravity field regularity on the target area , as described in section [ sec:5 ] .
finally , we evaluate the potential recovery quality for different data distribution sets , noise levels , and types of data , by comparing the statistics of the residuals @xmath11 between the estimated values @xmath10 and the reference model @xmath7 .
residuals @xmath11 ( higher than the machine precision ) can have several origins , depending on the parameters of the simulation that can be : 1 .
the modeled instrumental noise added to the reference model at step 2 .
this noise can be changed in order to determine , for instance , whether it is better to reduce gravimetry noise by one order of magnitude , rather than using clock measurements ; 2 .
the data distribution chosen in step 2 .
this is useful to check for instance the effect of the number of clock measurements on the residuals or to find an optimal coverage for the clock measurements ; 3 . the potential estimation error , due to the intrinsic imperfection of the chosen for the geopotential . in our case
, this is due to the low - frequency content of the covariance function chosen for the least - square collocation method ( see section [ sec:5 ] ) .
all these sources of errors are somewhat entangled with one another , such that a careful analysis must be done when varying the parameters of the simulation .
this is discussed in details in section [ sec:6 ] .
our study focuses on two different areas in france . the first region is the massif central located between n and e , and consists of plateaus and low mountain range , see figure [ fig : auv ] .
the second target area , much more hilly and mountainous , is the french alps with a portion of the mediterranean sea located at the limit of different countries and bounded by n and e , see figure [ fig : alpes ] .
topography is obtained from the 30 m digital elevation model over france by ign , completed with @xcite bathymetry and srtm data .
available surface gravity data in these areas , from the bgi ( international gravimetric bureau ) , are shown in figures [ fig : auv_bgi_dg_val][fig : alpes_bgi_dg_val ] .
note that the bgi gravity data values are not used in this study , but only their spatial distribution in order to generate realistic distribution in the synthetic tests . in these figures
, it is shown that the gravity data are sparsely distributed : the plain is densely surveyed while the mountainous regions are poorly covered because they are mostly inaccessible by the conventional gravity survey .
the range of free - air gravity anomalies ( see e.g * ? ? ?
* ; * ? ? ?
* ) which are quite large reflects the complex structure of the gravity field in these regions , which means that the gravitational field strength varies greatly from place to place at high - resolution .
the scarcity of gravity data in the hilly regions is thus a major limitation in deriving accurate high - resolution geopotential model . here
, we present the way to simulate our synthetic gravity disturbances @xmath6 and disturbing potentials @xmath7 by subtracting the gravity field long and short wavelengths influence of a high - resolution global geopotential model . the generation of the synthetic data @xmath6 and @xmath7 at the earth s topographic surface was carried out , in ellipsoidal approximation , with the fortran program geopot @xcite of the national geodetic survey ( ngs ) .
the ellipsoidal normal field is defined by the parameters of the geodetic reference system grs80 @xcite . as input
, we used the static global gravity field model eigen-6c4 @xcite .
it is a combined model up to degree and order ( d / o ) containing satellite , altimetry , terrestrial gravity and elevation data . by using the spherical harmonics ( sh )
coefficients up to d / o , it allows us to map gravity variations down to 10 km resolution . .
our objective is to study how clocks can advance knowledge of the geoid beyond the resolution of the satellites . in a first step ,
as illustrated in figure [ fig : filtren100 ] , the long wavelengths of the gravity field covered by the satellites are completely removed up to the degree @xmath12 ( 200 km resolution ) . between degree 101 and 583 ,
the gravity field is progressively filtered using 3 poisson wavelets spectra @xcite , while its full content is preserved above degree 583 . in this way we realize a smooth transition between the wavelengths covered by the satellites and those constrained from the surface data . to subtract the terrain effects included in eigen-6c4 , we used the topographic potential model dv_ell_ret2012 @xcite truncated at d / o . complete up to d / o
, this model provides in ellipsoidal approximation the gravitational attraction due to the topographic masses anywhere on the earth s surface .
the results of this data reduction yields to the reference fields @xmath6 and @xmath7 for both regions , shown in figures [ fig : auv_tdg_geopot][fig : alpes_tdg_geopot ] .
[ [ gravimetric - location - points - selection . ] ] * gravimetric location points selection*. + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + our goal is to reproduce a realistic spatial distribution of the gravity points .
the bgi gravity data sets contain hundreds of thousands points for the target regions ( see figure [ fig : auv_bgi_dg_val][fig : alpes_bgi_dg_val ] ) . in order to reduce the size of the problem and make it numerically more tractable
, we build a distribution with no more than several thousand points from the original one .
starting from the spatial distribution of the bgi gravity data sets , a grid @xmath6 of @xmath13 cells is built with a regular step of about 6.5 km .
each cell contains @xmath14 points with @xmath15 .
these @xmath14 points are replaced by one point which location is given by the geometric barycenter of the @xmath14 points , in the case that @xmath16 . if @xmath17 then there is no point in the cell @xmath18 .
figures [ fig : coverage_tdg ] show the new distributions of gravimetric data for the massif central and the alps regions ; they have , respectively , and location points .
these new spatial distributions reflect the initial bgi gravity data distribution but are be more homogeneous .
they will be used in what follows .
[ [ chronometric - location - points - selection . ] ] * chronometric location points selection*. + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we choose to put clock measurements only where existing land gravity data are located .
indeed , these data mainly follow the roads and valleys which could be accessible for a clock comparison .
then , we use a simple geometric approach in order to put clock measurements in regions where the gravity data coverage is poor .
since the potential varies smoothly compared to the gravity field , a clock measurement is affected by masses at a larger distance than in the case of a gravimetric measurement .
for that reason , a clock point will be able to constrain longer wavelengths of the geopotential than a gravimetric point . finally , in order to avoid having clocks too close to each other , we define a minimal distance @xmath19 between them .
we chose @xmath19 greater than the correlation length of the gravity covariance function ( in this work @xmath20 km , see table [ tab : cov_stats ] ) .
here we give more details about our algorithm to select the clock locations : 1 .
first , we initialize the clock locations on the nodes of a regular grid @xmath7 with a fixed interval @xmath19 .
this grid is included in the target region at a setback distance of about 30 km from each edge ( outside possible boundary effects ) ; 2 .
secondly , we change the positions of each clock point to the position of the nearest gravity point from the grid @xmath6 , located in cell @xmath18 ( see the previous paragraph ) ; in cell @xmath18 are located @xmath14 points of the initial bgi gravity data distribution ; 3 . finally , we remove all the clock points located in cells where @xmath21 .
this is a simple way to keep only the clock points located in areas with few gravimetric measurements .
this method allows to simulate different realistic clock measurement coverages by changing the values of @xmath19 and @xmath22 .
the number of clock measurements increases when the distance @xmath19 decreases or when the threshold @xmath22 increases , and vice versa .
it is also possible to obtain different spatial distributions but the same number of clock measurements for different sets of @xmath19 and @xmath22 . in figure
[ fig : coverage_tdg ] , we propose an example of clock coverage used hereafter for both target regions with 32 and 33 clock locations , respectively , in the massif central and the alps , corresponding to @xmath23 percent of the gravity data coverage . for the chosen distributions ,
the value of @xmath19 is about 60 km and @xmath24 .
[ [ synthetic - measurements - simulation . ] ] * synthetic measurements simulation*. + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + for each data point , the synthetic values of @xmath8 and @xmath9 are computed by applying the data reduction presented in section [ ssec : data_synt ] .
it is important to note that the location points of the simulated data @xmath9 are not necessarily at the same place than the estimated data @xmath7 .
a gaussian white noise model is used to simulate the instrumental noise of the measurements .
we chose , for the main tests in the next section , a standard deviation @xmath25 mgal for the gravity data and @xmath26
m@xmath27/s@xmath27 for the potential data . in terms of geoid height , the latter noise level is equivalent to 1 cm .
other tests with different noise levels are discussed in section [ sec:6 ] .
in this section , we present our numerical results showing the contribution of clock data in regional recovery of the geopotential from realistic data points distribution in the massif central and the alps .
the reconstruction of the disturbing potential is realized from the synthetic measurements @xmath8 and @xmath9 , and by applying the least - squares collocation ( lsc ) method . [ [ ssec : lsc ] ] * planar least - squares collocation*.
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the lsc method , described in @xcite , is a suitable tool in geodesy to combine heterogeneous data sets in gravity field modelling . assuming that the measured values are linear functionals of the disturbing potential @xmath9
, this approach allows us to estimate any gravity field parameter based on @xmath9 from many types of observables .
consider @xmath28 = l_k$ ] a data vector composed by @xmath29 data @xmath9 and @xmath30 data @xmath8 , affected by measurement errors @xmath31 , with @xmath32 .
the estimation of the disturbing potential @xmath33 at point @xmath34 from the data @xmath35 can be performed with the relation @xmath36 with @xmath37 the covariance matrix of the measurement vector @xmath35 , @xmath38 the covariance matrix of the noise , @xmath39 the cross covariance matrix between the estimated signal @xmath40 and the data @xmath35 , and @xmath41 the tikhonov regularization factor @xcite , also called weight factor . in practice ,
the data @xmath35 are synthesized as described in sections [ sec:3 ] and [ sec:4 ] .
therefore , the measurement noise is known to be a gaussian white noise .
noise and signal ( errorless part of @xmath42 ) are assumed to be uncorrelated , and the covariance matrix of the noise can be written as @xmath43\ ] ] with @xmath44 the identity matrix of size @xmath45 . because @xmath37 can be very ill - conditioned
, the matrix plays an important role in its regularization before inversion , since positive constant values are added to the elements of its main diagonal . to avoid any iterative process to find an optimum value of @xmath41 in case where this matrix @xmath37 is not definite positive , we chose to fix the weight factor @xmath46 and to apply a singular value decomposition ( svd ) to pseudo - inverse the matrix .
as shown in @xcite , these two approaches are similar .
[ [ ssec : efc ] ] * estimation of the covariance function*. + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + implementation of the collocation method requires to compute the covariance matrices @xmath47 and @xmath37 .
this step has been carried out using a logarithmic spatial covariance function from @xcite , see appendix [ anx : deffcf ] .
this stationary and isotropic model is well - adapted to our analysis .
indeed , it provides the auto - covariances ( acf ) and cross - covariances ( ccf ) of the disturbing potential @xmath9 and its derivatives in 3 dimensions with simple closed - form expressions .
the spatial correlations of the gravity field are analyzed with the program gpfit @xcite .
the variance @xmath48 is directly computed from the gravity data on the target area , and the parameters @xmath49 and @xmath50 ( see appendix [ anx : deffcf ] ) are estimated by fitting the a priori covariance function to the empirical acf of the gravity disturbances @xmath8 .
results of the optimal regression analysis for both regions are given in figure [ fig : covariogram ] and table [ tab : cov_stats ] .
knowing the parameter values of the covariance model , we can now estimate the potential anywhere on the earth s surface .
[ [ contribution - of - clocks . ] ] * contribution of clocks . * + + + + + + + + + + + + + + + + + + + + + + + + + the contribution of clock data in the potential recovery is evaluated by comparing the residuals of two solutions to the reference potential on a regular grid interval of 10 km .
the first solution corresponds to the errors between the estimated potential model computed solely from gravity data and the potential reference model , while the second solution uses combined gravimetric and clock data . to avoid boundary effects in the estimated potential recovery , a grid edge cutoff of 30
km has been removed in the solutions . for the massif central region ,
the disturbing potential is estimated with a bias @xmath51 ( 4.1 mm ) and a rms @xmath52 ( 2.5 cm ) using only the gravimetric data , see figure [ fig : auv_emc_tdg_from_t0dg_4art ] .
when we now reconstruct @xmath9 by adding the 33 potential measurements to the gravimetric measurements , the bias is improved by one order of magnitude ( @xmath53 or @xmath54 mm ) and the standard deviation by a factor 3 ( @xmath55 or 7 mm ) , see figure [ fig : auv_emc_tdg_from_t33dg_56_15_4art ] . for the alps ,
figure [ fig : alpes_t_recovery ] , the potential is estimated with a bias @xmath56 ( 2.3 cm ) and a standard deviation @xmath57 ( 3.9 cm ) using only the gravimetric data .
when adding the 32 potential measurements , we note that the bias is improved by a factor 4 ( @xmath58 or @xmath59 mm ) and the standard deviation by a factor 2 ( @xmath60 or 1.8 cm ) .
it should also be stressed that a trend appears in the reconstructed potential with respect to the original one when no clock data are added in both regions .
this effect is discussed in section [ sec:6 ] .
[ [ ssec : nbclockeffect ] ] * effect of the number of clock measurements . * + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + figure [ fig : effect_nb_clock ] shows the influence of the number of clock data in the potential recovery , and thereby of their spatial distribution density .
we vary the number and distribution of clock data by changing the mesh grid size @xmath19 , which represents the minimum distance between clock data points ( see section [ sec:4 ] ) .
the particular cases shown in detail in section [ sec:5 ] are included .
we characterize the performance of the potential reconstruction by the standard deviation and mean of the differences between the original potential on the regular grid and the reconstructed one .
when increasing the density of the clock network , the standard deviation of the differences tends toward the centimeter level , for the massif central case , and the bias can be reduced by up to 2 orders of magnitude .
note that we have not optimized the clock locations such as to maximize the improvement in potential recovery .
the chosen locations are simply based on a minimum distance and a maximum coverage of gravity data ( c.f .
section [ sec:4 ] ) .
an optimization of clock locations would likely lead to further improvement , but is beyond the scope of this work and will be the subject of future studies .
moreover , the results indicate that it is not necessary to have a large number of clock data to improve the reconstruction of the potential .
we can see that only a few tens of clock data , i.e. less than 1 percent of the gravity data coverage , are sufficient to obtain centimeter level standard deviations and large improvements in the bias . when continuing to increase the number of clock data the standard deviation curve seems to flatten at the cm level .
[ [ ssec : nbgraveffect ] ] * effect of the number of gravity measurements . * + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + [ [ ssec : syntdataeffect ] ] * covariance function consistency .
* + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in figures [ fig : auv_emc_tdg_from_t0dg_4art ] and [ fig : alpes_emc_tdg_from_t0dg_4art ] , a trend appears in the residuals , but disappears when gravimetric and clock data are combined .
this is due to the fact that the covariance function does not have the same spectral coverage as the data generated from the gravity field model eigen-6c4 .
indeed , the covariance function contains low frequencies while we have removed them for the synthetic data .
therefore , some low frequency content is present in the recovered potential .
whilst the issue could be avoided by using a covariance parametric model from which we can remove the low frequency content in a perfectly consistent way with the data generation ( e.g. a closed - form tscherning - rapp model @xcite ) , it is not obvious that the corresponding results would allows realistic conclusions .
indeed , the spectral content of real surface observations , after removal of lower frequencies from a global spherical harmonics model , may still retain some unknown low frequencies .
as consequence , it is not obvious to match to that of a single covariance function , while perfect consistency can only be achieved from synthetic data .
we chose to keep this mismatch , thereby investigating the interest of clocks for high - resolution geopotential determination when our prior knowledge on the surface data signal and noise components is not perfect .
more detailed studies on this issue are considered beyond the scope of our paper , which presents a first step to quantify the possible use of clock measurements in potential recovery .
[ [ ssec : noiseeffect ] ] * influence of the measurement noise . * + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + we have also investigated the effect of the noise levels applied to the synthetic data , see tables [ tab : effect_noise_auv][tab : effect_noise_alps ] , by using various standard deviations to simulate white noise of the measurements : @xmath61 for the clock measurements and @xmath62 mgal for the gravimetric measurements . these results were obtained for the same conditions as in section [ sec:5 ] , i.e. 33 ( resp .
32 ) clock data points and 4374 ( resp .
4959 ) gravity data points for the massif central ( resp .
alps ) .
we can see that adding clocks improves the potential recovery ( smaller standard deviation @xmath63 and bias @xmath64 of the residuals ) for both regions and whatever the noise of the gravimetric or clock measurements . decreasing the noise of the gravity data by up to two orders of magnitude only improves the standard deviation of the residuals @xmath63 of the recovered potential by comparatively small amounts ( less than a factor 2 ) .
this is probably due to the fact that the covariance function does not reflect the gravity field correctly in these regions , combined with a limited data coverage .
note that the low frequency content in the covariance function ( see above ) is unlikely to be the main cause here , as the comparatively small reduction of @xmath63 is also observed when clocks are present in spite of the fact that they remove the low frequency trend ( c.f . figures [ fig : auv_emc_tdg_from_t33dg_56_15_4art ] and [ fig : alpes_emc_tdg_from_t32dg_50_3_4art ] ) .
* 1c * 8s[table - format = -1.1e2 ] & & & & + ( r)1 - 1 ( r)2 - 3 ( r)4 - 5 ( r)6 - 7 ( ) 8 - 9 & & & & & & & & + ( r)1 - 1 ( lr)2 - 2(lr)3 - 3 ( lr)4 - 4(lr)5 - 5 ( lr)6 - 6(lr)7 - 7 ( lr)8 - 8(lr)9 - 9 & 2.2e-1 & 3.7e-1 & 4.1e-2 & 2.5e-1 & 1.5e-1 & 1.7e-1 & 2.6e-1 & 1.8e-1 + ( lr)1 - 1 ( lr)2 - 2(lr)3 - 3 ( lr)4 - 4(lr)5 - 5 ( lr)6 - 6(lr)7 - 7 ( lr)8 - 8(lr)9 - 9 1 & -4.4e-3 & 2.8e-1 & -1.8e-4 & 1.7e-1 & -1.1e-2 & 1.6e-1 & -2.0e-2 & 1.7e-1 + ( lr)1 - 1 ( lr)2 - 2(lr)3 - 3 ( lr)4 - 4(lr)5 - 5 ( lr)6 - 6(lr)7 - 7 ( lr)8 - 8(lr)9 - 9 0.1 & -1.4e-2 & 2.0e-1 & -2.4e-3 & 7.3e-2 & -6.7e-3 & 5.2e-2 & -1.1e-3 & 4.8e-2 + * 1c * 8s[table - format = -1.1e2 ] & & & & + ( r)1 - 1 ( r)2 - 3 ( r)4 - 5 ( r)6 - 7 ( ) 8 - 9 & & & & & & & & + ( r)1 - 1 ( lr)2 - 2(lr)3 - 3 ( lr)4 - 4(lr)5 - 5 ( lr)6 - 6(lr)7 - 7 ( lr)8 - 8(lr)9 - 9 & 5.8e-1 & 6.6e-1 & 2.2e-1 & 3.9e-1 & 2.1e-1 & 4.2e-1 & 2.1e-1 & 4.2e-1 + ( lr)1 - 1 ( lr)2 - 2(lr)3 - 3 ( lr)4 - 4(lr)5 - 5 ( lr)6 - 6(lr)7 - 7 ( lr)8 - 8(lr)9 - 9 1 & 1.8e-1 & 6.2e-1 & 1.4e-1 & 3.4e-1 & 1.2e-1 & 3.3e-1 & 1.2e-1 & 3.3e-1 + ( lr)1 - 1 ( lr)2 - 2(lr)3 - 3 ( lr)4 - 4(lr)5 - 5 ( lr)6 - 6(lr)7 - 7 ( lr)8 - 8(lr)9 - 9 0.1 & 2.0e-1 & 5.6e-1 & 6.8e-2 & 1.7e-1 & 4.7e-2 & 1.5e-1 & 1.7e-2 & 1.6e-1 + when adding clocks the standard deviations are decreased by up to a factor 3.7 with low clock noise ( 0.1 or 1 cm ) and a factor 1.5 with higher clock noise ( 1 or 10 cm ) . the effect is stronger in the massif central region than in the alps . we attribute this again to the mismatch between the covariance function and the complex structure of the gravity field , which is larger in the alps . thus , optical clocks with just an accuracy of 1 ( or 10 cm ) are interesting no matter what the gravity data quality . with an accuracy of 0.1 ( or 1 cm ) , we can expect a gain of up to a factor 4 in the estimated potential with respect to simulations using no clock data .
[ [ ssec : aliasing ] ] * aliasing of the very high - resolution components . *
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
optical clocks provide a tool to measure directly the potential and determine the geopotential at high spatial resolution .
we have shown that the recovery of the potential from gravity and clock data with the lsc method can improve the determination of geopotential at high spatial resolution , beyond what is available from satellites .
compared to a solution that does not use the clock data , the standard deviation can be improved by a factor 3 , and the bias can be reduced by up to 2 orders of magnitude with only a few tens of clock data . this demonstrates the benefit of this new potential geodetic observable , which could be put in practice in the medium term when the first transportable optical clocks and appropriate time transfer methods will be developed ( see * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
since clocks are sensitive to low frequencies of the gravity field , this method is particularly well - adapted in hilly and mountainous regions for which the gravity coverage is more sparsely distributed , allowing to fill areas not covered by the classical geodetic observables ( gravimetric measurements ) . additionally , adding new observables helps to reduce the modelling errors , e.g. coming from a mismatch between the covariance function used and the real gravity field . to our knowledge
, this is the first detailed quantitative study of the improvement in field determination that can be expected from chronometric geodesy observables .
it provides first estimates and paves the way for future more detailed and in depth works in this promising new field . to overcome some limitations in the a priori model ,
as discussed in the previous section , we intend in a forthcoming work to investigate in more details the imperfections of the covariance function model .
moreover , as the gravity field is in reality non - stationary in mountainous areas or near the coast , some numerical tests with non - stationary covariance functions will be conducted .
another promising source of improvement could be the optimization of the positioning of the clock data .
for example , the correlation lengths and the variations of the gravity field could be used as constraints .
a genetic algorithm could also be considered to solve this location problem .
finally , it will be interesting to focus on the improvement of the potential recovery quality by combining other types of observables such as leveling data and gradiometric measurements .
as knowledge of the geopotential provides access to height differences , this could be a way to estimate errors of the gnss technique for the vertical positioning , or contribute to regional height systems unification .
we thank ren forsberg for providing us the fortran code of the logarithmic covariance function .
we gratefully acknowledge financial support from labex first - tf , erc adoc ( grant n 617553 and emrp itoc ( emrp is jointly funded by the emrp participating countries within euramet and the european union ) .
we thank olivier jamet and matthias holschneider for discussions about the collocation method .
let us consider two points @xmath34 and @xmath65 with the cartesian coordinates @xmath66 and @xmath67 , respectively . to compute the acf and ccf of the disturbing potential @xmath9 and its derivatives , @xcite proposed a planar attenuated logarithm covariance model with upward continuation that can be expressed in the generic form @xmath68 with this model
is characterized by three parameters : @xmath48 the variance of the gravity disturbance @xmath8 and two scale factors acting as high and low frequency attenuators : @xmath49 the shallow depth parameter and @xmath50 the compensating depth , respectively .
the function @xmath70 is logarithmic function modelling the covariances between the gravity field quantities .
for example , by putting @xmath71 and @xmath72 , the acf of @xmath8 and @xmath9 can be evaluated respectively with @xmath73 and the ccf between @xmath9 and @xmath8 with @xmath74 | recent technological advances in optical atomic clocks are opening new perspectives for the direct determination of geopotential differences between any two points at a centimeter - level accuracy in geoid height .
however , so far detailed quantitative estimates of the possible improvement in geoid determination when adding such clock measurements to existing data are lacking .
we present a first step in that direction with the aim and hope of triggering further work and efforts in this emerging field of chronometric geodesy and geophysics .
we specifically focus on evaluating the contribution of this new kind of direct measurements in determining the geopotential at high spatial resolution ( @xmath0 ) .
we studied two test areas , both located in france and corresponding to a middle ( massif central ) and high ( alps ) mountainous terrain .
these regions are interesting because the gravitational field strength varies greatly from place to place at high spatial resolution due to the complex topography .
our method consists in first generating a synthetic high - resolution geopotential map , then drawing synthetic measurement data ( gravimetry and clock data ) from it , and finally reconstructing the geopotential map from that data using least squares collocation .
the quality of the reconstructed map is then assessed by comparing it to the original one used to generate the data .
we show that adding only a few clock data points ( less than 1 % of the gravimetry data ) reduces the bias significantly and improves the standard deviation by a factor 3 .
the effect of the data coverage and data quality on the results is investigated , and the trade - off between the measurement noise level and the number of data points is discussed .
* keywords . * chronometric geodesy ; high spatial resolution ; geopotential ; gravity field ; atomic clock ; least - squares collocation ( lsc ) ; stationary covariance function |
the notion of _ critical mass _ in research has been around for a long time without proper definition .
as governments , funding councils and universities seek indicators to measure research quality and to pursue greater efficiencies in the research sector , critical mass is becoming an increasingly important concept at managerial and policy - making level .
however , until very recently there have been no successful attempts to quantify this notion ( harrison , 2009 ) .
it has been described by evidence ( 2010 ) as `` some minimum size threshold for effective performance '' and , as such , has been linked to the idea that benefit accrues through increase of scale of research groups . however , although evidence ( 2010 ) demonstrated `` a relationship of some kind between larger units and relatively high citation impact '' , indications of such a threshold have been lacking .
we recently presented a model for the relationship between quality of research groups and their quantity ( kenna and berche , 2010a ) .
this model was inspired by mean - field theories of statistical physics and allowed for a quantitative definition of critical mass .
in fact there are two critical masses in research and their values are discipline dependent . instead of a threshold group size
above which research quality improves , we have shown that there is a breakpoint or _ upper critical mass _ beyond which the linear dependency of research quality on group quantity reduces . denoting this value by @xmath1
, we showed that the strength of the overall research sector in a given discipline is improved by supporting groups whose size are less than @xmath1 , provided they are bigger than a second critical mass , which we denote by @xmath2 .
groups whose size are smaller than @xmath2 are vunerable and should seek to achieve the lower critical mass for long - term viability .
the two critical masses are related by a scaling relation , @xmath3 we classify research groups of size @xmath4 within a given discipline as small , medium and large according to whether @xmath5 , @xmath6 or @xmath7 , respectively .
we recently determined the critical masses of a multitude of academic disciplines by applying statistical analyses to the results of the uk s most recent _ research assessment exercise _
( rae ) in which the quality of research groups were measured ( kenna and berche , 2010b ) . notably absent from our analaysis , however , were the statistics and operational research groups , as these were less straightforward to analyse than other subject areas .
here we rectify this omission by a careful analysis of these disciplines .
our main result is that the lower critical mass , which statistics and operational research groups should attain to be viable in the long term , is @xmath8 in section 2 we summarize our model and how we derive critical masses from it .
we also discuss the research assessment exercise . in section
3 we apply the model and statistical analysis to the results of the rae for statistics and operational research groups .
we conclude in section 4 , where implications for policy and management are briefly discussed .
our model is based on the idea that research groups are _ complex systems _ , for which the properties of the whole are not simple sums of the corresponding properties of the individual parts .
instead , interactions between individuals within research groups have to also be taken into account .
the strength of an individual within a research group is a function of many factors : their intrinsic calibre and training , their teaching and administrative loads , library facilities , journal access , extramural collaboration , the quality of management , and even confidence gained by previous successes as well as the prestige of the institution and other factors .
we denote the average individual research strength within the @xmath9 research group in a given academic discipline , resulting from all of these ( and any other ) factors by @xmath10 .
the overall calibre of a research group comprising @xmath4 individuals is also dependent on the extent of , and strength of , the communication links between them .
we denote the average strength of the @xmath11 interactions between the @xmath4 individuals in the @xmath9 group by @xmath12 .
the overall strength of the group is therefore given by @xmath13 however , once the size of a research group becomes too large ( say above a cutoff value @xmath1 ) , meaningful communication between _ all _ pairs of individuals becomes impossible . in this case
, the group may fragment into @xmath14 subgroups , of average size @xmath15 , say .
if the average strength of interaction between the subgroups is @xmath16 , the overall strength of the group becomes @xmath17 we denote by @xmath18 the _ expected _ strength of a group of size @xmath4 and we define the quality of such a research group to be the average strength per head : @xmath19 gathering terms of the same order in @xmath4 , we arrive at a form for the _ expected _ dependency of research - group quality on research - group quantity , @xmath20 we considered the effect on the overall strength of a discipline by adding new researchers ( kenna and berche , 2010a ) . asking the question whether it is better , on average , to allocate new researchers to a group with @xmath21 or @xmath22 members , we found that the latter is preferable provided @xmath23 , where @xmath2 is given by eq.([scaling ] ) .
this is equivalent to maximising the gradient of the strength function @xmath24 .
we also considered the consequences of transferring researchers from large to small / medium groups and found that such a movement is expected to be beneficial to society as a whole , provided the recipient group is not too small ( i.e. , provided , again , that it has over @xmath2 members ) .
thus there are two critical masses in research , which we name _ lower _
( @xmath2 ) and _ upper _ ( @xmath1 ) . of these ,
the former corresponds more closely to the traditional , intuitive notion of critical mass , although there is no threshold value beyond which research quality suddenly improves ( evidence , 2010 ) . to implement the model ( [ nc ] )
, we require a set of empirical data on the quality and quantity of research groups .
the rae is an evaluation process undertaken approximately every 5 years on behalf of the funding bodies for universities in the uk .
the results of the rae are used to allocate funding to such higher education institutes for the subsequent years .
the last rae was carried out in 2008 .
research groups were examined to determine the proportion of research submitted categorized as follows : * 4 * : quality that is world - leading in terms of originality , significance and rigour * 3 * : quality that is internationally excellent in terms of originality , significance and rigour but which nonetheless falls short of the highest standards of excellence * 2 * : quality that is recognised internationally in terms of originality , significance and rigour * 1 * : quality that is recognised nationally in terms of originality , significance and rigour * unclassified : quality that falls below the standard of nationally recognised work .
a formula is then used to determine how funding is distributed to research groups .
the 2009 formula used by the higher education funding council for england weighs each rank in such a way that 4 * and 3 * research respectively receive seven and three times the amount of funding allocated to 2 * research , and 1 * and unclassified research attract no funding .
this funding formula may therefore be considered to represent a measurement of quality of each research group .
( in 2010 , after lobbying by the larger , research intensive universities the english funding formula was changed so that 4 * research receives nine times the funding allocated to 2 * research .
we have checked that the 2010 formula produces no significant change to the results presented here . ) from the outset , we acknowledge that there are obvious assumptions underlying our analysis and limits to what can be achieved .
firstly , we use the term `` group '' in the sense of rae .
this means the collection of staff included in a submission to one of the 67 _ units of assessment _ ( uoa s ) .
rae groups are not always identical to administrative departments within universities , but we assume that they represent a coherent group for research purposes .
individuals submitted to rae are drawn from academic staff who were in post and on the payroll of the submitting higher education institution on the census date ( 31 october 2007 ) .
we assume that the rae process is fair and unbiased and that the scores are reasonably reliable and robust .
deviations from these assumptions contribute to noise in the system .
statistical analyses and a list of the critical masses for a variety of academic disciplines ( not including statistics and operational research ) are given in ( kenna and berche , 2010b ) .
in the next section , we perform a similar analysis for the statistics and operational research groups submitted to rae 2008 .
the statistics and operational research uoa at rae 2008 included theoretical , applied and methodological approaches to statistics , probability and operational research .
there were 30 submissions comprising 388.8 individuals ( with fractions corresponding to part - time staff ) and group sizes ranged from @xmath25 to @xmath26 , with mean group size 13 .
we find it useful to compare to the applied mathematics uoa because of the high degree of overlap between the two disciplines .
there were 45 submissions in applied mathematics entailing 850.05 individuals in groups of size @xmath27 to @xmath28 with mean group size @xmath29 .
the 30 submissions for statistics and operational research are listed in table 1 .
also listed are the numbers of staff submitted and the resultant quality score .
.universities which submitted to the statistics and operational research uoa at rae 2008 , listed alphabetically together with the numbers of staff submitted @xmath4 and quality measurements @xmath30 . [ cols= " > , < , > , > " , ] to further compare statistics and operational research to applied mathematics
, we plot the sets of data corresponding to both uoa s in fig .
2(b ) together with the fits coming from the model ( [ nc ] ) .
the similarities in their critical masses are evident , as are the similarities between slopes of the piecewise linear fits , although that for statistics and operational research is shifted slightly above that for applied mathematics , indicating a consistently better average performance for comparably sized groups or problems with the rae due to the absence of a systematic approach to normalize scores between disciplines .
we believe the latter is the more likely scenario . in any case , it is clear that in comparison to applied mathematics , there are relatively few statistics and operational research teams in the uk and , of those , there are even fewer which are supercritical ( and therefore operating with sufficient resources ) in size .
this suggests that greater investment in this subject area is required to achieve optimal research efficiency . given in eq.([nc ] ) .
the tighter distribution of the data about the line in ( b ) demonstrates the validity of the model . in both plots ,
the abscissae index the universities listed alphabetically in table 1.,title="fig : " ] given in eq.([nc ] ) .
the tighter distribution of the data about the line in ( b ) demonstrates the validity of the model . in both plots ,
the abscissae index the universities listed alphabetically in table 1.,title="fig : " ] to illustrate the superiority of the model over the alternative idea that there is no relationship between quality and quantity in research , we plot in fig .
3 the deviations of the data from the predictions coming from both scenarios . in each case
the data are plotted against the index values listed in table 1 , which correspond to an alphabetical ordering of the institutes which submitted to the statistics and operational research uoa . in fig .
3(a ) , the differences between the quality scores and the mean quality value of the 30 research groups are plotted .
the range and standard deviation corresponding to this plot are 43.6 and 10.5 respectively ( 43.6 and 10.7 if edinburgh / heriot - watt is excluded ) . in fig .
3(b ) , the deviations from the expectation values coming from the model ( [ nc ] ) are plotted .
the range and standard deviation associated with this plot ( excluding edinburgh / heriot - watt ) are 26.1 and 6.7 , respectively .
the tighter distribution of the data in fig .
3(b ) over fig .
3(a ) illustrates the validity of the model .
plots of the type given in fig .
3(a ) form the basis on which research groups are ranked post rae , with teams above and below the line deemed to be performing above and below average , respectively . however
, such rankings do not compare like with like as they fail to take size , and hence resources , into account .
we suggest that fig .
3(b ) forms the basis of a better system as in this plot , performances are compared to the averages for teams of given sizes .
3(b ) takes size into account and gives a better indication of which groups are punching above and below their weights .
to summarise , we have applied a mean - field inspired model to examine the relationship between the quality of research teams in statistics and operational research and the quantity of researchers in those teams .
our empirical data is taken from the most recent research assessment exercise in the uk .
we find that , when an outlying amalgamated group is omitted the dependency of quality upon quantity for this subject area is similar to , and consistent with , a multitude of other disciplines which were reported on in ( kenna and berche , 2010b ) .
the model allows the definition of two critical masses for the discipline .
the research quality of small ( @xmath31 ) and medium ( @xmath6 ) teams is strongly dependent on the number of researchers in the group . beyond @xmath1 ,
large teams tend to fragment and research quality is no longer correlated with group size .
the lower critical mass for statistics and operational research is determined to be @xmath32 , and the upper value is about twice that .
these values compare satisfactorily to the equivalent for applied mathematics which has @xmath33 .
to further contextualize these values , we quote from kenna and berche ( 2010b ) the results @xmath34 for pure mathematics ( a relatively solitary research discipline ) and @xmath35 for medical sciences ( a highly collaborative one ) . notwithstanding the fact that some statisticians and operational researchers were submitted to rae 2008 as part of teams in other disciplines such as business , economics , engineering and epidemiology , about a quarter of statistics / operational research _
groups _ submitted to rae are sub - critical , with @xmath36 , and therefore vulnerable .
these teams need to strive to attain critical mass . of the 29 teams excluding the edinburgh / heriot - watt combination ,
only five ( 17% ) have size above the upper critical mass of @xmath37 .
therefore the majority of statistics and operational research teams within the uk are under - resourced in terms of staff numbers .
we suggest that to increase research efficiency for this discipline investment is needed .
this conclusion parallels that of smith and staetsky ( 2007 ) for the teaching of statistics in the uk . *
references * harrison , m. ( 2009 ) does high quality research require critical mass ? in _ the question of r&d specialisation : perspectives and policy implications _ ( eds d. pontikakis , d. kriakou and r. van baval ) , pp 57 - 59 .
european commission : jrc technical and scientific reports . | using a recently developed model , inspired by mean field theory in statistical physics , and data from the uk s research assessment exercise , we analyse the relationship between the quality of statistics and operational research groups and the quantity researchers in them .
similar to other academic disciplines , we provide evidence for a linear dependency of quality on quantity up to an upper critical mass , which is interpreted as the average maximum number of colleagues with whom a researcher can communicate meaningfully within a research group .
the model also predicts a lower critical mass , which research groups should strive to achieve to avoid extinction . for statistics and operational research ,
the lower critical mass is estimated to be @xmath0 .
the upper critical mass , beyond which research quality does not significantly depend on group size , is about twice this value . |
the nature of magnetism in the recently discovered iron pnictides superconductors @xcite is an intensively studied topic , due to its possible coexistence with superconductivity and relation to the pairing mechanism @xcite . in the so - called `` 122 family '' , the parent compounds afe@xmath3as@xmath3 , where a = ba , ca , sr or eu , present a transition to a spin density wave ( sdw ) phase which is commensurate with the lattice spacing .
when substituting fe with another metal , like co or ni ( the dopant ) , the nel temperature gradually decreases and superconductivity eventually emerges at higher dopant concentrations . in ba(fe@xmath1co@xmath2)@xmath3as@xmath3 , superconductivity and magnetism coexist locally below the critical temperature @xmath6 @xcite , i.e. there occurs a so - called mixed phase with spatial coexistence of the two phases . by contrast , in some pnictides belonging to another family , e.g. lafeaso@xmath1f@xmath2 @xcite , magnetism vanishes when superconductivity sets in .
magnetic order in all afe@xmath3as@xmath3 is antiferromagnetic ( af ) and the spin arrangement is formed by af stripes along @xmath7 and ferromagnetic stripes along @xmath8 , in the orthorhombic notation @xcite .
the fe ordered moment is small , of the order of 1@xmath9 , which , together with the metallic behavior in the normal state , suggests the presence of itinerant magnetism .
this is believed to be induced by a fermi surface instability driven by the nesting between an electron and a hole pocket @xcite , as seen by angle - resolved photoemission studies ( arpes ) in bafe@xmath3as@xmath3 @xcite , as well as in electron - doped ba(fe@xmath1co@xmath2)@xmath3as@xmath3 @xcite and hole - doped ba@xmath1k@xmath2fe@xmath3as@xmath3 @xcite . in doped systems , it was suggested that an imperfect nesting between the hole and electron pockets should lead to the formation of an incommensurate spin density wave ( ic - sdw ) below the magnetic transition temperature @xcite , as occurs in cr @xcite .
several theoretical studies have shown that coexistence of superconductivity and magnetism is favoured by the presence of an ic - sdw @xcite .
therefore , the observation of an ic - sdw is an important issue to verify the various theories put forward in order to describe the physics of pnictide superconductors .
experimentally , a phase transition from commensurate to incommensurate sdw was recently reported in srfe@xmath3as@xmath3 under pressure @xcite , based on nmr measurements . in ba(fe@xmath1co@xmath2)@xmath3as@xmath3 ,
the presence of an ic - sdw was inferred from local techniques such as nmr @xcite , mssbauer spectroscopy @xcite and @xmath10sr @xcite , but was not observed by neutron scattering @xcite until recently , and only in a narrow concentration range close to a co doping level @xmath11=0.055 @xcite .
therefore , it was claimed that the perturbation of magnetic order induced by the dopant is responsible for the effects observed in refs.@xcite . in ba(fe@xmath1ni@xmath2)@xmath3as@xmath3
, @xmath5as nmr lineshapes were interpreted in the framework of an empirical model according to which the magnetic order is commensurate , and the dopant induces perturbations of the fe moment amplitude @xcite . in ba(fe@xmath1rh@xmath2)@xmath3as@xmath3 ,
mssbauer spectra were interpreted assuming either an ic - sdw or the presence of disorder @xcite , while neutron scattering did not detect any incommensurability in this family @xcite . in this paper
, we present mssbauer results obtained on a powder sample of bafe@xmath3as@xmath3 and in single crystal mosaics of ba(fe@xmath1ni@xmath2)@xmath3as@xmath3 with @xmath11=0.01 and 0.03 . in the doped systems ,
we show that a distribution of fe hyperfine fields is present in the magnetic phase , similar to what was observed in ba(fe@xmath1co@xmath2)@xmath3as@xmath3 @xcite .
we interpret the mssbauer data using the two models for an incommensurate spin density wave and for a dopant - induced fe moment perturbation .
then , we present simulations of nmr spectra from refs.@xcite assuming an ic - sdw modulation with the parameters obtained from the fit of the mssbauer spectra .
this is motivated by the observation made in ref.@xcite that the nmr data rule out the presence of ic - sdw in ni doped compounds .
bafe@xmath3as@xmath3 in powder form was prepared by solid - state reaction and it presents a magnetic transition at @xmath12k @xcite .
single crystals of ba(fe@xmath1ni@xmath2)@xmath3as@xmath3 with @xmath130.01 and 0.03 were grown by the self - flux method , as described in detail in ref.@xcite .
the system is not superconducting at @xmath4 , while at @xmath14 superconductivity occurs below @xmath15k .
susceptibility measurements on a single crystal show that the whole sample is superconducting below @xmath6 .
resistivity measurements performed on crystals from the same batches showed them to be very homogenous @xcite , with a magnetic transition towards a spin density wave at @xmath16k for @xmath4 and @xmath17k for @xmath14 .
ni - doped platelet - like crystals were glued in mosaic on aluminium scotch tape with the ( @xmath18 ) plane perpendicular to the @xmath19 beam .
the scotch tape contains iron impurities in very small amounts , which give rise to a quadrupolar doublet with 0.07% resonant absorption . because the resonant absorption due to the bafe@xmath3as@xmath3 phase is around 2% ,
the iron impurities give a negligible contribution to the spectra .
spectra were recorded using a commercial @xmath0co@xmath20:rh @xmath19-ray source mounted on an electromagnetic drive with linear velocity signal .
the width calibration of the source was done with a thin k@xmath21fe(cn)@xmath22 .
3h@xmath3o absorber and the half width at half maximum ( hwhm ) is found to be 0.128mm / s .
the velocity and linearity calibration were performed with an enriched @xmath23-fe@xmath3o@xmath24 absorber and a commercial wissel laser interferometer .
all spectra are shown in a velocity range of @xmath252mm / s .
however , the spectra in pure bafe@xmath3as@xmath3 and the 4.2k spectrum in ba(fe@xmath26ni@xmath27)@xmath3as@xmath3 were recorded in a velocity range of @xmath253mm / s .
fe in bafe@xmath3as@xmath3 in the sdw phase ( 4.2k ) and close to the sdw to paramagnetic transition ( 141k ) .
lines are fits as described in the text.,width=7 ] as@xmath3 around @xmath29 , extracted from the fit of the mssbauer spectra .
the vertical dashed lines indicate @xmath30139.5k .
inset : temperature dependence of the hyperfine field .
the lines are guides for the eye.,width=8 ] in the high temperature paramagnetic range , the mssbauer spectra are well fitted by a single lorentzian - shaped line of hwhm 0.15mm / s . in the magnetically ordered phase below @xmath31k ( fig.[pur_spectres ] ) and up to about 125k ,
spectra are well fitted to a 6-line pattern due to a single hyperfine field of 5.4 t at 4.2k .
this is expected for a commensurate magnetic order and it is in good agreement with similar measurements presented in refs.@xcite , as well as with neutron scattering @xcite and nmr @xcite results .
a peculiarity of these hyperfine field spectra is that the hwhm of the middle lines , 0.134(4)mm / s , is equal within uncertainties to that of the inner lines , while the hwhm of the outer lines is 0.168(4)mm / s , significantly larger .
this unusual spectral effect is observed in other published spectra at 4.2k @xcite and it can not be due to a mere distribution of hyperfine fields , which would yield increasing broadenings for the inner , middle and outer lines .
it can be accounted for by correlated distributions of quadrupolar interaction ( qi ) and isomer shift ( is ) .
the presence of a qi distribution at the @xmath0fe site is in line with the @xmath5as nmr spectra in bafe@xmath3as@xmath3 , which indicate a sizeable distribution of quadrupolar frequencies at the as site @xcite .
correlated distributions of qi and is have been observed in amorphous or glassy materials @xcite , but they are not expected in an ordered metallic material such as bafe@xmath3as@xmath3 .
we note that they can be inferred to occur in charge density wave ( cdw ) phases , but such a phase has not been reported in the iron pnictides .
the thermal variation of the hyperfine field is shown in the inset of fig.[pur_paramagfraction ] .
its weak temperature dependence suggests a transition with first order character . in the temperature range from 130 to 150k close to @xmath32
, the spectra can not be fitted to a single sextet .
an additional single - line paramagnetic component develops at the expense of the magnetically ordered one as temperature increases above 120k .
the fit with two components obtained at 141k is shown in fig.[pur_spectres ] .
the presence of this relatively narrow magnetic / paramagnetic coexistence region ( see fig.[pur_paramagfraction ] ) is another indication of the first order character of the transition .
this is in good agreement with the @xmath5as nmr data in bafe@xmath3as@xmath3 @xcite . in bafe@xmath3as@xmath3 ,
the paramagnetic fraction develops from 0 to 100% over a 15k range on each side of the transition ( @xmath250.1@xmath29 ) .
this is quite different from the behaviour in co - doped systems , where coexistence is observed to occur on a much wider temperature range @xcite , up to about 1.5 @xmath29 above the magnetic transition .
ni@xmath2)@xmath3as@xmath3 obtained at 4.2k for @xmath11=0.01 and 0.03 .
continuous lines are fits assuming an incommensurate modulation of fe moments . *
right : * modulations that fit well the spectra shown in the left panel , with parameters from table i. the thick red lines correspond to the fits shown on the left.,width=11 ] mosaic mssbauer spectra obtained for ba(fe@xmath1ni@xmath2)@xmath3as@xmath3 with @xmath11=0.01 and 0.03 with the @xmath19-beam along the * _ c _ * axis are shown in fig.[ni_spectres4k ] for @xmath34k .
these spectra do not display the six narrow line pattern expected for a single magnetic environment , as in the undoped compound .
this indicates a distribution of fe hyperfine fields , similar to what was observed in ba(fe@xmath1co@xmath2)@xmath3as@xmath3 @xcite .
our data is in agreement with that reported in refs.@xcite for the ni content @xmath11=0.016 and 0.024 . in these works ,
the spectra were fitted to a superposition of commensurate and sine - wave incommensurate subspectra @xcite . here
, we fit the spectra assuming either an incommensurate spin density wave with several harmonics , or a dopant - induced perturbation model .
we focus mostly on the @xmath4 ni concentration in order to take advantage of the good resolution of the spectra . for @xmath35 , due to the decrease of the fe magnetic moment , spectra become almost featureless and therefore lack the resolution necessary for our analysis .
spectra can be very well fitted by assuming an incommensurate modulation of the fe moments with several harmonics ( fig.[ni_spectres4k ] , left panel ) as described in ref.@xcite . according to this model , the hyperfine field @xmath36
can be developed as a fourier series along the propagation vector @xmath37 of the incommensurate modulation : @xmath38 . \label{hhf}\ ] ] the fit of the mssbauer spectra allows us to extract the parameters @xmath39 . due to the local character of the technique
, the vector @xmath40 can not be determined .
fourier coefficients obtained for @xmath11=0.01 and 0.03 are given in table i and the corresponding modulations are shown in fig .
[ ni_spectres4k ] ( right panel ) . note that there are several sets of parameters that yield equally good fits of the spectra .
the shape of the modulation appears to evolve with doping : for @xmath4 it is close to a `` square - wave '' while it approaches a sine - wave for @xmath14 , similar to ba(fe@xmath1co@xmath2)@xmath3as@xmath3 .
the relative intensities of the lines are close to 3:4:1:1:4:3 , as expected when the hyperfine field , and thus the fe magnetic moment , is perpendicular to the @xmath19-beam .
this is in good agreement with neutron scattering measurements in ba(fe@xmath1ni@xmath2)@xmath3as@xmath3 at low doping levels , showing that moments are parallel to * _ a _ * as in pure bafe@xmath3as@xmath3 .
.fourier coefficients ( in t ) obtained for ni - doping with @xmath4 ( three sets yielding equally good fits ) and @xmath14 ( only one set is shown , although several sets fit the data equally well ) . [ cols="^,^,^,^,^,^,^,^,^",options="header " , ] so far we have shown that the hypothesis of ic - sdw leads to very good fits of the mssbauer spectra .
we now show that good fits can also be obtained with the empirical model described in ref.@xcite , based on first - principle calculations @xcite , assuming that the dopant induces a perturbation of the fe moment amplitude . within this model , the fe moment value @xmath41(_*r * _ ) at distance _ * r * _ from the origin , is given by the expression : @xmath42 where @xmath43 exp[-@xmath44 . in this latter expression
, @xmath45 is the distance to the @xmath46th dopant atom , the sum runs over all dopant atoms and @xmath47 is the sdw vector for the commensurate pattern observed in bafe@xmath3as@xmath3 .
the parameter @xmath23 gives the reduction of the spin of the dopant atoms , while @xmath48 describes the spatial extent of the perturbation and @xmath49 is an unperturbed fe moment .
this model allows one to reproduce correctly the @xmath5as nmr spectra for different ni concentrations @xcite , with parameters @xmath50 and @xmath51 . in the following
, we shall refer to this model as the `` perturbation '' model .
= 0.01 and 0.03 , in the framework of the `` perturbation '' model along a line in the ( @xmath52,@xmath53 ) plane parallel to @xmath52 .
we use expression ( [ mdop ] ) with parameters @xmath23=0.6 and @xmath51 .
the red dashed lines mark the positions of the dopant atoms . according to this model , the periodicity of the magnetic structure should be completely broken for @xmath11=0.03 , in disagreement with neutron scattering results.,width=11 ] ni@xmath2)@xmath3as@xmath3 with @xmath4 , fitted to the `` perturbation '' model .
* right : * histogram of the reduced fe polarizations for the ic - sdw and `` perturbation '' models for @xmath11=0.01.,width=11 ] in order to visualize the effect of the dopant atom on its magnetic environment within this model , we chose a ( 300 @xmath54 300 ) point mesh representing one fe - as ( @xmath52,@xmath53 ) plane and placed ni atoms at random in this plane according to the concentration .
we compute the fe polarization along a given direction as a function of the distance from the origin for @xmath11=0.01 and 0.03 ( fig.[polarization ] ) . for @xmath4 ,
the polarization remains close to 1 for most sites , although destructive interferences lead to a few fe sites with strongly depleted or zero moments . for @xmath14 , this model leads to an average polarization much smaller than 1 and even negative for a few sites .
the disorder induced by the presence of ni dopants would be so strong that the periodicity of the magnetic structure would be completely broken .
if this were the case , no magnetic order could have been detected by neutron scattering .
however , bragg peaks are clearly observed at concentrations close to @xmath14 @xcite , which shows that the model is not valid beyond the very dilute limit .
note that for even larger dopant content , the polarization of all sites would turn negative and smaller than -1 , which is unphysical .
we therefore restricted our simulations of the mssbauer spectra to @xmath4 .
good agreement with the data was found for this concentration for a wide range of parameter values , with @xmath23 varying from 0.4 to 0.8 and @xmath48 from @xmath55 to @xmath56 .
the obtained fit is shown in the left panel of fig.[ni_spectre4khist ] for @xmath50 and @xmath51 .
note that even for @xmath4 , the rather large values of @xmath23 and @xmath48 seem to overestimate the influence of magnetic disorder with respect to what is predicted by theoretical calculations @xcite .
one can wonder why both the ic - sdw and `` perturbation '' models reproduce the mssbauer spectra well .
this is in fact due to similar features of the distributions of fe hyperfine fields for both models , as shown in the right panel of fig.[ni_spectre4khist ] .
both histograms are very asymmetric , with a quasi - divergence at the maximum field value and a continuous distribution extending down to zero polarisation . for the `` perturbation '' model ,
the noise is due to the random lattice positions of the dopant atoms .
ni@xmath2)@xmath3as@xmath3 with @xmath4 .
continuous lines are fits assuming either an incommensurate modulation ( left panel ) or an dopant - induced distribution of the fe hyperfine field ( right panel ) . at 116k , a lorentzian - shaped component was added to the fit in order to represent the paramagnetic fraction of the sample.,width=8 ] spectra recorded at higher temperatures for @xmath4 are shown in fig.[ni1p_spectres ] . all the spectra measured below @xmath16k
can be well fitted either with the ic - sdw or the `` perturbation '' model with the parameters @xmath50 and @xmath51 , as described above .
the ic - sdw model yields systematically better fits with @xmath571.10 - 1.25 , while fits with the `` perturbation '' model have @xmath571.5 - 2 .
however , this is not very significant because the ic - sdw model involves more parameters .
fits with the ic - sdw model yield similar shapes of the modulations in the temperature range from 4.2 to 110k .
above @xmath29 determined by resistivity measurements , a paramagnetic component develops at the expense of the magnetic one in a temperature range of 0.1@xmath29 , similar to pure bafe@xmath3as@xmath3 , indicative of a first order transition .
the paramagnetic fraction , shown as a function of temperature in fig.[ni1p_paramagfraction ] , grows rapidly with @xmath58 , so that at @xmath59k the sample is almost fully paramagnetic .
note that when the fraction of the magnetic phase is less than @xmath6050% , it is impossible to perform an accurate quantitative determination of the paramagnetic fraction , due to the lack of spectral resolution .
therefore , no experimental data are shown for temperatures above 118k . in the temperature range from 110 to 116k , fits with the ic - sdw model were performed by assuming that the shape of the modulation does not evolve with @xmath58 , in order to limit the number of fitting parameters .
hysteresis is observed between warming and cooling procedures .
this , together with the steep decrease of the hyperfine field when approaching @xmath29 ( fig.[ni1p_paramagfraction ] inset ) , confirms the first order nature of the transition . here
the hyperfine field @xmath61 was derived from the fit with both the `` perturbation '' and ic - sdw models . in the former case
, @xmath61 is a free parameter proportional to @xmath49 . in the latter case
, @xmath61 is the average over a period of the incommensurate modulation , given by the expression : @xmath62 ni@xmath2)@xmath3as@xmath3 with @xmath4 : temperature dependence of the paramagnetic fraction for increasing and decreasing temperature , derived from the fit of mssbauer spectra with an ic - sdw .
inset : temperature dependence of the hyperfine field , obtained from both fits with an ic - sdw and dopant - induced perturbation .
the vertical dashed lines indicate @xmath63k .
lines are guides for the eye.,width=8 ] we also fitted the spectrum recorded at 4.2k in ba(fe@xmath1co@xmath2)@xmath3as@xmath3 @xcite for @xmath11=0.014 to the `` perturbation '' model .
a rather good fit is obtained ( not shown ) with parameters @xmath64 and @xmath65 .
the value of @xmath48 which describes the extent of the spatial distribution is found to be the same as for ni doping .
we have shown in the previous section that , in bafe@xmath3as@xmath3 doped with co or ni , the @xmath0fe mssbauer spectra can be equally well reproduced assuming either an ic - sdw or a dopant - induced perturbation model .
so the mssbauer spectra do not allow us to distinguish between the two .
recently , it was claimed that the ic - sdw assumption yielded poor fits to the @xmath5as nmr spectra in ni doped bafe@xmath3as@xmath3 @xcite , and that this fact favoured the `` perturbation '' model . in these latter works ,
the simulations of the nmr spectra with the ic - sdw assumption were carried out using the incommensurate modulation inferred from the nmr data at @xmath11=0.06 co - doping @xcite . here , we perform simulations of the nmr spectra with the incommensurate modulations derived from the mssbauer spectra and we show that good agreement is achieved .
we consider a plane with ( 500 @xmath54 500 ) fe sites and calculate , for each as site , the corresponding hyperfine field .
the @xmath5as hyperfine hamiltonian can be written as : @xmath66 here the sum runs over the four fe nearest neighbors of an as nucleus with moment @xmath67 , @xmath19 is the gyromagnetic ratio of @xmath5as ( @xmath68mhz / t ) , @xmath69 is the nuclear spin operator and @xmath70 the hyperfine tensor .
four hyperfine tensor components are known : @xmath71t/@xmath9 , @xmath72t/@xmath9 and @xmath73t/@xmath9 , determined previously in pure bafe@xmath3as@xmath3 @xcite .
the fifth component @xmath74 is unknown , but it has to be taken into account when the external field is perpendicular to the @xmath75 axis . of the incommensurate modulation is parallel to the @xmath52 axis ,
the term in the hamiltonian involving @xmath74 is zero . ]
the fe magnetic moment at position @xmath76 is given by : @xmath77 , \label{mhk}\ ] ] the total propagation vector being @xmath78 .
the parameters @xmath39 are known from the mssbauer spectra , therefore the free parameters are @xmath40 , @xmath49 and @xmath74 . in fig.[rmn - moss ] we show the nmr spectra from refs.@xcite for two samples in the low doping regime , with the external field parallel to the @xmath75 axis or along [ 110 ] .
simulations with the dopant - induced perturbation model are shown as dashed lines , with parameters @xmath50 and @xmath51 .
the quadrupolar frequency was taken as @xmath79mhz and the magnetic moments were 0.77@xmath9 for @xmath80 and 0.65@xmath9 for @xmath81 .
as nmr spectra in ba(fe@xmath1ni@xmath2)@xmath3as@xmath3 taken from refs.@xcite . for @xmath11=0.0072 : * a * ) the external magnetic field applied parallel to the @xmath75 axis , and * b * ) along the [ 110 ] direction .
for @xmath81 : * c * ) the external magnetic field applied parallel to the @xmath75 axis .
the solid lines are simulations with the ic - sdw model and the dashed lines are simulations with the `` perturbation '' model.,width=9 ] for the simulations of the nmr spectra for @xmath80 assuming ic - sdw , we considered the modulation corresponding to the @xmath11=0.01 mssbauer spectrum .
the best fits to the nmr spectra were found with the set of parameters n@xmath82 1 in table i. regardless of the propagation vector direction ( @xmath7 , @xmath8 or [ 110 ] ) , we find very good agreement between simulations and the data for realistic values of @xmath83 , @xmath49 and @xmath74 .
continuous lines in figs.[rmn - moss ] * a * and * b * show the simulations obtained when the modulation is along [ 110 ] , with the external field parallel or perpendicular to the @xmath75 axis .
the parameters are @xmath84 , @xmath85@xmath9 and @xmath86t/@xmath9 .
the obtained distribution was convoluted with a gaussian with full width at half maximum ( fwhm ) 0.05 t , in order to take into account the disorder induced by the ni substitution .
simulations are not very sensitive to the value of the incommensurate parameter if it is kept small .
good fits are obtained in the range @xmath87 . in a similar manner , we simulated the nmr spectrum obtained for @xmath81 assuming ic - sdw with the fourier components for @xmath14 in table i and with the same @xmath83 and @xmath74 values as above .
figure [ rmn - moss ] c shows the simulation obtained for a modulation along [ 110 ] , with @xmath88@xmath9 , convoluted with a gaussian of fwhm 0.2 t .
it is quite close to the experimental lineshape .
our @xmath0fe mssbauer spectroscopy measurements on the pure and ni - doped bafe@xmath3as@xmath3 indicate the presence of a first order transition towards a spin density wave state . in pure bafe@xmath3as@xmath3 ,
the sdw is commensurate , in agreement with neutron scattering results , and the first order nature of the transition is in line with @xmath5as nmr results . in ba(fe@xmath1ni@xmath2)@xmath3as@xmath3 with @xmath4 and @xmath14 ,
the magnetic phase is characterized by a broad distribution of fe hyperfine fields .
we analyzed our results in relation to nmr measurements in the framework of two different models : an incommensurate spin density wave and a dopant - induced perturbation of the fe moments .
we have shown that both @xmath5as nmr and mssbauer spectra can be fitted well with either model , contrary to what is stated in refs.@xcite . therefore , at low doping it is not possible to conclude from the available nmr or mssbauer data whether the `` perturbation '' or the ic - sdw model is relevant in ni doped bafe@xmath3as@xmath3 . at larger doping ,
a more elaborate model is necessary in order to track the role of disorder , because the `` perturbation '' model collapses even for moderate concentrations such as @xmath14 . in the more studied ba(fe@xmath1co@xmath2)@xmath3as@xmath3 family ,
the presence of ic - sdw is still a matter of debate . while ic - sdw was inferred from local techniques , resonant x - ray scattering measurements for @xmath89 did not find any sign of incommensurability along @xmath7 or @xmath8 , down to relatively small values of @xmath83 @xcite .
neutron scattering detected an ic - sdw only recently , in a small concentration range above @xmath90 , with a propagation vector @xmath91 along @xmath8 and @xmath83 values ranging from 0.02 to 0.03 @xcite .
the mssbauer spectrum for this concentration is well fitted with these parameters .
one can wonder whether ic - sdw could be present at all concentrations in the low doping regime of ba(fe@xmath1co@xmath2)as@xmath3 , because the direction of the propagation vector of the modulation could change with @xmath11 , following variations in the fermi surface nesting . this would further complicate the detection of ic - sdw by neutron scattering and resonant x - ray measurements . while disorder due to substitutions is inherent in ba(fe@xmath1ni@xmath2)@xmath3as@xmath3 and ba(fe@xmath1co@xmath2)@xmath3as@xmath3 , there are a few stoichiometric systems in which the signature of incommensurability was observed and the presence of disorder can be ruled out . in the `` 122 family '' , the stoichiometric system srfe@xmath3as@xmath3 presents , under pressure , a crossover versus temperature from commensurate to incommensurate af , seen by @xmath5as nmr measurements @xcite . in nafeas ,
a similar crossover was evidenced at ambient pressure by @xmath92na and @xmath5as nmr @xcite .
this system presents filamentary superconductivity below 23k and , similar to other fe - based pnictides , shows a structural phase transition from tetragonal to orthorhombic at 50k , followed by a magnetic one towards an antiferromagnetically ordered state at 40k .
these observations suggests that incommensurability is not only related to doping but rather , might be connected to subtle changes of the fermi surface . in conclusion
, we have shown that at very low doping levels , both ic - sdw and dopant - induced perturbation models fit the nmr and mssbauer spectra well in ba(fe@xmath1ni@xmath2)@xmath3as@xmath3 .
however , the latter model collapses even for moderate dopant concentrations @xmath14 , because it overestimates the disorder effect .
further theoretical calculations are required in order to sort out the precise influence of the dopant on the magnetic properties of pnictides . on the other hand ,
the ic - sdw model successfully reproduces the spectra over the whole doping region .
this , together with the observation of ic - sdw in the stoichiometric systems mentioned above and the similarity with the co family , suggests that ic - sdw might also be present in the ni family .
authors wish to thank julien bobroff and henri alloul for helpful discussions and reading of the manuscript , and sylvie poissonnet ( srmp / cea ) for the chemical analysis of the samples .
this work was supported by the anr grant pnictides .
acknowledges financial support from `` triangle de la physique '' .
+ 45 url # 1#1urlprefix[2][]#2 kamihara y , watanabe t , hirano m and hosono h 2008 _ j. am .
_ mazin i i and schmalian j 2009 _ physica c _ * 469 * 614 laplace y , bobroff j , rullier - albenque f , colson d and forget a 2009 _ phys .
b _ * 80 * 140501 julien m h , mayaffre h , horvati m , berthier c , zhang x d , wu w , chen g f , wang
n l and luo j l 2009 _ europhys .
lett . _ * 87 * 37001 luetkens h , klauss h h , kraken m , litterst f j , dellmann t , klingeler r , hess c , khasanov r , amato a , baines c , kosmala m , schumann o j , braden m , hamann - borrero j , leps n , kondrat a , behr g , werner j and buchner b 2009 _ nat .
_ * 8 * 305 lumsden m d and christianson a d 2010 _ j. phys .
: condens . matter _ * 22 * 203203 singh d , du m h , zhang l , subedi a and an j 2009 _ physica c _ * 469 * 886 dong j , zhang h j , xu g , li z , li g , hu w z , wu d , chen g f , dai x , luo j l , fang z and wang n l 2008 _ europhys .
* 83 * 27006 fink j , thirupathaiah s , ovsyannikov r , drr h a , follath r , huang y , de jong s , golden m s , zhang y z , jeschke h o , valent r , felser c , dastjani farahani s , rotter m and johrendt d 2009 _ phys . rev .
b _ * 79 * 155118 brouet v , marsi m , mansart b , nicolaou a , taleb - ibrahimi a , le fvre p , bertran f , rullier - albenque f , forget a and colson d 2009 _ phys . rev .
b _ * 80 * 165115 ding h , richard p , nakayama k , sugawara k , arakane t , sekiba y , takayama a , souma s , sato t , takahashi t , wang z , dai x , fang z , chen g f , luo j l and wang n l 2008 _ europhys .
lett . _ * 83 * 47001 liu c , samolyuk g d , lee y , ni n , kondo t , santander - syro a f ,
budko s l , mcchesney j l , rotenberg e , valla t , fedorov a v , canfield p c , harmon b n and kaminski a 2008 _ phys .
lett . _ * 101 * 177005 park j t , inosov d s , yaresko a , graser s , sun d l , bourges p , sidis y , li y , kim j h , haug d , ivanov a , hradil k , schneidewind a , link p , faulhaber e , glavatskyy i , lin c t , keimer b and hinkov v 2010 _ phys . rev .
b _ * 82 * 134503 rice t m 1970 _ phys . rev .
b _ * 2 * 3619 vorontsov a b , vavilov m g and chubukov a v 2009 _ phys . rev . b _ * 79 * 060508 vorontsov a b , vavilov m g and chubukov a v 2010 _ phys .
b _ * 81 * 174538 cvetkovic v and tesanovic z 2009 _ phys . rev . b _ * 80 * 024512 kitagawa k , katayama n , gotou h , yagi t , ohgushi k , matsumoto t , uwatoko y and takigawa m 2009 _ phys . rev . lett . _ * 103 * 257002 bonville p , rullier - albenque f , colson d and forget a 2010 _ europhys .
lett . _ * 89 * 67008 marsik p , kim , k w , dubroka a , rssle m. , malik v k , schulz l , wang c n , niedermayer c , drew a j , willis m , wolf t and bernhard c 2010 _ phys .
lett . _ * 105 * 057001 christianson a d , lumsden m d , nagler s e , macdougall g j , mcguire m a , sefat a s , jin r , sales b c and mandrus d 2009 _ phys . rev
. lett . _ * 103 * 087002 pratt d k , tian w , kreyssig a , zarestky j l , nandi s , ni n , budko s l , canfield p c , goldman a i and mcqueeney r j 2009 _ phys . rev
* 087001 pratt d k , kim m g , kreyssig a , lee y b , tucker g s , thaler a , tian w , zarestky j l , budko s l , canfield p c , harmon b n , goldman a i and mcqueeney r j 2011 _ phys .
lett . _ * 106 * 257001 dioguardi a p , aproberts warren n , shockley a c , budko s l , ni n , canfield p c and curro n j 2010 _ phys .
b _ * 82 * 140411 dioguardi a p , aproberts warren n , shockley a c , budko s l , ni n , canfield p c and curro n j 2011 _ arxiv : 1102.1936 _ wang p , stadnik z m , zukrowski j , thaler a , budko s l , canfield p c 2011 _ phys .
b _ * 84 * 024509 kreyssig a , kim m g , nandi s , pratt d k , tian w , zarestky j l , ni n , thaler a , budko s l , canfield p c , mcqueeney r j , goldman a i 2010 _ phys .
b _ * 81 * 134512 rullier - albenque f , colson d , forget a and alloul h 2009 _ phys .
_ * 103 * 057001 olariu a , rullier - albenque f , colson d and forget a 2011 _ phys .
b _ * 83 * 054518 rotter m ,
tegel m , schellenberg i , schappacher f m , pttgen r , deisenhofer j , gnther a , schrettle f , loidl a and johrendt d 2009 _ new j. phys . _ * 11 * 025014 rotter m , tegel m , johrendt d , schellenberg i , hermes w and pttgen r 2008 _ phys .
b _ * 78 * 020503 nowik i , felner i , ni
n , budko s l and canfield p c 2010 _ j. phys . : condens
. matter _ * 22 * 355701 huang q , qiu y , bao w , green m a , lynn j w , gasparovic y c , wu t , wu g and chen x h 2008 _ phys .
lett . _ * 101 * 257003 kitagawa k , katayama n , ohgushi k , yoshida m and takigawa m 2008 _ j. phys . soc .
jpn . _ * 77 * 114709 dunlap r a , eelman d a and mackay g r 1998 _ j. non - cryst .
solids _ * 223 * 141 verma h r , sundqvist t , wppling r and arnberg l 1985 _ physica scripta _ * 32 * 155 felner i and nowik i 2011 _ j. supercond .
novel mag .
_ * 24*(4 ) 1363 nowik i and felner i 2009 _ physica c _ * 469 * 485 kemper a f , cao c , hirschfeld p j and cheng h p 2009 _ phys .
b _ * 80 * 104511 wang m , luo h , zhao j , zhang c , wang m , marty k , chi s , lynn j w , schneidewind a , li s and dai p 2010 _ phys .
b _ * 81 * 174524 kim m g , kreyssig a , lee y b , kim j w , pratt d k , thaler a , budko s l , canfield p c , harmon b n , mcqueeney r j and goldman a i 2010 _ phys .
b _ * 82 * 180412 kitagawa k , mezaki y , matsubayashi k , uwatoko y and takigawa m 2011 _ j. phys .
* 80 * 033705 klanjek m , jegli p , lv b , guloy a m , chu c w , aron d 2011 _ phys .
b _ * 84 * 054528 | we report @xmath0fe mssbauer spectral results in pure and doped ba(fe@xmath1ni@xmath2)@xmath3as@xmath3 with @xmath4 and 0.03 .
we show that all these materials present a first - order magnetic transition towards a magnetically ordered state . in the doped compounds ,
a broad distribution of fe hyperfine fields is present in the magnetic phase .
we successfully fit the mssbauer data in ba(fe@xmath1ni@xmath2)@xmath3as@xmath3 in the framework of two different models : 1 ) an incommensurate spin density wave ; 2 ) a dopant - induced perturbation of the fe polarization , recently proposed to interpret @xmath5as nmr data in ba(fe@xmath1ni@xmath2)@xmath3as@xmath3 , which is valid only in the very dilute limit @xmath4 .
moreover , we show here that these nmr data can also be successfully analysed in terms of the `` incommensurate model '' for all doping contents by using the parameters obtained from the mssbauer spectral analysis .
therefore it is not possible to rule out the presence of an incommensurate spin density wave on the basis of the @xmath5as nmr data . |
cp violation in the b - meson system has generated considerable experimental and theoretical interest , as potentially large effects are expected .
searches for cp(t ) violation using the small sample of @xmath4 decays at the lep collider at cern provide proofs of principle " for analysis techniques which will be employed by future b - factory experiments .
indirect cp violation is possible in the @xmath5 system provided the weak eigenstates @xmath5 and @xmath6 differ from the mass eigenstates @xmath7 and @xmath8 : @xmath9 where @xmath10 and @xmath11 parametrize cp and cpt violation , respectively@xcite .
these parameters have been investigated using semileptonic b hadron decays , resulting in limits of order @xmath12 on both @xmath10 and @xmath11@xcite . in the standard model , re(@xmath13 )
is expected to be around @xmath14@xcite , but it could be up to an order of magnitude larger in superweak models@xcite .
a non - zero value of @xmath13 gives rise to a time - dependent rate asymmetry , @xmath15 , in inclusive @xmath5 _ vs. _ inclusive @xmath6 decays@xcite , defined as : @xmath16 where @xmath17 and @xmath18 are the decay rates of @xmath5 and @xmath6 mesons .
for an unbiased selection of @xmath5 and @xmath19 mesons , the time - dependent inclusive decay rate asymmetry can be rewritten in terms of proper decay time @xmath20 : @xmath21 , \label{eq : asym2}\end{aligned}\ ] ] where @xmath22 is the cp - violating observable , @xmath23 is the @xmath5 oscillation frequency , and @xmath24 is the @xmath5 lifetime . + for @xmath25 , the parameter @xmath22 is related to @xmath26 by : @xmath27 .
furthermore , cpt invariance implies that @xmath28 .
if cpt violation occurred , the lifetimes of @xmath0 and @xmath1 hadrons could be different : @xmath29 \tau_{av},\end{aligned}\ ] ] where @xmath30 is the average and @xmath31 is the fractional difference in lifetimes .
the measurement of the time - dependent rate asymmetry , a(t ) , and the extraction of re(@xmath10 ) proceeds in several steps .
first , selected @xmath2 events are divided into 2 hemispheres defined by the plane @xmath32 to thrust axis and containing the @xmath33 interaction point .
a sample of about 400,000 @xmath34 events is identified using b - tagging techniques described in detail in @xcite .
the b - tags rely on the presence of a displaced secondary vertex or a high momentum lepton . in each event , the hemisphere containing the b - tag is referred to as the
t - tagged " hemisphere .
the b - tag has a @xmath35 of 37 % for @xmath36 events and a non - b impurity of 13% .
next , the b hadron proper decay time , t , in the opposite measurement " hemisphere is reconstructed by forming a secondary vertex , measuring the decay distance from the primary vertex , and estimating the b hadron energy .
the quantity @xmath22 is then extracted via a binned @xmath37-fit to the observed time - dependent asymmetry in bins of reconstructed proper time . the production flavor estimate , @xmath38 , in the tagged hemisphere is the output of a neural net with the following inputs : 1 .
jet charge of the highest energy jet , @xmath39 , with ( with momentum weight @xmath40 ) .
the jet charge is defined as : @xmath41 where @xmath42 is the longitudinal momentum component with respect to the jet axis and @xmath43 is the charge of track @xmath44 .
vertex charge , @xmath45 , where @xmath46 is the weight for track @xmath44 to have come from a secondary instead of a primary vertex .
weights are derived from an artificial neural network with three inputs : the momentum of track @xmath44 , the transverse momentum of track @xmath44 with respect to the jet axis , and the impact parameter .
3 . uncertainty on the vertex charge , @xmath47 .
the product of lepton charge and output of neural network used to select @xmath48 decays , a lepton tag . the jet charge and vertex charge are not charge symmetric due to detector effects resulting in the difference in the rate and reconstruction of positively and negatively charged tracks .
offsets are evaluated using inclusive samples of t - tagged events and are subtracted prior to constructing @xmath38 .
the lepton tag is diluted by @xmath0 mixing , cascade @xmath0 decays , @xmath49 decays , and fake leptons .
separate networks with fewer inputs are trained for events with a vertex or a lepton only .
the variable @xmath38 is defined as : @xmath50 are the numbers of mc @xmath0 and @xmath51 hadron hemispheres with a given value of artificial neural network output @xmath52 and @xmath53 is the mistag probability . if @xmath54 , then the tagged hemisphere is more likely to contain a @xmath55 than a @xmath56 , and vice versa
. if @xmath57 , both hypotheses are equally likely .
the @xmath38 flavor tag has some sensitivity to the _ decay _ flavor of the tagged hemisphere , which is not desirable in an inclusive measurement .
therefore , another tag @xmath58 is applied in the opposite , or measurement , hemisphere . the @xmath38 output in the t - tagged hemisphere , as well as
the jet charge in the measurement hemisphere , @xmath58 ( with momentum weight @xmath59 ) , are combined to construct the composite variable : @xmath60 -1 .
\end{aligned}\ ] ] again , if @xmath61 ( @xmath62 ) , the so - called m - tagged " hemisphere contains a @xmath0-hadron tag ( @xmath1-hadron ) tag .
the @xmath63 variable is designed to be sensitive to the production , but not the decay flavor of the @xmath0-hadron , thus avoiding biases to the reconstructed proper time measurement .
after flavor tagging , 394119 events remain in the data sample .
the cp - violating parameter @xmath22 can be extracted from the rate asymmetry distribution , a(t ) , as defined in equation [ eq : asym2 ] .
this is accomplished by calculating the number of @xmath0-hadron m - tags minus the number of @xmath51-hadron m - tags in bins of reconstructed proper time @xmath64 and performing a binned @xmath37 fit to measure @xmath22 .
the @xmath0-hadron proper time is defined as : @xmath65 where @xmath66 is the hadron decay length , @xmath67 is the @xmath0-hadron energy , and @xmath68 is the mass of the @xmath0-hadron , taken be that of the @xmath69 and @xmath5 ( 5.279 gev)@xcite .
the hadron decay length , @xmath66 , is reconstructed in the measurement hemisphere by first forming a seed " secondary vertex using the two tracks with the largest impact parameter , @xmath70 , relative to the primary vertex in the highest energy jet . all tracks with @xmath71 gev , @xmath72 cm , and @xmath73 cm , which are consistent with the seed " vertex
are then added to it via an interative procedure .
the secondary vertex must contain at least 3 tracks and have an invariant mass exceeding 0.8 gev , assuming all constituent tracks are pions . to further eliminate badly reconstructed or fake secondary vertices
, the secondary vertex must be kinematically consistent with a long - lived particle originating from the primary vertex .
secondary vertices meeting the above criteria are identified in approximately 70@xmath74 of m - tagged hemispheres for both signal and background .
the decay length @xmath66 between the primary and secondary vertices is then calculated using the jet axis as a constraint .
the @xmath0-hadron energy is computed by first estimating the energy of the @xmath0-jet by treating the event as a 2-body decay of a @xmath75 into a @xmath0-jet of mass @xmath76 and another object .
the charged and neutral fragmentation energy , @xmath77 , was estimated using the procedure described in @xcite , involving the charged track weights @xmath78 , and the unassociated electromagnetic calorimeter clusters weighted according to their angle with respect to the jet axis .
the @xmath0-hadron energy is then , @xmath79 .
the reconstructed proper time distribution described by equation [ eq : propt ] is convolved with 2 gaussians to account for detector resolution effects .
the rms widths of the resolution functions are 0.33 and 1.3 ps and are determined from monte carlo studies .
about 65@xmath74 of events lie within the narrower gaussian .
these resolution functions represent an average over all true decay proper times @xmath20 .
the non - gaussian effects apparent in small slices of @xmath20 due to contamination from primary vertex tracks are not critical in this analysis , as the result is not particularly dependent on accurate decay time resolution .
for each of 34 time bins @xmath80 ( in the range -2 to 15 ps ) , the asymmetry is calculated in 10 bins @xmath81 of @xmath82 : @xmath83 and the error @xmath84 is given by : @xmath85 where @xmath86 ( @xmath87 ) is the number of events with @xmath61 ( @xmath62 ) .
the factor @xmath88 corrects for the tagging dilution ( mis - tagging ) , which reduces the observed asymmetry for imperfectly tagged events .
the 10 estimates of @xmath89 in each bin @xmath44 are then averaged , weighting by @xmath90 to get @xmath91 .
a binned @xmath37-fit to the reconstructed proper time which accounts for non-@xmath5 background , the @xmath5 lifetime , and the gaussian time resolution functions yields : @xmath92 the asymmetry , a(t ) , as a function for reconstructed proper time @xmath20 is shown in figure [ fig : asym_fit ] .
the dots denote opal data , the solid line the fit result , and the dashed line the expected asymmetry for @xmath93 .
5.0 truein .2 cm the systematic uncertainties are summarized in table [ tab : syst ] .
detailed descriptions of the various contributions can be found in reference @xcite .
if the reconstruction efficiency for @xmath5 decays to different numbers of charm hadrons is not the same , the expected asymmetry could take the form : @xmath94 repeating the fit , letting both @xmath22 and @xmath95 vary gives : @xmath96 differences in efficiency are not significant , as @xmath22 does not change much .
the systematic uncertainties on the measurement of @xmath95 are listed in the second column of table [ tab : syst ] .
.systematic uncertainties on the measurements of @xmath22 and @xmath95 .
[ cols="<,^,^",options="header " , ] [ tab : syst ]
the fractional difference between the @xmath0 and @xmath1- hadron lifetimes is measured by dividing the data into 20 bins of @xmath98 ( related to @xmath0/@xmath1-hadron purity ) and performing a simultaneous fit of the reconstructed proper time distrbutions to the expected proper time distribution .
the fit yields values for both @xmath99 and @xmath97
. however , the result for @xmath100 has a large systematic uncertainty due to the time resolution function and should not be interpreted as a measurement of the average @xmath0-hadron lifetime .
the expected distribution accounts for time resolution effects , non-@xmath36 background , the lifetimes of @xmath0 and @xmath1-hadrons , and a background component with a lifetime @xmath101 .
the fit result is : @xmath102 the uncertainty in the flavor mistag rate dominates the systematic uncertainty on @xmath103 .
reconstructed proper time distributions in 4 ranges of @xmath63 are shown in figure [ fig : rpt ]
. 5.0 truein -.2 cm
an inclusive sample of @xmath0-hadron decays is used to search for cp and cpt violation effects .
no such effects are seen . from the time dependent asymmetry of inclusive
@xmath5 decays , the cp violation parameter is measured to be : @xmath104 this result agrees with the opal measurement using semileptonic b decays : @xmath105@xcite , and is also in agreement with other less precise results from cleo and cdf@xcite .
the fractional difference in the lifetimes of @xmath0 and @xmath1-hadrons is also measured to be : @xmath102 this is the first analysis accepted for publication which tests the equality of the @xmath0 and @xmath1-hadron lifetimes .
these results are summarized in figures [ fig : worldcp1 ] and [ fig : worldcp2 ] .
for example : r. hawkings , ` cp violation in b decays at lep ' , talk presented at the 3rd international conference on hyperons , charm and beauty hadrons , 2nd july 1998 , genova , italy , to be published in nucl . | a search for cp(t ) violation is performed and the fractional difference between the @xmath0 and @xmath1 hadron lifetimes is measured using reconstructed secondary vertices in inclusive b hadron decays selected from 3.1 million @xmath2 events .
the data were collected by the opal experiment at the lep collider at cern at @xmath3 gev from 1991 - 1995 . |
the development of methods based on the density functional theory@xcite and of the computer power has allowed to conceive calculations of phase diagram from first principles@xcite as an alternative to laboratory experimentations .
traditionally , only substitutional effects were considered , which was good enough to reproduce the topology of most phase diagrams .
so as to obtain a more quantitative agreement with experimental data , it has been shown more recently that electronic excitations@xcite as well as lattice vibrations@xcite could play an important part in the relative stability of different phases .
we chose to test the ability of these first principles methods to predict the solubility limit of zr in an aluminium solid solution this part of the al - zr phase diagram being interesting because of the presence of a metastable phase .
the al - richest intermediate phase of the al - zr phase diagram is al@xmath0zr .
this compound has the do@xmath1 structure which is body - centered tetragonal with eight atoms per unit cell , some of these atoms being allowed by symmetry to move along the main axis of the unit cell ( fig .
[ al3zr_fig ] ) .
it is stable up to 1580@xmath510@xmath6c .
the solubility limit of zr in al ( fcc ) is really low , the maximum solubility being 0.083 at.% at the peritectic reaction , liquid+zral@xmath0 @xmath7 ( al ) .
the solubilities of zr in both liquid and solid were definitively determined by fink and willey @xcite and the assessed phase diagram is based on their data .
the solid solubility was determined from resistivity data and checked by metallography .
solid solubilities were also reported by glazov _ _ et al . _ _
@xcite , drits _
_ et al . _ _
@xcite ( solubilities determined by means of microstructural analysis and electric resistivity measurements ) , and kuznetsov _ _ et al . _ _
@xcite ( determined from resistivity , microhardness and lattice constant measurements as well as metallography ) .
the solubilities reported in these last studies are higher than the ones of the assessed phase diagram .
( left ) and l1@xmath2 ( right).,scaledwidth=45.0% ] supersaturated solid solution of zr in ( al ) containing as much as 3 at.% zr can be prepared by rapid solidification .
a coherent metastable phase al@xmath0zr precipitates from the supersaturated solution@xcite .
this phase has the structure l1@xmath2 which is simple cubic with 4 atoms per unit cell ( fig .
[ al3zr_fig ] ) .
this metastable phase can also form from the melt as a primary phase during rapid solidification@xcite : al@xmath0zr acts as nuclei for solidification of ( al ) and zr can thus work as a grain refiner of al .
this metastable phase is also responsible for the effectiveness of zr to control recrystallization in al - based alloys : it leads to a more uniform distribution of fine precipitates that pins grains and subgrains boundaries .
moreover , this phase is really stable against coarsening and redissolution , all this making it highly desirable .
as few experimental data are available for this phase , it is hard to fit a thermodynamic model for it . in such a case ,
a first principles calculation of the phase diagram should allow us to obtain properties that are not available experimentally . in order to assess the metastable phase diagram , we studied with the same tools and approximations the l1@xmath2 and do@xmath1 phases
: this allows us to check first the agreement between the stable phase diagram obtained and experimental data , and then to compare it to the metastable phase diagram . in a first part
, we study the stability of ordered compounds based on an fcc underlying lattice in the al - zr system .
the energies of different structures were calculated using an ab - initio method , the full - potential - linear - muffin - tin - orbital ( fp - lmto ) .
the equilibrium parameters , like the volume , shape of the unit cell , or positions of atoms , were obtained . for stable structures
they can be compared to experimental data . using this whole set of calculations we applied the cluster expansion to deduce the energy of any structure based on the same underlying lattice in the al - zr system , examining carefully the way to include volume relaxations . at finite temperature , the electronic excitations ,
the vibrational free energy , and the configurational entropy have to be taken into account . at the end of this part , we are able to obtain a thermodynamic model written in the same way as in a calphad approach and to calculate the corresponding solubility limits .
formation energies were calculated at absolute zero temperature for 26 compounds in the al - zr binary system , all based on an fcc lattice .
calculations were carried out using a full - potential linear - muffin - tin - orbital ( fp - lmto ) method @xcite in the version developed by methfessel and van schilfgaarde @xcite .
the basis used contained 22 energy independent muffin - tin - orbitals ( mto ) per al and zr site : three @xmath8 values for the orbitals s and p and two @xmath8 values for the orbitals d where the corresponding kinetic energies were @xmath9 ry ( spd ) , @xmath10 ry ( spd ) , and @xmath11 ry ( sp ) .
a second panel with a basis composed of 22 energy independent mto with the same kinetic energies was used to make a correct treatment of the 4p semicore states of zr .
the same uniform mesh of points was used to make the integrations in the brillouin zone for valence and semicore states .
the number of divisions along reciprocal vectors was chosen big enough to ensure a convergence of the total energy of the order of 0.1 mry / atom for each structure .
the radii of the muffin - tin spheres were chosen to have a compactness of 47.6% for al sites and
58.4% for zr sites . inside the muffin - tin spheres ,
the potential is expanded in spherical harmonics up to @xmath12 and in the interstitial region spherical hankel functions of kinetic energies @xmath13
ry and @xmath14 ry were fitted up to @xmath12 .
the calculations were performed in the local density approximation ( lda ) @xcite and the parametrization used was the one of von barth - hedin @xcite .
_ et al . _ _
@xcite showed that generalized - gradient corrections have to be included in order to obtain a correct description of the stability of the different phases of pure zr , but as we were interested only in the al - rich part of the phase diagram we did not include these gradient corrections .
ground state energies at equilibrium @xmath15 , equilibrium volumes per atom @xmath16 , and bulk moduli @xmath17 were obtained by fitting the rose equation of state @xcite to the energies calculated for different volumes around the minimum @xmath18 where @xmath19 is the wigner - seitz radius associated to the atomic volume @xmath20 and @xmath21 is related to the bulk modulus @xmath17 through the relation @xmath22 for the different compounds , the energies were optimized with respect to the volume and all other degrees of freedom allowed by the symmetry , like the shape of the unit cell or some atomic positions .
the equilibrium volumes @xmath16 , bulk moduli @xmath17 , and formation energies @xmath23 relative to the fcc phases of both pure al and zr are presented in table [ alzr_relaxed_lda ] .
we note that all the formation energies are negative , and they thus characterize al - zr as an ordering system .
lll.-1.-1.-1 & & & & & + & & & & & + al ( fcc ) & cf4 & cu & 15.82 & 80.78 & 0 .
+ al@xmath24zr & cp32 & ? & 15.99 & 82.56 & -3.04 + al@xmath25zr & ci32 & ? & 16.10 & 84.27 & -6.99 + al@xmath26zr & ti18 & v@xmath27zn@xmath28 & 16.25 & 86.24 & -9.77 + al@xmath29zr ( d1 ) & cf32 & ca@xmath29ge & 16.30 & 87.32 & -14.33 + al@xmath27zr ( d1@xmath30 ) & ti10 & moni@xmath27 & 16.58 & 92.10 & -21.12 + al@xmath0zr ( l1@xmath2 ) & cp4 & cu@xmath0au & 16.12 & 99.59 & -39.00 + al@xmath0zr ( do@xmath31 ) & ti8 & al@xmath0ti & 16.60 & 99.65 & -39.04 + al@xmath0zr ( do@xmath1 ) & ti16 & al@xmath0zr & 16.35 & 100.05 & -40.72 + al@xmath2zr ( @xmath32 ) & hp3 & cdi@xmath2 & 18.01 & 87.16 & -11.73 + al@xmath2zr ( @xmath33 ) & ti6 & mosi@xmath2 & 17.13 & 96.40 & -26.19 + al@xmath2zr ( @xmath34 ) & oi6 & mopt@xmath2 & 17.15 & 96.51 & -26.08 + alzr ( l1@xmath35 ) & tp4 & aucu & 18.15 & 103.33 & -37.07 + alzr ( l1@xmath36 ) & hr32 & cupt & 19.04 & 93.29 & -16.50 + alzr ( ch40 ) & ti8 & nbp & 18.52 & 100.48 & -33.56 + alzr ( d4 ) & cf32 & ? & 18.49 & 92.58 & -14.78 + alzr ( z2 ) & tp8 & ? & 18.63 & 99.70 & -21.03 + zr@xmath2al ( @xmath32 ) & hp3 & cdi@xmath2 & 20.38 & 98.10 & -10.72 + zr@xmath2al ( @xmath33 ) & ti6 & mosi@xmath2 & 19.53 & 104.84 & -24.78 + zr@xmath2al ( @xmath34 ) & oi6 & mopt@xmath2 & 19.44 & 104.05 & -26.36 + zr@xmath0al ( l1@xmath2 ) & cp4 & cu@xmath0au & 19.71 & 107.67 & -27.11 + zr@xmath0al ( do@xmath31 ) & ti8 & al@xmath0ti & 19.88 & 105.14 & -23.49 + zr@xmath0al ( do@xmath1 ) & ti16 & al@xmath0zr & 19.80 & 102.77 & -25.18 + zr@xmath27al ( d1@xmath30 ) & ti10 & moni@xmath27 & 20.31 & 99.85 & -16.30 + zr@xmath29al ( d1 ) & cf32 & ca@xmath29ge & 20.93 & 101.66 & -7.96 + zr ( fcc ) & cf4 & cu & 21.70 & 98.74 & 0 .
+ we examined more closely the stability of the phases l1@xmath2 , do@xmath31 , and do@xmath1 of al@xmath0zr according to relaxations .
l1@xmath2 being cubic , its energy was optimized with respect to atomic volume only , whereas in the tetragonal do@xmath31 phase optimization was performed additionally with respect to the @xmath37 ratio and in the tetragonal do@xmath1 phase to the @xmath37 ratio and to the atomic displacements @xmath38 and @xmath39 ( @xmath38 and @xmath39 are defined on fig .
[ al3zr_fig ] ) . [ cols="<,<,^,^,^,^,^ " , ] al@xmath0zr in the do@xmath1 or l1@xmath2 structures can be considered as a line compound , _
_ i.e. _ _ perfectly ordered : both structures are composed of interpenetrating sublattices of pure al and pure zr .
the configurational entropy of such line compounds can be neglected and these structures only exist for a concentration @xmath40 .
we checked with a calculation using the previous cluster expansions of the formation and vibrational energies that this assumption was correct in the range of temperature of interest when looking at equilibrium with the solid solution .
the free energy of these compounds is then simply given by @xmath41 where @xmath42 ( in mry / atom ) is obtained from our previous calculations of the formation energy ( table [ alzr_relaxed_lda ] ) and of the excess vibrational free energy calculated from the phonon dos ( table [ hight_table ] ) , @xmath43 using the previous expressions for the free energies of the disordered phase and the line compounds al@xmath0zr , we obtained the solubility limit of zr in al ( fcc ) , both in the stable phase diagram when considering the structure do@xmath1 for al@xmath0zr and in the metastable one when considering the structure l1@xmath2 .
as we are in the case of a dilute solid solution , the solubility limit of zr in al ( fcc ) is an analytic function of the temperature@xcite , @xmath44 the solubility we obtained is too low : at the melting temperature of the perictectic it is equal to 0.0016 at.% zr , whereas the one deduced from experimental data is 0.08 at.% zr @xcite . when comparing the variation with respect to @xmath45 of @xmath46 with experimental measurements , we obtain a straight line having the same slope as fink data @xcite ( _ _ cf . _ _ fig .
[ fit ] ) .
this shows that our calculations provide an approximation of the enthalpy difference between the solid solution and the do@xmath1 ordered compound which is consistent with finks data , and that the discrepancy on the solubility limit only arises from an error on the estimation of the entropy difference .
computing the solubility limit of sc in al , asta _ _ et al . _ _
@xcite reached the same conclusion that _ _ ab - initio _ _ calculations correctly predicted the enthalpy difference between the ordered compound and the solid solution when compared to experimental data . in our case
, the error on the entropic terms may come from an overestimation of the vibrational free energy of the disordered phase due to the use of the debye model for this phase . as for the structures do@xmath1 and l1@xmath2 of al@xmath0zr , the debye model overestimates the excess vibrational free energy by a factor @xmath47 ( table [ hight_table ] ) , we can think that we get an error of the same range for the solid solution .
we correct the entropic part , leaving unchanged the enthalpic part , of the parameter @xmath48 defining the excess free energy of the solid solution so as to obtain a perfect agreement with fink data ( _ _ cf . _ _ fig .
[ fit ] ) , and we obtain @xmath49 we thus get a stable solubility limit that is consistent with fink measurements , and we are now able to predict the metastable limit using the expression ( [ omega_corrected ] ) to evaluate the excess free energy of the solid solution . as the structures do@xmath1 and l1@xmath2 of al@xmath0zr have the same vibrational free energy ,
the difference of solubility limit is only due to the difference of ground state energies of these two phases . at the melting temperature of the peritectic
, we obtain a maximal metastable solubility limit equal to 0.275 at.% zr .
this is to compare to the results obtained by a calphad method .
_ et al . _ _ @xcite modelled the metastable phase of al@xmath0zr as a line compound .
they assumed that only an enthalpy term , and no entropy term , contributes to the free energy difference between the stable and metastable phases .
this was done to guarantee that l1@xmath2 does not become stable at high temperature .
moreover there is no experimental data that allows to estimate the entropy of the l1@xmath2 phase .
our calculation of the vibrational free energy shows that such an approximation was correct .
the enthalpy difference between the two compounds was assumed to arise from the coherency of the phase l1@xmath2 with the matrix .
they calculated from the elastic properties of al and an estimate of the composition dependence of the lattice parameter an elastic energy of 1.52 mry / atom .
this estimation is quite close to our calculation ( @xmath50 mry / atom ) as well as to the experimental measurement of desch _ _ et al . _ _
@xcite ( @xmath51 mry / atom ) .
they thus obtained a solubility limit that is higher in the metastable phase diagram than in the stable one and their prevision is really close to our result : they predicted a maximal metastable solubility limit equal to 0.21 at.% . in another calphad study ,
saunders @xcite used the gibbs energy for the disordered al ( fcc ) solution , as derived from the stable equilibrium diagram , to construct the gibbs energy of the ordered l1@xmath2 phase in the bragg - williams approximation .
he found a higher solubility limit for zr in the metastable phase diagram than murray _ _ et al . _ _ as he predicted a metastable solvus composition of 0.3 at.% zr at the melting temperature of the peritectic .
our study thus allows one to estimate the free energy difference between the stable and metastable phases of al@xmath0zr , quantity which is not available experimentally and has to be guessed in these calphad studies .
one thus sees how it is possible to improve the thermodynamic database available to calphad methods .
the equation of state for several compounds in the al - zr system has been computed using the full potential linear muffin tin orbital method ( fp - lmto ) .
these _ _ ab - initio _ _ calculations correctly predict the stability of the phase do@xmath1 for al@xmath0zr if we consider the cell internal relaxations .
we made a cluster expansion of the results of _ _ ab - initio _ _ calculations to predict the formation energy of any compound in the al - zr system based on an fcc underlying lattice .
we showed that despite the size difference between al and zr a totally or globally relaxed expansion for the volume leads to the same result : there is no difference if we use the cluster expansion to predict the formation energy at the equilibrium volume of each structure or at a fixed volume , the energy being then minimized according to the volume . for finite temperature calculations
, we showed that the electronic excitations can be neglected .
the vibrational energy was studied in the harmonic model , using different levels of approximation : the debye model was compared to results obtained from a calculation of the phonon spectrum for al@xmath0zr in the structures do@xmath1 and l1@xmath2 , and it was found that the use of the debye model leads to an overestimation of the vibrational free energy .
so we preferred to use the results from the phonon spectrum to calculate the vibrational free energy of ordered compounds .
for the disordered phase , we chose to make a cluster expansion of the vibrational free energy .
it was only possible with the debye model as this requires less computational time .
we were able to calculate the solubility limit of zr in al ( fcc ) in the the bragg - williams approximation .
the solubility limit obtained is too low compared to experimental data .
we showed that this discrepancy is due to an error in the estimation of the entropy in our thermodynamic model .
this may be due to an overestimation of the vibrational free energy of the disordered phase because of the use of the debye model for this phase . correcting the vibrational entropy of the solid solution
so as to fit the experimental measurements of fink , we were able to predict the metastable solubility limit which lies between the estimation of murray and the one of saunders , both obtained by a calphad method .
we thus showed how first principles calculations can lead to an estimation of the phase diagram .
this approach has the advantage of not requiring any experimental input , and consequently this is not a problem to predict stability of metastable phases . | the experimental solubility limit of zr in al is well - known .
al@xmath0zr has a stable structure do@xmath1 and a metastable one l1@xmath2 .
consequently there is a metastable solubility limit for which only few experimental data are available .
the purpose of this study is to obtain by _ _ ab - initio _ _ calculations the solubility limit of zr in al for the stable as well as the metastable phase diagrams .
the formation energies of several ordered compounds al@xmath3zr@xmath4 , all based on an fcc underlying lattice , were calculated using the fp - lmto ( full potential linear muffin tin orbital ) method .
taking into account all the relaxations allowed by the symmetry , we found the do@xmath1 structure to be the stable one for al@xmath0zr .
this set of results was then used with the cluster expansion in order to fit a generalized ising model through the inverse method of connolly - williams .
different ways to consider volume relaxations were examined .
this allowed us to calculate in the bragg - williams approximation the configurational free energy at finite temperature . according to the previous fp - lmto calculations
the free energy due to electronic excitations can be neglected . for the vibrational free energy of ordered structures we compared results obtained from a calculation of the elastic constants used with the debye model and results obtained from a calculation of the phonon spectrum .
all these different steps lead to a calculation of the solubility limit of zr in al which is found to be lower than the experimental one .
the solubility limit in the metastable phase diagram is calculated in the same way and can thus be compared to the stable one . |
procuring accurate fundamental measurements of the masses and radii of stars remains one of the most outstanding questions in astronomy .
this is crucial in the high - mass regime , which is essential for determining the slope of the initial mass function ( imf ) for high - mass stars .
the sensitivity of stellar models to the initial mass can not be overstated .
this , then , necessitates fundamental determinations of stellar masses with errors @xmath513@xmath6 @xcite in order to reject models with incorrect underlying physics .
as emphasized in the review by @xcite , there are only 95 detached binary systems of any mass in which the mass and radius of both components are known to @xmath7 or better .
all but one are eclipsing binaries , with the sole exception being the sun s nearest neighbor , @xmath8 centauri .
of the 95 total systems there are only eight with @xmath9 , three with @xmath10 , and none with @xmath11 .
even when applying a less stringent uncertainty cutoff , only 14 stars with @xmath11 have both their mass and radius known to @xmath12 or better @xcite .
this paucity of accurately determined stellar masses and radii at the high - mass end requires an expedient redressing .
fundamental measurements of properties for a statistical sample of massive stars will allow us to constrain theoretical models , anchor the imf at the high - mass end , and better comprehend stellar structure and evolution . the cygnus ob2 association presents the ideal laboratory within which to learn about the fundamental properties of massive stars .
it is the nearest massive star - forming region , making it a prime target observationally , even in spite of the extinction in the larger cygnus region , which can range up to @xmath13 .
furthermore , it is known to be replete with an estimated 2,600 @xmath14 400 ob stars , including 120 @xmath14 20 o - type stars , and have a total inferred mass of @xmath15 @xcite .
the region has been the focus of several studies at a variety of wavelengths , dating back to @xcite , who initially noticed a clustering of 11 blue giants near @xmath16 cygni .
@xcite were the first to obtain accurate spectral types , finding 10 to be o - type and one to be an early b - type star .
the subsequent discovery of the heavily reddened and extremely luminous star cyg ob @xmath1712 @xcite was followed by seven @xcite and then 12 additional early - type stars @xcite .
the first comprehensive study of the morphology of the association estimated more than 3,000 total members , of which at least 300 are early - type ( ob ) stars , and a total mass of @xmath18 @xcite . the first synoptic optical study of vi cygnus produced _ ubv _ photometry of 799 stars in the region and obtained spectral types for 76 stars , most of which are the brightest and bluest members of the association @xcite .
they found a distance of 1.7 kpc and the slope of the imf to be @xmath19 .
this provided the first photometric catalog of cyg ob2 and spectroscopically identified the largest number of ob association members .
@xcite performed the first variability survey of the region , obtaining _
i_-band photometry of the central 6 @xmath20 12 arcmin@xmath21 .
they found 29 variables , five of which are o - type , and at least five eclipsing binaries .
@xcite studied the region using 2mass data to characterize the stellar population .
they aver an imf slope of @xmath22 and assert that the large total mass , high central density , and overall compactness of the region require that it be classified as a young globular cluster , a claim originally made by @xcite .
precise classification notwithstanding , early - type members continued to be discovered spectroscopically in the region .
@xcite used 2mass photometry to identify hundreds of potential early - type stars and spectroscopically found 77 of them to be early - type members .
of those , 31 had prior classifications , 24 of which are o - type and the remaining seven b - type , causing them to extrapolate that a total of 60 of their sample of 77 are o - type stars and that cyg ob2 contains between 90100 total o - type stars .
@xcite took spectra of 14 of the remaining 46 and identified all 14 to be early - type , perhaps indicating that the remaining 31 ( one was discovered to have been previously studied ) are also early - type .
@xcite then extrapolated the locally derived cluster luminosity function to argue that there could be tens or even one hundred similar massive open clusters within the milky way .
recently , @xcite built off the canonical optical study of @xcite , spectroscopically following up 183 of their photometrically observed cyg ob2 members .
seventy - three possessed extant spectral classifications as ob stars and 73 of the remaining 110 were also found to be ob stars . among these they flagged 36 probable and nine possible single - lined spectroscopic binaries ( sb1s ) as well as three new and eight candidate double - lined spectroscopic binaries ( sb2s ) .
several additional binaries have been discovered in the cyg ob2 region in the past few years .
@xcite presented orbital solutions for two sb1s and one sb2 that they had identified as sb - candidates in their earlier work .
cygnus ob2 @xmath23 , a non - thermal radio - emitter , was confirmed as a binary @xcite , supporting the theoretical wind - wind collision scenario .
two ebs in the cyg ob2 region were found via data in the nsvs survey @xcite .
@xcite determined the orbital parameters of four additional ob sbs .
this brings the total number of known ob binary systems with full photometric and/or spectroscopic solutions within cyg ob2 to 18 .
furthermore , the uncertainty regarding the age and star - formation history of cyg ob2 makes this cluster all the more intriguing .
the abundance of heavily - reddened early - type stars found by @xcite initially suggested a single epoch of star formation and a corresponding age of 13 myr .
several of the early - type candidates identified by @xcite were confirmed as ob stars by @xcite , but their dubious status as cluster members casts doubt on the 11.5 myr age that their membership would necessitate .
@xcite goes on to question the membership status of the remaining stars in the @xcite sample , and the presence of evolved o - stars such as the supergiant cyg ob2 @xmath179 requires an age of at least 3 myr .
more recently , iphas observations by @xcite revealed a clustering of @xmath3200 a - type dwarfs primarily to the south of cyg ob2 and within the 57 myr age range .
@xcite used the same method of @xcite to investigate an annular region between 1 and 2 degrees whose center lies slightly to the north of cyg ob2 .
they uncovered a diffuse spread of evolved early - type stars wtih ages up to @xmath310 myr that they identify as likely being field population stars .
finally , @xcite used isochrone fits to nir color - magnitude diagrams ( cmds ) to derive ages of 3.5 and 5.25 myr for their two fields .
given the large spread in ages , it is unclear whether the central core of cyg ob2 is merely the latest in a series of epochs of star formations , if it is the product of a parent cluster , or if the populations of older and more evolved stars are simply not cluster members .
the cygnus ob2 association is also interesting as a source of high - energy ( @xmath24mev ) emission . due to limitations of angular resolution , most observations thus far can not clearly separate cyg ob2 from nearby parts of the cygnus region more generally .
this region is bright in gamma rays from both point sources and diffuse emission , ranging from the mev to the gev to the tev @xcite .
neutrinos and gamma rays from this region will be important for testing models of cosmic ray production @xcite . with present and future instruments
, it should be possible to better resolve the young cyg ob2 , which would allow the isolation of the high - energy emission from pre - explosion massive stars as opposed to post - explosion remnants .
this paper describes the findings of the first deep optical two - band ( _ _ r__@xmath25 , _ _
i__@xmath25 ) photometric variability survey of the cygnus ob2 region .
we find 121 variables and provide variable classifications for them . in @xmath26 we give a full description of the photometric observations .
the details of the photometric calibration and light curve production are in @xmath27 .
we present the results of the variability search in @xmath28 and discuss the different variability classes separately .
we match our variability catalog to the 2mass point source catalog in order to find nir counterparts and then to a recent x - ray survey of cyg ob2 @xcite , both of which are discussed in @xmath29 .
finally , @xmath30 summarizes the results of the survey and future directions of the project .
the data were obtained on the mdm 1.3 m telescope using the 4k imager , which has a field of view ( fov ) of @xmath0 and a pixel scale of @xmath31 per pixel .
the observations consist of 60s exposures taken through the cousins _ r_-filter and 10s exposures taken through the cousins _
the data were collected over 19 nights during the fall of 2007 , spanning september 12 - october 24 , and 18 nights during the summer of 2009 , spanning may 30 - june 25 .
two pointings were used to provide greater spatial coverage of the cygnus ob2 region .
the first field was centered at ( @xmath32 ) and the second field to the north at ( @xmath33 ) , both chosen in order to cover the central core of the cluster and coincide with the study of @xcite .
table [ obs ] gives an overview of the observations .
figure [ fig : fichart ] shows a 2 @xmath20 2 deg@xmath21 image of the cygnus ob2 region with overlays of this study , the previous variability study of @xcite , and the x - ray survey of @xcite . we apply basic ccd reductions using a custom software pipeline developed by jason eastman .
the routine performs overscan subtraction , trims the images , and uses dome flats to flat field the data .
the 4k chip suffers from crosstalk contamination , a subtle but insidious effect , and so the procedure was modified to properly correct for this .
crosstalk results from the chip being divided into four quadrants of equal pixel size that are read out by four different amplifiers simultaneously . during the readout phase ,
signal from bright stars being read out on a given quadrant , referred to as the source quadrant , will reverberate through the electronics and leave residual signal on the other three quadrants , called victim quadrants .
the source signal is reflected across the three axes of symmetry of the chip , making the contamination predictable and predominantly tractable .
before the residual signal can be removed , a set of crosstalk coefficients , which specify the fraction of the source signal that is reflected onto each victim quadrant , must be calculated .
the coefficients are small enough ( @xmath34 ) that the problem can be sufficiently dealt with via a simple solution . using a set of a few hundred images of our data from both the 2007 and 2009 seasons , we find that the coefficients are not only constant in the count regime 45000 @xmath35 counts @xmath36 65535 , but also stable across the two - year baseline .
information can not be backed out of reflected signals due to saturated ( 65535 counts on the 4k imager ) sources , so pixels contaminated by saturated stars are set to zero , removed from the light curves , and otherwise ignored .
we use astrometry.net @xcite to do astrometry and we extract photometry using the isis package @xcite in conjunction with the daophot / allstar package @xcite , the procedural details of which are discussed in @xcite and @xcite
. we did not observe standard stars , so this requires that we match to other calibrated catalogs and apply a computed magnitude offset . for the _ r_-band data we match our full catalog to an overlapping region of stars observed by the iphas survey @xcite .
we uniquely match six stars from their catalog to ours using a matching radius of 1@xmath37 .
we then apply the transformation equation given by lupton et al .
( 2005 ) , @xmath38 to transform the iphas sdss magnitudes into cousins _
r_-band magnitudes . from this
we derive the magnitude offset and calibrate our _ r_-band data . to calibrate our _ i_-band data with the catalog of @xcite we use a more stringent star selection , as there is more spatial overlap between the two data sets . for a star to be used for the calibration
, it has to satisfy the following criteria : 1 ) have a unique match within 1@xmath37 , 2 ) have no additional matches within 5@xmath37 , 3 ) not be classified by them as a variable , and 4 ) not be saturated or blended in our own database .
this ultimately leaves 133 stars that we then use to calibrate our _ i_-band data .
as each of our fields has a different reference frame , there is a slight magnitude offset between fields that can be calculated using stars in the overlap region .
we perform a variability search using the vartools light curve analysis program @xcite .
we employ both analysis of variation ( aov ) @xcite and lomb - scargle ( ls ) @xcite period search algorithms .
we then inspect light curves with high significance by eye to find periodic variables and use the stetson variability index @xcite to select irregular variable stars .
figure [ fig : sigma ] shows the light curve rms as a function of reference magnitude for our 23,561 @xmath4-band light curves and 12,611 @xmath39-band light curves .
we find variables as deep as @xmath40 and @xmath41 and go fainter in @xmath4 due to the longer exposure time .
many stars with high rms are not listed as variable because we establish the criterion that stars must exhibit simultaneous two - band variability , which eliminates false positives and mitigates any potential residual effects of crosstalk contamination from saturated sources . as described in @xcite
, stars displaying variability in only one band will have a variability index near zero .
only stars that exhibited clear two - band variability are selected , 121 in total .
prior to discussing the parameters of the variables , the various naming conventions must be explained .
@xcite numbered the first 11 stars discovered in cyg ob2 , a system that @xcite expanded for star number 12 , @xcite for stars 1319 , and @xcite for stars 2085 .
we follow the convention of the literature , which is to precede these stars with schulte .
all variable stars in the gcvs are identified as such .
the primary numbering system is that of @xcite , for which the designation mt91 precedes star i d numbers , from 1799 .
the variability study of @xcite extends this system to 975 , and is referred to as mte ( mt91 extended ) .
the designations a36 and a45 ( gsc 0316100815 , 2mass j20294666 + 4105083 , respectively ) arise from the convention introduced by @xcite and followed by @xcite .
stars within our own variability catalog are given a varid , and numbers increase from brightest to faintest mean @xmath4-band magnitude .
@xcite performed the first variability survey of the region , detecting 455 stars in _ i_-band and finding 29 of them to be variable , including five ebs and five o - type stars .
our study goes much deeper , covers significantly more area , and furthermore provides a two - band photometric catalog .
we identify 121 stars that exhibit variability in both _
r_- and _ i_-band . of these , 10 are ebs , nine of them newly discovered , 17 are eb - candidates , 20 are potential herbig ae / be stars , and 52 are pulsating stars .
we first discuss the results of matching our catalog to the literature
. then we discuss each variability class individually : first the ebs and eb - candidates , then the pulsating variables , and finally the potential herbig ae / be stars .
we find 121 stars to be variable in both _
r_- and _ i_-band within the cygnus ob2 association .
after doing a literature search , we match our catalog to all previously known variable stars in cyg ob2 , listed in table [ photvar ] , using a matching radius of 1@xmath37 . of the 30 known variable stars in cyg ob2 , we recover variability in at least one band for nine of them .
seven stars are without matches while the remaining 14 are either saturated , photometrically constant , or outside our fov . we discuss those with matches here .
v2186 cyg is an eb of the algol type with a period of 2.9788 days @xcite .
though it is saturated in our _ r_-band data , we match to it in _ i_-band and successfully recover the period .
mt91 460 is an irregular variable with an _
i_-band amplitude of 0.28 mag @xcite .
we do not match to it in our _ r_-band data , but find it to exhibit a 0.7 mag drop in _
i_-band , with the overall shape of the light curve resembling that of a long period variable .
schulte 57 , varid 020 in our catalog , was previously classified as a variable of unknown type with a 0.728 day period .
we recover the period in both bands and classify it as a pulsating star .
the remainder of the stars discussed here are part of the mte system , introduced by @xcite to extend the catalog from 799 to 975 stars .
they report mte 831 to be irregular with an _
i_-band amplitude of 0.22 mag .
it matches to our varid 050 , with a 10 day period and a 0.3 mag amplitude in both _ r_- and _ i_-band .
mte 849 ( varid 021 ) is an eb of the algol type .
we report a period of 2.89 days , the first known for this system .
mte 876 ( varid 036 ) , previously a variable of unknown type , appears to have a characteristically pulsating light curve , with a period of 1.53 days and a 0.1 mag amplitude .
we agree with their classification of mte 900 ( varid 038 ) , which varies erratically with an amplitude of 1.0 mag on long timescales , perhaps indicating that it is an irregular or long - period variable .
mte 916 ( varid 011 ) similarly appears to be a long - period variable , though with a smaller 0.3 mag amplitude .
@xcite classify mte 973 as a potential eb , and we find it to be constant in _ r_-band but to have a 0.2 amplitude and a 0.88 day period in _
i_-band , though with no obvious eclipsing signature .
in addition to the eight previously known ebs within cyg ob2 , there are 13 ob sbs in cyg ob2 , four of which are single - lined and 11 of which are double - lined .
it should be noted that a few of these spectroscopic systems are also definite or possible eclipsing systems .
those that are confirmed ebs , schulte 27 and a36 , are treated here as photometric variables .
those that are unconfirmed , mt372 and schulte 3 , are included here only as sbs .
of the 121 stars we identify as photometrically variable , 10 are ebs and 17 are eb - candidates .
figure [ fig : lc.ebs ] shows sample light curves for four of the ebs and figure [ fig : lc.ebcands ] shows four of the eb - candidates . to determine if any of our variable stars are sbs , we match to all confirmed and potential sbs found in the literature , most of which are from the catalog of @xcite .
they present spectroscopic data for 146 ob stars within cyg ob2 , all initially discovered by @xcite , and find 73 new ob stars . restricting themselves to the 120 stars with the best data , they identify 36 probable and nine possible sb1s and three new and eight possible sb2s .
table [ specbin ] lists confirmed and potential sbs within cyg ob2 .
given that the catalog of @xcite reaches a limiting magnitude of @xmath42 and our upper limit is @xmath43 , the vast majority of the stars are saturated in our catalog , and thus we find no matches .
we find 52 of our stars to be pulsating variables , with @xmath4-band amplitudes ranging from 0.1 to 0.6 magnitudes .
figure [ fig : lc.pulse ] shows sample light curves of four of the pulsating variables .
the periods range from @xmath30.05 to @xmath360 days .
a histogram of the period distribution from 0 to 20 days is shown in figure [ fig : perhist ] .
initially identified by @xcite , herbig ae / be stars are pre - main sequence ( pms ) stars with masses @xmath44 and are the intermediate - mass analogs of t tauri stars . as a class their spectra display strong balmer emission lines and they additionally exhibit significant infrared excess relative to the stellar photospheric continuum , attributed to emission from circumstellar gas and dust @xcite .
herbig ae / be stars are further characterized by aperiodic large - amplitude ( up to 4 mag ) variability with minima that initially redden and subsequently undergo a color turnaroundas the star becomes bluer as it gets fainter @xcite .
they are catagorized within the class of uxors ( named after ux orion ) , since it is thought that there is no intrinsic difference between the two @xcite .
we flag 20 of our variables as potential herbig ae / be stars .
eight of these show large - amplitude ( @xmath452.0 mag ) variations while five of the remaining vary with lower amplitudes ( @xmath461.0 mag ) .
figure [ fig : herbig.amphist ] shows a histogram of @xmath4-band magnitude ranges for all 20 potential herbig ae / be stars .
the most interesting features are the dip in amplitude in the medium amplitude range ( 1.52.0 mag ) , the subsequent rise in amplitude ( 2.02.5 mag ) , and the presence of multiple ( two ) large - amplitude ( @xmath453.0 mag ) variables . considered as a whole , the amplitude distribution for these potential herbig ae / be stars in cyg ob2 is in agreement with that found by @xcite , including the three aforementioned features .
several of our candidates display variations strikingly similar to known herbig ae / be stars .
four follow the pattern of misv1147 @xcite , in which there are two distinct states . the stars exhibit a calm state with variations of small ( @xmath47 mag ) amplitudes and also experience fading episodes with deep minima of @xmath31 ( varid 041 ) , @xmath32 ( varids 087 , 100 ) , and even @xmath33 + ( varid 116 ) magnitudes . figure [ fig : herbig116 ] shows a two - band light curve of such an object .
many of the remaining candidates more closely resemble rr tau , spending roughly equal amounts of time at different magnitude levels @xcite .
figure [ fig : herbig038 ] shows an example light curve of a continuously active star .
the easiest way to confirm these as herbig ae / be stars would be to obtain spectra that display h@xmath8 emission , as well as other balmer emission lines .
spectral classification that would identify these as k0 or earlier would at the least confirm them as intermediate - mass stars , though the lack of periodic signals that we find in their light curves is consistent with the results of @xcite and perhaps indicates that these candidates similarly lack the surface hot or cool spots that are common among lower - mass pms stars and that evince themselves as periodic modulations in the light curves .
a population of herbig ae / be stars embedded within the central core of cyg ob2 would help to constrain the age of the cluster . without further spectral information , however , the true status of these stars remains unconfirmed .
cyg ob2 suffers from high and variable extinction , @xmath48 520 magnitudes @xcite , that is largely mitigated when observing in ir or nir wavelengths .
we extract all 2mass @xcite point sources within 30@xmath49 of the center of our survey and remove any with errors @xmath450.2 mag or a null value in any band ( _ j _ , _ h _ , or _ k _ ) , leaving 25,012 total sources . using a matching radius of 1@xmath37 we find unique matches to 99 ( @xmath382% ) stars in our variability catalog .
figure [ fig:2masscmdvartyp ] shows a cmd for all 2mass point sources remaining after the quality cut , with the different types of variability labeled in different shapes and colors .
the high extinction in the region can also be obviated with x - ray observations , as x - rays can penetrate extinction up to @xmath50 @xcite .
catalog of @xcite contains 1,696 x - ray sources detected within cyg ob2 , with @xmath361% of false sources .
a total of 97 of our variables lie within one of their two _ chandra _ fields , and using a matching radius of 1@xmath37 we obtain unique matches to 49 ( @xmath351% ) of those 97 stars in our variability catalog . they employ a one - dimensional kolmogorov - smirnov test to investigate x - ray variability , and of our 49 matched stars , seven display x - ray variability with three more being possibly variable . they find optical and nir counterparts for 1,501 of their sources . using a ( _ j _ , _ j__h _ ) cmd they calculate individual stellar masses by tracing extinction vectors .
figure [ fig : xray ] shows x - ray flux plotted as a function of the estimated stellar mass , with stellar variability classes overplotted in color .
it is clear that their mass estimates are reasonable for the six ob binaries , shown in purple , all of which have masses @xmath4515@xmath51 .
eight of our ebs and eb - candidates that match and have mass estimates appear to be low - mass ( @xmath36few m@xmath52 ) , while two are of intermediate ( 28@xmath51 ) mass .
the sample of our ebs and eb - candidates with mass estimates is incomplete , but such a diagnostic would be useful for constraining the binary fraction among high - mass early - type stars .
in this work we present a catalog of photometrically variable stars in the cygnus ob2 association . after generating _ r_- and _
i_-band light curves using isis , we run aov and ls period searches and apply the @xcite variability index , finding 121 stars with two - band variability , 116 of them new . of these , 99 have unique 2mass nir matches and 49 have _ chandra _ x - ray matches , with seven exhibiting x - ray variability .
we find 10 eclipsing binaries ( ebs ) , 17 eb - candidates , and 20 potential herbig ae / be stars , with most of the rest constituting an assortment of short and medium period pulsating variables .
follow - up work is underway to obtain spectra of the potential herbig ae / be stars , as h@xmath8 emission would be conclusive proof of their classification and help constrain the age of cyg ob2 .
additionally , obtaining spectral types of the ebs and eb - candidates would be useful for constraining the binary fraction of massive stars .
cyg ob2 has the largest number of spectroscopically identified o stars ( currently 65 ) as well as a high concentration of as many as 2,400 ob stars @xcite , making it an excellent target for finding massive binaries .
additionally , accurate mass determinations of these potential early - type stars could add to the total number of known ob stars in cyg ob2 and also provide evidence for or against the proposed propensity of massive binaries to form as twins , with a mass ratio near unity @xcite .
we thank jason eastman for supplying the mdm 4k photometric reduction pipeline and for helping to modify it to correct for crosstalk contamination .
we thank john beacom and alceste bonanos for their insight and stimulating discussion , matt kistler for his helpful comments , and don terndrup for his assistance with the observations .
cbh and kzs are supported in part by the nsf grant ast-0707982 .
jlp acknowledges support from nasa through hubble fellowship grant hf-51261.01-a awarded by the stsci , which is operated by aura , inc .
, for nasa , under contract nas 5 - 26555 .
ccccccc schulte 7 & irr & o3if & 10.55 & 1 & s & s + schulte 18 & irr & biib & 11.01 & 1 & s & s + schulte 27 & sb2/ew / ke & o9.5v & 12.32 & 4 & s & s + schulte 57 & unk & & & 1 & v & v + v0729 cyg & eb & o6.5 - 7 + early b ? & 9.21 & 2,3 & s & s + v2185 cyg & ea eb & o9v + b9v - a0v & 12.86 & 1 & s & c + v2186 cyg & ea eb & b0v + ? & 12.98 & 1 & s & v + v2187 cyg & @xmath53 ceph & & 15.39 & 1 & n & n + v2188 cyg & be & b1ve - b3ve & 14.88 & 1,8 & s & n + v2189 cyg & ea eb & & 16.42 & 1 & n & n + v2190 cyg & @xmath53 ceph cand & b2ve & 14.06 & 1,8 & c & c + v2191 cyg & ea eb & b4ve & 14.41 & 1,8 & c & c + mt91 512 & irr & & 12.73 & 1 & s & n + mt91 390 & irr & o8v & 12.95 & 1 & s & c + mt91 436 & unk & & 16.08 & 1 & n & n + mt91 448 & unk & o6v & 13.61 & 1 & s & n + mt91 456 & unk & & 13.53 & 1 & s & c + mt91 460 & irr & & 15.67 & 1 & n & v + mt91 503 & irr & & 15.56 & 1 & c & c + mte91 831 & irr & & & 1 & v & v + mte91 849 & ea eb & & & 1 & v & v + mte91 876 & unk & & & 1 & v & v + mte91 895 & irr / l & & & 1 & n & n + mte91 900 & irr / l & & & 1 & v & v + mte91 911 &
irr / l & & & 1 & n & n + mte91 916 & irr / l & & & 1 & v & v + mte91 973 & eb ? & & & 1 & c & v + mte91 974 & unk & & & 1 & n & n + mte91 975 & unk & & & 1 & n & n + a36 & sb2/ea & b0ib + b0iii & 11.4 & 5 - 7 & o & o + ccccccc schulte 1 & sb1 * & o8v & 11.18 & 2 & s & n + schulte 2 & sb1 * & b1i & 10.64 & 1 & s & n + schulte 3 & sb2/ea ?
& o6iv ? + o9iii & 10.35 & 3 & o & o + schulte 8a & sb2 & o5.5i + o6 ? & 8.99 & 6,7 & s & s + schulte 8c & sb1 ? & o5iii & 10.19 & 1 & s & s + schulte 9 & sb2 & o5 ? + o6 - 7 ? &
10.78 & 4 & s & s + schulte 11 & sb1 * & o5i & 10.03 & 1 & o & o + schulte 15 & sb1 * & o8v + b & 11.10 & 2 & s & s + schulte 18 & sb1 * & b1i & 11.01 & 1 & s & s + schulte 20 & sb1 * & o9iii + mid b & 11.52 & 3 & s & s + schulte 21 & sb1 * & b0ib & 11.42 & 1 & s & s + schulte 26 & sb1 * & b1iii & 11.78 & 1 & s & n + schulte 29 & sb1 ? & o7v & 11.91 & 1 & o & o + schulte 41 & sb1*/sb2 ? & b0v & 13.49 & 1 & s & c + schulte 73 & sb2 & o8iii + o8iii & 12.5 & 3 & o & o + schulte 74 & sb1 * & o8v & 12.51 & 1 & o & o + v1393 cyg & sb1 * & b0iab & 11.07 & 1 & s & s + v1827 cyg & sb2 & o7 ?
+ o9 ? & & 9,10 & o & o + mt91 021 & sb1 * & b2ii & 13.74 & 1 & o & o + mt91 138 & sb1 * & o8i & 12.26 & 1 & s & s + mt91 174 & sb1 * & b2iv & 12.55 & 1 & s & n + mt91 196 & sb1 * & b6v & 14.81 & 1 & c & n + mt91 202 & sb1 * & b2v & 14.40 & 1 & c & c + mt91 234 & sb1 * & b2v & 13.25 & 1 & s & c + mt91 238 & sb1 ? & b1v & 14.91 & 1 & n & c + mt91 241 & sb1 * & b2v & 13.41 & 1 & s & c + mt91 252 & sb2 & b2iii + b1v & 14.15 & 2 & c & c + mt91 268 & sb1 * & b2.5v & 14.38 & 1 & c & c + mt91 292 & sb1*/sb2 ? & b2v & 13.08 & 1 & s & c + mt91 298 & sb1 ? & b3v & 14.43 & 1 & c & c + mt91 311 & sb1 * & b2v & 13.87 & 1 & c & c + mt91 336 & sb1 * & b3iii & 14.13 & 1 & o & o + mt91 365 & sb2 ? & b1v & 13.81 & 1 & s & n + mt91 372 & ea*/sb1 & b0v + b2?v & 14.97 & 3,5 & o & o + mt91 403 & sb1 * & b1v & 12.94 & 1 & o & o + mt91 428 & sb1 * & b1v & 14.06 & 1 & s & c + mt91 448 & sb1 * & o6v & 13.61 & 1 & s & n + mt91 485 & sb1 ? &
o8v & 12.06 & 1 & s & c + mt91 492 & sb1 * & b1v & 14.85 & 1 & n & n + mt91 493 & sb1 * & b5iv & 14.99 & 1 & o & o + mt91 513 & sb1 * & b2v & 14.26 & 1 & s & c + mt91 517 & sb1 ? & b1v & 13.74 & 1 & s & n + mt91 561 & sb1*/sb2 ? & b2v & 13.73 & 1 & s & c + mt91 573 & sb1 ? &
b3i & 13.87 & 1 & s & n + mt91 605 & sb2 ? & b1v & 11.78 & 1 & s & c + mt91 620 & sb1 ? & b0v & 13.89 & 1 & s & n + mt91 639 & sb2 ? & b2v & 14.37 & 1 & c & c + mt91 692 & sb2 ? & b0v & 13.61 & 1 & s & n + mt91 720 & sb2 & early b + early b & 13.59 & 2 & o & o + mt91 759 & sb2 ? & b1v & 14.65 & 1 & o & o + mt91 771 & sb2 & o7v + o9v & 12.06 & 2 & o & o + a45 & sb2 & b0.5v + b2v ?- b3v ? & & 3,8 & o & o + cccccccccccc 001 & 20:31:00.18 & 41:34:30.07 & & 20310017 + 4134300 & 13.58 & 13.29 & 12.523 & 12.089 & 11.979 & o & + 002 & 20:33:40.09 & 41:08:29.65 & pul & & 13.76 & 13.39 & & & & n & + 003 & 20:33:39.61 & 41:10:19.26 & & 20333961 + 4110192 & 13.91 & 11.50 & 6.728 & 5.296 & 4.621 & n & + 004 & 20:33:42.13 & 41:22:22.77 & eb & 20334212 + 4122227 & 14.01 & 13.51 & 12.313 & 11.905 & 11.675 & o & + 005 & 20:34:04.04 & 41:14:43.04 & pul & 20340404 + 4114430 & 14.10 & 12.95 & 10.660 & 9.999 & 9.598 & o & + 006 & 20:31:42.01 & 41:21:58.29 & pul & 20314200 + 4121582 & 14.22 & 13.98 & 13.298 & 12.925 & 12.857 & o & + 007 & 20:32:19.94 & 41:33:55.08 & pul & 20321994 + 4133550 & 14.52 & 13.88 & 12.520 & 12.122 & 11.917 & y & c + 008 & 20:33:56.32 & 41:18:17.07 & pul & 20335631 + 4118170 & 14.59 & 13.88 & 12.345 & 11.928 & 11.647 & n & + 009 & 20:33:04.37 & 41:24:39.96 & pul & & 14.84 & 14.54 & & & & o & + 010 & 20:32:34.97 & 41:35:03.36 & eb ? & 20323497 + 4135033 & 14.84 & 14.10 & 12.666 & 12.177 & 11.941 & n & + 011 & 20:33:31.45 & 41:20:57.62 & pul & 20333144 + 4120576 & 15.23 & 14.32 & 12.270 & 11.424 & 10.651 & n & + 012 & 20:34:05.78 & 41:19:33.26 & & & 15.31 & 14.84 & & & & n & + 013 & 20:33:52.24 & 41:23:21.50 & pul & & 15.32 & 14.64 & & & & o & + 014 & 20:33:23.06 & 41:10:13.86 & pul & 20332306 + 4110138 & 15.41 & 14.45 & 12.378 & 11.790 & 11.475 & n & + 015 & 20:32:09.99 & 41:18:22.87 & & & 15.42 & 14.98 & & & & o & + 016 & 20:31:23.59 & 41:29:48.97 & pul & & 15.60 & 14.91 & & & & y & c + 017 & 20:31:37.32 & 41:24:38.21 & pul & 20313732 + 4124382 & 15.71 & 14.91 & 13.129 & 12.585 & 12.323 & n & + 018 & 20:33:39.86 & 41:21:45.19 & pul & 20333985 + 4121451 & 15.83 & 14.93 & 12.988 & 12.384 & 12.017 & n & + 019 & 20:31:25.74 & 41:29:15.77 & pul & 20312573 + 4129157 & 15.85 & 15.40 & 14.376 & 13.877 & 13.878 & n & + 020 & 20:33:08.73 & 41:18:51.55 & pul & & 15.94 & 15.07 & & & & n & + 021 & 20:33:08.33 & 41:15:43.26 & eb & 20330833 + 4115432 & 16.08 & 15.36 & 13.679 & 13.095 & 12.850 & y & c + 022 & 20:32:36.58 & 41:32:18.58 & eb & 20323658 + 4132185 & 16.09 & 15.05 & 13.148 & 12.533 & 12.239 & y & v + 023 & 20:32:17.79 & 41:20:00.85 & pul & 20321779 + 4120008 & 16.29 & 15.20 & 12.891 & 11.838 & 11.362 & o & + 024 & 20:32:30.88 & 41:18:10.75 & pul & 20323087 + 4118107 & 16.30 & 15.66 & 14.563 & 13.892 & 13.738 & n & + 025 & 20:32:37.86 & 41:28:52.60 & eb ? & 20323786 + 4128525 & 16.32 & 15.29 & 13.263 & 12.361 & 12.047 & y & c + 026 & 20:32:29.87 & 41:32:10.16 & pul & 20322986 + 4132101 & 16.42 & 15.24 & 12.986 & 12.010 & 11.624 & y & c + 027 & 20:33:31.78 & 41:20:47.35 & pul & 20333177 + 4120473 & 16.45 & 15.41 & 13.270 & 12.361 & 12.038 & y & c + 028 & 20:32:41.11 & 41:24:12.53 & & 20324111 + 4124125 & 16.46 & 15.79 & 14.221 & 13.557 & 13.355 & n & + 029 & 20:33:50.90 & 41:19:03.50 & pul & 20335089 + 4119034 & 16.50 & 14.02 & 9.434 & 7.713 & 6.944 & n & + 030 & 20:32:25.91 & 41:35:58.92 & pul & 20322590 + 4135589 & 16.50 & 15.41 & 13.279 & 12.282 & 11.779 & y & c + 031 & 20:33:47.13 & 41:13:16.59 & & 20334713 + 4113165 & 16.51 & 15.82 & 14.259 & 13.734 & 13.573 & y & c + 032 & 20:33:21.96 & 41:11:40.91 & pul & 20332196 + 4111409 & 16.58 & 15.43 & 13.008 & 12.204 & 11.816 & n & + 033 & 20:33:53.10 & 41:11:19.40 & eb ? & 20335310 + 4111193 & 16.59 & 15.36 & 13.823 & 13.241 & 13.068 & y & c + 034 & 20:33:48.81 & 41:23:08.26 & pul & 20334880 + 4123082 & 16.67 & 15.74 & 13.773 & 13.203 & 12.896 & o & + 035 & 20:33:15.59 & 41:21:33.58 & & & 16.75 & 15.70 & & & & y & c + 036 & 20:33:22.25 & 41:17:44.60 & pul & 20332224 + 4117446 & 16.77 & 15.55 & 13.037 & 12.005 & 11.605 & y & v + 037 & 20:32:51.37 & 41:08:40.07 & her ? & 20325137 + 4108400 & 16.84 & 15.50 & 12.492 & 10.988 & 9.706 & n & + 038 & 20:33:08.60 & 41:19:15.41 & her ? & 20330859 + 4119154 & 16.86 & 15.76 & 13.168 & 11.970 & 10.932 & y & c + 039 & 20:32:23.76 & 41:29:13.03 & eb ? & 20322376 + 4129130 & 16.86 & 15.85 & 13.843 & 13.053 & 12.757 & y & c + 040 & 20:32:30.17 & 41:34:49.27 & eb & 20323017 + 4134492 & 16.86 & 15.85 & 13.854 & 13.242 & 12.925 &
n & + 041 & 20:32:23.88 & 41:20:59.79 & her ? & 20322387 + 4120597 & 16.93 & 15.77 & 13.738 & 12.699 & 12.028 & n & + 042 & 20:32:25.06 & 41:17:21.93 & her ? & 20322506 + 4117219 & 16.95 & 15.79 & 13.090 & 12.018 & 11.311 & o & + 043 & 20:32:10.88 & 41:22:05.57 & & 20321087 + 4122055 & 16.96 & 15.80 & 13.506 & 12.597 & 12.193 & n & + 044 & 20:33:36.95 & 41:23:21.65 & eb ? & 20333694 + 4123216 & 16.97 & 15.88 & 13.691 & 12.803 & 12.395 & o & + 045 & 20:32:26.23 & 41:29:49.51 & pul & 20322622 + 4129495 & 17.05 & 16.06 & 13.995 & 13.188 & 12.913 & y & c + 046 & 20:33:18.00 & 41:10:20.67 & pul & 20331800 + 4110206 & 17.06 & 15.94 & 13.652 & 12.909 & 12.608 &
n & + 047 & 20:33:49.00 & 41:14:09.00 & eb ? & 20334899 + 4114089 & 17.07 & 15.56 & 12.634 & 11.737 & 11.163 & n & + 048 & 20:33:01.78 & 41:11:10.77 & cv & & 17.14 & 16.59 & & & & y & v + 049 & 20:32:19.92 & 41:21:11.68 & eb ?
& 20321992 + 4121116 & 17.15 & 15.98 & 13.644 & 12.677 & 12.313 & n & + 050 & 20:33:18.27 & 41:14:11.49 & & 20331826 + 4114114 & 17.16 & 15.92 & 13.323 & 11.933 & 10.816 & n & + 051 & 20:32:11.73 & 41:26:12.21 & eb ?
& & 17.21 & 16.06 & & & & y & c + 052 & 20:33:58.35 & 41:20:56.26 & & & 17.24 & 16.08 & & & & y & v + 053 & 20:31:53.13 & 41:33:31.20 & & 20315312 + 4133311 & 17.43 & 16.53 & 14.754 & 13.875 & 13.561 & y & c + 054 & 20:33:08.18 & 41:06:26.55 & pul & 20330817 + 4106265 & 17.44 & 16.12 & 13.278 & 12.118 & 11.699 & n & + 055 & 20:32:29.05 & 41:25:37.07 & pul & 20322904 + 4125370 & 17.44 & 16.31 & 14.093 & 13.185 & 12.784 & y & c + 056 & 20:32:45.02 & 41:23:49.18 & pul & & 17.48 & 16.71 & & & & n & + 057 & 20:33:16.01 & 41:17:27.26 & pul & 20331600 + 4117272 & 17.49 & 16.28 & 13.671 & 12.740 & 12.200 &
n & + 058 & 20:32:08.03 & 41:27:14.22 & eb & 20320802 + 4127142 & 17.50 & 16.53 & 14.622 & 13.878 & 13.651 & y & c + 059 & 20:32:25.76 & 41:28:42.51 & pul & 20322575 + 4128425 & 17.52 & 16.42 & 13.354 & 12.386 & 11.871 & y & c + 060 & 20:33:29.65 & 41:18:45.88 & her ?
& 20332964 + 4118458 & 17.60 & 16.39 & 13.812 & 12.300 & 11.073 & n & + 061 & 20:33:40.81 & 41:21:14.65 & eb ?
& 20334081 + 4121146 & 17.65 & 16.44 & 14.076 & 13.117 & 12.683 & y & p + 062 & 20:32:20.31 & 41:23:23.81 & pul & & 17.68 & 16.55 & & & & y & c + 063 & 20:32:25.23 & 41:24:27.65 & & 20322522 + 4124276 & 17.72 & 15.23 & 10.379 & 8.621 & 7.803 & n & + 064 & 20:32:07.49 & 41:24:35.49 & eb ? & 20320748 + 4124354 & 17.73 & 16.61 & 14.014 & 13.125 & 12.873 & y & c + 065 & 20:33:27.54 & 41:20:15.88 & pul & & 17.78 & 16.66 & & & & y & c + 066 & 20:33:20.16 & 41:19:43.28 & pul & 20332016 + 4119432 & 17.83 & 16.57 & 13.942 & 12.822 & 12.411 & y & c + 067 & 20:31:48.69 & 41:21:54.97 & eb ?
& 20314869 + 4121549 & 17.84 & 16.49 & 13.844 & 12.979 & 12.528 & o & + 068 & 20:32:01.92 & 41:31:36.74 & pul & & 17.87 & 16.82 & & & & y & c + 069 & 20:33:57.42 & 41:21:48.22 & pul & & 17.95 & 16.81 & & & & o & + 070 & 20:32:28.42 & 41:20:17.48 & her ? &
20322842 + 4120174 & 18.00 & 16.75 & 14.384 & 13.258 & 12.475 & n & + 071 & 20:32:26.34 & 41:24:09.54 & eb ? & 20322633 + 4124095 & 18.05 & 16.83 & 14.540 & 13.588 & 13.126 & n & + 072 & 20:31:22.63 & 41:28:33.76 & eb & 20312263 + 4128337 & 18.13 & 16.87 & 14.192 & 13.368 & 12.954 & n & + 073 & 20:32:31.81 & 41:27:14.99 & & & 18.24 & 17.08 & & & & y & p + 074 & 20:33:41.86 & 41:10:20.35 & pul & & 18.25 & 16.83 & & & & y & v + 075 & 20:32:37.87 & 41:21:19.40 & & 20323787 + 4121194 & 18.25 & 16.88 & 14.374 & 13.212 & 12.805 & y & c + 076 & 20:33:23.36 & 41:19:12.63 & pul & 20332336 + 4119126 & 18.38 & 17.22 & 14.807 & 13.749 & 13.355 & y & c + 077 & 20:33:04.24 & 41:06:51.43 & pul & 20330423 + 4106514 & 18.39 & 16.95 & 13.945 & 12.741 & 12.290 & n & + 078 & 20:33:57.05 & 41:06:54.89 & eb & 20335704 + 4106548 & 18.40 & 16.99 & 14.151 & 13.155 & 12.713 & o & + 079 & 20:33:28.11 & 41:21:40.26 & & 20332810 + 4121402 & 18.43 & 17.31 & 14.876 & 13.899 & 13.494 & n & + 080 & 20:31:04.79 & 41:20:26.10 & pul & 20310478 + 4120261 & 18.45 & 17.11 & 14.458 & 13.393 & 12.940 & o & + 081 & 20:33:26.34 & 41:12:38.85 & pul & 20332633 + 4112388 & 18.46 & 17.13 & 14.423 & 13.380 & 12.914 & y & c + 082 & 20:31:53.46 & 41:23:48.71 & her ? & & 18.46 & 17.19 & & & & n & + 083 & 20:32:53.77 & 41:15:13.40 & eb ? & 20325377 + 4115134 & 18.52 & 17.09 & 14.122 & 12.961 & 12.488 & y & v + 084 & 20:33:12.73 & 41:11:33.02 & & 20331273 + 4111330 & 18.55 & 17.12 & 13.983 & 12.674 & 11.989 & y & c + 085 & 20:34:08.62 & 41:06:59.75 & eb & 20340861 + 4106597 & 18.58 & 17.26 & 14.366 & 13.461 & 13.084 & o & + 086 & 20:32:31.04 & 41:18:55.09 & pul & 20323104 + 4118550 & 18.58 & 17.25 & 14.817 & 13.713 & 13.282 & n & + 087 & 20:31:53.57 & 41:25:34.16 & her ? & 20315356 + 4125341 & 18.66 & 17.67 & 14.721 & 13.440 & 12.545 & n & + 088 & 20:32:32.93 & 41:18:23.14 & & 20323292 + 4118231 & 18.70 & 17.33 & 14.617 & 13.422 & 12.737 & n & + 089 & 20:32:19.01 & 41:22:24.63 & & 20321901 + 4122246 & 18.76 & 17.48 & 15.063 & 13.948 & 13.629 & y & c + 090 & 20:33:02.33 & 41:23:45.52 & pul & 20330232 + 4123455 & 18.84 & 17.56 & 14.775 & 13.721 & 13.292 &
n & + 091 & 20:32:23.90 & 41:16:48.49 & pul & 20322389 + 4116484 & 18.85 & 17.26 & 15.002 & 14.018 & 13.821 & o & + 092 & 20:32:26.80 & 41:30:38.82 & pul & 20322680 + 4130388 & 18.88 & 17.52 & 14.875 & 13.748 & 13.455 & y & c + 093 & 20:32:33.60 & 41:21:02.56 & eb ? & 20323360 + 4121025 & 18.94 & 17.55 & 14.909 & 13.716 & 13.314 & n & + 094 & 20:31:16.89 & 41:30:01.23 & her ? & 20311688 + 4130012 & 18.94 & 17.89 & 15.651 & 14.498 & 13.924 & o & + 095 & 20:31:51.49 & 41:23:54.33 & eb & 20315149 + 4123543 & 18.97 & 17.84 & 15.243 & 14.371 & 14.091 & n & + 096 & 20:34:10.98 & 41:08:14.63 & her ? & 20341097 + 4108146 & 19.11 & 17.68 & 14.326 & 12.957 & 12.181 & o & + 097 & 20:33:04.50 & 41:24:16.25 & her ? & & 19.11 & 17.93 & & & & n & + 098 & 20:33:09.93 & 41:17:56.39 & pul & 20330992 + 4117563 & 19.15 & 17.82 & 14.827 & 13.538 & 12.793 & y & c + 099 & 20:31:59.75 & 41:27:23.16 & eb ? & 20315974 + 4127231 & 19.17 & 17.82 & 15.069 & 13.971 & 13.645 & y & c + 100 & 20:33:18.89 & 41:14:37.15 & her ? & 20331888 + 4114371 & 19.23 & 17.62 & 14.971 & 13.170 & 12.009 & y & v + 101 & 20:31:18.74 & 41:34:55.14 & pul & 20311873 + 4134551 & 19.30 & 18.08 & 15.279 & 14.208 & 13.862 & o & + 102 & 20:32:31.77 & 41:24:31.37 & pul & 20323177 + 4124313 & 19.33 & 18.04 & 15.184 & 14.084 & 13.605 & n & + 103 & 20:32:55.97 & 41:22:34.33 & pul & 20325597 + 4122343 & 19.40 & 18.11 & 15.452 & 14.283 & 13.802 & n & + 104 & 20:33:11.47 & 41:12:56.87 & pul & & 19.48 & 18.07 & & & & y & c + 105 & 20:33:55.21 & 41:20:01.19 & her ? & 20335520 + 4120011 & 19.49 & 17.97 & 14.929 & 13.659 & 12.862 & n & + 106 & 20:32:15.63 & 41:20:09.69 & her ? & 20321563 + 4120096 & 19.55 & 18.07 & 15.208 & 14.049 & 13.761 & o & + 107 & 20:31:47.13 & 41:26:36.18 & pul & 20314713 + 4126361 & 19.66 & 18.33 & 15.081 & 14.022 & 13.569 & y & c + 108 & 20:31:30.27 & 41:28:49.52 & eb ? & 20313027 + 4128495 & 19.67 & 18.47 & 15.866 & 14.850 & 14.650 & n & + 109 & 20:32:36.81 & 41:19:44.74 & pul & 20323680 + 4119447 & 19.69 & 18.13 & 15.446 & 14.233 & 13.764 & y & c + 110 & 20:33:14.35 & 41:12:21.89 & her ? & 20331434 + 4112218 & 19.71 & 18.27 & 15.390 & 13.908 & 13.076 & y & c + 111 & 20:33:11.52 & 41:07:19.73 & & 20331151 + 4107197 & 19.75 & 18.46 & 15.707 & 14.532 & 14.279 & y & c + 112 & 20:34:02.52 & 41:11:15.63 & eb ? & 20340252 + 4111156 & 19.76 & 18.28 & 15.161 & 13.884 & 13.366 & o & + 113 & 20:33:08.69 & 41:21:10.25 & her ? & 20330868 + 4121102 & 19.89 & 18.47 & 15.994 & 14.376 & 13.409 & n & + 114 & 20:31:56.79 & 41:21:53.64 & pul & & 19.96 & 18.44 & & & & o & + 115 & 20:32:22.24 & 41:22:45.45 & eb & 20322224 + 4122454 & 19.99 & 18.72 & 16.243 & 15.127 & 15.069 & n & + 116 & 20:31:52.41 & 41:26:10.28 & her ? & 20315240 + 4126102 & 20.00 & 18.88 & 16.340 & 15.044 & 14.408 & y & c + 117 & 20:32:57.47 & 41:04:55.19 & & 20325746 + 4104551 & 20.02 & 18.57 & 15.295 & 14.060 & 13.642 & o & + 118 & 20:33:12.84 & 41:14:40.91 & her ? & 20331283 + 4114409 & 20.11 & 18.53 & 15.378 & 13.799 & 13.104 & n & + 119 & 20:33:21.85 & 41:09:44.55 & her ? & 20332184 + 4109445 & 20.14 & 18.54 & 15.144 & 13.682 & 12.979 & y & c + 120 & 20:32:25.03 & 41:25:38.52 & her ? & 20322503 + 4125385 & 20.31 & 18.87 & 15.877 & 14.738 & 14.363 & y & c + 121 & 20:33:40.93 & 41:09:36.29 & & 20334092 + 4109362 & 21.14 & 19.50 & 16.128 & 14.904 & 14.239 & y & p + , a. a. , allen , b. , berley , d. , blaufuss , e. , casanova , s. , chen , c. , coyne , d. g. , delay , r. s. , dingus , b. l. , ellsworth , r. w. , fleysher , l. , fleysher , r. , gebauer , i. , gonzalez , m. m. , goodman , j. a. , hays , e. , hoffman , c. m. , kolterman , b. e. , kelley , l. a. , lansdell , c. p. , linnemann , j. t. , mcenery , j. e. , mincer , a. i. , moskalenko , i. v. , nemethy , p. , noyes , d. , ryan , j. m. , samuelson , f. w. , saz parkinson , p. m. , schneider , m. , shoup , a. , sinnis , g. , smith , a. j. , strong , a. w. , sullivan , g. w. , vasileiou , v. , walker , g. p. , williams , d. a. , xu , x. w. , & yodh , g. b. 2007 , , 658 , l33 , j. , aliu , e. , anderhub , h. , antoranz , p. , baixeras , c. , barrio , j. a. , bartko , h. , bastieri , d. , becker , j. k. , bednarek , w. , berger , k. , bigongiari , c. , biland , a. , bock , r. k. , bonnoli , g. , bordas , p. , bosch - ramon , v. , bretz , t. , britvitch , i. , camara , m. , carmona , e. , chilingarian , a. , commichau , s. , contreras , j. l. , cortina , j. , costado , m. t. , curtef , v. , dazzi , f. , de angelis , a. , delgado , c. , de los reyes , r. , domingo - santamara , e. , de lotto , b. , de maria , m. , de sabata , f. , dorner , d. , doro , m. , errando , m. , fagiolini , m. , ferenc , d. , fernndez , e. , firpo , r. , fonseca , m. v. , font , l. , galante , n. , garca - lpez , r. j. , garczarczyk , m. , gaug , m. , goebel , f. , hayashida , m. , herrero , a. , hhne , d. , hose , j. , hsu , c. c. , huber , s. , jogler , t. , kosyra , r. , kranich , d. , laille , a. , leonardo , e. , lindfors , e. , lombardi , s. , longo , f. , lpez , m. , lorenz , e. , majumdar , p. , maneva , g. , mankuzhiyil , n. , mannheim , k. , mariotti , m. , martnez , m. , mazin , d. , merck , c. , meucci , m. , meyer , m. , miranda , j. m. , mirzoyan , r. , mizobuchi , s. , moralejo , a. , nieto , d. , nilsson , k. , ninkovic , j. , oa - wilhelmi , e. , otte , n. , oya , i. , panniello , m. , paoletti , r. , paredes , j. m. , pasanen , m. , pascoli , d. , pauss , f. , pegna , r. , persic , m. , peruzzo , l. , piccioli , a. , prandini , e. , puchades , n. , raymers , a. , rhode , w. , rib , m. , rico , j. , rissi , m. , robert , a. , rgamer , s. , saggion , a. , saito , t. y. , snchez , a. , sartori , p. , scalzotto , v. , scapin , v. , schmitt , r. , schweizer , t. , shayduk , m. , shinozaki , k. , shore , s. n. , sidro , n. , sillanp , a. , sobczynska , d. , spanier , f. , stamerra , a. , stark , l. s. , takalo , l. , temnikov , p. , tescaro , d. , teshima , m. , torres , d. f. , turini , n. , vankov , h. , venturini , a. , vitale , v. , wagner , r. m. , wittek , w. , zandanel , f. , zanin , r. , & zapatero , j. 2008 , , 675 , l25 , j. e. , greimel , r. , irwin , m. j. , aungwerojwit , a. , barlow , m. j. , corradi , r. l. m. , drake , j. j. , gnsicke , b. t. , groot , p. , hales , a. , hopewell , e. c. , irwin , j. , knigge , c. , leisy , p. , lennon , d. j. , mampaso , a. , masheder , m. r. w. , matsuura , m. , morales - rueda , l. , morris , r. a. h. , parker , q. a. , phillipps , s. , rodriguez - gil , p. , roelofs , g. , skillen , i. , sokoloski , j. l. , steeghs , d. , unruh , y. c. , viironen , k. , vink , j. s. , walton , n. a. , witham , a. , wright , n. , zijlstra , a. a. , & zurita , a. 2005 , , 362 , 753 , n. , feigelson , e. d. , getman , k. v. , townsley , l. , broos , p. , flaccomio , e. , mccaughrean , m. j. , micela , g. , sciortino , s. , bally , j. , smith , n. , muench , a. a. , garmire , g. p. , & palla , f. 2005 , , 160 , 530 , r. c. , bertsch , d. l. , bloom , s. d. , chen , a. w. , deines - jones , p. , esposito , j. a. , fichtel , c. e. , friedlander , d. p. , hunter , s. d. , mcdonald , l. m. , sreekumar , p. , thompson , d. j. , jones , b. b. , lin , y. c. , michelson , p. f. , nolan , p. l. , tompkins , w. f. , kanbach , g. , mayer - hasselwander , h. a. , mcke , a. , pohl , m. , reimer , o. , kniffen , d. a. , schneid , e. j. , von montigny , c. , mukherjee , r. , & dingus , b. l. 1999 , , 123 , 79 , m. f. , cutri , r. m. , stiening , r. , weinberg , m. d. , schneider , s. , carpenter , j. m. , beichman , c. , capps , r. , chester , t. , elias , j. , huchra , j. , liebert , j. , lonsdale , c. , monet , d. g. , price , s. , seitzer , p. , jarrett , t. , kirkpatrick , j. d. , gizis , j. e. , howard , e. , evans , t. , fowler , j. , fullmer , l. , hurt , r. , light , r. , kopan , e. l. , marsh , k. a. , mccallon , h. l. , tam , r. , van dyk , s. , & wheelock , s. 2006 , , 131 , 1163 , m. , kato , t. , ishioka , r. , yoshida , s. , kadota , k. , ohkura , n. , henden , a. , pejcha , o. , kinugasa , k. , fujii , m. , simonsen , m. , greaves , j. , dubovsky , p. a. , poyner , g. , west , d. , stine , r. j. , taylor , d. , poxon , m. , muyllaert , e. , ripero , j. , reszelski , m. , & jones , c. p. 2004 , , 56 , 183 , p. r. , vestrand , w. t. , akerlof , c. w. , balsano , r. , bloch , j. , casperson , d. , fletcher , s. , gisler , g. , kehoe , r. , kinemuchi , k. , lee , b. c. , marshall , s. , mcgowan , k. e. , mckay , t. a. , rykoff , e. s. , smith , d. a. , szymanski , j. , & wren , j. 2004 , , 127 , 2436 | we present a catalog of photometrically variable stars discovered within two @xmath0 fields centered on the cygnus ob2 association .
there have hitherto been no deep optical variability studies of cyg ob2 despite it being replete with early - type massive stars , perhaps due to the high and variable extinction ( up to @xmath1 @xmath2 20 ) that permeates much of the region .
here we provide results of the first variability study with this combination of spatial coverage ( @xmath3 0.5 deg ) and photometric depth ( @xmath4 @xmath3 21 mag ) .
we find 121 stars to be variable in both _
r_- and _ i_-band , 116 of them newly discovered . of the 121 variables , we identify 27 eclipsing binaries ( ebs ) and eclipsing binary candidates , 20 potential herbig ae / be stars , and 52 pulsating variables .
confirming both the status and the cluster membership of the herbig ae / be stars would address the uncertainty regarding the age and star formation history of cyg ob2 .
we match our catalog to known variables and binaries in the region , 2mass near - ir ( nir ) data , and _
chandra _ x - ray observations to find counterparts to new variables in other wavelengths . |
the standard model ( sm ) of the elementary particles is now undergoing the greatest challenges ever by the successful running of the large hadron collider ( lhc ) at cern .
the last missing piece of the sm , the higgs boson , would be discovered or ruled out sooner or later and new physics signals beyond the sm are expected to appear for the first time . among the most striking events that the lhc would reveal
is the formation of mini black holes
. if there are extra dimensions @xcite the fundamental energy scale for strong gravitation @xmath0 in @xmath1-spacetime dimensions can be much smaller than the planck mass . in the case of @xmath2 tev
, the lhc experiments can probe the strong gravity regime and possibly produce black holes @xcite .
typical size of the schwarzschild radius of these black holes is @xmath3 fm and they are called as mini black holes .
mini black holes also evaporate through hawking radiation @xcite just as ordinary black holes .
the lifetime of a mini black hole is typically about @xmath4 so the black hole decays instantaneously .
once the black hole is produced at the lhc one can detect its creation by observing hawking radiation .
it is also possible that rotating or charged black holes would be produced . for these reasons
there have been many studies on the emission of particles by mini black holes , both static and rotating ones in extra dimensions @xcite .
when particles are scattered off by rotating ( or kerr ) black holes a very interesting phenomenon known as superradiance occurs .
if a field of the form @xmath5 is incident on a rotating black hole of angular velocity @xmath6 , the field is amplified if @xmath7 . in this case
the rotation energy is transferred to the scattered field . for
any rotating objects the superradiance can occur .
if the rotating black hole is surrounded by a mirror which reflects the scattered field back to the black hole , the field bounces back and forth between the black hole and the mirror . during the process the field amplifies itself and the extracted energy from the rotating black hole grows exponentially .
this is so called the press - teukolsky black - hole bomb @xcite .
the origin of the black - hole bomb is the small imaginary part of @xmath8 . in the case of superradiance where @xmath7 , the imaginary part is positive
so the field contains the term of exponential increasing with time , @xmath9 . the mirror which reflects the field needs not be a real one .
for example , a mass term of the field in 4-dimensional spacetime can play a role of the mirror . in this case
the massive field forms a bound state with the black hole , and the superradiance results in the instability of the black hole .
there have been many works on the black - hole bomb where a massive particle is bound to a celestial black hole under superradiance @xcite . for the higher dimensional kerr black holes
, it was shown that they do not exhibit such instabilities for massive fields because there are no bound geodesics in the spacetime around the black hole @xcite .
in a recent work @xcite i have investigated the possibility of black - hole bombs at the lhc .
the work is a higher dimensional version of @xcite . in @xcite
the superradiance of a massive scalar field by a rotating black hole in higher dimensions with the mirror boundary condition is analyzed .
as shown in @xcite the mass term would not be enough for the role of mirror and probably electromagnetic fields might work out for charged particles , but to focus on the possibility of forming a black - hole bomb , the mirror boundary condition was simply imposed without considering any technical concerns about it .
it was assumed that the mirror boundary condition is simply given by whatever reflects the scattered particle back to the black hole .
current work is an extension of @xcite . in this work we analyzed the bulk emission as well as the brane emission .
intuitively the bulk emission modes are expected to be suppressed compared with the brane emission , and we will see that this is really the case by a factor of @xmath10 .
we also investigate the different angular modes for different masses . in the next section the superradiance of massive scalar particles by rotating black holes in higher dimensions with the mirror boundary condition
is described .
both of the bulk emission and the brane emission are considered . in sec .
iii the results are summarized and discussions are given .
the conclusions appear in sec .
in this section , we describe the rotating black holes in higher @xmath11 dimensions just as in the way of @xcite .
the procedure is much the same as that of @xcite .
we assume that the angular momentum lies parallel to our brane .
a proper background geometry is the myers - perry solution @xcite @xmath12 where @xmath13 and @xmath14 is a unit @xmath15-sphere .
the mass parameter @xmath16 is related to the black hole mass @xmath17 and the new fundamental scale @xmath0 in @xmath18 dimensions via @xmath19 and @xmath20 is proportional to the black hole angular momentum @xmath21 as @xmath22 the radius of the black hole horizon @xmath23 is determined by @xmath24 which makes @xmath25 , and @xmath26 .
a massive scalar field @xmath27 of mass @xmath28 under the gravitational background ( [ mp ] ) satisfies the klein - gordon equation in curved spacetime @xmath29 where @xmath30 is the @xmath31-dimensional metric tensor of ( [ mp ] ) and @xmath32 is its determinant .
the equation can be solved by the separation of variables for @xmath27 as @xmath33 where @xmath34 are the hyperspherical harmonics on the @xmath15-sphere .
the radial function @xmath35 satisfies @xmath36 while the angular function @xmath37 satisfies @xmath38 where @xmath39 and @xmath40 is expanded around small @xmath41 as @xmath42 to investigate the superradiance regime , it is enough to concentrate on the radial function @xmath35 since the absorption and reflection probabilities of scattering are determined by @xmath35 . to solve the radial equation ,
first consider the near - horizon region where @xmath43 .
after changing the variable as @xmath44 , the radial equation becomes @xmath45r=0~ , \label{rf}\ ] ] where @xmath46(1+a_*^2)~,\\ d_*&=&1-\frac{4a_*^2}{a_*^2}~.\end{aligned}\ ] ] the solution for @xmath47 in eq .
( [ rf ] ) is proportional to the hypergeometric function @xmath48 as @xmath49 where @xmath50~,\end{aligned}\ ] ] with @xmath51 and @xmath52 is the integration constant . in the far - field region where @xmath53
, the radial equation becomes @xmath54\tr=0~,\ ] ] where @xmath55 and @xmath56 , and the solution is the bessel function , @xmath57 here @xmath58 and @xmath59 are the first and second kind of the bessel functions , where @xmath60 matching the two solutions @xmath61 and @xmath62 gives the ratio of @xmath63 to @xmath64 as @xcite @xmath65^{2j+n+1 } \frac{1}{\nu\gamma^2(\nu)}\nn\\ & & \times \frac{\gamma(2\beta+d_*-2)\gamma(2+\alpha-\beta - d_*)\gamma(1+\alpha-\beta ) } { \gamma(\alpha+\beta+d_*-1)\gamma(\alpha+\beta)\gamma(2 - 2\beta - d_*)}~. \label{match}\end{aligned}\ ] ] the results up to now hold for any scalar particles embedded in the geometry of ( [ mp ] ) through eq .
( [ kg ] ) .
our concern is the system surrounded by a mirror which reflects the scalar particle back to the black hole .
if the mirror is located at @xmath66 , then the mirror boundary condition is @xmath67 , which means @xcite @xmath68 combining eqs .
( [ match ] ) and ( [ bc ] ) gives @xmath69^{2j+n+1 } \frac{1}{\nu\gamma^2(\nu)}\nn\\ & & \times \frac{\gamma(2\beta+d_*-2)\gamma(2+\alpha-\beta - d_*)\gamma(1+\alpha-\beta ) } { \gamma(\alpha+\beta+d_*-1)\gamma(\alpha+\beta)\gamma(2 - 2\beta - d_*)}~. \label{bc2}\end{aligned}\ ] ] in the limit of @xmath70 the right - hand - side of eq .
( [ bc2 ] ) is @xmath71 so that @xmath72 in this limit . to a good approximation ,
the value of @xmath73 is equal to the zeros of the bessel function @xmath74 , @xmath75 where @xmath76 .
assume that @xmath77 , the solution of eq .
( [ bc2 ] ) now can be written as @xmath78 where @xmath79 is introduced as a small imaginary part of @xmath80 . for small @xmath81 , @xmath82 , and
one can extract the expression for @xmath83 : @xmath84^{2j+n+1 } \frac{1}{\nu\gamma^2(\nu)}\nn\\ & & \times \frac{\gamma(2\beta+d_*-2)\gamma(2+\alpha-\beta - d_*)\gamma(1+\alpha-\beta ) } { \gamma(\alpha+\beta+d_*-1)\gamma(\alpha+\beta)\gamma(2 - 2\beta - d_*)}~. \label{td}\end{aligned}\ ] ] on the right - hand - side of the above equation the imaginary part comes from the terms containing @xmath85 .
the right - hand - side also has a nonzero real part .
it can be considered as a correction to eq .
( [ besselzero ] ) , and for the brane emission the size of it is typical around 1% . for the bulk emission we found
that it is much more suppressed because of the factor @xmath86 , so it will be neglected . using the properties of the gamma functions ,
one arrives at the final result @xmath87^{2j+n+1 } \frac{1}{2}\sinh\pi\left(\frac{2k_*}{a_*}\right)\nn\\ & & \times \frac{\left[\beta^2+\frac{k_*^2}{a_*^2}\right]\left|\gamma\left(-\beta+i\frac{k_*}{a_*}\right)\right|^2 \left[(1-\beta - d_*)^2+\frac{k_*^2}{a_*^2}\right]\left|\gamma\left(1-\beta - d_*+i\frac{k_*}{a_*}\right)\right|^2 } { \nu\gamma^2(\nu)(2 - 2\beta - d_*)\gamma^2(2 - 2\beta - d_*)}~.\nn\\\end{aligned}\ ] ] the prefactor of @xmath88 is negative for our relevant range of @xmath8 , so the sign of @xmath81 is opposite to that of @xmath89 .
( especially , for small @xmath90 , @xmath91 . )
more specifically , @xmath92 for @xmath93 , which is the superradiant condition .
for the brane emission , one can follow the procedure of @xcite and use its results .
it is not difficult to see that the energy emission rate of the black hole @xmath94 is proportional to @xmath95 @xcite .
thus a larger value of @xmath95 means the stronger instability of the black hole via superradiance .
in our analysis we fix @xmath96 tev and @xmath97 tev . the relevant range of @xmath8 for the superradiance is @xmath98 since @xmath99 from the mirror boundary condition ( [ besselzero ] ) , one has @xmath100 we only consider the case of @xmath101 for simplicity .
the size of the horizon radius @xmath23 is @xmath102 for @xmath103 and @xmath104 . as for @xmath105 ,
a dimensionless parameter @xmath106 is almost @xmath15-independent for the brane emission with sufficiently small @xmath28 .
but in the case of bulk emission , the parameter @xmath107 itself is @xmath15-dependent .
figure 1 shows @xmath95 as a function of @xmath108 in the bulk emission for @xmath109 , @xmath110 gev and for the modes @xmath111 , and the details of information are summarized in tables [ bulk101pi]-[bulk202120 ] . in these tables
, @xmath112 is the value of @xmath8 which makes @xmath95 maximum , so @xmath113 , and @xmath114 is the @xmath115-folding time .
note that @xmath116 gets larger for smaller @xmath15 and for larger @xmath117 ( but when @xmath118 one should be cautious ; see below . ) , and for lighter mass @xmath28 .
the suppression of @xmath116 is severe for large @xmath15 because @xmath119 .
also @xmath120 generally , with the only exception of @xmath121 .
but it must be noted that for @xmath122 mode with @xmath118 , the value of @xmath123 for @xmath124 .
( @xmath125 itself is @xmath126 in this case . ) for the region @xmath127 the series of eq .
( [ te ] ) is not a good expansion so we can not rely on the validity of the whole analysis . for @xmath118 ,
the analysis is reliable when @xmath128 . in this case
the maximum value of @xmath95 is obtained for @xmath129 . in tables
[ bulk202pi ] and [ bulk202120 ] all the parameters for @xmath118 are calculated for this region .
for higher modes of @xmath130 , the value of @xmath131 exceeds unity even for @xmath132 , so we do not consider the higher modes . as a comparison
, we provide the results for the emission `` on the brane '' in fig .
[ brane ] and tables [ brane11120]-[brane22120 ] .
( for the details of the brane emission , see @xcite ) . in general
, @xmath116 on the brane is much larger ( about @xmath133 ) than that in the bulk , since in the case of brane emission @xmath134 . as in the bulk emission , @xmath116 gets larger for smaller @xmath15 and for larger @xmath117 , and for lighter mass @xmath28 .
one exception is that @xmath135 for @xmath109 .
one different feature of brane emission is that @xmath136 mode for @xmath110 gev ( one exception is @xmath137 ) while @xmath138 for @xmath109 . in the brane emission ,
one encounters the same problem when @xmath118 for @xmath139 , as in the bulk emission .
the value of @xmath8 should be restricted as @xmath128 , and the maximum value of @xmath95 is given when @xmath129 .
see tables [ brane22pi ] and [ brane22120 ] .
higher @xmath140 modes are not considered as before . in the current analysis @xmath95
tends to decrease for heavier @xmath28 .
it is because it would be more difficult to amplify heavier particles .
this feature is also verified in @xcite . in the case of 4-dimensional scattering ,
the presence of the mass term makes it possible to form a bound state of the black hole and the scattered particle , only when @xmath141 . in this case
the instability of the black hole gets stronger as @xmath28 increases because the depth of the effective potential well becomes larger , as pointed out in @xcite .
but in the current work we only consider the region of @xmath142 to make use of the expansion like eq .
( [ te ] ) , with the assumption of the mirror boundary condition .
thus we can not expect a smiliar behavior to that of @xcite .
for all the cases considered in this paper , the largest value of @xmath116 is @xmath143 gev ( brane emission of @xmath144 for @xmath118 and @xmath110 gev ) , which is smaller than the largest @xmath116 of @xcite , @xmath145 gev ( brane emission of @xmath144 for @xmath118 and @xmath109 gev ) . and in both cases of bulk emission and brane emission , one can easily distinguish the modes @xmath146 from @xmath147 , or @xmath144 from @xmath148 , since for @xmath149 or @xmath150 modes @xmath151 while for @xmath147 or @xmath148 modes @xmath152 .
actually @xmath151 for @xmath149 or @xmath150 modes is obvious because @xmath153 from eq .
( [ wmax ] ) .
the @xmath115-folding time @xmath114 is a quantity such that for every @xmath154 for @xmath155 the field amplitude is amplified by @xmath115-times and the energy gets larger by an amount of @xmath156 .
kerr instability is comparable to hawking radiation when @xmath115-folding time is shorter than the hawking evaporation lifetime @xcite . for the bulk emission @xmath157 while @xmath158 for the brane emission .
these values are quite large compared with the typical mini black hole lifetime @xcite @xmath159 . for this value of the black hole lifetime , the amplification by the superradiance is very small , @xmath160 at best .
but it is not impossible that @xmath161 could be much longer , say , @xmath162 @xcite . even for the bulk emission case the scattered particles can experience large amplifications if the black hole lifetime is long enough .
thus there is a chance to observe the black hole bombs at the lhc near future . as in the brane emission case
we can estimated the relative energy extracted from the black hole when the mirror is put near its minimum value @xcite . from the first law of thermodynamics of black holes ,
one has @xmath163 as the scattering process proceeds , the black hole loses its rotational energy and @xmath117 gets smaller .
note that @xmath105 is minimum at @xmath132 ( see eqs .
( [ wmax ] ) and([r0min ] ) ) and becomes larger for smaller @xmath117 .
for example , consider the bulk emission of mode @xmath122 with @xmath109 and @xmath164 .
see the table [ bulk202pi ] . in this case @xmath165
, so put the mirror at @xmath166 at the beginning . during the scattering @xmath117 decreases while @xmath105 increases , and eventually there is a point where @xmath167 . if @xmath105 exceeds the mirror location , @xmath166 , the superradiance stops .
this happens when @xmath168 where @xmath169 .
for this process the energy fraction transferred from the black hole is @xmath170 for @xmath171 from eq .
( [ deltam ] ) .
this is a half value of the fraction for the brane emission case considered in @xcite .
similar calculation shows that for the case of @xmath146 with @xmath109 , @xmath164 in the bulk emission , @xmath172 .
if one puts the mirror far enough from @xmath105 , one can extract much more energy from the black hole @xcite .
in this paper the superradiant scattering by rotating black holes in higher dimensions with mirror boundary conditions ( the so called _ black - hole bomb _ ) is investigated .
the work is an extension of the previous one @xcite in that in this version the bulk emission case is also analyzed , and other modes of @xmath130 or @xmath140 are compared , and different mass effects are considered . among various cases in the analyses we found that the largest value of @xmath116 is obtained for the brane emission with @xmath144 for smaller scalar mass and larger angular momentum of the black hole . in the bulk emission @xmath116
is suppressed by a factor of @xmath173 or more . for the brane emission
the shortest @xmath115-folding time is about @xmath174 which is much longer than the typical lifetime of mini black holes .
but it is possible that the long - lived mini black holes might exist and could provide a chance for forming the black - hole bombs .
any technical details on how to realize the mirror to bounce off the scattered particles back to the black hole are beyond the scope of this work and are not considered here .
this work was supported by the basic science research program through the national research foundation of korea ( nrf ) funded by the korean ministry of education , science and technology ( 2009 - 00888396 ) .
99 n. arkani - hamed , s. dimopoulos and g. r. dvali , phys .
b * 429 * , 263 ( 1998 ) ; l. randall and r. sundrum , phys . rev . lett . * 83 * , 3370 ( 1999 ) .
t. banks and w. fischler , arxiv : hep - th/9906038 ; s. b. giddings and s. d. thomas , phys .
d * 65 * , 056010 ( 2002 ) ; s. dimopoulos and g. l. landsberg , phys . rev .
lett . * 87 * , 161602 ( 2001 ) .
s. w. hawking , commun . math .
phys . *
43 * , 199 ( 1975 ) [ erratum - ibid . *
46 * , 206 ( 1976 ) ] .
v. p. frolov and d. stojkovic , phys .
d * 67 * , 084004 ( 2003 ) .
d. ida , k. y. oda and s. c. park , phys .
d * 67 * , 064025 ( 2003 ) [ erratum - ibid .
d * 69 * , 049901 ( 2004 ) ] . c. m. harris and p. kanti , phys .
b * 633 * , 106 ( 2006 ) .
p. kanti and n. pappas , phys .
d * 82 * , 024039 ( 2010 ) .
w. h. press and s. a. teukolsky , nature * 238 * , 211 ( 1972 ) .
v. cardoso , o. j. c. dias , j. p. s. lemos and s. yoshida , phys .
d * 70 * , 044039 ( 2004 ) [ erratum - ibid .
d * 70 * , 049903 ( 2004 ) ] .
s. r. dolan , phys .
d * 76 * , 084001 ( 2007 ) .
s. hod and o. hod , phys .
d * 81 * , 061502 ( 2010 ) .
j. g. rosa , jhep * 1006 * , 015 ( 2010 ) .
v. cardoso and s. yoshida , jhep * 0507 * , 009 ( 2005 ) .
lee , arxiv:1104.0496 [ hep - ph ] .
r. c. myers and m. j. perry , annals phys .
* 172 * , 304 ( 1986 ) .
p. kanti , lect .
notes phys .
* 769 * , 387 ( 2009 ) .
p. c. argyres , s. dimopoulos and j. march - russell , phys .
b * 441 * , 96 ( 1998 ) .
r. casadio and b. harms , entropy * 13 * , 502 ( 2011 ) ; int . j. mod
. phys .
a * 17 * , 4635 ( 2002 ) . | the superradiant scattering of massive scalar particles by a rotating mini black hole is investigated . imposing the mirror boundary condition ,
the system becomes the so called black - hole bomb where the rotation energy of the black hole is transferred to the scattered particle exponentially with time .
bulk emissions as well as brane emissions are considered altogether .
it is found that the largest effects are expected for the brane emission of lower angular modes with lighter mass and larger angular momentum of the black hole .
possibilities of the forming the black - hole bomb at the lhc are discussed . |
galaxy clusters , the largest mass concentrations in the universe , have long played significant roles in the study of galaxy evolution and the determination of cosmological parameters . in both cases ,
the leverage and advantages offered by having a sample with a large redshift grasp are significant . in the case of cluster evolution , at the redshift of @xmath0 we are approaching the formation epoch of clusters , often put at @xmath10 to 5 , so that evolutionary effects can be more easily discerned . for the determination of cosmological parameters , a large baseline of redshift
allows us to measure the evolution of the mass function , breaking the degeneracy between @xmath2 and @xmath3 .
the measurement of the mass function at @xmath8 of a sufficiently large sample will also provide a chance to constrain @xmath9 .
the red - sequence method ( gladders & yee 2000 ) was developed with the specific aim to create large , robust galaxy cluster samples efficiently out to @xmath11 using optical data . this technique uses the red - sequence ( early - type ) galaxies as markers for clusters , searching for enhanced galaxy surface densities in successive cuts in the color - magnitude plane , which act to provide a discrimination in the third dimension , mitigating the projection contamination problem inherent in searching for high redshift galaxy clusters .
this technique is inexpensive , as it requires photometry from only two filters , and can be used to map large regions of the sky efficiently . in this paper
we describe the application of the first red - sequence cluster survey ( rcs-1 ) to the determination of @xmath2 and @xmath3 , and the on - going 1000 square degree rcs-2 survey , with the goal of measuring @xmath9 .
the creation of these large cluster samples to @xmath8 also enables us to carry out many other studies , some of which are presented in these proceedings , including an analysis of the butcher - oemler effect using over 1000 clusters ( ellingson et al . , these proceedings ) , and the discovery of a large number of strong gravitational lenses ( gilbank et al . , these proceedings ) .
the rcs-1 provides for the first time a cluster sample which is large , well - characterized , and covering a sufficiently large redshift range to measure @xmath2 and @xmath3 directly using cluster abundance .
the details are reported in gladders et al .
( 2007 ) ; we present a brief description of the results here .
the rcs-1 survey contains a total of @xmath12 sq .
deg of imaging data obtained using the cfht-12k camera and the mosaic - ii camera on the ctio 4 m with coverage in the @xmath13 and @xmath4 bands ( see gladders & yee 2005 for details ) .
both cameras provide a field of view of @xmath141/3 sq .
we select a ` best ' sample using 72 sq .
deg based on depth and photometric calibration accuracy .
the sample contains 956 clusters in the redshift range of @xmath15 .
we use a somewhat restricted redshift range which provides the most robust red - sequence photo-@xmath16 and cluster richness estimates .
we use the cluster richness parameter @xmath17 ( see gladders & yee 2005 ) , computed using the net galaxy count in the red - sequence , as a mass proxy .
cluster richness , or other similar measurements such as cluster light , has been shown to be correlated with mass , @xmath18 , and @xmath19 in a number of investigations ( e.g. , yee & ellingson 2003 ; hansen et al .
we link the observable to mass ( @xmath20 ) using a power - law @xmath21 , where @xmath22 allows for possible evolution in the relationship .
we use the self - calibration analysis technique suggested by @xcite , in which the parameters of the mass - observable relation are simultaneously fitted with the cosmological parameters .
the cosmological parameters are estimated by fitting the number of clusters at different redshifts ( @xmath23 ) to a mass limit @xmath24 to that expected from the jenkins et al .
( 2001 ) mass function , using a markov - chain monte - carlo ( mcmc ) method .
we effectively fit two cosmological parameters ( @xmath2 and @xmath3 ) and three cluster parameters ( the mass limit as a combination of the @xmath25 and @xmath26 parameters , @xmath22 , and the fractional scatter @xmath27 in the mass-@xmath17 relation ) .
we use wmap priors for @xmath28 ( @xmath29 ) , @xmath30 ( 0.046 , fixed ) , and @xmath31 ( 0.99 , fixed ) .
the results are shown in figure 1 .
we obtain @xmath32 and @xmath33 , which are in excellent agreement with those from the year-3 wmap results @xcite . an equally interesting result in our analysis
is the comparison between the cluster parameters derived from the self - calibration cosmology fit to those measured explicitly from cluster samples , specifically those obtained from the cnoc1 survey ( yee & ellingson 2003 ) and a sample of 33 @xmath34 rcs-1 clusters with dynamical mass measurements ( blindert et al .
we perform this comparison of the @xmath35-mass relation by fitting with 4 cluster parameters , replacing @xmath36 with @xmath25 and @xmath26 .
we find that the richness to mass relation as derived from the cosmology fit produces entirely consistent results to those from direct measurements of the relation .
we obtain @xmath25 and @xmath26 values of 10.55@xmath37 and @xmath38 , respectively , compared to @xmath39 and @xmath40 using the cnoc1 sample ( yee & ellingson 2003 ) .
similar results are also obtained by blindert et al .
( 2007 ) from the rcs-1 spectroscopic sample .
furthermore , the best fit scatter @xmath41 is also in excellent agreement with that determined from the blindert et al .
sample of @xmath42 .
the fact that the fitting results for both the cosmological parameters and the cluster parameters are in excellent agreement with other completely independent measurements is a strong endorsement of both the use of cluster mass function , including that derived from an optical sample such as the rcs , as a cosmological probe , and the self - calibration methodology .
the rcs-2 is the first large scale imaging survey with sufficient area and depth for constraining @xmath9 using cluster abundance .
the recent commissioning of the megaprime camera ( megacam ) on cfht provides the field of view on a sufficiently large telescope required to carry out such a survey for the first time .
the megacam is a one - square - degree camera consisting of a mosaic of thirty - six 4612@xmath432048 pixels , with a scale of 0.18@xmath44/pix .
the survey will image 820 sq .
deg using @xmath4 , @xmath5 , and @xmath6 filters of relatively shallow depths with integration times of 6 , 8 and 4 minutes , attaining @xmath45 magnitudes of 22.5 , 24.8 , and 25.3 , respectively . a total coverage of @xmath141000 sq .
deg is obtained when combined with the 170 sq .
deg of the wide component of the cfht legacy survey .
the survey area for the 820 sq .
deg is divided up into 13 patches of 9@xmath439 or 6@xmath436 pointings of the megacam .
the survey is expected to be completed by the end of 2007 .
the depths of the survey allow us to be able to detect galaxy clusters of richness of about half that of coma out to redshift 1 .
the inclusion of the @xmath6 band filter will allow for the proper characterization of the galaxy sample down to redshift as low as 0.1 . in any survey of significant size ,
the only feasible way to obtain the mass observable is from the survey data themselves , in our case , the optical richness .
a large observational effort is also underway , using primarily clusters from rcs-1 , to calibrate and provide strong priors on the cluster mass - richness relation .
having strong priors to the mass - richness relation over several redshift bins ( to estimate @xmath22 ) , and including an estimate of the scatter , will provide much tighter constraints on the cosmological parameters , and is essential for the measurement of @xmath9 .
our calibration efforts include using weak lensing cluster mass as our primary mass calibrator , augmented by dynamical mass measurements from multi - object spectroscopy for a core sample of @xmath46 clusters , and x - ray observations .
the study of cluster galaxy evolution , using multi - color optical and ir photometry and spectroscopy , will also allow us to account for galaxy evolution when computing the richness of clusters at different redshifts .
the results from rcs-1 bode well for the use of rcs-2 as a probe for @xmath9 . scaling with the rcs-1 as a guide
, we expect to obtain a cluster sample of the order of 20,000 clusters with @xmath47 of similar richness . with priors on the mass - richness relation from our follow - up program ,
we expect to be able to measure to an accuracy of @xmath1410% in @xmath9 , @xmath140.02 in @xmath2 , and @xmath140.05 in @xmath3 . ,
k. , et al . 2007 , to be submitted to , m. d. & yee , h. k. c. 2000 , , 120 , 2148 , m. d. & yee , h. k. c. 2005 , , 157 , 1 , m. d. , yee , h. k. c. , majumdar , s. , barrientos , l.f . , hoekstra , h. , hall , p.b .
, & infante , l. 2007 , , 655 , 128 , m. , et al . 2005 , , 633 , 122 , a. et al .
2001 , , 321 , 372 , s. & mohr , j.j .
2003 , , 585 , 603 , d.n .
2006 , astro - ph/0603449 , h.k.c . &
ellingson , e. 2003 , , 585 , 215 | the red - sequence cluster surveys ( rcs-1 and rcs-2 ) are large optical imaging surveys optimized to create well - characterized catalogs of clusters of galaxies up to a redshift of @xmath0 .
we describe our first cosmological analysis , using the self - calibration technique , of a cluster sample of @xmath1 from the 90 sq .
deg rcs-1 , using optical richness as a mass proxy .
we obtain values for the cosmological parameters @xmath2 and @xmath3 that are in excellent agreement with the year - three wmap results .
furthermore , the derived cluster richness - mass relationship is entirely consistent with those measured directly using dynamical mass measurements .
we describe briefly the on - going rcs-2 , which is a 1000 sq .
deg survey carried out using the one - square - degree megacam on the cfht .
this survey , in @xmath4 , @xmath5 and @xmath6 bands , will produce a cluster sample of several 10@xmath7 clusters up to @xmath8 , and will allow us to constrain the dark energy equation of state , @xmath9 , and provide a large cluster sample for the studies of cluster evolution and lensing . |
it is widely believed that most complex systems are unpredictable , with concrete implications in earthquake prediction , in engineering failure @xcite , and in financial markets @xcite to cite a few .
in addition to the persistent failures of predictive schemes for these systems , concepts such as self - organized criticality @xcite suggest an intrinsic impossibility for the prediction of catastrophes . several recent works suggest a different picture : catastrophes may result from novel mechanisms amplifying their size @xcite and may thus be less unpredictable than previously thought .
this idea has been mostly explored in material failure @xcite , in earthquakes @xcite , and in financial markets and emerged in the latter from the analysis of cumulative losses ( drawdowns ) @xcite , from measures of algorithmic complexity @xcite , and from agent - based models @xcite .
we present novel empirical tests that provide a strong support for the hypothesis that large events can be predicted .
we focus our analysis on financial indices ( typically the daily dow jones industrial average ( djia ) from 26-may-1896 to 11-mar-2003 ) as they provide perhaps the best data sets that can be taken as proxies for other complex systems .
our methodology is based on the assumption that fast large market drops ( crashes ) are the results of interactions of market players resulting in herding behavior : exogenous shocks ( due to changes in market fundamentals ) often do not play an important role , which is at odds with standard economy theory .
the fact that exogenous shocks may not be the most important driving causes of the structures found in financial time series has been shown to be the case for volatility shocks @xcite : most of the bursts of volatility in major us indices can be explained by an endogenous organization involving long - memory processes , while only a few major shocks such as 9/11/2001 or the coup against gorbachev in 1991 are found to leave observable signatures . concerning financial crashes in indices , bonds and currencies ,
@xcite have performed an extended analysis of the distribution of drawdowns ( cumulative losses ) in the two leading exchange markets ( us dollar against the deutsch and against the yen ) , in the major world stock markets , in the u.s . and japanese bond market and in the gold market and
have shown the existence of `` outliers , '' in the sense that the few largest drawdowns do not belong to the same distribution .
for each identified outlier , @xcite have checked whether the formula ( [ eq : lnpt ] ) given below , expressing a so - called log - periodic power law signature ( lppl ) , could fit the price time series preceding them ; if yes , the existence of the lppl was taken as the qualifying signature for an endogenous crash ( a drawdown outlier was seen as the end of a speculative unsustainable accelerating bubble generated endogenously ) . in the absence of lppl
, @xcite were able to identify the relevant historical event , _
i.e. _ , a new piece of information of such magnitude and impact that it is reasonable to attribute the crash to it , following the standard view of the efficient market hypothesis .
such drawdown outliers were thus classified as having an exogenous origin .
globally over all the markets analyzed , @xcite identified 49 outliers , of which 25 were classified as endogenous , 22 as exogenous and 2 as associated with the japanese `` anti - bubble '' starting in jan .
1990 . restricting to the world market indices
, @xcite found 31 outliers , of which 19 are endogenous , 10 are exogenous and 2 are associated with the japanese anti - bubble .
the combination of the two proposed detection techniques , one for outliers in the distribution of drawdowns and the second one for lppl , provided a systematic taxonomy of crashes .
the present paper goes one step further to propose a prediction scheme for crashes , and by extension large and rapid changes of regime .
our proposed learning algorithm belongs to a recent body of literature that challenges the efficient market hypothesis ( emh ) . in the context of information
theory which is related to the present pattern recognition approach , see for instance @xcite .
our approach is also related to the field of nonlinearity applied to financial time series : see @xcite for a review and @xcite for tests and models of nonlinearity in financial time series . however , while we use a deterministic function to fit bubble periods , our approach is not deterministic because our underlying rational expectation bubble theory acknowledges the stochastic component introduced by the crash hazard rate and by the random regime shifts associated with the nucleation of bubble periods .
in addition , the lppl pattern that we propose to characterize these bubble periods , while rather rigid in its geometrical proportions exemplifying the concept of discrete scale symmetry ( see below ) , is different from one bubble to the next as its time span and amplitude can change .
this paper is organized as follows .
section [ s1:theory ] presents the theoretical foundation of our approach .
section [ s1:pr ] describes the definition of the pattern recognition method and its implementation and tests on the djia .
section [ s1:hsi ] shows further tests on the hong kong hang - seng index and section [ s1:conclude ] concludes .
our key idea is to test for signatures of collective behaviors similar to those well - known in condensed - matter and statistical physics .
the existence of similar collective behaviors in physics and in markets may be surprising to those not involved in the study of complex systems . indeed , in physics , the governing laws are well established and tested whereas one could argue that there is no well - established fundamental law in stock markets .
however , the science of complexity developed in the last two decades has shown that the emergence of collective phenomena proceeds similarly in physics , biology or social sciences as long as one is interested in the coarse - grained properties of large assemblies of constituting elements or agents ( see for instance @xcite and references therein ) .
collective behaviors in population of agents can emerge through the forces of imitation leading to herding @xcite .
herding behavior of investors is reflected in significant deviations of financial prices from their fundamental values @xcite , leading to so - called speculative bubbles @xcite and excess volatility @xcite .
the similarity between herding and statistical physics models , such as the ising model , has been noted by many authors @xcite .
we use this similarity to model a speculative bubble as resulting from positive feedback investing , leading to a faster - than - exponential power law growth of the price @xcite .
in addition , the competition between such nonlinear positive feedbacks , negative feedbacks due to investors trading on fundamental value and the inertia of decision processes lead to nonlinear oscillations ( nonlinear in the sense of amplitude - dependence frequencies ) @xcite that can be captured approximately by adding an imaginary part to the exponent of the power law price growth ( by power law price growth , we refer to the super - exponential growth typically associated with the solution of the equation @xmath2 , with @xmath3 and @xmath4 . for @xmath5 ,
one recovers the standard exponential growth . for @xmath4
, @xmath6 can approach a singularity in finite time like @xmath7 such that it is written @xmath8 describes the existence of accelerating log - periodic oscillations obeying a discrete scale invariance ( dsi ) symmetry @xcite .
this can be seen from the fact that @xmath9 $ ] .
more generally , it can be shown that the first terms in a systematic expansion can be written as @xmath10 = a - b\tau^m + c\tau^m \cos\left[\omega\ln(\tau)-\phi\right]~ , \label{eq : lnpt}\ ] ] where @xmath11 is the price of a specific asset ( it can be a specific stock , commodity or index ) , @xmath12 is the logarithm of the price at @xmath13 , @xmath14 is the angular log - frequency , @xmath15 and @xmath16 for an acceleration to exist , @xmath17 quantifies the amplitude of the log - periodic oscillations , @xmath18 is an arbitrary phase determining the unit of the time , @xmath19 is the current time , and @xmath20 is the distance to the critical time @xmath13 defined as the end of the bubble .
equation ( [ eq : lnpt ] ) does not hold for any time but rather applies to those non - stationary transient phases , otherwise known as bubbles .
expression ( [ eq : lnpt ] ) has been used to fit bubbles according to the following algorithm described in @xcite , that we summarize here : ( i ) the crashes were identified within a given financial time series as the largest cumulative losses ( largest drawdowns ) according to the methodology of @xcite ; ( ii ) the bubble preceding a given crash was then identified as the price trajectory starting at the most obvious well - defined minimum before a ramp - up ending at the highest value before the crash .
the starting value ( local minimum ) of the bubble is unambiguous for large developed markets but it was sometimes necessary to change the interval fitted due to some ambiguity with respect to the choice of the first point in these smaller emergent markets @xcite .
this algorithm was applied to over thirty crashes in the major financial markets and the fits of their corresponding bubbles with eqn.([eq : lnpt ] ) showed that the dsi parameter @xmath14 exhibits a remarkable universality , with a mean of @xmath21 and standard deviation of @xmath22 @xcite , suggesting that it is a significant distinguishing feature of bubble regimes . in these previous
look back studies , the knowledge of the time of a given crash was used to identify its associated bubble with the purpose of testing for a common behavior .
this was done in the usual spirit of a scientific endeavor which consists in general in the following steps : ( a ) collect samples ; ( b ) test for regularities and commonalities in the samples to establish a robust classification of a new phenomenon ; ( c ) develop a model accounting for this classification and calibrate it ; ( d ) put the model to test using its prediction that are applied to novel data , that have not be used in the model construction .
the previous works mentioned above @xcite cover the steps ( a - c ) .
the present paper addresses step ( d ) by developing a systematic look forward approach in which the goal is to identify `` in real time '' the existence of a developing bubble and to predict times of increased crash probability . the physics - type construction ( i.e. , emphasizing interactions and collective behavior ) leading to eqn.([eq : lnpt ] ) is however missing an essential ingredient distinguishing the natural and physical from the social sciences , namely the existence of expectations : market participants are trying to discount a future that is itself shaped by market expectations . as exemplified by the famous parable of keynes beauty contest , in order to predict the winner , recognizing objective beauty is not very important , but knowledge or prediction of others prediction of beauty is much more relevant . similarly , mass psychology and investors expectations influence financial markets significantly .
we use the rational - expectation ( re ) blanchard - watson model of speculative bubbles @xcite to describe concisely such feedback loops of prices on agent expectation and vice - versa . applied to a given bubble price trajectory
, agents form an expectation of the sustainability of such bubble and of its potential burst . by this process , to each price @xmath11 is associated a crash hazard rate @xmath23 , such that @xmath24 is the probability for a crash or large change of regime of average amplitude @xmath25 to occur between @xmath19 and @xmath26 conditioned on the fact that it has not happened . the crash hazard rate @xmath23 quantifies the general belief of agents in the non - sustainability of the bubble .
the re - bubble model predicts the remarkable relationship @xcite @xmath27 = \kappa\int_{t_0}^t h(t')dt ' \label{positcond}\ ] ] linking price to crash hazard rate .
this equation is the solution of the model describing the dynamics of the price increment @xmath28 over a time increment dt : @xmath29 , where @xmath30 is the instantaneous return , @xmath31 is a jump process equal to @xmath32 when a crash occurs with amplitude @xmath25 and zero otherwise .
the no - arbitrage condition @xmath33=0 $ ] implies that @xmath34 $ ] , where @xmath35 $ ] denotes the expectation conditional on all information available up to the present time . by definition of the crash hazard rate , @xmath36 $ ] , this gives @xmath37 .
thus , conditioned on no crash having occurred until time @xmath19 , the integration of the model yields ( [ positcond ] ) . substituting eqn.([eq : lnpt ] ) in eqn.([positcond ] ) and using the fact , that by definition @xmath38 , gives the constraint @xmath39 this condition has been found to provide a significant feature for distinguishing market phases with bubbles ending with crashes from regimes without bubbles @xcite . note
that @xmath23 is not normalized to @xmath32 : @xmath40 gives the total conditional probability for a crash to occur , which is less than @xmath32 for the re bubble model to hold .
there is thus a finite probability @xmath41 for the bubble to land smoothly without a crash .
this is necessary for the re bubble model , as otherwise the crash would be certain . by having a non - zero probability for no crash to occur ,
it remains rational for investors to remain in the market and harvest the risk premium ( the bubble ) for the risk of a possible ( but not certain ) crash .
technically , using the standard rule of conditional probabilities , @xmath23 is related to the unconditional probability @xmath6 for a crash per unit time by @xmath42 . by integrating , rearranging and differentiating
, this gives @xmath43 $ ] .
let us now build on these insights and construct a systematic predictor of bubble regimes and their associated large changes . due to the complexity of financial time series , to the existence of many different regimes , and to the constant action of investors arbitraging gain opportunities , it is widely held that the prediction of crashes is an impossible task .
the truth is that financial markets do not behave like statistically stationary physical systems and exhibit a variety of non - stationarities , influenced both by exogenous and endogenous shocks @xcite . to obtain robustness and to extract information systematically in the approximately self - similar price series , it is useful to use a multiscale analysis : the calibrations of eqn.([eq : lnpt ] ) and eqn.([positcond ] )
are thus performed for each trading day over nine different time scales of @xmath44 , @xmath45 , @xmath46 , @xmath47 , @xmath48 , @xmath49 , @xmath50 , @xmath51 and @xmath52 trading days . each trading day
ending an interval of length @xmath52 or smaller is called an object and there are a total of 27428 objects from 06-feb-1902 till 11-mar-2003 .
we define a map from the calendar time @xmath19 to an ordered index @xmath53 of the trading day , such that @xmath54 if @xmath19 is a trading date and @xmath55 ( empty set ) otherwise .
this mapping is just a convenient way for defining the set of objects to which our method is applied
. it does not produce any bias or additional dynamics since it is used only in the classification scheme .
we denote @xmath56 a list of targets ( crashes ) that happened at @xmath57 .
the total number of targets is thus @xmath58 .
all objects are partitioned into two classes @xmath59 and @xmath60 , where class @xmath61 contains all the objects in @xmath62 $ ] for @xmath63 and class @xmath64 includes all the remaining , where @xmath65 is a given number in unit of calender day . in the following ,
our target set are three well - known speculative bubbles that culminated on 14-sep-1929 , 10-oct-1987 and 18-jul-1998 , respectively , before crashing in the following week or month .
the first two bubbles are perhaps the most famous or infamous examples discussed in the literature .
the last one is more recent and has been associated with the crisis of the russian default on its debt and ltcm . of course , the choice of this triplet is otherwise arbitrary .
our goal is just to show that our method can generalize to other bubbles beyond those which have been used for training .
figure [ fig1 ] presents the distributions of the @xmath14 s ( defined in eqn.([eq : lnpt ] ) ) and @xmath66 s ( defined in eqn.([defb ] ) ) , obtained from fits with eqn.([eq : lnpt ] ) in intervals ending on objects for all time scales and for three different values of @xmath65 .
the distributions for classes @xmath59 and @xmath60 , which are very robust with respect to changes in @xmath65 , are very different : for @xmath67 $ ] ( which corresponds to the values obtained in previous works ) , @xmath68 is larger than @xmath69 ; @xmath70 of objects in class @xmath59 obey the constraint @xmath71 compared to @xmath72 in class @xmath60 .
a standard kolmogorov - smirnov test gives a probability that this difference results from noise much less than @xmath73 .
the very significant differences between the distributions in classes @xmath59 and @xmath60 confirm that the dsi and constraint parameters are distinguishing patterns of the three bubbles .
figure [ fig2 ] plots a vertical line segment for each time scale at the date @xmath19 of the object that fulfills simultaneously three criteria : @xmath74 , @xmath75 and @xmath76 .
the last criterion ensures a smooth super - exponential acceleration .
each stripe thus indicates a potentially dangerous time for the end of a bubble and for the coming of a crash .
the number of dangerous objects decreases with the increase of the scale of analysis , forming a hierarchical structure .
notice the clustering of alarms , especially at large scales , around the times of the three targets .
altogether , figs .
[ fig1 ] and [ fig2 ] suggest that there is valuable information for crash prediction but the challenge remains to decipher the relevant patterns in the changing financial environment . for this , we call in the statistical analysis of sparse data developed by the russian mathematical school @xcite in order to construct a systematic classifier of bubbles . the modified `` cora-3 '' algorithm @xcite that we briefly describe below allows us to construct an alarm index that can then be tested out - of - sample .
each object is characterized by five quantities , which are believed to carry relevant information : @xmath14 , @xmath66 , @xmath77 , the r.m.s . of the residuals of the fits with eqn.([eq : lnpt ] ) and the height of the power spectrum @xmath78 of log - periodicity at @xmath14 @xcite .
the residuals exhibit much less structure that the initial time series but some dependence remains as the fits with eqn.([eq : lnpt ] ) account only for the large time - scale structure .
indeed , using the r.m.s .
of the residuals in the discrimination of the objects is useful only if some dependence remains in them . in other words ,
the residuals contain all the information at the smaller time scales and it is hoped that these sub - structures may be linked to the log - periodic structure at the largest time scale .
this has actually been shown to be the case by @xcite in a specific case and by the construction of trading methods at short time scales [ _ johansen and sornette _ , unpublished ] .
this also follows from the full solution of the underlying so - called renormalization group equation embodying the critical nature of a bubble and crash @xcite .
the use of the power spectrum @xmath78 follows the same logic .
integrated with the 9 scales , there are totally @xmath79 parameters for each object .
one could worry that using so many parameters is meaningless in the deterministic framework of eqn.([eq : lnpt ] ) because the large number of degrees of freedom and their associated uncertainties lead to an effective randomness , but this view does not recognize that this approach amounts to a decomposition in multiple scales : ( a ) it is allowed for or recognized that the price trajectory has patterns at many scales ; ( ii ) for each pre - selected scale , a rather parsimonious deterministic fit is obtained .
the selection of the relevant parameters discussed below amounts actually to searching for those time scales that are the most relevant , since the data should decide and not some a priori preconception .
we set @xmath80 but have checked that all results are robust with large changes of @xmath65 .
class i has 444 objects . by proceeding as in fig .
[ fig1 ] , for each of the 45 parameters , we qualify it as relevant if its distributions in classes i and ii are statistically different .
we have played with several ways of quantifying this difference and find again a strong robustness . out of the 45 parameters , 31 are found to be informative and we have determined their respective intervals over which the two distributions are significantly different .
we use these 31 parameters to construct a questionnaire of 31 questions asked to each object .
the @xmath81th question on a given object asks if the value of the @xmath81th parameter is within its qualifying interval .
each object @xmath12 can then be coded as a sequence of binary decisions so that @xmath82 where @xmath83 is the answer ( y / n ) to the @xmath81th question .
a distinctive property of the pattern recognition approach is to strive for robustness notwithstanding the lack of sufficient training data ( here only three bubbles ) @xcite . in this spirit , we define a trait as an array @xmath84 where @xmath85 , @xmath86 , @xmath87 and @xmath88 y or n. there are @xmath89 possible traits . if @xmath90 , @xmath91 and @xmath92 , we say that the object @xmath12 exhibits a trait @xmath84 .
the number of questions ( three are used here ) defining a trait is a compromise between significance and robustness .
a feature is a trait that is present relatively frequently in class i and relatively infrequently in class ii .
specifically , if there are no less than @xmath93 objects in class @xmath59 and no more than @xmath94 objects in class @xmath60 that exhibit a same trait , then this trait is said to be a feature of class @xmath59 . calling @xmath95 the number of distinctive features of a given object at time @xmath19 , we define the alarm index @xmath96 , here amounting to perform a moving average of @xmath95 over a time window of width @xmath97 .
we then normalize it : @xmath98 .
let us summarize our construction leading to the alarm index .
* we select a few targets ( the three well - known speculative bubbles ending in a crash at oct . 1929 , oct . 1987 and aug .
1998 for the us index and two targets for the hang seng ) , which serve to train our system ( independently for the two indices ) . *
an object is defined simply as a trading day ending a block of trading days of a pre - defined duration . *
those objects which are in a neighborhood of the crashes of the targets are defined to belong to class i. all other objects are said to belong to class ii . * for each object , we fit the price trajectory with expression ( [ eq : lnpt ] ) over a given time scale ( defining the duration of the block of days ending with the object used in the fit ) and obtain the corresponding values of the parameters , which are considered as characteristic of the object .
this step is repeated nine times , once for each of the nine time scales used in the analysis .
we keep @xmath99 parameters of the fit by expression ( [ eq : lnpt ] ) for each time scale , thus giving a total of @xmath100 parameters characterizing each object .
* we construct the probability density functions ( pdf ) of each parameter over all objects of class i and of class ii separately .
those parameters which are found to exhibit sufficiently distinct pdf s over the two classes are kept as being sufficiently discriminating . in this way , out of the total of @xmath101 parameters , @xmath102 are kept . *
each selected parameter gives a binary information , y or n , as whether its value for a given object falls within a qualifying interval for class i. * as a compromise between robustness and an exhausive description , we group the parameters in triplets ( called traits ) to obtain a characterization of objects .
ideally , one would like to use all @xmath102 parameters simultaneously for each object , but the curse of dimensionality prevents doing this .
* we study the statistics of all traits and look for those which are frequent in objects of class i and unfrequent in objects of class ii .
such traits are called features . *
the alarm index at a given time is defined as a moving average number of distinctive features found at that time .
large values of the alarm index are proposed to be predictors of changes of regime in the stock market .
figure [ fig3 ] shows the alarm index @xmath96 for @xmath103 with @xmath104 ( corresponding to at least @xmath105 of the objects in class i ) and @xmath106 ( corresponding to no more than @xmath107 of the objects in class ii ) . the alarm index is found to be highly discriminative as only 13.0% ( respectively 9.5% , and 4.7% ) of the total time is occupied by a high level of alarm larger than @xmath108 ( respectively @xmath109 , and @xmath110 ) .
we have performed extensive experiments to test the robustness of the results .
we have varied simultaneously @xmath93 from 100 to 400 with spacing 50 and @xmath94 from @xmath111 to @xmath112 with spacing @xmath113 .
the results are remarkably robust with no change of the major peaks .
the three bubbles used in the learning process are predicted ( this is expected of course ) : peak 3 is on 01-jul-1929 ( market maximum on 14-sep-1929 ) ; peak 9 is on 10-oct-1987 ( market maximum on 04-sep-1987 and crash on 19-oct-1987 ) ; peak 13 is on 23-apr-1998 ( market maximum on 18-jul-1998 ) .
this timing is a good compromise for a prudent investor to exit or short the market without losing much momentum .
this success is however not a big surprise because this corresponds to in - sample training .
the most remarkable property of our algorithm is its ability for generalizing from the three bubbles ending with crashes to the detection of significant changes of regimes that may take a large variety of shapes ( for instance , a crash is not a one - day drop of , say , + 15% but can be spread over several days or a few weeks with rather complex trajectories , leading to very large cumulative losses ) .
the following peaks are predictors of crashes or of strong corrections with an @xmath114 alarm threshold : peak 1 ( jan , 1907 ) , peak 4 ( apr , 1937 ) , peak 5 ( aug , 1946 ) , peak 6 ( jan , 1952 ) , peak 7 ( sep , 1956 ) , peak 8 ( jan , 1966 ) , peak 12 ( jul , 1997 ) and peak 14 ( sep , 1999 ) .
the more than @xmath115 price drops in 1907 ( peak 1 ) and 1937 ( peak 4 ) have been previously identified as crashes occurring over less than three months @xcite .
peak 12 is occurring three months before the turmoil on the stock market ( one day @xmath116 drop on 27-oct-1997 ) that was followed by a three month plateau ( see discussion in chapter 9 of ref .
peak 10 is a false alarm .
peaks 2 and 11 can be regarded as early alarms of the following 1929 crash and 1997 , 1998 , and 2000 descents .
peak 15 - 17 are smaller peaks just barely above @xmath117 .
peaks 15 and 16 are false alarms , while peak 17 ( 31-may-2000 ) is just two months before the onset of the global anti - bubble starting in august of 2000 @xcite .
figure [ fig4 ] is constructed as follows .
each alarm index peak is characterized by the time @xmath13 of its maximum and by its corresponding log of price @xmath118 $ ] .
we stack the index time series by synchronizing all the @xmath13 for each peak at the origin of time and by translating vertically the prices so that they coincide at that time .
figure [ fig4 ] shows four years of the djia before and two years after the time of each alarm index maximum .
we choose this representation to avoid the delicate and still only partly resolved task of defining what is a crash , a large drop or a significant change of regime . on this problem ,
the literature is still vague and not systematic .
previous systematic classifications @xcite miss events that are obvious to the trained eyes of professional investors or on the contrary may incorporate events that are controversial .
figure [ fig4 ] shows that these peaks all correspond to a time shortly before sharp drops or strong change of market regimes . in a real - time situation ,
a similar prediction quality can be obtained in identifying the peaks by waiting a few weeks for the alarm index to drop or alternatively by declaring an alarm when @xmath96 reaches above a level in the range @xmath119 .
the sharpness and the large amplitudes of the peaks of the alarm index ensures a strong robustness of this latter approach .
our method however misses several large price drops , such as those in 1920 , 1962 , 1969 , 1974 @xcite .
this may be attributed to a combination of the fact that some large market drops are not due to the collective self - organization of agents but result from strong exogenous shocks , as argued in refs .
@xcite and that the method needs to be improved .
however , the generalizing ability of our learning algorithm strengthens the hypothesis that herding behavior leads to speculative bubbles and sharp changes of regimes .
this brief presentation is borrowed from chap .
9 of the book in @xcite . in evaluating predictions and their impact on ( investment ) decisions
, one must weight the relative cost of false alarms with respect to the gain resulting from correct predictions .
the neyman - pearson diagram , also called the decision quality diagram , is used in optimizing decision strategies with a single test statistic .
the assumption is that samples of events or probability density functions are available both for correct signals ( the crashes ) and for the background noise ( false alarms ) ; a suitable test statistic is then sought which optimally distinguishes between the two . using a given test statistic ( or discriminant function )
, one can introduce a cut which separates an acceptance region ( dominated by correct predictions ) from a rejection region ( dominated by false alarms ) . the neyman - pearson diagram plots contamination ( misclassified events , i.e. , classified as predictions which are thus false alarms ) against losses ( misclassified signal events , i.e. , classified as background or failure - to - predict ) , both as fractions of the total sample .
an ideal test statistic corresponds to a diagram where the `` acceptance of prediction '' is plotted as a function of the `` acceptance of false alarm '' in which the acceptance is close to @xmath32 for the real signals , and close to @xmath120 for the false alarms .
different strategies are possible : a `` liberal '' strategy favors minimal loss ( i.e. , high acceptance of signal , i.e. , almost no failure to catch the real events but many false alarms ) , a `` conservative '' one favors minimal contamination ( i.e. , high purity of signal and almost no false alarms but many possible misses of true events ) .
@xcite has shown that the task of predicting an event in continuous time can be mapped onto the neyman - pearson procedure .
he has introduced the `` error diagram '' which plots the rate of failure - to - predict ( the number of missed events divided by the total number of events in the total time interval ) as a function of the rate of time alarms ( the total time of alarms divided by the total time , in other words the fraction of time we declare that a crash or large correction is looming ) ( see also @xcite for extensions and reviews ) .
the best predictor corresponds to a point close to the origin in this diagram , with almost no failure - to - predict and with a small fraction of time declared as dangerous : in other words , this ideal strategy misses no event and does not declare false alarms !
these considerations teach us that making a prediction is one thing , using it is another which corresponds to solving a control optimization problem @xcite . to assess quantitatively the prediction procedure described above
, we thus construct an error diagram , plotting the failure to predict as a function of the total alarm duration .
the targets to be predicted are defined as large drawdowns ( cumulative losses as defined in @xcite ) whose absolute values are greater than some given value @xmath121 .
each drawdown has a duration covering a time interval @xmath122 $ ] . for a given alarm index threshold @xmath123 ,
if there exist a time @xmath124 so that @xmath125 , then the drawdown is said to be successfully predicted ; in contrast , if @xmath126 for all @xmath127 , we say it is a failure .
this definition ensures that the forecast occurs before the occurrence of the drawdown .
the error diagram is obtained by varying the decision threshold @xmath123 .
the quantity `` failure to predict '' is the ratio of the number of failures over the number of targets as said above .
the total alarm duration is the total time covered by the objects whose alarm index is larger than @xmath123 divided by the total time . by varying the value of @xmath123 , we obtain different pairs of failure to predict and total alarm duration .
figure [ fig5 ] presents the error diagram for two target definitions ( @xmath128 and @xmath129 ) for the djia with @xmath130 , @xmath131 and @xmath132 .
the results do not change significantly as long as @xmath93 is much smaller than @xmath133 .
the anti - diagonal corresponds to the completely random prediction scheme .
the error diagrams with different @xmath134 , @xmath93 and @xmath133 are found to be similar , with small variations , showing a strong robustness of our construction .
the inset shows the prediction gain , defined as the ratio of the fraction of targets correctly predicted ( equal to one minus the fraction of missed targets ) to the total alarm duration .
theoretically , the prediction gain of a random prediction strategy is @xmath32 .
the error diagram shows that half of the targets with @xmath129 are predicted with a very small alarm duration and all targets are predicted for an alarm duration less than 40% .
this gives rise to large prediction gains for small alarm durations , confirming the very good quality of our predictors .
confidence in this conclusion is enhanced by finding that the quality of the prediction increases as the targets are evolved toward largest losses ( @xmath135 loss for @xmath128 to @xmath136 loss for @xmath129 ) .
this trend continues for largest losses but there are too few events to draw statistically meaningful conclusions for larger drawdowns of @xmath115 or more .
the validity of our construction should be ascertained by testing it without modification on other independent time series . here , we present similar promising results obtained for the hong kong hang seng composite index ( hsi ) from 24-nov-1969 to present .
the hong kong hang seng composite index is particularly interesting as it can be considered as a `` textbook '' example of an unending succession of bubbles and crashes @xcite .
the nine biggest crashes since 24-nov-1969 were triggered approximately at the following dates : 20-sep-1971 , 5-mar-1973 , 04-sep-1978 , 13-nov-1980 , 01-oct-1987 , 15-may-1989 , 4-jan-1994 , 8-aug-1997 , and 28-mar-2000 . except for the last one which was posterior to the study published in @xcite ,
all of them have been studied previously in @xcite .
the distinctive properties of the first eight bubbles are consistent with those reported previously @xcite ( see page 461 concerning the parameters @xmath14 and @xmath77 ) .
in contrast with ref .
@xcite in which the positivity constraint was not considered , we find here that it plays a significant role in screening out solutions .
we use the two crashes in 1987 and 1997 to train the parameters of our algorithm because they give the best defined log - periodic power law patterns over the longest time intervals .
figure [ fig6 ] illustrates the typical result for the alarm index obtained with @xmath80 , @xmath137 and @xmath138 .
four out of the five crashes that were not used in the training phase are identified very clearly by the alarm index peaks ( note that the first two crashes in 20-sep-1971 and 5-mar-1973 are omitted in our counting since they occur before the time when our alarm index can be constructed ) .
this encouraging result seems to confirm the hypothesis that the seven crashes , posterior to 1976 where the alarm index can be defined , are triggered by endogenous stock market instabilities preceded by log - periodic power - law speculative bubbles @xcite .
varying the values of the parameters @xmath65 , @xmath93 and @xmath133 of our algorithm gives results which are robust .
in addition , one can identify more peaks in fig .
[ fig6 ] and it is an interesting question to test whether they are associated with an anomaly in the market . in total , we count 15 peaks in fig .
[ fig6 ] around 15-sep-1978 ( s ) , 09-apr-1980 ( ? ) , 29-sep-1981 ( f ) , 30-nov-1982 ( f ) , 01-oct-1986 ( ? ) , 30-sep-1987 ( s ) , 20-may-1989 ( s ) , 18-aug-1990 ( f ) , 09-oct-1991 ( f ) , 19-oct-1992 ( f ) , 28-oct-1993 ( s ) , 05-may-1996 ( ? ) , 06-aug-1997 ( s ) , 10-apr-1999 ( f ) , and 25-mar-2000 ( s ) , whose alarm indexes are greater than @xmath139 .
six peaks whose dates are identified by `` s '' correspond to successful predictions , among which two are trivial since they were used in the training process .
six peaks identified with `` f '' are false alarms . the remaining three peaks marked with `` ? ''
are neither successful predictions nor complete false alarms as they fall close to crashes .
the date with an alarm index of 0.1 on the east side of the first `` ? '' alarm peak is 31-jul-1980 and can probably be interpreted as a forerunner of the crash on 13-nov-1980 . the two other `` ? ''
alarm peaks can probably interpreted as `` fore - alarms '' of two main alarm peaks used for training the algorithm .
similar to fig .
[ fig4 ] , fig .
[ fig7 ] shows a superposed epoch analysis of the price trajectories a few years before and one year after these 15 peaks , to see whether our algorithm is able to generalize beyond the detection of crashes to detect changes of regimes .
the ability to generalize is less obvious for the hang - seng index that it was for the djia shown in fig .
[ fig4 ] .
figure [ fig8 ] presents the error diagram for three target definitions ( @xmath128 , @xmath129 , and @xmath140 ) for the hsi with @xmath130 , @xmath141 and @xmath142 .
the anti - diagonal corresponds to the random prediction scheme .
again , the error diagrams with different @xmath134 , @xmath93 and @xmath133 are robust .
the inset plots the corresponding prediction gains .
more than half of the targets with @xmath143 are predicted with a very small alarm time and all targets are predicted for a finite alarm time less than 50% .
these results are associated with a very significant predictability and strong prediction gains for small alarm durations .
we find that different stock markets have slightly different characteristics : for instance , the parameters trained on the djia are not optimal for the hang - seng index .
when we use the parameters obtained from the training of our algorithm on the djia on the hsi , we find that two of the seven crashes in the hsi are identified accurately ( 01-oct-1987 and 4-jan-1994 ) while the five other crashes are missed .
interestingly , these two crashes are the most significant .
this suggests the existence of a universal behavior only for the `` purest '' event cases , while smaller events exhibit more idiosyncratic behavior .
as we have shown , training our algorithm on these two events on the hsi improves very significantly the prediction of the other events .
this shows both a degree of universality of our procedure and the need to adapt to the idiosyncratic structure of each market .
this situation is similar to that documented in earthquake predictions based on pattern recognition techniques @xcite . how does the performance of our system compare with those of commonly used stochastic models of stock indices , such as garch ? to answer this question , we have used a standard method to calibrate a garch(1,1 ) model to the s&p500 .
then , the garch model was run to provide a prediction of the volatility for the next period ( here taken in unit of days ) , but not of the sign of the next return ( which is unpredictable according to the garch model )
. we found a quite strong predictability of the volatility `` predicted '' by garch by preceding large drops ( and often large ups ) of the market but not the reverse .
in other words , garch is not predicting in any way future large drops , it is the other way around : large realized drops predict future large garch - predicted values of the volatility .
actually , this is not surprising for two reasons : ( i ) from the structure of the garch model , a previous large daily drop ( or gain ) immediately increases the present amplitude of the volatility ; ( ii ) the fact that losses predict future increases of volatility ( and not the reverse ) has a rich literature associated with the `` leverage '' effect ( see for instance @xcite and references therein ) . thus , we can conclude that the garch model can in no way reproduce or even approach the detection of impending changes of regime as we have documented here . in summary , we have developed a multiscale analysis on the analysis of stock market bubbles and crashes .
based on a theory of investor imitation , we have integrated the log - periodic power - law patterns characteristic of speculative bubbles preceding financial crashes within a general pattern recognition approach .
we have applied our approach to two financial time series , djia ( dow jones industrial average ) and hsi ( hang - seng hong kong index ) .
training our algorithm on only a few crashes in each market , we have been able to predict the otherwise crashes not used in the training set .
our work provides new evidence of the relevance of log - periodic power - law patterns in the evolution of speculative bubbles supporting the imitation theory .
anifrani , j .- c . ,
lefloch , c. , sornette , d. , & souillard , b. ( 1995 ) .
universal log - periodic correction to renormalization - group scaling for rupture stress prediction from acoustic emissions , _ journal de physique i _ , 5 , 631 - 638 .
blanchard , o.j .
, & watson , m.w .
bubbles , rational expectations and speculative markets , in : p. wachtel ( ed . ) , _ crisis in economic and financial structure : bubbles , bursts , and shocks_. lexington : lexington books .
cecen , a.a . and erkal , c. ( 1996 ) .
distinguishing between stochastic and deterministic behavior in high frequency foreign exchange rate returns : can non - linear dynamics help forecasting ?
. j. forecasting _
12(4 ) , 465 - 473 .
gelfand , i.m . ,
guberman , s.a .
, keilis - borok , v.i . , knopoff , l. , press , f. , ranzman , e.y . ,
rotwain , i.m . , & sadovsky , a.m. ( 1976 ) .
pattern - recognition applied to earthquake epicenters in california , _ physics of the earth and planetary interiors _ , 11 , 227 - 283 .
johansen , a. , & sornette , d. ( 2001 ) .
bubbles and anti - bubbles in latin - american , asian and western stock markets : an empirical study , international journal of theoretical and applied finance 4 ( 6 ) , 853 - 920 .
johansen , a. , & sornette , d. ( 2004 ) .
endogenous versus exogenous crashes in financial markets , in press in `` contemporary issues in international finance '' ( nova science publishers ) .
( http://arxiv.org/abs/cond-mat/0210509 ) sornette , d. ( 2002 ) predictability of catastrophic events : material rupture , earthquakes , turbulence , financial crashes , and human birth , _ procedings of the national academic sciences of the united states of america _ , 99 , 2522 - 2529 .
zhou , w .- x . , &
d. sornette ( 2003b ) .
renormalization group analysis of the 2000 - 2002 anti - bubble in the us s&p 500 index : explanation of the hierarchy of five crashes and prediction , _ physica a _ 330 , 584 - 604 .
density distribution @xmath144 of the dsi parameter @xmath14 obtained from eqn.([eq : lnpt ] ) and complementary cumulative distribution @xmath145 of the constraint parameter @xmath66 obtained from eqn.([positcond ] ) for the objects in classes i ( dotted , dashed , and dotted - dashed ) and ii ( continuous ) for three different values of @xmath65.,width=491 ] alarm times @xmath19 ( or dangerous objects ) obtained by the multiscale analysis .
the alarms satisfy @xmath146 , @xmath74 and @xmath147 simultaneously .
the ordinate is the investigation `` scale '' in trading day unit .
the results are robust with reasonable changes of these bounds . , width=491 ] ( color online ) superposed epoch analysis of the 11 time intervals , each of 6 years long , of the djia index centered on the time of the maxima of the 11 predictor peaks above @xmath114 of the alarm index shown in fig .
[ fig3].,width=491 ] error diagram for our predictions for two definitions of targets to be predicted @xmath128 and @xmath129 obtained for the djia .
the anti - diagonal line corresponds to the random prediction result .
the inset shows the prediction gain.,width=491 ] ( color online ) alarm index @xmath96 ( upper panel ) and the hong kong hang seng composite index from 1975 to 2003 ( lower panel ) .
the vertical lines indicate the timing of the seven largest crashes .
note that the first two crashes are not included in the analysis since the longest window used in our multiscale analysis is seven years.,width=491 ] ( color online ) superposed epoch analysis of the 15 time intervals of the hsi index centered on the time of the maxima of the 15 peaks above @xmath148 of the alarm index shown in fig .
[ fig6].,width=491 ] error diagram of the predictions for three definitions of targets to be predicted with @xmath128 , @xmath129 , and @xmath140 in regard to hsi .
the diagonal line is a random prediction . in the inset
shows the prediction gains.,width=491 ] | we present a systematic algorithm testing for the existence of collective self - organization in the behavior of agents in social systems , with a concrete empirical implementation on the dow jones industrial average index ( djia ) over the 20th century and on hong kong hang seng composite index ( hsi ) since 1969 .
the algorithm combines ideas from critical phenomena , the impact of agents expectation , multi - scale analysis and the mathematical method of pattern recognition of sparse data . trained on the three major crashes in djia of the century ,
our algorithm exhibits a remarkable ability for generalization and detects in advance 8 other significant drops or changes of regimes .
an application to hsi gives promising results as well .
the results are robust with respect to the variations of the recognition algorithm .
we quantify the prediction procedure with error diagrams .
predictability of large future changes in major financial indices didier sornette@xmath0 , wei - xing zhou@xmath1 1 .
institute of geophysics and planetary physics , university of california , los angeles , ca 90095[igpp ] 2 .
department of earth and space sciences , university of california , los angeles , ca 90095[ess ] 3 . laboratoire de physique de la matire condense , cnrs umr 6622 and universit de nice - sophia antipolis , 06108 nice cedex 2 , france[lpec ] correspondence should be address to d. sornette : 1693 geology building + institute of geophysics and planetary physics + university of california , los angeles + ca 90095 - 1567 , usa .
tel : + 1 - 310 - 825 - 2863 + fax : + 1 - 310 - 206 - 3051 .
+ _
e - mail address:[email protected] ( d. sornette ) predictability of large future changes in major financial indices _ keywords _ : econophysics ; multiscale analysis ; pattern recognition ; predictability |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.