article
stringlengths 0
745k
| abstract
stringlengths 0
92.7k
|
|---|---|
the past decade has seen a series of anomalies emerge in short baseline ( sbl ) neutrino oscillation experiments which can not be explained within the three active neutrino framework of the standard model ( sm ) . here
, sbl refers to experiments with the ratio of the oscillation distance over the neutrino energy , @xmath1 , which are sensitive to neutrino oscillations involving mass squared splittings @xmath2 .
the lsnd experiment @xcite reports evidence of @xmath3 oscillation consistent with @xmath2 , as well as a less dramatic excess for @xmath4 oscillation @xcite .
the miniboone collaboration also searched for the same signal , reporting excesses in both electron and anti - electron neutrino events @xcite , again suggesting oscillations of the form @xmath4 and @xmath3 , consistent with the lsnd results .
together , these observations lead to the tantalizing suggestion of additional `` sterile '' neutrino flavors at a mass scale of @xmath5 .
many schemes have been considered to fit the excess , including 3 active plus @xmath6 sterile neutrino oscillation schemes ( 3+@xmath6 ) , with most of the attention being focused on @xmath7 and @xmath8 @xcite .
while even the simple 3 + 1 scheme can provide a good fit to the @xmath4 ( @xmath3 ) appearance excesses , these fits are in tension with @xmath9 , @xmath10 and @xmath11 disappearance constraints from miniboone+sciboone @xcite and lsnd+karmen @xcite , respectively . to ameliorate the disappearance constraint ,
some authors have also considered fairly prompt decay of sterile neutrinos @xcite ( _ i.e. _ @xmath12 ) to allow the decay of sterile neutrinos to active neutrinos within the experimental lengths . in most cases ,
a decay of this form requires the coupling of neutrinos to a new light state @xmath13 ( potentially a majoron @xcite ) , which enables the sterile neutrinos to decay through the process @xmath14 . while some authors have also considered decays of the form @xmath15 @xcite to explain the miniboone signal , this decay can not explain the lsnd excess and we will not consider it here
. very little attention has been focused on the 3 + 3 oscillation scheme , mainly in the interest of minimality and because there was no clear indication that adding a third sterile neutrino would improve the 3 + 2 fit . however
, when the sterile neutrino sector is embedded within a model that mirrors " the sm particle content @xcite , the 3 + 3 scenario becomes natural to consider .
one well motivated model of this type is the twin higgs " model @xcite , although others have been considered @xcite .
the twin higgs model contains a full or partial copy of the sm gauge group and particle content , with couplings in the two sectors related by a discrete @xmath16 symmetry .
the particle content in each sector , usually denoted @xmath17 and @xmath18 , transforms under its own gauge group and is sterile with respect to the other sector .
an attractive feature of the twin higgs model is that it provides a solution to the little hierarchy problem without requiring new particles charged under the sm gauge group , at least below the cutoff of the effective theory . in this model ,
the higgs field is a pseudo - nambu goldstone boson ( pngb ) associated with spontaneous breaking of an approximate global @xmath19 symmetry .
a twin @xmath16 symmetry is introduced to constrain the form of corrections to the pngb higgs potential , allowing natural electroweak symmetry breaking with no quadratically divergent corrections to the higgs mass at one - loop level . in this paper , we construct a 3 + 3 neutrino model within the twin higgs framework , although many of our phenomenological studies can be applied to other models with similar flavor structures .
two higher dimensional operators turn out to be relevant for the neutrino sectors .
the first operator is dimension - five and respects both the @xmath16 and @xmath19 symmetries .
after the higgs fields develop their vacuum expectation values ( vev s ) , three out of the total six neutrinos become massive and can be identified as the three sterile neutrinos . because of @xmath16-enforced yukawa alignment between the two sectors , only one universal mixing angle in addition to the usual pontecorvo - maki - nakagawa - sakata ( pmns ) matrix is required to describe how the three sterile neutrinos interact with the sm charged leptons .
this mixing angle @xmath20 is predicted to be the ratio of two higgs vev s , @xmath21 , and has the right order of magnitude to fit the sbl excesses .
the second relevant operator is dimension - six , which is @xmath16-conserving and @xmath19-breaking .
it is responsible for coupling the majoron to neutrinos and additionally for providing mass to the light active neutrinos .
we will show that after satisfying various constraints , the three heavy sterile neutrinos can decay into active neutrinos plus one majoron with the decay distance within the experimental lengths . in what follows
, we will analyze oscillation and decay of dirac and majorana sterile neutrinos within the context of this 3 + 3 twin neutrino " model .
we will show that promptly decaying sterile neutrinos in this model can provide a good fit to the lsnd and miniboone anomalies .
motivated by the twin higgs model , we consider a global non - abelian @xmath19 symmetry in the electroweak parts of both the sm and twin sectors .
the two higgs doublets @xmath22 and @xmath23 , which transform under @xmath24 and @xmath25 gauge symmetries , can be grouped together as a quadruplet of @xmath19 : @xmath26 . at the minimum of its @xmath19 invariant potential ,
the quadruplet develops a vev of @xmath27 , spontaneously breaking @xmath19 down to its @xmath28 subgroup . as a result , there are seven nambu - goldstone - bosons ( ngb s ) in the low energy theory below the cutoff @xmath29 . turning on electroweak gauge interactions in both sectors
, the quadruplet vev breaks the twin electroweak gauge symmetry @xmath30 to a single @xmath31 with three ngb s eaten by the three massive gauge bosons @xmath32 and @xmath33 .
the remaining four ngb s can be identified as the sm higgs doublet , @xmath34 , which acquire mass and become pngb s after turning on @xmath19 breaking gauge or yukawa interactions .
the little hierarchy problem can be alleviated by imposing an additional @xmath35 symmetry between the two sectors which forces all couplings to be the same .
this is because in the gauge sector , the one loop corrections to the higgs mass which are quadratic in @xmath36 have the form @xmath37 and are independent of the pngb higgs field . in addition , the logarithmically divergent part contributes to the coefficient of the @xmath19-breaking operator @xmath38 at the order of @xmath39 , so the higgs field mass is generically suppressed compared to the vev @xmath40 tev . to obtain the lighter higgs boson mass at 125 gev
, the coefficient is needed to be around 1/4 , which suggests additional @xmath19 breaking terms in the scalar potential . minimizing the potential for the two higgs doublets with small @xmath16-breaking terms , the ratio of the two vev s is = ~o(0.1 ) , with the electroweak vev , @xmath41 gev .
later we will show that this ratio will be crucial to determine the fermion mass spectrum in the twin sector .
the fermion yukawa couplings explicitly break the global @xmath42 ) symmetry .
the @xmath16 twin parity is required to ensure that there are no corrections to the sm higgs mass proportional to @xmath43 at the one - loop level .
therefore , we keep the yukawa couplings in both sectors to be exactly the same , both for charged leptons and neutrinos . for the charged leptons , the @xmath16-invariant and
@xmath19-breaking yukawa couplings in the lagrangian are y_e^ij ( h_a _ a i e_a j + h_b _ b
i e_b j ) + . here
, the indexes @xmath44 " denote lepton flavors .
after inputting the scalar vev s , we have the three twin charged - lepton masses exactly proportional to the sm ones : @xmath45 .
for instance , the twin electron is anticipated to have a mass of @xmath46 . in the neutrino sector
, we will consider both majorana and dirac neutrino cases and will only focus on the spectrum with normal ordering , which is preferred for the majorana case . in this subsection
, we first study the majorana neutrinos for both sm and twin sectors .
different from the charged - lepton sector , we can have the following @xmath16-invariant and @xmath19-conserving majorana mass operators l^t_i c
^t l_j + [ eq : majorana - mass - operator-1 ] here , @xmath47 and @xmath48 with @xmath49 . the cutoff , @xmath50 ,
could be related to the some heavy right - handed neutrino masses to realize the see - saw mechanism .
after higgs doublets get their vev s , the linear combinations @xmath51 are massive and are approximately the three sterile neutrino states . to provide mass for other combinations , we introduce the following @xmath16-invariant and @xmath19-breaking dimension - six operator l^t_ai _ a c ^t_b l_bj + [ eq : majorana - mass - operator-2 ] here
, the new gauge - singlet scalar @xmath52 carries both sm and twin lepton numbers .
furthermore , one could define a discrete symmetry in the twin sector with @xmath53 , @xmath54 and @xmath55 , such that additional operators for @xmath52 coupling to only sm leptons are forbidden .
this discrete symmetry is important for our later discussion of @xmath52-related phenomenology .
the yukawa couplings are chosen to be identical for eq .
( [ eq : majorana - mass - operator-1 ] ) and eq . ( [ eq : majorana - mass - operator-2 ] ) , which could originate from a uv theory at a much higher than tev scale .
after @xmath52 develops a vev with @xmath56 , to be a real number .
its complex phase is physical , but will not change our phenomenological study later . ]
the @xmath57 neutrino mass matrix is m & = & y_^33 + & = & u^t ( m__a^1 , m__a^2 , m__a^3 , m__s^1 , m__s^2 , m__s^3 ) u . in the leading order of @xmath58 and @xmath59 , the three heavy ( sterile ) neutrino masses are m__s^i |y^i _ , where @xmath60 is the eigenvalue of the yukawa matrix @xmath61 . because of flavor alignment in the sm and twin sectors ,
the ratios of the neutrino masses satisfy r = 2 , to the leading order of the small parameter , @xmath58 , in our model . for a normal ordering mass spectrum of active neutrinos , @xmath62 , and from ref .
@xcite , we have @xmath63 ev@xmath64 and @xmath65 ev@xmath64 .
the twin neutrino masses are shown in fig .
[ fig : mass - spectrum ] for two different values of @xmath66 . in the charged - lepton mass - eigenstate basis
, the diagonalization unitary matrix @xmath67 can be written in terms of a tensor product ^66 = o^22 u^33_pmns , where @xmath68 is the pmns matrix in the sm ( the experimental values are taken from ref .
@xcite ) and @xmath69 is a rotation matrix o^22 = ( cc & + - & ) . to the leading order of @xmath70 , the new mixing angle between the sm and twin neutrino sectors is = o(0.1 ) .
[ eq : o - matrix ] consequently , the three active neutrinos ( @xmath71 ) interact with the sm charged leptons with a strength proportional to @xmath72 , while the three sterile neutrinos ( @xmath73 ) have interactions suppressed by @xmath74 .
_ the sterile neutrino interaction strengths with sm charged leptons are therefore related to the fine - tuning problem for the sm higgs boson .
_ in this model , there is a pngb or majoron @xcite associated with the global symmetry breaking of @xmath75 .
the relevant yukawa couplings for the majoron particle , @xmath13 , defined as @xmath76 , in our models is flavor diagonal and are _ a s^i i _ a , l^it c _ s , l^i & & |y_^i i_a , l^it c _ s , l^i , + _ a a^i i _ a , l^it c _ a , l^i & & |y_^i i_a , l^it c _ a , l^i .
[ eq : transition - yukawa ] the sterile neutrinos are not stable particles and have the decay widths of & = & 2 + & & = . in our numerical study
later , we will focus on the parameter region of @xmath77 or @xmath78 and @xmath79 . since the mass operators in eq .
( [ eq : majorana - mass - operator-1 ] ) explicitly break the @xmath80 symmetry , we anticipate a non - zero mass for the majoron filed , @xmath13 .
the one - loop diagram mediated by @xmath81 and @xmath82 generates a mass for @xmath13 m^2_&~ & ( y_y_^y_y_^ ) + & ~ & ( ^3_as ) ^2 ^2m__s^3 ^ 2 . for a normal ordering neutrino mass spectrum
, we have @xmath83 ev for @xmath84 kev , @xmath85 and @xmath86 .
the majoron decay width is _
i ( 2 _ a^i ) = _ i m _
( m_- 2 m__a^i ) . for a normal ordering neutrino mass spectrum , @xmath83 ev , @xmath86 and @xmath87 ,
the decay width at rest is @xmath88 ev . for the dirac neutrino case , one need to introduce an additional set of right - handed neutrinos in both the sm and twin sectors .
the @xmath35-invariant and @xmath19-conserving dirac mass operators are y_^ij ( _ a , r + _ b , r ) + furthermore , the following @xmath35 and @xmath19-breaking dimension - five operator is introduced to provide light neutrino masses and decay couplings for the heavy neutrinos _
a _ a _ b , r + after @xmath52 develops a vev with @xmath56 , the @xmath57 neutrino mass matrix is m & = & y_^33 + & = & u^t ( m__a^1 , m__a^2 , m__a^3 , m__s^1 , m__s^2 , m__s^3 ) w . with the left - handed rotation matrix as ^66 = o^22_l
u^33_pmns . using the same parametrization in eq . ( [ eq : o - matrix ] ) and in the limit of @xmath89 , we still have @xmath90 . the mass ratios for this dirac neutrino model is r = . in the dirac neutrino model
, we also have a pngb associated with the symmetry breaking of @xmath91 .
the couplings of the majoron , @xmath13 , parametrized by @xmath76 are ^i_as i ^i_s , r & & |y^i _
i ^i_s , r , + ^i_aa i ^i_a , r & & |y^i _
i ^i_a , r .
the sterile neutrino decay widths are m__s^i = . in our model ,
the active neutrinos from sterile neutrino decays are left - handed .
for both the majorana and dirac neutrino models , we have the model parameters : @xmath92 , @xmath93 ( related to the coupling @xmath94 ) , @xmath20 and @xmath66 . in this section ,
we study the existing constraints on our model parameters from unitarity of the active neutrino mixing matrix , neutrinoless double beta decay and meson decays .
the mixing between active and twin neutrinos reduces the couplings of active neutrinos to charged leptons in the sm .
the @xmath57 mixing matrix , @xmath95 , is a unitary matrix in our model , but the @xmath96 mixing matrix , @xmath97 , is not unitary and has the normalization property of @xmath98 . using the results in ref .
@xcite and neglecting the effects of sterile decay products , we have found that 0.20 , at @xmath99 c.l .
there are additional bounds on the sterile neutrino decay widths from the pion and kaon three - body decay into the new majoron state and the electron - muon universality tests of their total leptonic widths ( see ref .
@xcite for a recent summary ) . using the analysis in ref .
@xcite and the measurement of @xmath100 branching ratio @xcite , the predicted deviation from @xmath101 universality is @xmath102 with the experimental value of @xmath103 ( the updated value of @xmath104 provides a similar bound ) , so the bound on our model parameters is ( g^2)_ee = c_i=1 ^ 3 u_e i u^t_i e < 3.0 10 ^ -5 , at 90% c.l . here , @xmath105 for the majorana(dirac ) neutrino case . in fig .
[ fig : constraints - meson - decay ] , we show the constraints on our model parameters in the @xmath106 and @xmath92 plane .
universality of pion decays .
the dirac phase @xmath107.,scaledwidth=45.0% ] for the majorana neutrino case , searches for the neutrinoless double beta decay @xmath108 can also impose bounds on our model parameter space . the amplitude of @xmath108 is proportional to the effective majorana mass |m _
i=1 ^ 6 m_iu^2_ei | = ( r ^2 + ^2 ) | _
i=1 ^ 3 m_3+iu^2_ei | .
the @xmath109-violating phases can affect the predicted effective majorana mass . in our model ,
we have identical majorana and dirac phases for the sm and twin sectors . in fig .
[ fig : constraints - neutrinoless ] , we show the allowed parameter space by allowing arbitrary majorana phases but fixed dirac phase of @xmath107 . and
@xmath74 from @xmath110 decay with @xmath111 ev @xcite .
we have chosen a normal ordering mass spectrum for the three active neutrinos and the dirac phase @xmath107.,scaledwidth=45.0% ] we did not find additional processes like @xmath112 or @xmath113 provide more stringent bounds than from @xmath110 and meson decays .
because of the fairly large interaction strength ( @xmath114 ) among sterile neutrinos , active neutrinos and the majoron particle , both sterile neutrinos and the majoron particles stay within the supernova core .
our model does not have an additional energy loss problem for supernova sn1987a .
for sterile neutrinos with a mass below around 1 mev , they contribute to additional relativistic degrees of freedom and are constrained from bbn @xcite .
however , to have a conclusive statement , additional analysis or non - standard cosmology should be taken into account .
we do not explore these directions here .
in this section , we write down general neutrino appearance and disappearance formulas for our model .
since we have both oscillation and decay , we will keep both effects for neutrino appearance and disappearance . for the short - baseline experiments ,
the differential probability for a neutrino of flavor @xmath115 with energy @xmath116 converting into a neutrino of flavor @xmath117 with energy in the interval of @xmath118 is @xcite & = & | _
i=1 ^ 6 u_i u^*_i e^ - i - |^2 + & & ( e__-e _ _ ) + w_e__e _ _ + & & ^l_0 dl^ | _
j=4 ^ 6 u_(j-3 ) u^*_j e^ - i - |^2 .
[ eq : decay - plus - oscillation ] here , the neutrino decay widths , @xmath119 for @xmath120 and @xmath121 for @xmath122 , are defined in the rest frame . @xmath105 for the majorana(dirac ) neutrino case . for the case of neutrino
goes to neutrino , the energy spectrum function is @xmath123 . for the majorana model ,
the helicity - flip formula for @xmath124 has only the second term ( the decaying part ) with a different energy spectrum function , @xmath125 . for the case with initial anti - neutrinos
, one should replace the elements of @xmath67 by their complex conjugates . in our numerical studies
, we will focus on the pure oscillation case as well as the case where the decay effect dominates . for the pure oscillation case with @xmath126 and in the limits of @xmath127 and @xmath128 ( the short - baseline approximation ) ,
the neutrino appearance probability is p(__)&= & 4 ^4 _
k=1^j-1 |u_ju^*_j u^*_k u_k | + & & , [ eq : oscillation - app ] for @xmath129 . here ,
the phase @xmath130 .
the formula for the anti - neutrino case can be obtained by a replacement of @xmath131 . in our model ,
only a single dirac @xmath109 phase enters both sectors . in the small mixing angle limit of @xmath132 and @xmath133 order of unity , we have @xmath134 , @xmath135 .
so , a large @xmath109-violating phase can affect the ( anti-)neutrino appearance probabilities .
the disappearance probability is independent of @xmath109-violating phase and has p ( _ _ ) = 1 - ^2(2 ) _
j=1 ^ 3 |u_j|^2 ^2 ( ) .
[ eq : oscillation - disapp ] comparing eqs .
( [ eq : oscillation - app ] ) and ( [ eq : oscillation - disapp ] ) , one can see that the appearance probability is suppressed by @xmath136 , while the disappearance is only suppressed by @xmath137 .
we will later show because of this fact it is challenging to only use oscillation to explain lsnd and miniboone anomalies .
for the case in which sterile neutrino decay effects are dominant and in the limit of @xmath138 , the appearance probability is & = & ^2w_e__e _ _ + & & _ j=1 ^ 3 |u_j|^2 |u_j|^2 ( 1 - e^- ) + o(^4 ) .
[ eq : decay - appearance ] for the majorana model , the helicity - flip formula for @xmath124 is similar but using @xmath139 .
the disappearance has contributions from both terms in eq .
( [ eq : decay - plus - oscillation ] ) and is & = & + & & ( e _ _ - e__^ ) + & & + w_e__e__^ ^2 _
j=1 ^ 3 |u_j|^4 ( 1 - e^- ) , ignoring additional terms suppressed by @xmath136 .
one can see that for the decay case both appearance and disappearance are suppressed by the same power of @xmath74 .
this fact will make the decay model a better fit to lsnd and miniboone data .
in this section , we will consider both cases for interpreting short - baseline experimental data with or without sterile neutrino decays .
for the first case without sterile neutrino decays , the energy spectrum of the neutrinos near the far detector follows the initial injected neutrino spectrum . on the other hand , for the second case with
sterile neutrino decays , additional care should be taken to account for the energy spectrum change . to interpret the lsnd event excess of anti - neutrino appearance , @xmath140 @xcite
( we will ignore the less significant excess in the neutrino appearance observed by lsnd @xcite ) , and to derive the preferred model parameter space , we follow ref .
@xcite to account the energy spectrum as well as the conversion of the neutrino energy spectrum to measured positron energy spectrum , @xmath141 , from the inverse neutron decay process @xmath142 .
the measured positron energy has the range of @xmath143 . for the simplest @xmath144 sterile neutrino model , we have reproduced the lsnd contours in ref .
@xcite . for the miniboone @xmath140 and @xmath145 appearance analysis based on initial anti - neutrino and neutrino fluxes @xcite
, we use the data released by miniboone collaboration @xcite to derive the preferred contours in our model parameter space .
we combine the two data sets in the ( anti-)neutrino energy range of @xmath146 . again , for the simplest @xmath144 sterile neutrino model , we reproduce the contours in the miniboone publication @xcite . for the constraints from ( anti-)neutrino disappearance , we use the combined sciboone and miniboone analysis @xcite for the anti - neutrino disappearance measurement , which provides more stringent constraints than the ones from the neutrino disappearance measurement @xcite .
we use the publicly available data @xcite to constrain our model parameter space .
we have also checked additional constraints from appearance searches by karmen @xcite but found them to be less stringent , and we will not report them here .
is the constraint line from unitarity of the three active neutrinos .
the yellow star is the best fit point of our model.,scaledwidth=45.0% ] in fig .
[ fig : oscillation ] , we show the lsnd and mniboone preferred contours in terms of our model parameters : @xmath74 and @xmath66 by fixing @xmath147 ev and assuming @xmath148 , which provides a better fit than no @xmath109 violation with @xmath107 . from a three - dimensional parameter scan , we have found the point with the smallest @xmath149 at @xmath150 , @xmath151 and @xmath152 ev .
this is an improvement by @xmath153 compared to the no oscillation fit .
unfortunately , the disappearance constraints from sciboone and miniboone exclude all the @xmath154 appearance - data - preferred region . furthermore , the constraints from unitarity of the three active neutrinos also exclude the lsnd and mniboone preferred contours .
this can be understood by the formulas in section [ sec : neutrino - formula ] , which show that the appearance probabilities for the pure oscillation case have an additional @xmath137 with respect to the disappearance probabilities .
the tension for appearance and disappearance data is a general feature even for a general @xmath155 global fit @xcite . for the second case
with the decay effects dominant , we need to know more information about the detectors
. the first important piece of information is to know whether the sterile neutrinos generated at the source location have already decayed or not .
it can be seen from eq .
( [ eq : decay - appearance ] ) that in order to have a larger appearance probability , it is preferable to have all sterile neutrinos decay within the experimental lengths .
taking the lorentz boost into account , this means @xmath156 .
lsnd has the distance range of @xmath157 m and the energy range of @xmath158 mev ; miniboone has the distance around 540 m and the energy range of @xmath159 mev ; sciboone has the distance around 100 m and the energy range of @xmath160 mev for the disappearance analysis @xcite .
altogether , if @xmath161 ev@xmath64 , the majority of sterile neutrinos have already decayed before reaching the detector .
we also note that for large values of @xmath162 , all sterile neutrinos decay promptly and only @xmath20 and @xmath162 are relevant parameters for both appearance and disappearance experiments . both lsnd and miniboone are able to generate @xmath9 or @xmath163 initial fluxes . on the other hand , for the appeared electron neutrino or anti - neutrinos in the detectors ,
the lsnd detectors are different from miniboone and sciboone .
the lsnd experiment can distinguish @xmath164 and @xmath11 because after @xmath164 interacts via inverse beta decay in the mineral oil target of lsnd , both a prompt positron and a correlated 2.2 mev photon from neutron capture appear .
this twofold signature is not true for @xmath11 . on the other hand
, the miniboone and sciboone can not distinguish @xmath165 and @xmath166 . in order to compare to the miniboone and sciboone data ,
we need to add both @xmath165 and @xmath166 .
furthermore , we also note that the quasi - elastic cross sections for @xmath165 and @xmath166 interacting with ch@xmath167 in miniboone are different . using the cross sections in genie @xcite and for the relevant energy range of @xmath159 mev at miniboone , we have ^quasi - elastic_+ _ 2 > ^quasi - elastic _ + _ 2 , [ eq : quasi - elastic ] which is mainly due to @xmath168 from a negative axial - vector form factor @xcite
. we will show later that because of different scattering cross sections the majorana and dirac models have different features for fitting appearance data .
before we present our results , we also want to comment on the fact that the lsnd or the miniboone measured neutrino transition probability is not simply eq .
( [ eq : decay - plus - oscillation ] ) for the sterile neutrino decay case .
they have the measured probability for each energy bin to be @xmath169 @xcite . for the full oscillation model , their measured probability matches to the theoretical @xmath170 oscillation probability . for the decay case at hand ,
the energy spectra of the initial neutrino flux at the source and the final flux at the detector are different .
to compare to the lsnd and miniboone @xmath170 data plots , we use the following probability p^exp._model ( _ _ + _ ) ( e _ _ ) & = & + & & .
[ eq : exp - probability ] here , the symbol
@xmath171 " means function convolution . for the majorana model
, we have @xmath172 for lsnd and @xmath173 for miniboone . for the dirac model
, we have @xmath172 for both lsnd and miniboone . for the initial anti - neutrino experiments ,
one just makes the interchange of @xmath174 . in deriving the above equation ,
we have also made the quasi - elastic approximation with the outgoing electron(positron ) energy equal to the incoming neutrino(anti - neutrino ) energy .
based on eqs .
( [ eq : quasi - elastic])([eq : exp - probability ] ) , there is an interesting observation about our majorana model prediction for miniboone . under the approximation of the same energy spectra of the initial fluxes , @xmath175 , one has p^ _ ( _ _ + _ ) < p^ _ ( _ _ + _ ) , simply from the fact that the neutrino quasi - elastic cross section is larger than the anti - neutrino one . in fig .
[ fig : decay - lsnd ] , we show a comparison of a benchmark model point with @xmath176 , @xmath177 kev , @xmath178 and @xmath179 ev@xmath64 , to the lsnd data .
the dirac model has probabilities higher than the majorana model simply by a factor of two .
note that the starting points of the model curves are higher than the actual data starting point .
this is due to the approximation of using the shortest distance @xmath180 m and the maximal neutrino energy of @xmath181 mev from muon decays at rest @xcite .
similarly for miniboone , in fig .
[ fig : decay - mb ] we show both majorana and dirac model predictions for the same benchmark model point . as we argued before
, the majorana model has a larger excess for the initial @xmath163 run than the initial @xmath9 run at miniboone . for the dirac model
, there is also some difference between the initial @xmath163 and @xmath9 runs , which comes from the slightly different energy spectra for the initial fluxes . in fig .
[ fig : decay - contour ] and for the majorana model , we show the contour plot for the two most relevant model parameters , @xmath74 and @xmath182 , to fit the miniboone and lsnd appearance data . also shown in this plot
is the 90% c.l .
constraints from miniboone plus sciboone .
different from the pure oscillation case in fig .
[ fig : oscillation ] , the constraint from the disappearance data is not stringent enough to rule out the best fit region for appearance data . for @xmath176 and @xmath177 kev ( allowed from the neutrinoless double beta decay constraints in fig .
[ fig : constraints - neutrinoless ] ) , the smallest @xmath149 to fit both miniboone ( using their monte carlo samples ) and lsnd ( using their energy resolution ) has @xmath183 and @xmath184 ev@xmath64 .
compared to the fit without new physics , the difference is @xmath185 . for two degrees of freedom
this means that the background - only fit has a @xmath149-probability of @xmath186 ( or @xmath187 ) relative to our twin neutrino decay model . and @xmath182 for fixed values of @xmath66 and @xmath92 .
the vertical dashed line at @xmath188 is the constraint line ( its left side is allowed ) from unitarity of the three active neutrinos .
the yellow star is the best fit point of our model , which has @xmath185 compared to the no twin neutrino assumption.,scaledwidth=48.0% ] for the best fit point of the majorana neutrino model , the six neutrino masses are 0.004 ev , 0.0096 ev , 0.050 ev , 4 kev , 9.56 kev , 50.0 kev .
the three sterile neutrino widths are @xmath189 ev , @xmath190 ev and @xmath191 ev . the decay yukawa coupling defined in eq .
( [ eq : transition - yukawa ] ) is @xmath192 . as a result ,
the majoron receives a loop - generated contribution to its mass with a value of around 3 ev , which means that the majoron can decay into all three active neutrinos . for the typical neutrino energy of 30 mev at lsnd ( 500 mev at miniboone )
, the majoron travels far enough that it can be treated as an invisible particle for both experiments . for the dirac neutrino case , we show the allowed parameter region in fig . [
fig : decay - contour - dirac ] for fixed values of @xmath176 and @xmath193 kev .
the best fit point is at @xmath194 and @xmath195 ev@xmath64 with @xmath196 ( a probability of @xmath197 or @xmath198 ) compared to the fit without the three twin neutrinos .
the dirac model provides a slightly worse fit than the majorana model , because the majorana model has a higher model prediction for the anti - neutrino appearance probability than the neutrino one .
we also note that the anti - neutrino disappearance constraint from miniboone plus sciboone for the dirac model is more stringent than for the majorana model .
this is simply due to the larger scattering cross section of neutrinos than anti - neutrinos .
but for the dirac neutrino model .
the best fit point has @xmath196 compared to the no twin neutrino assumption.,scaledwidth=48.0% ]
the twin neutrino scenario is very predictive .
the neutrino mixing matrix is fixed by the usual pmns matrix and the ratio of the higgs vev s in the two sectors .
additionally , the signals observed by lsnd and miniboone dictates either @xmath199 or @xmath200 to be around @xmath201 . in the majorana sterile neutrino decay scenario , the model also requires a normal ordering mass spectrum for the three active neutrinos to avoid the @xmath202 decay constraint .
because the sm and twin sectors are closely related by the twin @xmath35 symmetry , the number of free parameters is greatly reduced .
the next generation neutrino experiments probing the twin sector neutrinos can also provide information about the sm active neutrino sector . combining the experimental information from both sectors ,
it is very likely that we can completely determine all the parameters in the neutrino sector .
the decay sterile neutrino scenario provides a novel explanation to the lsnd and miniboone anomalies . depending on whether neutrinos are dirac or majorana
, one could have either only @xmath11 or both @xmath11 and @xmath164 to be decay products of the twin neutrino components in @xmath9 .
for the majorana case , the miniboone @xmath163 run can have a larger excess than the @xmath9 run .
this is because @xmath11 as a product of the @xmath163 twin partner has a larger scattering cross section than @xmath164 . our explanation for the larger @xmath163 run excess does not require @xmath109 violation , which is necessary for the pure oscillation explanation .
future experiments like microboone @xcite ( see also ref .
@xcite for other interesting proposals ) may provide decisive tests for the lsnd and miniboone excesses .
their results will not just cover our twin neutrino decay scenario , but also the pure oscillation interpretation .
we also note that the icecube collaboration is finalizing their @xmath203(ev ) sterile neutrino searches .
some preliminary results have shown significant improvement on constraining oscillation parameters @xcite .
the icecube bound is based on oscillation effects with a dramatical enhancement of the oscillation amplitude due to matter effects at tev energies .
although our twin neutrino decay scenario is unlikely to be constrained by the icecube search , it is interesting to explore how to search for the three twin neutrinos at large cosmic - ray neutrino experiments . in summary , motivated by the lsnd and miniboone excesses
, we have constructed an interesting neutrino model to link the sterile neutrino phenomenology to the new tev - scale physics associated with electroweak symmetry breaking . _
* acknowledgments . *
_ we would like to thank vernon barger , josh berger and lisa everett for useful discussion and comments .
this work is supported by the u. s. department of energy under the contract de - fg-02 - 95er40896 .
jordi salvado is also supported by fpa2011 - 29678 and fpa2014 - 57816-p .
collaboration , a. aguilar - arevalo et al . ,
_ evidence for neutrino oscillations from the observation of anti - neutrino(electron ) appearance in a anti - neutrino(muon ) beam _ , _ phys . rev . _
* d64 * ( 2001 ) 112007 , [ http://arxiv.org/abs/hep-ex/0104049 [ hep - ex/0104049 ] ] .
collaboration , c. athanassopoulos et al .
, _ evidence for muon - neutrino @xmath204 electron - neutrino oscillations from pion decay in flight neutrinos _ , _ phys .
* c58 * ( 1998 ) 24892511 , [ http://arxiv.org/abs/nucl-ex/9706006[nucl-ex/9706006 ] ] .
collaboration , a. a. aguilar - arevalo et al .
, _ improved search for @xmath205 oscillations in the miniboone experiment _ , _ phys .
* 110 * ( 2013 ) 161801 , [ http://arxiv.org/abs/1207.4809[arxiv:1207.4809 ] ] .
j. m. conrad , w. c. louis , and m. h. shaevitz , _ the lsnd and miniboone oscillation searches at high @xmath206 _ , _ ann .
nucl . part .
_ * 63 * ( 2013 ) 4567 , [ http://arxiv.org/abs/1306.6494 [ arxiv:1306.6494 ] ] .
g. karagiorgi , a. aguilar - arevalo , j. m. conrad , m. h. shaevitz , k. whisnant , m. sorel , and v. barger , _ leptonic cp violation studies at miniboone in the ( 3 + 2 ) sterile neutrino oscillation hypothesis _ , _ phys .
* d75 * ( 2007 ) 013011 , [ http://arxiv.org/abs/hep-ph/0609177 [ hep - ph/0609177 ] ] .
[ erratum : phys .
rev.d80,099902(2009 ) ] . a. e. nelson , _ effects of cp violation from neutral heavy fermions on neutrino oscillations , and the lsnd / miniboone anomalies _ , _ phys . rev .
_ * d84 * ( 2011 ) 053001 , [ http://arxiv.org/abs/1010.3970 [ arxiv:1010.3970 ] ] .
j. m. conrad and m. h. shaevitz , _ limits on electron neutrino disappearance from the karmen and lsnd @xmath11 - carbon cross section data _ , _ phys .
* d85 * ( 2012 ) 013017 , [ http://arxiv.org/abs/1106.5552[arxiv:1106.5552 ] ] .
z. g. berezhiani and r. n. mohapatra , _ reconciling present neutrino puzzles : sterile neutrinos as mirror neutrinos _ , _ phys .
rev . _ * d52 * ( 1995 ) 66076611 , [ http://arxiv.org/abs/hep-ph/9505385 [ hep - ph/9505385 ] ] .
p. guzowski , l. barnes , j. evans , g. karagiorgi , n. mccabe , and s. soldner - rembold , _ combined limit on the neutrino mass from neutrinoless double-@xmath117 decay and constraints on sterile majorana neutrinos _ , _ phys .
* d92 * ( 2015 ) , no . 1 012002 , [ http://arxiv.org/abs/1504.03600[arxiv:1504.03600 ] ] .
v. barger , j. p. kneller , h .- s .
lee , d. marfatia , and g. steigman , _ effective number of neutrinos and baryon asymmetry from bbn and wmap _ , _ phys .
lett . _ * b566 * ( 2003 ) 818 , [ http://arxiv.org/abs/hep-ph/0305075[hep-ph/0305075 ] ] .
collaboration , b. armbruster et al . , _
upper limits for neutrino oscillations muon - anti - neutrino
@xmath204 electron - anti - neutrino from muon decay at rest _ , _ phys .
_ * d65 * ( 2002 ) 112001 , [ http://arxiv.org/abs/hep-ex/0203021[hep-ex/0203021 ] ] . c. andreopoulos ,
c. barry , s. dytman , h. gallagher , t. golan , r. hatcher , g. perdue , and j. yarba , _ the genie neutrino monte carlo generator : physics and user manual _ , http://arxiv.org/abs/1510.05494 [ arxiv:1510.05494 ] .
http://www.genie-mc.org .
b. dutta , y. gao , r. mahapatra , n. mirabolfathi , l. e. strigari , and j. w. walker , _ sensitivity to oscillation with a sterile fourth generation neutrino from ultra - low threshold neutrino - nucleus coherent scattering _ , http://arxiv.org/abs/1511.02834[arxiv:1511.02834 ] . | we construct a neutrino model of three twin neutrinos in light of the neutrino appearance excesses at lsnd and miniboone .
the model , which includes a twin parity , naturally predicts identical lepton yukawa structures in the standard model and the twin sectors . as a result , a universal mixing angle controls all three twin neutrino couplings to the standard model charged leptons
this mixing angle is predicted to be the ratio of the electroweak scale over the composite scale of the higgs boson and has the right order of magnitude to fit the data .
the heavy twin neutrinos decay within the experimental lengths into active neutrinos plus a long - lived majoron and can provide a good fit , at around @xmath0 confidence level , to the lsnd and miniboone appearance data while simultaneously satisfying the disappearance constraints .
for the majorana neutrino case , the fact that neutrinos have a larger scattering cross section than anti - neutrinos provides a natural explanation for miniboone s observation of a larger anti - neutrino appearance excess . |
quantum entanglement plays a vital role in many practical quantum information system , such as quantum teleportation @xcite , quantum dense coding @xcite , and quantum cryptography @xcite .
entangled states of higher - dimensional systems are of great interest owing to the extended possibilities they provide , which including higher information density coding @xcite , stronger violations of local realism @xcite , and more resilience to error @xcite than two dimensional system . over the past few years , fairish attention has been paid to implement higher - dimensional entanglement with trapped ions @xcite , photons @xcite , and cavity qed @xcite .
atoms trapped in separated cavities connected by optical fiber is a good candidate to create distant entanglement @xcite .
the main problems in entangling atoms in these schemes are the decoherence due to leakage of photons from the cavity and fiber mode , and spontaneous radiation of the atoms @xcite . by using the stimulated raman adiabatic passage ( stirap ) @xcite
, our scheme can overcome these problems .
recently , remote entanglement between a single atom and a bose - einstein condensate ( bec ) was experimentally realized @xcite . but the efficiency is very low due to the photon loss . in this paper , we takes both the advantages of cavity qed and stirap in order to create three - dimensional entanglement state between a single @xmath0 atom and a @xmath0 bec at a distance .
the entanglement state can be generated with highly fidelity even in the range that the cavity decay and spontaneous radiation of the atoms are comparable with the atom - cavity coupling .
our scheme is also robust to the variation of atom number in the bec . as a result
, the highly fidelity three - dimensional entanglement state of the bec and atom can be realized base on our proposed scheme .
we consider the situation describe in fig .
1 , where a single @xmath1 atom and a @xmath1 bec are trapped in two distant double - mode optical cavities , which are connected by an optical fiber ( see fig .
the @xmath1 atomic levels and transitions are also depicted in this figure @xcite .
the states @xmath2 , @xmath3 , @xmath4 and @xmath5 correspond to @xmath6 , @xmath7 , @xmath8 of @xmath9 and @xmath10 of @xmath9 , while @xmath11 , @xmath12 and @xmath13 correspond to @xmath6 , @xmath7 and @xmath8 of @xmath14 . the atomic transition @xmath15 of atom in cavity @xmath16 is driven resonantly by a @xmath17-polarized classical field with rabi frequency @xmath18 ; @xmath19 @xmath20 is resonantly coupled to the cavity mode @xmath21 @xmath22 with coupling constant @xmath23 . the atomic transition @xmath24 @xmath25 of bec in cavity @xmath26 is driven resonantly by a @xmath17-polarized classical field with rabi frequency @xmath27 ; @xmath28 @xmath29 is resonantly coupled to the cavity mode @xmath21 @xmath30 with coupling constant @xmath31 . here
we consider bec for a single excitation , the single excitation states is described by the state vectors @xmath32 and @xmath33 @xmath34 , where @xmath35 describe the state of the @xmath36th atom in the bec @xcite .
initially , if the atom and bec are prepared in the state @xmath37 and @xmath38 respectively , and the cavity mode is in the vacuum state . in the rotating wave approximation , the interaction hamiltonian of the bec - cavity system can be written as ( setting @xmath39 ) @xcite @xmath40 in the short fibre limit , the coupling between the cavity fields and the fiber modes can be written as the interaction hamiltonian @xcite @xmath41 \text { , } \label{2}\ ] ] in the interaction picture the total hamiltonian now becomes@xmath42
in this section , we begin to investigate the generation of the three - dimensional entangled state in detail .
the time evolution of the whole system state is governed by the schrdinger equation @xmath43 @xmath44 commutes with @xmath45 so that the excitation number is conserved during the evolution .
the subspace with @xmath46 is spanned by the state vectors @xcite @xmath47 where @xmath48 denotes the field state with @xmath49 ( @xmath50 , @xmath51 ) photons in the @xmath52 polarized mode of cavity @xmath16 , @xmath53 in the @xmath52 polarized mode of cavity @xmath26 , and @xmath54 represents @xmath55 photons in @xmath52 polarized mode of the fiber .
the hamiltonian @xmath45 has the following dark state : @xmath56 \nonumber\\ & -g_{b}(t)\omega_{a}(t)\left [ \left\vert \phi_{11}\right\rangle + \left\vert \phi_{12}\right\rangle \right ] \}\text{,}\label{6}\ ] ] which is the eigenstate of the hamiltonian corresponding to zero eigenvalue . here and in the following @xmath57 , @xmath58 are real , and @xmath59 . under the condition @xmath60
we have@xmath61 \text{,}\label{8}\ ] ]
suppose the initial state of the system is @xmath62 , if we design pulse shapes such that @xmath63 we can adiabatically transfer the initial state @xmath64 to a superposition of @xmath65 , @xmath66 and @xmath67 , i.e. , @xmath68 , which is a product state of the three - dimensional atom - bec entangled state , the cavity mode state , and the fiber mode state .
the pulse shapes and sequence can be designed by an appropriate choice of the parameters .
the coupling rates are chosen such that @xmath69 , @xmath70 , @xmath71 , laser rabi frequencies are chosen as @xmath72 $ ] and @xmath73 + \frac{\omega_{0}}{2}\exp\left [ -\left ( t - t_{0}\right ) ^{2}/200\tau^{2}\right ] $ ] , with @xmath74 being the delay between pulses @xcite . with this choice , conditions ( 5 ) and ( 6 ) .
figure 2 shows the simulation results of the entanglement generation process , where we choose @xmath75 , @xmath76 @xmath77 .
the rabi frequencies of @xmath78 , @xmath79 are shown in fig .
2(a ) . fig .
2(b ) and 2(c ) shows the time evolution of populations . in fig .
2(b ) @xmath80 , @xmath81 , and @xmath82 denote the populations of the states @xmath83 , @xmath66 , and @xmath84 .
fig.2 ( c ) show the time evolution of populations of the states @xmath85 , @xmath86 , @xmath87 , @xmath88 , @xmath89 , @xmath90 , @xmath91 , @xmath92 , @xmath93 , which are almost zero during the whole dynamics .
finally @xmath80 , @xmath81 , and @xmath82 arrive at @xmath94 , which means the successful generation of the 3-dimensional entangled state .
figure 2(d ) shows the error probability defined by @xcite : @xmath95 here @xmath96 is the state obtained by numerical simulation of hamiltonian ( 3 ) and @xmath97 is the dark state defined by eq .
( 6 ) . from the fig .
2(a)-(d ) we conclude that we can prepare the 3d entanglement state between single atom and a bec with high success probability .
to evaluate the performance of our scheme , we now consider the dissipative processes due to spontaneous decay of the atoms from the excited states and the decay of cavity . we assess the effects through the numerical integration of the master equation for the system in the lindblad form .
the master equation for the density matrix of whole system can be expressed as @xcite @xmath98 -\sum_{k = l\text { , } r}[\frac{\kappa_{fk}}{2}\left ( b_{k}^{+}b_{k}\rho-2b_{k}\rho b_{k}^{+}+b_{k}^{+}b_{k}\rho\right ) \nonumber\\ & -\sum_{i = a\text { , } b}\frac{\kappa_{ik}}{2}\left ( a_{ik}^{+}a_{ik}\rho-2a_{ik}^{+}\rho a_{ik}+\rho a_{ik}^{+}a_{ik}\right ) ] \nonumber\\ & -\sum_{j = a\text { , } l\text { , } r}\frac{\gamma_{aa}^{ej}}{2}\left ( \sigma_{e_{0}e_{0}}^{a}\rho-2\sigma_{g_{j}e_{0}}^{a}\rho\sigma_{e_{0}g_{j}}^{a}+\rho\sigma_{e_{0}e_{0}}^{a}\right ) \nonumber\\ & -\sum_{k = l\text { , } r}\sum_{j = k\text { , } 0}\frac{\gamma_{ba}^{e_{k}j}}{2}\left ( \sigma_{e_{k}e_{k}}^{b}\rho-2\sigma_{g_{j}e_{k}}^{b}\rho \sigma_{e_{k}g_{j}}^{b}+\rho\sigma_{e_{k}e_{k}}^{b}\right ) \text{,}\label{11}\ ] ] where @xmath99 and @xmath100 denote the spontaneous decay rates from state @xmath101 to @xmath102 and @xmath103 to @xmath104 , respectively ; @xmath105 and @xmath106 denote the decay rates of cavity fields and fiber modes , respectively ; @xmath107 ( @xmath108 , @xmath109 , @xmath110 , @xmath111 ) are the usual pauli matrices . starting with the initial density matrix @xmath112 , by solving numerically eq .
( 11 ) in the subspace spanned by the vectors ( 5 ) and @xmath113 , @xmath114 .
we plot the fidelity of the entanglement state as a function of the atom decay rate @xmath115 @xmath116 in fig .
3 . in the calculation , for simplicity
we choose @xmath117 @xmath118 , @xmath119 , @xmath120 , the other parameters same as in fig . 2 , and @xmath121 @xmath122 , @xmath123 , @xmath124 , @xmath125 , @xmath126 , @xmath127 ( from the top to the bottom ) . from the fig .
we can see that the entanglement state can be generated with highly fidelity even in the range of @xmath115 , @xmath128 .
it is necessary to briefly discuss the experimental feasibility of our scheme . trapping @xmath1 bec in cavity qed
has been realized in recently experiment @xcite . in this experiment
, the atom number can be selected between @xmath129 and @xmath130 and the relevant cavity qed parameter @xmath131 , @xmath132 , @xmath133 is realizable .
so the condition @xmath115 , @xmath134 can be satisfied with these system parameters for entangling the bec and atom with fidelity larger than @xmath135 .
the bec spatial atomic mode and cavity mode have the overlap @xmath136 , so the coupling strength @xmath137 will decrease with increasing atom number @xmath138 .
we can increase the @xmath79 accordingly to compensate this .
the fidelity versus the atom number @xmath138 of the bec is plotted in fig .
4 with the parameters @xmath139 , and the other parameters same as in fig .
2 . in summary ,
based on the stirap techniques , we propose a scheme to prepare three - dimensional entanglement state between a bec and a atom . in this scheme ,
the atomic spontaneous decay and photon leakage can be efficiently suppressed , since the populations of the excited states of atoms and cavity ( fiber ) modes are almost zero in the whole process .
we also show that this scheme is highly stable to the variation of atom number in the bec . recently , strong atom field coupling for bose
einstein condensates in an optical cavity on a chip @xcite and strong coupling between distant photonic nanocavities @xcite have been experimentally realized .
so our scheme is considered as a promising scheme for realizing entanglement between bec and atom .
* acknowledgments : * this work was supported by the national natural science foundation of china ( grant no .
60677044 ) and by the fundamental research funds for the central universities ( grant no .
. l. chen was also supported in part by the government of china through csc ( grant no .
2009633075 ) .
1 a single @xmath1 atom and a @xmath1 bec are trapped in two distant double - mode optical cavities , which are connected by an optical fiber .
the states @xmath2 , @xmath3 , @xmath4 and @xmath5 correspond to @xmath6 , @xmath7 , @xmath8 of @xmath9 and @xmath10 of @xmath9 , while @xmath11 , @xmath12 and @xmath13 correspond to @xmath6 , @xmath7 and @xmath8 of @xmath14 . the atomic transition @xmath15 of atom in cavity @xmath16 is driven resonantly by a @xmath17-polarized classical field with rabi frequency @xmath18 ; @xmath19 @xmath20 is resonantly coupled to the cavity mode @xmath21 @xmath22 with coupling constant @xmath23 . the atomic transition @xmath24 @xmath25 of bec in cavity @xmath26 is driven resonantly by a @xmath17-polarized classical field with rabi frequency @xmath27 ; @xmath28 @xmath29 is resonantly coupled to the cavity mode @xmath21 @xmath30 with coupling constant @xmath31 .
2 the numerical simulation of hamiltonian ( 3 ) in the entanglement generation process , where we choose @xmath75 , @xmath76 @xmath77 . figure 2(a ) : the rabi frequency of @xmath78 , @xmath79 .
2(b ) : the time evolution of populations of the states @xmath64 , @xmath66 , and @xmath84 is denoted by @xmath80 , @xmath81 , and @xmath82 respectively
. fig.2 ( c ) : time evolution of populations of the states @xmath85 , @xmath140 , @xmath87 , @xmath141 , @xmath89 , @xmath142 , @xmath91 , @xmath143 , @xmath93 are almost zero during the whole dynamics .
figure 2(d ) : error probability @xmath144 defined by eq .
( 6 ) . fig
. 3 fidelity of the entanglement state as a function of the atom decay rate @xmath115 ( obtained by numerical simulation of master equation ( 8) ) and the atom decay rate @xmath145 , @xmath123 , @xmath124 , @xmath125 , @xmath126 , @xmath127 ( from the top to the bottom ) . | we propose a robust scheme to prepare three - dimensional entanglement states between a single atom and a bose - einstein condensate ( bec ) via stimulated raman adiabatic passage ( stirap ) techniques .
the atomic spontaneous radiation , the cavity decay , and the fiber loss are efficiently suppressed by the engineering adiabatic passage .
our scheme is also robust to the variation of atom number in the bec . |
electrical impedance tomography ( eit ) is a recently developed non - invasive imaging technique , where the inner structure of a reference object can be recovered from the current and voltage measurements on the object s surface .
it is fast , inexpensive , portable and requires no ionizing radiation . for these reasons
, eit qualifies for continuous real time visualization right at the bedside . in clinical eit applications ,
the reconstructed images are usually obtained by minimizing the linearized - data - fit residuum @xcite .
these algorithms are fast and simple . however , to the best of the authors knowledge , there is no rigorous global convergence results that have been proved so far .
moreover , the reconstructed images usually tend to contain ringing artifacts .
recently , seo and one of the author have shown in @xcite that a single linearized step can give the correct shape of the conductivity contrast .
this result raises a question that whether to regularize the linearized - data - fit functional such that the corresponding minimizer yields a good approximation of the conductivity contrast .
an affirmative answer has been proved in @xcite for the continuum boundary data . in the present paper
, we shall apply this new algorithm to the real electrode setting and test with standard phantom experiment data .
numerical results later on show that this new algorithm helps to improve the quality of the reconstructed images as well as reduce the ringing artifacts .
it is worth to mention that our new algorithm is non - iterative , hence , it does not depend on an initial guess and does not require expensive computation .
other non - iterative algorithms , for example , the factorization method @xcite and the monotonicity - based method @xcite , on the other hand , are much more sensitive to measurement errors than our new algorithm when phantom data or real data are applied @xcite .
the paper is organized as follows . in section [ sec : setting ] we introduce the mathematical setting , describe how the measured data can be collected and set up a link between the mathematical setting and the measured data .
section [ sec : algorithm ] presents our new algorithm and the numerical results were shown in section [ sec : num ] .
we conclude this paper with a brief discussion in section
let @xmath0 describe the imaging subject and @xmath1 be the unknown conductivity distribution inside @xmath2 .
we assume that @xmath2 is a bounded domain with smooth boundary @xmath3 and that the function @xmath4 is real - valued , strictly positive and bounded .
electrical impedance tomography ( eit ) aims at recovering @xmath4 using voltage and current measurements on the boundary of @xmath2 .
there are several ways to inject currents and measure voltages .
we shall follow the _ neighboring method _ ( aka adjacent method ) which was suggested by brown and segar in 1987 @xcite and is still widely being used by practitioners . in this method ,
electrodes are attached on the object s surface , and an electrical current is applied through a pair of adjacent electrodes whilst the voltage is measured on all other pairs of adjacent electrodes excluding those pairs containing at least one electrode with injected current .
figure [ fig : m1 ] illustrates the first and second current patterns for a @xmath5-electrode eit system . at the first current pattern ( figure [ fig : m1]a ) ,
small currents of intensity @xmath6 and @xmath7 are applied through electrodes @xmath8 and @xmath9 respectively , and the voltage differences @xmath10 are measured successively on electrode pairs @xmath11 . in general , for a @xmath12-electrode eit system , at the @xmath13-th current pattern , by injecting currents @xmath14 and @xmath15 to electrodes @xmath16 and @xmath17 respectively , one gets @xmath18 voltage measurements @xmath19 , where @xmath20 and @xmath21 . note that here and throughout the paper , the electrode index is always considered modulo @xmath12 , i.e. the index @xmath22 also refers to the first electrode , etc .
= [ circle , draw , inner sep=0mm , minimum size=0.5cm , font= ] \(c ) at ( , ) ; ( , ) circle ( ) ; ( omega ) at ( -3 , 4 ) @xmath2 ; ( text ) at ( -5 , 6 ) a ) ; in 0,22.5,45,67.5,90,112.5,135,157.5,180,202.5,225,247.5,270,292.5,315,337.5 ( , ) (,)(,)(,)cycle ; /in 0/e_1 , 22.5/e_2 , 45/e_3 , 67.5/e_4 , 90/e_5 , 112.5/e_6 , 135/e_7 , 157.5/e_8 , 180/e_9 , 202.5/e_10 , 225/e_11 , 247.5/e_12 , 270/e_13 , 292.5/e_14 , 315/e_15 , 337.5/e_16 ( t ) at ( @xmath23 ) ; ( t ) node ; ( -3.5 , 5 ) to [ out=200 , in=120 ] ( -2.2,4.3 ) to [ out=20 , in=50 ] ( -3.5,5 ) ; ( -3.5 , 3 ) to [ out=20 , in=10 ] ( -2.5,3.8 ) to [ out=200 , in=150 ] ( -3.5,3 ) ; in 0,22.5,45,67.5,90,112.5,135,157.5,180,202.5,225,247.5,270,292.5,315,337.5 ( , )
( , ) ; ( , ) circle(0.5 mm ) ; ( , ) ( , ) ; ( , ) ( , ) ; ( n0 ) at ( , ) [ place ] @xmath24 ; ( , ) ( n0 ) ( n0 ) ( , ) ; in 45,67.5,90,112.5,135,157.5,180,202.5,225,247.5,270,292.5,315 \(n ) at ( , ) [ place ] @xmath25 ; ( , ) ( n ) ( n ) ( , ) ; ( cr ) at ( , ) ; ( , ) circle ( ) ; ( omegar ) at ( , ) @xmath2 ; ( text ) at ( 0 , 6 ) b ) ; in 0,22.5,45,67.5,90,112.5,135,157.5,180,202.5,225,247.5,270,292.5,315,337.5 ( , ) (,)(,)(,)cycle ; /in 0/e_1 , 22.5/e_2 , 45/e_3 , 67.5/e_4 , 90/e_5 , 112.5/e_6 , 135/e_7 , 157.5/e_8 , 180/e_9 , 202.5/e_10 , 225/e_11 , 247.5/e_12 , 270/e_13 , 292.5/e_14 , 315/e_15 , 337.5/e_16 ( tr ) at ( @xmath26 ) ; ( tr ) node ; ( 1.5 , 5 ) to [ out=200 , in=120 ] ( 2.8,4.3 ) to [ out=20 , in=50 ] ( 1.5,5 ) ; ( 1.5 , 3 ) to [ out=20 , in=10 ] ( 2.5,3.8 ) to [ out=200 , in=150 ] ( 1.5,3 ) ; in 0,22.5,45,67.5,90,112.5,135,157.5,180,202.5,225,247.5,270,292.5,315,337.5 ( , ) ( , ) ; ( , ) circle(0.5 mm ) ; ( , )
( , ) ; ( , ) ( , ) ; ( n0 ) at ( , ) [ place ] @xmath27 ; ( , ) ( n0 ) ( n0 ) ( , ) ; in 67.5,90,112.5,135,157.5,180,202.5,225,247.5,270,292.5,315,337.5 \(n ) at ( , ) [ place ] @xmath28 ; ( , ) ( n ) ( n ) ( , ) ; assuming that the electrodes @xmath29 are relatively open and connected subsets of @xmath3 , that they are perfectly conducting and that contact impedances are negligible , the resulting electric potential @xmath30 at the @xmath13-th current pattern obeys the following mathematical model ( the so - called _ shunt model _
@xcite ) : @xmath31 here @xmath32 is the unit normal vector on @xmath33 pointing outward and @xmath34 describes the @xmath13-th applied current pattern where a current of strength @xmath35 is driven through the @xmath13-th and @xmath36-th electrode . notice
that @xmath37 satisfy the conservation of charge @xmath38 , and that the electric potential @xmath30 is uniquely determined by ( [ eqn : shunt ] ) only up to the addition of a constant .
the voltage measurements are given by @xmath39 the herein used shunt model ignores the effect of contact impedances between the electrodes and the imaging domain .
this is only valid when voltages are measured on small ( see @xcite ) and current - free electrodes , so that ( [ eqn : shunt ] ) correctly models only the measurements @xmath40 with @xmath41 . for difference measurements ,
the missing elements @xmath42 with @xmath43 , on the other hand , can be calculated by interpolation taking into account reciprocity , conservation of voltages and the geometry - specific smoothness of difference eit data , cf .
@xcite . for an imaging subject with unknown conductivity @xmath4 , one thus obtains a full matrix of measurements @xmath44 .
in difference eit , the measurements @xmath45 are compared with measurements @xmath46 for some reference conductivity distribution @xmath47 in order to reconstruct the conductivity difference @xmath48 .
this is usually done by a single linearization step @xmath49 where @xmath50 is the frchet derivative of the voltage measurements @xmath51 we discretize the reference domain @xmath52 into @xmath53 disjoint open pixels @xmath54 and make the piecewise - constant ansatz @xmath55 this approach leads to the linear equation @xmath56 where @xmath57 and the columns of the _ sensitivity matrix _
@xmath58 contain the entries of the measurements @xmath59 and the discretized frchet derivative , resp .
, written as long vectors , i.e. , @xmath60 most practically used eit algorithms are based on solving a regularized variant of ( [ eqn : one_step_linearized ] ) to obtain an approximation @xmath61 to the conductivity difference @xmath48 .
the popular algorithms noser @xcite and greit @xcite use ( generalized ) tikhonov regularization and minimize @xmath62 with ( heuristically chosen ) weighted euclidian norms @xmath63 and @xmath64 in the residuum and penalty term .
it has been shown in @xcite that shape information in eit is invariant under linearization .
thus one - step linearization methods are principally capable of reconstructing the correct ( outer ) support of the conductivity difference even though they ignore the non - linearity of the eit measurement process . in @xcite the authors developed a monotonicity - based regularization method for the linearized eit equation for which ( in the continuum model ) it can be guaranteed that the regularized solutions converge against a function that shows the correct outer shape . in this section ,
we formulate and analyze this new method for real electrode measurements , and in the next section we will apply it to real data from a phantom experiment and compare it with the greit method .
the main idea of monotonicity - based regularization is to minimize the residual of the linearized equation ( [ eqn : one_step_linearized ] ) @xmath65 with constraints on the entries of @xmath66 that are obtained from monotonicity tests . for the following ,
we assume that the background is homogeneous and that all anomalies are more conductive , or all anomalies are less conductive than the background , i.e. , @xmath47 is constant , and either @xmath67 @xmath68 is an open set denoting the conductivity anomalies , and @xmath69 is the contrast of the anomalies .
we furthermore assume that we are given a lower bound @xmath70 of the anomaly contrast , i.e. @xmath71 . for the monotonicity tests it is crucial to consider the measurements and the columns of the sensitivity matrix @xmath58 as matrices and compare them in terms of matrix definiteness , cf .
@xcite for the origins of this sensitivity matrix based approach .
let @xmath72 denote the eit difference measurements written as @xmath73-matrix , and @xmath74 denote the @xmath13-th column of the sensitivity matrix written as @xmath73-matrix , i.e. the @xmath75-th entry of @xmath76 is given by @xmath77 we then define for each pixel @xmath78 @xmath79 where @xmath80 denotes the matrix absolute value of @xmath81 , and the comparison @xmath82 is to be understood in the sense of matrix definiteness , i.e. @xmath82 holds if and only if all eigenvalues of @xmath83 are non - negative .
following @xcite we then solve the linearized eit equation ( [ eqn : one_step_linearized ] ) using the monotonicity constraints @xmath84 .
we minimize the euclidean norm of the residuum @xmath85 under the constraints that 1 .
in the case @xmath86 : @xmath87 , and 2 . in the case
@xmath88 : @xmath89 . where @xmath90 , and @xmath91 .
for noisy data @xmath92 with @xmath93 this approach can be regularized by replacing @xmath84 with @xmath94 where @xmath95 is the identity matrix . for the implementation of @xmath96 see section [ sec : num ] . for the continuum model , and under the assumption that @xmath68 has connected complement , the authors @xcite showed that for exact data this monotonicity - constrained minimization of the linearized eit residuum admits a unique solution and that the support of the solution agrees with the anomalies support @xmath68 up to the pixel partition .
moreover , @xcite also shows that for noisy data and using the regularized constraints @xmath96 , minimizers exist and that , for @xmath97 , they converge to the minimizer with the correct support . since practical electrode measurements can be regarded as an approximation to the continuum model , we therefore expect that the above approach will also well approximate the anomaly support for real electrode data . in the continuum model , the constraints @xmath84 will be zero outside the support of the anomaly and positive for each pixel inside the anomaly .
the first property relies on the existence of localized potentials @xcite and is only true in the limit of infinitely many , infinitely small electrodes .
the latter property is however true for any number of electrodes as the following result shows : [ thm : beta_k ] if @xmath98 , then a. in the case @xmath86 the constraint @xmath84 fulfills @xmath99 , and b. in the case @xmath88 the constraint @xmath84 fulfills @xmath100 . if @xmath98 and @xmath86 then @xmath101 and if @xmath98 and @xmath88 then @xmath102 hence , it suffices to show that @xmath103 holds for all @xmath104 that fulfill a. @xmath105 , or b. @xmath106 .
we use the following monotonicity relation from ( * ? ? ?
* lemma 3.1 ) ( see also @xcite for the origin of this estimate ) : for any vector @xmath107 we have that @xmath108 with @xmath109 . if @xmath105 , then @xmath110 which shows that @xmath111 . if @xmath106 , then @xmath112 which shows that @xmath113 .
in this section , we will test our algorithm on the data set ` iirc`@xmath114`data`@xmath114`2006 ` measured by professor eung
je woo s eit research group in korea @xcite . `
iirc ` stands for impedance imaging research center .
the data set ` iirc`@xmath114`data`@xmath114`2006 ` is publicly available as part of the open source software framework eidors @xcite ( electrical impedance and diffused optical reconstruction software ) .
since ` iirc`@xmath114`data`@xmath114`2006 ` is also frequently used in the eidors tutorials , we believe that this is a good benchmark example to test our new algorithm . the data set ` iirc`@xmath114`data`@xmath114`2006 `
was collected using the 16-electrode eit system khu mark1 ( see @xcite for more information of this system ) .
the reference object was a plexiglas tank filled with saline .
the tank was a cylinder of diameter 0.2 m with 0.01 m diameter round electrodes attached on its boundary .
saline was filled to about 0.06 m depth . inside the tank ,
one put a plexiglas rod of diameter 0.02 m .
the conductivity of the saline was 0.15 s / m and the plexiglas rod was basically non - conductive .
data acquisition protocol was adjacent stimulation , adjacent measurement with data acquired on all electrodes . the data set ` iirc`@xmath114`data`@xmath114`2006 `
contains the voltage measurements for both homogeneous and non - homogeneous cases .
measurements for the homogeneous case were obtained when the plexiglas rod was taken away ( reference conductivity in this case is 0.15 s / m ) . in the non - homogeneous case ,
@xmath115 different voltage measurements were measured corresponding to @xmath115 different positions of the plexiglas rod .
eidors @xcite ( electrical impedance and diffused optical reconstruction software ) is an open source software that is widely used to reconstruct images in electrical impedance tomography and diffuse optical tomography . to reconstruct images with eidors ,
one first needs to build an eidors model that fits with the measured data . in this paper
, we shall use the same eidors model described in the eidors tutorial web - page : ` http://eidors3d.sourceforge.net/tutorial/eidors_basics/tutorial110.shtml ` figure [ fig:3 ] shows the reconstructed images of the @xmath116th - inhomogeneous measurements with different regularization parameters using the eidors built - in command ` inv\_solve ` , which follows the algorithm proposed in @xcite .
we emphasize that , figure [ fig:3b ] ( regularization parameter is chosen as @xmath117 by default ) was considered at the eidors tutorial web - page , we show them here again in order to easily compare them with the reconstructed images using our new method later on
. in the eidors model suggested in the eidors tutorial web - page , the reference body was chosen by default as a disk of diameter @xmath118 m and the default reference conductivity was @xmath118 s / m .
however , in the experiment setting , the reference body was a cylinder of diameter @xmath119 m and the reference conductivity was @xmath120 s / m .
hence , an appropriate scaling factor should be applied to the measurements , to make sure that the eidors model fits with these measurements . in the eidors
tutorial web - page , the measurements were scaled by multiplying by a factor @xmath121 . in this paper , to increase the precision of the model , we shall find the best scaling factor that minimizes the error between the measured data and the data generated by the eidors model .
more precisely , let call ` vh ` the measured data for homogeneous case and ` vh\_model ` the homogeneous data generated by the eidors model , the best scaling factor is a minimizer of the following problem @xmath122 for this experiment setting , the best factor is @xmath123 .
from now on , by measured data we always refer to scaled measured data with respect to this best factor .
+ the next step is to recover the missing measurements on the driving electrodes .
we shall follow the result in @xcite to obtain an approximation for these missing measurements using interpolation .
now we are in a position to minimize the problem ( [ eq : min_lin_res ] ) under the linear constraint ( c1 ) or ( c2 ) .
to do this , we need to clarify @xmath124 and @xmath84 in the linear constraints . after scaling ,
the reference conductivity is @xmath125 s / m , and @xmath68 still denotes the plexiglas rod with conductivity @xmath126 s / m .
thus , @xmath127 , @xmath128 and @xmath84 is calculated using ( [ betak ] ) . in practice , there is no way to obtain the exact value of the matrix @xmath81 in ( [ betak ] ) .
indeed , what we know is just the measured data @xmath129 , where @xmath130 denotes the noise level .
when replacing @xmath80 by the noisy version @xmath131 , it may happen that there is no @xmath104 so that the matrix @xmath132 is still positive semi - definite . therefore , instead of using ( [ betak ] )
, we shall calculate @xmath84 from @xmath133 here , @xmath134 represents the identity matrix , and @xmath130 is chosen as the absolute value of the smallest eigenvalue of @xmath92 .
notice that , in the presence of noise , @xmath135 plays the role of the positive semi - definite matrix @xmath80 .
we shall follow the argument in @xcite to calculate @xmath84 .
let @xmath12 be the lower triangular cholesky decomposition matrix of @xmath135 , and let @xmath136 be the smallest eigenvalue of the matrix @xmath137 . since @xmath76 is negative semi - definite ,
so is @xmath137 .
thus , @xmath138 .
arguing in the same manner as in @xcite , we get @xmath139 the minimizer of ( [ eq : min_lin_res ] ) is then obtained using two different approaches : one employs ` cvx ` ( figure [ fig:4a ] ) , a package for specifying and solving convex programs @xcite , the other ( figure [ fig:4b ] ) uses the ` matlab ` built - in function ` quadprog ` ( ` trust - region - reflective ` algorithm ) .
we also show the reconstructed result using the built - in function ` inv\_solve ` of eidors @xcite ( figure [ fig:4c ] ) with the default regularization parameter @xmath117 and with greit algorithm @xcite ( figure [ fig:4d ] ) to see that scaling the measured data with the best scaling factor will improve a little bit the reconstructed image .
notice that reconstructed images are highly affected by the choice of the minimization algorithms , and we will see from figure [ fig:4 ] that the images obtained by ` cvx ` has less artifacts than the others . it is worth to emphasize that although each eidors model is assigned to a default regularization parameter , when using the eidors built - in function ` inv\_solve ` @xcite , in order to obtain a good reconstruction ( figure [ fig:3 ] ) one has to manually choose a regularization parameter , whilst the regularization parameters @xmath140 and @xmath84 in our method are known a - priori provided the information of the conductivity @xmath4 and the reference conductivity @xmath47 exists .
besides , if we manually choose the parameters @xmath141 , we even get much better reconstructed images ( figure [ fig:5 ] ) .
@l*15@l algorithm & runtime ( second ) + ` cvx ` & 839.3892 + ` quadprog ( trust - region - reflective ) ` & 5.4467 + eidors ( ` inv\_solve ` ) & 0.0231 + greit & 0.0120 + last but not least , our new method proves its advantage when there are more than one inclusions ( figure [ fig:6 ] ) .
in this paper , we have presented a new algorithm to reconstruct images in eit in the real electrode setting .
numerical results show that this new algorithm helps to reduce the ringing artifacts in the reconstructed images .
global convergence result of this algorithm has been proved in @xcite for the continuum model . in future works
, we shall prove global convergence result for the shunt model setting as well as reduce the runtime to fit with real - time applications .
the authors thank professor eung
je woo s eit research group for the ` iirc ` phantom data set .
mnm thanks the academy of finland ( finnish center of excellence in inverse problems research ) for financial support of the project number 273979 . during part of the preparation of this work , mnm worked at the department of mathematics of the goethe university frankfurt , germany .
10 a. adler , j. h. arnold , r. bayford , a. borsic , b. brown , p. dixon , t. j. faes , i. frerichs , h. gagnon , y. grber , et al . : a unified approach to 2d linear eit reconstruction of lung images .
, 30(6):s35 , 2009 .
m. grant and s. boyd .
graph implementations for nonsmooth convex programs . in v.
blondel , s. boyd , and h. kimura , editors , _ recent advances in learning and control _ , lecture notes in control and information sciences , pages 95110 .
springer - verlag limited , 2008 . | in electrical impedance tomography , algorithms based on minimizing the linearized - data - fit residuum have been widely used due to their real - time implementation and satisfactory reconstructed images .
however , the resulting images usually tend to contain ringing artifacts . in this work
, we shall minimize the linearized - data - fit functional with respect to a linear constraint defined by the monotonicity relation in the framework of real electrode setting .
numerical results of standard phantom experiment data confirm that this new algorithm improves the quality of the reconstructed images as well as reduce the ringing artifacts . |
proton capture reactions at a very low temperature play an important role in nucleosynthesis process .
most importantly , in explosive nucleosynthesis ( _ e.g. _ an x - ray burst ) , the rapid proton capture ( @xmath0 ) process is responsible for the production of proton - rich isotopes upto mass 100 region . in nature , the proton capture reactions , important for nucleosynthesis , usually involve certain nuclei as targets which are not available on earth or can not be produced in terrestrial laboratories with our present day technology .
therefore , theory remains the sole guide to extract the physics . in our present work ,
we have studied the endpoint of the @xmath0 process in a microscopic approach using a new phenomenological mass formula @xcite . in a similar work ,
schatz _ et al . _
@xcite calculated the endpoint of @xmath0 process using mass values from a finite range droplet model ( frdm)@xcite calculation and proton capture rates from hauser - feshbach code non - smoker@xcite .
we will show that the results of our calculation are different from their observations . in the present work
, we will concentrate only on x - ray burst scenarios , which have typical timescale of 100 seconds and a peak proton flux density of the order of @xmath1 .
this type of burst provides a highly proton - rich environment around the peak temperatures 1 - 2 gk .
we try to look at different models of the x - ray burster and find out the endpoint of the @xmath0 process nucleosynthesis .
when an x - ray burst takes place , a proton - rich high temperature environment , which triggers the @xmath0 process , is created .
the process passes through nuclei near the proton drip line , not available on earth . in regions
far from the stability valley , rates derived from phenomenological calculations may not represent the reality very well , leading to considerable uncertainty in the process .
very often , the reaction rates are varied by a large factor to study their effects . on the other hand , in a microscopic calculation , uncertainty in reaction rates can be reduced and therefore , this approach is expected to give a more accurate result for unknown mass regions . in a previous work@xcite , we have shown that the rates may vary at most by a factor less than two when the cross - section values range over four orders of magnitude .
a microscopic calculation has been performed to evaluate proton capture rates for the nuclei involve in the @xmath0 process in the present work .
we use the spherical optical model to calculate the rates of the relevant reactions . as most of the nuclei involved in the process
lie around the drip line , experimental density information are not available .
hence , theoretical density profiles have been calculated using relativistic mean field ( rmf ) theory . in the present work ,
we use the fsu gold lagrangian density@xcite and solve the rmf equations in the spherical approximation for the nuclei involved in the @xmath0 process . this lagrangian density , containing additional nonlinear terms for the vector isoscalar meson self interaction and an interaction between the isoscalar vector and
the isovector vector mesons , has been found to be very useful in describing nuclear properties throughout the mass table [ see _ e.g. _ bhattacharya @xcite and references therein ] .
the microscopic optical model potential for capture reactions are obtained using effective interactions derived from the nuclear matter calculation in local density approximation , _
i.e. _ by substituting the nuclear matter density with the density distribution of the finite nucleus . in the present work ,
we have constructed the potential by folding the density dependent m3y ( ddm3y)@xcite interaction with densities from rmf approach .
this interaction was extracted from a finite - range energy - independent @xmath2-matrix element of the reid potential by adding a zero - range energy - dependent pseudo - potential and a density - dependent factor .
the interaction at the point @xmath3 is , thus , given by @xmath4 where @xmath5 is the incident energy and @xmath6 , the nuclear density .
the @xmath7 interaction is given by @xmath8 for @xmath9 in @xmath10 , and @xmath11 is the zero range pseudo potential , @xmath12 the density dependent factor @xmath13 has been chosen of the form @xmath14 from the work by chaudhuri@xcite where the constants were obtained from a nuclear matter calculation as @xmath15 = 2.07 and @xmath16 = 1.624 @xmath17 .
we have used this form in our calculation without changing any of the above parameters .
we have also fixed the various parameters and prescriptions in the hauser - feshbach calculation for the relevant mass region by comparing our results to the experimental low energy proton capture cross sections for these nuclei .
our method of calculation for mean field and proton capture rates has been described in our earlier works@xcite in detail .
the computer code talys@xcite has been used for rate calculation .
binding energy of nuclear ground state is one of the most important inputs in the study of astrophysical reactions .
experimental measurements are very difficult to perform in nuclei far from the stability line .
therefore , one has to take recourse to theoretical predictions . though we have used a mean field calculation to extract the nuclear density profiles , no rmf calculation has been able to achieve a prediction of mass values with accuracy sufficient for studying the proton drip line .
in fact , even skyrme hartree - fock calculations can predict the mass values with an root mean square ( rms ) error slightly less than 0.6 mev only .
thus , in the present work , we have obtained the mass values from a newly developed mass formula@xcite .
it uses a purely phenomenological form with empirically fitted parameters and predicts the known mass values of 2140 nuclei with an rms error of 0.376 mev . in view of the success of the formula to predict the proton dripline and @xmath0 process upto mass 80 region@xcite and to predict the peaks in @xmath9 process@xcite quite well ,
it will be interesting to see the effect of this mass formula on @xmath0 process beyond mass 80 region and to the endpoint of the @xmath0 process . in an x - ray burst environment
, a nucleus ( @xmath18 ) may capture a proton to form the nucleus ( @xmath19 ) .
however , this process has to compete with its inverse , _ i.e. _ photodisintegration by emitting a proton at high temperature [ @xmath20 reaction ] .
a negative or a small positive value of the proton separation energy implies that the inverse reaction dominates and the @xmath0 process stalls at that point , the so - called waiting point .
therefore , only a two proton capture process can bridge the waiting point nuclei .
the bridging mechanism has been discussed in standard text books ( for example the book by illiadis@xcite ) . for the @xmath21 reaction
, the rate @xmath22 of the inverse process @xmath23 is related to the the proton capture rate by the reciprocity theorem and is of the form @xcite @xmath24 in the unit of @xmath25 for @xmath21 process .
the forward reaction rate , denoted by @xmath26 , is expressed in @xmath27 .
the temperature @xmath28 is in gk ( @xmath29k ) and @xmath30 is the ground state q - value of the @xmath31 reaction expressed in mev . the normalized partition functions , @xmath32 and @xmath33 , have been obtained from rauscher .@xcite for protons , we use the standard values , @xmath34 and @xmath35 . as evident from the above expression , the q - value , appearing in the exponent , plays a vital role in the whole process .
apart from the above processes , a nucleus can decay by emitting beta particles while , for higher mass isotopes , @xmath36-decay is another probable channel . in this work , the measured half life values for @xmath16-decay
have been taken from the compilation by audi _
et al._@xcite except in the case of @xmath37as , which is taken from the experimental measurements by lpez _ et al._@xcite in absence of experimental data , half life values have been taken from the calculation by mller _
et al._@xcite both for @xmath16- and @xmath36- decay .
taking into account all the above processes , we have constructed a network extended upto a=110 region to study the nucleosynthesis path and relative abundances of the elements at any instant .
the @xmath0 process paths are shown in fig . 1 and fig . 2 for temperatures 1.2 and 1.5 gk , respectively , using a constant proton flux density of @xmath38 , a proton fraction of 0.7 and 100 seconds burst duration . here ,
black lines indicate the path along which the major portion of the total flux flow whereas gray lines indicate the minor paths . the filled boxes in figures indicate the waiting points . as evident from these figures ,
@xmath0 process paths depend on the temperature of the environment to some extent . for example , at t= 1.2 gk , major portions of the @xmath0 process flux at the waiting point nucleus @xmath39ge convert to @xmath40se by two - proton capture process .
less than 1% of the total flux flows through the @xmath16-decay channel and follow the paths showed by gray lines in fig .
1 . in contrast , at temperature t= 1.5 gk ( in fig .
2 ) , the probability of two - proton capture of @xmath39ge gets reduced and the @xmath0 process path bifurcates from the waiting point nucleus almost in an equal proportion .
this suggests that , at 1.2 gk , proton capture by @xmath39ge dominates over its inverse process , i.e. photodisintegration . as the temperature increases , contributions of photodisintegration process increases and therefore a large fraction of the total abundance chooses another path through more stable nuclei such as @xmath37ge , @xmath40as , etc .
similarly , near other waiting points ( @xmath41 @xmath42kr , @xmath43sr etc . )
the abundance flow pattern changes with changing temperature .
it is also evident from above the figures that , beyond mass 80 , temperature change can hardly affect the scenario . as shown in fig . 1 and fig
. 2 , above a=100 region , the @xmath0 process continues through proton capture by in isotopes and @xmath16-decay of sn isotopes . here
, @xmath44 in captures a proton to form @xmath45sn which completely decays to @xmath45 in as @xmath46sb is proton unbound . in turn , @xmath45 in undergoes proton capture and further exhibits @xmath16-decay .
the process of proton capture followed by consecutive @xmath16-decay continues and relative abundances of nuclei decrease as one proceeds towards higher mass region .
ultimately , less than 0.001% of the total flux can reach @xmath47 in , according to our present calculation . according to the calculations of schatz _
et al._@xcite , a significant portion of the @xmath48sn captures a proton to form @xmath47sb , as @xmath47sb has a sufficient positive proton separation energy ( 0.59 mev from frdm@xcite calculation ) .
another proton capture leads to @xmath49te , which instantly undergoes @xmath36-decay to @xmath50sn .
thus , the @xmath0 process ends in the snsbte cycle .
in contrast , our calculation does not go through the snsbte cycle .
it is clear that the snsbte cycle may not occur under two scenarios .
firstly , if @xmath47sb isotope be very loosely bound ( proton separation energy 0.119 mev according to the mass formula@xcite ) , any @xmath47sb that is formed by a proton capture instantly reverts back to @xmath48sn .
second scenario occurs when the proton capture rates are too small for @xmath48sn to initiate a proton capture process . in a previous work@xcite
, we have shown that a small fluctuation in the mass values of waiting point nuclei in mass 60 - 80 region may affect the effective half life and thus , proper knowledge of ground state binding energy is necessary to understand the bridging phenomena of a waiting point nucleus below mass 80 region . however , as we move towards the higher mass region , we find that small variations in binding energy do not affect the @xmath0 process path significantly . taking the rms error into account for the proton separation energy of @xmath47sb @xmath51mev , we have checked whether the proton capture on @xmath48sn can dominate over its inverse process .
we find that the fraction of the initial flux entering into the snsbte - cycle is negligibly small .
we have repeated our calculation with the proton separation energy of @xmath47sb isotope from a recent experiment by elomma _
et al._@xcite , _ viz .
_ 0.428(0.008)mev .
we find that our results remain almost invariant .
repeating the entire calculation with ground state binding energies from frdm@xcite calculation or the duflo - zuker@xcite mass formula do not alter this conclusion .
hence , we see that for a reasonable variation of mass values the @xmath0 process fails to enter into the snsbte - cycle for an x - ray burst of 100 seconds duration .
if we consider an x - ray burst having duration greater than 150 seconds , a very small fraction of total flux , less than 0.001% of the original , enters the cycle .
cm schatz _ et al._@xcite have used capture rates different from ours . in fig .
3 , we have plotted proton capture rates with temperature for the isotopes @xmath48sn and @xmath47sb respectively .
the different values have been indicated as follows : pres - present work and non - smoker - non - smoker@xcite results used by schatz _ et al._@xcite in case of @xmath48sn nuclei , it is evident from fig .
3 that proton capture rates from our calculation are approximately @xmath52 times smaller than the rates from non - smoker calculation .
it seems , the difference is mainly due to the fact that they have used the form of the interaction from jeukenne _
et al._@xcite which is different from our case . for better understanding
, we have repeated the entire calculation using reaction rates from hauser - feshbach code non - smoker@xcite . in this case , considerable fraction of the initial flux ( @xmath53 0.1% ) enters into the snsbte cycle .
thus , we may conclude that the proton capture rates , and not the mass values , determine the endpoint of @xmath0 process . in our opinion
, our approach in fixing the parameters in the reaction calculation by fitting known reaction rates , and extrapolating the calculation to unknown reactions , may be relied upon to predict the endpoint correctly .
there are many x - ray burster models in literature with various density - temperature profiles .
our present goal is to study how different temperature - density profiles can affect the relative abundances of the nuclei produced by @xmath0 process and the endpoint . in fig .
4 , we have plotted the relative abundances of elements versus mass number at 100th second of the burst in various models , as described below .
-0.4 cm -0.35 cm in the first case ( model - i ) , we assume a constant density - temperature framework with temperature @xmath54 gk , density @xmath55 and proton fraction value 0.7 .
model - ii@xcite describes a situation where ignition takes place at a constant density of @xmath56 and the burst reaches a peak temperature @xmath57 = 1.9 gk after 4 seconds while the cooling phase lasts for approximately 200 seconds . in another example , taken from the book by illiadis@xcite ( model - iii ) , nuclear burning starts with temperature and density values of @xmath58 gk and @xmath55 , respectively .
after 4 seconds , a maximum temperature of @xmath57 = 1.36 gk and a minimum density of @xmath59 are achieved .
after 100 seconds , the temperature drops to @xmath60 gk and the density increases to @xmath61 .
for all the above models , it is assumed that the x - ray burst environment is sufficiently proton - rich to maintain the number 0.7 as the constant proton fraction . as a significant amount of proton flux
is consumed during the thermonuclear reaction , another situation may arise , where the proton fraction decreases gradually with time .
such a situation is given in model - iv where the proton fraction decreases to 0.16 after 100 seconds with temperature - density profile same as that of the model - iii .
it is evident from fig .
4 that abundance peaks in all cases are obtained at mass values of 72 , 76 and 80 , as a result of the existence of waiting point nuclei @xmath42kr , @xmath43sr and @xmath62zr , respectively . other peaks ( for example , peaks at mass 85 , 93 and 95 for isotopes @xmath63mo , @xmath64pd , @xmath65cd , respectively ) suggest that the @xmath0 process flux gets accumulated at these points due to very small positive or negative proton separation energies of those isotopes . we have considered the mass number in each case for which the flux amount drops below 1% of the initial flux .
it is evident from fig .
4 that for model - i , a @xmath66 93 is the region above which the @xmath0 process flux fall below the range of our interest , whereas for model - ii , the region is around a @xmath66 95 .
other abundance peaks are observed at masses @xmath6797 and 101 for the isotopes @xmath68cd and @xmath45sn respectively , though , the fractions of the total flux accumulated at those isotopes are less than 0.1% of the initial value . in case of models iii and iv , the locations to be studied are around a @xmath66 93 and 91 respectively . from above observations
, it can be concluded that , for various density - temperature profiles , the @xmath0 process flux falls below a significant amount near mass 90 - 95 .
the observations from fig .
4 suggest that the end points of the @xmath0 process has a rather weak dependence on different x - ray burster models .
the location above which the @xmath0 process flux falls below an insignificant amount is calculated using the microscopic optical model utilizing the densities from the rmf approach and with a new phenomenological mass formula .
present result is compared with result obtained from another existing work@xcite and the reason for these indifferences between the results are discussed .
our results do not significantly depend on the mass models . for different x - ray burster models ,
endpoints are calculated .
this work has been carried out with financial assistance of the ugc sponsored drs programme of the department of physics of the university of calcutta .
chirashree lahiri acknowledges the grant of a fellowship awarded by the ugc .
99 g. gangopadhyay , _ int . j. mod
_ e * 20 * ( 2011 ) 179 .
h. schatz _
( 2001 ) 3471 .
p. mller , j. r. nix , and k. l. kratz , ( 1997 ) 131 .
t. rauscher and f .- k .
thielmann , ( 2000 ) 1 .
t. rauscher and f .- k .
thielmann , ( 2001 ) 47 .
c. lahiri and g. gangopadhyay , _ phys .
_ c * 84 * ( 2011 ) 057601 .
b. g. todd - rutel and j. piekarewicz , ( 2005 ) 122501 .
m. bhattacharya and g. gangopadhyay , c * 77 * ( 2008 ) 047302 .
m. bhattacharya , g. gangopadhyay and s. roy , c * 85 * ( 2012 ) 034312
. a.m. kobos , b.a .
brown , r. lindsay , g.r . and
satchler , a * 425 * ( 1984 ) 205 .
chaudhuri , a * 449 * ( 1986 ) 243 .
chaudhuri , a * 459 * ( 1986 ) 417 .
c. lahiri and g. gangopadhyay , a * 47 * ( 2011 ) 87 .
a. j. koning , s. hilaire , and m. duijvestijn , proceedings of the international conference on nuclear data for science and technology , april 2227 , 2007 , nice , france , edited by o. bersillon , f. gunsing , e. bauge , r. jacqmin , and s. leray ( edp sciences , paris , 2008 ) , pp .
c. lahiri and g. gangopadhyay , _ int
_ e * 20 * ( 2011 ) 2417 . c. lahiri and g. gangopadhyay , to appear in _ int .
_ e ( 2012 ) . c. illiadis , _ nuclear physics of the stars _
( wiley - vch verlag gmbh , weinheim , 2007 ) . t. rauscher and f - k thielemann , ( 2000 ) 1 . g. audi , o. bersillon , j. blachot and a.h .
wapstra , _ nucl .
phys . _ a * 729 * ( 2003 ) 3 .
lpez jimnez _
_ , _ phys . rev .
_ c * 66 * ( 2002 ) 025803 .
v.- v. elomaa _
( 2009 ) 252501 .
j. duflo and a.p .
zuker , _ phys .
_ c * 52 * ( 1995 ) r23 .
j. jeukenne , a. lejeune , and c. mahaux , c * 16 * ( 1977 ) 80 . | densities from relativistic mean field calculations are applied to construct the optical potential and , hence calculate the endpoint of the rapid proton capture ( @xmath0 ) process .
mass values are taken from a new phenomenological mass formula .
endpoints are calculated for different temperature - density profiles of various x - ray bursters .
we find that the @xmath0 process can produce significant quantities of nuclei upto around mass 95 .
our results differ from existing works to some extent . |
the inflation theory is one of the most successful theories of modern cosmology . having a period of very rapidly accelerating expansion
, it can not only solve many theoretical problems in cosmology , such as flatness , horizon , monopole and so on , but also gives the right amount of primordial fluctuations with nearly scale - invariant power spectrum , which fits the data very well in structure formation @xcite .
there are many ways to construct inflation models , one of which is to introduce a scalar field called inflaton " @xmath6 ( see @xcite ) . moreover
, one may expect that inflaton could have non - minimal coupling to ricci scalar @xmath0 .
the most usual coupling form is @xmath7 , which was initially studied for new inflation scenario @xcite and chaotic inflation scenario @xcite .
later on , various models have been taken on with deeply and wildly investigations . with a non - minimal coupling term
, inflation can be easily obtained and an attractor solution is also available @xcite .
perturbations based on non - minimal coupling inflation are discussed in @xcite , where the coupling term may give rise to corrections on power spectrum which can be used to fit the data or constrain the parameters .
non - minimal couplings can be extended to multifields , see @xcite , or kinetic term coupling @xcite .
the constraints from observational data were also performed , e.g. in @xcite , where the authors claimed that for non - minimal coupling chaotic inflation models , a tiny tensor to scalar ratio will be obtained .
other applications of non - minimal coupling inflation include the realization of warm inflation @xcite and the avoidance of the so called @xmath8 " problem @xcite in the framework of string theory @xcite .
one can also see @xcite and also @xcite for comprehensive reviews of non - minimal coupling theories .
the non - gaussianity of the primordial perturbation has been widely acknowledged to be an important probe in the early universe @xcite .
experimentally , more and more accurate data allow us to study the non - linear properties of the fluctuation in cosmic microwave background ( cmb ) and large scale structure ( lss ) @xcite ; theoretically , the redundance of inflation models requires more information than those of linear perturbations only to have them distinguished .
the non - gaussianity of the fluctuations was first considered in @xcite , and it was further shown in @xcite that the canonical single field slow roll inflation can only give rise to negligible amount of non - gaussianity . to get large non - gaussianity people need to find new inflation models , an incomplete list and references of which include : multi - field models @xcite , k - inflation @xcite , dbi - type inflation @xcite , curvaton scenario @xcite , ghost inflation @xcite , warm inflation @xcite , non - bunch - davies vacuum scenario @xcite , bounce scenario @xcite , island cosmology @xcite , loop correction @xcite , non - commutativity @xcite , string gas scenario @xcite , cosmic string @xcite , end - in - inflation " scenario @xcite , ekpyrotic scenario @xcite , vector field @xcite , ho@xmath9ava theories @xcite and so on and so forth . in this note , we investigate the non - gaussianity of inflation driven by a general single field @xmath10 coupling non - minimally to the einstein gravity
. some specific examples of non - gaussianities of non - minimal coupled field has been studied in , e.g. , @xcite and non - gaussianity generated by modified gravity is expected to have effects that can be tested by cmb anisotropies @xcite . by taking a subclass of linear coupling , we calculated various shapes depending on the ratio between two slow roll parameters @xmath2 and @xmath3 , which describe the evolution of cosmic expansion and the non - minimal correction , respectively
. the power spectrum will deviate from scale - invariance due to the existence of non - minimal coupling @xcite , and the shape of the 3-point correlation function are correspondingly affected . in this paper
we find that for different ( red or blue ) tilt of the power spectrum , the shape will include different parts which will obtain different amplitude of non - gaussianities .
however , since we have only calculated up to leading order in the slow - roll parameter , this conclusion has not been so unambiguous yet .
nevertheless , if it can be verified after a complete consideration to all the orders , one can find the relations between 2- and 3-point correlation functions which can be used to constrain non - minimal coupling models .
this will be one of our future works .
this paper is organized as follows : sec.ii briefly reviews the preliminaries and basic equations of the general non - minimal coupling single field inflation .
we study the non - gaussianities of the general non - scalar field with linear coupling in sec .
iii , which is the main part of the paper .
we first study the perturbed action of the system up to 3rd order , and obtained the mode solution at the quadratic level .
after that , we calculate various shapes of the 3-point correlation functions using the mode solution .
we also study their equilateral limit and the relation with slow - roll parameters at their leading order .
in the last part of this section , we present the non - gaussianities of non - minimal coupling chaotic inflation as an explicit example using numerical calculations .
iv is the conclusion and discussions .
to begin with , let s consider the most general action of a single scalar field with non - minimal coupling : [ action ] s = dtd^3x[f(r,)+2p(x , ) ] , where @xmath11 is the kinetic term and the metric @xmath12 $ ] with @xmath13 the scale factor of the universe .
for the background evolution , one can vary the action ( [ action ] ) with respect to the field @xmath6 and the metric @xmath14 to get the equation of motion for @xmath6 : [ eom ] f_+2p_+2(p_xx^x + p_x^)_+2p_x=0 , and the einstein equations : _ = t_^ ( ) , where _ & & f_rg_-__f_r+f_rr_-fg _ , + t_^()&&p_x__+pg _ , with @xmath15 being the covariant derivative with respect to the metric @xmath14 and @xmath16
. the evolution of the universe can be described by the slow roll parameter : [ epsilonh ] _ h - where @xmath17 is the hubble parameter and dot means the time derivative . in the inflation case , we require that @xmath2 be small . moreover , we can define two more parameters following @xcite:&&xp_x+2x^2p_xx , + & & x^2p_xx+x^3p_xxx , which will be used in later parts of the paper . to study non - gaussianities , we adopt the usual convention of using the _ arnowitt - deser - misner _ ( adm ) metric @xcite as follows : [ metric ] ds^2=-n^2dt^2+h_ij(dx^i+n^idt)(dx^j+n^jdt ) , where @xmath18 and @xmath19 are the lapse function and the shift vector , respectively .
it is useful to decompose the action ( [ action ] ) into the 3 + 1 form , saying : [ actionadm ] s&=&dtd^3xn\{f[r^(3)+k_ijk^ij - k^2+_t(k)-_i(kn^i+h^ij_jn ) , ] + & & + 2p(x , ) }
.the extrinsic tensor @xmath20 is defined as : k_ij(_ij-_in_j-_jn_i ) where @xmath21 is the covariant derivative with respect to the metric @xmath22 and its indices can be raised and lowered by @xmath22 .
the contraction @xmath23 .
the three - dimensional ricci scalar @xmath24 is computed from the metric @xmath22 . from this action , we are able to obtain the equations of motion for @xmath25 and @xmath26 ( constraint equations ) as : [ constraint1 ] & & f-2f_r[k_ijk^ij - k^2+_t(k)-_i(kn^i+h^ij_jn)]+(_tf_r - n^i_if_r ) + & & -_j(h^ij_if_r)+2p-2p_xv^2n^-2=0 , [ constraint2 ] _ j(f_rk^ij)-h^ij_j(f_rk)-h^ij_j(n^-1_tf_r)+h^ij_j(n^-1n^l_lf_r)+h^ij_jf_rk - a^-2n^-1p_xv^i=0 , respectively , where @xmath27 . moreover , a gauge choice is needed to eliminate the redundant degrees of freedom . here
we choose the uniform density ( comoving ) gauge , where the perturbations of the scalar field and metric take the following form : [ gauge ] = 0 , h_ij = a^2e^2_ij .we also need to expand the constraining variables @xmath25 and @xmath26 .
it is only needed to expand them up to the @xmath28-th order when we calculate @xmath29-th order perturbation @xcite .
so here we expand them to 1st order as follows : [ expansion ] n=1 + , n_i=_i+_i , ^i_i=0 , where @xmath30 , @xmath31 and @xmath32 are in the first order of @xmath33 .
since it is very complicated to consider the full types of non - minimal coupling theory , in this paper we will only pick up a subclass where the ricci scalar is coupled linearly to the scalar field , saying @xmath34 . this type of non - minimal coupling is often referred to as scalar - tensor theory @xcite or linear coupling @xcite . for the non - linear coupling case , the constraint equations are listed in * appendix a * and the solutions will be postponed to the future studies . considering all the equations from ( [ constraint1 ] ) to ( [ expansion ] ) in the linear coupling case , one can get:&&-a^-2f_r0 ^ 2^i+a^-2(2f_r0h+)^i-2a^-2f_r0^i=0 , + & & 4 + 12hf_r0 - 12h^2f_r0 - 4a^-2f_r0h^2 - 4a^-2f_r0 ^ 2-(12h-6 + 2a^-2 ^ 2)=0 , where @xmath35 is the background value of @xmath36 and for the current case @xmath37 .
the equations above give the specific solution at first order in @xmath38 : ^i&=&0 , = , + & = & + a^2 ^ -2 .
we define @xmath39 , so it will be followed that @xmath40 and @xmath41 .
note that when @xmath42 , the system will return to the minimal coupling case and @xmath43 coincides with the hubble parameter .
it is also convenient to rewrite @xmath30 and @xmath32 in terms of @xmath44 as:[solution ] = , = -+ , = [ 3a^2(1-)^2+a^4 ^ -2e^-2]^-2 .hereafter , we will use this parameter for convenience throughout the paper . expanding the action ( [ actionadm ] ) to 3rd order of @xmath38 and substituting eq .
( [ solution ] ) into the action , one can get the expanded action in each order : [ s0 ] s_0=dtd^3xa^3(f_0 + 2p_0 ) , [ s1 ] s_1=0 , [ s2 ] s_2=dtd^3xae^2\{3(-2 + 1)^2+a^2e^-2+a^-2(-1+)()^2 } , [ s3 ] s_3&=&dtd^3xae^2\{+6(+2 - 1-)+18(-2 + 1+)^2 + 2a^-2(-1)()^2 + & & + 2a^-2()^2 - 4a^-4()^2 + 3a^-4_i_j^j^i - a^-4_i_j^i^j-3a^-4(^2)^2 + & & + a^-4(^2)^2}+dtd^3xa^3\{6 - 2(+2 ) } .
this process is very straightforward , but rather tedious .
the physical meaning of each equation is easy understanding : the 0th order expansion ( [ s0 ] ) is just the background part of the original action ( [ action ] ) where subscript ` 0 ' denotes the background value ; the 1st order expansion ( [ s1 ] ) is just the background equation of motion . the 2nd and 3rd order expansions are only deviate from the gr case due to the difference of the parameter @xmath43 from @xmath45 , which will coincide when @xmath46 .
however , as will be seen below , this deviation causes very different results of non - gaussianity of our case from that of gr .
first of all , let s consider the solution of the 2nd order action ( [ s2 ] ) .
this is the most important and kernel step in the calculation of the bispectrum which will be performed later .
the 2nd order action can be written as : s_2&=&dtd^3xae^2\{3(-2 + 1)^2+a^2e^-2+a^-2(-1+)()^2 } + & = & dd^3xe^2\{3(-2 + 1)u_^2()+a^2e^-2 ^ 2()-(-1+)k^2u_^2 ( ) } , where in the second step we have used conformal time @xmath47 and transformed the variable @xmath48 into its fourier form , namely : ( , ) = [ u_()a_e^-i+u_^()a_^e^i ] .
it is convenient to define another variable @xmath49 as @xmath50 , where z= , to let the equation of motion for @xmath49 be in a canonical form , which is : [ vk ] v_k^+(c_s^2k^2-)v_k=0 .here the effective sound speed squared @xmath51 is defined as @xmath52 , and thus one can have @xmath53 . in the inflation period , the hubble parameter changes slowly so as to have enough fast expansion . here in order to solve the equation above , we need to introduce another parameter : _ - , which describes the variation of @xmath35 with respect to time . in case that @xmath2 ( see eq .
( [ epsilonh ] ) ) and @xmath3 are both small and @xmath54 , we can take the leading order of @xmath2 and @xmath3 so that eq .
( [ vk ] ) becomes : [ mode ] v_^+(c_s^2k^2-)v_=0 , where @xmath55 is the ratio of the two slow roll parameters .
the solution of eq .
( [ mode ] ) can be presented in form of hankal function : [ hankel ] v_()=ch_(+)(c_sk|| ) , where @xmath56 is an undetermined constant denoting the amplitude of the solution . in deriving this ,
the approximation of slow - varying sound speed @xmath57 is also taken .
the solution ( [ hankel ] ) can be splitted into two limits corresponding to the subhubble and superhubble regions respectively . in the superhubble region
, we can take the limit to be : v_&&((c_sk||)^++(c_sk||)^-(+ ) ) + [ super1]&=&(c_sk||)^+1+(c_sk||)^- , and from the relation @xmath58 one have : u_&=&\{(c_sk||)^+1+(c_sk||)^- } + [ super2]&=&\{(c_sk||)^2 + 1+}where @xmath59 denotes the value of @xmath35 at @xmath60 and @xmath35 can be parameterized as @xmath61 . in deriving these equations
we also used the approximation @xmath62 .
the solution contains a constant mode and a decaying mode , the latter of which is irrelevant and should be discarded . in the subhubble region , we can take the limit as : [ sub1 ] v _
ce^ic_sk||e^i , where we also discarded the @xmath63 branch .
on the other hand , one can use wkb method to calculate the subhubble solution of eq .
( [ mode ] ) , which is : [ sub2 ] v _ e^-ic_sk[(1+)+2ic_sk ] .comparing eqs .
( [ sub1 ] ) and ( [ sub2 ] ) at @xmath64 , one can determine the coefficient @xmath56 as:[c ] c = e^-i .with this in hand , we can have the exact solution of @xmath65 in both superhubble and subhubble limits . substituting this back to ( [ super2 ] )
, we get the final form of the superhubble solution : u_=e^-i .it is a time - independent mode and thus can be applied to far future where @xmath66 . from this
we can also obtain the power spectrum of @xmath38 , which is : _ k^&&|u_|^2
+ [ spectrum]&= & , and the spectrum index is @xmath67 .
one can see from this that the power gets a red spectrum ( @xmath68 ) when @xmath69 while a blue spectrum ( @xmath70 ) will be obtained at @xmath71 . moreover , the constraints that the primordial spectrum must be nearly scale invariant requires that @xmath72 .
furthermore , using eq .
( [ sub1 ] ) , the subhubble solution can be solved as : u _ & & e^-ic_s||[(1+)+2ic_s|| ] + & = & e^-ic_sk[(1+)+2ic_sk ] + & = & e^-ic_sk[(1+)||^-1 - 2ic_sk||^ ] , and u_^()=||^ , where we keep only the leading order terms in terms of @xmath33 .
the results above will be useful for our analysis of non - gaussianity in the next paragraph . according to in - in " formalism @xcite
, the 3-point correlation function is characterized in the interaction picture as : [ inin ] < |(,)(,)(,)|>=-i_t_0^tdt^<|[(t,)(t,)(t,),h_int^p(t^)]| > , where @xmath73 is the 3rd order interaction hamiltonian and @xmath74 is the time - ordering operator . from the 3rd order action ( [ s3 ] ) , we can write down the 3rd order hamiltonian as : h_int&=&dtd^3xae^2\{[a^2e^-2(+2)+3(-h)^2]-3[a^2e^-2 + 3(-h)^2]-a^-2(-1+)()^2 + & & -a^-4 ^ -1[3a^2(1-)^2+a^4 ^ -2e^-2]()^2 + 2a^-4[3a^2(1-)^2+a^4 ^ -2e^-2 ] } , or , if changed to momentum space , h_int^p&=&(2)^3 ^ 3(++)ae^2\{[a^2e^-2(+2)+3(-h)^2 ] + & & -3[a^2e^-2 + 3(-h)^2]-a^-2(-1+)()(t,)(t,)(t , ) + & & -a^-4 ^ -1[3a^2(1-)^2+a^4 ^ -2e^-2]()(t,)(t,)(t , ) + & & + 2a^-4[3a^2(1-)^2+a^4 ^ -2e^-2]()(t,)(t,)(t , ) } .there are five terms of 3rd order , each containing a long prefactor . using eq .
( [ inin ] ) , we can calculate their contributions to non - gaussianities . neglecting the detailed calculating process
, we only give the final results of the contributions to non - gaussianities from each term as follows . the contribution from @xmath75 : & & -6iu_(0)u_(0)u_(0)_-^0d(2)^3 ^ 3(_i)u_-^()u_-^()u_-^ ( ) + & & -18iu_(0)u_(0)u_(0)_-^0||^1 - 2d(2)^3 ^ 3(_i)u_-^()u_-^()u_-^ ( ) + [ con1]&&+c.c.,the contribution from @xmath76 : & & 6iu_(0)u_(0)u_(0)_-^0d(2)^3 ^ 3(_i)[u_-^()u_-^()u_-^()+2perms .
] + & & 18iu_(0)u_(0)u_(0)_-^0||^-2d(2)^3 ^ 3(_i)[u_-^()u_-^()u_-^()+2perms . ]
+ [ con2]&&+c.c . , the contribution from @xmath77 : & & 2iu_(0)u_(0)u_(0)_-^0||^-2d(2)^3 ^ 3(_i)[()u_-^()u_-^()u_-^ ( ) + [ con3]&&+2perms.]+c.c.,the contribution from @xmath78 : & & -6iu_(0)u_(0)u_(0)_-^0||^1 - 2d(2)^3 ^ 3(_i)[()u_-^()u_-^()u_-^()+2perms ] + [ con4]&&-2iu_(0)u_(0)u_(0)_-^0d(2)^3 ^ 3(_i)[()u_-^()u_-^()u_-^()+2perms]+c.c.,the contribution from @xmath79 : & & -18iu_(0)u_(0)u_(0)_-^0||^-2d(2)^3 ^ 3(_i)[u_-^()u_-^()u_-^()+5perms ] + & & -2iu_(0)u_(0)u_(0)_-^0||^2 - 4d(2)^3 ^ 3(_i)[u_-^()u_-^()u_-^()+5perms ] + [ con5]&&-12iu_(0)u_(0)u_(0)_-^0d(2)^3 ^ 3(_i)[u_-^()u_-^()u_-^()+5perms]+c.c . in all these contributions
above , we can substitute the explicit forms of @xmath80 into the equation above to get a lot of integrals with @xmath81 .
it s straightforward but the result is rather boring and page - wasting , so we would put them into * appendix b*. actually , it has 9 terms differing from each an order of @xmath81 sequently , plus permutations and complex conjugates . at the end we will compile all these terms coming from contributions of all the terms in the hamiltonian according to their indices in a clearer form in order to make our study comfortable . from eqs .
( [ con1]-[con5 ] ) and also ( [ con1]-[con5 ] ) , we can integrate them out to have different shapes .
since the power - law indices of @xmath81 in each integration depend on the value of @xmath4 , one may worry that for indices less than @xmath82 , the infrared ( ir ) divergence will occur .
however , that is not the case . as has already been shown in sec .
iii b , the mode solution is frozen outside of horizon and there will be no more ir evolution .
actually , the ir divergences will all be canceled with each other and we do not see any real singularity , though it is tedious to check analytically and even difficult numerically .
the same argument can be found in the paper jcap * 1004 * , 027 ( 2010 ) as cited in @xcite .
furthermore , we can replace the value of @xmath65 at far future @xmath66 to that at horizon - crossing time .
taking these into consideration , we first write down all the possible shapes of the bispectrum . here
we define = ( 2)^3 ^ 3(_i)(k_1,k_2,k_3)where @xmath83 are the shapes of non - gaussianity
. there are totally 10 shapes at the leading order : _ 3 - 3&= & + & & [ b3mu-3]((1+)(3 ^ 2 + -12(-1)^2)_i >
jk_i^2k_j^2-(1+)(-6(-1)^2)_ik_i^4 ) , _ 3 - 2&= & + & & ( ( -6(-1)^2)(_i)_ij(k_i^2k_j^3-k_ik_j^4)+(6 ^ 2 + -12(-1)^2)_i >
jk_ik_j ) , _ 3 - 1&=&((_i)(_ijk_ik_j^2-_ik_i^3)-3(+2 ) ) , _ -3&=&_ik_i^2 , _ -2&=&-_ik_i^2
, _ -1&= & + & & ( 9(-1)(2(-1)^2-^2)_i > jk_i^2k_j^2 - 9(-1)^3_ik_i^4+(_i > jk_ik_j)_ik_i^2 ) , _ & = & + & & ( ( -1)(_ik_i)(9(-1)(2 - 1)_i > jk_ik_j+_ik_i^2)+_ij(k_i^2k_j^3-k_ik_j^4 ) ) , _ + 1&= & , _ 5 - 5&=&(2_i > jk_i^2k_j^2-_ik_i^4 ) , [ b5mu-4 ] _ 5 - 4&=&(_ijk_i^2k_j^3+(_ik_i)_i > jk_ik_j-_ijk_ik_j^4 ) . for the next step
, we have to discard the terms that will be divergent as the indices in the original integral becomes less than @xmath82 , and sum up all the convergent terms to give the total shape .
it depends on the value of @xmath4 , obviously . in the following
we give the total shape for different values of @xmath4 .
the final result is _
total=_y(y+1)_y , where @xmath84 is the heaviside step function defined as : ( x)=\ { c 0 , for x<0 , + + 1 , for x>0 , + . and @xmath85 for all the subscripts of @xmath86 above . from this result , we can also divide all the shapes to four classes : + i ) @xmath87 , @xmath88 , @xmath89 , @xmath90 : since we assume that the slow - roll parameters @xmath2 and @xmath3 larger than 0 , thus the parameter @xmath91 , and the indices of these shapes are larger than @xmath82 .
so they will definitely contribute to @xmath92 ; + ii ) @xmath93 , @xmath94 , @xmath95 , @xmath96 : if @xmath69 , i.e. , the power spectrum gets a red index , the indices of these shapes are larger than @xmath82 and they will contribute to @xmath92 ; + iii ) @xmath97 : if @xmath71 , i.e. , the power spectrum gets a blue index , the indices of these shapes are smaller than @xmath82 and they will not contribute to @xmath92 ; and + iv ) @xmath98 : the indices of these shapes are at divide " values .
so when @xmath69 , they will contribute to @xmath92 while when @xmath71 , they will not . we can also define the estimator through ( k_1,k_2,k_3)=f_nl\{_k_1^_k_2^+2perms . } , so , f_nl&= & + [ fnl]&=&(k_1,k_2,k_3 ) for each shape listed above .
the result is rather obvious by just substituting each shape into eq .
( [ fnl ] ) , so we will not list them here in order to save the page .
we will only show the equilateral limits of these @xmath5 s , of which @xmath99 , in the next paragraph . in this section
, we can take the equilateral limit , namely , @xmath99 , which gives simpler form of @xmath5 .
following the last paragraph , we have : [ f3mu-3 ] ( f_nl)_3 - 3^equil = k^2 - 2(1+)(3 ^ 2 + -6(-1)^2 ) , ( f_nl)_3 - 2^equil = k^2 - 2(6 ^ 2 + -12(-1)^2 ) , ( f_nl)_3 - 1^equil = k^2 - 2((-1)-2 ) , ( f_nl)_-3^equil= , ( f_nl)_-2^equil=- , ( f_nl)_-1^equil=(3(-1)^3 - 3(-1)^2 + ) , ( f_nl)_^equil=(-1)(^3 - 18(-1)(2 - 1 ) ) , ( f_nl)_+1^equil= , ( f_nl)_5 - 5^equil = k^4 - 4 , [ f5mu-4 ] ( f_nl)_5 - 4^equil = k^4 - 4 , and same as the shape , we have for total non - linear parameter : ( f_nl)_total^equil=_y(y+1)(f_nl)_y^equil .
from the result above we can see that @xmath100 has a slight running behavior due to the deviation of @xmath4 from 1 , that is , due to the non - minimal coupling behavior .
this is different from the usual minimal coupling case which was studied in @xcite where the equilateral limit of @xmath5 is independent of @xmath101 .
this is another result in this paper and will later be confirmed with numerical calculations .
one can also obtain @xmath5 in the local limit ( @xmath102 ) and folded limit ( @xmath103 ) .
whichever limits they are in , all the values of @xmath5 with different indices can also be divided into four classes by the same criteria used for @xmath104 . in order to support our long analytical derivation ,
we in this section focus on an explicit model of non - minimal coupling inflation . for simplicity but without losing generality , we consider chaotic inflation , of which the potential has a quadratic form as @xmath105 where @xmath106 is the coupling coefficients .
furthermore , we set the non - minimal coupling term @xmath107 where @xmath108 is the newtonian gravitational constant and @xmath109 is the non - minimal coupling coefficient .
this model is indeed very interesting since with the presence of non - minimal coupling term , the coefficient @xmath106 does nt need to go to incredibly small value ( fakir and unruh , @xcite ) to meet the observational constraint , and this property has been used to construct higgs inflation models , which connects inflation theory to particle physics in the standard model @xcite . in higgs inflation ,
the higgs potential asymptotically coincides with the chaotic potential in inflation period where the scalar field is in a high energy region with some large value , and a too small value of @xmath106 will make the models inconsistent with the constraints from standard model @xcite . putting aside its physical motivation , in this paper we study whether it can give rise to large non - gaussianities . from the original action ( [ action ] )
, one can obtain the equation of motion for the fields : + 3h-6(h+2h^2)+=0 , and friedmann equation : 3h^2(+^2)=^2+v()-6h , and the above two equations can be combined to give another equation : h=(+^2)=-^2+h-^2- .
firstly , we draw the background evolution of the system in figs .
[ epsilon ] and [ efold ] .
from the plots we can see that with natural choice of initial conditions and parameters , a period of inflation can be easily constructed with enough amount of number of e - folds . setting @xmath110 ,
the parameter @xmath106 could be raised up to @xmath111 , compared to the unnatural choice of @xmath112 in the case with @xmath113 ( fakir and unruh , @xcite ) .
[ spectrumplot ] and [ mu ] shows the amplitude and the @xmath114dependence of its quadratic perturbation spectrum , whose analytical form has already been given in eq .
( [ spectrum ] ) . for the given initial conditions and parameters
, we can see that the spectrum behaves nearly @xmath101-independent , with a slight tilt caused by the deviation of @xmath4 from 1 , which is mildly favored by the wmap-7 data @xcite .
the amplitude of the spectrum is also consistent with the observations .
cosmic time @xmath115 .
the arrival of @xmath33 at 1 stops the inflation .
parameters and initial values : @xmath110 , @xmath116 , @xmath117 , @xmath118 .
the normalization is @xmath119 . ]
cosmic time @xmath115 .
parameters and initial values are the same as in fig .
[ epsilon ] . ] . ] w.r.t .
cosmic time @xmath115 .
slightly deviation from 1 is obtained due to the non - minimal coupling . ]
next let s move on to the non - gaussianities that this model can give rise to .
since the power spectrum of this model has a blue tilt , the first two classes of the total four in the shape @xmath83 as well as the estimator @xmath5 which was shown in the last paragraphs will be applied . from our numerical calculations ,
we can obtain the values of every parameter that appears in eqs .
( [ b3mu-3])-([b5mu-4 ] ) as well as ( [ f3mu-3])-([f5mu-4 ] ) . with this in hand
, we can easily obtain the numerical results of the non - gaussianities generated in this model . the total shape ( of leading order ) of the non - gaussianities and the estimator in the equilateral limit @xmath100
are shown in figs .
[ shape ] and [ fnl ] . from the plots we can see that the shape of the non - gaussianities are well within the constraints of the observational data .
note that the estimator @xmath100 shows a running behavior with respect to @xmath120 , with a positive sign .
this is because of the effect of non - minimal coupling which makes the parameter @xmath4 deviate from @xmath121 and thus @xmath100 will be dependent on @xmath101 . in the gr limit @xmath122 , @xmath100 will have a constant value , as shown in @xcite . .
here we renormalize @xmath123 , @xmath124 , and set @xmath125 . ] . running behavior
is obtained due to the non - minimal coupling effect . ]
in this paper , we performed the non - gaussianities of a general single scalar field which linearly couples to gravity .
our result shows that due to the non - minimal coupling , the power spectrum will deviate from scale - invariance , which in order lead to the complicated non - gaussianities in the 3rd order .
we obtained all the possible shapes of the 3-point correlation functions and for different tilt of power spectrum , we showed that different shapes will be involved in to give rise to non - gaussianities .
our calculation presents the description in general non - minimal coupling inflation and this result , if verified to all the orders , can provide relation between 2- and 3-point correlation functions and can be used to constrain non - minimal coupling models .
another result that was presented in this paper is that there is some running behavior of the estimator @xmath5 in the equilateral limit with respect to @xmath101 which is different from the normal minimal coupling case .
this behavior is due to the non - minimal coupling , and are expected to have signature on observations in order to distinguish minimal and non - minimal cases .
besides the analytical calculations , we also performed numerical computations on a specific example of non - minimal coupling chaotic inflation model .
this model is extendedly studied with the application of higgs inflation .
we obtained the behavior of background , 2-point power spectrum as well as the shape and estimator of its non - gaussianity .
we showed that the non - gaussianities are well within the observational constraint , with the running behavior of @xmath126 w.r.t .
@xmath101 .
other than inflation , such non - minimal coupling system can also be applied to other aspect in cosmology .
for example , non - minimal coupling theory can act as dark energy @xcite or give rise to a bouncing / cyclic universe @xcite .
moreover , non - minimal coupling can be used to make up an open / closed universe @xcite , while dualities of einstein s gravity in the presence of a non - minimal coupling was taken on in @xcite .
the stabilities and singularities in superacceleration phases was discussed in @xcite , and the removal of singularities in loop quantum gravity with non - minimal coupling was studied in @xcite .
our calculation of non - gaussianities are also expected to be applied to these interesting fields .
tq owes many thanks to dr .
xian gao , dr .
yi wang and prof . xingang chen for useful help and guidance .
tq also thanks dr .
yi - fu cai and dr .
antonio de felice for discussions .
tq is grateful to the high energy accelerator research organization ( kek ) and institute of the physic and mathematics of the universe ( ipmu ) in japan for their hospitalities during his visit , and the horiba international conference cosmo / cospa 2010 " and the following string workshop held in tokyo university for stimulating talks and discussions when this work was processing , as well as the lecospa group meeting in national taiwan university for helpful discussions .
this research is supported in parts by the national science council of r.o.c . under grant
nsc99 - 2112-m-033 - 005-my3 and no .
nsc99 - 2811-m-033 - 008 and by the national center for theoretical sciences .
here , we only list the constraint equations for the general form of non - minimal coupling single scalar field , including the non - linear coupling case .
the constraints from @xmath25 and @xmath26 are : & & 4 + 12hf_r0 - 12h^2f_r0 - 4a^-2f_r0h^2 + & & -4a^-2f_r0 ^ 2-(12h-6 + 2a^-2 ^
2)_tf_r0 + 72(-h^2)(+2h^2)f_rr0 - 36(-h^2)f_rr0 + & & -36(-h^2)hf_rr0(4-)+12a^-2(-h^2)f_rr0(^2 + 2h^2+^2 + 2 ^ 2 ) + & & + 6h_t[-12(+2h^2)f_rr0 + 6hf_rr0(4-)+6f_rr0 - 2a^-2f_rr0(^2 + 2h^2+^2 + 2 ^ 2 ) ] + & & + 24a^-2(+2h^2)f_rr0 ^ 2 - 12a^-2hf_rr0 ^ 2(4-)-12a^-2f_rr0 ^ 2 + & & + 4a^-4f_rr0 ^ 2(^2 + 2h^2+^2 + 2 ^ 2)=0and & & -a^-2f_r0 ^ 2^i-12a^-2h(+2h^2)f_rr0^i+6a^-2h^2f_rr0^i(4-)+6a^-2hf_rr0^i + & & -2a^-4hf_rr0^i(^2 + 2h^2+^2 + 2 ^ 2)+a^-2_t[12(+2h^2)f_rr0^i-6hf_rr0^i(4- ) + & & -6f_rr0^i+2a^-2f_rr0^i(^2 + 2h^2+^2 + 2 ^ 2 ) ] + & & + 2a^-2f_r0h^i+a^-2_tf_r0^i-2a^-2f_r0^i=0respectively , where @xmath35 and @xmath127 denotes the background value of the first and second derivatives of @xmath128 with respect to @xmath0 .
here , we list the contributions from terms in @xmath73 , i.e. eq . ( [ con1]-[con5 ] ) , in terms of @xmath81 by substituting @xmath80 in .
one can see that there contains integrals of different power - laws of @xmath81 , each differing one order from the other in every term .
one can combine the integrals of the same order power - law to have more neat forms , as in * sec .
iii d*. the contribution from @xmath76 : & & -6 + & & _
-^0de^ic_sk[\{+4c_s^4(1+)k_2 ^ 2k_3 ^ 2||^3 - 3 + 8ic_s^5k_1k_2 ^ 2k_3 ^ 2||^3 - 2}+2perms . ] + & & -18 + [ con2]&&_-^0de^ic_sk[\{+4c_s^4(1+)k_2 ^ 2k_3 ^ 2||^-1 + 8ic_s^5k_1k_2 ^ 2k_3 ^ 2||^}+2perms.]+c.c . the contribution from @xmath79 : & & + & & _
-^0de^ic_sk[\{+4c_s^4(1+)k_1 ^ 2k_3 ^ 2||^-1 + 8ic_s^5k_2k_1 ^ 2k_3 ^ 2||^}+5perms ] + & & + + & & _
-^0de^ic_sk[\{+4c_s^4(1+)k_1 ^ 2k_3 ^ 2||^5 - 5 + 8ic_s^5k_2k_1 ^ 2k_3 ^ 2||^5 - 4}+5perms ] + & & + + [ con5]&&_-^0de^ic_sk[\{+4c_s^4(1+)k_1 ^ 2k_3 ^ 2||^3 - 3 + 8ic_s^5k_2k_1 ^ 2k_3 ^ 2||^3 - 2}+5perms]+c.c .
t. futamase , t. rothman and r. matzner , phys .
d * 39 * , 405 ( 1989 ) ; l. amendola , d. bellisai and f. occhionero , phys .
d * 47 * , 4267 ( 1993 ) [ arxiv : gr - qc/9303023 ] ; v. faraoni , phys . rev .
d * 53 * , 6813 ( 1996 ) [ arxiv : astro - ph/9602111 ] ; t. futamase and m. tanaka , phys .
d * 60 * , 063511 ( 1999 ) [ arxiv : hep - ph/9704303 ] ; v. faraoni , phys .
d * 62 * , 023504 ( 2000 ) [ arxiv : gr - qc/0002091 ] ; v. faraoni , phys .
a * 269 * , 209 ( 2000 ) [ arxiv : gr - qc/0004007 ] ; t. chiba and m. yamaguchi , jcap * 0810 * , 021 ( 2008 ) [ arxiv:0807.4965 [ astro - ph ] ] .
d. s. salopek , j. r. bond and j. m. bardeen , phys .
d * 40 * , 1753 ( 1989 ) ; r. fakir and w. g. unruh , phys .
d * 41 * , 1783 ( 1990 ) ; d. i. kaiser , phys .
d * 52 * , 4295 ( 1995 ) [ arxiv : astro - ph/9408044 ] ; d. hochberg and t. w. kephart , phys .
d * 51 * , 2687 ( 1995 ) [ arxiv : gr - qc/9501022 ] ; m. jankiewicz and t. w. kephart , phys .
d * 73 * , 123514 ( 2006 ) [ arxiv : hep - ph/0510009 ] ; k. nozari and s. shafizadeh , phys .
scripta * 82 * , 015901 ( 2010 ) [ arxiv:1006.1027 [ gr - qc ] ] ; j. weenink and t. prokopec , arxiv:1007.2133 [ hep - th ] .
s. tsujikawa , phys .
d * 62 * , 043512 ( 2000 ) [ arxiv : hep - ph/0004088 ] ; s. tsujikawa and h. yajima , phys .
d * 62 * , 123512 ( 2000 ) [ arxiv : hep - ph/0007351 ] ; a. a. starobinsky , s. tsujikawa and j. yokoyama , nucl .
b * 610 * , 383 ( 2001 ) [ arxiv : astro - ph/0107555 ] ; l. brenig , a. figueiredo , e. gunzig , t. m. rocha filho and a. saa , int .
j. theor .
phys . *
42 * , 1153 ( 2003 ) [ arxiv : gr - qc/0305011 ] ; d. i. kaiser , phys
. rev .
d * 81 * , 084044 ( 2010 ) [ arxiv:1003.1159 [ gr - qc ] ] ; d. i. kaiser and a. t. todhunter , phys .
d * 81 * , 124037 ( 2010 ) [ arxiv:1004.3805 [ astro-ph.co ] ] .
e. komatsu and t. futamase , phys .
d * 58 * , 023004 ( 1998 ) [ arxiv : astro - ph/9711340 ] ; e. komatsu and t. futamase , phys .
d * 59 * , 064029 ( 1999 ) [ arxiv : astro - ph/9901127 ] ; j. c. hwang and h. noh , phys .
d * 60 * , 123001 ( 1999 ) [ arxiv : astro - ph/9908340 ] .
t. j. allen , b. grinstein and m. b. wise , phys .
b * 197 * , 66 ( 1987 ) ; h. m. hodges , g. r. blumenthal , l. a. kofman and j. r. primack , nucl . phys .
b * 335 * , 197 ( 1990 ) ; d. s. salopek and j. r. bond , phys .
d * 42 * , 3936 ( 1990 ) ; t. falk , r. rangarajan and m. srednicki , astrophys
. j. * 403 * , l1 ( 1993 ) [ arxiv : astro - ph/9208001 ] ; a. gangui , f. lucchin , s. matarrese and s. mollerach , astrophys .
j. * 430 * , 447 ( 1994 ) [ arxiv : astro - ph/9312033 ] ; e. komatsu and d. n. spergel , phys .
d * 63 * , 063002 ( 2001 ) [ arxiv : astro - ph/0005036 ] .
v. acquaviva , n. bartolo , s. matarrese and a. riotto , nucl . phys .
b * 667 * , 119 ( 2003 ) [ arxiv : astro - ph/0209156 ] ; j. m. maldacena , jhep * 0305 * , 013 ( 2003 ) [ arxiv : astro - ph/0210603 ] ; x. chen , m. x. huang , s. kachru and g. shiu , jcap * 0701 * , 002 ( 2007 ) [ arxiv : hep - th/0605045 ] .
a. d. linde and v. f. mukhanov , phys .
d * 56 * , 535 ( 1997 ) [ arxiv : astro - ph/9610219 ] ; n. bartolo , s. matarrese and a. riotto , phys .
d * 65 * , 103505 ( 2002 ) [ arxiv : hep - ph/0112261 ] ; f. bernardeau and j. p. uzan , phys .
d * 66 * , 103506 ( 2002 ) [ arxiv : hep - ph/0207295 ] ; g. i. rigopoulos , e. p. s. shellard and b. j. w. van tent , phys .
d * 73 * , 083521 ( 2006 ) [ arxiv : astro - ph/0504508 ] ; d. seery and j. e. lidsey , jcap * 0509 * , 011 ( 2005 ) [ arxiv : astro - ph/0506056 ] ; k. a. malik , jcap * 0511 * , 005 ( 2005 ) [ arxiv : astro - ph/0506532 ] ; g. i. rigopoulos , e. p. s. shellard and b. j. w. van tent , phys .
d * 73 * , 083522 ( 2006 ) [ arxiv : astro - ph/0506704 ] ; l. e. allen , s. gupta and d. wands , jcap * 0601 * , 006 ( 2006 ) [ arxiv : astro - ph/0509719 ] ; c. y. sun and d. h. zhang , int .
j. mod .
d * 15 * , 1259 ( 2006 ) [ arxiv : astro - ph/0510709 ] ; g. i. rigopoulos , e. p. s. shellard and b. j. w. van tent , phys .
d * 76 * , 083512 ( 2007 ) [ arxiv : astro - ph/0511041 ] ; a. jokinen and a. mazumdar , jcap * 0604 * , 003 ( 2006 ) [ arxiv : astro - ph/0512368 ] ; f. vernizzi and d. wands , jcap * 0605 * , 019 ( 2006 ) [ arxiv : astro - ph/0603799 ] ; t. battefeld and r. easther , jcap * 0703 * , 020 ( 2007 ) [ arxiv : astro - ph/0610296 ] ; k. y. choi , l. m. h. hall and c. van de bruck , jcap * 0702 * , 029 ( 2007 ) [ arxiv : astro - ph/0701247 ] ; d. battefeld and t. battefeld , jcap * 0705 * , 012 ( 2007 ) [ arxiv : hep - th/0703012 ] ; s. yokoyama , t. suyama and t. tanaka , jcap * 0707 * , 013 ( 2007 ) [ arxiv:0705.3178 [ astro - ph ] ] ; d. a. easson , r. gregory , d. f. mota , g. tasinato and i. zavala , jcap * 0802 * , 010 ( 2008 ) [ arxiv:0709.2666 [ hep - th ] ] ; s. yokoyama , t. suyama and t. tanaka , phys . rev .
d * 77 * , 083511 ( 2008 ) [ arxiv:0711.2920 [ astro - ph ] ] ; m. x. huang , g. shiu and b. underwood , phys .
d * 77 * , 023511 ( 2008 ) [ arxiv:0709.3299 [ hep - th ] ] ; j. ward , jhep * 0712 * , 045 ( 2007 ) [ arxiv:0711.0760 [ hep - th ] ] ; s. w. li and w. xue , arxiv:0804.0574 [ astro - ph ] ; s. yokoyama , t. suyama and t. tanaka , phys .
d * 77 * , 083511 ( 2008 ) [ arxiv:0711.2920 [ astro - ph ] ] ; x. gao , jcap * 0806 * , 029 ( 2008 ) [ arxiv:0804.1055 [ astro - ph ] ] ; d. langlois , s. renaux - petel , d. a. steer and t. tanaka , phys . rev .
* 101 * , 061301 ( 2008 ) [ arxiv:0804.3139 [ hep - th ] ] ; d. langlois , s. renaux - petel , d. a. steer and t. tanaka , phys . rev .
d * 78 * , 063523 ( 2008 ) [ arxiv:0806.0336 [ hep - th ] ] ; f. arroja , s. mizuno and k. koyama , jcap * 0808 * , 015 ( 2008 ) [ arxiv:0806.0619 [ astro - ph ] ] ; c. t. byrnes , k. y. choi and l. m. h. hall , jcap * 0810 * , 008 ( 2008 ) [ arxiv:0807.1101 [ astro - ph ] ] ; a. misra and p. shukla , nucl . phys .
b * 810 * , 174 ( 2009 ) [ arxiv:0807.0996 [ hep - th ] ] ; m. kawasaki , k. nakayama , t. sekiguchi , t. suyama and f. takahashi , jcap * 0811 * , 019 ( 2008 ) [ arxiv:0808.0009 [ astro - ph ] ] ; d. langlois , f. vernizzi and d. wands , jcap * 0812 * , 004 ( 2008 ) [ arxiv:0809.4646 [ astro - ph ] ] ; y. f. cai and w. xue , phys .
b * 680 * , 395 ( 2009 ) [ arxiv:0809.4134 [ hep - th ] ] ; c. t. byrnes , k. y. choi and l. m. h. hall , jcap * 0902 * , 017 ( 2009 ) [ arxiv:0812.0807 [ astro - ph ] ] ; s. h. tye , j. xu and y. zhang , jcap * 0904 * , 018 ( 2009 ) [ arxiv:0812.1944 [ hep - th ] ] ; d. langlois , s. renaux - petel and d. a. steer , jcap * 0904 * , 021 ( 2009 ) [ arxiv:0902.2941 [ hep - th ] ] ; x. gao and b. hu , jcap * 0908 * , 012 ( 2009 ) [ arxiv:0903.1920 [ astro-ph.co ] ] ; y. f. cai and h. y. xia , phys .
b * 677 * , 226 ( 2009 ) [ arxiv:0904.0062 [ hep - th ] ] ; e. kawakami , m. kawasaki , k. nakayama and f. takahashi , jcap * 0909 * , 002 ( 2009 ) [ arxiv:0905.1552 [ astro-ph.co ] ] ; s. pi and t. wang , phys .
d * 80 * , 043503 ( 2009 ) [ arxiv:0905.3470 [ astro-ph.co ] ] ; s. mizuno , f. arroja , k. koyama and t. tanaka , phys .
d * 80 * , 023530 ( 2009 ) [ arxiv:0905.4557 [ hep - th ] ] ; x. gao , m. li and c. lin , jcap * 0911 * , 007 ( 2009 ) [ arxiv:0906.1345 [ astro-ph.co ] ] ; j. l. lehners and s. renaux - petel , phys .
d * 80 * , 063503 ( 2009 ) [ arxiv:0906.0530 [ hep - th ] ] ; s. mizuno , f. arroja and k. koyama , phys .
d * 80 * , 083517 ( 2009 ) [ arxiv:0907.2439 [ hep - th ] ] ; s. renaux - petel , jcap * 0910 * , 012 ( 2009 ) [ arxiv:0907.2476 [ hep - th ] ] ; d. battefeld and t. battefeld , jcap * 0911 * , 010 ( 2009 ) [ arxiv:0908.4269 [ hep - th ] ] ; t. wang , arxiv:1008.3198 [ astro-ph.co ] ; x. gao and c. lin , arxiv:1009.1311 [ hep - th ] .
p. creminelli , jcap * 0310 * , 003 ( 2003 ) [ arxiv : astro - ph/0306122 ] ; d. babich , p. creminelli and m. zaldarriaga , jcap * 0408 * , 009 ( 2004 ) [ arxiv : astro - ph/0405356 ] ; g. calcagni , jcap * 0510 * , 009 ( 2005 ) [ arxiv : astro - ph/0411773 ] ; d. seery and j. e. lidsey , jcap * 0506 * , 003 ( 2005 ) [ arxiv : astro - ph/0503692 ] ; d. seery , j. e. lidsey and m. s. sloth , jcap * 0701 * , 027 ( 2007 ) [ arxiv : astro - ph/0610210 ] ; d. seery and j. e. lidsey , jcap * 0701 * , 008 ( 2007 ) [ arxiv : astro - ph/0611034 ] ; c. t. byrnes , m. sasaki and d. wands , phys .
d * 74 * , 123519 ( 2006 ) [ arxiv : astro - ph/0611075 ] ; x. chen , r. easther and e. a. lim , jcap * 0706 * , 023 ( 2007 ) [ arxiv : astro - ph/0611645 ] ; c. cheung , p. creminelli , a. l. fitzpatrick , j. kaplan and l. senatore , jhep * 0803 * , 014 ( 2008 ) [ arxiv:0709.0293 [ hep - th ] ] . m. li , t. wang and y. wang , jcap * 0803 * , 028 ( 2008 ) [ arxiv:0801.0040 [ astro - ph ] ] ; x. chen , r. easther and e. a. lim , jcap * 0804 * , 010 ( 2008 ) [ arxiv:0801.3295 [ astro - ph ] ] ; f. arroja and k. koyama , phys . rev .
d * 77 * , 083517 ( 2008 ) [ arxiv:0802.1167 [ hep - th ] ] ; d. seery , m. s. sloth and f. vernizzi , jcap * 0903 * , 018 ( 2009 ) [ arxiv:0811.3934 [ astro - ph ] ] ; x. chen , b. hu , m. x. huang , g. shiu and y. wang , jcap * 0908 * , 008 ( 2009 ) [ arxiv:0905.3494 [ astro-ph.co ] ] ; x. chen and y. wang , phys .
d * 81 * , 063511 ( 2010 ) [ arxiv:0909.0496 [ astro-ph.co ] ] ; x. chen and y. wang , jcap * 1004 * , 027 ( 2010 ) [ arxiv:0911.3380 [ hep - th ] ] ; n. bartolo , m. fasiello , s. matarrese and a. riotto , jcap * 1008 * , 008 ( 2010 ) [ arxiv:1004.0893 [ astro-ph.co ] ] ; n. bartolo , m. fasiello , s. matarrese and a. riotto , jcap * 1009 * , 035 ( 2010 ) [ arxiv:1006.5411 [ astro-ph.co ] ] .
d. h. lyth , c. ungarelli and d. wands , phys .
d * 67 * , 023503 ( 2003 ) [ arxiv : astro - ph/0208055 ] ; n. bartolo , s. matarrese and a. riotto , phys .
d * 69 * , 043503 ( 2004 ) [ arxiv : hep - ph/0309033 ] ; k. enqvist and s. nurmi , jcap * 0510 * , 013 ( 2005 ) [ arxiv : astro - ph/0508573 ] ; d. h. lyth , jcap * 0606 * , 015 ( 2006 ) [ arxiv : astro - ph/0602285 ] ; k. a. malik and d. h. lyth , jcap * 0609 * , 008 ( 2006 ) [ arxiv : astro - ph/0604387 ] ; m. sasaki , j. valiviita and d. wands , phys .
d * 74 * , 103003 ( 2006 ) [ arxiv : astro - ph/0607627 ] ; j. valiviita , m. sasaki and d. wands , arxiv : astro - ph/0610001 ; h. assadullahi , j. valiviita and d. wands , phys .
d * 76 * , 103003 ( 2007 ) [ arxiv:0708.0223 [ hep - ph ] ] ; q. g. huang , phys .
b * 669 * , 260 ( 2008 ) [ arxiv:0801.0467 [ hep - th ] ] ; k. ichikawa , t. suyama , t. takahashi and m. yamaguchi , phys .
d * 78 * , 023513 ( 2008 ) [ arxiv:0802.4138 [ astro - ph ] ] ; t. multamaki , j. sainio and i. vilja , phys .
d * 79 * , 103516 ( 2009 ) [ arxiv:0803.2637 [ astro - ph ] ] ; q. g. huang , jcap * 0809 * , 017 ( 2008 ) [ arxiv:0807.1567 [ hep - th ] ] ; q. g. huang and y. wang , jcap * 0809 * , 025 ( 2008 ) [ arxiv:0808.1168 [ hep - th ] ] ; m. li , c. lin , t. wang and y. wang , phys .
d * 79 * , 063526 ( 2009 ) [ arxiv:0805.1299 [ astro - ph ] ] ; y. f. cai , s. li and y. s. piao , phys .
b * 671 * , 423 ( 2009 ) [ arxiv:0806.2363 [ hep - ph ] ] ; q. g. huang , phys .
d * 78 * , 043515 ( 2008 ) [ arxiv:0807.0050 [ hep - th ] ] ; p. chingangbam and q. g. huang , jcap * 0904 * , 031 ( 2009 ) [ arxiv:0902.2619 [ astro-ph.co ] ] .
t. kobayashi and s. mukohyama , jcap * 0907 * , 032 ( 2009 ) [ arxiv:0905.2835 [ hep - th ] ] ; j. o. gong , c. lin and y. wang , jcap * 1003 * , 004 ( 2010 ) [ arxiv:0912.2796 [ astro-ph.co ] ] ; j. zhang , y. f. cai and y. s. piao , jcap * 1005 * , 001 ( 2010 ) [ arxiv:0912.0791 [ hep - th ] ] ; c. lin and y. wang , jcap * 1007 * , 011 ( 2010 ) [ arxiv:1004.0461 [ astro-ph.co ] ] .
y. f. cai and y. wang , arxiv:1005.0127 [ hep - th ] .
m. alishahiha , e. silverstein and d. tong , phys .
d * 70 * , 123505 ( 2004 ) [ arxiv : hep - th/0404084 ] ; x. chen , phys .
d * 72 * , 123518 ( 2005 ) [ arxiv : astro - ph/0507053 ] ; x. chen , m. x. huang and g. shiu , phys .
d * 74 * , 121301 ( 2006 ) [ arxiv : hep - th/0610235 ] ; j. khoury and f. piazza , jcap * 0907 * , 026 ( 2009 ) [ arxiv:0811.3633 [ hep - th ] ] ; f. arroja , s. mizuno , k. koyama and t. tanaka , phys .
d * 80 * , 043527 ( 2009 ) [ arxiv:0905.3641 [ hep - th ] ] .
n. arkani - hamed , p. creminelli , s. mukohyama and m. zaldarriaga , jcap * 0404 * , 001 ( 2004 ) [ arxiv : hep - th/0312100 ] ; q. g. huang , jcap * 1007 * , 025 ( 2010 ) [ arxiv:1004.0808 [ astro-ph.co ] ] ; k. izumi and s. mukohyama , jcap * 1006 * , 016 ( 2010 ) [ arxiv:1004.1776 [ hep - th ] ] .
s. gupta , a. berera , a. f. heavens and s. matarrese , phys .
d * 66 * , 043510 ( 2002 ) [ arxiv : astro - ph/0205152 ] ; s. gupta , phys .
d * 73 * , 083514 ( 2006 ) [ arxiv : astro - ph/0509676 ] ; i. g. moss and c. xiong , jcap * 0704 * , 007 ( 2007 ) [ arxiv : astro - ph/0701302 ] ; i. g. moss and c. m. graham , jcap * 0711 * , 004 ( 2007 ) [ arxiv:0707.1647 [ astro - ph ] ] ; y. f. cai , j. b. dent and d. a. easson , arxiv:1011.4074 [ hep - th ] . r. holman and a. j. tolley , jcap * 0805 * , 001 ( 2008 ) [ arxiv:0710.1302 [ hep - th ] ] ; b. chen , y. wang and w. xue , jcap * 0805 * , 014 ( 2008 ) [ arxiv:0712.2345 [ hep - th ] ] ; w. xue and b. chen , phys
. rev .
d * 79 * , 043518 ( 2009 ) [ arxiv:0806.4109 [ hep - th ] ] ; y. ling and j. p.
wu , phys .
b * 675 * , 151 ( 2009 ) [ arxiv:0809.3398 [ hep - th ] ] ; p. d. meerburg , j. p. van der schaar and p. s. corasaniti , jcap * 0905 * , 018 ( 2009 ) [ arxiv:0901.4044 [ hep - th ] ] ; p.
d. meerburg , j. p.
van der schaar and m. g. jackson , jcap * 1002 * , 001 ( 2010 ) [ arxiv:0910.4986 [ hep - th ] ] ; x. chen , arxiv:1008.2485 [ hep - th ] ; y. f. cai and e. n. saridakis , arxiv:1011.1245 [ hep - th ] .
y. f. cai and x. zhang , jcap * 0906 * , 003 ( 2009 ) [ arxiv:0808.2551 [ astro - ph ] ] ; y. f. cai , w. xue , r. brandenberger and x. zhang , jcap * 0905 * , 011 ( 2009 ) [ arxiv:0903.0631 [ astro-ph.co ] ] ; y. f. cai , w. xue , r. brandenberger and x. m. zhang , jcap * 0906 * , 037 ( 2009 ) [ arxiv:0903.4938 [ hep - th ] ] .
d. seery , jcap * 0711 * , 025 ( 2007 ) [ arxiv:0707.3377 [ astro - ph ] ] ; d. seery , jcap * 0802 * , 006 ( 2008 ) [ arxiv:0707.3378 [ astro - ph ] ] ; e. dimastrogiovanni and n. bartolo , jcap * 0811 * , 016 ( 2008 ) [ arxiv:0807.2790 [ astro - ph ] ] ; h. r. s. cogollo , y. rodriguez and c. a. valenzuela - toledo , jcap * 0808 * , 029 ( 2008 ) [ arxiv:0806.1546 [ astro - ph ] ] ; y. rodriguez and c. a. valenzuela - toledo , phys .
d * 81 * , 023531 ( 2010 ) [ arxiv:0811.4092 [ astro - ph ] ] ; x. gao and f. xu , jcap * 0907 * , 042 ( 2009 ) [ arxiv:0905.0405 [ hep - th ] ] .
j. kumar , l. leblond and a. rajaraman , jcap * 1004 * , 024 ( 2010 ) [ arxiv:0909.2040 [ astro-ph.co ] ] .
m. hindmarsh , c. ringeval and t. suyama , phys .
d * 80 * , 083501 ( 2009 ) [ arxiv:0908.0432 [ astro-ph.co ] ] ; m. hindmarsh , c. ringeval and t. suyama , phys .
d * 81 * , 063505 ( 2010 ) [ arxiv:0911.1241 [ astro-ph.co ] ] ; d. m. regan and e. p. s. shellard , phys .
d * 82 * , 063527 ( 2010 ) [ arxiv:0911.2491 [ astro-ph.co ] ] .
d. h. lyth , jcap * 0511 * , 006 ( 2005 ) [ arxiv : astro - ph/0510443 ] ; l. alabidi and d. lyth , jcap * 0608 * , 006 ( 2006 ) [ arxiv : astro - ph/0604569 ] ; l. alabidi , jcap * 0610 * , 015 ( 2006 ) [ arxiv : astro - ph/0604611 ] ; n. barnaby and j. m. cline , phys . rev .
d * 73 * , 106012 ( 2006 ) [ arxiv : astro - ph/0601481 ] ; n. barnaby and j. m. cline , phys .
d * 75 * , 086004 ( 2007 ) [ arxiv : astro - ph/0611750 ] ; m. sasaki , prog .
phys . * 120 * , 159 ( 2008 ) [ arxiv:0805.0974 [ astro - ph ] ] ; a. naruko and m. sasaki , prog .
phys . * 121 * , 193 ( 2009 ) [ arxiv:0807.0180 [ astro - ph ] ] ; q. g. huang , jcap * 0905 * , 005 ( 2009 ) [ arxiv:0903.1542 [ hep - th ] ] ; q. g. huang , jcap * 0906 * , 035 ( 2009 ) [ arxiv:0904.2649 [ hep - th ] ] .
k. koyama , s. mizuno , f. vernizzi and d. wands , jcap * 0711 * , 024 ( 2007 ) [ arxiv:0708.4321 [ hep - th ] ] ; e. i. buchbinder , j. khoury and b. a. ovrut , phys
. rev .
lett . * 100 * , 171302 ( 2008 ) [ arxiv:0710.5172 [ hep - th ] ] ; j. l. lehners and p. j. steinhardt , phys .
d * 77 * , 063533 ( 2008 ) [ erratum - ibid .
d * 79 * , 129903 ( 2009 ) ] [ arxiv:0712.3779 [ hep - th ] ] ; j. l. lehners and p. j. steinhardt , phys .
d * 78 * , 023506 ( 2008 ) [ erratum - ibid .
d * 79 * , 129902 ( 2009 ) ] [ arxiv:0804.1293 [ hep - th ] ] ; j. l. lehners and s. renaux - petel , phys .
d * 80 * , 063503 ( 2009 ) [ arxiv:0906.0530 [ hep - th ] ] .
m. karciauskas , k. dimopoulos and d. h. lyth , phys .
d * 80 * , 023509 ( 2009 ) [ arxiv:0812.0264 [ astro - ph ] ] ; n. bartolo , e. dimastrogiovanni , s. matarrese and a. riotto , jcap * 0910 * , 015 ( 2009 ) [ arxiv:0906.4944 [ astro-ph.co ] ] ; c. a. valenzuela - toledo , y. rodriguez and d. h. lyth , phys .
d * 80 * , 103519 ( 2009 ) [ arxiv:0909.4064 [ astro-ph.co ] ] ; e. dimastrogiovanni , n. bartolo , s. matarrese and a. riotto , adv .
astron . *
2010 * , 752670 ( 2010 ) [ arxiv:1001.4049 [ astro-ph.co ] ] .
a. volovich and c. wen , jhep * 0905 * , 087 ( 2009 ) [ arxiv:0903.2455 [ hep - th ] ] ; x. gao , arxiv:0904.4187 [ hep - th ] ; k. izumi , t. kobayashi and s. mukohyama , jcap * 1010 * , 031 ( 2010 ) [ arxiv:1008.1406 [ hep - th ] ] ; t. p. sotiriou , arxiv:1010.3218 [ hep - th ] .
g. dvali , a. gruzinov and m. zaldarriaga , phys .
d * 69 * , 023505 ( 2004 ) [ arxiv : astro - ph/0303591 ] ; t. suyama and f. takahashi , jcap * 0809 * , 007 ( 2008 ) [ arxiv:0804.0425 [ astro - ph ] ] ; r. flauger and e. pajer , arxiv:1002.0833 [ hep - th ] .
s. koh , s. p. kim and d. j. song , phys .
d * 72 * , 043523 ( 2005 ) [ arxiv : astro - ph/0505188 ] ; s. koh , j. korean phys .
soc . * 49 * , s787 ( 2006 ) [ arxiv : astro - ph/0510030 ] ; d. a. easson , s. mukohyama and b. a. powell , phys .
d * 81 * , 023512 ( 2010 ) [ arxiv:0910.1353 [ astro-ph.co ] ] ; c. pallis , phys .
b * 692 * , 287 ( 2010 ) [ arxiv:1002.4765 [ astro-ph.co ] ] .
y. fujii and k. maeda , _ the scalar - tensor theory of gravitation _ , cambridge university press ( 2003 ) ; b. boisseau , g. esposito - farese , d. polarski and a. a. starobinsky , phys .
lett . * 85 * , 2236 ( 2000 ) .
j. p. uzan , phys .
d * 59 * , 123510 ( 1999 ) [ arxiv : gr - qc/9903004 ] ; t. chiba , phys .
d * 60 * , 083508 ( 1999 ) [ arxiv : gr - qc/9903094 ] ; l. amendola , phys .
d * 60 * , 043501 ( 1999 ) [ arxiv : astro - ph/9904120 ] ; l. amendola , phys .
d * 62 * , 043511 ( 2000 ) [ arxiv : astro - ph/9908023 ] ; v. faraoni , class .
* 22 * , 3235 ( 2005 ) [ arxiv : gr - qc/0506095 ] ; m. szydlowski , o. hrycyna and a. kurek , phys .
d * 77 * , 027302 ( 2008 ) [ arxiv:0710.0366 [ astro - ph ] ] ; l. n. granda , jcap * 1007 * , 006 ( 2010 ) [ arxiv:0911.3702 [ hep - th ] ] ; l. n. granda and w. cardona , jcap * 1007 * , 021 ( 2010 ) [ arxiv:1005.2716 [ hep - th ] ] .
j. p. abreu , p. crawford and j. p. mimoso , class .
grav . * 11 * , 1919 ( 1994 ) [ arxiv : gr - qc/9401024 ] ; s. s. bayin , f. i. cooperstock and v. faraoni , astrophys . j. * 428 * , 439 ( 1994 ) [ arxiv : astro - ph/9402033 ] ; r. fakir , arxiv : gr - qc/9810054 ; j. w. lee , s. koh , c. park , s. j. sin and c. h. lee , phys . rev .
d * 61 * , 027301 ( 2000 ) [ arxiv : hep - th/9909106 ] .
a. d. linde and a. mezhlumian , phys .
d * 52 * , 6789 ( 1995 ) [ arxiv : astro - ph/9506017 ] ; a. d. linde , arxiv : gr - qc/9508019 ; a. d. linde , arxiv : astro - ph/9601004 ; a. d. linde , jcap * 0305 * , 002 ( 2003 ) [ arxiv : astro - ph/0303245 ] .
e. gunzig , a. saa , l. brenig , v. faraoni , t. m. rocha filho and a. figueiredo , phys .
d * 63 * , 067301 ( 2001 ) [ arxiv : gr - qc/0012085 ] ; l. r. abramo , l. brenig and e. gunzig , phys .
b * 549 * , 13 ( 2002 ) [ arxiv : gr - qc/0205022 ] ; l. r. abramo , l. brenig , e. gunzig and a. saa , phys .
d * 67 * , 027301 ( 2003 ) [ arxiv : gr - qc/0210069 ] ; v. faraoni , phys .
d * 70 * , 047301 ( 2004 ) [ arxiv : gr - qc/0403020 ] . | we investigate the non - gaussianities of inflation driven by a single scalar field coupling non - minimally to the einstein gravity .
we assume that the form of the scalar field is very general with an arbitrary sound speed . for convenience to study , we take the subclass that the non - minimal coupling term is linear to the ricci scalar @xmath0 . we define a parameter @xmath1 where @xmath2 and @xmath3 are two kinds of slow - roll parameters , and obtain the dependence of the shape of the 3-point correlation function on @xmath4 .
we also show the estimator @xmath5 in the equilateral limit . finally , based on numerical calculations ,
we present the non - gaussianities of non - minimal coupling chaotic inflation as an explicit example . |
the study of stellar - forming regions ( sfr ) and young clusters is key for a complete understanding of cloud collapse and for evaluating star - formation mechanisms .
one of the main aims is the search for patterns in the phase - space ( in the classical dynamical sense of the term ) and its subsequent temporal evolution .
the spatial part of the phase space has been widely studied through a variety of studies and statistical tools ( * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* among others ) .
however , only a few clusters have been studied considering kinematic data as well @xcite , and all of them used customized exploratory hand - made analyses .
this lack of knowledge of the structure associated with the kinematic dimensions of the phase - space is due both to the scarcity of accurate and complete kinematic data and also to the absence of reliable statistical tools specifically designed for this purpose . in our previous work , @xcite
, we addressed this issue and presented a statistical tool to analyse the kinematic structure of a dataset , which we call spectrum of kinematic groupings or skg .
we showed the capabilities of this tool for distinguishing structure associated with radial velocity ( rv ) using a set of test cases with a wide range of configurations .
the skg can be easily implemented in any pipeline developed to mine large databases and leads to a quantitative description of the kinematic pattern allowing a comparative analysis between different clusters , environments and datasets in a homogeneous way .
this makes the skg a suitable procedure for analysing the amount of data expected either from ground - based projects such as apogee @xcite , lamost @xcite , ges @xcite , and weave @xcite , or from the gaia space mission @xcite . in this work
we will show the potential of this method for analysing a set of real data from the star - forming region ngc 2264 .
we have chosen this area for two main reasons : first of all , it is a particularly interesting region of the sky where the existence of a spatial pattern of rv has previously been detected @xcite , and at the same time it is one of the regions with most active star formation , containing ob stars , pms stars , ionized , neutral and molecular gas interacting in the complex process of forming stars .
ngc 2264 is a very interesting region , located in the monoceros ob1 association , in the third galactic quadrant .
this is one of the least obscured regions in the galactic plane @xcite .
its line of sight may be crossing different galactic features , such as the local arm , the perseus arm , the outer arm and the monoceros ring .
ngc 2264 is relatively near , at a distance between 750 and 900 pc @xcite .
@xcite discovered for the first time a well - populated sequence of pre - main - sequence stars in this region , in accordance with the young age of its members , between 1 and 6 myr @xcite and other hints indicating that star formation has occurred in different regions of the molecular cloud over the last several myr @xcite .
the large amount of molecular gas remaining in the various cloud cores indicates that star formation may continue in the region for several myr . despite the gas , ngc
2264 shows low reddening @xcite , which makes the study of its members easier .
based on optical photometric data of the region , @xcite identified two dense sfrs ( s mon and cone in the northern and southern parts of the area , respectively ) , and a low density halo surrounding them . later work by @xcite further refined the cone sfr , finding two substructures , the spokes cluster and the cone core .
all these reasons have made ngc 2264 a traditional laboratory for studying star formation , and there is a long list of works on this object , focusing on different topics , such as the properties of the gas @xcite , stellar content @xcite , and their interaction and feedback in the cluster region ( see * ? ? ?
* ; * ? ? ?
* and references therein ) .
in this work we will focus on the results obtained by @xcite and @xcite , which showed a well - structured spatial pattern associated with the radial velocity ( rv ) .
@xcite performed a careful and individualized analysis of the data , finding a north - south gradient in rv and spatial substructure associated with rv consistent with the structure of the molecular gas in the region .
@xcite expanded the sample from @xcite to perform a new analysis , and found a new population of stars , in this case systematically blueshifted from the molecular gas .
we will use the complete sample from @xcite to check the quality of the results obtained with the skg , comparing them with the results from their careful , individualized analysis .
the paper is divided into four sections , the first being this introduction .
the description of the procedure is shown in section 2 , and its application to the star - forming region ngc 2264 and the main conclusions of the study are presented in sections 3 and 4 , respectively .
the search for phase - space structure in stellar systems requires specific tools that respond to different concepts of what a stellar grouping is .
here we consider the existence of a clumpy velocity pattern where there are velocity ranges ( channels ) whose spatial distribution is more concentrated than that of the whole kinematic interval . in this section
we briefly describe the method , but refer to @xcite for a thorough description of the foundation and procedure .
the essence of the method is to examine a group of stars associated with a radial velocity channel , and calculate its _ kinematic index _
@xmath0 , which compares its concentration with that of a group of the same size , representative of the whole sample . to measure the concentration of a group of stars
we will use the median edge length of its euclidean minimum spanning tree ( henceforth mst ) .
the procedure designed to search for kinematic segregation starts from a sample of stars in a region of the sky , including both potential cluster members and field stars , preferably with accurate spatial coordinates and precise radial velocity data .
first we sort the data by radial velocity values and separate the sample into bins of size @xmath1 with an incremental step @xmath2 , so we scan the whole rv domain .
we then obtain the mst of each bin and calculate its median edge length .
we also calculate the median edge length of the mst of 500 groups of size @xmath1 randomly extracted from the whole sample , and consider the mean of the 500 medians as the size of a group of @xmath1 elements representative of the whole sample .
we estimate the kinematic index associated with each velocity channel , @xmath3 , @xmath4 , as @xmath5 where @xmath6 the median edge length of the mst of the bin @xmath7 , and @xmath8 is the mean of the median edge length of the 500 stochastic groups of size @xmath1 .
when @xmath9 then the bin @xmath7 is more concentrated than the representative group .
a plot of @xmath10 against the median of the radial velocity @xmath11 in the @xmath3 interval for all j , constitutes the _ spectrum of kinematic groupings , or skg , _ which provides an interesting and useful tool for the exploratory analysis of the stellar - system phase space .
those @xmath10 that verify the inequality @xmath12 mark the radial velocity channels @xmath3 where kinematic structure or segregation has been detected .
. [ eq3 ] is a conservative criterion where @xmath10 is larger than unity with confidence level greater than 95% . after applying this method
, we get eight variables associated with each rv bin : four of them correspond to its location in the cluster and its spatial extent ; another two are the radial velocity median and dispersion ; and the last two are the kinematic index and its uncertainty .
we now apply the method described in the previous section to the case of ngc 2264 .
our sample consists of the ( ra , dec , rv ) data from @xcite , downloaded from the vizier catalogue @xcite , for 995 stars ( of which 407 are very likely members according to the authors ) in the field of ngc 2264 with a very broad range of rv .
the stars are mainly distributed along a north - south strip with two apparent concentrations or lobes , consistent with the structure described in @xcite .
the s mon sfr would correspond to the northern lobe and the cone sfr to the southern .
there is also a medium density structure in the central western area of the region .
the first step was changing the coordinates to angular distances , @xmath13ra and @xmath13dec , to a centroid of the area , in this case the average position in ( ra , dec)=(100.189 , 9.691 ) .
this was done so the position coordinates used are more coherent with the euclidean distances we use in the mst calculations .
we started by feeding our algorithm with all the spatial and rv data of all the stars independently of the rv measurement quality or membership probability , to check the performance of the method in a realistic situation of contaminated , possibly biased samples with error .
in addition , we set the bin size to @xmath14 , the nearest integer to the square root of the number of stars to analyse , and step @xmath15 , as would be done in an automatic pipeline implementation .
.characteristics of selected groups a and b , and the separated components of a : a1 , a2 , and a3 . from left to right : number of stars , mean radial velocity , median radial velocity and velocity dispersion [ cols="<,^,^,^,^",options="header " , ] [ tabla3 ]
* with the skg we can visualize , at a glance , the kinematic behaviour of the population in a stellar system .
the algorithm is easy to implement in any pipeline aimed at analysing the phase space of stellar systems and has been specially designed for the study of large amounts of quality data , such as those expected from the gaia mission .
although in this paper we have focused on the analysis of the radial velocity , the method can be easily extrapolated to other velocity components , such as the proper motions . * the application of the skg method to the star - forming region ngc 2264 enabled us to highlight the excellent capabilities of the method for analysing the phase - space structure of a real dataset of stars .
moreover , the iterated hierarchical application of the algorithm shows the kinematic structure at different scales .
* when we automatically calculate the skg of the raw data we find two large significant structures ( groups a and b ) with very different distributions both in space and rv .
group a is clearly associated with the central positions and radial velocities of ngc 2264 and shows hints of internal substructure , and group b is associated with large radial velocities and spread towards the outskirts of the spatial domain .
the appearance of substructure in the outer area was an unexpected result from our method . *
the skg allows us to distinguish structure associated with kinematic populations in star - forming regions . depending on the scale of the sample
, we can find structure associated with stellar clusters , substructure in stellar clusters or even structure in field - related velocity channels .
* when we hierarchically iterate the analysis and study the members of group a we find at least four different subgroups separated in radial velocity , consistent with the complex kinematic structure manually found by previous works @xcite , which also found dynamical coupling of some of the structures with the surrounding gas .
the results obtained by our analysis are also selfconsistent . *
the structures found in the northern and southern lobes of ngc 2264 can be associated with older and younger regions , respectively , found by @xcite .
these resulst suggest that the kinematic structure of the phase space can be used to trace the particularities of star formation in a specific region . *
the members of group b are in the red - giant locus of the nir colour - colour diagram of the whole sample
. this may hint at b group members being giants far away in the background of ngc 2264 .
we have roughly estimated the distance to these stars with two different methods and obtained a median kinematic distance of 6.4 kpc and a median photometric distance of 7.0 kpc , both with large uncertainty .
these rough distance estimates and the rv values are compatible with distant structures such as the outer arm or the monoceros ring , but we can not distinguish between them with our current data .
* in an area as complex as this one , the spatial structure associated with radial velocity found with our method could be roughly associated with the tridimensional structure of the galactic disk . in this case
, the skg would work as a raw tomography of the galactic structure in the direction of the line of sight .
* the structure associated with the phase - space can be used to discover populations associated with other variables not included in the analysed dataset , such as age or distance , given the intimate relationship it has with star - formation processes and the dynamical evolution of clustered stars .
we acknowledge support from the spanish ministry for economy and competitiveness and feder funds through grant aya2013 - 40611-p . this research has made use of the simbad database and the vizier catalogue access tool , cds , strasbourg , france .
this work makes use of euro - vo software , tools or services .
the euro - vo has been funded by the european commission through contracts ri031675 ( dca ) and 011892 ( vo - tech ) underthe 6th framework program and contracts 212104 ( aida ) , 261541 ( vo - ice ) , 312559 ( cosadie ) under the 7th frame - work program . | in this work we analyse the structure of a subspace of the phase space of the star - forming region ngc 2264 using the spectrum of kinematic groupings ( skg ) .
we show that the skg can be used to process a collection of star data to find substructure at different scales . we have found structure associated with the ngc 2264 region and also with the background area . in the ngc 2264 region ,
a hierarchical analysis shows substructure compatible with that found in previous specific studies of the area but with an objective , compact methodology that allows us to homogeneously compare the structure of different clusters and star - forming regions .
moreover , this structure is compatible with the different ages of the main ngc 2264 star - forming populations .
the structure found in the field can be roughly associated with giant stars far in the background , dynamically decoupled from ngc 2264 , which could be related either with the outer arm or monoceros ring .
the results in this paper confirm the relationship between structure in the rv phase - space subspace and different kinds of populations , defined by other variables not necessarily analysed with the skg , such as age or distance , showing the importance of detecting phase - space substructure in order to trace stellar populations in the broadest sense of the word .
[ firstpage ] stellar clusters
star - forming regions radial velocity kinematic structures minimum spanning tree ngc 2264 . |
the investigation of the magnetic properties of ferromagnetic ultrathin films is a field of intense current interest.@xcite among the different experimental methods the measurement of the magnetic susceptibility @xmath0 is a very powerful method for the analysis of such thin film systems.@xcite the singularity of @xmath0 corresponds to the onset of a ferromagnetic state , i.e. to the occurrence of a nonvanishing magnetization @xmath1 for temperatures below the ( ferromagnetic ) curie temperature @xmath2 .
for @xmath3 the _ inverse susceptibility _ @xmath4 exhibits the linear ( curie - weiss- ) behavior : @xmath5 .
the paramagnetic curie temperature @xmath6 is obtained from the extrapolation of this linear behavior to @xmath7 , which corresponds to the curie temperature calculated in the mean field approximation.@xcite for an isotropic ferromagnet the behavior of @xmath0 does not depend on the lattice orientation . in the collectively ordered magnetic state the direction of the magnetization is determined by magnetic anisotropies , which are the _ free energy differences _ between the hard and easy magnetic directions . due to their relativistic origin resulting from the spin - orbit interaction
they are usually much smaller than the isotropic exchange .
as obtained in experiments the anisotropies depend on temperature and are expected to vanish above the curie temperature.@xcite it is known from general considerations@xcite that the mentioned singularity ( or maximum ) of the susceptibility is only observed if @xmath0 is measured along _ easy _ magnetic directions .
corresponding experiments have been performed for bulk systems,@xcite a thin film investigation has been reported for the fe / w(110 ) system.@xcite thus , a signature of the anisotropy is also present in the paramagnetic state above @xmath2 .
at first we comment on the fact that the anisotropy is noticeable also for @xmath8 .
we like to stress the fact that the _ microscopic _ anisotropy , e.g. the single - ion uniaxial anisotropy @xmath9 as present in a heisenberg hamiltonian , is different from the _ effective _ , temperature dependent anisotropy @xmath10 as measured for a collectively ordered magnetic state.@xcite the effective anisotropy is equal to the microscopic one for @xmath11 , thus @xmath12 .
when treated as a _ perturbation _ to the exchange interaction , @xmath10 indeed vanishes at @xmath2.@xcite however , a vanishing effective anisotropy for @xmath8 does not indicate that the microscopic anisotropy vanishes either , or that the underlying spin - orbit coupling is strongly varying with temperature .
a noticeable drop of the spin - orbit coupling is expected on a larger temperature scale.@xcite thus , a single magnetic moment in the paramagnetic state is still subject to the anisotropy even if the net magnetization is zero . here
a free energy difference between the easy and hard magnetic directions is also present ( paramagnetic anisotropy ) , exhibiting a temperature behavior as @xmath13 for @xmath14 , with @xmath15 the boltzmann constant.@xcite evidently , the paramagnetic anisotropy is rather small as @xmath9 is small compared to the exchange interaction @xmath16 . in the present study
we will show that the anisotropy although small has a sizable effect on the susceptibilities in the paramagnetic state of ultrathin films in particular when approaching the curie temperature . whereas a vast amount of susceptibility data are available for various systems,@xcite to our knowledge
the different behavior of @xmath0 measured along the easy and hard magnetic directions has not been exploited to gain information about thin films . in this paper
we report measurements of @xmath0 for ferromagnetic ultrathin co film grown on a vicinal cu substrate .
this system exhibits an _ in - plane _ two - fold ( uniaxial ) magnetic anisotropy due to the presence of regularly distributed steps in the cu surface , with the easy axis directed along the steps.@xcite we find strong differences for the magnetic susceptibilities along the easy ( @xmath17- ) and hard ( @xmath18- ) in - plane magnetic directions . with the help of an anisotropic heisenberg model solved within a many - body green s function method we are able to perform a quantitative comparison with experiments .
furthermore , we demonstrate how the exchange interaction and magnetic anisotropy can be extracted from these susceptibilities . in the previous study on
fe / w(100 ) films only a qualitative comparison in the framework of a renormalization treatment was possible.@xcite our present approach represents a new method to study quantitatively magnetic properties of ultrathin ferromagnetic films with the help of high - accuracy susceptibility measurements above @xmath2 . the paper is organized as follows . in secs.ii and
iii we describe the experimental methods and the theoretical model .
results from measurements and calculations are presented in sec.iv .
a discussion and conclusion is given in sec.v .
the experiments , including film preparation and investigation of the magnetic properties , were performed under ultra high vacuum ( uhv ) conditions ( base pressure @xmath19 torr ) in the same chamber .
the film characterization and surface preparation were made via auger electron spectroscopy ( aes ) , low energy electron diffraction ( leed ) and medium energy electron diffraction ( meed ) .
vicinal cu(11n ) surfaces with @xmath20 are used as templates for the ultrathin co films .
these surfaces have been well studied by means of helium scattering@xcite and scanning tunneling microscopy ( stm).@xcite microscopically , the surfaces consist of terraces with the normal oriented along the ( 001 ) direction , and an average terrace width of @xmath21 atomic distances .
the terraces are separated by mono - atomic steps which are aligned along the [ @xmath22 in - plane direction .
step bunching has not been observed .
the substrate was cleaned and prepared by cycles of ar sputtering ( 600 ev ) and subsequent annealing ( @xmath23c ) .
the quality of the surface structure was confirmed via leed and meed .
pronounced splitting of regular lattice spots were found , indicating the periodic step arrangement on the surface .
the co films were grown at @xmath24c with a rate of one monolayer ( ml ) per minute . during electron beam evaporation
the pressure did not exceed @xmath25 torr .
the growth process was monitored by measuring meed intensity oscillations , which were used for thickness calibration .
the magnetic characterization has been performed _ in situ _ using the longitudinal magneto - optic kerr effect ( moke).@xcite hysteresis loops were obtained in static magnetic fields up to @xmath26 mt . with the same optical setup the magnetic susceptibility @xmath0 was studied . for that purpose ,
the change of the kerr ellipticity in an applied _ ac _ magnetic field has been determined via phase sensitive detection .
no additional bias fields were used , hence the zero field _ ac _
susceptibility is measured.@xcite in this paper we monitor the static susceptibility , which is obtained for sufficiently low magnetic fields and frequencies .
the modulation frequency was set to approximately 110 hz , while the modulation amplitude @xmath27 has been varied on purpose .
@xmath0 was measured for different in - plane directions of the applied magnetic field , with angular uncertainties of about @xmath28 . due to the optical setup
we have monitored the magnetic response along the magnetic field direction ( longitudinal susceptibility ) .
sample heating is a very delicate issue while performing zero - field measurements . at first
the heating has to be performed quasi - statically to achieve an equilibrium phase transition .
secondly , any spurious magnetic fields have to be prevented as they will alter the manifestation of the phase transition , particularly when investigating the zero field susceptibility . for the latter reason we have used an external light source for heating the front side of the sample . due to this arrangement
some scattering of our data appear as the film warms up quicker than the thermocouple , located at the rear side of the sample , on every change of the radiation intensity .
this affects the measurements and causes some of the apparently periodic modulations in the plots for @xmath4 .
to calculate the susceptibility of the ferromagnetic ultrathin film with an in - plane uniaxial anisotropy we apply an anisotropic heisenberg hamiltonian with localized three - component spins @xmath29 with spin quantum number @xmath30 on lattice sites @xmath31 : @xmath32 a thin film with @xmath33 atomic layers is assumed , spanned by the @xmath34- plane .
@xmath35 is the isotropic exchange interaction between spins @xmath31 and @xmath36 .
the last term is the zeeman energy , with the magnetic field @xmath37 confined to the film plane , where @xmath38 is the land factor and @xmath39 the bohr magneton .
the uniaxial magnetic anisotropy within a heisenberg hamiltonian is usually represented by a _ magneto - crystalline single - ion anisotropy _ , @xmath40 , impling a spin quantum number @xmath41 .
however , such a single - ion anisotropy complicates considerably the solution with the method applied in this study.@xcite thus , for simplicity we consider an _ exchange anisotropy _
@xmath42 between nearest neighbor spins .
although originating from very different physical mechanisms , the anisotropic properties obtained from a single - ion term and an exchange anisotropy are quite similar , if one assumes @xmath43 , with @xmath44 and @xmath45 the coordination number.@xcite a positive value for @xmath46 indicates the easy direction to be parallel to the @xmath17- axis .
note that ferromagnetic thin films with a strong surface anisotropy sometimes exhibit a magnetization perpendicular to the film plane .
the interpretation of the susceptibility of such a system is more complicated due to the shape anisotropy resulting from the dipole interaction .
the magnetic ground state in this case is a stripe - domain structure , and not the single - domain state.@xcite this complication vanishes for an _ in - plane _ uniaxial anisotropy , since then the dipole interaction favors a single - domain , ferromagnetic ground state .
thus , this coupling is not considered explicitely in the present study .
the hamiltonian eq.([e1 ] ) is solved by a many - body green s function approach,@xcite which is suited to consider simultaneously both expectation values @xmath47 and @xmath48.@xcite furthermore , collective magnetic excitations ( spin waves ) are taken into account , which are particularly important for low - dimensional systems . the long range magnetic order of an isotropic two - dimensional heisenberg magnet becomes unstable against collective magnetic excitations with long wavelengths ( mermin - wagner - theorem).@xcite already weak anisotropies , however , induce a magnetization with a curie temperature of the order of the exchange coupling.@xcite we consider the set of anticommutator green s functions in frequency space , @xmath49 , where the ladder operators @xmath50 have been introduced , and @xmath51 .
these green s functions are solved in the usual way by the equation of motion.@xcite the vanishing eigenvalues occurring in the set of equations motivate the use of _ anticommutator _ green s functions .
the higher - order green s functions appearing within this procedure are approximated by the generalized tyablikov - decoupling ( rpa ) for @xmath52,@xcite @xmath53 interactions between spin waves are partly taken into account , allowing for the determination of the magnetic properties with a reasonable accuracy in the whole temperature range .
it has been shown recently that the magnetization and the curie temperature of a weakly anisotropic @xmath54- heisenberg monolayer calculated by this approach agrees very well with the values as obtained from a quantum monte carlo method.@xcite the set of linear equations of the corresponding correlation functions @xmath55 can be solved numerically for films with an arbitrary number @xmath33 of layers , for inhomogeneous coupling constants @xmath35 and @xmath56 , and for arbitrary spin quantum numbers @xmath30 .
this will be investigated in a forthcoming study.@xcite in the remainder of this section we present two cases for which analytical solutions can be derived .
first , to give some insight in the structure of the solutions we consider a homogeneous square ( 001 ) ferromagnetic monolayer ( @xmath57 and @xmath58 ) with spins @xmath59 .
the coupling constants are put equal to @xmath60 and @xmath61 if @xmath31 and @xmath36 are nearest neighbors , and zero otherwise .
a fourier transformation into the two - dimensional wave vector space @xmath62 is applied . by considering the properties for @xmath59- spin operators ,
the magnetization components @xmath63 and @xmath64 are given implicitely by the equations @xmath65 with the denotations @xmath66+g\,\mu_b\,b_z \ , , \label{e5 } \\
\tilde h_x & = & m_x\,j\,(q_\|-\gamma_\mathbf{k})+g\,\mu_b\,b_x \ , , \label{e6 } \\
h_x & = & m_x\,[j\,(q_\|-\gamma_\mathbf{k})-d\,\gamma_\mathbf{k } ] + g\,\mu_b\,b_x \ , , \label{e7 } \end{aligned}\ ] ] where @xmath67 $ ] , @xmath68 the lattice constant , and @xmath69 the number of * k*-points in the first brillouin zone . from eqs.([e3a ] ) and ( [ e3b ] ) the susceptibilities @xmath70 and @xmath71 along the easy and hard axes , which we denote by parallel and transverse susceptibilities , will be determined numerically .
the curie temperature @xmath72 is calculated from @xmath73 note that the value of @xmath2 is determined not only by the isotropic exchange interaction @xmath74 but depends also on the exchange anisotropy @xmath46.@xcite the mean field approximation ( mfa ) is obtained by putting @xmath75 in eqs.([e5 ] ) ( [ e8 ] ) , yielding the corresponding ordering temperature @xmath76 .
we point out that the paramagnetic curie temperature @xmath77 calculated within the green s function method is equal to @xmath78.@xcite secondly , to allow for a quantitative comparison with the 2 ml co / cu thin film system as investigated experimentally in the present study , we consider a homogeneous fcc ( 001 ) film with @xmath79 layers and spins @xmath59 . for a magnetic field along the easy ( @xmath17- ) axis ( @xmath80 )
we calculate the magnetization component @xmath64 from @xmath81 denoting @xmath82 where @xmath83 is the coordination number between neighboring layers and @xmath84 . for comparison a simple - cubic ( 001 ) film
is characterized by @xmath85 .
the curie temperature for this two - layer film is given by @xmath86 from a fit to experimental data the coupling constants are determined .
this can be done by using both @xmath87 and @xmath88 , see eqs.([e3a ] ) and ( [ e3b ] ) .
alternatively , on can employ solely @xmath87 , since the increase of @xmath89 for @xmath90 depends sensitively on the anisotropy .
the latter method is used for the determination of @xmath74 and @xmath46 of the present 2 ml case , since for 2 ml an analytical expression is only available for @xmath64 , see eq.([e9 ] ) .
a corresponding expression for @xmath63 along the hard - axis needs for additional numerical work .
the transition temperatures for the co / cu(001 ) and the co / cu(1 1 17 ) thin films exhibit a similar dependence on film thickness.@xcite due to the high instability against surface diffusion of co films on cu,@xcite and the steep increase of @xmath2 with increasing film thickness,@xcite a very thin co film of about 2.2 ml was chosen with @xmath2 well below @xmath91c.@xcite the real part of the susceptibility @xmath0 in si units as a function of the temperature is shown in fig.1 . both parallel and transverse susceptibilities for magnetic field directions along the easy and hard axes are displayed .
the semi - logarithmic plot allows for a comparison of both quantities .
the inset shows the corresponding linear plot .
the parallel susceptibility exhibits a strong peak at @xmath92 k , with a full width half maximum ( fwhm ) of 2.7 k for the actual modulation amplitude @xmath93 t .
the fwhm can be reduced to values around 1.5 k for smaller modulation amplitudes.@xcite for @xmath94 t an imaginary part of the parallel susceptibility is observed , while for the smallest applied modulation field ( @xmath95 t ) the imaginary part vanishes.@xcite while a peak was found in the susceptibility @xmath70 along the easy axis , the magnetic response @xmath71 along the hard axis was not detectable for small @xmath96 t . in order to increase the magneto - optical signal and thus the detection limit ,
the modulation amplitude along the hard axis was increased by a factor of five .
however , the response is still small , exhibiting a weak maximum near @xmath2 , as shown in fig.1 .
obviously , the phase transition is reflected only weakly in the transverse susceptibility signal .
thus , the phase transition for this thin film system with an in - plane magnetization exhibits a strongly anisotropic behavior .
[ fig2 ] fig.2 displays the temperature dependence of the _ inverse _ parallel and transverse susceptibilities @xmath89 and @xmath97 around the curie temperature . for elevated temperatures @xmath98c the scattering of the data points is strong , and the temperature dependence of @xmath89 and @xmath97 can not be given precisely .
as mentioned in sec.ii this scattering is due to the fact that the signal becomes very small , and that the measuring time can not be increased appropriately as the system properties might change due to the onset of surface diffusion .
the oscillations obtained for @xmath97 are most likely caused by too large steps of changes of the heating power , which were indeed larger than in case of @xmath89 .
we have not systematically explored these effects . on the other hand , within the temperature range between @xmath99 and @xmath100c the behavior of @xmath89 and @xmath97
is clearly resolved .
we remark that the inverse susceptibilities can not be described by straight lines in this temperature range , rather they exhibit an upward curvature .
evidently , @xmath97 is shifted upwards with respect to @xmath89 by an almost constant , temperature independent amount .
thus , at any temperature above @xmath2 the inverse susceptibility @xmath97 along the hard axis is larger than the corresponding value @xmath89 along the easy axis .
this is in accordance with measurements for bulk magnets,@xcite and is also expected theoretically.@xcite we will show that the temperature range as displayed in fig.2 is still far below the linear regime of the inverse susceptibilities .
[ fig3 ] at first we have carried out corresponding calculations for the parallel susceptibility of a fcc ( 001 ) ferromagnetic film with two atomic layers using eq.([e9 ] ) .
the atomic magnetic moment @xmath101 and the atomic volume @xmath102 appropriate for bulk co are assumed .
the exchange coupling @xmath74 and the exchange anisotropy @xmath46 are obtained by fitting the results calculated with eq.([e9 ] ) to the measured @xmath87 in the temperature range between @xmath99 and @xmath103c .
we obtain @xmath104 mev / bond , which is close to the co bulk value @xmath105 mev / bond .
furthermore , we get @xmath106 mk / atom for the exchange anisotropy , which corresponds to the single - ion in - plane uniaxial anisotropy @xmath107 . from this value
the anisotropy energy density is calculated to be @xmath108 .
this value should be compared with the one obtained from independent measurements of the effective anisotropy for the same system.@xcite by means of a thermodynamic perturbation theory@xcite we calculate the _ microscopic _ anisotropy energy density at @xmath11 to be @xmath109 from the data of ref.@xcite , which is in reasonable agreement with the result based on the measured susceptibilities presented here .
the resulting small value of @xmath110 is comparable to the 6th - order anisotropy energy density @xmath111 in the hexagonal plane , and is about 20 times smaller than the 2nd - order uniaxial anisotropy energy density @xmath112 of bulk hcp co.@xcite the calculated inverse parallel susceptibility @xmath89 is depicted in fig.2 .
a good agreement with experiment is obtained .
we note that an increase of the anisotropy @xmath46 will result in a corresponding decrease of @xmath87 for @xmath8 .
we emphasize that the susceptibilities measured in the accessible temperature range up to 155@xmath113c is still far below the linear ( curie - weiss- ) range .
as mentioned in the introduction , the paramagnetic curie temperature extrapolated from the inverse susceptibilities as calculated by the green s function method is equal to the curie temperature @xmath78 obtained from the mean field approximation.@xcite for the parameters @xmath74 and @xmath46 as given above we obtain @xmath114 k. only for temperatures above @xmath78 a curie - weiss behavior will emerge .
the difference between @xmath2 and @xmath78 and thus the range of the curved behavior of the inverse susceptibilities depicted in fig.2 is determined to be very large for this ultrathin film .
the reason is that the influence of collective magnetic fluctuations is much stronger in such two - dimensional systems as compared to bulk magnets . finally , we like to demonstrate the behavior of @xmath89 and @xmath97 in a large temperature range above @xmath2 . in fig.3
the calculated inverse parallel and transverse suceptibilities for a _ single _ square ferromagnetic layer are shown , using eqs.([e3a ] ) and ( [ e3b ] ) . in order to reveal clearly the difference between @xmath89 and @xmath97 a strong exchange anisotropy @xmath115 is assumed for these calculations .
the temperature is given in units of @xmath116 .
for @xmath8 both susceptibilities differ by a temperature - independent shift , exhibiting the same curvature .
the linear behavior of @xmath89 and @xmath97 is reached for elevated temperatures @xmath3 , where the inverse susceptibilities approach the ones obtained from the mfa .
their extrapolations yield the characteristic temperatures @xmath117 and @xmath118 , which differ by the anisotropy @xmath119 .
the larger value @xmath120 is the paramagnetic curie temperature @xmath121 .
we point out that the behavior of the inverse susceptibilites as calculated by the mfa is even qualitatively wrong , since it predicts a curie - weiss behavior for @xmath4 for all temperatures above the curie temperature .
in contrast , both inverse susceptibilities exhibit a considerably different behavior for @xmath122 .
@xmath89 vanishes at @xmath99 for an infinitely small modulation amplitude , and increases strongly for @xmath123 . on the other hand
, @xmath97 merely exhibits a cusp at @xmath2 , and assumes the constant value @xmath124 in the ferromagnetic phase .
thus , one could directly extract the anisotropy @xmath46 by measuring @xmath125 , and determine @xmath74 from @xmath2 .
this is an alternative treatment to the one as applied for the 2 ml case where both coupling constants are derived from the parallel susceptibility @xmath87 alone .
however , one has to make sure that secondary processes in the ordered phase are not effective .
the apparent peak of the transverse susceptibility observed in ref.@xcite was attributed to vortex- and domain wall - excitations .
these result in a non - constant behavior of @xmath97 for @xmath123 , as is also seen in our experiments presented in fig.2 . for a quantitative comparison such domain processes or multi - domain states have to be considered as well .
these complications are not expected to occur for @xmath8 .
we have measured the parallel and transverse magnetic susceptibilities @xmath87 and @xmath88 of a 2.2 ml co film grown on a vicinal cu(1 1 17 ) surface .
corresponding calculations have been performed within an anisotropic heisenberg model by applying a many - body green s function method .
we have demonstrated that the curie temperature @xmath2 and the susceptibility in the paramagnetic regime ( @xmath126 ) gives access to the exchange interaction @xmath74 and the anisotropy @xmath46 of ultrathin films .
using these coupling parameters a quantitative agreement between experiment and theory is obtained at least in the limited temperature range accessible by our measurements .
it is evident that this temperature range is far below the crossover to the curie - weiss behavior .
any extrapolation @xmath127 from the experimental data , assuming apparently linear parts in fig.2 , will yield a wrong value for the paramagnetic curie temperature @xmath128 . to reach the linear behavior of the susceptibilities the measurements have to be extended to temperatures around @xmath129c , which is impossible for the co / cu thin film system due to its instability against surface diffusion and alloying .
the good agreement between theory and experiment justifies a methodological generalization for exploring magnetic properties by investigating the susceptibility in the paramagnetic regime .
while the effect of the anisotropy in the paramagnetic regime for bulk systems is known for a long time,@xcite for ultrathin ferromagnetic films improved theoretical approaches considering collective magnetic fluctuations have to be applied .
a successful realization of corresponding experiments is challenging as the susceptibilities have to be measured in very small modulation fields with very high accuracy .
the measurements should be extended to as large temperatures as possible in order to allow for a comparison with theory over a wide temperature range . from such
a comparison values for isotropic exchange interactions and anisotropies as present in the heisenberg hamiltonian eq.([e1 ] ) can be extracted .
note that these quantities refer to constant _ microscopic _ parameters and not to _ effective _ anisotropies measured at finite temperatures.@xcite in the previous study on fe / w(110 ) films a quantitative comparison between experiment and theory has not been drawn.@xcite here a different theoretical approach has been applied , namely a renormalization group treatment , allowing at the most only for a qualitative comparison .
thus the coupling constants for thin films can not be determined by this method .
it should be mentioned that the polyakov renormalization scheme applied in ref.@xcite has been strongly criticized .
several authors argue that this scheme might not be applicable for two - dimensional ferromagnets.@xcite in future theoretical work we will explore the behavior of the magnetic susceptibility with increasing film thickness @xmath33 .
the range of the curved behavior of @xmath0 for @xmath130 , which is very pronounced for ultrathin films as considered in the present study , is expected to reduce for thick films , approaching the one of the corresponding bulk system . with an improved theory for general spin quantum numbers @xmath30
the consideration of single - ion anisotropies becomes feasible .
such anisotropies are more appropriate for @xmath131- transition metal magnets .
a similar treatment for a _ perpendicular _ anisotropy needs the additional consideration of the magnetic dipole coupling .
anisotropies with a different symmetry , e.g. a _
quartic _ in - plane anisotropy , are also accessible within the scope of such a treatment , resulting in a considerably different behavior of the susceptibilities with respect to the uniaxial case .
99 see for example , _ ultrathin magnetic structures i + ii _ , edited by b. heinrich and j. a. c. bland ( springer verlag , berlin , 1994 ) .
w. wulfhekel , s. knappmann , b. gehring , and h. p. oepen , phys .
b * 50 * , 16074 ( 1994 ) ; a. berger , a. w. pang , and h. hopster , j. magn . magn . mater .
* 137 * , l1 ( 1994 ) ; a. aspelmeier , m. tischer , m. farle , m. russo , k. baberschke , and d. arvantis , j. magn .
magn . mater . * 146 * , 256 ( 1995 ) ; d. venus , c. s. arnold , and m. dunlavy , phys .
b * 60 * , 9607 ( 1999 ) ; f. beusch , g. garreau , r. moosbhler , and g. bayreuther , j. appl .
phys . * 89 * , 7133 ( 2001 ) .
s. v. tyablikov , _ methods in the quantum theory of magnetism _ ( plenum press , new york , 1967 ) ; w. nolting , _ quantentheorie des magnetismus _
( b. g. teubner , stuttgart , 1986 ) , vol.2 . h. b. callen and e. r. callen , j. phys .
solids * 27 * , 1271 ( 1966 ) ; y. millev and m. fhnle , phys .
b * 51 * , 2937 ( 1995 ) ; p. j. jensen and k. h. bennemann , in _ magnetism and electronic correlations in local - moment systems : rare earth elements and compounds , _ edited by m. donath , p. a. dowben , and w. nolting ( world scientific , singapore , 1998 ) , pp . 113 140
. p. pfeuty and g. toulouse , _ introduction to the renormalisation group and to critical phenomena _ ( wiley , london , 1977 ) .
h. suzuki and t. watanabe , j. phys .
japan * 30 * , 367 ( 1970 ) . c. h. back , a. kashuba , and d. pescia , phys . low - dim
* 2 * , 9 ( 1994 ) .
t. moos , w. hbner , and k. h. bennemann , solid state commun .
* 98 * , 639 ( 1996 ) .
p. j. jensen et al . , unpublished .
a. berger , u. linke , and h. p. oepen , phys .
lett . * 68 * , 839 ( 1992 ) .
j. lapujoulade , j. perreau , and a. kara , surf .
sci . * 129 * , 59 ( 1983 ) .
j. frohn , m. giesen , m. poensgen , j. f. wolf , and h. ibach , phys .
* 67 * , 3543 ( 1991 ) .
a. berger , s. knappmann , and h. p. oepen , j. appl .
phys . * 75 * , 5598 ( 1994 ) .
h. p. oepen , s. knappmann , and w. wulfhekel , j. magn .
* 148 * , 90 ( 1995 ) .
w. wulfhekel , s. knappmann , and h. p. oepen , j. magn .
magn . mater . *
163 * , 267 ( 1996 ) .
p. frbrich , p. j. jensen , and p. j. kuntz , eur .
j. b * 13 * , 477 ( 2000 ) .
m. e. lines , phys .
b * 3 * , 1749 ( 1971 ) .
y. yafet and e. m. gyorgy , phys .
b * 38 * , 9145 ( 1988 ) . n. m. mermin and h. wagner , phys
* 17 * , 1133 ( 1966 ) . c. herring and c. kittel , phys
. rev . * 81 * , 869 ( 1951 ) ; w. dring , z. naturforsch .
* a16 * , 1008 ( 1961 ) ; s. v. maleev , sov .
jetp * 43 * , 1240 ( 1976 ) [ zh .
teor . fiz . *
70 * , 2374 ( 1976 ) ] ; d. a. yablonskiy , phys .
b * 44 * , 4467 ( 1991 ) .
p. henelius , p. frbrich , p. j. kuntz , and p. j. jensen , phys .
b * 66 * , 094407 ( 2002 ) .
s. knappmann , diplomathesis , rwth aachen , germany ( 1994 ) .
a. schmid , d. atlan , h. itoh , b. heinrich , t. ichinokawa , and j. kirschner , phys .
b * 48 * , 2855 ( 1993 ) .
m. giesen , f. schmitz , and h. ibach , surf .
sci . * 336 * , 269 ( 1995 ) ; f. nouvertn , u. may , m. bamming , a. rampe , u. korte , g. gntherodt , r. pentcheva , and m. scheffler , phys .
b * 60 * , 14382 ( 1999 ) . c. m. schneider , p. bressler , p. schuster , j. kirschner , j. j. de miguel , and r. miranda , phys .
lett . * 64 * , 1059 ( 1990 ) .
j. p. rebouillat , ieee trans . magn .
* 8 * , 630 ( 1972 ) ; d. m. paige , b. szpunar , and b. k. tanner , j. magn .
magn . mater . *
44 * , 239 ( 1984 ) .
a. p. levanyuk and n. garcia , j. phys .
: condens .
matter * 4 * , 10227 ( 1992 ) ; r. p. erickson and d. l. mills , phys .
b * 46 * , 861 ( 1992 ) . | we measure the magnetic susceptibility of ultrathin co films with an in - plane uniaxial magnetic anisotropy grown on a vicinal cu substrate . above the curie temperature the influence of the magnetic anisotropy can be investigated by means of the parallel and transverse susceptibilities along the easy and hard axes . by comparison with a theoretical analysis of the susceptibilities we determine the isotropic exchange interaction and the magnetic anisotropy .
these calculations are performed in the framework of a heisenberg model by means of a many - body green s function method , since collective magnetic excitations are very important in two - dimensional magnets . |
on the space @xmath5 consider the topology of uniform convergence on compact sets , the corresponding @xmath6-field of the borel sets , and @xmath7 the measure on @xmath8 under which the coordinate the processes @xmath9 are independent standard brownian motions . also let @xmath10 , and equip it with the @xmath6-field of borel sets derived from the topology of uniform convergence on compact sets . for @xmath11 ,
we denote by @xmath12 the probability measure on @xmath13 such that @xmath14 is a diffusion with @xmath15 and generator @xmath16 the construction of such a diffusion is done with scale and time transformation from a one - dimensional brownian motion ( see , e.g. , @xcite , @xcite ) . using this construction
, it is easy to see that for @xmath7-almost all @xmath11 the diffusion does not explode in finite time ; and on the same set of @xmath17 s it satisfies the formal sde @xmath18 where @xmath19 is a one - dimensional standard brownian motion . then consider the space @xmath20 , equip it with the product @xmath6-field , and take the probability measure defined by @xmath21 the marginal of @xmath22 in @xmath13 gives a process that is known as diffusion in a random environment ; the environment being the function @xmath17 .
s. schumacher ( @xcite ) proved the following result .
[ schumprop ] there is a process @xmath23 such that for the formal solution @xmath24 of it holds @xmath25 where for @xmath26 we let @xmath27 we will define the process @xmath3 soon .
this result shows the dominant effect of the environment , through the process @xmath3 , on the asymptotic behavior of the diffusion .
the results we prove in this paper concern the process @xmath3 . in subsection [ connection ] we commend on their implications for the behavior of the diffusion itself . besides this diffusion model
, there is a discrete time and space analog , known as sinai s walk , which was studied first .
sinai s pioneering paper @xcite identified the role of the process @xmath3 in the analogous to limit theorem for the walk .
then s. schumacher proved in @xcite ( see also @xcite for the results without the proofs ) a more general statement than the above proposition where the environment @xmath17 was not necessarily a two sided brownian motion , while t. brox @xcite gave a different proof in the brownian case .
h. kesten @xcite computed the density of @xmath28 in the case we consider , and tanaka @xcite generalized the computation to the case that @xmath17 is a two sided symmetric stable process .
localization results have been given for the sinai walk by golosov ( @xcite , actually , for the reflected walk ) and for the diffusion model by tanaka @xcite .
also tanaka ( @xcite , @xcite ) studied the cases where the environment is non - positive reflecting brownian motion , non - negative reflecting brownian motion , or brownian motion with drift .
finer results on the asymptotics of sinai s have been obtained by z. shi and y. hu .
a survey of some of them as well as a connection between sinai s walk and diffusion in random environment is given in @xcite .
another connection is established in @xcite . in @xcite ,
le dousal , c. monthus , and d. fisher proposed a new method for tackling questions related to asymptotic properties of sinai s walk , and using it they gave a host of results .
the method is a renormalization group analysis and it has consequences agreeing with rigorously proved results ( e.g. , @xcite , @xcite ) .
this is the starting point of the present paper . in the context of diffusion in random environment ,
we show how one can justify the renormalization group method using two tools .
the first is a path decomposition for a two sided standard brownian motion ; the second is the renewal theorem .
our main results illustrate the use of the method and the way we justify it .
the structure of the paper is as follows . in the remaining of the introduction we state our results . in section [ preparation ]
we provide all the necessary machinery for the proofs , which are given in section [ proofs ] .
some technical lemmata that we use are proved in section [ lemmata ] .
we begin by defining the process @xmath3 . for a function @xmath29 and @xmath30
we say that * @xmath17 admits an @xmath31-minimum at @xmath32 * if there are @xmath33 with @xmath34 , @xmath35\}$ ] and @xmath36 , @xmath37 .
we say that * @xmath17 admits an @xmath31-maximum at @xmath32 * if @xmath38 admits an @xmath31-minimum at @xmath32 . for convenience
, we will call a point where @xmath17 admits an @xmath31-maximum or @xmath31-minimum an @xmath31-maximum or an @xmath31-minimum respectively .
we denote by @xmath39 the set of @xmath31-extrema of @xmath17 and define @xmath40 thus , for @xmath41 and @xmath42 we can write @xmath43 with @xmath44 strictly increasing , @xmath45 , @xmath46 .
it holds that @xmath47 , and the easy proof of this fact is given in lemma [ discrete ] . the process @xmath48 is defined for @xmath42 and @xmath49 as @xmath50 and @xmath51 if @xmath52 or @xmath53 . in the definition of @xmath54
we do not make use of the entire sequence of @xmath31-extrema .
the reason we introduce this sequence is that we plan to study the evolution of the process @xmath3 as @xmath31 increases . since @xmath55 for @xmath56 , the later values of @xmath57 are elements of @xmath39 .
for @xmath58 large enough , the points @xmath59 , @xmath60 will not be @xmath58-extrema .
we will decompose the process @xmath17 at the endpoints of the intervals @xmath61:k\in\d{z}\}$ ] and study its restriction to each of them .
of course @xmath62 $ ] has a particular importance for the process @xmath3 ; and it is in the study of @xmath63 $ ] that the renewal theorem enters ( see lemma [ cslope ] ) .
[ bscaling ] it is clear that @xmath3 satisfies @xmath64 for all @xmath65 , and @xmath11 . so that the quantity @xmath66 appearing in equals also @xmath67 . for @xmath68
define on @xmath69 the random variable @xmath70.\ ] ] the main result of the paper is the computation of the generating function of @xmath71 .
[ thm1 ] for @xmath68 and @xmath72 with @xmath73 , it holds @xmath74 where @xmath75 and @xmath76 from this we extract several corollaries .
corollary [ nozero ] and corollary [ nzeros ] are immediate , while the rest require some work and are proved in section [ proofs ] .
corollary [ nozero ] follows by taking @xmath77 in .
[ nozero ] @xmath78)/ x^{(-3+\sqrt{5})/2}\to 1/2 + 7\sqrt{5}/30\ ] ] as @xmath79 .
[ markov ] the increasing process of points @xmath80 where @xmath3 changes sign in @xmath81 has the form @xmath82 , @xmath83 , where @xmath84 is the smallest such point and the @xmath85 s are i.i.d with density @xmath86 where @xmath87 . now observe that @xmath88 since @xmath89 and @xmath90 is finite a.s .
( e.g. , by corollary [ nozero ] , @xmath91 ) , the next statement follows from renewal theory .
[ nzeros ] @xmath92 as @xmath93 @xmath7-a.s .
corollary [ markov ] allows the following strengthening of the above theorem .
[ strenghening ] relation holds for all @xmath94.$ ] for @xmath95 consider the random variable @xmath96 and let @xmath97 be its distribution measure .
then the following holds .
[ ldp1 ] the family of measures @xmath98 satisfies a large deviation principle with speed @xmath99 and good rate function @xmath100 in @xcite , corollary [ markov ] appears in paragraph iv.b with a different justification .
we state it here because we need it for the proof of corollary [ ldp1 ] .
the large deviation result of corollary [ ldp1 ] is the precise mathematical interpretation of the discussion in paragraph iv.a of the same paper . in corollary [ nozero ]
, the condition @xmath101 means that the process @xmath3 tends to keep the diffusion away form zero in the time interval @xmath102 $ ] ( since the diffusion localizes around @xmath3 , and @xmath3 keeps sign on @xmath4 $ ] ) . for the event that the diffusion _ hits _
zero there are two interesting relevant papers .
the first one is by y. hu @xcite who treats the annealed asymptotics of the first time of hitting zero after time @xmath99 as @xmath103 .
the second is by f. comets and s. popov @xcite and refers to a related model .
that is , it considers a process @xmath104 on @xmath105 that runs in continuous time in an environment @xmath24 satisfying the conditions of the sinai model and studies , among other things , the asymptotics of the quenched probability @xmath106 as @xmath103 .
the results of the previous subsection concern the process @xmath3 , which is a functional of the environment . and our motivation for studying @xmath3 was the localization results involving this process ( the simplest being fact [ schumprop ] , and keep in mind remark [ bscaling ] ) .
an obvious question is what we can infer about the behavior of the diffusion from our results .
using the representation of the diffusion as a time and scale change of brownian motion , one can show easily that the diffusion is recurrent and 0 is a regular point for @xmath107 and @xmath108 .
consequently , for all @xmath109 , the diffusion visits 0 in the time interval @xmath110 and in its first visit there it scores an infinite number of sign changes .
so there can be no direct connection with the corresponding number for the process @xmath3 .
one can consider , say , the number of sign changes of the diffusion between times where the diffusion achieves a positive or negative record value .
this number is finite on compact intervals of @xmath111 but still not related with the sign changes of @xmath3 ( and it is not hard to see this ) .
when @xmath112 jumps to a new value , what happens is not that just the diffusion goes through that value shortly before or after that . as it is known ( see , e.g. , @xcite , @xcite ) , the impact of the jump is that the diffusion goes to that value and it is trapped in its neighborhood for a large amount of time .
the way to detect the approximate location of such values when we observe the diffusion in real time ( i.e. , at time @xmath99 we know @xmath113 $ ] ) is to find the site the diffusion has spent the most time thus far . to make the last statement precise
, we use the local time process @xmath114 that corresponds to the diffusion @xmath24 .
this process is jointly continuous , and with probability one satisfies @xmath115 for all @xmath116 and any bounded borel function @xmath117 . for a fixed @xmath118 , the set @xmath119 of the points with the most local time at time
@xmath99 is nonempty and compact .
any point there is called a * favorite point * of the diffusion at time @xmath99 .
define @xmath120 with @xmath121 , the smallest favorite point at time @xmath99 .
also , for @xmath122 define the interval @xmath123 . in a work in progress
we expect to prove the following .
[ fplocal ] with @xmath22 probability one , there is a strictly increasing sequence @xmath124 converging to infinity so that if we denote by @xmath125 the sequence of consecutive values that @xmath112 takes on the interval @xmath126 ( remember that @xmath3 is a step function in any interval @xmath127 with @xmath42 ) , then @xmath128 and @xmath129 for some @xmath130 .
we abbreviated @xmath131 to @xmath132 .
observe that for big @xmath31 , the interval @xmath133 is a relatively small neighborhood of @xmath31 .
thus , the claim says that , after some point , the function @xmath134 `` almost tracks '' the values of the process @xmath112 with the same order and at about the same time .
consequently , the number of sign changes of @xmath135 on an interval @xmath136 ( with @xmath137 large ) would correspond to the number of sign changes of @xmath134 on approximately the same interval . or , more precisely , the number of sign changes of @xmath134 on @xmath136 and the corresponding number for @xmath135 differ by at most two .
it is easy to see that the last statement follows from the claim above .
as a first step towards the study of the process @xmath3 , we look at the law of the brownian path between two consecutive @xmath31-extrema as well as the way these pieces are put together to constitute the entire path . the first piece of information is provided by proposition of 1 in @xcite .
[ renewal ] for every @xmath42 , the times of x - extrema of a brownian motion @xmath138 build a stationary renewal process @xmath139 with @xmath140 strictly increasing and @xmath141 the trajectories between consecutive x - extrema @xmath142 ) , k\in\mathbb{z}$ ] are independent and for @xmath143 identically distributed ( up to changes of sign ) . in lemma of 1 of the same paper @xcite a description of each such trajectory is given which we quote ( see figure [ p2 ] ) .
[ p2 ] for @xmath144 let @xmath145\},\\ \tau:&=\min\{t : m_t = w_t+x\ } , \\
\beta:&=m_\tau , \\ \sigma:&=\max\{s\in[0,\tau]:w_s=\beta\}.\end{aligned}\ ] ] [ nplaplace ] the following hold : 1 . the two trajectories @xmath146)\text{~~ and~~ } ( \beta - w_{\sigma+t}:t\in[0,\tau-\sigma])$ ] are independent .
@xmath19 has exponential distribution with mean @xmath31 .
the laplace transform of the law of @xmath147 given @xmath19 is @xmath148=\exp\{-(y / x)\phi(sx^2)\ } \text{\qquad,~s}>0.\ ] ] 4 .
the laplace transform of the law of @xmath149 is @xmath150=\psi(sx^2 ) \text{\qquad,~s}>0,\ ] ] where @xmath151 we call the translation @xmath152 $ ] of the trajectory of @xmath17 between two consecutive @xmath31-extrema an * @xmath31-slope * ( or a slope , when the value of @xmath31 is clear or irrelevant ) , a slope that takes only non - negative values an * upward slope * , and a slope taking only non - positive values a * downward slope*. we call @xmath153 $ ] the * central @xmath31-slope*. in the following we will use the operation of `` gluing together '' functions defined on compact intervals . for two functions @xmath154\to\d{r}$ ] , @xmath155\to\d{r}$ ] , by gluing @xmath156 to the right of @xmath157 we mean that we define a new function @xmath158\to\d{r}$ ] with @xmath159 , \\
f(\gb)+g(x-\gb+\gamma)-g(\gamma ) & \mbox { for } x\in [ \gb , \gb+\gd-\gamma ] . \end{cases}\ ] ] it is clear that for all @xmath160 , if @xmath152 $ ] is an upward @xmath31-slope , then it is obtained by gluing together a trajectory of type @xmath146)$ ] to the right of a trajectory of type @xmath161)$ ] and then translating the resulting path so that it starts at @xmath162 . similarly for a downward slope . for any @xmath31-slope @xmath163\to\d{r}$ ] we call @xmath164 , @xmath165 the * length * and the * height * of the slope respectively , and we denote by @xmath166 the `` * * excess height * * '' of @xmath167 . also we denote by @xmath168 , @xmath169 , [ theta ] the slopes with domains @xmath170 $ ] , @xmath171 $ ] and values @xmath172 , @xmath173 respectively .
for any slope @xmath167 , the slope @xmath174 is in the set @xmath175 defined by @xmath176 $ , $ f$ continuous , } \\ \text { and } 0=f(0)\le f(x ) \le f\big(l(f)\big ) \quad \forall x \in [ 0,l(f ) ] \end{array } \right\}.\ ] ] on @xmath175 we define a topology for which the base of neighborhoods of an element @xmath177 is the collection of all sets of the form @xmath178$}\}.\ ] ] with this topology , @xmath175 is a polish space .
equip @xmath175 with the borel @xmath6-algebra , and define the measures @xmath179 , @xmath180 the first to be the distribution of @xmath181)$ ] and the second to be the distribution of @xmath182)$ ] ( the superscripts @xmath183 and @xmath184 standing for renewal and central ) . in the remaining part of this section we compute the distribution of the length and excess height of a slope picked from @xmath179 or @xmath180 .
we assume @xmath185 since the scaling property of brownian motion gives the corresponding results for the case @xmath186 .
let @xmath167 be a slope picked from @xmath187 .
earlier we described the way an upward slope is formed .
its length is the sum of two independent independent random variables @xmath188 with @xmath189 ( where we take @xmath185 in their definitions just before fact [ nplaplace ] ) .
but from ( i ) of fact [ nplaplace ] , @xmath6 and @xmath190 are independent .
thus , @xmath191 . by definition , @xmath192 is the time where the reflected process @xmath193 hits one .
this reflected process has the same law as @xmath194 , and the laplace transform of the time it first hits one is known as @xmath195=(\cosh \sqrt{2\lambda})^{-1 } \text { for } \gl>0.\ ] ] also @xmath196 . using the laplace inversion formula
( see @xcite , pg 531 ) we find that the density of @xmath197 is @xmath198 \quad , ~x>0.\ ] ] we note for future reference that , by ( ii ) of fact [ nplaplace ] , for any @xmath199 , the excess height of a slope picked from @xmath200 is exponential with mean @xmath201 ; i.e. , it has density @xmath202 now let @xmath203 be a slope picked from @xmath204 . more specifically ,
take @xmath203 to be the central 1-slope .
observe that , by virtue of fact [ renewal ] , we can `` start a renewal process at @xmath205 '' with i.i.d . alternating upward and downward slopes , and ask what the characteristics are of the slope covering zero .
the renewal theorem says that the length of the slope covering zero is picked from the distribution of @xmath206 given in with size - biased sampling .
once the length , say @xmath207 , is picked , we expect that the remaining characteristics of the slope , ignoring direction ( upward or downward ) , are determined by the law of @xmath208 under @xmath187 . we give a formal proof of this .
notice that a regular conditional distribution for the random variable @xmath167 given the @xmath147-field @xmath209 exists because the space @xmath175 is polish .
[ cslope ] for any measurable subset a of @xmath175 , it holds @xmath210 where @xmath167 has under @xmath7 distribution @xmath187 .
let @xmath211 be the distribution function of @xmath197 , and for @xmath212 let @xmath213 be the 1-slope around @xmath99 .
that is , the slope whose domain contains @xmath99 .
then @xmath214 for all @xmath118 because @xmath215 is the same as the image under @xmath216 of the slope around @xmath99 for @xmath217 , and the latter process is again a standard two sided brownian motion .
now let @xmath218 be an independent sequence of slopes with @xmath219 .
glue them sequentially to get a function @xmath157 in @xmath220 with @xmath221 , and denote by @xmath222 the slope around @xmath99 , for @xmath118 .
if we take @xmath6 as defined just before fact [ nplaplace ] with @xmath185 in the definition of @xmath192 there , then it holds @xmath223 and @xmath224 where @xmath225 is the density of @xmath147
. we will take @xmath226 and finish with the proof by showing that the limit @xmath227 exists and @xmath228 to see this , define @xmath229 for @xmath116 . then @xmath230 the distribution of @xmath206 is nonarithmetic with mean value 1 . by the renewal theorem ( @xcite ,
chapter 3 , statement ( 4.9 ) ) , it follows that the @xmath231 exists and @xmath232 now we apply lemma [ cslope ] to obtain the distribution of the length and height of the central 1-slope .
@xmath233 regarding the length of @xmath203 , observe that for @xmath42 the set @xmath234 is open and @xmath235 so that @xmath236 has the density @xmath237 which is the size - biased sampling formula from renewal theory .
@xmath233 regarding the excess height of @xmath203 , observe that for @xmath42 the set @xmath238 is open and @xmath239 differentiating with respect to @xmath31 we get the density of @xmath240 as @xmath241 from ( iii ) , ( iv ) of fact [ nplaplace ] @xmath242 after some calculations , @xmath243 .
the derivative can move inside the expectation on the left hand side of the above relation , giving @xmath244 , due to the monotone convergence theorem ; which applies because @xmath245 and the function @xmath246 is nonnegative and decreasing in @xmath247 for any @xmath199 .
therefore the density of @xmath240 is @xmath248 the last bit of information needed to achieve our goal , stated at the first sentence of this section , is the direction of the central 1-slope ( i.e. , upward or downward ) and its location with respect to zero . by symmetry
, @xmath203 is an upward slope with probability 1/2 , and from exercise 3.4.7 of @xcite it follows easily that given the length @xmath206 of the slope @xmath249 $ ] around zero , the distance of zero from @xmath250 is uniformly distributed in @xmath251 $ ] .
for @xmath42 and @xmath49 with set of @xmath31-extrema @xmath252 define @xmath253:k\in\d{z}\big\}.\ ] ] we refer to the parameter @xmath31 as time since we are going to study the evolution of @xmath254 as @xmath31 increases .
@xmath254 is the set of slopes at time @xmath31 .
roughly , as @xmath31 increases , the slopes that have height smaller than @xmath31 are absorbed into greater ones . for any lebesgue measurable set @xmath255 , @xmath68 , and @xmath256 we define @xmath257$~~}\end{array}\right).\ ] ] it is important to note that the values of @xmath3 up to time @xmath31 are `` encoded '' in the central slope of @xmath254 .
so the number of sign changes of @xmath57 in @xmath4 $ ] can be inferred by that slope . for @xmath258
fixed , @xmath259 is a measure that satisfies @xmath260 and the right - hand side , considered as a function of @xmath261 , is a measure absolutely continuous with respect to the lebesgue measure with density @xmath262 , where @xmath263 is given in ( [ biasheight ] ) .
therefore the measure on the left - hand side has a density also ( an element of @xmath264 ) , call it @xmath265 , and for lebesgue almost all @xmath266 it holds @xmath267 define @xmath268 with @xmath269 + @xmath270 is continuous as is proved in lemma [ cont ] in section [ lemmata ] .
we plan to establish a pde for @xmath270 . to do this
, we look at @xmath254 and try to predict how @xmath271 should look like . in the transition from @xmath254 to @xmath271 ,
a slope around a point remains the same if the excess height of this and the two neighboring slopes are greater than @xmath272 . in case
one slope has excess height in @xmath273 , it does not appear in @xmath271 .
for example , in figure [ p3 ] the slope @xmath203 has a height say @xmath274 with @xmath275 , and the two neighboring slopes @xmath276 , have height greater than @xmath277 .
assume that @xmath278 , @xmath279 , have heights @xmath280 , @xmath281 respectively .
in @xmath271 we know that the slopes @xmath282 will merge to constitute a new slope with height @xmath283 ; i.e. , with excess height @xmath284 . the slopes @xmath285 can stay as they are in the transition from @xmath254 to @xmath286 or they can be extended if some of @xmath287 has excess height in @xmath273 . in any case , they do nt interfere with @xmath288 .
this simple observation , combined with the renewal structure of the sets @xmath254 , is the basis for the next lemma , which is the first step towards establishing a pde that @xmath270 solves .
we denote by @xmath289 the @xmath266 derivative of @xmath290 , and recall that @xmath291 was defined in as the density of an exponential with mean @xmath31 .
[ step1 ] for @xmath292 we have @xmath293 where for @xmath294 we assume that @xmath295 is differentiable in @xmath266 with continuous derivative .
the term @xmath296 depends on @xmath297 .
the left - hand side of the equation is the probability of an event referring to @xmath271 , and we express it in terms of probabilities referring to @xmath254 . in @xmath254
we focus our attention on the seven @xmath31-slopes closest to zero .
denote them by @xmath298 in the order they appear in the path of @xmath17 from left to right , with @xmath203 being the central one .
the slopes @xmath299 , @xmath300 are independent having for @xmath301 law @xmath179 , and @xmath302 having law @xmath180 .
the probability of the event that at least two of them have excess height in @xmath273 is bounded by @xmath303 , and this is accounted for in the @xmath296 term in . in the complement of this event ,
the event whose probability appears in the left - hand side of happens if and only if in @xmath254 one of the following three holds ( considering what slope , if any , among the seven has excess height in @xmath273 ) .
* at most one of @xmath304 has excess height in @xmath273 , the excess height of each of @xmath305 is at least @xmath272 , and @xmath3 has changed sign @xmath306 times in @xmath4 $ ] . in this case
the slope around zero is the same for both @xmath307 . *
exactly one of @xmath276 has excess height in @xmath273 , @xmath203 has excess height at least @xmath272 , and @xmath3 has changed sign @xmath306 times in @xmath308 $ ] . in this case
@xmath309 has the same sign as @xmath310 .
for example , assume that @xmath279 has excess height in @xmath273 . if @xmath311 then @xmath312 , while if @xmath313 then @xmath314 and simply the central slope in @xmath271 results from merging @xmath315 .
* @xmath294 , @xmath203 has excess height in @xmath316 , and @xmath3 has changed sign @xmath317 times in @xmath4 $ ] . in this case @xmath318 because in @xmath271 the central slope results from merging @xmath282 and it has different direction than @xmath203 ( i.e. , if for example @xmath203 is an upward slope , then the central slope in @xmath271 is a downward slope ) .
the first case has probability @xmath319 with @xmath320 ( resp .
@xmath321 ) denoting the excess height of @xmath278 ( resp .
@xmath322 ) .
this follows by the independence mentioned above and by the fact that @xmath320 and @xmath321 have exponential distribution with mean @xmath31 , and it expresses the demand that both of the @xmath31-slopes neighboring the central @xmath31-slope have excess height greater than @xmath272 . in this case
, the excess heights of @xmath304 do not matter .
they ca nt influence the central slope in @xmath271 .
the second case has probability @xmath323 because , say in the event that @xmath279 has excess height in @xmath273 , the central slope in @xmath271 comes from merging @xmath324 of @xmath254 .
assume that they have excess heights @xmath325 respectively .
then the central slope in @xmath271 will have excess height @xmath326 , and the requirement that this is greater than @xmath266 translates to @xmath327 being greater than @xmath328 . and
of course , by assumption , @xmath203 and @xmath329 have excess height greater than @xmath272 .
@xmath270 is continuous as it is proved in lemma [ cont ] .
thus , dividing with @xmath272 the previous integral and taking @xmath330 we get as limit @xmath331 from this procedure we pick up another @xmath296 term . the last case has probability @xmath332 by assumption , @xmath333 exists and it is the density of the measure @xmath334 .
the dummy variables @xmath335 stand for the excess heights of @xmath336 respectively , and in this case the central slope in @xmath271 has excess height @xmath337 , giving the restriction @xmath338 . since @xmath339 is continuous , dividing by @xmath272 and taking @xmath340 we get @xmath341 and again we pick up an @xmath296 term .
the double integral equals @xmath342 . before getting to the actual proof of our theorem
, we give a non - rigorous short derivation to illustrate its main steps .
the main problem is that we do nt know if @xmath270 is differentiable in the @xmath343 variables for every @xmath344 .
assume for the moment that it is .
for @xmath345 , as usual , we define @xmath346 by @xmath347 for all @xmath348
. the above lemma would give for @xmath270 the pde @xmath349 let @xmath350 ; i.e. , @xmath351 + then @xmath157 should satisfy @xmath352 while the conditions at @xmath185 are @xmath353 the first equation comes from , the second is clear . for @xmath354 , the generating function @xmath355 is well defined . assuming that @xmath356 is differentiable with respect to @xmath343 and its @xmath343 derivatives are obtained with term by term differentiation , we see that @xmath356 satisfies the pde problem @xmath357 we try for a solution of the form @xmath358\,e^{-y}.\ ] ] substituting this into we see that @xmath359 factors out in both sides , and after cancellation
, we arrive in an equality of two first degree polynomials in @xmath266 with coefficients depending on @xmath360 . equating the coefficients in equal powers of @xmath266 in the two sides of the equation
, we arrive at the following system of odes for @xmath361 .
@xmath362 the initial condition for @xmath356 , expressed in terms of @xmath363 , becomes @xmath364 for all @xmath365 .
we easily see that the only solution of the system satisfying these conditions is [ alfa ] @xmath366 where @xmath75 and @xmath367 now @xmath368 , which is what we want .
* proof of the theorem : * the proof is done by taking the steps of the above `` proof '' in reverse order .
this time all the steps can be justified .
we will need three lemmata .
two of them are non - trivial , and their proof is given in section [ lemmata ] .
[ powerseries ] the solution @xmath356 of , obtained above is analytic as a function of @xmath207 in @xmath369 .
the coefficients of its development as a power series around zero are differentiable with respect to @xmath31 and @xmath266 in @xmath370 , and its @xmath343-derivatives are continuous and can be found with term by term differentiation . using this lemma ,
we write @xmath356 as @xmath371 differentiating @xmath356 term by term and equating the coefficients of equal powers of @xmath207 in the two sides of , we see that the sequence of functions @xmath372 satisfies the pdes with conditions at @xmath185 given by , . for @xmath344 define @xmath373 by @xmath374 for @xmath375 .
the sequence of functions @xmath376 satisfies the pdes with conditions at @xmath185 given by , .
the proof is finished by showing that the sequence @xmath377 satisfies a weak form of these pdes , and then a uniqueness result will identify @xmath270 as @xmath378 . for fixed @xmath379 ,
define @xmath380 for @xmath381.$ ] we state as a lemma an equation that @xmath382 satisfies .
the proof is straightforward from lemma [ step1 ] .
[ odes ] the function @xmath382 is differentiable in @xmath383 and satisfies @xmath384 where for @xmath294 we assume that @xmath295 is differentiable in @xmath266 with continuous derivative .
and the promised uniqueness result is the following .
[ uniq ] let @xmath385 be a continuous function such that for every @xmath379 the function @xmath386\rightarrow\mathbb{r}$ ] with @xmath387 is continuous on @xmath388 $ ] and differentiable on @xmath383 , with @xmath389 and @xmath390
. then @xmath391 .
now using induction we show that @xmath392 the function @xmath393 satisfies the assumptions of lemma [ uniq ] because of the pde problem that @xmath394 solves and lemma [ odes ] . for @xmath294 , assuming the statement true for @xmath317 , the same argument works for @xmath395 , where now the assumption on @xmath396 required by lemma [ odes ] is provided by the inductive hypothesis .
therefore , @xmath397 for @xmath398 and @xmath399 , proving for @xmath400 .
@xmath401 * proof of corollary [ markov ] : * our theorem gives that @xmath402 does nt change sign in @xmath4)=a(x,0)$ ] , where the function @xmath201 is defined on page , and by scaling @xmath403)=a(y / x,0 ) \text{\qquad for $ 0<x < y$.}\ ] ] consequently , the density of the last point before @xmath266 that we have sign change is @xmath404 , where we use @xmath405 here and below to denote derivative with respect to the first argument .
differentiating with respect to @xmath266 , we get the density of the event that @xmath406 are consecutive times of sign change as @xmath407 which , after using the expression for @xmath408 , becomes @xmath409 the event that @xmath3 changes sign at @xmath31 translates to the central @xmath31-slope having excess height 0 .
this has density @xmath410 due to the scaling property of the @xmath31-slopes and relation , which refers to the central 1-slope .
thus , the density of the time @xmath411 where the next sign change after @xmath31 happens , given that there was a sign change at @xmath31 , is @xmath412 the quotient @xmath413 , given that there was a sign change at @xmath31 , has the density of the @xmath85 s given in .
@xmath401 * proof of corollary [ strenghening ] : * observe that the right - hand side of is a function analytic in @xmath414.$ ] as for the left - hand side , we have the following lemma .
[ entire ] for any @xmath122 , the power series @xmath415 defines an entire function . and from a basic property of analytic functions
, it follows that the quantities @xmath416 , @xmath417 agree for all @xmath418 $ ] .
@xmath401 of course , lemma [ entire ] implies that the right hand side of can be extended to an entire function .
however , the way is written does nt allow as to claim that it holds for all @xmath72 because the function @xmath419 does nt have an entire extension .
* proof of corollary [ ldp1 ] : * we apply the gartner - ellis theorem ( theorem 2.3.6 in @xcite ) . the moment generating function of @xmath96 is given for any @xmath420 by @xmath421 .
so @xmath422 , and using corollary [ strenghening ] we see that @xmath423 the fenchel - legendre transform @xmath424 of @xmath425 , defined by @xmath426 for all @xmath427 , is found to be the function @xmath428 defined in the statement of the proposition . also @xmath429 and @xmath430
is strictly convex and differentiable in @xmath431 .
the result follows from the gartner - ellis theorem .
[ cont ] for any @xmath68 , @xmath432 @xmath433 , and @xmath434 we have latexmath:[\[\begin{aligned } call @xmath436 and @xmath437 the two events whose probabilities are @xmath438 respectively . then @xmath439 [ in @xmath254 at least one of the three slopes neighboring zero has excess height @xmath440 ] .
denote by @xmath441 the central slope and the slope to the right to it in @xmath254 .
then @xmath442 the other inequality follows because the density of the measure @xmath443 is bounded by @xmath444 ( see ( [ bound ] ) ) .
* proof of lemma [ powerseries ] : * define @xmath445 with @xmath446 for @xmath447 .
clearly , @xmath448 is a holomorphic function ( choose an analytic branch of the square root function defined on @xmath449 ; the number @xmath450 is there for @xmath400 ) and has a power series development centered at zero that converges in @xmath451 ( see , e.g. , @xcite , proposition 2.3.16 ) .
the claims of the lemma follow by the relation @xmath452 and standard properties of power series .
@xmath401 * proof of lemma [ uniq ] : * for @xmath379 and @xmath453 $ ] , define @xmath454\}$ ] .
@xmath455 is well defined because @xmath157 is bounded on compact sets .
+ from ( [ pde ] ) , for @xmath456 one has @xmath457 and since @xmath390 , we get after integrating @xmath458 which implies @xmath459 and @xmath460 the function @xmath461 ( @xmath462 $ ] ) is continuous ( because @xmath157 is ) and has @xmath463 . an application of gronwall s lemma to @xmath436 gives @xmath464 for @xmath453,\ d\in[1,c]$ ] ; i.e. , @xmath465 for all @xmath466 .
@xmath401 * proof of lemma [ entire ] : * let @xmath467 be as in corollary [ markov ] .
then @xmath468 for @xmath469 set @xmath470 , @xmath471 , and let @xmath472 be the distribution measure of @xmath473 . by cramer s theorem ( @xcite theorem 2.2.3 ) , the sequence @xmath474 satisfies a large deviation principle with rate function @xmath475 for @xmath427 .
clearly , for any @xmath476 we have @xmath477 and @xmath478 because @xmath428 is lower semicontinuous and @xmath479 . to see the last point , observe that @xmath480 because @xmath481 with probability one and the bounded convergence theorem applies .
thus @xmath482 , proving that the radius of convergence for the power series is infinite .
first we prove that for fixed @xmath42 , the set @xmath483 has @xmath484 .
observe that for @xmath207 a point of @xmath31-minimum and @xmath485 , @xmath486 it holds that @xmath487 because @xmath17 is continuous at @xmath207 .
and there is no other point of @xmath31-minimum in @xmath488 since if @xmath489 is such a point , say in @xmath490 , then assuming @xmath491 we get a contradiction with the definition of @xmath492 while assuming @xmath493 we get that @xmath17 takes the same value in two local minima , which has probability zero .
now assume that there is a strictly monotone , say increasing , sequence @xmath494 of @xmath31-minima converging to @xmath495 . by the above observations we get @xmath496 contradicting the continuity of @xmath17 at @xmath497 . similarly if @xmath494 is decreasing .
so in a set of @xmath17 s in @xmath498 with probability 1 , it holds that the set of @xmath31-minima of @xmath17 has no accumulation point .
the same holds for the set of @xmath31-maxima and as a result also for @xmath39 . since @xmath499 , @xmath500
, it follows that @xmath501 .
it is clear that between two consecutive @xmath31-maxima ( resp .
minima ) there is exactly one @xmath31-minimum ( resp .
@xmath31-maximum ) . consequently , @xmath484 .
* acknowledgments . *
i thank my advisor amir dembo for his help throughout the time this research was conducted and for his comments on previous drafts of the paper .
i also thank ofer zeitouni for his comments on an earlier draft , daniel fisher for a useful discussion we had during a visit of his to stanford , and francis comets for providing me with a preprint of @xcite .
comets , francis and popov , serguei ( 2003 ) .
limit law for transition probabilities and moderate deviations for sinai s random walk in random environment , probab . theory related fields * 126 * , no . 4 : 571 - 609 . dembo , a. , guionnet , a. , and zeitouni , o. ( 2001 ) , aging properties of sinai s model of random walk in random environment . in saint flour summer school 2001 lecture notes by o. zeitouni , url http://www.math.umn.edu/
zeitouni / technion/. neveu , j. and pitman , j. ( 1989 ) .
renewal property of the extrema and tree property of the excursion of a one - dimensional brownian motion , in sminaire de probabilits xxiii , lecture notes in math .
1372 , 239247 , springer , berlin .
tanaka , hiroshi ( 1987 ) .
limit distributions for one - dimensional diffusion processes in self - similar random environments , in hydrodynamic behavior and interacting particle systems , vol . 9 of i m a vol .
appl . , 189 - 210 , springer | according to a theorem of s. schumacher and t. brox , for a diffusion @xmath0 in a brownian environment it holds that @xmath1 in probability , as @xmath2 , where @xmath3 is a stochastic process having an explicit description and depending only on the environment .
we compute the distribution of the number of sign changes for @xmath3 on an interval @xmath4 $ ] and study some of the consequences of the computation ; in particular we get the probability of @xmath3 keeping the same sign on that interval .
these results have been announced in 1999 in a non - rigorous paper by p. le doussal , c. monthus , and d. fisher and were treated with a renormalization group analysis .
we prove that this analysis can be made rigorous using a path decomposition for the brownian environment and renewal theory .
finally , we comment on the information these results give about the behavior of the diffusion . |
the analyses reported in this talk were performed using either a sample of @xmath9 @xmath7 events or a sample of @xmath10 @xmath8 events collected with the upgraded beijing spectrometer ( besii ) detector @xcite at the beijing electron - positron collider ( bepc ) .
a new structure , denoted as @xmath0 and with mass @xmath11 gev/@xmath12 and width @xmath13 mev/@xmath12 , was observed by the babar experiment in the @xmath14 initial - state radiation process @xcite .
this observation stimulated some theoretical speculation that this @xmath15 state may be an @xmath16-quark version of the @xmath17 since both of them are produced in @xmath18 annihilation and exhibit similar decay patterns @xcite .
here we report the observation of the @xmath0 in the decays of @xmath19 , with @xmath20 , @xmath21 , @xmath22 .
a four - constraint energy - momentum conservation kinematic fit is performed to the @xmath23 hypothesis for the selected four charged tracks and two photons .
@xmath24 candidates are defined as @xmath25-pairs with @xmath26 gev/@xmath12 , a @xmath6 signal is defined as @xmath27 gev/@xmath12 , and in the @xmath28 invariant mass spectrum , candidate @xmath29 mesons are defined by @xmath30 gev/@xmath12 .
the @xmath31 invariant mass spectrum for the selected events is shown in fig .
[ draft - fit ] , where a clear enhancement is seen around 2.18 gev/@xmath12 .
fit with a breit - wigner and a polynomial background yields @xmath32 signal events and the statistical significance is found to be @xmath33 for the signal .
the mass of the structure is determined to be @xmath34 gev/@xmath12 , the width is @xmath35
gev/@xmath12 , and the product branching ratio is @xmath36 .
the mass and width are consistent with babar s results .
invariant mass distribution of the data ( points with error bars ) and the fit ( solid curve ) with a breit - wigner function and polynomial background ; the dashed curve indicates the background function.,scaledwidth=40.0% ]
structures in the @xmath38 invariant - mass spectrum have been observed by several experiments both in the reaction @xmath39
@xcite and in radiative @xmath7 decays @xcite .
the @xmath2 was first observed by the mark - iii collaboration in @xmath7 radiative decays @xmath40 .
a fit to the @xmath38 invariant - mass spectrum gave a mass of 2.22 gev/@xmath12 and a width of 150 mev/@xmath12 @xcite .
an angular analysis of the structure found it to be consistent with a @xmath41 assignment .
it was subsequently observed by the dm2 collaboration , also in @xmath42 decays @xcite .
we present results from a high statistics study of @xmath43 in the @xmath44 final state , with the @xmath45 missing and reconstructed with a one - constraint kinematic fit .
after kinematic fit , we require both the @xmath46 and @xmath47 invariant masses lie within the @xmath6 mass region ( @xmath48 mev/@xmath12 and @xmath49 mev/@xmath12 ) .
the @xmath38 invariant mass distribution is shown in fig .
[ dalitz ] .
there are a total of 508 events with a prominent structure around 2.24 gev/@xmath12 .
invariant mass distribution for @xmath50 candidate events .
the dashed histogram is the phase space invariant mass distribution , and the dotted curve indicates how the acceptance varies with the @xmath38 invariant mass.,scaledwidth=40.0% ] a partial wave analysis of the events with @xmath51 2.7 gev/@xmath12 was performed .
the two - body decay amplitudes in the sequential decay process @xmath52 , @xmath53 and @xmath54 are constructed using the covariant helicity coupling amplitude method .
the intermediate resonance @xmath55 is described with the normal breit - wigner propagator @xmath56 , where @xmath16 is the @xmath38 invariant mass - squared and @xmath57 and @xmath58 are the resonance s mass and width . when @xmath59 , @xmath60 is fitted with both the @xmath38 and @xmath61 systems in a @xmath62-wave , which corresponds to a pseudoscalar @xmath55 state , the fit gives @xmath63 events with mass @xmath64 gev/@xmath12 , width @xmath65 gev/@xmath12 , and a statistical significance larger than @xmath66 , and a product branching fraction of : @xmath67 .
the presence of a signal around 2.24 gev/@xmath12 and its pseudoscalar character are confirmed , and the mass , width , and branching fraction are in good agreement with previous experiments .
a pseudoscalar gluonium candidate , the so - called @xmath68 , was observed in @xmath69 annihilation in 1967 @xcite and in @xmath7 radiative decays in the 1980 s @xcite .
the study of the decays @xmath70 \{@xmath5 , @xmath6}@xmath71 is a useful tool in the investigation of quark and possible gluonium content of the states around 1.44 gev/@xmath72 .
here we investigate the possible structure in the @xmath71 final state in @xmath7 hadronic decays at around @xmath73 gev/@xmath72 . in this analysis , @xmath5 mesons
are observed in the @xmath74 decay , @xmath6 mesons in the @xmath75 decay , and other mesons are detected in the decays : @xmath76 , @xmath77 .
@xmath71 could be @xmath78 or @xmath79 .
figures [ fig : w - x1440-recoiling ] and [ fig : x1440-phikksp ] show the @xmath80 and @xmath81 invariant mass spectra after @xmath5 selection ( @xmath82 gev / c@xmath83 ) or @xmath6 signal selection ( @xmath84 gev/@xmath72 ) .
clear @xmath4 signal is observed recoiling against the @xmath5 , and there is no significant signal recoiling against a @xmath6 . the @xmath80 invariant mass distribution in @xmath85 ( fig .
[ fig : w - x1440-recoiling](b ) ) is fitted with a bw function convoluted with a gaussian mass resolution function ( @xmath86 mev/@xmath72 ) to represent the @xmath4 signal and a third - order polynomial background function .
the mass and width obtained from the fit are @xmath87 mev/@xmath72 and @xmath88 mev/@xmath72 , and the fit yields @xmath89 events . using the efficiency of @xmath90 determined from a uniform phase space mc simulation ,
we obtain the branching fraction to be @xmath91 , where the first error is statistical and the second one systematic . for @xmath92 mode , by fitting the @xmath81 mass spectrum in fig . [
fig : w - x1440-recoiling](c ) with same functions , we obtain the mass and width of @xmath93 mev/@xmath72 and @xmath94 mev/@xmath72 , and the number of events from the fit is @xmath95 .
the efficiency is determined to be @xmath96 from a phase space mc simulation , and the branching fraction is @xmath97 , in good agreement with the isospin symmetry expectation from @xmath85 mode . the distribution of @xmath98 and @xmath99 invariant mass spectra recoiling against the @xmath6 signal are shown in fig .
[ fig : x1440-phikksp ] , and there is no evidence for @xmath4 . the upper limits on the branching fractions at the @xmath100 c.l .
are @xmath101 and @xmath102 . in conclusion , the mass and width of the @xmath4
are measured , which are in agreement with previous measurements ; the branching fractions we measured are also in agreement with the dm2 and mark - iii results .
the significant signal in @xmath103 mode and the missing signal in @xmath104 mode may indicate the @xmath105 component in the @xmath4 is not significant .
besides conventional meson and baryon states , qcd also predicts a rich spectrum of glueballs , hybrids , and multi - quark states in the 1.0 to 2.5 @xmath107 mass region .
therefore , searches for the evidence of these exotic states play an important role in testing qcd .
the radiative decays of @xmath106 to hadrons are expected to contribute about 1% to the total @xmath106 decay width @xcite .
however , the measured channels only sum up to about 0.05% @xcite .
we measured the decays of @xmath106 into @xmath108 , @xmath109 , @xmath110 , @xmath111 , @xmath112 , @xmath113 , @xmath114 , @xmath115 , @xmath116 , and @xmath117 , with the invariant mass of the hadrons ( @xmath118 ) less than 2.9 @xmath107 for each decay mode @xcite .
the differential branching fractions are shown in fig .
[ difbr ] .
the branching fractions below @xmath119 @xmath120 are given in table [ tot - nev ] , which sum up to @xmath121 of the total @xmath106 decay width .
we also analyzed @xmath122 and @xmath123 modes to study the resonances in @xmath124 and @xmath125 invariant mass spectrum .
significant signals for @xmath126 and @xmath127 were observed , but the low statistics prevent us from drawing solid conclusion on the other resonances @xcite .
differential branching fractions for @xmath128 , @xmath129 , @xmath111 , and @xmath110 . here
@xmath118 is the invariant mass of the hadrons . for each point , the smaller longitudinal error is the statistical error , while the bigger one is the total error . ,
scaledwidth=48.0% ] .[tot - nev ] branching fractions for @xmath130 with @xmath119 @xmath107 , where the upper limits are determined at the 90% c.l . [ cols="<,<",options="header " , ]
using the 58 m @xmath131 and 14 m @xmath106 events samples taken with the besii detector at the bepc storage ring , bes experiment provided many interesting results in charmonium decays , including the observation of the @xmath0 , @xmath2 , @xmath4 , and many @xmath106 radiative decays .
these results shed light on the understanding of strong interaction sector of the standard model .
a. etkin _ et al .
_ , phys .
b * 201 * , 568 ( 1988 ) .
z. bai _ et al . _ [ mark - iii collaboration ] , phys .
lett . * 65 * , 1309 ( 1990 ) .
d. bisello _ et al . _
[ dm2 collaboration ] , phys .
b * 179 * , 294 ( 1986 ) .
d. bisello _ et al . _
[ dm2 collaboration ] , phys .
b * 241 * , 617 ( 1990 ) .
m. ablikim _ et al . _
[ bes collaboration ] , phys .
d * 77 * , 032005 ( 2008 ) . | we present recent results from the bes experiment on the observation of the @xmath0 in @xmath1 , study of @xmath2 in @xmath3 , and the production of @xmath4 recoiling against an @xmath5 or a @xmath6 in @xmath7 hadronic decays .
the observation of @xmath8 radiative decays is also presented . |
the following attributes all make an optimization problem more difficult : having an objective function with an unknown and possibly large number of local minima , being constrained , having nonlinear constraints , having inequality constraints , having both discrete and continuous variables .
unfortunately , faithfully modeling an application tends to introduce many of these attributes . as a result
, optimization problems are usually linearized , discretized , relaxed , or otherwise modified to make them feasible according to conventional methods .
one of the most exciting prospects of constraint programming is that such difficult optimization problems can be solved without these possibly invalidating modifications .
moreover , constraint programming solutions are of known quality : they yield intervals guaranteed to contain all solutions .
equally important , constraint programming can prove the absence of solutions . in this paper we only consider the core of the constraint programming approach to optimization , which is to solve a system of nonlinear inequalities : @xmath0
it is understood that it may happen that @xmath1 for some pairs @xmath2 and @xmath3 , so that equalities are a special case .
if this occurs , then certain obvious optimizations are possible in the methods described here .
the ability to solve systems such as ( [ nonlinsys ] ) supports optimization in more ways than one . in the first place
, these systems occur as conditions in some constrained optimized problems .
moreover , one of @xmath4 could be defined as @xmath5 , where @xmath6 is the objective function and where @xmath7 is a constant . by repeatedly solving such a system for suitably chosen @xmath7 , one can find the greatest value of @xmath7 for which ( [ nonlinsys ] ) is found to have no solution .
that value is a lower bound for the global minimum @xcite .
this approach handles nonlinear inequalities with real variables .
it also allows some or all variables to be integer by regarding integrality as a constraint on a real variable @xcite .
all constraint programming work in this direction has been based on interval arithmetic .
the earliest work @xcite used a generic propagation algorithm based directly on domain reduction operators for primitive arithmetic constraints .
these constraints included @xmath8 defined as @xmath9 for all reals @xmath10 , @xmath11 , and @xmath12 .
also included was @xmath13 defined as @xmath14 for all reals @xmath10 , @xmath11 , and @xmath12 .
this was criticized in @xcite which advocated the use of composite arithmetic expression directly rather than reducing them to primitive arithmetic constraints . in @xcite
it was acknowledged that the generic propagation algorithm is not satisfactory for csps that derive from composite arithmetic expressions .
these papers describe propagation algorithms that exploit the structure of such expressions and thereby improve on what is attainable by evaluating such expressions in interval arithmetic .
selective initialization was first described in @xcite .
this was done under the tacit assumption that all default domains are @xmath15 $ ] . as a result
some of the theorems in that paper are not as widely applicable as claimed .
all of these researches are motivated by the severe difficulties experienced by conventional numerical analysis to solve practical optimization problems .
they can be regarded as attempts to fully exploit the potential of interval arithmetic . in this paper
we also take this point of view .
we show that , though equation ( [ nonlinsys ] ) can contain arbitrarily large expressions , only a small modification of the generic propagation algorithm is needed to optimally exploit the structure of these expressions .
this is made possible by a new canonical form for equation ( [ nonlinsys ] ) that we introduce in this paper .
in addition to supporting our application of constraint processing to solving systems similar to equation ( [ nonlinsys ] ) , this canonical form exploits the potential for parallelism in such systems .
in figure [ softarch ] we propose a hierarchical software architecture for optimization problems . each layer
is implemented in terms of the layer below . in the introduction we briefly remarked on how layer
4 can be reduced to layer 3 .
more detail is given in @xcite .
for the transition between layers 0 and 1 there is much material in the interval arithmetic literature .
the part that is relevant to constraint processing can be found in @xcite . in the present paper
we present a new method for implementing layer 3 in terms of layer 2 .
but first we review the transition between layers 1 and 2 .
in this section we provide background by reviewing some basic concepts . these reviews also serve to establish the terminology and notation used in this paper .
the first few sections apply to all constraint satisfaction problems , not only to numerical ones . a _ constraint satisfaction problem ( csp ) _ consists of a set of _
constraints_. each of the variables in the constraint is associated with a _ domain _ , which is the set of values that are possible for the variable concerned .
typically , not all sets of values can be domains .
for example , sets of real values are restricted to intervals , as described later .
a _ valuation _ is a tuple indexed by variables where the component indexed by @xmath16 is an element of the domain of @xmath16 .
solution _ is a valuation such that each constraint is true if every variable in the constraint is substituted by the component of the valuation indexed by the variable .
the set of solutions is a set of valuations ; hence a set of tuples ; hence a _
relation_. we regard this relation as the relation defined by the csp . in this way
the relation that is the meaning of a constraint in one csp can be defined by another .
this gives csps an hierarchical structure . with each constraint
, there is an associated _ domain reduction operator _ ; dro for short .
this operator may remove from the domains of each of the variables in the constraint certain values that do not satisfy the constraint , given that the other variables of the constraint are restricted to their associated domains .
any dro is contracting , monotonic , and idempotent .
when the dros of the constraints are applied in a `` fair '' order , the domains converge to a limit or one of the domains becomes empty .
a sequence of dros activations is _ fair _ if every one of them occurs an infinite number of times .
the resulting cartesian product of the domains becomes the greatest common fixpoint of the dros @xcite .
if one of the domains becomes empty , it follows that no solutions exist within the initial domains . in practice ,
we are restricted to the domains that are representable in a computer . as there are only a finite number of these
, any fair sequence of dro applications yields domains that remain constant from a certain point onwards . according to the usual terminology in constraint programming ,
a constraint states that a certain relation holds between its arguments .
but in first - order predicate logic the same role is played by an _
atomic formula_. in this paper we adopt the terminology of first - order predicate logic for the meaning of `` atomic formula '' and we reserve `` constraint '' for a special case .
thus an atomic formula consists of a predicate symbol with terms as arguments .
a _ term _ is a function symbol with terms as arguments .
what makes an atomic formula first - order is that a predicate symbol can only occur as the outermost symbol in the formula . at first sight , the inequalities in equation ( [ nonlinsys ] ) are atomic formulas .
in fact , they follow the related , but different , usage that prevails in informal mathematics .
the inequality @xmath17 is an atomic formula where @xmath18 is the predicate symbol with two arguments , which are the terms @xmath19 and @xmath20 .
a possible source of confusion is that in mathematics @xmath19 is not necessarily interpreted as a syntactically concrete term , but as an abstractly conceived function @xmath21 applied to the arguments @xmath22 .
the function could be defined by means of a term that looks quite different ; such a term could be nested and contain several function symbols .
for example , equation ( [ ineq ] ) could be @xmath23 with @xmath21 defined as @xmath24 for all @xmath10 and @xmath11 .
accordingly , the atomic formula corresponding to equation ( [ ineq ] ) is @xmath25 taking advantage of infix and postfix notation this becomes @xmath26 .
a _ constraint _ is an atomic formula without function symbols .
an example of such an atomic formula is @xmath8 , which is a ternary constraint whose relation is defined by @xmath9 for all reals @xmath10 , @xmath11 , and @xmath12 . in this paper
we translate equation ( [ ineqex ] ) to a csp with the set of constraints @xmath27 consider a constraint @xmath28 .
the meaning of predicate symbol @xmath7 is a relation , say @xmath29 .
for all @xmath30 , a value @xmath31 for variable @xmath32 is _ inconsistent _ with respect to @xmath29 and domains @xmath33 iff it is not the case that @xmath34 a dro for @xmath7 may replace one or more of the domains @xmath35 by a subset of it if the set difference between the old and the new domains contains inconsistent values only .
a dro is _ optimal _ if every domain is replaced by the smallest domain containing all its consistent values .
we call a constraint _ primitive _ if an optimal dro is available for it that is sufficiently efficiently computed .
what is sufficient depends on the context . to gain information about the solution set ,
inconsistent values are removed as much as possible with modest computation effort .
for example , dros can be applied as long as they remove inconsistent values .
it is the task of a constraint propagation algorithm to reach as quickly as possible a set of domains that can not be reduced by any dro .
many versions of this algorithm exist @xcite .
they can be regarded as refinements of the algorithm in figure [ lgpa ] , which we refer to as the _ generic propagation algorithm _ ; gpa for short .
gpa maintains a pool of dros , called _
active set_. no order is specified for applying these operators .
notice that the active set @xmath36 is initialized to contain all constraints . put all constraints into the active set @xmath36 + while ( @xmath37 ) @xmath38 +
choose a constraint @xmath39 from @xmath36 + apply the dro associated with @xmath39 + if one of the domains has become empty , then stop + add to @xmath36 all constraints involving variables whose domains have changed , if any + remove @xmath39 from @xmath36 + @xmath40 a _ floating - point number _ is any element of @xmath41 , where @xmath42 is a finite set of reals .
a _ floating - point interval _ is a closed connected set of reals , where the bounds , in so far as they exist , are floating - point numbers . when we write `` interval '' without qualification , we mean a floating - point interval . for every bounded non - empty interval @xmath43 , @xmath44 and @xmath45
denote the least and the greatest element of @xmath43 respectively .
they are referred to as the left and the right bound of @xmath43 . if @xmath43 is not bounded from below , then @xmath46 .
similarly , if @xmath43 is not bounded from above , then @xmath47 .
thus , @xmath48 $ ] is a convenient notation for all non - empty intervals , bounded or not .
a _ box _ is a cartesian product of intervals .
moore s idea of solving inequalities such as those in equation ( [ nonlinsys ] ) by means of interval arithmetic is at least as important as the subsequent applications of interval constraints to this problem .
suppose we wish to investigate the presence of solutions of a single inequality in equation ( [ nonlinsys ] ) in a box @xmath49 .
then one evaluates in interval arithmetic the expression in the left - hand side . as values for the variables
@xmath50 one uses the intervals @xmath51 .
suppose the result is the interval @xmath52 $ ] .
we have exactly one of the following three cases . if @xmath53 for at least one @xmath2 , then there are no solutions .
if @xmath54 for all @xmath2 , then all tuples in @xmath49 are solutions . if @xmath55 for at least one @xmath2 , then either of the above may be true .
in this same case of @xmath55 , it may also be that some of the tuples in @xmath49 are solutions , while others are not .
again , in the case where @xmath55 , it may be possible to split @xmath49 . in this way
, a more informative outcome may be obtained for one or both of the results of splitting .
such splits can be repeated as long as possible and necessary .
problems in a wide variety of application areas can be expressed as csps .
domains can be as different as booleans , integers , finite symbolic domains , and reals . in this paper
we consider _ interval csps _
( icsps ) , which are csps where all domains are intervals and all constraints are primitive .
icsps are important because they encapsulate what can be efficiently computed ; they represent layer 2 in the software architecture of figure [ softarch ] .
the layer is distinct from layer 3 because in equation ( [ nonlinsys ] ) there typically occur atomic formulas that contain function symbols . to emphasize the role of icsps as a layer of software architecture ,
we view them as a _ virtual machine _ , with a function that is similar to those for prolog or java .
just as a program in prolog or java is translated to virtual machine instructions , a system such as equation ( [ nonlinsys ] ) can be translated to an icsp , as described in a later section .
the instructions of the icsp level are dros , one for each constraint . as an example of such an icsp virtual machine instruction , consider the dro for product constraint .
it reduces the box @xmath56\times [ c , d]\times [ e , f]$ ] to the box that has the projections @xmath57 & \cap & ( [ e , f]/[c , d ] ) ) \nonumber\\ \varphi([c , d ] & \cap & ( [ e , f]/[a , b ] ) ) \nonumber\\ \varphi([e , f ] & \cap & ( [ a , b]*[c , d ] ) ) \nonumber\end{aligned}\ ] ] here @xmath58 is the function that yields the smallest interval containing its argument .
of particular interest is the effect of the dro when all variables have @xmath15 $ ] as domain . for each argument
, the domain after application of the dro is defined as the _ default domain _ of that argument .
typically , default domains are @xmath15 $ ] .
notable exceptions include the constraint @xmath59 ( defined as @xmath60 ) , where the default domain of @xmath11 is @xmath61 $ ] .
another is @xmath62 ( defined as @xmath63 ) , where the default domain of @xmath11 is @xmath64 $ ] .
a difference with other virtual machines is that a program for the icsp virtual machine is an unordered collection of dros .
programs for other virtual machines are ordered sequences of instructions . in those other virtual machines
, the typical instruction does not specify the successor instruction . by default this is taken to be the next one in textual order .
execution of the successor is implemented by incrementing the instruction counter by one .
the simplicity of the instruction sequencing in conventional virtual ( and actual ) machines is misleading .
many instruction executions concern untypical instructions , where the next instruction is specified to be another than the default next instruction .
examples of such untypical instructions are branches ( conditional or unconditional ) and subroutine jumps . in the icsp virtual machine , the dros are the instructions , and they form an unordered set . instead of an instruction counter specifying the next instruction , there is the active set of gpa containing the set of possible next instructions . instead of an instruction or a default rule determining the next instruction to be executed
, gpa selects in an unspecified way which of the dros in the active set to execute . in this way
, programs can be declarative : instructions have only meaning in terms of _ what _ is to be computed . _ how _ it is done ( instruction sequencing ) ,
is the exclusive task of the virtual machine . equation ( [ nonlinsys ] ) may have multiple occurrences of variables in the same formula .
as there are certain advantages in avoiding such occurrences , we rewrite without loss of generality the system in equation ( [ nonlinsys ] ) to the canonical form shown in figure [ singlesys ] .
@xmath65 in figure [ singlesys ] , the expressions for the functions @xmath66 have no multiple occurrences of variables . as a result
, they have variables @xmath67 instead of @xmath22 , with @xmath68 as in equation ( [ nonlinsys ] ) .
this canonical form is obtained by associating with each of the variables @xmath69 in equation ( [ nonlinsys ] ) an equivalence class of the variables in figure [ singlesys ] .
this is done by replacing in equation ( [ nonlinsys ] ) each occurrence of a variable by a different element of the corresponding equivalence class .
this is possible by ensuring that each equivalence class is as large as the largest number of multiple occurrences .
the predicate @xmath70 is true if and only if all its real - valued arguments are equal .
an advantage of this translation is that evaluation in interval arithmetic of each expression gives the best possible result , namely the range of the function values . at the same time
, the @xmath70 constraint is easy to enforce by making all intervals of the variables in the constraint equal to their common intersection .
this takes information into account from all @xmath71 inequalities .
if the system in its original form as in equation [ nonlinsys ] , with multiple occurrences , would be translated to a csp , then only multiple occurrences in a single expression would be exploited at one time . in the coming sections and without loss of generality
, we will only consider expressions without multiple occurrences of variables .
icsps represent what we _ can _ solve .
they consist of atomic formulas without function symbols that , moreover , have efficient dros .
equation ( [ nonlinsys ] ) exemplifies what we _ want _ to solve : it consists of atomic formulas typically containing deeply nested terms .
[ [ the - tree - form - of - a - formula ] ] the tree form of a formula + + + + + + + + + + + + + + + + + + + + + + + + + + we regard a first - order atomic formula as a tree .
the unique predicate symbol is the root .
the terms that are the arguments of the formula are also trees and they are the subtrees of the root . if the term is a variable , then the tree only has a root , which is that variable .
a term may also be a function symbol with one or more arguments , which are terms . in that case , the function symbol is the root with the argument terms as subtrees . in the tree form of a formula
the leaves are variables . in addition
, we label every node that is a function symbol with a unique variable .
any constants that may occur in the formula are replaced by unique variables .
we ensure that the associated domains contain the constants and are as small as possible .
[ [ translating - a - formula - to - an - icsp ] ] translating a formula to an icsp + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the tree form of a formula thus labeled is readily translated to an icsp .
the translation has a set of constraints in which each element is obtained by translating an internal node of the tree .
the root translates to @xmath72 where @xmath73 is the predicate symbol that is the root and @xmath74 are the variables labeling the children of the root .
a non - root internal node of the form @xmath75 translates to @xmath76 , where * @xmath11 is the variable labeling the node * @xmath74 are the variables labeling the child nodes * @xmath42 is the relation defined by @xmath77 iff @xmath78 for all @xmath79
. propagation may terminate with small intervals for all variables of interest .
this is rare .
more likely , propagation leaves a large box as containing all solutions , if any . to obtain more information about possibly existing solutions ,
it is necessary to split an icsp into two icsps and to apply propagation to both .
an icsp @xmath80 is a result of splitting an icsp @xmath81 if @xmath80 has the same constraints as @xmath81 and differs only in the domain for one variable , say , @xmath10 .
the domain for @xmath10 in @xmath80 is the left or right half of the domain for @xmath10 in @xmath81 .
a _ search strategy _ for an icsp is a binary tree representing the result of successive splits .
search strategies can differ greatly in the effort required to carry them to completion .
the most obvious search strategy is the _ greedy strategy _ : the one that ensures that all intervals become small enough by choosing a widest domain as the one to be split .
this is a plausible strategy in the case where the icsp has a few point solutions . in general ,
the set of solutions is a continuum : a line segment , a piece of a surface , or a variety in a higher dimensional space that has positive volume . in such cases
we prefer the search to result in a single box containing all solutions .
of course we also prefer such a box to be as small as possible .
the greedy search strategy splits the continuum of solutions into an unmanageably large number of small boxes .
it is not clear that the greedy strategy is preferable even in the case of a few well - isolated point solutions . in general
we need a search strategy other than the greedy one .
a more promising search strategy was first described in the ` absolve ` predicate of the bnr prolog system @xcite and by @xcite , where it is called _ box consistency_. the box consistency search strategy selects a variable and a domain bound .
box consistency uses binary search to determine a boundary interval that can be shown to contain no solutions .
this boundary interval can then be removed from the domain , thus shrinking the domain .
this is repeated until a boundary interval with width less than a certain tolerance is found that can not be shown to contain no solutions .
when this is the case for both boundaries of all variables , the domains are said to be _ box consistent _ with respect to the tolerance used and with respect to the method for showing inconsistency .
when this method is interval arithmetic , we obtain _ functional box consistency_. when it is propagation , then it is called _ relational box consistency _ @xcite .
all we need to know about search in this paper is that greedy search and box consistency are both search strategies and that both can be based on propagation .
box consistency is the more promising search strategy .
thus we need to compare interval arithmetic and propagation as ways of showing that a nonlinear inequality has no solutions in a given box .
this we do in section [ psi ] .
suppose we have a term that can be evaluated in interval arithmetic .
let us compare the interval that is the result of such an evaluation with the effect of gpa on the icsp obtained by translating the term as described in section [ translation ] . to make the comparison possible we define evaluation of a term in interval arithmetic .
the definition follows the recursive structure of the term : a term is either a variable or it is a function symbol with terms as arguments .
if the term is an interval , then the result is that interval .
if the argument is function @xmath6 applied to arguments , then the result is @xmath6 evaluated in interval arithmetic applied to the results of evaluating the arguments in interval arithmetic .
this assumes that every function symbol denotes a function that is defined on reals as well as on intervals .
the latter is called the _ interval extension _ of the former . for a full treatment of interval extensions , see @xcite .
the following lemma appears substantially as theorem 2.6 in @xcite .
[ basic ] let @xmath82 be a term that can be evaluated in interval arithmetic .
let the variables of @xmath82 be @xmath83 .
let @xmath11 be the variable associated with the root of the tree form of @xmath82 .
let @xmath81 be the icsp that results from translating @xmath82 , where the domains of @xmath83 are @xmath84 and where the domains of the internal variables are @xmath15 $ ] .
after termination of gpa applied to @xmath81 , the domain for @xmath11 is the interval that results from interval arithmetic evaluation of @xmath82 .
suppose that a variable of a constraint has domain @xmath15 $ ] .
after applying the dro for that constraint , this domain has become the result of the interval arithmetic operation that obtains the domain for this variable from the domains of the other variables of the constraint .
+ according to @xcite , every fair sequence of dros converges to the same domains for the variables .
these are also the domains on termination of gpa .
let us consider a fair sequence that begins with a sequence @xmath85 of dros that mimics the evaluation of @xmath82 in interval arithmetic . at the end of this
, @xmath11 has the value computed by interval arithmetic .
this shows that gpa gives a result that is a subinterval of the result obtained by interval arithmetic .
+ gpa terminates after activating the dros in @xmath85 .
this is because in the interval arithmetic evaluation of @xmath82 an operation is only performed when its arguments have been evaluated .
this means that the corresponding dro only changes one domain .
this domain is the domain of a unique variable that occurs in only one constraint that is already in the active set .
therefore none of the dro activations adds a constraint to the active set , which is empty after @xmath85 .
gpa yields the same result whatever the way constraints are selected in the active set .
therefore gpa always gives the result of interval arithmetic evaluation .
however , gpa may obtain this result in an inefficient way by selecting constraints that have no effect .
this suggests that the active set be structured in a way that reflects the structure of @xmath82 .
this approach has been taken in @xcite .
the proof shows that , if the active set had not contained any of the constraints only involving internal variables , these constraints would have been added to the active set by gpa .
this is the main idea of selective initialization . by initializing and ordering the active set in a suitable way and leaving gpa otherwise unchanged
, it will obtain the interval arithmetic result with no more operations than interval arithmetic .
this assumes the optimization implied by the totality theorem in @xcite .
a constraint is a _ seed constraint _
iff at least one of its variables has a domain that differs from the default domain assigned to that variable .
for example , the term @xmath86 translates to an icsp with constraints @xmath87 , @xmath88 , and @xmath89 . when the domains are @xmath15 $ ] for all variables , then the seed constraints are @xmath88 and @xmath89 .
when the domains are @xmath15 $ ] for @xmath90 , @xmath91 , and @xmath11 ; @xmath61 $ ] for @xmath92 and @xmath16 , then @xmath87 is the one seed constraint .
when the domains are @xmath15 $ ] for @xmath91 and @xmath11 ; @xmath61 $ ] for @xmath90 , @xmath92 and @xmath16 , then the seed constraints are @xmath87 and @xmath88 .
let psi ( propagation with selective initialization ) be gpa except for the following modifications .
+ ( a ) psi only applies to icsps generated by translation from an atomic formula .
+ ( b ) the active set is a priority queue that is ordered according to the distance from the root of the node that generated the constraint .
the greater that distance , the earlier the item is removed from the queue .
+ ( c ) the active set contains all seed constraints and no other ones .
lemma [ basic ] says that gpa simulates interval arithmetic as far as the result is concerned .
it does not say anything about the efficiency with which the result is obtained .
theorem [ simulate ] says that psi obtains the result as efficiently as it is done in interval arithmetic .
this assumes the functionality optimization in the dros @xcite .
[ simulate ] let @xmath81 be the icsp obtained by translating a term @xmath82 in variables @xmath93 , where these variables have domains @xmath35 . applying psi to s terminates after selecting no constraint more than once .
moreover , the root variable ends up with @xmath94 as domain where @xmath94 is the interval resulting from evaluating @xmath82 with @xmath93 substituted by @xmath35 .
suppose gpa is applied to @xmath81 in such a way that all non - seed constraints are selected first .
the execution of the dro corresponding to these non - seed constraints does not change any domains .
therefore these dro executions do not add any constraints . as a result
, the effect of applying gpa is the same as when the active set would have been initialized with only the seed constraints .
+ suppose the seed constraints are selected according to priority order .
this ensures that no future constraint selection re - introduces a constraint previously selected .
thus gpa terminates after activating every seed constraint exactly once .
+ such an execution of gpa coincides step by step with that of psi .
as gpa terminates with the correct result , so does psi .
we briefly reviewed how interval arithmetic can solve systems of nonlinear inequalities .
the fundamental capability turned out to be that of evaluating a term in interval arithmetic .
we saw that this can also be done by applying propagation to icsps generated from arithmetic terms .
we now investigate how to extend this to the use of icsps to solve nonlinear inequalities .
suppose @xmath81 is the icsp resulting from translating @xmath95 let @xmath11 be the variable labeling the left child of the root ; that is , the variable representing the value of the left - hand side .
let @xmath96 be the domains in @xmath81 of @xmath22 , respectively .
now suppose that gpa is applied to @xmath81 .
one possible initial sequence of dro activations is the equivalent of interval arithmetic evaluation of the left - hand side , leaving @xmath97 as the only constraint in the active set with the domain for @xmath11 equal to @xmath52 $ ] , the value in interval arithmetic of @xmath98 . at this stage
the dro for @xmath97 is executed . if @xmath53 , then failure occurs . if @xmath54 , then the domain for @xmath11 is unchanged .
therefore , no constraint is added to the active set .
termination occurs with nonfailure .
there is no change in the domain of any of @xmath22 .
the third possibility is that @xmath55 . in this case , the domain for @xmath11 shrinks : the upper bound decreases from @xmath99 to @xmath20 .
this causes the constraints to be brought into the active set that correspond to nodes at the next lower level in the tree .
this propagation may continue all the way down to the lowest level in the tree , resulting in shrinking of the domain of one or more of @xmath22 .
let us compare this behaviour with the use of interval arithmetic to solve the same inequality .
in all three cases , gpa gives the same outcome as interval arithmetic : failure or nonfailure .
in the first two cases , gpa gives no more information than interval arithmetic .
it also does no more work . in the third case
, gpa may give more information than interval arithmetic : in addition to the nonfailure outcome , it may shrink the domain of one or more of @xmath22 .
this is beyond the capabilities of interval arithmetic , which is restricted to transmit information about arguments of a function to information about the value of the function .
it can not transmit information in the reverse direction . to achieve this extra capability , gpa needs to do more work than the equivalent of interval arithmetic evaluation . in the above
, we have assumed that gpa magically avoids selecting constraints in a way that is not optimal .
in such an execution of gpa we can recognize two phases : an initial phase that corresponds to evaluating the left - hand side in interval arithmetic , followed by a second phase that starts with the active set containing only the constraint @xmath97 .
when we consider the nodes in the tree that correspond to the constraints that are selected , then it is natural to call the first phase bottom - up ( it starts at the leaves and ends at the root ) and the second phase top - down ( it starts at the root and may go down as far as to touch some of the leaves ) .
the bottom - up phase can be performed automatically by the psi algorithm .
the start of the top - down phase is similar to the situation that occurs in search . in both search and in the top - down
phase a different form of selective initialization can be used , shown in the next section . the bottom - up phase and the top - down phase
are separated by a state in which the active set only contains @xmath97 . for reasons that become apparent in the next section
, we prefer a separate treatment of this constraint : not to add it to the active set and to execute the shrinking of the domain for @xmath11 as an extraneous event .
this is then a special case of termination of gpa , or its equivalent psi , followed by the extraneous event of shrinking one domain .
the pseudo - code for psi algorithm is given in figure [ psialg ] .
let the active set @xmath36 be a priority queue in which the constraints are + @xmath100 ordered according to the level they occupy in the tree , + @xmath100 with those that are further away from the root placed nearer to the front of the queue + put only * seed * constraints into @xmath36 + while ( @xmath37 ) @xmath38 + choose a constraint @xmath39 from @xmath36 + apply the dro associated with @xmath39 + if one of the domains is empty , then stop + add to @xmath36 all constraints involving variables whose domains have changed , if any + remove @xmath39 from @xmath36 + @xmath40 the correctness of psi algorithm can be easily deduced from the following theorem .
[ modeval ] consider the icsp @xmath101 obtained from the tree @xmath102 of the atomic formula @xmath103 .
suppose we modify gpa so that the active set is initialized to contain instead of all constraints only seed constraints .
suppose also that the active set is a priority queue in which the constraints are ordered according to the level they occupy in the tree @xmath102 , with those that are further away from the root placed nearer to the front of the queue .
then gpa terminates with the same result as when the active set would have been initialized to contain all constraints . as we did before , suppose that in gpa the active set @xmath36 is initialized with all constraints such that the seed constraints are at the end of the active set . applying any dro of a constraint that is
not a seed constraint will not affect any domain .
thus , the constraints that are not seed constraints can be removed from the active set without changing the result of gpa . since the gpa does not specify any order , @xmath36 can be ordered as desired . here
we choose to order it in such a way we get an efficient gpa when used to evaluate an expression ( see previous section ) .
often we find that after applying gpa to an icsp @xmath81 , the domain @xmath43 for one of the variables , say @xmath10 , is too wide .
search is then necessary .
this can take the form of splitting @xmath81 on the domain for @xmath10 .
the results of such a split are two separate icsps @xmath104 and @xmath105 that are the same as @xmath81 except for the domain of @xmath10 . in @xmath104
, @xmath10 has as domain the left half of @xmath43 ; in @xmath105 , it is the right half of @xmath43 .
however , applying gpa to @xmath104 and @xmath105 entails duplicating work already done when gpa was applied to @xmath81 .
when splitting on @xmath10 after termination of the application of gpa to @xmath81 , we have the same situation as at the beginning of the downward phase of applying gpa to an inequality : the active set is empty and an extraneous event changes the domain of one variable to a proper subset .
the following theorem justifies a form of the psi algorithm where the active set is initialized with what is typically a small subset of all constraints .
[ modsolvegeneral ] let @xmath102 be the tree obtained from the atomic formula @xmath103 .
let @xmath101 be the icsp obtained from @xmath102 .
let @xmath10 be a variable in @xmath101 .
suppose we apply gpa to @xmath101 .
after the termination of gpa , suppose the domain of @xmath10 is changed to an interval that is a proper subset of it .
if we apply gpa to @xmath101 with an active set initialized with the constraints only involving @xmath10 , then gpa terminates with the same result as when the active set would have been initialized to contain all constraints . to prove theorem [ modsolvegeneral ]
, we should show that initializing gpa with all constraints gives the same results as when it is initialized with only the constraints involving @xmath10 . since no ordering is specified for the active set of gpa , we choose an order in which the constraints involving @xmath10 are at the end of the active set .
because dros are idempotent , all constraints at the front of the active set , different from those involving @xmath10 , do not affect any domain . thus removing them from the active set in the initialization process does not change the fixpoint of the gpa .
thus , theorem [ modsolvegeneral ] is proved .
we have only considered the application of selective initialization to solve a single inequality .
a conjunction of inequalities such as equation ( [ nonlinsys ] ) can be solved by solving each in turn .
this has to be iterated because the solving of another inequality affects the domain of an inequality already solved .
this suggests performing the solving of all inequalities in parallel .
doing so avoids the waste of completing an iteration on the basis of unnecessarily wide intervals .
it also promises speed - up because many of the dro activations only involve variables that are unique to the inequality . in the current version of the design of our algorithm
, we combine this parallelization with a method of minimizing the complexity usually caused by multiple occurrences of variables .
before interval methods it was not clear how to tackle numerically realistic optimization models . only with the advent of interval arithmetic in the 1960s @xcite one could for the first time at least say : `` if only we had so much memory and so much time , then we could solve this problem . ''
interval arithmetic has been slow in developing . since the 1980s constraint programming has added fresh impetus to interval methods .
conjunctions of nonlinear inequalities , the basis for optimization , can be solved both with interval arithmetic and with constraint programming . in this paper
we relate these two approaches .
it was known that constraint propagation subsumes interval arithmetic .
it was also clear that using propagation for the special case of interval arithmetic evaluation is wasteful . in this paper
we present an algorithm for propagation by selective initialization that ensures that propagation is as efficient in the special case of interval arithmetic evaluation .
we also apply selective initialization for search and for solving inequalities .
preliminary results on a parallel version of the methods presented here suggest that realistic optimization models will soon be within reach of modest computing resources .
we acknowledge generous support by the university of victoria , the natural science and engineering research council nserc , the centrum voor wiskunde en informatica cwi , and the nederlandse organisatie voor wetenschappelijk onderzoek nwo .
frdric benhamou , frdric goualard , laurent granvilliers , and jean - franois puget .
revising hull and box consistency . in _ proceedings of the 16th international conference on logic programming _ , pages 230244 .
mit press , 1999 .
hickey , m.h .
van emden , and h. wu . a unified framework for interval constraints and interval arithmetic . in michael maher and jean - franois puget , editors , _ principles and practice of constraint programming cp98 _ , pages 250 264 .
springer - verlag , 1998 .
lecture notes in computer science 1520 .
van emden .
computing functional and relational box consistency by structured propagation in atomic constraint systems . in _ proc .
6th annual workshop of the ercim working group on constraints ; downloadable from corr _ , 2001 .
van emden and b. moa . using propagation for solving complex arithmetic constraints .
technical report dcs - xxx - ir , department of computer science , university of victoria .
paper cs.na/0309018 in computing research repository ( corr ) ,
september 2003 . | numerical analysis has no satisfactory method for the more realistic optimization models .
however , with constraint programming one can compute a cover for the solution set to arbitrarily close approximation . because the use of constraint propagation for composite arithmetic expressions is computationally expensive ,
consistency is computed with interval arithmetic . in this paper
we present theorems that support , selective initialization , a simple modification of constraint propagation that allows composite arithmetic expressions to be handled efficiently . |
nmt would like thank hitesh changlani , cyrus umrigar , adam holmes , bryan ogorman , bryan clark , and jonathan moussa for useful discussions .
this work was supported through the scientific discovery through advanced computing ( scidac ) program funded by the u.s .
department of energy , office of science , advanced scientific computing research and basic energy sciences .
we used the extreme science and engineering discovery environment ( xsede ) , which is supported by the national science foundation grant no .
oci-1053575 and resources of the oak ridge leadership computing facility ( olcf ) at the oak ridge national laboratory , which is supported by the office of science of the u.s .
department of energy under contract no .
de - ac0500or22725 .
59ifxundefined [ 1 ] ifx#1 ifnum [ 1 ] # 1firstoftwo secondoftwo ifx [ 1 ] # 1firstoftwo secondoftwo `` `` # 1''''@noop [ 0]secondoftwosanitize@url [ 0 ]
+ 12$12 & 12#1212_12%12@startlink[1]@endlink[0]@bib@innerbibempty in link:\doibase 10.1109/sc.2005.17 [ _ _ ] ( ) pp .
\doibase http://dx.doi.org/10.1063/1.2335446 [ * * , ( ) , http://dx.doi.org/10.1063/1.2335446 ] \doibase http://dx.doi.org/10.1063/1.472352 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.481764 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.471518 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.3193710 [ * * , ( ) , http://dx.doi.org/10.1063/1.3193710 ] link:\doibase 10.1103/physrevb.48.10345 [ * * , ( ) ] link:\doibase 10.1063/1.478295 [ * * , ( ) ] link:\doibase 10.1063/1.1449459 [ * * , ( ) ] link:\doibase 10.1103/revmodphys.77.259 [ * * , ( ) ] link:\doibase 10.1146/annurev - conmatphys-020911 - 125018 [ * * , ( ) ] , http://dx.doi.org/10.1038/nature11770 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.4720076 [ * * , ( ) , http://dx.doi.org/10.1063/1.4720076 ] link:\doibase 10.1038/nature11770 [ * * , ( ) ] link:\doibase 10.1021/ct300486d [ * * , ( ) ] link:\doibase 10.1103/physrevlett.114.033001 [ * * , ( ) ] link:\doibase 10.1021/ct300504f [ * * , ( ) ] link:\doibase 10.1063/1.2768362 [ * * , ( ) ] link:\doibase 10.1021/ct3008974 [ * * , ( ) ] link:\doibase 10.1038/nchem.2041 [ * * , ( ) ] link:\doibase 10.1103/physrevlett.109.230201 [ * * , ( ) ] @noop ( ) , \doibase http://dx.doi.org/10.1063/1.4773819 [ * * , ( ) , http://dx.doi.org/10.1063/1.4773819 ] \doibase http://dx.doi.org/10.1063/1.3302277 [ * * , ( ) , http://dx.doi.org/10.1063/1.3302277 ] \doibase http://dx.doi.org/10.1063/1.4905329 [ * * , ( ) , http://dx.doi.org/10.1063/1.4905329 ] ( , ) pp .
\doibase http://dx.doi.org/10.1063/1.4869192 [ * * , ( ) , http://dx.doi.org/10.1063/1.4869192 ] link:\doibase 10.1103/physrev.183.23 [ * * , ( ) ]
link:\doibase 10.1002/qua.560070515 [ * * , ( ) ] link:\doibase 10.1007/bf02394557 [ * * , ( ) ] link:\doibase 10.1080/00268977800100581 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.1679199 [ * * , ( ) ] \doibase http://dx.doi.org/10.1016/0301-0104(83)85011-3 [ * * , ( ) ] link:\doibase 10.1002/jcc.540080105 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.461037 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.460537 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.465368 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.1615956 [ * * , ( ) ] link:\doibase 10.1103/physrevc.79.064324 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.3600351 [ * * , ( ) , http://dx.doi.org/10.1063/1.3600351 ] \doibase http://dx.doi.org/10.1063/1.4905528 [ * * , ( ) , http://dx.doi.org/10.1063/1.4905528 ] link:\doibase 10.1021/acs.jctc.5b01099 [ * * , ( ) ] , \doibase http://dx.doi.org/10.1063/1.4766327 [ * * , ( ) , http://dx.doi.org/10.1063/1.4766327 ] link:\doibase 10.1139/cjc-2013 - 0017 [ * * , ( ) ] , @noop _ _ ( , ) @noop ( ) , @noop ( ) , link:\doibase 10.1137/s0895479898334605 [ * * , ( ) ] , link:\doibase 10.1002/wcms.93 [ * * , ( ) ] link:\doibase 10.1002/wcms.81 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.463096 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.1804498 [ * * , ( ) ] \doibase http://dx.doi.org/10.1063/1.4905124 [ * * , ( ) , http://dx.doi.org/10.1063/1.4905124 ] link:\doibase 10.1021/ct5004252 [ * * , ( ) ] , \doibase http://dx.doi.org/10.1063/1.1829045 [ * * , ( ) , http://dx.doi.org/10.1063/1.1829045 ] link:\doibase 10.1021/cr200137a [ * * , ( ) ] , @noop * * , ( ) \doibase http://dx.doi.org/10.1063/1.4927594 [ * * , ( ) , http://dx.doi.org/10.1063/1.4927594 ] \doibase http://dx.doi.org/10.1063/1.4932595 [ * * , ( ) , http://dx.doi.org/10.1063/1.4932595 ] | development of exponentially scaling methods has seen great progress in tackling larger systems than previously thought possible .
one such technique , full configuration interaction quantum monte carlo , is a useful algorithm that allows exact diagonalization through stochastically sampling determinants .
the method derives its utility from the information in the matrix elements of the hamiltonian , along with a stochastic projected wave function , to find the important parts of hilbert space .
however , the stochastic representation of the wave function is not required to search hilbert space efficiently , and here we describe a highly efficient deterministic method to achieve chemical accuracy for a wide range of systems , including the difficult cr@xmath0 dimer . we demonstrate that such calculations for systems like cr@xmath0 can be performed in just a few cpu hours . in addition
our method also allows efficient calculation of excited state energies , for which we illustrate with benchmark results for the excited states of c@xmath0 .
_ introduction : _ the scope of traditional approaches to full configuration interaction ( fci ) has been limited to simple diatomic molecules @xcite , and there has been little progress in diagonalizing spaces much larger than a billion determinants in recent times @xcite . however , recent progress in alternative approaches to fci problems has increased the scope of fci beyond simple diatomic molecules .
two techniques in particular have been important in this progress , full configuration quantum monte carlo ( fciqmc ) @xcite , and density matrix renormalization group ( dmrg ) @xcite .
both algorithms provide unique advantages , with dmrg being the definitive method for systems in which one can identify degrees of freedom with low levels of entanglement @xcite , and fciqmc showing promise for molecules and extended systems in two or more dimensions @xcite .
the success of dmrg and fciqmc in quantum chemistry is highlighted by their recent applications to unprecedented large - size determinant spaces while also achieving chemical accuracy @xcite .
the fciqmc method is a useful technique with a few limitations which include biased sampling and comparatively computationally expensive simulations .
the biased sampling is a result of the initiator approximation @xcite , generally employed in fciqmc calculations which additionally limits the space in which determinants can be sampled .
the initiator approximation can cause errors in fciqmc calculations that can be unexpectedly large @xcite .
the need for this approximation is related to the monte carlo sampling and not necessarily related to power of the technique which we suggest is finding important determinants . in this letter we suggest an alternative to stochastic sampling in favor of a completely deterministic version of the fciqmc technique that efficiently samples the determinant space .
we denote this deterministic fciqmc algorithm by adaptive sampling ci ( asci ) .
our approach should be contrasted to other techniques for finding energetically important determinant subspaces .
the majority of traditional ci methods encode relevant physical degrees of freedom based on excitation levels from a reference determinant @xcite .
these excitation - based methods can also suffer from inaccuracy as they miss important parts of determinant space . within the ci framework
several promising ways to circumvent this problem have been suggested that focus on selected ci approaches where one selects relevant determinants based on different criteria @xcite . after presenting the asci method ,
we establish a connection between fciqmc and various selected ci approaches .
we then apply the asci method to the cr@xmath0 dimer , a classic hard problem for many computational electronic structure methods @xcite .
finally , we demonstrate that excited states are straightforward to calculate with asci .
we note that the calculation of excited states within stochastic fciqmc is possible , but require specialized techniques @xcite , or stochastic orthogonalization between walker sets , which is quite different from the method described here . _ a path to a deterministic algorithm _ : in the initial development of fciqmc , one of the original improvements on the method was to take part of the projection step and make it deterministic @xcite . in this work ,
we go further and develop a completely deterministic algorithm .
our approach here is to find important determinants in the same manner as fciqmc , i.e. to sample determinants based on the absolute value of the ground state wave function amplitudes .
the fciqmc technique was originally presented as a projector method in imaginary time and we use this approach to motivate our method .
we start by expanding a wave function in the space of determinants , @xmath1 and the propagator in imaginary time , @xmath2 which has an asymptotic solution of a stationary state with @xmath3 . in fciqmc ,
the parameter @xmath4 is a free parameter that controls the population of walkers . here
we will consider @xmath4 to be the ground state energy or our best approximation thereof .
the power of fciqmc is that it ignores the unimportant parts of determinant space , finds important determinants , and samples them according to their amplitudes . for a stationary state we can solve for the individual coefficients as @xmath5 .
the rhs of this equation captures all aspects of the fciqmc algorithm .
the @xmath6 in the numerator corresponds to the spawning step ( transition moves between determinants ) , the sign of @xmath6 in the numerator and specifically summing over positive and negative terms corresponds to the annihilation step ( cancelation of positive and negative walkers ) , and the denominator corresponds to the death / cloning step ( adding or removing walkers from the simulation ) .
the key to turning fciqmc into a non - stochastic algorithm is to remove the stochastic sampling and replace it with a deterministic ranking of important determinants . in both approaches
the hamiltonian matrix elements and a wave function are needed . in fciqmc
the wave function is represented by the distribution of walkers at any given step , whereas for our deterministic algorithm , we use an approximate wave function at each iteration as follows , @xmath7 in this equation , we have labeled the coefficients of the initial wave function as @xmath8 , and the output coefficients as @xmath9 .
thus for any good approximation to the ground state wave function we can use eq .
[ trialwfs ] to determine the importance of determinants in a much larger space that what is initially included in @xmath8 .
this @xmath9 is essentially a first - order perturbation estimate for ci coefficients in the epstein - nesbet perturbation theory @xcite . _
the deterministic algorithm _ : the technique described in this section can be considered a variant of cipsi ( configuration interaction by perturbation with multiconfigurational zeroth - order wave functions selected by iterative process ) @xcite , with a modified search procedure . for the largest systems considered here ,
only a modest amount of memory is needed , and all parts except the diagonalization step can be trivially parallelized .
the algorithm is defined by two determinant subspaces : a core space of size _ cdets _ and a target space of size _
tdets_. the _ core space _ determines the number of terms @xmath10 we include in the sum in eq .
[ trialwfs ] . this is to say we select @xmath10 determinants with the largest c@xmath11 , and consider only those to be non - zero in the sum of equation eq .
[ trialwfs ] .
the space to be searched is the set of single and double excitations of the core set of determinants . since our objective is to find the determinants with the largest amplitudes , we in general only need to search determinants connected to those with large amplitudes .
we illustrate the use of this approximation with numerical tests in the next section .
the _ target space _ contains the top _ tdet _
determinants , as determined from the ranking , and is the rank of the matrix diagonalized in each iteration .
initalize : set size of core ( _ cdets _ ) and target space ( _ tdets _ ) . set the starting wave function to the hartree fock wave function ( c@xmath12 = 1 , c@xmath13 ; e = e@xmath14 ) .
\(1 ) evaluate the perturbed wave function amplitudes over al the single and double substitutions from the core space . @xmath15
\(2 ) for the core determinants of the current wave function \{c@xmath16}@xmath17 , and the current perturbed wave function \{a@xmath16}@xmath18 , select the _
tdet _ largest absolute values to define the new target space .
\(3 ) form and diagonalize @xmath19 in the target space .
\(4 ) the lowest eigenvalue is the new energy @xmath4 .
the largest _ cdet _ amplitudes , by magnitude , define the new core space . if the energy is not converged , return to step ( 1 ) .
the determinants found at the end of the simulation will in general be the most important determinants for the ground state wave function . unlike fciqmc
this technique has no population bias , initiator bias , and no sign problem .
the technique described here provides an inherently variational energy .
however , it is possible to extend the accuracy of the technique with perturbation theory , which comes at the expense of the energy no longer being variational @xcite .
_ discussion _ : when performing an asci calculation the first few steps involve the exploration of higher order excitations . if the starting wave function is a single determinant , then each step increases the maximum number of excitations that have been explored by 2 .
the coefficients @xmath20 , as calculated in step ( 1 ) , can span all single and double excitations from the current wave function , and is generally very large .
truncating the coefficients that are calculated for @xmath20 in the main part of the self - consistent loop allows us to consider applying our algorithm to large systems . to maintain size consistency , the value of _ tdets _ and _ cdets _ will have to grow with system size . for the systems considered in this work , it was possible to converge the value of _ cdets _ , as the energy with respect to this parameter can be extrapolated by running with a few different values .
apart from the diagonalization step , the most computationally expensive task is forming the @xmath20 matrix and finding its largest values .
this can be split among many processors , with the only non - trivial communication occurring during the final aggregation of the final values .
further improvements are possible by generating the natural orbitals after an initial run and using them to recalculate the electron integrals . the natural orbitals can be generated at any point in the simulation from the current best wave function .
the natural orbitals are generally thought to produce highly compact representations of a wave function @xcite . for a difficult system like cr@xmath0
, we find the use of natural orbitals to be crucial to obtain accurate energies .
before we present our results we consider numerical tests for our approximation of the matrix @xmath20 . in general , the coefficients of the ground state wave function will span many orders of magnitude , and searching over the determinants with the smallest amplitude coefficients is inefficient . using _
cdets _ as a parameter to limit the search to only the important determinants is a well controlled approximation that can be converged .
this approach has similarities to the fciqmc initiator approximation @xcite .
we demonstrate the accuracy of this approximation by considering cn in an sto-3 g basis ( 6240 determinants ) . we show that we can find the most important determinants in hilbert space without having to perform a diagonalization over a larger determinant space .
table [ tab3 ] provides a comparison of the most important determinants found with different values of _ cdets _ and _ tdets _ to those obtained from a full diagonalization over the entire space . for a simulation of 100 core determinants and a target space of 200 determinants ( 100/200 ) we found 182 of the top 200 determinants ( 91% ) .
the remaining 9% of missed determinants were found to be close in amplitude to the determinants that replaced them in the target space .
similar result can be seen for for all the simulations presented .
these results suggest that for some simulations that the search algorithm is nt highly dependent on the size of the core space and the small percentage of determinants that are missed by the algorithm are replaced by determinants that are similar in importance .
thus we argue that extremely high accuracy is not needed in determining the ranking order .
this approximation is even further reduced when perturbation corrections are used , which can correct for any important determinants that were missed .
.test of the cn dimer using the full search algorithm .
the energies are in units of ha .
the columns with top 200 ; is the number of determinants we found that agree with the top 200 determinants , by amplitude , of the exact answer . likewise for top 400 and top 800.
the total fci space is 6240 determinants [ tab3 ] [ cols="<,^,>,>,>,>",options="header " , ] for c@xmath0 simulations , the convergence of energies to chemical accuracy was easily achieved , and neither the use of natural orbitals nor perturbation corrections are needed . in comparison with the exact results
@xcite , we were able to achieve an accuracy of 1 mha using a diagonalization no larger than 200,000 determinants .
the total computer time for a simulation of this size was less than 2 cpu hours .
results are presented in table [ c2tab ] , using different values of _ tdets _ and _
cdets_. _ cr@xmath0 with the sv basis set _ : fig .
[ cr2ene ] shows the convergence of our results for cr@xmath0 . in order to make a comparison with previous studies @xcite ,
cr@xmath0 calculations were carried out with the sv basis set @xcite at 1.5 with 24 active electrons in 30 orbitals and a frozen core .
a compact representation of the wave function in this system is dependent on having a good set of orbitals . as part of our algorithm
, we run a preliminary calculation with a target space of 100,000 determinants , after which we calculated the natural orbitals .
the resulting natural orbitals were used to recalculate the integrals for the production run .
the total energy for our most accurate simulation converged to within 16 mha of the predicted full ci basis set energy @xcite .
energy as a function of tdet size , with the largest target space going up to 1 million determinants .
the plotted energies have been shifted by 2086 ha .
the pt correction line represents our result with the added perturbation theory correction , which brings our final energy to within 1 mha of the predicted exact result .
our best energy without the perturbation correction is -2086.40388 ( ha ) . with the perturbation correction it is -2086.4203 ( ha ) .
the dmrg benchmark is -2086.420948 ( ha ) @xcite . ]
a perturbation theory analysis @xcite was performed bringing the final energy within 1 mha of the predicted exact result .
the timing for the largest simulation ( _ _ tdets__=10@xmath21 ) , including the initial run for calculating the natural orbitals but not including the perturbation correction , requires approximately 7 cpu hours when run on a single core of a 2.40 ghz intel xeon processor .
while this is a relatively fast calculation , we expect further improvements will speed up the simulation significantly . to demonstrate the distribution of the determinants located by asci we plot a histogram of excitations from the dominant determinants for cr@xmath0 in fig .
[ crdist ] .
the ratio between different excitations sectors does not change much in increasing _ tdet _
values from 10@xmath22 to 10@xmath21 . for this range ,
the quadruple excitations generally make up half the wave function , while higher excitations above the quadruples make up roughly 30% of the wave function . by excitations from the dominant determinant .
the x - axis is excitation level from the dominant determinant , and the y - axis is the fraction of determinants .
plots ( a ) , ( b ) , ( c ) , ( d ) have tdets = 100k , 400k , 800k , and 1 million respectively .
simulations with different tdet spaces ( for the ones shown here ) have roughly the same fractional importance in the different excitation levels .
thus even for our smallest tdet simulations , large determinant excitations are important .
, title="fig : " ] by excitations from the dominant determinant .
the x - axis is excitation level from the dominant determinant , and the y - axis is the fraction of determinants .
plots ( a ) , ( b ) , ( c ) , ( d ) have tdets = 100k , 400k , 800k , and 1 million respectively .
simulations with different tdet spaces ( for the ones shown here ) have roughly the same fractional importance in the different excitation levels .
thus even for our smallest tdet simulations , large determinant excitations are important .
, title="fig : " ] by excitations from the dominant determinant .
the x - axis is excitation level from the dominant determinant , and the y - axis is the fraction of determinants .
plots ( a ) , ( b ) , ( c ) , ( d ) have tdets = 100k , 400k , 800k , and 1 million respectively .
simulations with different tdet spaces ( for the ones shown here ) have roughly the same fractional importance in the different excitation levels .
thus even for our smallest tdet simulations , large determinant excitations are important .
, title="fig : " ] by excitations from the dominant determinant .
the x - axis is excitation level from the dominant determinant , and the y - axis is the fraction of determinants .
plots ( a ) , ( b ) , ( c ) , ( d ) have tdets = 100k , 400k , 800k , and 1 million respectively .
simulations with different tdet spaces ( for the ones shown here ) have roughly the same fractional importance in the different excitation levels .
thus even for our smallest tdet simulations , large determinant excitations are important .
, title="fig : " ] _ excited states _ : we are also able to calculate excited states with our technique , as they are obtained automatically within the diagonalization procedure . for a 6 - 31 g * basis of c@xmath0 , we compare against previous fci simulations for the first two excited states @xcite . for a simulation at a distance of 1.25 with cdets=@xmath23 and tdets=800000
, we have the following energies for the ground state and the first two excited states ( @xmath24,@xmath25,@xmath26 ) in units of ha .
the exact results are ( @xmath27,@xmath28,@xmath29 ) @xcite .
thus , although the accuracy of the excited states is not as good as the ground state , it is still straightforward to achieve chemical accuracy ( which is generally defined as 0.0016 ha ) .
we can improve the accuracy of excited states by noting that any eigenstate can be used in eq .
[ trialwfs ] .
thus we can find and rank determinants by their importance to individual excited states .
the excited state optimization can be done simultaneously with the ground state method , or in a state - by - state bootstrap method . for the simultaneous optimization algorithm
, we determine a set of important determinants to retain for each excited state .
there is a separate search step for each excited state , but one diagonalization step that combines all retained determinants .
in contrast , the bootstrap method would converge each excited state one by one , where at each step determinants would be added in specifically for the targeted excited state .
the use of natural orbitals averaged over various excited states may also be used to improve the algorithm @xcite .
a detailed study of the targeted excited state technique will be presented elsewhere .
_ connection to selected ci _ : as mentioned earlier , the asci method may also be considered a variant of selected ci techniques @xcite in which there is considerable current interest @xcite .
our method employs first - order perturbation coefficients for selecting determinants , and is thus closely resembles the ci method , cipsi .
another related technique is the @xmath30+sd - ci @xcite method which uses a one - step energy criteria , and a one - step approach together with our eq .
[ trialwfs ] in order to find important determinants . despite these similarities ,
none of these algorithms have been pushed to achieve chemical accuracy for hard systems , and do not appear to have been benchmarked against fciqmc or dmrg .
the largest @xmath30+sd - ci calculations included roughly 50,000 determinants and attained 13 mha accuracy for c@xmath0 6 - 31 g * ( comparable to our results in table [ c2tab ] ) .
as shown in this work , it is easy to go more than an order of magnitude in accuracy using our iterative scheme , without significantly increasing the computational effort .
the largest selected ci techniques we are aware of have been extended up to 4 million determinants @xcite , but for systems in which no benchmarks exist . for the future development of asci and other selected ci techniques , it is important to consider how such methods are different from standard ci methods .
the difference is largely due to the construction of the hamiltonian .
selected ci techniques need unique data structures in order to construct the hamiltonian efficiently @xcite .
a previous study demonstrated that much larger scale simulations , than what we presented here , is possible for selected ci techniques @xcite .
we are currently considering various data structures used previously @xcite and new structures , to determine the best way to scale up our simulations .
_ conclusions _
: we have shown that the underlying dynamics of fciqmc can be used to generate a deterministic algorithm that can be efficiently used to calculate both ground and excited states of chemical systems .
we have applied this technique to a known difficult problem in electronic structure theory , the cr@xmath0 molecule , and shown that chemical accuracy can be achieved with the cpu power available on any modern computer .
our results suggest that the asci method ( and selected ci methods in general ) should be considered as a state of the art ci method in both accuracy and efficiency
. it will be interesting to determine where asci stands in relationship to dmrg and fciqmc , as all these methods have different strengths .
certainly the use of asci and fciqmc is currently important since dmrg and post - dmrg methods are not yet well suited for simulations in two and three dimensions .
the asci method also distinguishes itself from fciqmc in that excited states and other properties , such as the 2-rdm , are inherently easy to calculate @xcite . |
the nuclear dynamics is a complex problem joining all the puzzling aspects of quantum mechanics to the difficulties of many - body systems . besides these factors
, the strong interaction , which is up to date not completely understood , adds new challenges for calculations in the nonperturbative regime .
collective nuclear phenomena , as fission , particle or cluster evaporation and nuclear fragmentation , offer the possibility of studying those complex features of the nuclear dynamics . aside the interest from the fundamental nuclear physics , there are many applications where the knowledge of fragment formation would be helpful .
for instance , information on intermediate mass fragments ( imf ) cross section is relevant for the design of accelerator - driven systems ( ads ) and radioactive ion - beam ( rib ) facilities and also in the study of resistance of materials to radiation .
imf are particles with @xmath1 4 but lighter than fission fragments , i.e. , @xmath2100
. fragments in high energy nuclear collisions are produced by spallation , fission , and multifragmentation .
j. hufner @xcite using as classification parameters the mass number of the fragments , @xmath3 , and their multiplicity , @xmath4 , defined the processes in the following way : * spallation is the process in which only one heavy fragment with mass close to the target mass @xmath5 is formed ( a special case of spallation is the so - called deep spallation where @xmath6 but @xmath7 ) ; * fission is the process in which @xmath8 and @xmath3 is arround @xmath9 ; * multifragmentation is the process where @xmath10 and @xmath11 .
emission of light particles , with atomic number @xmath12 , usually dominates the yield of reaction products for light target nuclei , while for heavy targets also spallation and fission residua give significant contribution . in this scenario
the formation of imf can happen through one of the following processes : * fission of nuclei mass number in the range 120 - 130 @xcite .
* spallation including the emission of imf , the so - called associated spallation @xcite .
* multifragmentation of heavy nuclei @xcite .
following this reasoning , as discussed below , for heavy targets the multifragmentation would be the only mechanism for the formation of imf . indeed in ref .
@xcite it was found that in the reaction of 3.65@xmath3 @xmath13pb with @xmath14h the formation of @xmath15c nuclei presents the characteristics of multifragmentation , with a possible small contribution of binary process .
however the formation of imf was observed even at lower energies @xcite and in this case the dynamics was characteristic of a binary process with no evidences for multifragmantation .
hence the study of production of imf from heavy target nuclei at intermediate energies can give new information on the nuclear dynamics .
in the present work our objective is the measurement of the cross sections of residual nuclei in the imf region for reactions on heavy nuclei induced by 660 mev protons .
the experimental data will be analysed by comparison with monte carlo calculations with the crisp code @xcite , as described below .
it is generally assumed that at intermediate energies the nuclear reaction proceeds in two stages : 1 ) the incoming fast projectile collides with a single nucleon or with several nucleons transferring momentum and energy to the nucleus and leaves the nucleus accompanied by several light particles ; 2 ) the residual nucleus achieves statistical equilibrium and some nucleons or clusters are emitted from the excited nucleus . at higher energies , when the excitation energy of the residual nucleus per nucleon is @xmath16 3.5 mev / a , the multifragmentation of the nucleus can take place .
this reaction mechanism differs from evaporation since it describes a sudden breakup of the nucleus instead of the successive emission of particles . in the framework described above the formation of imf from heavy targets at intermediate energies
could only be attributed to fission inside a long evaporation chain ( both pre- and post - fission ) , which is very unlike .
in fact the fission probability of heavy nuclei drops very fast as mass number decreases @xcite , so long evaporation chains would lead to lower fissility nuclei .
another possibility would be a very long evaporation chain leading to light spallation products .
this mechanism is limited by the maximum excitation energy allowed before multifragmentation becomes dominant , because the evaporation would cool down the nucleus before the imf region is reached . increasing the excitation energy above the 3.5 mev /
a threshold would only increase the contribution from multifragmentation and in this way imf formed in reactions with heavy targets should be dominated by fragmentation products .
hence , for excitation energies below the multifragmentation threshold the formation of imf from heavy nuclei would be very unlike .
a comprehensive set of data on the proton - induced reaction of @xmath17u at 1 agev @xcite presents the cross sections of a 254 light nuclides in the element range @xmath18 measured in inverse kinematics . on the base of a detailed study the experimental kinematical information the authors qualified such nuclides as binary decay products of a fully equilibrated compound nucleus , whereas clear indications for fast breakup processes was absent . these result
are corroborated by those from ref .
@xcite and @xcite and are in contradiction with the scenario described in ref .
@xcite .
one way to overcome the problem posed by binary production of imf from heavy targets is by including the possibility of highly asymmetric fission with fragments which can still undergo evaporation to form at the end a nuclide in the region of imf .
this process corresponds to a modification in the classification given by hufner for fission by using a less restrictive definition for fission , since in this case the fragment would have mass number quite different from @xmath19 .
this superasymmetric mechanism would corroborate the conjecture that evaporation and fission are manifestation of a single mechanism , called binary decay @xcite .
a complete study of this possibility involves the description of the entire process from the primary interaction of the incident proton up to the evaporation of nucleons from the fission fragments .
a task like this can only be performed through the monte carlo method . in this work
we use the crisp code to calculate all the features of the nuclear dynamics during the reaction .
crisp is a monte carlo code for simulating nuclear reactions @xcite that assumes that nuclear reactions can be separated in the two stages mentioned above : the intranuclear cascade and the evaporation / fission process .
this code has been developed during the last 25 years , while it has successfully described many different reactions .
the main characteristic of the intranuclear cascade calculation with crisp is the multi - collisional approach @xcite , where the full nuclear dynamics is considered in each step of the cascade . in this process
the nucleus is modeled by an infinite square - potential which determines the level structure for protons and neutrons .
the effects of the nuclear potential are present in the transmission of the particles through the nuclear surface or through an effective mass according to wallecka mean field approximation @xcite .
this multicollisional calculation is accomplished by constantly updating all the kinematical variables of all particles inside the nucleus , what opens the possibility of treating more realistically many nuclear phenomena .
for instance , the antisymmetrization criteria , which determines a strict observation of the pauli principle is performed through a strict observance of the pauli principle , allowing the use of physical criteria for separating the intranuclear cascade from the thermalization process @xcite .
the effectiveness of such approach can be verified in processes which are predominantly dependent on the intranuclear cascade step , as kaon production and hypernuclei decay , which have been studied with crisp leading to results compatible with experiments @xcite .
also the fission of several nuclei has been studied in the quasi - deuteron region @xcite , where pauli blocking mechanism is very important in the determination of the residual nucleus .
the evaporation / fission competition was first studied in ref .
@xcite , giving for the first time an explanation to the saturation of the fissility of heavy nuclei below the unit observed in photofission experiments at intermediate energies @xcite , and afterward it was extended to energies up to 3.5 gev @xcite , showing also in this case good agreement with experimental data from ref @xcite .
after the nuclear thermalization , the competition between fission and evaporation processes , which includes neutrons , protons and alpha - particles , is determined through the ratios between their respective widths according to the weisskopf model for evaporation and to the bohr - wheeler model for fission .
resulting in @xmath20 \right\rbrace , \label{gamap}\end{aligned}\ ] ] and @xmath21 \right\rbrace .
\label{gamaa}\ ] ] for evaporation and by @xmath22 \right\rbrace , \label{gamaf}\end{aligned}\ ] ] where , @xmath23}{(4a^{2/3}a_f e_n ) } , \label{kf}\end{aligned}\ ] ] for fission .
the parameters @xmath24 stand for the density levels calculated by the dostrovsky s parametrization @xcite and @xmath25 is given by @xmath26 where @xmath27 , @xmath28 and @xmath29 are the separation energies for neutrons , protons and alphas , respectively and @xmath30 is the fission barrier .
@xmath31 stands for the coulomb potential . at each step @xmath32 of the evaporation
the excitation energy of the compound nucleus left is modified by @xmath33 where @xmath34 is the kinetic energy of the emitted particle .
if the nucleus undergoes fission , the the fragments are determined according to the multimodal - random neck rupture model(mm - nrm ) @xcite , which takes into account the collective effects of nuclear deformation during fission through a liquid - drop model and single - particle effects through microscopic shell - model corrections .
the microscopic corrections create valleys in the space of elongation and mass number , each valley corresponding to one different fission mode @xcite . according to the mm - nrm the fragment mass distributions
are determined by the uncorrelated sum of the different fission modes . at present
it is supposed that there are three distinct fission modes for the heavy nuclei : symmetric syperlong ( s ) mode and two asymmetric modes standard i ( s1 ) , standard ii ( s2 ) . in the superlong mode
the fissioning system with mass @xmath35 presents itself at the saddle - point in an extremely deformed shape with a long neck connecting the two forming fragments , which will have masses around @xmath36 .
the standard i mode is characterized by influence of the spherical neutron shell @xmath37 and of the proton shell @xmath38 in the heavy fragments with masses @xmath39 .
the standard ii mode is characterized by influence the deformed neutron shell closure @xmath40 and proton shell @xmath41 in the heavy fragments with masses @xmath42 .
the fission cross section as a function of mass number is then obtained by the sum of three gaussian functions , corresponding to the three modes mentioned above @xcite : @xmath43 , \label{masstotal } \end{split}\end{aligned}\ ] ] where @xmath44 is the mean mass number determining the center of gaussian functions ; and @xmath45 , @xmath46 , and d@xmath47 are the intensity , dispersion and position parameters of the @xmath48 gaussian functions . the indexes @xmath49 , @xmath50 designate the asymmetric and symmetric components . the crisp code works on an event - by - event basis , and therefore the parameter @xmath44 in eq.3 is completely determined by the mass of the fissioning nucleus @xmath35 , that is , @xmath51 . the quantities
@xmath44 + d@xmath52 = a@xmath53 and @xmath44 - d@xmath52 = a@xmath54 , where @xmath55 and @xmath56 are the masses of the heavy and light fragment , respectively , determine the positions of the heavy and light peaks of the asymmetric components on the mass scale .
the values of a@xmath53 + a@xmath54=2@xmath44 is treated as the masses of nuclei that undergo fission in the respective channel .
one important observable of fission is the charge distribution of a given isobaric chain with mass number @xmath3 .
it is assumed that the charge distribution of the fission fragments is well described by a gaussian function characterized by the most probable charge , @xmath57 , of an isobaric chains @xmath3 ( centroid of the gaussian function ) and the associate width parameter , @xmath58 of the distribution as following @xcite : @xmath59 where @xmath60 is the independent cross section of the nuclide , @xmath61 .
@xmath57 stands for the most probable charge for the charge distribution of an isobar chain with mass number @xmath3 and @xmath58 is the corresponding width parameter .
the values @xmath57 and @xmath58 can be represented as a linear functions of the mass numbers of fission fragments , @xmath62 and @xmath63 here , @xmath64 and @xmath65 are parameters fixed through a systematic analysis of atomic number distributions of fission fragments .
the values for all parameters used in the prsent work are reported in table [ table1 ] . with the code described above ,
many spallation and fission calculations were performed , showing good agreement with the available experimental data .
we emphasize that the crisp code is able to simulate nuclear reactions of several kinds , as those initiated by protons @xcite , photons @xcite , electrons @xcite or hypernuclei @xcite , with energies from 50 mev up to 3.5 gev , and on nuclei with masses going from @xmath66 12 up to @xmath66240 and with several observables : spallation products , strange particles , fission products , hyperon - decay particles , fragment mass and atomic number distributions .
the code has been applied in the study for development of nuclear reactors @xcite thus the crisp code is a reliable tool to investigate properties of nuclear reactions .
a natural uranium target of 0.164 g and 0.0487 mm thick and neptunium target of 0.742 g and 0.193 mm thick were exposed to an accelerated proton beam of 660 mev in energy from the lnr phasotron , joint institute for nuclear research ( jinr ) , dubna , russia @xcite .
the proton flux was determined by the use of an al monitor @xcite .
the monitor was irradiated with targets and had the same size as the targets .
the cross section for the @xmath67al(p,3pn)@xmath68na reaction equals [email protected] mb at 600 mev of proton energy was used .
the irradiation time was 27 min at an proton beam intensity of about 3 @xmath70 10@xmath71 protons per min .
the induced activity of the targets was measured by two detectors , an hpge detector with the 20% efficiency and an energy resolution of 1.8 kev ( 1332 kev @xmath72co ) for the @xmath17u target and a ge(li ) detector with the 4.8% efficiency and a resolution of 2.6 kev ( 1332 kev @xmath72co ) for the @xmath73np target .
recording of the g spectrum was carried out with a high - rate 4-input multichannel buffer master 919 , which automaticaly determined the dead time .
the identification of the reaction products , and the determination of their production cross section , were performed using data from @xcite by means of half - lives , energies and intensities of @xmath74-transition of radioactive fragments . in the absence of a parent isotope
, the cross section of fragment production for each fragment is determined using the following equation : @xmath75 where @xmath76 denotes the cross section of the reaction fragment production ( mb ) ; @xmath77 is the area under the photopeak ; @xmath78 is the projectile beam intensity ( min@xmath79 ) ; @xmath80 is the number of target nuclei ( in 1/@xmath81 units ) ; @xmath82 is the irradiation time ; @xmath83 is the time of exposure between the end of the irradiation and the beginning of the measurement ; @xmath84 is the time measurement ; @xmath85 is the decay constant ( min@xmath79 ) ; @xmath86 is the intensity of @xmath74-transitions ; @xmath87 is the total coefficient of @xmath74-ray absorption in target and detector materials , and @xmath88 is the @xmath74-ray - detection efficiency . usually , the cross section of an isotope production in the reaction under investigation is direct and independent ( i ) of the parent nuclei decay , and the cross section is determined by eq .
( [ g1 ] ) .
if the yield of a given isotope receives a contribution from the @xmath89-decay of neighboring unstable isobars , the cross section calculation becomes more complicated @xcite . if the formation probability for the parent isotope is known from experimental data or if it can be estimated on the basis of other sources , then the independent cross sections of daughter nuclei can be calculated by the relation : @xmath90,\end{aligned}\ ] ] where the subscripts @xmath3 and @xmath91 label variables referring to , respectively , the parent and the daughter nucleus ; the coefficient @xmath92 specifies the fraction of nuclei @xmath3 decaying to nuclei @xmath91 ( this coefficient provides the idea of how much the @xmath93-decay affects our data ; and @xmath94 , when the contribution from the @xmath93-decay corresponds 100% ) ; and @xmath77 is the total photo peak area associated with the decays of the daughter and parent isotopes .
the effect of the forerunner can be negligible in some limit cases for example , in the case where the half - life of the parent nucleus is very long , or in the case where the fraction of its contribution is very small . in the case
when parent and daughter isotopes could not be separated experimentally , the calculated cross sections are classified as cumulative ones ( c ) .
the cross sections for fragment production from reactions induced by 660 mev protons on uranium and neptunium are shown in fig .
1 . in both cases
a prominent peak is observed around the symmetric fragment mass , which is indeed composed by fragments from the symmetric fission mode , but also presents a contribution from two asymmetric modes @xcite .
some analysis on the mass distribution for the p+237np and p+238u systems have been performed for the mass range 70@xmath95 150 @xcite . in this work
we have added data in the region of intermediate mass fragments corresponding to @xmath96 and in the region @xmath97 .
the measured cross sections for the nuclei in this mass range are listed in table i , where the quoted errors include contributions from those associated with the statistical significance of experimental results ( 2 - 3% ) , those in measuring the target thickness ( 3% ) , and those in determining the detector efficiency ( 10% ) .
usually , studies on the production of fission fragments do not extends to light nuclei [ 5 ] and the inclusion of this region in our analysis can bring up interesting features of the dynamics for fission fragments production .
in fact , theoretical calculations , based on the mass asymmetry parameter and fission barrier height @xcite have shown that for heavy targets , and at intermediate or low energies , the cross sections for imf are very small . as a consequence most of the experimental observations available in the literature for fission seems to die out for atomic numbers below z=28 . in fig .
[ figmassdist ] we can observe a shoulder formed in the region @xmath96 for both 238u and 237np distributions .
the presence of imf in reactions at a relatively low energy as in the case of the present study can hardly be attributed to multi - fragmentation .
the observation of another shoulder at the region of @xmath98 , for both distributions , reinforces the idea of a binary process as the origin of the imf .
these observations , therefore , are in agreement with the results obtained by ricciard et al .
@xcite . in this work we present the results of a study , performed with the simulation code crisp , for the possibility of describing the new experimental data on the light mass region as a result from fission or spallation . to this end a super - asymmetric fission mode , as described in the next section , was included in the code . in fig .
1 we observe that for both @xmath99 and @xmath100 there is a shoulder formed by fragments in the region 30@xmath10170 . the presence of imf in reactions at a relatively low energy as in the case of the present study can hardly be attributed to multifragmentation .
the fact that in both cases we observe also another shoulder at the region 170@xmath101200 reinforces the idea of a binary process as the origin of the imf and therefore our results confirm the findings in ref .
the two known binary processes in nuclear reactions are evaporation and fission . in this work
we study the possibility of describing the new experimental data presented here as resulting from fission or from spallation according to the description given in the crisp code . to this end
a superasymmetric fission mode was included , as described below .
to take into account the possibility of a super - asymetric fission we included in the crisp code another mode , s3 , described by the usual gaussian shape from mm - nrm , @xmath102 , \label{massasym}\end{aligned}\ ] ] as in the case of the three modes previously analized in ref .
@xcite , @xmath103 , @xmath104 and @xmath105 are fitting parameters which allow us to describe the experimental data for fragments produced through the fission channel .
in addition to fission we calculated the mass distributions for fragments produced in the deep - spallation process . the results are shown in figs .
[ figmassdist : subfig1:a ] and [ figmassdist : subfig1:b ] where we observe that with the inclusion of the superasymmetric mode the experimental data is well described by the fission mechanism according to the crisp calculations .
the deep - spallation mechanism gives only a very small contribution in the heavy fragments region , showing that in fact the superasymmetric fission is the relevant mechanism for the production of fragments in the region 160@xmath101200 .
the best - fit values for the parameters used in the nn - rnm approach are shown in table [ table2 ] .
the parameters for the modes s , s1 and s2 were already discussed in ref .
@xcite therefore we focuss on the parameters for s3 .
the superasymmetric mode contributes with 0.6% and 1.2% of the total fission cross section for @xmath17u and for @xmath73np , respectively .
the total fission cross sections are 1140 mb for @xmath17u and 1360 mb for @xmath73np .
the s3 width is somewhat larger than those for s1 and s2 but smaller than that for s mode .
the most stricking feature of the superasymmetric mode is the mass number gap around 60 a.m.u . in respect with the symmetric fragment for both cases studied here .
our results confirm that imf at intermediate energies are formed predominantly through a binary process , and that it is described by a superasymmetric fission mode .
although a good description of the fragment production was obtained here with the fisson mechanism , it is not possible to exclude that the same description of the experiemntal data would not be obtained by a mechanism of evaporation which includes the associated spallation , with emissions of fragments heavier than the alpha - particle .
in fact it was shown in ref .
@xcite that , at least in the ground - state , a global description of fission and cluster emission would be possible in a single theoretical framework . in
would be interesting to investigate if such an approach would be valid also for the decay of excited compound nuclei .
the cross sections for fragment production in proton - induced fission on @xmath106u and @xmath107np at 660 mev were measured at the lnr phasatron ( jinr ) .
the fragment mass distributions covering the region 20@xmath101200 , allowed the study of intermediate mass fragments ( imf ) with 20@xmath10170 .
it was found that to the imf observed there is a heavier counterpart in the region 170@xmath101200 , indicating that the formation of imf could be due to a binary process .
this hypothesis was tested with the use of the crisp code by including an addition superasymmetric fission mode described according to the mm - nrm approach .
the results show that it is indeed possible to give an accurate description of the fragment production in the entire region , 20@xmath101200 by the evaporation / fission mechanism adopted in the crisp code with the usual fission modes , namely , one symmetric and two asymmetric , and including a forth superasymmetric mode .
this last mode produces fragments that are around 60 a.m.u .
far from the symmetric fragment mass and contributes with 0.6% and 1.2% of the total fission cross section for @xmath106u and @xmath107np , respectively .
our results are in agreement with previous results @xcite evidencing the binary production of imf at intermediate energy nbuclear reaction .
g. karapetyan is grateful to fundao de amparo pesquisa do estado de so paulo ( fapesp ) 2011/00314 - 0 and 2013/01754 - 9 , and also to international centre for theoretical physics ( ictp ) under the associate grant scheme . a.deppman
acknowledge the partial support from cnpq under grant 305639/2010 - 2 and fapesp under grant 2010/16641 - 7 .
e. andrade acknowledge the support from fapesp under grant 2012/13337 - 0 .
99 j. hufner , phys .
@xmath108 , 129 ( 1985 ) .
w. loveland @xmath109 , phys . rev .
c @xmath110 , 464 ( 1981 ) . y. yariv and z. fraenkel , phys .
c @xmath111 , 2227 ( 1979 ) .
b. grabez , phys .
c @xmath112 , r2144 ( 1993 ) .
m. v. ricciardi , p. armbruster , j. benlliure @xmath109 , phys .
c @xmath113 , 014607 ( 2006 ) .
a. a. kotov , l. n. andronenko , m. n. andronenko @xmath109 , nucl .
phys . a @xmath114 , 575 ( 1995 ) .
h. w. barz , j. p. bondorf , h. schulz @xmath109 , nucl .
phys . a @xmath115 , 714 ( 1986 ) .
a. deppman , e. andrade - ii , v. guimaraes , g. s. karapetyan @xmath109 , phys .
c 87 ( 2013 ) 054604 .
a. deppman , e. andrade - ii , v. guimaraes , g. s. karapetyan @xmath109 , phys .
c 88 ( 2013 ) 024608 .
d. a. de lima , j. b. martins and o. a. p. tavares , il nuovo cim .
a @xmath116 , 701 ( 1990 ) .
m. l. terranova and o. a. p. tavares .
physica scripta 49 ( 1994 ) 267 - 279 .
t. fukahori , o. iwamoto , and s. chiba , _ in proc . of the 7th intern
. conf . on nuclear criticality safety , icnc 2003 , tokai - mura , japan , 144 ( 2003)_. l. g. moretto , nucl
a @xmath117 , 211 ( 1975 ) .
s. b. duarte , o. a. p. tavares , f. guzman et al . ,
data and nucl .
data tables 80 ( 2002 ) 235 - 299 .
a. deppman , s. b. duarte , g. silva @xmath109 , j. phys .
@xmath118 , ( 2004 ) 1991 - 2002 .
t. kodama , s. b. duarte , k. c. chung @xmath109 , phys .
@xmath119 , 536 ( 1982 ) .
m. goncalves , s. depina , d. a. lima , @xmath109 , phys .
b @xmath120 , 1 ( 1997 ) . b. d. serot and j. d. walecka , adv .
@xmath121 , 1 ( 1986 ) .
a. deppman @xmath109 , j. phys .
g : nucl . part .
@xmath122 1991 ( 2004 ) .
s. de pina @xmath109 , phys .
b @xmath123 , 1 ( 1998 ) . i. gonzalez @xmath109 , j. phys .
g : nucl . part .
@xmath124 , 115105 ( 2011 ) .
a. deppman @xmath109 , phys .
@xmath125 , 182701 ( 2001 ) .
a. deppman @xmath109 , nucl .
instr . and meth .
b @xmath126 , 15 ( 2003 ) .
a. deppman @xmath109 , comp .
@xmath127 , 385 ( 2002 ) . n. bianchi @xmath109 , phys .
b @xmath128 , 219 ( 1993 ) . j. c. sanabria @xmath109 , phys .
c @xmath129 , 034604 ( 2000 ) . c. cetina @xmath109 , phys
c @xmath130 , 044622 ( 2002 ) .
a. deppman @xmath109 , phys .
c @xmath113 , 064607 ( 2006 ) .
i. dostrovsky , p. rabinowitz and r. bivins , phys .
111 ( 1958 ) 1659 - 1676 .
u.brosa , s. grossman , and a. muller , z. naturforschung @xmath131 , 1341 ( 1986 ) .
w. younes , j. a. becker , l. a. bernstein _
_ et al.__@xmath132_nuclear physics in the 21st century : international nuclear physics conference(inpc 2001)(aip , new york , 2001 ) _ [ aip conf .
@xmath133 , 673 ( 2001 ) ] .
h. kudo , m. maruyama , m. tanikawa @xmath109 , phys . rev .
c * 57 * , 178 ( 1998 ) .
m. c. duijvestijn , a. j. koning @xmath109 , phys .
c @xmath134 , 776 ( 1999 ) . s. a. pereira et al .
sci . and eng .
159 ( 2008 ) 159 .
e. andrade - ii @xmath109 , j. phys .
series 312 ( 2011 ) 082010 .
e. andrade - ii @xmath109 , j. phys .
g- nucl . and part .
38 ( 2011 ) 085104 .
e. andrade - ii @xmath109 , epj wb conf . 21
( 2012 ) 10001 .
a. deppman @xmath109 , phys .
c 66 ( 2002 ) 067601 .
e. andrade - ii @xmath109 , aip conf .
proc . 1139 ( 2009 ) 64 - 69 .
a. deppman @xmath109 , phys .
c 73 ( 2006 ) 064607 .
v. p. likhachev @xmath109 , phys .
c 65 ( 2002 ) 044611
. v. p. likhachev @xmath109 , nucl .
( 2003 ) 24 - 44 .
likhachev @xmath109 , phys .
c 68 ( 2003 ) 014615 .
i. gonzales @xmath109 , j. phys .
series 312 ( 2011 ) 022017 .
f. krmpotic @xmath109 , aip conf .
proc . 1491 ( 2012 ) 117 - 120 .
s. anefalos @xmath109 , braz .
34 ( 2004 ) 966 - 968 . s. anefalos @xmath109 , aip conf .
( 2005 ) 1299 - 1302 .
s. anefalos @xmath109 , nucl .
sci . and eng . 151 ( 2005 ) 82 - 87 .
s. t. mogelli @xmath109 , braz .
35 ( 2005 ) 894 - 897 .
a. deppman @xmath109 , sci . and tech .
( 2012 ) 480343 .
j. adam , k. katovsky , r. michel , a. balabekyan , _ proceding of aip international conference on nuclear data for science and tecnology may 24 , 2005 volume 769 , pp .
1043 - 1046_. j. b. cumming , ann .
@xmath135 , 261 ( 1963 ) .
r. b. firestone , in tables of isotopes , 8th ed . :
1998 update ( with cd rom ) , edited by s. y. frank chu ( cd - rom editor ) and c. m. baglin ( wiley interscience , new york , 1996 ) . h. baba , j. sanada , h. araki @xmath109 , nucl . instrum .
methods a @xmath136 , 301 ( 1998 ) . g. s. karapetyan , a. r. balabekyan , n. a. demekhina , and j. adam , phys . at .
@xmath137 , 911 ( 2009 ) .
a. r. balabekyan , g. s. karapetyan , n. a. demekhina , @xmath109 , phys . at .
@xmath113 , 1814 ( 2010 ) . | the investigation of intermediate - mass fragments ( imf ) production with the proton - induced reaction at 660 mev , on 238u and 237np target , is present .
the data were obtained with the lnr phasotron u-400 m cyclotron at joint institute for nuclear research ( jinr ) , dubna , russia .
a total of 93 isotopes , in the range @xmath0 , were unambiguously identified with high precision .
the fragment production cross sections were obtained by the means of the induced - activation method in an off - line analysis .
mass - yield distributions were derived from the data and compared with the results of the simulation code crisp for multimodal fission .
a discussion on the superasymmetric fragment production mechanism is also present . |
core - collapse supernova remnants ( snrs ) are more or less aspherical in their morphology ( lopez et al .
2011 ) . the asymmetries could be caused by external shaping from non - uniform ambient medium ( tenorio - tagle et al .
1985 ) , the dense slow winds of progenitor stars ( blondin et al . 1996 ) , the runaway progenitors ( meyer et al . 2015 ) , and the galactic magnetic field ( gaensler 1998 ; west et al . 2015 ) .
intrinsic asymmetries of the explosion can also impact on the morphologies of snrs , with increasing evidences provided by studying the distribution and physical states of the ejecta .
the historical snr cas a shows fast moving ejecta knots outside the main shell ( e.g. , fesen & gunderson 1996 ) and non - uniform distribution of heavy elements ( e.g. hwang et al .
2000 ; @xmath12ti recently reported by grefenstette et al .
high - velocity ejecta `` shrapnels '' have been discovered in the evolved snr vela ( aschenbach et al .
the accumulating observations of asymmetric snrs challenge the standard spherical pictures of sn explosion and snr evolution . in light of this ,
the environmental and spatially resolved study of asphercial snrs becomes more and more important .
kes 79 ( a.k.a . g33.6 + 0.1 ) is a galactic snr with a round western boundary and deformed eastern boundary in radio band ( frail & clifton 1989 ) . the radio morphology is characterized by multiple concentric shells or filaments ( velusamy et al .
an early rosat x - ray observation showed that most of the diffuse x - ray emission is from a bright inner region and some faint x - ray emission is extended to the outer region ( seward & velusamy 1995 ) .
the 30 ks _ chandra _ observation revealed rich structures , such as many filaments and a `` protrusion , '' and a constant temperature ( 0.7 kev ) across the snr ( sun et al .
2004 ; hereafter s04 ) .
the spectral results of the global snr were next supported with the spectral study using two epochs of _ xmm - newton _ observations ( giacani et al .
2009 ) . using _
xmm - newton _ observations spanning 2004 and 2009 , the spatially resolved studies provided further information on the hot gas , where two thermal components are required to explain the observed spectra ( auchettl et al .
2014 , hereafter a14 ) .
kes 79 hosts a central compact object ( cco ) psr j1852 + 0040 ( seward et al . 2003 ) , which was discovered as a 105 ms x - ray pulsar ( gotthelf et al .
2005 ) with a weak magnetic field ( halpern et al . 2007 ;
halpern & gotthelf 2010 ) . in the south , an 11.56 s low - b magnetar , 3xmm j185246.6 + 003317 , was found at a similar distance to kes 79 ( zhou et al . 2014 ; rea et al .
2014 ) .
kes 79 is classified as a thermal composite ( or mixed - morphology ) snr presenting a centrally filled morphology in x - rays and shell - like in the radio band ( rho & petre 1998 ) .
thermal composite snrs generally display good correlation with or molecular clouds ( mcs ; rho & petre 1998 ; zhang et al .
2015 ) and @xmath13-ray emission ( check the snr catalog in ferrand & safi - harb 2012 ) , and are believed to be the best targets to study hadronic cosmic rays .
green & dewdney ( 1992 ) performed @xmath14co @xmath00 and hco@xmath15 @xmath00 observations toward kes 79 and found a morphological coincidence of the snr with the mcs in the east and southeast at the local standard of rest ( lsr ) velocity ( @xmath16 ) of 105 @xmath4 . at this velocity ,
broad 1667 mhz oh absorption ( green 1989 ) and absorption were detected ( frail & clifton 1989 ) .
green et al . ( 1997 ) reported the detection of 1720 mhz oh line emission toward kes 79 , although no oh maser was found .
hence , kes 79 is very likely interacting with mcs at around 105 @xmath4 , but direct physical evidence is still lacking .
the lsr velocity corresponds to a kinetic distance of 7.1 kpc according to the rotation curve of the galaxy ( frail & clifton 1989 ; case & bhattacharya 1998 ) .
gev @xmath13-ray
emission is also detected with _
fermi _ east of kes 79 at a significance of @xmath17 , where bright co emission is present ( a14 ) .
in order to study the origin of asymmetries and thermal composite morphology of kes 79 and to find physical evidence for the snr mc interaction , we performed new multi - transition co observations ( see also chen et al .
2014 ) and revisited the _ xmm - newton _ data .
this paper is organized as follows . in section [ s :
obs ] , we describe the multi - wavelength observations and data reduction . our results are shown in section [ s : result ] and the discussion is presented in section [ s : discussion ] .
a summary is given in section [ s : summary ] .
the observations of @xmath14co @xmath00 ( at 115.271 ghz ) and @xmath18co @xmath00 ( at 110.201 ghz ) were taken during 2011 may with the 13.7 m millimeter - wavelength telescope of the purple mountain observatory at delingha ( pmod ) , china .
the new superconducting spectroscopic array receiver with @xmath19 beams was used as the front end ( shan et al .
2012 ) . the @xmath14co @xmath00 and @xmath18co @xmath00 lines were configured at at the upper and lower sidebands , respectively , and observed simultaneously .
the fast fourier transform spectrometers with 1 ghz bandwidth and 16,384 channels were used as the back ends , providing a velocity resolution of 0.16 @xmath4 for @xmath14co @xmath00 and 0.17 @xmath4 for @xmath18co @xmath00 .
the full width at half maximum ( fwhm ) of the main beam was @xmath20 , the main beam efficiency was 0.48 in the observation epoch , and the pointing accuracy was better than @xmath21 .
we mapped a @xmath22 region centered at ( @xmath23 , @xmath24 , j2000 ) in on - the - fly observing mode .
the data were then converted to @xmath14co @xmath00 and @xmath18co @xmath00 main - beam temperature ( @xmath25 ) cubes with a grid spacing of @xmath26 and a velocity resolution of 0.5 @xmath4 .
the corresponding mean rms of the spectra in each pixel is 0.41 k and 0.26 k per channel , respectively .
the @xmath14co @xmath11 ( at 230.538 ghz ) observation of kes 79 was carried out in 2010 january with klner observatory for submillimeter astronomy ( kosma ; now renamed ccosma ) .
we mapped an @xmath27 area centered at ( @xmath28 , @xmath29 , j2000 ) with a grid spacing of @xmath30 using a superconductor insulator superconductor receiver and a medium - resolution acousto - optical spectrometer .
the fwhm of the main beam was @xmath31 and the velocity resolution was @xmath32 @xmath4 at 230 ghz .
the main beam efficiency was @xmath33 .
we retrieve the @xmath14co @xmath22 ( at 345.796 ghz ) data of kes 79 from the co high resolution survey ( dempsey et al .
2013 ) observed with heterodyne array receiver program ( harp ) on board the james clerk maxwell telescope ( jcmt ) .
the first release of the survey covers @xmath34 between @xmath35 , and @xmath36 between @xmath37 .
the fwhm of the main beam at the frequency was @xmath38 , and the main beam efficiency was 0.63 .
we use the rebinned data , which has a pixel size of 6@xmath39 and a velocity resolution 1 @xmath4 . the @xmath00 and @xmath11 data were reduced with the gildas / class package . for the data analysis and visualization , we made use of the miriad ( sault et al .
1995 ) , karma ( gooch 1996 ) and idl software packages .
there are 21 archival _ xmm - newton _ observations toward kes 79 , which were taken in 2004 ( pi : f. seward ) , 2006 - 2007 ( pi : e. gotthelf ) , and 2008 - 2009 ( pi : j. halpern ) .
we excluded five observations ( obs .
i d : 550671401,550671501 , 55061601 , 55061701 , and 55062001 ) since they were performed with filter cal - closed ( affected by radiation ) and observation 550671101 suffered from flares for most of the observation time .
we use epic - mos data which covered the whole snr in full frame mode with medium filter , while epic - pn data are not used here since the pn detector only covered the inner part the remnant in small window mode .
the epic - mos data were reduced using the science analysis system software ( sas ) .
we used the _ xmm - newton _ extended source analysis software , xmm - esas , in the sas package to filter out the time intervals with soft proton contaminations , detect point sources in the field of view , and create particle background images . after removing time intervals with heavy proton flarings , the total effective exposures are 376 ks and 380 ks for mos1 and mos2 , respectively .
detailed information , including the observation i d , observing date , and net exposure of the observations , is summarized in table [ t : xmmobs ] .
we retrieve _ spitzer _
@xmath40 post - basic calibrated data from the _ spitzer _ archive .
the mid - ir observation was performed as a 24 micron survey of the inner galactic disk program ( pid : 20597 ; pi : s. carey ) .
lcccccc 0204970201 & 2004 oct 17 & 18:52:43.87 & 00:39:07.0 & 30.2 & 29.8 + 0204970301 & 2004 oct 23 & 18:52:44.07 & 00:39:08.5 & 27.6 & 27.9 + 0400390201 & 2006 oct 08 & 18:52:43.03 & 00:39:00.0 & 25.1 & 24.9 + 0400390301 & 2007 mar 20 & 18:52:34.77 & 00:41:47.3 & 30.3 & 30.7 + 0550670201 & 2008 sep 19 & 18:52:42.42 & 00:38:50.8 & 20.4 & 20.4 + 0550670301 & 2008 sep 21 & 18:52:42.38 & 00:38:50.1 & 29.3 & 29.2 + 0550670401 & 2008 sep 23 & 18:52:42.46 & 00:38:53.6 & 34.7 & 34.2 + 0550670501 & 2008 sep 29 & 18:52:42.65 & 00:38:53.5 & 31.8 & 32.4 + 0550670601 & 2008 oct 10 & 18:52:43.03 & 00:39:04.2 & 30.8 & 30.5 + 0550670901 & 2009 mar 17 & 18:52:34.94 & 00:41:45.8 & 22.0 & 23.3 + 0550671001 & 2009 mar 16 & 18:52:34.88 & 00:41:48.1 & 19.1 & 20.0 + 0550671201 & 2009 mar 23 & 18:52:34.62 & 00:41:44.5 & 15.2 & 15.7 + 0550671301 & 2009 apr 04 & 18:52:34.11 & 00:41:40.4 & 19.1 & 20.0 + 0550671801 & 2009 apr 22 & 18:52:33.49 & 00:41:29.6 & 27.1 & 27.1 + 0550671901 & 2009 apr 10 & 18:52:33.84 & 00:41:35.7 & 13.0 & 14.4 + 0550671101@xmath41 & 2009 mar 25 & 18:52:34.39 & 00:41:41.0 & @xmath42 & @xmath43 + 0550671401@xmath44 & 2009 mar 17 & 18:52:34.95 & 00:41:45.7 & 0 & 0 + 0550671501@xmath44 & 2009 mar 25 & 18:52:34.36 & 00:41:42.6 & 0 & 0 + 0550671601@xmath44 & 2009 mar 23 & 18:52:34.61 & 00:41:44.3 & 0 & 0 + 0550671701@xmath44 & 2009 apr 04 & 18:52:34.11 & 00:41:40.2 & 0 & 0 + 0550672001@xmath44 & 2009 apr 10 & 18:52:33.85 & 00:41:35.6 & 0 & 0 [ t : xmmobs ] kes 79 has been suggested to expand into the denser ism in the east , since the eastern radio shells are deformed and has a prominent protrusion in the northeast ( velusamy et al .
previous co @xmath00 observations showed that the snr morphology is correlated with the eastern molecular gas at @xmath46 ( green & dewdney 1992 ) .
in this paper , we present the direct kinematic evidence with multiple transitions of co. we first search for morphological coincidences between the co emission and the snr shells , especially for the deformed eastern and northeastern shells , which might be shaped by dense ambient medium .
figure [ f:12co10_grid ] displays the channel maps of pmod @xmath14co @xmath00 in the velocity ranges of @xmath47124 @xmath4 with a velocity step of 6 @xmath4 .
each channel shows the velocity - integrated @xmath48 in a @xmath49 mapping area and the velocity labeled in the channel image indicates the central velocity . in the upper - left panel ,
we define several radio shells ( also referring to the definition in s04 ) , which will be used for multi - wavelength comparison in the following study .
located on the inner galactic plane , the sky region of kes 79 is rich in molecular gas in its line of sight .
there is bright @xmath14co emission in the eastern and northeastern shells in the velocity range @xmath50112 @xmath4 , as noted by green & dewdney ( 1992 ) , while the co emission fades out to the northwest .
we also notice that a molecular shell at @xmath51 @xmath4spatially matches the snr s middle and outer radio shells in the west , and the inner radio shell in the east . at @xmath52 and @xmath53 ,
the mcs are spatially close to the eastern double shell of the remnant as well ; however , the gas does not match the deformed northeastern shell .
the co emission at other velocities is either outside the snr or revealing complicated morphology inside the remnant . the open shell detected in the north and west at @xmath54 ( giacani et al .
2009 ) , however , seems to not correspond to shell structure of co emission . from both the morphology and the spectral lines ,
we can not find a signal for mc
snr interaction below 95 @xmath4 .
the jcmt high - resolution observations in the @xmath14co @xmath22 line ( with beamwidth of @xmath38 and pixel size of @xmath55 ) provide more details on the gas morphology and can be compared with the vla 20 cm continuum data with a synthesis beam size of @xmath56 . since the @xmath14co @xmath22 transition has a relatively high critical density of @xmath57 and a high upper state temperature of 33 k ( schier et al . 2005 ) , it can trace dense and warm gas better than @xmath14co @xmath00 , and thus help to reveal shocked gas .
figure [ f:12co32_grid ] shows harp @xmath14co @xmath22 channel maps in the velocity interval 88118 @xmath4 , overlaid with the 20 cm radio continuum contours .
the velocity step of the channels is 2 @xmath4 and the maps cover a @xmath49 area .
the brightest @xmath14co @xmath22 emission is shown near the northeast radio protrusion , and along the eastern double shell . notably , at @xmath46 , two thin , clumpy @xmath14co @xmath22 filaments mainly align with the boundary of the eastern double shell , indicating a relation between the dense and warm molecular filaments and the eastern double - shell .
we also found that the @xmath14co @xmath22 emission shifts from the northeast to the west with increasing velocity ( up to 115 @xmath4 ) .
co @xmath22 spectrum of the clump at around the radio protrusion ( denoted by the arrow in fig [ f:12co32_grid ] ) fit with a two - gaussian model ( solid line ) .
the narrow component with @xmath58 and the broad component with @xmath59 are shown with dotted line and dashed line , respectively .
, scaledwidth=47.0% ] broad molecular line broadening or asymmetric profile is an evidence of shock perturbation of molecular gas ( e.g. , denoyer 1979 ; chen et al .
a broad @xmath14co @xmath22 line is found in the bright clump at the southern boundary of the radio protrusion ( @xmath60 , @xmath61 , j2000 ; with size of @xmath62 and denoted by an arrow in figure [ f:12co32_grid ] ) .
as shown in figure [ f:2gaussfit ] , the spectrum consists of a sharp peak at 103 @xmath4 and a broad wing spanning a velocity range over 20 @xmath4 , which can not be explained with a single excitation component .
we find that the profile can be well described by two gaussian components : a narrow line with line width ( fwhm ) @xmath63
@xmath4 at @xmath64 @xmath4 ; and a broad line with @xmath65 @xmath4 at @xmath66 @xmath4 .
the narrow line corresponds to the dense , quiescent gas , while the broad component is probably emitted by the shocked gas .
the broad @xmath14co @xmath22 line detected at the protrusion position , together with the morphological agreement between the two molecular filaments in the east and the radio double shell , supports that the systemic velocity of kes 79 is at @xmath3 @xmath4 .
we subsequently search for shocked molecular gas over the entire snr .
figure [ f : linegrid ] shows a grid of line profiles of @xmath14co @xmath00 , @xmath18co @xmath00 , @xmath14co @xmath11 in the 75125 @xmath4 velocity range across the whole snr .
the radio image is displayed as a background . due to a low abundance of @xmath18co in the interstellar molecular gas ( [ @xmath18co]/[h@xmath67 \sim 2{\times 10^{-6}}$ ] ; dickman 1978 ) , the @xmath18co emission is normally optically thin and traces the quiescent gas
. the molecular gas shocked by the snr may show asymmetric or broad @xmath14co lines deviating from the line profiles of @xmath18co . the @xmath18co emission associated with the snr peaks at @xmath68 across the remnant . in the western hemisphere of the remnant , the @xmath14co
lines shift to higher velocity ( red - shift ) relative to the @xmath18co line peaks .
the @xmath14co line profiles in the northeast of the remnant ( in the red rectangle in figure [ f : linegrid ] ) show strongly asymmetric features and appear to extend to 120 @xmath4 figure [ f : moment1 ] reveals the distribution intensity weighted mean velocity ( first moment ; velocity field ) of @xmath18co and @xmath14co lines .
the @xmath14co emission , especially @xmath14co @xmath11 , has a higher mean velocity than the systemic velocity given by the @xmath18co emission in the west .
notably , the high-@xmath16 @xmath14co emission delineates the outer , middle , and the eastern inner radio shells , suggesting this high-@xmath16 emission is excited by the shock at the snr shells .
we merged the data of the 15 epic - mos observations and show a tri - color image of kes 79 in figure [ f : rgb](a ) , in which the x - ray emission in the soft ( 0.31.2 kev ) , medium ( 1.22.0 kev ) , and hard ( 2.07.0 kev ) bands is colored red , green , and blue , respectively .
the energy bands are chosen to obtain similar counts in these images .
each of the three band images was exposure - corrected and adaptively smoothed with a gaussian kernel to achieve a minimum significance of 3 and a maximum significance of 4 using the `` csmooth '' command in ciao .
the x - ray emission of kes 79 is highly filamentary and clumpy . in the east ,
two striking x - ray filaments are distorted and composed of with a string of small x - ray clumps in scale of less than @xmath69 ( @xmath70 pc at a distance of 7.1 kpc ) .
the v - shaped structure in the southern part appears to be the softest x - ray emitter ( due to its relatively red color ) . in the snr interior , two jet - like filaments
stretch from the southeast to the northwest and across the cco . near the northwestern end
, there is a patch of hard x - ray emission , which is discussed in section [ s : fragment ] .
the two compact sources , the magnetar 3xmm j185246.6 + 003317 and the cco psr j1852 + 0040 , are located in the south and the center of the snr , respectively . the _ spitzer _ @xmath71 ir image reveals many thin filaments inside the remnant which are well matched with the bright x - ray filaments ( see figure [ f : rgb]b ) . here
we use the unsharp mask ir image instead of the original emission image in order to highlight the sharp structures . in figure
[ f : rgb]d , we compare the distribution of the high-@xmath16 mcs ( 110120 @xmath4 ; above the system velocity 105 @xmath4 ; green ) with that of the snr s x - ray ( blue ) and radio ( red ) emission . a molecular arc delineates the western outer radio shell and confines the bright x - ray emission in the snr interior .
meanwhile , a northeast - south oriented molecular ridge also spatially matches the bright x - ray filaments and eastern inner radio emission .
a faint x - ray halo surrounding the filamentary structures appears to stretch to the outer radio boundary , as shown in the 0.37.0 kev x - ray image ( figure [ f : rgb](c ) ) .
the x - ray image was exposure - corrected , smoothed with a gaussian kernel of @xmath72 , and overlaid with contours of the 20 cm radio continuum emission . here
we display the x - ray image with a histogram equalization scale to highlight the faint emission .
unlike the filaments , the faint x - ray halo has no evident ir counterpart .
the properties of the x - ray halo and the filaments will be compared in section [ s : xspec ] .
we investigate the distribution of metal lines firstly by using the ew images of mg he@xmath73 at @xmath74 kev , si he@xmath73 at @xmath75 kev , s he@xmath73 at @xmath76 kev , and ar he@xmath73 at @xmath77 kev ( see figure [ f : ewmap ] ) .
the ew of an emission line is defined as @xmath78 , where @xmath79 and @xmath80 are the energy - integrated intensity of the line component and the intensity of the underlying continuum at the line center , respectively .
the ew values of metal lines depend on the abundances , and are also affected by the temperature and ionization states .
we define the energy bands of the lines , and the left and right continuum shoulders as shown in table [ t : ewmap ] , according to the global spectra of kes 79 ( see section [ s : result_global ] below ) .
the continuum emission of each line is estimated by interpolating the left and right shoulders emission .
the background of each image is subtracted before producing the ew map .
the quiescent particle background ( qpb ) is first subtracted from each image since it is apparently spatially variant at @xmath81 kev and @xmath82 kev , which would affect the ews of the mg and si lines .
the qpb images are created from the filter - wheel closed ( fwc ) data with the xmm - esas software .
we then estimate the local background level from a region near the remnant and subtract it from the qpb - subtracted image of each band .
we adaptively bin the background - subtracted ew images with signal - to - noise ratio ( s / n ) @xmath83 by applying the weighted voronoi tessellations binning algorithm ( diehl & statler 2006 ) , which is a generalization of cappellari & copin s ( 2003 ) voronoi binning algorithm .
the bins with s / n@xmath84 and low continuum fluxes were set to zero .
the ews of the mg line are nearly uniform across the remnant . here
the mg he@xmath73 line s soft shoulder is strongly affected by the variation of the interstellar absorption , which could also bring uncertainties to the estimation of the continuum .
asymmetric distribution is clearly revealed in both ew images of si and s. the largest ew values are found in a patch in the northwest .
generally , the ews at the x - ray filaments are larger than those in the faint halo regions .
a similar trend can also be found in the ew map of ar .
the asymmetric ew distribution of the metal lines suggests that the hot gas properties ( abundances , temperature , or ionization states ) are not uniform in the snr .
the halo and filamentary regions may not be in the same physical states , as confirmed below according to the spectral analysis .
the @xmath85 375 ks _ xmm - newton _ observations allow us to quantitatively investigate the hot gas in small - scale regions in kes 79 .
we define source ( on - snr ) and background ( off - snr ) regions for spectral analysis after removing the central bright x - ray source psr j1852 + 0040 by excluding a circular region with a radius of @xmath86 centered at ( @xmath87 , @xmath88 , j2000 ) ( see figure [ f : reg ] ) .
we select a large region with a radius of @xmath89 that covers the whole snr , 14 small regions for the bright filamentary structures ( `` f114 '' ) and 5 smaller regions ( `` c '' , `` me '' , `` mw '' , `` ow '' , and `` on '' ) for the faint halo gas .
the background region `` bg '' is selected from the nearby sky with the same galactic latitude as that of the remnant to minimize contamination by the galactic ridge emission .
p2.cmccc mg he@xmath73 & 1.161.22 & 1.251.42 & 1.451.51 + si he@xmath73 & 1.451.51 & 1.751.95 & 1.962.08 + s he@xmath73 & 1.962.08 & 2.302.60 & 2.632.78 + ar he@xmath73 & 2.632.78 & 2.983.22 & 3.283.70 [ t : ewmap ] we apply a double - subtraction method that takes into account vignetting effects and the spatially variable instrumental background as follows : ( 1 ) subtract respective instrumental background contribution from the raw on- and off - snr spectra by using the fwc data of mos1 ; ( 2 ) construct a background model by fitting the off - snr spectra ( instrumental background - subtracted ) , which is subsequently normalized by area and added to the source model to describe the on - snr spectra .
fwc data are selected in the epochs close to those of the source observations in order to reduce variation of instrumental background . using the task @xmath90 in heasoft , we merged the spectra taken within two months to increase the statistical quality for the hard x - ray band
. hence , the spectra of the two 2004 observations , five 2008 observations , and six 2009 observations are merged respectively .
finally , the 10 source spectra ( 5 mos1 plus 5 mos2 spectra taken from the 2004 , 2006 , 2007 , 2008 , and 2009 observations ) for each on - snr region are then instrumental background - subtracted and binned to reach an s / n of @xmath91 in the following joint - fit .
we only use the six off - snr spectra ( region `` bg '' ; 3 mos1 plus 3 mos2 ) taken in 2004 , 2006 and 2008 to construct the background model , since the region `` bg '' was at the unusable ccd of mos1 during the observations in 2007 and 2009 .
xspec ( version 12.9 ) is used for spectral fitting .
the spectra in the off - snr region `` bg '' is phenomenologically described with an absorbed @xmath92+@xmath93-@xmath94 plus an unabsorbed @xmath95 + @xmath96 model ( @xmath97 ; see figure [ f : snrspec ] ) .
the on - snr sky background is determined by normalizing the model according to the region sizes .
+ @xmath98 model ( dashed lines ) plus the background model ( dotted lines ; phenomenologically describing the `` bg '' spectra ; note that the area response is different at the the on - snr and off - snr regions ) .
the spectral fitting results are summarized in table [ t : nei_vnei ] .
, scaledwidth=50.0% ] figure [ f : snrspec ] shows the spectra extracted from source region `` snr '' ( upper ) and background region `` bg '' ( bottom ; normalized by area ) .
the global spectra of the snr in 0.58 kev reveals distinct he - like lines of mg ( @xmath74 kev ) , si ( @xmath75 kev and @xmath99 kev ) , s ( @xmath76 kev ) , and ar ( @xmath77 kev ) , and several faint h - like lines of ne at 0.9 kev and s at 2.9 kev .
we first applied an absorbed non - equilibrium ionization ( nei ) model ( plus the background model with fixed values ) to jointly fit the 10 source spectra in 0.5 - 8.0 kev .
for the foreground absorption , we used the @xmath100 model with the anders & grevesse ( 1989 ) abundances and photoelectric cross - section from balucinska - church & mccammon ( 1992 ) .
the nei model @xmath98 ( nei version 3.0 in xspec ) is firstly applied , with the abundances of ne , mg , si , s , ar , and fe ( the abundance of ni is tied to fe ) in the models allowed to vary in the spectral fit . here
@xmath98 describes an nei plasma with single and uniform ionization parameters
. however , the single nei model fails to describe the spectra in the hard x - ray band ( @xmath101 ) as also pointed out by a14 .
the elevated tail of the observed spectra compared to the model above 3 kev is also seen in single thermal model spectral fittings in previous studies ( e.g. s04 , and giacani et al .
2009 ) , implying that the global spectrum of kes 79 should contain more than one component . since some thermal composite snrs show recombining plasma , we tried an absorbed recombining plasma model ( @xmath102 in xspec ) to fit the spectra .
however , the spectral fit ( @xmath103 ) gives an initial plasma temperature ( 0.12 kev ) much smaller than the current plasma temperature ( 0.75 kev ) , suggesting that the gas is under - ionized rather than over - ionized .
an absorbed @xmath92+@xmath98 model is subsequently used , which substantially improves the spectral fit ( @xmath104 ; see table [ t : nei_vnei ] for the spectral fit results and figure [ f : snrspec ] for the spectra ) .
the abundance of the cool component is set to solar since variable abundance does not apparently improve the spectral fit .
the model is still not good enough to reproduce the spectra , especially in the 1 - 3 kev band , which suggests that the physical properties of the x - ray emitting gas are not uniform across the remnant .
however , the two - temperature model , to some extent , provides us with general properties of the hot plasma .
lccccccccccccccc snr & 1.92 /2654 & 1.707@xmath105 & 0.195@xmath106 & 6.4@xmath107 & 0.80@xmath108 & 8.1@xmath109 & 1.83@xmath110 & 1.56@xmath111 & 1.40@xmath112 & 1.72@xmath111 & 1.8@xmath109 & 0.93@xmath113 & 2.1 & 0.5 & 4.2 + f1 & 1.01 / 501 & 1.8@xmath109 & 0.21@xmath114 & 6.4 ( @xmath115 ) & 0.7@xmath116 & 6.6@xmath117 & 2.0@xmath118 & 1.8@xmath119 & 1.5@xmath120 & 1.8@xmath119 & 1 ( fixed ) & 1.5@xmath121 & 5.7 & 1.7 & 1.0 + f2 & 1.06 / 697 & 1.77@xmath122 & 0.19@xmath123 & 6.7 ( @xmath124 ) & 0.73@xmath125 & 7.7@xmath126 & 2.5@xmath127 & 1.6@xmath120 & 1.6@xmath128 & 2.1@xmath120 & 1 ( fixed ) & 1.2@xmath129 & 6.0 & 1.6 & 1.3 + f3 & 1.19 /1249 & 1.84@xmath130 & 0.20@xmath108 & 6.2@xmath131 & 0.75@xmath113 & 7.6@xmath132 & 2.2@xmath111 & 1.6@xmath112 & 1.6@xmath116 & 1.9@xmath133 & 2.8@xmath134 & 1.1@xmath128 & 3.8 & 1.0 & 2.0 + f4 & 1.28 /1622 & 1.63@xmath135 & 0.197@xmath136 & 5.4@xmath137 & 0.91@xmath138 & 6.4@xmath139 & 2.5@xmath133 & 1.8@xmath109 & 1.48@xmath140 & 1.82@xmath141 & 1.5@xmath120 & 0.94@xmath142 & 8.1 & 1.8 & 1.0 + f5 & 0.97 / 614 & 1.6@xmath109 & 0.19@xmath108 & 7.5 ( @xmath143 ) & 1.4@xmath128 & 2.3@xmath129 & 2.8@xmath144 & 1.9@xmath129 & 1.7@xmath145 & 1.8@xmath145 & 1 ( fixed ) & 1.1@xmath146 & 6.2 & 0.8 & 0.8 + f6 & 1.18 /1093 & 1.58@xmath130 & 0.200@xmath147 & 7.2@xmath148 & 0.82@xmath149 & 9.6@xmath150 & 2.6@xmath151 & 1.9@xmath133 & 1.7@xmath152 & 2.0@xmath133 & 1.6@xmath151 & 1.0@xmath133 & 13.0 & 3.2 & 0.8 + f7 & 1.14 /1075 & 1.68@xmath153 & 0.20@xmath108 & 13.4 ( @xmath154 ) & 0.91@xmath155 & 7.2@xmath156 & 2.3@xmath146 & 1.8@xmath133 & 1.5@xmath133 & 1.8@xmath133 & 1.1@xmath151 & 0.9@xmath133 & 13.2 & 2.8 & 0.7 + f8 & 1.20 /1060 & 1.81@xmath157 & 0.21@xmath158 & & 0.91@xmath111 & 6.8@xmath159 & 2.1@xmath145 & 2.0@xmath133 & 1.6@xmath133 & 1.9@xmath133 & 1.3@xmath146 & 1.1@xmath128 & 8.8 & 2.2 & 0.9 + f9 & 1.07 / 620 & 1.9@xmath109 & 0.21@xmath111 & 7.0 ( @xmath160 ) & 0.9@xmath109 & 6.7@xmath161 & 2.0@xmath162 & 2.2@xmath151 & 1.7@xmath145 & 1.9@xmath129 & 2.6@xmath163 & 1.5@xmath164 & 7.6 & 1.8 & 1.0 + f10 & 1.17 /1064 & 1.90@xmath111 & 0.22@xmath108 & & 1.18@xmath165 & 2.4@xmath133 & 2.8@xmath146 & 2.8@xmath145 & 3.1@xmath120 & 3.4@xmath107 & 3.6@xmath132 & 1.0@xmath120 & 6.6 & 1.2 & 0.5 + f11 & 1.03 / 629 & 2.0@xmath109 & 0.19@xmath108 & 5.9@xmath166 & 1.6@xmath167 & 3.2@xmath168 & 4.4@xmath169 & 3.6@xmath170 & 5.2@xmath171 & 5.2@xmath172 & 4.2@xmath173 & 0.6@xmath174 & 7.9 & 1.0 & 0.9 + f12 & 1.11 /1066 & 1.82@xmath175 & 0.20@xmath176 & 7.5@xmath177 & 0.74@xmath155 & 9.9@xmath178 & 1.8@xmath107 & 1.7@xmath133 & 1.4@xmath133 & 1.8@xmath133 & 1.2@xmath179 & 1.1@xmath128 & 6.0 & 1.7 & 1.6 + f13 & 1.07 / 998 & 1.66@xmath113 & 0.18@xmath108 & & 0.89@xmath175 & 10.0@xmath180 & 2.1@xmath120 & 1.9@xmath133 & 1.5@xmath116 & 1.6@xmath133 & 1.7@xmath151 & 1.0@xmath128 & 12.4 & 2.5 & 1.0 + f14 & 1.10 / 924 & 1.83@xmath157 & 0.22@xmath108 & & 0.88@xmath141 & 7.2@xmath181 & 2.3@xmath151 & 1.9@xmath128 & 1.6@xmath152 & 1.8@xmath128 & 1.2@xmath179 & 0.7@xmath120 & 9.1 & 2.2 & 0.9 + c & 0.98 / 723 & 1.9@xmath109 & 0.19@xmath108 & 4.8@xmath182 & 0.9@xmath109 & 7.8@xmath183 & 3.0@xmath184 & 1.5@xmath120 & 1.3@xmath128 & 1.6@xmath128 & 0.9@xmath168 & 0.5@xmath185 & 2.4 & 0.5 & 4.1 + me & 1.15 / 351 & 1.62@xmath186 & 0.17@xmath130 & & 0.7@xmath187 & 7.2@xmath188 & 0.9@xmath189 & 1.2@xmath190 & 1.1@xmath191 & 1.3@xmath129 & 1 ( fixed ) & 1.1@xmath119 & 1.4 & 0.3 & 6.3 + mw & 1.15 / 836 & 1.66@xmath192 & 0.20@xmath176 & 7.8 ( @xmath193 ) & 1.0@xmath116 & 5.4@xmath194 & 2.3@xmath195 & 1.8@xmath145 & 1.5@xmath128 & 1.7@xmath145 & 1.6@xmath132 & 1.1@xmath107 & 2.1 & 0.4 & 3.5 + ow & 1.07 / 804 & 1.72@xmath196 & 0.18@xmath176 & 2.1@xmath197 & 0.68@xmath198 & 12.4@xmath199 & 1.3@xmath200 & 1.4@xmath191 & 1.2@xmath191 & 1.5@xmath190 & 1 ( fixed ) & 1.7@xmath164 & 1.2 & 0.3 & 10.9 + on & 1.15 / 406 & 1.7@xmath109 & 0.19@xmath201 & 4.5 ( @xmath202 ) & 0.7@xmath116 & 7.6@xmath203 & 1.9@xmath204 & 1.8@xmath205 & 1.5@xmath206 & 1.7@xmath207 & 1 ( fixed ) & 1.3@xmath162 & 1.8 & 0.5 & 4.0 [ t : nei_vnei ] according to the @xmath92+@xmath98 model , the x - rays from the remnant suffers an interstellar absorption @xmath208 and the x - ray - emitting plasma can be roughly described as under - ionized two - temperature gas .
the cool phase of the gas has a temperature @xmath209 kev , an ionization timescale @xmath210 , and solar abundances ( so fixed to 1 ) , while the hot phase has a temperature @xmath211 kev , an ionization timescale @xmath212 and enriched metal abundances ( [ ne]/[ne]@xmath213 , [ mg]/[mg]@xmath214 , [ s]/[s]@xmath215 , and [ ar]/[ar]@xmath216 ) . the column density ( inferred from the @xmath100 model ) is larger than that obtained by s04 ( @xmath217 , @xmath218 model ) and giacani et al .
( 2009 ; @xmath219 , @xmath218 model ) , but smaller than that reported by a14 ( @xmath220 ; @xmath100 model ) .
some other spectral fit results , such as @xmath221 and abundances , are also different from those obtained by a14 .
we believe that the background selection is the main reason for the discrepancy between our work and a14 .
in particular , some of the background regions selected by a14 are well inside the snr where halo x - ray emission is present ( see figure [ f : rgb]c in this paper in comparison to figure 3 in a14 ) .
we assume that the two - temperature gas fills the whole volume ( @xmath222 ) and is in pressure balance ( @xmath223 ) , where @xmath224 and @xmath225 are the filling factor and hydrogen density ( @xmath226 ) , respectively .
the subscripts `` c '' and `` h '' denote the parameters for the cool and hot phases , respectively . the @xmath92+@xmath98 fit results ( see table [ t : nei_vnei ] ) give an emission measure @xmath227 for the cool phase gas , and @xmath228 for the hot phase , where @xmath229 is the volume of the sphere and @xmath230 is the distance scaled to 7.1 kpc .
the cool component is found to fill 62% of the volume , with a larger density @xmath231 , while the hot component has a density @xmath232 .
then we can obtain the ionization age for the two components @xmath233 kyr and @xmath234 kyr .
the larger ionization age of the cool phase is possibly due to a variable @xmath235 ( e.g. , decreasing with time ) . the x - ray image of kes 79 ( see figure [ f : rgb]a ) has revealed a non - uniform nature of the hot gas
. the ew maps of metal lines ( figure [ f : ewmap ] ) also indicate a nonuniform distribution of the ejecta .
the gas properties are thus expected to be spatially variant , and the global spectra may not be well described with a simple one - component or two - component model .
we thus compare the spectral analysis results of the 19 defined small regions to investigate the variation of gas properties across the snr .
a two - component thermal model is needed to describe the spectra well for the 19 small - scale regions .
adding an extra model to a single thermal model is also feasible according to the _ f_-test null hypothesis probabilities .
we first use @xmath92(soft)+ @xmath98(hard ) to fit the spectra , and then replace @xmath92 with @xmath236 ( apec thermal model describing the plasma in collisional ionization equilibrium ; atomdb version 3.0.2 ) for the cool phase if its ionization timescale reaches higher than @xmath237 . here we allow the ne , mg , si , s , ar , and fe abundances of the hot component to vary as long as they can be well constrained . the abundance of the cool component is set to solar since variable abundance does not apparently improve the spectral fit .
the best - fit results for the @xmath92/@xmath236+@xmath98 model are tabulated in table [ t : nei_vnei]./@xmath236+@xmath238 model for a comparison , where @xmath239 is a plane - parallel shock plasma model with a range of ionization parameters and solar abundances , while @xmath238 has variable abundances .
the two models result in almost identical temperature and abundance values , with larger @xmath240 and @xmath241 obtained with the latter model .
the larger @xmath241 value is interpretable given that in @xmath238 it stands for the upper limit of the ionization scale while in @xmath98a single ionization scale is adopted . in this paper , we use the fit results of the @xmath92/@xmath236+@xmath98 model because it is based on a simpler uniform assumption and generally provides a good characterization of the gas s general properties . ] the interstellar column density @xmath240 of the x - ray emission varies from @xmath242 in the south to @xmath243 in the northwest , consistent with the fact that the southern portion emits softer x - rays ( as shown in figure [ f : rgb]a ) . generally , both the bright filamentary ( except regions `` f5 '' , `` f10 '' and `` f11 '' ) and faint halo gas hold spectral properties similar to the global snr gas , including the temperature of the cool phase ( @xmath244 kev ) , and the temperature ( @xmath2451.0 kev ) , ionization timescale ( @xmath246@xmath247 ) , and metal abundances of the hot phase .
nevertheless , there is a trend that the gas near the snr edge has lower @xmath248 ( @xmath249 kev ) than in the snr center ( @xmath250 kev ) . to estimate the density of the plasma , we assume several three - dimensional shapes for the two - dimensional regions : ( 1 ) oblate spheroids for the elliptical regions `` f112 '' ; ( 2 ) oblate cylinders for the rectangular regions `` f1314 '' ; ( 3 ) sections of spherical layers for the shell - like halo regions `` mw '' and `` ow '' ; 4 ) for the halo regions `` c '' , `` me '' and `` on '' , the volumes are expressed as @xmath251 ; where @xmath252 is the area of the region and @xmath253 is the line of sight length along the snr sphere ( depending on the distance of the region to the snr center ) .
we assume that @xmath222 and @xmath254 , similar to the calculation for the global gas .
the calculated densities of the two - phase gas are summarized in table [ t : nei_vnei ] .
using the multi - transition co observations , we have provided kinematic and new morphological evidence to support the scenario that kes 79 is interacting with mcs at a systemic lsr velocity of @xmath255 .
the column density and mass of the surrounding molecular gas can be estimated by using the @xmath18co @xmath00 lines spanning the velocity range of 95115 @xmath4 . assuming that the rotational levels of the @xmath18co molecules are in local thermodynamic equilibrium
, we obtain the column density of @xmath18co molecules as follows ( kawamura et al .
1998 ) : @xmath256 where @xmath257 and @xmath258 are the excitation temperature and line width of @xmath18co @xmath00 , respectively .
when the @xmath18co lines are optically thin ( optical depth @xmath259 1 ) , the main beam brightness temperature of @xmath18co is @xmath260 . considering that @xmath261 and taking the @xmath262 to @xmath263 ratio as @xmath264 ( dickman 1978 )
, we obtain the column density of h@xmath265 : @xmath266 a typical mcs temperature of 10 k is adopted here for @xmath257 , which is also close to the peak main beam temperature of @xmath14co @xmath00 lines in the vicinity of kes 79 ( see figure [ f : linegrid ] ) . the @xmath18co emission integrated in the velocity range 95115 @xmath4 ( associated with the snr )
is then used to derive the @xmath262 and the distribution of @xmath262 near the remnant is revealed in figure [ f : nh2 ] . column density ( estimated from @xmath18co @xmath00 in the velocity range 95115 @xmath4 ) in the vicinity of kes 79 , vla 20 cm
radio contours are overlaid .
, scaledwidth=45.0% ] the mass of the environmental molecular gas in the @xmath267 ( or @xmath268 pc @xmath269 pc ) region ( field of view in figure [ f : nh2 ] ) , is obtained to be @xmath270\sim5.7{d_{7.1}}^2{\times 10^{4}}~{m_{\odot}}$ ] , where the mean molecular weight @xmath271 , @xmath272 is the mass of the hydrogen atom , and @xmath273 is the size of the gas ( @xmath268 pc ) . the size and mass values of the cloud is typical for giant mcs ( gmcs ; dame et al .
the mean molecular density of the gmc can be estimated as @xmath274 , where @xmath275 is the mean column density .
co @xmath22 image at 105 @xmath4 ; green : unsharp mask _ spitzer _ 24 @xmath9 image ; blue : _ xmm - newton _ mos 0.37.0 kev .
, scaledwidth=45.0% ] a good spatial correlation is present between the mid - ir filaments and the x - ray filaments ( see figure [ f : rgb ] ) .
previous ir studies toward middle - age snrs summarized three main origins of the emission at the _ spitzer _
24 @xmath9 band : ( 1 ) continuum emission of dust grains ( e.g. , andersen et al .
2011 ) ; ( 2 ) atomic lines such as [ ] at 25.988 @xmath9 and [ ] at 25.890 @xmath9 ( e.g. , reach & rho 2000 ) , and ( 3 ) molecular lines such as the vibrational rotational line of h@xmath265 at 28.2 @xmath9 ( e.g. , neufeld et al .
2007 ) . the ir emission at @xmath276 @xmath9 is usually dominated by the continuum of dust grains with typical temperature @xmath277100 k ( seok et al.2013 ) , while at the shorter wavelengths , the ir emission of snrs can be dominated by ionic / molecular lines ( e.g. , ic 443 , cesarsky et al .
1999 ; 3c 391 , reach et al .
the warm dust could be heated by either collisions or radiation . in the collisions case ,
the swept - up dust grains are collisionally heated by the electrons in the hot plasma ( dwek 1987 ) , which results in a good spatial correlation between the ir and x - ray emission .
the dust grains could also be heated by the uv photons from the postshock gas in the radiative shock ( hollenbach & mckee 1979 ) , which may be efficient for the snrs interacting with mcs . considering the mc
ir - x ray correlation revealed in kes 79 ( figure [ f : rgb ] ) , the two cases may both work here .
the snr has been considered to be in the sedov phase of evolution ( s04 ; see also discussion in section [ s : global ] ) and no optical emission has been detected to support the radiative shock scenario ( could possibly be due to heavy absorption ) . however , small - scale radiative shock in mcs can not be excluded . since the ir emission is filamentary and a strong ir - x - ray correlation
is widely seen in the whole remnant ( see figure [ f : rgb]b ) , it is more likely that the @xmath71 emission comes from dust heated by the hot plasma .
further ir spectroscopic observations are required to provide a firm conclusion .
the most prominent ir filaments emerge in the eastern section of the snr , where there is good correlation with the x - ray structures , including the enhancements at several clumps . in figure
[ f : twinfilaments ] , we show the two filaments in the eastern part of the snr revealed in @xmath14co @xmath22 , 24 @xmath9 , and the 0.37.0 kev x - ray bands .
the three bands correspond to three hierachical gas phases : cold dense gas , warm gas / dust , and hot tenuous plasma .
there is a trend wherein the colder gas is located east of the hotter gas . in particular , the thin ir filaments with a width of @xmath55 ( 0.2 pc at a distance of 7.1 kpc ) are just east of the x - ray filaments ( @xmath278 ) .
the trend is consistent with a scenario where the shock is interacting with multi - phase ism in the east , which then generates different types of shocks .
the shock is damped while it propagates into denser medium .
the fast shocks heat the moderate - density cloudlet ( @xmath279 ) to a temperature of @xmath280 kev and the low - density inter - cloud medium ( @xmath281@xmath282 ) to a temperature of @xmath245 kev ( see table [ t : nei_vnei ] ) .
the nondissociative shock transmitted into the dense molecular clumps ( @xmath283 for the clumps emitting @xmath14co @xmath22 emission ) has a low velocity ( @xmath284 , draine & mckee 1993 ) and generates the broad / asymmetric @xmath14co lines .
therefore , the multiple shocks in clumpy ism explains the bright filamentary radiation in multi - wavelengths .
the interior x - ray filaments probably arise from snr interaction with relatively dense gas ( @xmath285@xmath286 for the cold component ) , as compared to the low density of the halo gas ( @xmath287@xmath282 for the cold component ; see table [ t : nei_vnei ] ) .
kes 79 was classified as a mixed - morphology or thermal composite snr ( rho & petre 1998 ) , as its brightest x - ray emission arises from the interior and the radio emission is shell - like , different from most other thermal composite - type snrs , which have a smooth , centrally filled x - ray morphology , kes 79 has a highly structured x - ray morphology , and the interior x - ray and radio emission are correlated ( see figure [ f : rgb ] ) .
as s04 note , the inner x - ray shell and outer x - ray halo have clear - cut edges and may both come from shock surfaces . in section [ s : xfilaments ] , we have provided evidence that the x - ray and ir filaments , and the radio shells in the snr interior result from the interaction of the shock with a dense medium .
hence , projection effect is the most plausible origin of the multiple - shell structure mixed - morphology nature .
kes 79 may have a double - hemisphere morphology viewed essentially along the symmetric axis .
snrs breaking out from an mc into a low - density region can produce such a double - hemisphere morphology as predicted by hydrodynamic simulations ( e.g. , tenorio - tagle et al .
1985 ; cho et al . 2015 ) .
there are a number of snrs showing such double - hemisphere morphology due to shock breaking out from a dense medium , such as ic 443 ( e.g. , troja et al .
2006 ) , w28 ( e.g. , dubner et al . 2000 ) , 3c391 ( e.g. , reynolds & moffett 1993 , chen et al .
2004 ) , vro 42.05.01 ( e.g. , pineault et al .
1987 ) , g349.7@xmath288 ( e.g. , lazendic et al .
2005 ) , etc .
all of these snrs are associated with mcs ( see jiang et al .
2010 and references therein ) . here
we suggest that kes 79 is among the group of double - hemisphere snrs interacting with mcs .
the bright filaments correspond to the smaller hemisphere ( radius @xmath289 ) evolving into the denser ism , while the shock breaking out in a tenuous medium forms a larger hemisphere ( radius @xmath290 ) filled with hot halo gas .
the larger hemisphere or the blow - out part is the blueshifted one essentially heading toward us . this picture can be confirmed based on the distribution of the foreground absorption , @xmath240 , and the molecular line analysis .
as summarized in table [ t : nei_vnei ] , @xmath240 is lowest in the southeast ( `` f5 '' and `` me '' ) and southwest ( `` f6 '' and `` mw '' ) of the remnant . since the column density of mc at these regions can reach as high as @xmath291 ( see figure [ f : nh2 ] ) , heavy absorption would have been present if the mcs were in the foreground .
hence , we suggest a picture where most of the molecular gas ( except that in the east ) is behind the larger hemisphere ( halo gas ) and the x - ray filaments correspond to the snr dense - gas interacting regions . this scenario is also in agreement with the @xmath14co shell being red - shifted due to shock perturbation ( @xmath292 - 120 @xmath4 ; see the detailed description in section [ s : mol ] ) .
most small - scale regions have temperatures @xmath221 and @xmath248 similar to those of the global gas ( see table [ t : nei_vnei ] ) , while the filling factors @xmath293 could be varying across the remnant . hence , the overall spectral results can represent the average properties of the snr s x - ray - emitting plasma .
the masses for the cool and hot phases ( @xmath294 ) are @xmath295 and @xmath296 , respectively .
both masses are too high to be produced by the sn ejecta , and are most likely due to the shock - heated ism .
the enriched s and ar in the hot component indicates that part of the mass in the hot phase must be also contributed by the ejecta .
hence , the hot phase with lower density ( @xmath297 ) and elevated metal abundances is related to the emission in the inter - cloud medium , while the cool phase with higher density ( @xmath282 ) and solar abundances is suggested to come from the shocked denser cloud .
we use the parameters of the hot phase gas to investigate the snr evolution in the inter - cloud medium .
the bright x - ray emission and low ionization age ( @xmath298 kyr ) for the hot component suggest that kes 79 has not yet entered radiative phase ( at least globally ) .
hence , we adopt the sedov evolutionary phase ( sedov 1959 ) for kes 79 as done in previous studies of the remnant ( e.g. s04 and a14 ) . adopting @xmath299 kev
as the emission measure weighted temperature for the whole remnant , the post - shock temperature can be estimated as @xmath300 ( for ion electron equipartition ; borkowski et al .
2001 ) , which is @xmath301 kev .
the shock velocity can be derived as @xmath302^{1/2}=727\pm 5 $ ] @xmath4 , where @xmath303 the mean atomic weight for fully ionized plasma .
taking the curvature radius of the western shell @xmath304 pc ( @xmath305 ) , we estimate the explosion energy as @xmath306 erg , where @xmath307 and the preshock diffuse gas density @xmath308 .
the sedov age of the remnant is estimated as @xmath309 kyr .
the real age might deviate from 6.7 kyr to some extent because of the non - spherical evolution of kes 79 .
varying @xmath310 between @xmath311 ( @xmath312 ) and @xmath313 ( @xmath314 ) gives an age range between 4.46.7 kyr .
the lower limit 4.4 kyr is close to the ionization age of the hot component ( 4.2 kyr ) . the asymmetric metal distribution in kes 79 is supported by both the ew maps of si and s ( see figure [ f : ewmap ] ) and the spatially resolved spectral results ( see table [ t : nei_vnei ] ) .
notably , a bright patch is revealed in the si and s ew maps .
the northwestern regions `` f11 '' , `` f10 '' have a distinctly higher @xmath248 and abundances compared to other filamentary regions .
the regions overlap the bright patch in the ew maps of si and s. the metal - rich patch mainly corresponds to the spectral extraction region `` f11 '' at @xmath315 , @xmath316 ( j2000 ) , where the hot component gas has a high @xmath248 ( @xmath317 kev ) and the largest abundances of ne ( @xmath318 ) , mg ( @xmath319 ) , si ( @xmath320 ) , s ( @xmath321 ) and ar ( @xmath322 ) .
the patch is a protrusion of the filaments in the northwest ( see figure [ f : rgb]a ) and may be an ejecta fragment or a conglomeration of ejecta fragments at a high velocity .
using the hot component temperature ( @xmath323 kev ) , the velocity of the fragment is estimated to be @xmath324^{1/2}=1.0^{+0.2}_{-0.1}{\times 10^{3}}$]@xmath4 , which is 41% faster than the mean shock velocity @xmath325 @xmath4 derived in section [ s : global ] . in the opposite direction , the plasma in region `` f5 '' has a high temperature ( @xmath326 kev ; about twice the average value ) and a lower ionization timescale ( @xmath327 ) in the hot component , while the metal abundances are not significantly elevated .
the gas in region `` f5 '' thus could be fast ejecta clump well mixed with the ambient medium .
the non - uniform distribution of metal species in kes 79 and the presence of high - velocity ejecta fragement(s ) reflect the intrinsic asymmetries of the sn explosion .
, scaledwidth=50.0% ] kes 79 hosts a cco , psr j1852 + 0040 , which is strong evidence of core - collapse explosion . therefore
, its progenitor is a massive ( @xmath328 ) star born in the mcs . during its short lifetime ( @xmath329@xmath330 years in the main - sequence stage ; schaller et al .
1992 ) , the massive progenitor launches a strong stellar wind which blows a stellar wind bubble and sculpts the parent mcs .
the snr expanding inside the bubble may finally interact with the molecular cavity wall if the mc has not been dissipated , and actually several such snrs have been revealed during the past several decades ( see chen et al .
2014 ; zhang et al .
2015 and references therein ) .
chen et al . ( 2013 ) found a linear relationship between the size of the massive star s molecular cavity and the star s initial mass ( @xmath331@xmath332 relation ) : @xmath333 pc , where @xmath334 , @xmath335 , @xmath336 , with the mean pressure @xmath337 in the mcs about @xmath338 .
this relation can be used to estimate the initial masses of the progenitors for snrs interacting with molecular cavity walls or shells .
kes 79 is associated with the mcs in the velocity range of 95 - 115 @xmath4 , and the molecular shell delineates the western radio outer shell and confines the middle shell . if the smaller hemisphere corresponding to the bright x - ray / ir filaments is in contact with the molecular cavity created by the progenitor s stellar wind , we can adopt @xmath339 ( 8.3 pc ; we use @xmath340 here since part of the wind energy may leak into the tenuous medium ) .
the progenitor mass of kes 79 is thus estimated to be @xmath341 .
another method to estimate the progenitor mass is to compare the metal compositions inferred from the x - ray spectra with those predicted in the nucleosynthesis models .
this method requires that different metal species are well mixed and observed , if the abundances from the global gas properties are used for comparison .
according to the _ xmm - newton _ spectral analysis of the global hot gas , kes 79 has enriched ne ( 1.8 ) , mg ( 1.6 ) , si ( 1.4 ) , s ( 1.7 ) , and ar ( @xmath342 ) .
we first compare the best - fit abundances of ne , mg , s , and ar relative to si ( e.g , [ ar / si]/[ar / si]@xmath343 ) in kes 79 to those produced by the progenitors with different masses based on the spherical core - collapse nucleosynthesis models of nomoto et al .
( 2006 ) and woosley & weaver ( 1995 ) .
differences between the two models are discussed by kumar et al .
( 2014 ; see also kumar et al .
2012 ) . however , we find that none of the available nucleosynthesis models can explain all of the observed metal abundance ratios for this snr .
we subsequently compare the element masses in the x - ray - emitting gas with the predicted sn yields in the two nucleosynthesis models , which can provide at least a lower limit of the snr s progenitor mass .
the element ( @xmath344 ) mass @xmath345 is estimated from the abundance @xmath346 and the mass of the hot component @xmath347 ( @xmath348 ) : @xmath349 , where @xmath350 is the solar mass fraction of the metal @xmath344 adopted from anders & grevesse ( 1989 ) . as shown in figure [ f : yield ] , a progenitor mass less than @xmath351 can be excluded ( spherical explosion models ) , since the mass of observed ne exceeds that predicted in the @xmath352 yield models .
it is consistent with the value @xmath353 given by the @xmath331@xmath332 relation ( discussed above ) . the observed metal masses are no larger than the 20 @xmath354 yield models for either the spherical or bipolar explosion ( hypernova model with kinetic energy 1030 times of the typical value of @xmath355 erg ) scenarios . under the assumption that most of the ejecta are observed in the x - ray band and that the ne and mg are well mixed in the snr
, the progenitor mass of kes 79 is likely between 15 and 20 @xmath354 .
however , the element masses are not consistent with any group of the modeled nucleosynthesis yields .
the first explanation is that the ejecta are not well mixed or asymmetrically distributed in the snr .
asymmetric distribution of ejecta is present in the snr as discussed in section [ s : fragment ] .
an alternative reason might be that kes 79 is born from an asymmetric explosion with normal kinetic energy ( @xmath356 erg ) and the ejecta abundances can not be explained with current spherical explosion or hypernova bipolar explosion model .
last but not least , the available nucleosynthesis models do not consistently provide similar yields , partly because of the assumptions made in the calculations .
future detailed models incorporating mildly asymmetric explosions and binary evolution are desirable to provide a more secure constraint on the progenitor of kes 79 and other snrs .
we have investigated the multi - wavelength emission from the thermal composite snr kes 79 . using the multi - transition co data covering the whole snr , we study the large - scale molecular environment as well as the small - scale structures which cause the asymmetries of the snr .
we also revisit the 380 ks _ xmm - newton _ mos data and carry out imaging and spectroscopic analysis of the x - ray - emitting plasma .
the combined long - exposure x - ray data allow us to study the detailed distribution of different elements and the hot gas properties across the snr .
the main results are summarized as follows . 1 .
we provide kinematic and morphological evidence to support the interaction of snr kes 79 with the mcs in the velocity range 95 - 115 @xmath4 : ( 1 ) broadened @xmath14co @xmath22 line ( @xmath59 ) is detected for the first time near the protrusion at the northeastern radio boundary ; ( 2 ) the morphology agreement between the two @xmath14co @xmath22 filaments and the radio / ir / x - ray filaments in the east , and between the western molecular shell and the radio shells ; and ( 3 ) the red - shifted @xmath14co lines relative to @xmath18co , suggesting an interaction with the mcs from the foreground snr .
the molecular gas region ( mostly inside a @xmath357 region ) near kes 79 has a mass of @xmath358 , which is typical for a gmc .
the overall x - ray - emitting gas can be characterized by a cool ( @xmath359 kev ) under - ionized plasma with @xmath360 and solar abundances , plus a hot ( @xmath299 kev ) plasma with ionization timescale of @xmath361 and elevated ne ( @xmath362 ) , mg ( @xmath363 ) , si ( @xmath364 ) , s ( @xmath365 ) , and ar ( @xmath366 ) abundances .
the average densities of the two components are @xmath367 and @xmath368 , respectively .
the masses are @xmath369 and @xmath370 , respectively .
hence , most of the x - ray emission is contributed by the shocked ism .
kes 79 has a sedov age of 4.46.7 kyr .
the mean shock velocity is @xmath371 @xmath4 .
the _ xmm - newton _ image of kes 79 reveals many bright x - ray filaments embedded in a faint halo . a two - temperature model ( generally @xmath244 kev and @xmath2451.0 kev )
is required to describe the small - scale regions spectra all over the snr .
the filamentary gas has densities @xmath285@xmath372 and @xmath287@xmath282 for the cool and hot components , respectively , which are much larger than those of the halo gas ( @xmath373@xmath374 and @xmath375@xmath297 for the cool and hot components , respectively ) .
the ionization ages of the hot component in the filaments ( 0.52.0 kyr ) are smaller than in the halo ( @xmath28511 kyr ) .
the x - ray - bright filaments are probably produced by the snr interaction with the dense ambient gas , while the halo forms from snr breaking out to a low - density medium .
the snr shock propagating into the dense gas produces bright filamentary radiation in multiple wavelengths .
the x - ray filaments show good spatial correlation with the 24 @xmath9 ir filaments and part of the radio shells .
the filamentary mid - ir emission may come from the dust grains collisionally heated by the hot plasma .
5 . due to shaping by the inhomogeneous environment ,
kes 79 is likely to have a double - hemisphere morphology .
the smaller hemisphere containing bright filaments is at the back side and projected into the snr interior .
projection effect can explain the multiple - shell structures and the thermal composite morphology of kes 79 .
we find a high - velocity ejecta fragment which shows distinctly high temperature ( @xmath376 kev ) and abundances of ne ( @xmath318 ) , mg ( @xmath319 ) , si ( @xmath320 ) , s ( @xmath321 ) , and ar ( @xmath322 ) .
its velocity ( @xmath377 ) is 32% larger than the average velocity of the blast wave .
the high - velocity ejecta fragment , in addition to the asymmetric metal distribution across the remnant , supports the idea that the sn explosion is intrinsically asymmetric .
the progenitor mass of kes 79 is estimated to be 15@xmath378 by using two methods : ( 1 ) the linear relation between the progenitor mass and the wind blown bubble size ( @xmath331@xmath332 relation ) , and ( 2 ) a comparison between the metal masses and the yields predicted by nucleosynthesis models .
is thankful to randall smith for helpful discussions on the nei models .
p.z . and y.c .
thank the support of nsfc grants 11503008 , 11590781 , and 11233001 and the 973 program grant 2015cb857100 .
s.s.h . acknowledges support from the canada research chairs program , the nserc discovery grants program , and the canadian space agency .
z.y.z . acknowledges support from the european research council ( erc ) in the form of an advanced grant , cosmicism .
part of this research is conducted during a china scholarship council award held by p.z . at the u. of manitoba .
this research made use of nasa s astrophysics data system ( ads ) and the high - energy science archive research center ( heasarc ) maintained at nasa s goddard space flight center . | kes 79 ( g33.6 + 0.1 ) is an aspherical thermal composite supernova remnant ( snr ) observed across the electromagnetic spectrum and showing an unusual highly structured morphology , in addition to harboring a central compact object ( cco ) . using the co @xmath00 , @xmath11 , and @xmath22 data , we provide the first direct evidence and new morphological evidence to support the physical interaction between the snr and the molecular cloud in the local standard of rest velocity @xmath3 @xmath4 .
we revisit the 380 ks _ xmm - newton _ observations and perform a dedicated spatially resolved x - ray spectroscopic study with careful background subtraction .
the overall x - ray - emitting gas is characterized by an under - ionized ( @xmath5 ) cool ( @xmath6 kev ) plasma with solar abundances , plus an under - ionized ( @xmath7 ) hot ( @xmath8 kev ) plasma with elevated ne , mg , si , s and ar abundances .
the x - ray filaments , spatially correlated with the 24 @xmath9 ir filaments , are suggested to be due to the snr shock interaction with dense gas , while the halo forms from snr breaking out into a tenuous medium .
kes 79 appears to have a double - hemisphere morphology viewed along the symmetric axis .
projection effect can explain the multiple - shell structures and the thermal composite morphology .
the high - velocity , hot ( @xmath101.6 kev ) ejecta patch with high metal abundances , together with the non - uniform metal distribution across the snr , indicates an asymmetric sn explosion of kes 79 .
we refine the sedov age to 4.46.7 kyr and the mean shock velocity to 730 @xmath4 .
our multi - wavelength study suggests a progenitor mass of @xmath1120 solar masses for the core - collapse explosion that formed kes 79and its cco , psr j1852 + 0040 . |
magnetic fields on the sun exist in a large variety of phenomena and interact in various ways with the plasma and the radiation . in the convection zone
large and small scale magnetic fields are generated .
these magnetic fields are partially transported into the outer layers of the sun , i.e. , into the chromosphere and the corona .
the most prominent example of a magnetic phenomenon is a sunspot as seen in the photosphere .
a typical sunspot has a lifetime of a few weeks and has a size of about 30 granules .
the magnetic field strength spans from 1000 to 3000 gauss in the deep photosphere , summing up to a magnetic flux of some @xmath0 mx .
the magnetic field of a sunspot extends into the interior as well as into the outer layers of the sun .
the most detailed information of sunspots is obtained in the photosphere .
the topology of the magnetic field above and beneath the photosphere is poorly understood .
in particular our knowledge of the magnetic field extension into the interior presents a theoretical challenge .
direct measurements of the sub - photospheric structure are impossible , but at least for the larger scales , indirect methods are being explored in the framework of local helioseismology ( cf .
gizon , these proceedings ) . [
[ time - scales ] ] time scales : + + + + + + + + + + + + although the sunspot is a coherent phenomenon on large spatial and temporal scales , it seems crucial to realize that it is not static , but finds a dynamical equilibrium : a variety of small - scale features evolve on a dynamic time scale to produce a large - scale coherent structure on long time scales . this
fine structure is complex and is seen in white light images in form of umbral dots , light bridges , bright and dark penumbral filaments , penumbral grains , dark - cored bright filaments , penumbral twists , and other features .
this intensity fine structure corresponds to a fine structure of the velocity field and the magnetic field , which will be described below .
the dynamic fine structure forms a globally stable sunspot and it is the goal of sunspot physics to understand how an ensemble of short - lived features with small scales is organized to form a coherent large and long - living sunspot .
the coolness of sunspots relative to the surrounding quiet sun is readily explained by the tension of the magnetic field which tends to suppress convective motions . it is more difficult to understand why sunspots are as hot as they are :
neither radiative transport nor heat conduction can account for the surface brightness of sunspots .
hence convection can not be fully suppressed and the energy must be transported by convective flows . indeed , the fine structure manifests the inhomogeneities of the magnetic and velocity field and testifies that the energy transport in sunspots happens on small spatial scales by the motion of plasma . yet
, the crucial question is about the interaction between convective flows , the magnetic field , and the radiation .
are the flows non - magnetic or magnetic ? what is their intrinsic spatial scale ?
do coherent up- and downflows exist , similar to the granulation in the quiet sun ? [ [ jelly - fish - and - field - free - gaps ] ] jelly fish and field - free gaps : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + parker ( 1979 ) has introduced the jelly fish model in which the sub - photospheric magnetic field separates into individual bundles of field lines , resulting in gaps free of magnetic field .
the gaps between these bundles open up into very deep layers , being connected to the quiet sun convection . within these cracks ,
the field - free plasma would convect and transport heat upwards .
an umbral dot would correspond to the peak of a field - free gap .
more recently , spruit & scharmer ( 2006 ) suggested that such field - free gaps in the inclined magnetic field of the penumbra may result in elongated bright filaments , instead of in point - like dots , thereby proposing an explanation for the brightness of the penumbra .
the surplus brightness of the penumbra relative to the umbra would then be due to the fact that the convective cell can become larger in the more inclined and weaker magnetic field as in the less inclined ( more vertical ) and stronger field of the umbra .
[ [ stability - of - sunspots - and - monolithic - models ] ] stability of sunspots and monolithic models : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + sunspots are stable relative to the dynamical time , i.e. , alfvn waves are estimated to travel across a spot in about 1h , while the life time is in order of weeks .
how can it be that all this dynamic fine structure constitutes a spot which is stable ?
the question of stability can be addressed if one assumes a `` simple '' vertical magnetohydrostatic magnetic flux tube that fans out with heigth . in such models
the heat transport is attributed to ( magneto- ) convection , but is parametrized by a reduced mixing length parameter ( jahn 1989 , deinzer 1965 ) .
the dynamic fine structure is ignored and only their averaged effect on the stratification for umbra and penumbra is accounted for .
the configuration is in magneto - static equilibrium together with a hydrostatic equilibrium vertically and with a total pressure balance between the umbra , penumbra , and quiet sun horizontally ( see e.g. jahn & schmidt 1994 , pizzo 1990 )
. this configuration can be stable against the interchange instability ( meyer et al .
1977 ) , at least in the first 5 mm or so beneath the photosphere ( jahn 1997 ) . in these upper layers of the convection zone
the inclination of the interface between spot and surrounding is so large that buoyancy forces make the spot to float on the granulation . in deeper layers , beyond 5 mm
, the inclination of the outermost magnetic field line , i.e. , the magnetopause , is small relative to the vertical .
there , interchange ( fluting ) instability is no longer suppressed by buoyancy effects , and the magnetic configuration of a monolithic sunspot is unstable .
indeed , it has been proposed that the magnetic field strength progressively weakens in these deep layers shortly after the formation of a sunspot .
the decreasing field strength , the convective motions , and the interchange instability dynamically disrupt the sunspot magnetic field from the deeper roots ( schssler & rempel 2005 ) .
hence , the magnetic field in the deeper layers may be dispersed , but the floating part of the sunspot is stable .
for an extensive review of the sunspot structure , we refer the reader to an instructive overview by solanki ( 2003 ) .
the umbra of a sunspot harbors dynamic inhomogeneities .
they are observed as dot - like bright spots with typical sizes of half an arcsec or less , embedded in a more uniform and darker background .
these umbral dots seem to be present in all sunspots , although their intensity varies a lot
. in some spots they can be almost as bright as bright penumbral filaments , in other spots their intensity is much smaller . in the latter case ,
the dot - like intensity variations occurs in a background that also shows a lower intensity .
[ [ the - physics - of - umbral - dots ] ] the physics of umbral dots : + + + + + + + + + + + + + + + + + + + + + + + + + + + umbral dots are an obvious signature of convection , yet it is not so obvious to understand the type of convection that leads to umbral dots . in the field - free gap idea of parker , the convection is confined by the strong surrounding magnetic field , such that the column of convection narrows upwards and only a small brightening is seen at the surface .
observationally , it is established that the magnetic field in umbral dots is weaker than in the surroundings and that an upflow of at least a few hundred m / s is associated with them ( socas - navarro et al .
2004 , rimmele 2004 , 2008 , bharti et al .
the latter two observations also establish the presence of dark lanes across umbral dots .
the most recent simulations of radiatively driven magneto - convection in strong vertical magnetic field ( schssler & vgler 2006 ) result in local convective cells which produce umbral dots as well as their dark lanes .
these cells barely touch the photosphere , similar as in parker s idea .
the cells extend downward for a few mm , in the first 1 mm the cells have a weak magnetic field strength .
the weak field strength is caused by magnetic flux expulsion , i.e. convection advects the magnetic field ( as in weiss 1964 ) . in the simulations the magnetic field strength in the cells amounts to a few hundred gauss , but this number may decrease if magnetic dissipation is reduced in more advanced simulation runs . in any case ,
in deeper layers the magnetic field strength increases considerably .
hence the cells do not connect to field - free plasma in deeper layers . in this respect these new simulations change our model vision of umbral dots .
now , we may conceive that the umbra is an overall monolithic fully magnetic structure , in which the fine structure is a local disturbance .
the dots are produced locally by magneto - convection processes , which are needed for the energy transport .
the penumbra is a manifestation of small - scale structure .
the variety of the penumbral intensity fine structure is described in detail in the contribution to this volume by gran scharmer . in essence , there are bright and dark filaments , as well as penumbral grains .
it turns out that bright filaments have dark cores ( scharmer et al .
2002 , stterlin et al .
2004 ) and that intensity twists exist along bright filaments ( ichimoto et al .
2007b ) on spatial scales of about 0.2 arcsec .
the challenge consists in measuring the spectroscopic and spectropolarimetric signatures of this fine structure in order to derive their thermodynamic properties as well as their velocity and magnetic field . only recently , with the technological advance of adaptive optics and with observations from space , it has become possible to acquire such high spatial resolution data for exposure times as long as 5 sec or more .
this is a necessity to collect enough photons to have high spatial , spectral , and polarimetric resolution . at a spatial resolution of better than half an arcsec
, it can be demonstrated that not only the intensity and velocity , but also the magnetic field concists of a filamentary structure ( title et al .
1993 , langhans et al . 2005 , tritschler et al .
2007 , ichimoto et al . 2007a , 2008 ) .
actually , at a spatial resolution of better than half an arcsec , all physical quantities in the penumbra show small - scale variations and predominantly filamentary ( radially elongated ) features . however , the penumbra looks fairly uniform at a spatial resolution worse than 1 arc sec . at this lower spatial resolution ,
i.e. in average , the penumbra is brighter than the umbra , but less bright than in the surrounding granulation .
but even if the penumbra is less bright in average , the small scale peak - to - peak intensity variation in the penumbra is larger than in the granulation , and the spatial scales of the variations are smaller than in the granulation .
the same is true for velocities in the penumbra .
line - of - sight velocities in the penumbra of more than 5km / s have been derived from doppler shifts of photospheric lines ( e.g. , wiehr 1995 ) and radial flow channels with widths of less than half an arcsec are observed ( e.g. , tritschler et al .
2004 , rimmele & marino 2006 ) .
+ + + for understanding the nature of the penumbral fine structure , it is essential to know the topology of the velocity field and the magnetic field .
the first attempt to measure the flow field was undertaken by evershed in 1908 ( see evershed , 1909 ) in order to test hale s tornado theory of sunspots . yet , instead of a circular flow , evershed found a radial outflow of plasma , and until today we lack a consistent theory for sunspots .
before we discuss the progress in modeling the characteristic feature of the penumbra , we discuss important observational aspects .
[ [ the - flow - field ] ] the flow field : + + + + + + + + + + + + + + + with high spatial resolution , it is now established that the flow has a filamentary structure ( tritschler et al .
2004 , rimmele & marino 2006 ) . on average
, the flow has a small upward component in the inner penumbra and a small downward component in the outer penumbra ( schlichenmaier & schmidt 2000 , schmidt & schlichenmaier 2000 , tritschler et al .
2004 , langhans et al .
recent observations have revealed that radially aligned up- and downflows exist on small scales next to each other ( sainz dalda & bellot rubio 2008 ) .
regarding the photospheric height at which the flow exists , there is convincing evidence that the flow is predominantly present in the very deep photosphere , i.e. , beneath @xmath1 ( maltby 1964 , schlichenmaier et al .
2004 , bellot rubio et al .
the flow velocities measured in the penumbra are substantially larger than what is measured in the granulation .
individual penumbral profiles exhibit line satellites that are doppler shifted by up to 8 km / s ( e.g. , wiehr 1995 ) . from inversions ,
velocities well above 10 km / s have been found by del toro iniesta et al .
bellot rubio et al .
( 2004 ) find an azimuthally averaged evershed flow velocity of about 6.5 km / s , with local peaks of more than 10 km / s , based on two component inversions ( see below ) .
the small - scale flow field of dark cored bright filaments is discussed in the context of convective roll models ( at the end of sect .
[ sec.convective ] ) .
[ [ the - magnetic - field ] ] the magnetic field : + + + + + + + + + + + + + + + + + + + attempts to describe the magnetic field as being uniform along the line of sight are clearly inconsistent with the measured stokes @xmath2 , @xmath3 , and @xmath4 profiles ( e.g. , westendorp plaza et al .
2001a , 2001b ) .
in particular , the penumbral v - profiles with 3 or more lobes can not be explained by one component , even if unresolved doppler - shifted components are assumed ( schlichenmaier & collados 2002 ) .
therefore , it was proposed that the magnetic field is interlocked or in other words uncombed ( solanki & montavon 1993 ) . in order to keep things as simple as possible ,
the magnetic field is assumed to have two components with different directions .
indeed , if the observed stokes profiles with a spatial resolution of about 1 arcsec are interpreted with two components by means of inversions techniques , the fit to the observations is much better than with only one component ( bellot rubio 2004 , bellot rubio et al .
2003 , 2004 , borrero et al .
2004 , 2005 , beck 2008 ) .
such inversions yield one less inclined magnetic component that is only slightly doppler shifted , and a second magnetic field component that is somewhat weaker and more inclined , i.e. , approximately horizontal .
this second component carries the evershed flow , with spatially averaged flow speeds of about 6.5 km / s .
these inversions also show that the magnetic field of the second component is aligned with the associated flow , pointing slightly upwards in the inner and slightly downward in the outer penumbra . the inclination of the first magnetic field component increases from some 30 degree at the umbral - penumbral boundary to some 60 degrees at the outer penumbral boundary .
inversions which are optimized to locate the width and height of the flow layer find that the flow is present in the very deep atmosphere , in the continuum forming layers ( bellot rubio 2003 , borrero et al .
2006 , jurcak et al . 2007 ,
jurcak & bellot rubio 2008 ) .
indeed , the width can hardly be determined , since the lower end of the flow layer is found to be beneath @xmath5 .
at 0.3 arcsec spatial resolution , spectropolarimetric measurements reveal that , at least in the inner penumbra , the more inclined magnetic component which carries the flow is associated with the dark cored bright filaments .
individual dark cores have a smaller degree of circular polarization than their lateral brightenings ( langhans et al .
a thorough analysis shows that the latter statement is also true for the total polarization and that the dark core magnetic field is weaker and more inclined than in the lateral brightenings ( bellot rubio et al .
additionally , these studies confirm that the dark cores harbor strong evershed flows . [ [ the - magnetic - canopy ] ] the magnetic canopy : + + + + + + + + + + + + + + + + + + + + outside the white - light boundary of the penumbra , the inclined magnetic field continues into the chromosphere , forming a magnetic chromospheric canopy in the surroundings of the sunspot , rising with distance to the spot up to a height of approximately 800 km ( solanki et al .
in the canopy a radial outflow is present which is interpreted as the continuation of the evershed flow ( solanki et al .
1992 , rezaei et al .
2006 ) . however , it is estimated that only a few tenth of the flow mass is seen in the canopy .
the rest of the penumbral evershed flow must disappear within the penumbral downflow regions .
[ [ the - net - circular - polarization - ncp ] ] the net circular polarization ( ncp ) : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the ncp , @xmath6 , is a quantity that intimately links the flow and the magnetic field : ncp can only be non - zero , if and only if velocity gradients along the line of sight are present ( e.g. , sanchez almeida & lites 1992 ) .
the magnitude and the size of the ncp depends on the gradient of the line of sight velocity , but also on the gradients in the magnetic field strength , inclination , and azimuth ( landolfi & landi deglinnocenti 1996 , mller et al .
2002 , 2006 , borrero et al .
a predominantly horizontal flow channel embedded in a less inclined background magnetic field successfully explains some symmetry properties of ncp maps of sunspots ( schlichenmaier et al . 2002 ) as well as some properties of the center to limb variation of ncp ( martinez pillet 2000 , borrero et al .
2007 ) . yet
, some recent interpretations of ncp maps require that the flow component should be associated with stronger magnetic field ( tritschler et al .
2007 , ichimoto et al . 2008 ) , rather than being associated with the same or weaker magnetic field in the flow channels , as we would expect from the models . since there are also other indications for these stronger magnetic fields ( e.g. , bellot rubio 2003 , cabrera solana et al .
2008 , borrero & solanki 2008 ) , the concept of embedded flow channels will need to be reviewed taking into account these new measurements .
[ [ magnetized - or - non - magnetized - flow ] ] magnetized or non - magnetized flow : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + in terms of modeling the evershed flow , it is crucial to know whether or not the flow is magnetized . while ncp can be generated by a field - free flow in a magnetized environment ( e.g. , steiner 2000 ) , the observed @xmath7 profiles in certain locations in the penumbra show more than two lobes ( e.g. , schlichenmaier & collados 2002 , beck 2008 ) .
these additional lobes must be generated by an additional magnetic component : a non - magnetic component may produce a line asymmetry of stokes-@xmath8 and a non - zero ncp , but it can not produce additional lobes in stokes-@xmath7 . for this it needs to be magnetized ! and after all , the inversion results based on two components ( see above ) demonstrate that the doppler - shifted `` second '' component is magnetized .
hence , we are convinced that any model for the penumbra needs to account for a magnetized evershed flow .
the previous section stresses the point that the penumbra is a phenomenon of complex interaction of magneto - convective forces and radiation in a regime of inclined magnetic field of intermediate strength .
one simplified view on this problem is to consider a separation between convective plumes and a magnetic configuration as it is done in the field - free gap model .
another simplified view is by dealing with the problem in ideal mhd , in which the thin flux tube approximation is applicable .
the latter perspective is taken in the siphon flow model and the dynamic extension of it , the moving tube model . yet , for a full understanding it seems necessary to take into account dissipative magneto - convection driven by radiation .
but we want to stress that simplified models often help to isolate the dominating physical processes , and to understand the essentials .
originally proposed by parker to explain the umbral dots , spruit & scharmer ( 2006 ) and scharmer & spruit ( 2006 ) extended the concept of the field - free gaps to explain the bright penumbral filaments and they realized that such a configuration may also produce the dark cores within bright filaments , caused by a subtle radiative effect at the top of the field - free gap .
the idea of field - free gaps in the penumbra is that the inclined penumbral magnetic field produces bright elongations instead of dots .
the gaps are supposed to be void of magnetic field and to be connected to the surrounding quiet sun . within the gaps , overturning convection transports ample amounts of heat which would account for the brightness of the penumbra .
the convective flow field is directed upwards along the central lane of the filament and downward at the edges of the long sides of the filaments . within the field - free gap there
may exist a radial outflow that corresponds to the evershed flow .
the problem with this description is that the evershed flow which is observed to be magnetized need to be non - magnetized in the field - free gap .
the field - free gap model is in many respects similar to the model of the convective rolls proposed in 1961 by danielson ( see also grosser 1991 for a numerical investigation on this model ) .
convective rolls lie radially aligned next to each other .
two such rolls would form one filament as they rotate in opposite direction , producing an upflow in the central lane and a downflow at the lateral lanes .
danielson assumed that a horizontal magnetic field component would be associated with the rolls .
this model has been discarded for two reasons : ( 1 ) there was no evidence for the corresponding convective flow field , and ( 2 ) a major fraction of the magnetic flux in the penumbra is directed upwards , and not horizontal .
however , reason ( 1 ) depends on spatial resolution and the issue is not settled yet , as we can not rule out the existence small amplitude vertical motions of a few hundred m / s .
reason ( 2 ) could be overcome by assuming that the rolls are separated by less inclined ( more vertical ) magnetic field lines , which constitute a more or less static background magnetic field
. and , magnetized rolls interlaced by a static background field that is less inclined relative to the vertical would also meet the observational requirements of two magnetic components in the penumbra . in this respect , at least in principle , it is possible that the horizontal magnetic component carries an evershed flow .
the problem here is that up to now
there is only little support for downflows along the edges of bright filaments , although there is some indication for weak upflows within the dark cores ( e.g. , zakharov et al .
rimmele ( 2008 ) does find a convective - roll like flow field in a filament that extends into the umbra for a sunspot close to disk center , while bellot rubio et al .
( 2005 ) did not find indications for up and down flows associated with a dark - cored bright filament at disk center .
zakharov et al .
( 2008 ) observe a very small downward velocity component .
the latter authors argue that the downflow may be obscured by the upflows , and that convective rolls exist .
hence , the crucial question of vertical flows in the penumbra needs to be reconsidered . in order to minimize the effects of the horizontal radial outflow and of possible flows in azimuthal direction ,
sunspot observations at disk center are needed to learn about the presumably small vertical flow component .
another simplified view of the problem is by restricting oneself to ideal mhd . in the self - consistent magneto - static tripartite sunspot model of jahn & schmidt ( 1994 ) the surplus brightness of the penumbra relative to the umbra
is produced by a heat transfer through the magnetopause , i.e. through the interface between the quiet sun and the penumbra .
this additional heat is thought to be distributed horizontally by interchange convection of magnetic flux tubes .
the idea of dynamic magnetic flux tubes is compatible with the observationally finding of multiple magnetic components in the penumbra .
this motivated the study of the dynamics of a single thin magnetic flux tube as it evolves in a 2d static model background ( schlichenmaier et al .
1998a , b ) .
however , these studies did not confirm the concept of interchanging magnetic flux tubes which distribute heat horizontally .
instead , these studies created a new picture : the simulated tube lies along the magnetopause of the tripartite sunspot model and is taken to be a bundle of magnetic field lines with penumbral properties .
initially the tube is in magneto - static equilibrium . however , at a magnetopause that is sufficiently inclined , radiative heat exchange between the tube and the hotter quiet sun triggers an instability : a thin magnetic flux tube that initially lies along the magnetopause , ( a ) feels the hotter quiet sun , ( b ) heats up by radiation most effectively just beneath the photosphere , ( c ) expands , ( d ) rises through the subphotospheric convectively unstable stratification , and ( e ) develops an upflow along the tube , which brings hot subphotospheric plasma into the photosphere .
( f ) this hot upflow cools radiatively in the photosphere and streams radially outwards with supercritical velocity .
the radiative cooling sustains the gas pressure gradient that drives the flow .
( g ) the ouflow intrudes the convectively stable photosphere up to a height of some 50 to 100 km .
the equilibrium height is determined by the balance of the diamagnetic force which pulls the conducting tube upwards toward decreasing magnetic field strength and the downward acting buoyancy which increases as the tube is being pulled up in a convectively stable stratification .
[ [ weak - magnetic - field - at - footpoint ] ] weak magnetic field at footpoint : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the gas pressure gradient that drives the flow is caused by a surplus gas pressure building up inside the part of the tube that rises through the subphotospheric stratification . at the foot - point ,
i.e. , the intersection of the tube with the transition layer from convectively unstable to stable , the gas pressure is high , and in order to balance the total pressure with the surroundings , the magnetic field strength is strongly decreased relative to the surroundings . in this sense
the upflow foot - points can be considered as regions of weak magnetic field strength . in other words ,
the moving tube model is a magneto - convective mode which consists of a region of weak - field plasma that harbors hot up - flows and that travels inwards . in principle
the effect leading to the up and out flow works like an inverse convective collapse : in the classical convective collapse the plasma in the tube is cooled and a downflow occurs . here , the heating of the plasma results in an upflow , and consequently the magnetic field strength in the tube decreases as the flow continues . in the photosphere , the gas pressure gradient is sustained by radiative cooling .
the moving tube scenario successfully explains a number of observational findings : ( i ) penumbral grains are the photospheric footpoints of the tube , where the hot and bright plasma enters the photosphere .
( ii ) the upflow turns horizontally outwards in the photosphere and cools radiatively until it reaches temperature equilibrium .
this determines the length of the penumbral grains .
( iii ) the footpoints migrate inward , as many observed penumbral grains do ( e.g. , sobotka & stterlin 2001 ) .
( iv ) the horizontal outflow corresponds to the evershed flow .
( v ) the tube constitutes a flow channel being embedded in a background magnetic field .
this is in agreement with the uncombed penumbra , and produces realistic maps of ncp .
[ [ magneto - convective - overshoot ] ] magneto - convective overshoot : + + + + + + + + + + + + + + + + + + + + + + + + + + + + + an interesting effect that can be studied with the idealized moving tube model , is related to overshooting ( schlichenmaier 2002 , 2003 ) .
the upflow shoots into the convectively stable photosphere , and is turned horizontally by the magnetic curvature forces along the tube .
the dominating forces here are the centrifugal force of the flow , @xmath9 and the magnetic curvature force , @xmath10 , with @xmath11 being the curvature . in equilibrium
@xmath12 equals @xmath13 , with @xmath13 being the alfvn velocity . during the evolution of the tube
the velocity is roughly constant , but the magnetic field strength and hence @xmath13 decreases leading to an overshoot , which creates an oscillation of the outflow around its equilibrium position such that the tube adopts a wave - like shape , i.e. the plasma first shoots up and then down , again passing the equilibrium position . such a wave can be considered quasi - stationary , and the crest of such a wave can be compared with the properties of a siphon flow ( see below ) .
hence , the flow yields a serpentine shape , looking like a sea serpent , and evidence for such radially aligned up and down flows has been presented by sainz dalda & bellot rubio ( 2008 ) .
the amplitude of this wave increases as the magnetic field strength decreases , and eventually the downflow part dives in the sub - photosphere .
there the stratification is convectively unstable and the magnetic flux tube experiences a dynamic evolution , that produces outward propagating waves .
this scenario produces down - flows and makes the tube to disappear within the penumbra .
thereby it would solve a problem of the moving tube model : the out - flow would not extend into the surrounding canopy , but would disappear within the penumbra , as it is observed .
[ [ serpentine - flow ] ] serpentine flow : + + + + + + + + + + + + + + + + such a two - dimensional serpentine solution was criticized to be unstable in three dimensions ( thomas 2005 ) , arguing that buoyancy forces make the wavy tube to fall over sideways . but this argument is not valid , since the influence of the upflow at the footpoint of the tube is not taken into account . at the footpoint
the plasma is ejected upwards into the photosphere and due to conservation of momentum , the plasma overshoots and follows an up- and down wavy behavior .
the fact that the density at the upper crest is larger than in the surroundings does not make the tube to fall over . as an analogy
, one may think of a jet of water directed upwards with a garden hose .
as long as the jet is pointing upwards with the hose ( at the footpoint ) , the jet of water will not fall over .
the jet of water will not fall over , even though the density of water is larger than that of the surrounding air .
since the footpoint of the up - flowing flux tube and its inclination is constrained , the boundary condition circumvents the wavy flow to fall over .
therefore the argument of thomas ( 2005 ) is only true for a serpentine flow without a footpoint , and is not applicable here . [
[ siphon - flows ] ] siphon flows : + + + + + + + + + + + + + siphon flow arches are stationary magnetic flux tube models , which were proposed to explain the evershed flow ( e.g. , meyer & schmidt 1968 , thomas 1988 , degenhardt 1991 , thomas & montesinos 1991 ) .
this class of models makes the ad hoc assumption of different magnetic field strengths at the two foot - points of a magnetic arch , which is responsible for a gas pressure gradient along the tube driving the flow . in the dynamic sea - serpent solutions ( see above ) a quasi - stationary solution exists ( schlichenmaier 2003 ) .
this solution corresponds to one ( out of four ) particular siphon flow solution : a flow with a supercritical flow speed along the arch .
[ [ heat - transport ] ] heat transport : + + + + + + + + + + + + + + + temporal measurements of the intensity evolution rule out the existence of interchange convection ( solanki & redi 2003 ) , and also , the numerical work of the moving tube model did not confirm the concept of interchange convection of magnetic flux tubes as the heating mechanism for the surplus brightness of the penumbra : a crucial result of the numerical investigation is that a tube rises and develops an upflow , but the upflow does not stop nor does the tube sink back down to the magnetopause .
hence , instead of interchange convection the moving tube simulations suggests that the heating occurs in form of upflow channels along magnetic field lines .
ruiz cobo & bellot rubio ( 2008 ) demonstrate that such an up - flow is capable to account for the brightness of the penumbra and that such up - flows can produce dark - cored bright filaments with a length of up to 3 mm .
yet , even if such up flows can transport enough heat to account for the brightness of the penumbra , schlichenmaier & solanki ( 2003 ) have shown , that downflows within the penumbra are obligatory : there is not enough space for the magnetic flux associated with the up flows , such that down flows must remove the magnetic flux from the photosphere . in this respect , the overshoot scenario ( serpentine flow ) may help : the hot up - flow cools and the cool down - flow heats up in the hot sub - photosphere , and re - enters the photosphere as a hot upflow .
hence , the moving tube scenario encounters problems in accounting for sufficient heat transport , but there are ways to solve the heat transport problem with channeled flows . and these channeled flows are driven by radiative cooling .
before we continue with well accepted model descriptions of the penumbra , we also want to mention an off - track approach by kuhn & morgan ( 2006 ) who argue that in the photosphere of a cool spot a large fraction of the plasma is neutral . a simplified consideration with two fluid components , one neutral and one ionized , yields an outward plasma flow driven by osmotic pressure . whether or not such an effect and such a non - magnetic flow
is realized on the sun is not known , but it does not explain the observed evershed flow , since the observed flow is magnetized .
a better understanding of the sunspot penumbra is expected from numerical simulations of radiative magneto - convection in inclined magnetic fields .
first results of such simulations ( heinemann et al . 2007 ;
scharmer et al .
2008 , rempel , schssler , & knlker 2008 ) consider 3d boxes solving for the full set of mhd equations including the ( grey ) radiative transport .
these simulations consider a slice through the diameter of a round sunspot , including the umbra , the penumbra , and the surrounding quiet sun . assuming that the penumbral filament width is very small relative to the radius
, the slice has a rectangular geometry with periodic boundary conditions in the horizontal directions .
these simulations are still not able to produce a mature penumbra , but they succeed in reproducing single elongated filaments with lengths of up to a few mm which resemble in many ways what is observed as thin light bridges and penumbral filaments of the inner penumbra
. the heating of these filaments does not occur by a single hot upflow channel , but rather in a form of a vertically elongated convective roll : a central lane of upflow , associated with two adjacent lanes of donwnflow .
one convective cell has a vertical extension of some 500 km , while its lateral thickness is little less than 500 km .
the vertical component of the magnetic field seems to become expelled by the convective flow , such that the convective cell is associated with more horizontal magnetic field . in this sense , these simulation results are a revival of the convective rolls proposed by danielson in 1961 .
but in contrast to the danielson rolls , the filaments in the simulations are interlaced with less inclined stronger magnetic field than in the filaments , and the rolls are elongated in depth .
the simulated penumbral filaments resemble light bridges with an inner upflow and two lateral downflows forming two apparent rolls as observed e.g. by rimmele ( 2008 ) .
as the magnetic field becomes more inclined relative to the vertical , the upflow has an increasing horizontal outflow component .
this horizontal outflow component increases with heigth , culminating in the photosphere .
this horizontal flow is already present in the umbral dot simulations of schssler & vgler ( 2006 ) , but in an environment of more inclined magnetic fields this horizontal flow component becomes stronger .
however , at this stage , the horizontal velocities in the simulations are only a little larger than the vertical velocities of the convective roll , while the state - of - the - art observations retrieve a horizontal velocity that is roughly a factor of 20 ( 10km / s compared to @xmath14km / s ) larger than the vertical velocity . in that sense
the simulation fail to reproduce the evershed effect , but there are indications that the simulated horizontal velocity component increases with more inclined magnetic field , and future simulations that may exhibit a fully developed penumbra and more inclined magnetic fields are expected to develop stronger horizontal flows , thereby reproducing the evershed flow . in the present mhd simulations the energy transport in umbral dots , light bridges , and filaments in the inner penumbra
is accomplished by a magneto - convective mode , which may be characterized as convective elongated cells . yet , these simulations do not exhibit a mature penumbra and the associated evershed flow .
it remains to be seen whether this magnetoconvective mode is also capable to reproduce a mature penumbra , or wether another magnetoconvective mode exists in the outer penumbra .
to explain different aspects of the penumbral properties two `` simple '' model classes have been proposed for the penumbra : ( a ) the moving magnetic flux tube models assumes ideal mhd , in which flows channeled by magnetic fields account for the filamentation , the evershed flow , and the line asymmetries .
( b ) the gappy penumbra and convective rolls , which assume elongated convection cells to account for the surplus brightness of the penumbra .
neither of these simple models can account for all observational aspects : the moving tube scenario has problems to reproduce the overall down - flow in the outer penumbra and to account for all of the energy transport , while the elongated convective cells fail to produce an evershed flow .
the recent 3d box simulations of the full set of mhd equations show that magneto - convective heat transport may take place in elongated pancakes similar to what was proposed by danielson .
these simulations produce elongated convective cells which are associated with horizontal magnetic fields with weaker strengths , but in contrast to danielson s proposal the horizontal rolls with horizontal magnetic fields are embedded in stronger and more vertical magnetic field .
this latter magnetic field component would form a more or less static background .
considering the state - of - the - art of observational results and theoretical modeling , we conclude that there is evidence for both , a channeled flow with velocities of more than 5km / s producing the evershed flow , while the new 3d simulations suggest the existence of elongated convective rolls with much smaller up- and downflow velocities in the order 0.5 km / s
. it may well be that the penumbra is a superposition of both , channeled flows above convective rolls , but at this point , we do not know .
more advanced simulations will ultimately produce a fully developed penumbra , and by then it will be possible to understand how an ensemble of dynamic filaments is capable to form a stable penumbra , and how the penumbra is heated .
finally , the questions of sunspot stability and how sunspots form , evolve , and decay can be addressed . yet , at this point , the models are not fully consistent with obervational facts . in particular
it remains to be seen whether the flow pattern of convective rolls can be measured , and whether the observed penumbral line asymmetries in the stokes parameters including the ncp can be reproduced by such models .
spectropolarimetric measurements need to have a spatial resolution of better than 0.1 arcsec to be comparable to the models .
l. r. bellot rubio , r. schlichenmaier , and a. tritschler , _ two - dimensional spectroscopy of a sunspot .
thermal and kinematic structure of the penumbra at 0.5 arcsec resolution _ , * 453 * ( 2006 ) , 11171127 .
l. r. bellot rubio , s. tsuneta , k. ichimoto , y. katsukawa , b. w. lites , s. nagata , t. shimizu , r. a. shine , y. suematsu , t. d. tarbell , a. m. title , and j. c. del toro iniesta , _ vector spectropolarimetry of dark - cored penumbral filaments with hinode _ , * 668 * ( 2007 ) , l91l94 . j. m. borrero , s. k. solanki , l. r. bellot rubio , a. lagg , and s. k. mathew , _ on the fine structure of sunspot penumbrae .
i. a quantitative comparison of two semiempirical models with implications for the evershed effect _ ,
* 422 * ( 2004 ) , 10931104 .
d. cabrera solana , l. r. bellot rubio , j. m. borrero , j .- c .
del toro iniesta , _ temporal evolution of the evershed flow in sunspots .
ii . physical properties and nature of evershed clouds _ , * 477 * ( 2008 ) , 273283 .
k. ichimoto , r. a. shine , b. lites , m. kubo , t. shimizu , y. suematsu , s. tsuneta , y. katsukawa , t. d. tarbell , a. m. title , s. nagata , t. yokoyama , and m. shimojo , _ fine - scale structures of the evershed effect observed by the solar optical telescope aboard hinode _ , * 59 * ( 2007a ) , 593 .
k. ichimoto , y. suematsu , s. tsuneta , y. katsukawa , t. shimizu , r. a. shine , t. d. tarbell , a. m. title , b. w. lites , m. kubo , and s. nagata , _ twisting motions of sunspot penumbral filaments _ , science * 318 * ( 2007b ) , 1597 .
k. ichimoto , s. tsuneta , y. suematsu , y. katsukawa , t. shimizu , b. w. lites , m. kubo , t. d. tarbell , r. a. shine , a. m. title , and s. nagata , _ net circular polarization of sunspots in high spatial resolution _ ,
* 481 * ( 2008 ) , l9l12 .
k. jahn , _ physical models of sunspots _ , asp conf .
ser . 118 : 1st advances in solar physics euroconference
. advances in physics of sunspots ( b. schmieder , j. c. del toro iniesta , and m. vazquez , eds . ) , 1997 , p. 122 . j. jurcak
. l. r. bellot rubio , k. ichimoto , y. katsukawa , b. w. lites , s. nagata , t. shimizu , y. suematsu , t. d. tarbell , a. m. title , and s. tsuneta , _ the analysis of penumbral fine structure using an advanced inversion technique _
, pasj * 59 * ( 2007 ) , s601s606 .
j. r. kuhn and h. morgan , _ osmotically driven neutral sunspot winds _ , solar mhd theory and observations : a high spatial resolution perspective ( j. leibacher , r. f. stein , and h. uitenbroek , eds . ) , astronomical society of the pacific conference series , vol . 354 , december 2006 , pp . 230+ .
r. schlichenmaier , _ the sunspot penumbra : new developments _ , asp conf .
286 : current theoretical models and future high resolution solar observations ( a. pevtsov and h. uitenbroek , eds . ) , 2003 , pp .
211226 .
h. socas - navarro , v. martinez pillet , m. sobotka , m. vazquez , _ the thermal and magnetic structure of umbral dots form the inversion of high - resolution full stokes observations _ , * 614 * , ( 2004 ) , 448456 .
c. westendorp plaza , j. c. del toro iniesta , b. ruiz cobo , v. martnez pillet , b. w. lites , and a. skumanich , _ optical tomography of a sunspot .
vector magnetic field and temperature stratification _ , * 547 * ( 2001 ) , 11301147 . | the penumbra of a sunspot is a fascinating phenomenon featuring complex velocity and magnetic fields .
it challenges both our understanding of radiative magneto - convection and our means to measure and derive the actual geometry of the magnetic and velocity fields . in this contribution
we attempt to summarize the present state - of - the - art from an observational and a theoretical perspective .
we describe spectro - polarimetric measurements which reveal that the penumbra is inhomogeneous , changing the modulus and the direction of the velocity , and the strength and the inclination of the magnetic field with depth , i.e. , along the line - of - sight , and on spatial scales below 0.5 arcsec . yet
, many details of the small - scale geometry of the fields are still unclear such that the small scale inhomogeneities await a consistent explanation . a simple model which relies on magnetic flux tubes evolving in a penumbral `` background '' reproduces some properties of sunspot inhomogeneities , like its filamentation , its strong ( evershed- ) outflows , and its uncombed geometry , but it encounters some problems in explaining the penumbral heat transport . another model approach , which can explain the heat transport and long bright filaments , but fails to explain the evershed flow , relies on elongated convective cells , either field - free as in the gappy penumbra or filled with horizontal magnetic field as in danielson s convective rolls .
such simplified models fail to give a consistent picture of all observational aspects , and it is clear that we need a more sophisticated description of the penumbra , that must result from simulations of radiative magneto - convection in inclined magnetic fields .
first results of such simulations are discussed .
the understanding of the small - scales will then be the key to understand the global structure and the large - scale stability of sunspots . |
j. a. noble is a royal commission for the exhibition of 1851 research fellow
* corresponding author : s. coussan , [email protected] | in the quest to understand the formation of the building blocks of life , amorphous solid water ( asw ) is one of the most widely studied molecular systems . indeed , asw is ubiquitous in the cold interstellar medium ( ism ) , where asw - coated dust grains provide a catalytic surface for solid phase chemistry , and is believed to be present in the earth s atmosphere at high altitudes .
it has been shown that the ice surface adsorbs small molecules such as co , n@xmath0 , or ch@xmath1 , most likely at oh groups dangling from the surface .
our study presents completely new insights concerning the behaviour of asw upon selective infrared ( ir ) irradiation of its dangling modes .
when irradiated , these surface h@xmath0o molecules reorganise , predominantly forming a stabilised monomer - like water mode on the ice surface .
we show that we systematically provoke `` hole - burning '' effects ( or net loss of oscillators ) at the wavelength of irradiation and reproduce the same absorbed water monomer on the asw surface .
our study suggests that all dangling modes share one common channel of vibrational relaxation ; the ice remains amorphous but with a reduced range of binding sites , and thus an altered catalytic capacity .
+ asw is a molecular system which has long provoked interest due , in part , to its role in the formation of molecules key to the origins of life@xcite .
asw has long been known to accrete small molecules such as co , h@xmath0o , n@xmath0 , or ch@xmath1@xcite , initiating chemical and photochemical surface reactivity@xcite . in the ism ,
water in the form of asw is the most abundant solid phase molecular species@xcite .
the production of molecules , from the most simple , h@xmath0@xcite , to the more complex ch@xmath2oh@xcite , and even precursors to the simplest amino acid , glycine@xcite , is catalysed by the asw surface@xcite ; both the outer surface and surfaces within its porous structure are involved .
the selective ir irradiation of crystalline ice@xcite and water clusters@xcite has already been studied . in the former case , the desorption of h@xmath0o molecules , and in the latter , the dissociation of clusters , was stimulated .
we are interested in the behaviour of asw upon selective ir irradiation and have studied the irradiation of the four surface modes of this ice , assigned in the literature@xcite and illustrated in figure [ fig : cartoon ] .
theoretical calculations , supported by experimental studies , suggest that water molecules in the dh mode are bi- or tri - coordinated , presenting one free oh bond dangling at the surface ; do molecules present a free oxygen electronic doublet ; and s4 molecules have a tetrahedral structure at the surface , which is distorted compared to the tetrahedra of bulk asw . ) , do ( 3549 @xmath3 ) , and s4 ( 3503 @xmath3 ) are illustrated on a representative asw sample.,scaledwidth=50.0% ] in these experiments we prepared a pure asw sample ( figure [ fig : cartoon ] ) as follows : deionised water was subjected to multiple freeze - pump - thaw cycles under vacuum to remove dissolved gases .
mixtures of purified h@xmath0o and helium ( air liquide , @xmath4 99.9999 % ) gas were prepared in a stainless steel dosing line with base pressure 10@xmath5 mbar in ratios of @xmath6 .
ices were produced by depositing the gas mixture directly onto a gold - plated copper surface held at 50 k ( to avoid trapping of the vector gas or nitrogen ) then cooled to 3.7 k ( the cooling typically takes around five minutes due to the high cryogenic power of 0.5 w at 4 k ) .
the cooled surface is located in a high vacuum chamber with a base pressure of 10@xmath7 mbar at 3.7 k. ir spectra were recorded in reflection mode using a bruker 66/s ftir spectrometer equipped with a mct detector ( 4000 800 @xmath3 ) .
full details of the experimental setup are given in coussan _
_ et al.__@xcite .
the ices grown in our study were characterised as purely amorphous in nature due to the position of the bulk oh stretch and the characteristic dangling modes at 3720 and 3698 @xmath3 ( see figure [ fig : cartoon])@xcite .
some previous studies using a helium carrier gas have reported the formation of ice nanocrystals or clusters@xcite , but this can be ruled out due to the absence of an absorption feature centred at 3692 @xmath3 . this band was never observed in any deposition of a h@xmath0o : he mixture during the work detailed here .
after deposition , ices were selectively irradiated using a tunable ir opo laserspec ( 1.5 4 @xmath8 m ) , pumped at 10 hz by a pulsed nd : yag quantel brilliant b laser ( 1064 nm , pulse duration 6 ns ) .
the average laser power is @xmath9 35 mw in the @xmath10 domain , except in the range 3520 3500 @xmath3 where it is @xmath9 10 mw , with a fwhm @xmath4 1.5 @xmath3 .
each irradiation was performed for one hour to ensure saturation of the effects . , ( b ) 3698 @xmath3 , ( c ) 3549 @xmath3 , and ( d ) 3503 @xmath3 .
the spectra are coloured as follows : before irradiation , blue ; after irradiation , green ; difference spectrum , red .
the insets magnify the irradiated range.,scaledwidth=50.0% ] figure [ fig : irr ] shows the results of irradiations carried out on the four dangling mode bands .
the irradiations provoke permanent `` hole - burnings '' in the irradiated band which are slightly shifted ( by up to 3 @xmath3 ) compared to the irradiation frequency .
the `` hole - burning '' is clearest for the irradiation of dh at 3698 @xmath3 but is also easily visible for the irradiations at 3720 and 3549 @xmath3 .
interestingly , upon each irradiation , we observe the growth of a new band centred at @xmath11 3725 @xmath3 , with fwhm @xmath9 5 @xmath3 .
both the shifts in frequency upon irradiation , and the narrowness of the `` hole - burning '' effects and the newly created bands at 3725 @xmath3 illustrate the inhomogeneity of the bands ; each band contains a distribution of oscillators , but only one class of oscillator isomerises upon irradiation at a given frequency , producing one new oscillator class .
after irradiation at 3698 @xmath3 , a second , smaller peak at 3638 @xmath3 is also observed ( see figure [ fig : depot]a ) . unirradiated asw samples and their ir spectra were stable and remained unchanged over the timescale of an irradiation study .
the common feature of all the irradiations in figure [ fig : irr ] is the growth of the 3725 @xmath3 band .
what is the source of this new band , previously unidentified in asw spectra ?
considering energetics only , irradiating between 3720 and 3503 @xmath3 could potentially break one or two hydrogen bonds , as the average weak h - bond strength is around 1800 @xmath3 .
is the band , therefore , due to h@xmath0o surface molecules in a different conformation than those in dh , do and s4 modes , or is it due to water molecules which have desorbed then re - adsorbed at the surface ?
the latter response can be immediately discarded based on the results of molecular adsorption studies which have shown a red - shift of the oh dangling frequency upon adsorption of multiple molecular species at the dangling modes@xcite .
for example , in the case of nitrogen adsorption onto asw , manca _ _ et al.__@xcite observed a red shift of 22 @xmath3 of the dh mode@xcite . in our experiments , the new band at 3725
@xmath3 is blue - shifted with respect to the dh modes .
moreover , the dynamic vacuum of 10@xmath7 mbar rapidly evacuates any desorbing molecules , such that the residual pressure is too low to allow redeposition .
thermal effects are discounted based upon annealing of asw samples , which provoked a global decrease of the dangling bonds and no production of narrow peaks at 3725 @xmath3 , in agreement with previous studies@xcite . upon irradiation ,
the newly produced band is narrow , indicating a single , homogeneous vibrational mode .
thus , the 3725 @xmath3 band is clearly not due to a perturbed dh mode , but it rather has dangling oh character , and we propose that it is samples a bi - coordinated h@xmath0o molecule , with two dangling oh and two co - ordinated electron pairs ( d2h ) . this structure explains the blue shift of the peak with respect to the dh mode at 3720 @xmath3 as , because neither of the two oh oscillators is directly hydrogen bonded , the free oh oscillators are less perturbed than the surface dh modes . ,
reproduced from figure [ fig : irr ] ; ( b ) an asw sample plus background - deposited water ; ( c ) spectrum b plus background - deposited nitrogen ; ( d ) an asw sample plus background - deposited nitrogen ; ( e ) spectrum d plus background - deposited h@xmath0o : n@xmath0 mixture.,scaledwidth=50.0% ] we performed further experiments to verify the source of the two new peaks at 3725 and 3638 @xmath3 , as illustrated in figure [ fig : depot ] .
the background deposition of water on a pure asw sample does not result in a new band , but rather the global growth of the dh and bulk oh bands ( figure [ fig : depot]b ) .
however , subsequent background deposition of pure nitrogen , a molecule present as a low - level source of pollution in the chamber , provokes the appearance of a band at 3726 @xmath3 ( figure [ fig : depot]c ) , revealed thanks to the magnifying effect of nitrogen@xcite .
the background deposition of n@xmath0 upon a new asw sample does not produce a clear peak ( figure [ fig : depot]d ) , but the background deposition of a h@xmath0o : n@xmath0 ( 1:10 ) mixture provokes the appearance of two narrow bands at 3725 @xmath3 and 3638 @xmath3 ( figure [ fig : depot ] ) , with a peak area ratio of approximately 10:1 , compared to 6:1 seen after irradiation at 3698 @xmath3 . when this deposition spectrum is compared to the spectrum of a gas - phase water monomer ( see table [ table : frequencies ] ) , the two peaks show red shifts of nearly 200 @xmath3 with respect to the @xmath12 modes of the monomer ( one always compares gas - phase experimental results with those of a `` classical '' theoretical calculation , as classical calculations are carried out for isolated species .
it also allows us to approximate the perturbation induced by the medium ) .
these red shifts are almost precisely those observed in the solid phase for water in a nitrogen matrix , where the @xmath13 ratio is 8:1 , and a h@xmath0o - n@xmath0 complex.@xcite considering our experimental results and the literature , the band at 3725 @xmath3 is positively attributed to a water monomer interacting with the surface via its two electronic doublets ; its large intensity compared to the other dh bands is explained by the magnifying effect of nitrogen , as extensively investigated by hujo _ _ et al.__@xcite .
we suggest that nitrogen molecules , present as a low - level pollutant in the chamber , serendipitously complex the water molecule , as illustrated in figure [ fig : depot ] , stabilising the molecule , preventing any further adsorption , and magnifying the oh stretching bands
. however , the nitrogen must only be present as a trace pollutant on the surface , as no shifting of the dh peak positions of the asw is seen , either during cooling of the water ice sample from 50 k to 3.7 k , or during the irradiation period .
if nitrogen were present at multilayer concentrations , we would expect to see a red shift@xcite of up to 22 @xmath3 .
the @xmath14 peak at 3638 @xmath3 is not observed after irradiation of the other three surface modes , but this is likely due to its low intensity compared to that of @xmath15 .
.comparison between calculated and observed @xmath15 and @xmath14 water monomer frequencies .
frequencies are given in @xmath3 . [ cols="^,^,^,^,^,^ " , ] one explanation for the observed `` hole - burning '' is that amorphous ice is unable to relax all of the vibrational energy injected at the surface through bulk relaxation channels .
as a result , some fraction of this energy is accumulated at the surface and in the immediate sub - layers , where it induces reconstruction of the surface .
we observe saturation of the `` hole - burning '' events within the timescale of the irradiations performed , suggesting that the rearrangement of surface molecules is not an efficient relaxation channel , but is a minority effect compared to the main relaxation channels via the bulk ice .
thus , the production of the 3725 @xmath3 band is not the major result of irradiation , but is due to the inability of the ice to fully dissipate the injected energy .
it has previously been suggested that asw is a disordered material which has no long - range organisation@xcite , which could help to explain the lack of efficiency in the bulk relaxation channels . although it is possible that some h@xmath0o molecules desorb upon irradiation , as in the studies of focsa _ _ et al.__@xcite , we consider this effect to be minor because no increase in pressure is observed in the chamber and the energy injected is not enough to break more than two h - bonds .
it is curious that the absorption band at 3725 @xmath3 is produced upon irradiation of all dangling bonds .
the irradiation effects at 3720 and 3698 @xmath3 , in particular , are very similar , except that the `` hole - burning '' of the doubly - coordinated dh is less pronounced than that of the triply - coordinated dh .
such molecular rearrangement requires only a reorientation of the water molecule and thus is not `` energy consuming '' compared to the energy injected into the system .
these results also provide evidence of a local ordering to the asw structure . upon irradiation of the dh modes ,
both the `` hole - burning '' in absorption bands and the newly produced monomer band at 3725 @xmath3 are relatively narrow ( fwhm @xmath9 5 @xmath3 ) .
this suggests that each surface molecule is surrounded by a locally ordered oscillator network , resulting reproducibly in the production of one oscillator class upon selective irradiation .
if we consider irradiation of the s4 band , centred at 3503 @xmath3 , we were unable to observe a definitive `` hole - burning '' event , likely due to the width of the band .
however , we observed an increase at 3725 @xmath3 , as for the dh modes .
it is unlikely that the tetra - coordinated s4 molecules are themselves ejected from within the surface layer , as this would be highly energetically unfavourable .
however , as we see an increase at 3725 @xmath3 , it is likely that the relaxation channel involves the breaking of h - bonds , with two breaks being enough to `` transform '' a s4 molecule into a monomer - like water molecule .
the case of do is likely intermediate between those of dh and s4 . during each of these irradiations
, we see no interconversion between the modes , suggesting that there is only a single `` surface '' channel for the release of excess vibrational energy . in this work we have provided new insights into the bonding and structure of the surface molecules in amorphous solid water .
the ensemble of our results show that surface modes , in particular the dh dangling bonds , are sensitive to photo - induced rearrangement due to competition between surface reorganisation and the main relaxation channels in the bulk water ice . rather than desorbing from the surface , molecules embedded in the surface layer
become loosely associated with the surface in the form of monomer - like structures interacting through their two free electron pairs .
the fortuitous presence of nitrogen in the chamber both promoted the magnification of the oh stretching mode of the monomer - like molecule and stabilised it on the ice surface .
inducing such conformational changes in an asw surface potentially alters its physicochemical properties , most notably its catalytic potential . 100 buch , v. ; devlin , j. p. spectra of dangling oh bonds in amorphous ice : assignment to 2 and 3 coordinated surface molecules .
_ j. chem .
phys . _ * 1991 * , _ 94 _ , 4091 - 4092 .
devlin , j. p. ; buch , v. surface of ice as viewed from combined spectroscopic and computer modeling studies .
_ j. phys .
chem . _ * 1995 * , _ 99 _ , 16534 - 16548 .
devlin , j. p. ; buch , v. vibrational spectroscopy and modeling of the surface and substructure of ice and of ice - adsorbate interactions .
_ j. phys .
chem .
b _ * 1997 * , _ 101 _ , 6095 - 6098 .
manca , c. ; allouche , a. quantum study of the adsorption of small molecules on ice : the infrared frequency of the surface hydroxyl group and the vibrational stark effect .
_ j. chem .
phys .
_ , * 2001 * , _ 114 _ , 4226 - 4234 .
manca , c. ; martin , c. ; roubin , p. spectroscopic and volumetric characterization of a non - microporous amorphous ice .
_ chem .
phys .
lett .
_ * 2002 * , _ 364 _ , 220 - 224 .
manca , c. ; roubin , p. ; martin , c. volumetric and infrared co - measurements of ch@xmath1 and co isotherms on microporous ice .
_ chem .
phys .
lett .
_ * 2000 * , _ 330 _ , 21 - 26 .
collings , m. p. ; anderson m. a. ; chen , r. ; dewer j. w. ; viti , s. ; williams d. a. ; mc coustra m. r. s. laboratory survey of the thermal desorption of astrophysically relevant molecules . _
monthly notices royal astronomical society _ * 2004 * , _ 354 _ , 1133 - 1140 .
watanabe , n. ; kouchi , a. efficient formation of formaldehyde and methanol by the addition of hydrogen atoms to co in h@xmath0o - co ice at 10 k. _ astrophysical journal lett . _ * 2002 * , _ 571 _ , l173-l176 .
rowland , b. ; fisher , m. ; devlin , j. p. probing icy surfaces with the dangling oh mode absorption : large ice clusters and microporous amorphous ice . _ j. chem .
phys . _ * 1991 * , _ 95 _ , 1378 - 1384 .
leger , a. ; klein , j. ; cheveigne s. d. ; guinet , c. ; defourneau d. ; belin , m. the 3.1 micron absorption in molecular clouds is probably due to amorphous h@xmath0o ice . _
astronomy & astrophysics _ * 1979 * , _ 79 _ , 256 - 259 .
whittet , d. c. b. ; bode , m. f. ; longmore , a. j. ; baines , d. w. t. ; evans , a. interstellar ice grains in the taurus molecular clouds .
_ nature _ * 1983 * , _ 303 _ , 218 - 221 .
hollenbach , d. j. ; salpeter , e. e. surface recombination of hydrogen molecules .
_ astrophysical journal _ * 1971 * , _ 163 _ , 155 - 164 .
manic , g. ; ragun , g. ; pirronello , v. ; roser , j. e. ; vidali , g. laboratory measurements of molecular hydrogen formation on amorphous water ice .
_ astrophysical journal lett . _ * 2001 * , _ 548 _ , l253-l256 .
danger , g. ; duvernay , f. ; theul , p. ; borget , f. ; chiavassa , t. hydroxyacetonitrile ( hoch@xmath0cn ) formation in astrophysical conditions .
competition with the aminomethanol , a glycine precursor .
_ astrophysical journal _ * 2012 * , _ 756 _ , 11 - 20 .
williams , d. a. ; taylor , s. d. the chemical role of cosmic dust .
_ quarterly journal royal astronomical society _ * 1996 * , _ 37 _ , 565 - 592 .
focsa , c. ; chazallon , b. ; destombes , j. l. resonant desorption of ice with a tunable linbo@xmath2 optical parametric oscillator .
_ surface science _ * 2003 * , _ 528 _ , 189 - 195 .
mihesan , c. ; ziskind m. ; chazallon b. ; therssen e. ; desgroux p. ; gurlui s. ; focsa c. ; ir wavelength - selective laser desorption via o - h and c - h stretching modes .
_ applied surface science _ * 2006 * , _ 253 _ , 1090 - 1094 .
buck , u. ; ettischer , i. ; melzer , m. ; buch , v. ; sadlej , j. structure and spectra of three - dimensional ( h@xmath0o)@xmath16 clusters , n = 8 , 9 , 10 .
_ physical review lett . _ * 1998 * , _ 80 _ , 2578 - 2581 .
wassermann , t. n. ; suhm , m. a. ; roubin , p. ; coussan , s. isomerization around c c and c o bonds in 1-propanol : collisional relaxation in supersonic jets and selective ir photo - isomerization in cryogenic matrices .
_ j. mol .
struct . _ * 2012 * , _ 1025 _ , 20 - 32 .
rowland , b. ; devlin , j. p. spectra of dangling oh groups at ice cluster surfaces and within pores of amorphous ice .
_ j. chem .
phys . _ * 1991 * , _ 94 _ , 812 - 813 .
manca , c. ; martin , c. ; roubin , p. comparative study of gas adsorption on amorphous ice : thermodynamic and spectroscopic features of the adlayer and the surface . _ j. phys .
chem .
b _ * 2003 * , _ 107 _ , 8929 - 8934 .
sadlej , j. ; rowland , b. ; devlin , j. p. ; buch , v. vibrational - spectra of water complexes with h-2 , n-2 , and co. _ j. chem . phys .
_ * 1995 * , _ 102 _ , 4804 - 4818 .
buch , v. ; bauerecker , s. ; devlin , j. p. solid water clusters in the size range of tens - thousands of h2o : a combined computational / spectroscopic outlook .
_ int .
rev .
phys .
chem . _ * 2004 * , _ 23 _ , 375 - 433 .
hujo , w. ; gaus , m. ; schultze , m. ; kubar , t. ; grunenberg , j. ; elstner , m. ; bauerecker , s. effect of nitrogen adsorption on the mid - infrared spectrum of water clusters .
_ j. phys .
chem .
a _
* 2011 * , _ 115 _ , 6218 - 6225 .
coussan , s. ; loutellier , a. ; perchard , j. p. ; racine , s. ; bouteiller , y. matrix isolation infrared spectroscopy and dft calculations of complexes between water and nitrogen .
_ j. mol .
struct . _ * 1998 * , _ 471 _ , 37 - 47 .
benedict , w. s. ; gaillard , n. ; plyler , e. k. rotation vibration spectra of deuterated water vapor .
_ j. chem .
phys . _ * 1956 * , _ 24 _ , 1139 - 1165 .
coussan , s. ; roubin , p. ; perchard , j. p. infrared induced isomerizations of water polymers trapped in nitrogen matrix . _
chem .
phys .
_ * 2006 * , _ 324 _ , 527 - 540 . |
superconductivity ( s ) and ferromagnetism ( f ) are two competing phenomena . on one hand
a bulk superconductor expels the magnetic field ( meissner effect ) . on the other hand the magnetic field for @xmath4
destroys the superconductivity .
this fact is due to the unequal symmetry in time : ferromagnetic order breaks the time - reversal symmetry , whereas conventional superconductivity relies on the pairing of time - reversed states .
it turns out that the combination of both , superconductor and ferromagnet , leads to rich and interesting physics .
one particular example the phase oscillations of the superconducting ginzburg - landau order parameter inside the ferromagnet will play a major role for the devices discussed in this work .
the current - phase relation @xmath5 of a conventional sis josephson junction ( jj ) is given by @xmath6 .
@xmath7 is the phase difference of the macroscopic superconducting wave functions @xmath8 ( order - parameters of each electrode ) across the junction , @xmath9 is the critical current .
usually @xmath9 is positive and the minimum of the josephson energy @xmath10 , @xmath11 is at @xmath12 .
however , bulaevski _ et al . _
@xcite calculated the supercurrent through a jj with ferromagnetic impurities in the tunnel barrier and predicted a negative supercurrent , @xmath13 . for @xmath14
the solution @xmath12 is unstable and corresponds to the maximum energy @xmath15 , while @xmath16 is stable and corresponds to the ground state .
such jjs with @xmath16 in ground state are called @xmath2 junctions , in contrast to conventional @xmath1 junctions with @xmath12 . in case of a @xmath2 josephson junction the first josephson relation
is modified to @xmath17 . in experiment
the measured critical current in a single junction is always positive and is equal to @xmath18 .
it is not possible to distinguish @xmath1 jjs from @xmath2 jjs from the current - voltage characteristic ( ivc ) of a single junction .
the particular @xmath19 @xcite and @xmath20 @xcite dependencies for sfs / sifs type junction are used to determine the @xmath2 coupled state . for low - transparency sifs junctions the @xmath20 dependence is given by @xmath21 where @xmath22 are the decay and oscillation lengths of critical current and @xmath23 is the dead magnetic layer thickness @xcite . for @xmath24 the coupling in ground state of jjs
is shifted by @xmath2 . in a second work bulaevski _
_ @xcite predicted the appearance of a _ spontaneous _ supercurrent at the boundary between a @xmath1 and a @xmath2 coupled long jj ( ljj ) .
this supercurrent emerges in the absence of a driving bias current or an external field @xmath25 , i.e. in the ground state .
depending on the length of the junction @xmath26 the supercurrent carries one half of the flux quantum , i.e. @xmath27 ( called _ semifluxon _ _ ) , or less .
[ sketch_0pisifs](a ) depicts the cross section of a symmetric @xmath1@xmath2 _ long _ jj .
the spontaneous supercurrent @xmath28 flows either clockwise or counterclockwise , creating the magnetic field of @xmath29 .
the current density jumps from maximum positive to maximum negative value at the @xmath1@xmath2 phase boundary . a theoretical analysis based on the perturbed sine - gordon equation
is given in ref .
below we will first discuss the properties of the spontaneous supercurrent and , second , various systems having @xmath1@xmath2 phase boundaries . _
_ _ [ [ spontaneous - supercurrent ] ] spontaneous supercurrent + + + + + + + + + + + + + + + + + + + + + + + + kirtley _ et al . _
@xcite calculated the free energy of @xmath1@xmath2 jjs for various lengths of the @xmath1 and @xmath2 parts as a function of the normalized length @xmath30 and the degree of asymmetry @xmath31 , where @xmath32 are the critical current densities and @xmath33 are the lengths of @xmath1 and @xmath2 parts respectively , so that @xmath34 .
the state of a _ symmetric _ @xmath1@xmath2 junction ( @xmath35 ) with spontaneous flux has lower energy than the states @xmath12 or @xmath16 without flux .
symmetric @xmath1@xmath2 junctions have _ always _ some self - generated spontaneous flux , although its amplitude vanishes for @xmath36 as @xmath37 .
for example , a symmetric @xmath1@xmath2 jj of the total length @xmath38 has a spontaneous magnetic flux @xmath39 and a symmetric @xmath1@xmath2 jj with @xmath40 has a spontaneous flux of some @xmath41 below @xmath27 . only in case of a infinitely long jj we refer to the spontaneous flux as _ semifluxons _ , for shorter jjs it is named _
fractional vortex_. + the supercurrent or magnetic flux can be directly detected by measuring @xmath42 @xcite , by scanning squid ( superconducting quantum interference device ) microscopy ( in the ljj limit , see @xcite ) or by ltsem ( low temperature scanning electron microscopy ) @xcite .
_ _ [ [ pi - junctions - technology ] ] @xmath1@xmath2 junctions technology + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + @xmath1@xmath2 josephson junctions with a spontaneous flux in the ground state are realized with various technologies .
the presence of fractional vortex has been demonstrated experimentally in @xmath43-wave superconductor based ramp zigzag junctions @xcite , in long josephson @xmath1@xmath2 junctions fabricated using the conventional @xmath44 technology with a pair of current injectors @xcite , in the so - called tricrystal grain - boundary ljjs @xcite or in sfs / sifs jjs @xcite with _ stepped
_ ferromagnetic barrier as in fig .
[ sketch_0pisifs ] . in the latter systems
the josephson phase in the ground state is set to @xmath1 or @xmath2 by choosing proper f - layer thicknesses @xmath45 , @xmath46 for @xmath1 and @xmath2 parts , i.e. the amplitude of the critical current densities @xmath47 and @xmath48 can be controlled to some degree .
the advantages of this system are that it can be prepared in a multilayer geometry ( allowing topological flexibility ) and it can be easily combined with the well - developed @xmath49 technology . +
the starting point for estimation of the ground state of a _ stepped _
jj is studying the ivcs and @xmath42 for the _ planar _ reference @xmath1 and @xmath2 jjs . from this one
can calculate important parameters such as the critical current densities @xmath50 , @xmath51 , the josephson penetration depths @xmath52 , @xmath53 and the ratio of asymmetry @xmath54 . for
@xmath1@xmath2 junctions one needs @xmath1 and @xmath2 coupling in _
one _ junction , setting high demands on the fabrication process .
the ideal @xmath1@xmath2 jj would have equal @xmath55 and a @xmath1@xmath2 phase boundary in its center to have a symmetric situation .
furthermore the junctions should be underdamped ( sifs structure ) since low dissipation is necessary to study dynamics and eventually macroscopic quantum effects .
the junctions should have a high @xmath56 ( and hence small @xmath57 ) to reach the ljj limit and to keep high @xmath58 products , where @xmath59 is the characteristic voltage and @xmath60 the normal state resistance .
+ previous experimental works on @xmath1@xmath2 jjs based on sfs technology @xcite gave no information about @xmath47 and @xmath48 .
hence , the josephson penetration depth @xmath61 could not be calculated for these samples and the ratio of asymmetry @xmath54 was unknown .
the first intentionally made symmetric @xmath1@xmath2 tunnel jj of sifs type with a large @xmath59 was realized by the authors @xcite , making direct transport measurements of @xmath42 and calculation of the ground state with spontaneous flux feasible .
within this paper we _ review _ the physics of @xmath1@xmath2 coupled sifs - type josephson junctions and give an overview on our experimental results .
special focus is put on the fabrication of sifs junctions having a planar or stepped - typed ferromagnetic layer ( @xmath62 ) , the determination of ground state ( @xmath1 or @xmath2 for planar jjs ) and asymmetry of critical currents ( stepped jjs ) .
finally we give an estimation of the spontaneous magnetic flux in the ferromagnetic @xmath1@xmath2 jjs .
the fabrication process for _ planar _ junctions is based on @xmath63 stacks , deposited by dc magnetron sputtering @xcite .
thermally oxidized @xmath64-inch @xmath65 wafer served as substrate .
first of all , a @xmath66 nm thick @xmath67 bottom electrode and a @xmath68 nm thick @xmath69 layer were deposited .
second , the aluminium was oxidized for @xmath70 min at room temperature in a separate chamber .
third , the ferromagnet ( i.e. @xmath71 alloy , @xmath72 ) was deposited .
to have many structures with different thicknesses in one fabrication run , we decided to deposit a _ wedge_-shaped f - layer .
for this the substrate and sputter target were shifted about half of the substrate diameter .
this allowed the preparation of sifs junctions with a gradient in f - layer thickness in order to minimize inevitable run - to - run variations .
the sputtering rates for @xmath62 along the gradient were determined by thickness measurements on reference samples using a dektak profiler . at the end
a @xmath73 nm @xmath67 cap layer was deposited .
the tunnel junctions were patterned using a three level optical photolithographic mask procedure and @xmath74 ion - beam milling @xcite .
the insulation between top and bottom electrode is done by a self - aligned growth of @xmath75 insulator by anodic oxidation of @xmath67 after the ion - beam etching .
the @xmath75 exhibited a defect free insulation between the superconducting electrodes .
+ topological and electrical measurements , see ref .
@xcite , indicated that the direct deposition of @xmath62 on the tunnel barrier ( sifs - stacks ) led to an anomalous @xmath42 dependence such as shown in fig .
[ fig : ich_sinfs](a ) , which is an indication for an inhomogeneous current transport .
an additional @xmath76 thin @xmath77 layer between the @xmath78 tunnel barrier and the ferromagnetic @xmath62 ( sinfs - stacks ) brought considerable benefits , as it ensured a homogeneous current transport , see fig .
[ fig : ich_sinfs](b ) . in this way a high number of functioning devices with @xmath56 spreads less than 2% was obtained .
the variation of the f - layer thickness over a length of one junction diameter is less than @xmath79 .
for simplification we refer in the following to sifs stacks , although the actual multilayer is sinfs - type .
+ the patterning of _ stepped _
junctions was done after the complete deposition of the planar sifs stack and before the definition of the junction mesa by argon - etching and @xmath75 insulation .
the detailed process is published in ref .
the junction was partly protected with photoresist to define the step location in the f - layer , followed by i ) _ selective reactive etching _ of the @xmath67 , ii ) _ ion - etching _ of the @xmath62 by @xmath80 and iii ) subsequent _ in situ _ deposition of @xmath67 . to our knowledge , this was the first controlled patterning of @xmath1@xmath2 jjs based on a ferromagnetic interlayer .
+ the planar @xmath1 , @xmath2 reference junctions and the stepped @xmath1@xmath2 junctions were fabricated from a single trilayer .
all investigated junctions had an area of @xmath81 , but the length and width were different for different junctions .
the length was comparable or shorter than the josephson penetration depth @xmath61 .
we investigated the thickness dependence of the critical current @xmath20 . to produce the @xmath78 barrier the @xmath69 layer was oxidized at @xmath82 yielding @xmath83 for the reference superconductor - insulator - superconductor ( sis ) jjs .
then sifs stacks with wedge - like f - layer were fabricated in another run .
taking the jjs of the same geometry ( @xmath84 ) , but situated at different places on the wafer ( i.e. different @xmath85 ) we have measured the nonmonotonic @xmath20 dependence shown in fig .
[ fig : icdf ] . as a result
the fitted parameters are @xmath86 , @xmath87 and @xmath88 .
the coupling changed from @xmath1 to @xmath2 at the crossover thickness @xmath89 @xcite .
the magnetic and spin - orbit scattering in the f - layer mixes the up and down spin states of electrons in the conduction bands .
if the spin - flip scattering time @xmath90 is short @xmath91 , like in @xmath62 alloys , the temperature dependence of scattering provides the dominant mechanism for the @xmath19 dependence @xcite . the oscillation period @xmath92 becomes shorter for decreasing temperature , thus the whole @xmath20 dependence is squeezed to thinner f - layer thicknesses .
hence , the temperature dependence of the critical current @xmath19 is an interplay between an increasing component due to an increasing gap and a magnetic coupling dependent contribution which may de- or increase @xmath9 .
the @xmath19 relations for two jjs ( one @xmath1 , one @xmath2 ) are shown in the inset of fig .
[ fig : icdf ] . at @xmath93
the jj is @xmath1 coupled , but one can relate the nearly constant @xmath9 below @xmath94 to the interplay between an increasing gap and a decreasing oscillation length @xmath95 .
the @xmath96 jj is @xmath2 coupled and showed a linearly increasing @xmath9 with decreasing temperature .
various structures on the wafer were placed within a narrow ribbon perpendicular to the gradient in the f - layer thickness and were replicated along this gradient .
one ribbon contained reference jjs with the uniform f - layer thickness @xmath45 ( uniformly etched ) and @xmath46 ( non - etched ) and a jj with a step @xmath80 in the f - layer thickness from @xmath45 to @xmath46 .
the lengths @xmath97 and @xmath98 are both equal to @xmath99 .
the lithographic accuracy is of the order of @xmath100 .
a set of structures with difference in @xmath85 between neighboring ribbons of @xmath101 was obtained . comparing the critical currents @xmath9 of non - etched jjs ( dots ) , see fig .
[ fig : icdf_etch ] with the experimental @xmath20 data for the etched samples ( stars ) we estimate the etched - away f - layer thickness as @xmath102 .
the stars in fig .
[ fig : icdf_etch ] are shown already shifted by this amount .
now we choose the set of junctions which have the thickness @xmath46 and critical current @xmath103 ( @xmath2 junction ) before etching and have the thickness @xmath104 and critical current @xmath105 ( @xmath1 junction ) after etching .
one option is to choose the junction set denoted by closed circles around the data points in fig .
[ fig : icdf_etch ] , i.e. @xmath106 and @xmath107 .
the _ i - v _ characteristics and the magnetic field dependence of the critical current @xmath42 was measured for all three junctions : @xmath1 jj with @xmath108 , @xmath2 jj with @xmath109 and @xmath1@xmath2 jjs with stepped f - layer ( @xmath45 and @xmath46 in each half ) . the magnetic diffraction pattern @xmath42 of the @xmath1@xmath2 jj and the @xmath1 and @xmath2 reference jjs are plotted in fig . [ fig : ich-0-pi ] .
the magnetic field @xmath25 was applied in - plane of the sample and parallel to the step in the f - layer . due to a small net magnetization of the f
- layers the @xmath42 of references junctions were slightly shifted along the @xmath25 axis . nevertheless , both had the same oscillation period @xmath110 . at @xmath111
the @xmath1@xmath2 jjs was slightly asymmetric with @xmath112 and @xmath113 ( data of reference jjs ) . to achieve a more symmetric configuration ,
the bath temperature was reduced , because a decrease in temperature should increase @xmath114 more than @xmath115 , like for the @xmath1 and @xmath2 samples in the inset of fig .
[ fig : icdf ] . as a result
, both @xmath116 and @xmath117 were increasing when decreasing the temperature , but with different rates . at @xmath118 the critical currents @xmath119 and @xmath120 became approximately equal , see fig .
[ fig : ich-0-pi ] .
the magnetic field dependence of the planar reference junctions @xmath121 and @xmath122 look like perfect fraunhofer patterns .
one can see that the @xmath123 and @xmath124 measurements almost coincide , having the form of a symmetric fraunhofer pattern with the critical currents @xmath125 , @xmath126 and the same oscillation period .
the stepped @xmath1@xmath2 junction had a magnetic field dependence @xmath127 with a clear minimum near zero field and almost no asymmetry .
the critical currents at the left and right maxima ( @xmath128 and @xmath129 ) differ by less than @xmath130 , i.e. the @xmath1@xmath2 junction is symmetric , and its ground state in absence of a driving bias or magnetic field ( @xmath131 ) can be calculated @xcite .
our symmetric @xmath1@xmath2 ljj had an normalized length of @xmath132 , with a spontaneous flux in the ground state of@xmath133 being equal to @xmath134 of @xmath27 .
a detailed calculation taking several deviations from the ideal short jj model into account can be found elsewhere @xcite .
the concept and realization of @xmath1-@xmath2 junction based on sifs stacks has been presented .
the realization of @xmath2 coupling in sifs junctions and the precise combination of @xmath1 and @xmath2 coupled parts in a single junction has been shown .
the coupling of the ferromagnetic josephson tunnel junctions was investigated by means of transport measurements .
the emergence of a spontaneous flux , which was calculated as @xmath135 of half a flux quantum @xmath27 , was observed in the magnetic field dependence of the current - voltage characteristics of the @xmath1@xmath2 jj . + as an outlook , the ferromagnetic @xmath1@xmath2 josephson junctions allow to study the physics of fractional vortices with a good temperature control of the symmetry between @xmath1 and @xmath2 parts .
we note that symmetry is only needed for jj lengths @xmath136 . for longer jjs
the semifluxon appears even in rather asymmetric jjs , and @xmath137 can be varied in a wide range affecting the semifluxon properties only weakly .
the presented sifs technology allows us to construct @xmath1 , @xmath2 and @xmath1@xmath2 jjs with comparable @xmath50 and @xmath138 in a single fabrication run .
such jjs may be used to construct classical and quantum devices such as oscillators , memory cells , @xmath2 flux qubits @xcite or semifluxon based qubits @xcite .
this work was suportted by heraeus foundation and the deutsche forschungsgemeinschaft ( projects sfb / tr 21 ) . | we present a study on low-@xmath0 superconductor - insulator - ferromagnet - superconductor ( sifs ) josephson junctions .
sifs junctions have gained considerable interest in recent years because they show a number of interesting properties for future classical and quantum computing devices .
we optimized the fabrication process of these junctions to achieve a homogeneous current transport , ending up with high - quality samples .
depending on the thickness of the ferromagnetic layer and on temperature , the sifs junctions are in the ground state with a phase drop either @xmath1 or @xmath2 . by using a ferromagnetic layer with variable step - like thickness along the junction
, we obtained a so - called @xmath1-@xmath2 josephson junction , in which @xmath1 and @xmath2 ground states compete with each other . at a certain temperature
the @xmath1 and @xmath2 parts of the junction are perfectly symmetric , i.e. the absolute critical current densities are equal . in this case
the degenerate ground state corresponds to a vortex of supercurrent circulating clock- or counterclockwise and creating a magnetic flux which carries a fraction of the magnetic flux quantum @xmath3 . |
dna microarrays allow the comparison of the expression levels of all genes in an organism in a single experiment , which often involve different conditions ( _ i.e. _ health - illness , normal - stress ) , or different discrete time points ( _ i.e. _ cell cycle ) @xcite . among other applications ,
they provide clues about how genes interact with each other , which genes are part of the same metabolic pathway or which could be the possible role for those genes without a previously assigned function .
dna microarrays also have been used to obtain accurate disease classifications at the molecular level @xcite . however , transforming the huge amount of data produced by microarrays into useful knowledge has proven to be a difficult key step @xcite . on the other hand ,
clustering techniques have several applications , ranging from bioinformatics to economy @xcite .
particularly , data clustering is probably the most popular unsupervised technique for analyzing microarray data sets as a first approach .
many algorithms have been proposed , hierarchical clustering , k - means and self - organizing maps being the most known @xcite .
clustering consists of grouping items together based on a similarity measure in such a way that elements in a group must be more similar between them than between elements belonging to different groups .
the similarity measure definition , which quantifies the affinity between pairs of elements , introduces _ a priori _ information that determines the clustering solution .
therefore , this similarity measure could be optimized taking into account additional data acquired , for example , from real experiments . some works with
_ a priori _ inclusion of bioinformation in clustering models can be found in @xcite . in the case of gene expression clustering ,
the behavior of the genes reported by microarray experiments is represented as @xmath0 points in a @xmath1-dimensional space , being @xmath0 the total number of genes , and @xmath1 the number of conditions .
each gene behavior ( or point ) is then described by its coordinates ( its expression value for each condition ) .
genes whose expression pattern is similar will appear closer in the @xmath1-space , a characteristic that is used to classify data in groups . in our case , we have used the superparamagnetic clustering algorithm ( spc ) @xcite , which was proposed in 1996 by domany and collaborators as a new approach for grouping data sets .
however , this methodology has difficulties dealing with different density clusters , and in order to ameliorate this , we report here some modifications of the original algorithm that improve cluster detection .
our main contribution consists on increasing the similarity measure between genes by taking advantage of transcription factors , special proteins involved in the regulation of gene expression .
the present paper is organized as follows : in section 2 , the spc algorithm is introduced , as well as our proposal to include further biological information and our considerations for the selection of the most natural clusters .
results for a real data set , as well as performance comparisons , are presented in section 3 . finally , section 4 is dedicated to a summary of our results and conclusions .
a potts model can be used to simulate the collective behavior of a set of interacting sites using a statistical mechanics formalism . in the more general inhomogeneous potts model , the sites
are placed on an irregular lattice .
next , in the spc idea of domany _ et al . _
@xcite , each gene s expression pattern is represented as a site in an inhomogeneus potts model , whose coordinates are given by the microarray expression values . in this way , a particular lattice arrangement is spanned for the entire data set being analyzed .
a spin value @xmath2 , arbitrarily chosen from @xmath3 possibilities , is assigned to each site , where @xmath4 corresponds to the site of the lattice @xmath5 .
the main idea is to characterize the resulting spin configuration by the ferromagnetic hamiltonian : @xmath6 where the sum goes over all neighboring pairs , @xmath2 and @xmath7 are spin values of site @xmath4 and site @xmath8 respectively , and @xmath9 is their ferromagnetic interaction strength .
each site interacts only with its neighbors , however since the lattice is irregular , it is necessary to assign the set of nearest - neighbors of each site using the so - called @xmath10-mutual - nearest - neighbor criterion @xcite .
the original interaction strength is as follows : @xmath11 with @xmath12 the average number of neighbors per site and @xmath13 the average distance between neighbors .
the interaction strength between two neighboring sites decreases in a gaussian way with distance @xmath14 and therefore , sites that are separated by a small distance have more probability of sharing the same spin value during the simulation than the distant sites . on the other hand , said probability , @xmath15 , also depends on the temperature @xmath16 , which acts as a control parameter . at low temperatures ,
the sites tend to have the same spin values , forming a ferromagnetic system .
this configuration is preferred over others because it minimizes the total energy . however , the probability of encountering aligned spins diminishes as temperature increases , and the system could experience either a single transition to a totally disordered state ( paramagnetic phase ) , or pass through an intermediate phase in which the system is partially ordered , which is known as the superparamagnetic phase . in the latter case ,
varios regions of sites sharing the same spin value emerge .
sites within these regions interact among them with a stronger force , exhibiting at the same time weak interactions with sites outside the region .
these regions could fragment into smaller grains , leading to a chain of transitions within the superparamagnetic phase until the temperature is so high that the system enters the paramagnetic phase , where each spin behaves independently .
this hierarchical subdivision in magnetic grains reflects the organization of data into categories and subcategories .
regions of aligned spins emerging during simulation correspond to groups of points with similar coordinates , _
i.e. _ , similar gene expression patterns @xcite .
this subdivision can be simulated , for example , by using the monte carlo approach , by which one can compute and follow the evolution of system properties such as energy , magnetization and susceptibility , while the temperature is modified .
in addition , the temperature ranges in which each phase transition takes place can be localized . rather than thresholding the distances between pairs of sites to decide their assignment to clusters , the pair correlation @xmath17 , indicating a collective aspect of the data distribution , is preferred .
it can be calculated as follows @xcite @xmath18 in this way , @xmath17 is the normalized probability for finding two potts spins @xmath2 and @xmath7 sharing the same value for a given temperature step . if both spins belong to the same ordered region , their correlation value would be close to one , otherwise their correlation would be close to zero @xcite .
thus , for each temperature step , two sites are assigned to the same cluster if their correlation exceeds a threshold value of @xmath19 . if a site does not have a single correlation value greater than @xmath20 , it is joined with its neighbor showing the highest value .
+ for our spctf algorithm , we also accept sites whose @xmath17 are larger than @xmath20 in order to build a cluster .
however , differently from the traditional spc algorithm @xcite , if two sites do not reach the @xmath17 value greater than @xmath20 they are not connected . this is because with our data we have found that the original condition led to unnatural growth of some clusters when the temperature is increased .
as already mentioned , the data are fragmented in various clusters for each temperature value , and for higher temperatures , the number of clusters increases due to finer and finer segmentation . in order to select the more representative clusters through all temperature steps ,
we assign a stability value to each obtained cluster , based on its evolution .
we define @xmath21 as the number of temperature steps until the system reaches the paramagnetic phase and @xmath22 as the number of temperature steps a cluster @xmath23 survives , while @xmath24 and @xmath25 are defined as the total number of sites and the number of elements in a given cluster , respectively .
we assign a stability parameter @xmath26 to each cluster , as follows : @xmath27 where @xmath28 is the fraction of temperature steps a cluster @xmath23 survives , while @xmath29 is the fraction of total elements belonging to @xmath23 .
the advantage of using the stability parameter @xmath26 is that it gives preference to clusters that survive several temperatures , but also have an acceptable number of elements .
we added a small positive real number @xmath30 to the denominator in the expression of @xmath26 for the special case when @xmath31 , where @xmath32 belongs to the range @xmath33 $ ] , leading to @xmath34 instead of the infinity .
it has been reported that the main drawback of the spc algorithm consists of dealing with data showing regions of different density @xcite . in this case , either depending on temperature or the number of neighbors selected , some clusters will easily get prominent whereas the detection of others will be hindered . to overcome this problem ,
at least two techniques have been proposed _
e.g. _ , sequential superparamagnetic clustering @xcite and a modularity approach @xcite . our idea is to take advantage of already available biological information to improve lattice connectivity in such a way that biologically significant clusters have more probability of being detected by the algorithm .
indeed , at the transcriptional level , the expression of a gene could be promoted / suppressed by the binding of the proteins named transcription factors to specific sequences on the gene promoter region .
then , if a group of genes shows the same expression behavior in a microarray experiment , it is quite possible that they are being regulated by a specific transcription factor , forming a group of coregulated genes @xcite .
thus , available information about which genes are targeted by the same transcription factors may be useful in the detection of groups of genes with similar expression profiles . to make effective this idea ,
we downloaded from _ www.yeastract.com _ a list of yeast transcription factors that are well documented , and
whenever two neighboring genes are controlled by the same transcription factor , we increased their interaction strength .
it is important to note that the list provided by _ www.yeastract.com _
includes transcription factors associated with several processes and are not only cell cycle related .
the formula that takes this into account replaces eq .
( [ eq : js ] ) of the original algorithm , and has the following form : @xmath35 here , @xmath36 is the number of common transcription factors shared by @xmath4 and @xmath8 ( @xmath32 , which varies for each pair of neighboring genes ) , multiplied by a factor @xmath37 which was chosen to be 2.0 after comparing the results obtained with several other values .
the selected value has the characteristic of preserving well - defined susceptibility peaks as well as obtaining larger clusters .
the objective is to strengthen some connections without preventing the natural fragmentation of clusters caused by the temperature parameter . if two elements do not share a transcription factor , then @xmath38 , recovering the original spc formula
therefore , the modified interaction strength between each site and its neighbors is governed by two aspects : the distance between them , which comes from gene expression values generated through microarray experiments , and the number of transcription factors regulating both genes , obtained from documented biological data . any time
two genes share a transcription factor , their interaction strength becomes larger , and this favors that the clusters including these sites remain stable for longer temperature ranges , with the corresponding increase of their stability values .
we analyzed spellman _ et al . _
@xcite microarray data in which gene expression values from synchronized yeast cultures were obtained at various time moments , aiming to identify cell cycle genes .
yeast cultures were synchronized by three methods : adding alpha pheromone , which arrests cells in the g1 phase ; using centrifugal elutration for separating small g1 cells ; and using a mutation that arrests cells late in mitosis at a given temperature . combining the three experiments and using fourier and correlation algorithms , spellman _
@xcite reported @xmath39 cell cycle regulated genes .
the goal was to compare the performance of spc and spc with transcription factors ( spctf ) , which are algorithms that do not make assumptions about periodicity .
nonetheless , the overall analysis is time consuming and we only selected the data set treated with the alpha pheromone , available at _ http://cellcycle - www.stanford.edu_. genes with missing values were discarded , leaving an input matrix of @xmath40 genes and @xmath41 time courses that included only @xmath42 of the genes reported by spellman _
_ @xcite . furthermore ,
as we do not include the other two synchronization experiments , we expect to loose some of their cell cycle genes .
it is worth mentioning that getz _
_ @xcite also analyzed the spellman alpha synchronized set with the spc algorithm .
they took @xmath43 genes which have characterized functions and introduced a fourier transform to take into account the oscillatory nature of the cell cycle . in our case , however , we decided not to introduce any considerations about the periodicity of the data , mainly because the time series cover only two cell cycle periods @xcite .
we obtain compact gene clusters implementing spc original algorithm and spctf , both with parameter values @xmath44 and @xmath45 .
the cluster with the highest stability value contains an extremely large number of elements without a clear biological linkage between them .
it is mainly composed of genes whose expression do not change significantly over time , thus it is possible that they are included here for this very reason .
we discard this cluster from our analysis , although it could always be taken apart and analyzed again with spctf by choosing the appropiate number of neighbors to obtain more information .
to compare in more detail both approaches , it is necessary to correlate each cluster in the spc method with its equivalent in spctf . in order to do this , we calculate the euclidian distance between the mean position vector of every cluster in each approach , and choose the pairs with the shortest distance between them .
( we recall that the mean position vector of a cluster is obtained by averaging each coordinate between all its elements ) .
although different measures could have been used , this one performed adequately , as can be seen in the supplementary information file , where we provide a more detailed comparison between spctf and spc clusters . in table
[ tabla1 ] , we present the differences in cluster size as well as the hits , the number of genes reported by spellman _
@xcite , which have been included in the clusters .
when going through the spctf approach , one can see that the first largest cluster looses some genes , while the number of the rest of the clusters augments .
besides , hits or coincidences with spellman _ et al . _ @xcite cell cycle genes in clusters of six or more elements
increase by @xmath46 , from @xmath47 to @xmath48 .
therefore , we were able to incorporate several genes to these clusters , mainly from outliers . * comparison between spc and spctf * in the following analysis , we focus on clusters of six or more elements , because we are interested in finding groups of several genes sharing the same expression pattern ( coregulated genes ) . results of the comparison for the first @xmath49 most stable clusters , discarding the first one ,
are shown in fig .
[ fig : todos_cc ] . generally , these clusters incorporate more elements with spctf , including more cell cycle genes as those reported by spellman _
_ @xcite and thus improving the matching .
clusters , discarding the first one .
gray bars correspond to the clusters obtained with the spc algorithm and black bars to the equivalent clusters in spctf .
groups tend to increase in size and also in hits with cell cycle genes reported by spellman _
@xcite , with the exception of cluster @xmath50.,height=211 ] depending on the available information about the genes , we classify the clusters in three groups .
the first cluster type , cell cycle genes , cc , corresponds to groups formed in their majority ( @xmath51 ) by already reported cell cycle genes ( fig .
[ fig : cc_sp ] ) .
the second type , mixed genes , m , contains clusters with non - reported genes as well as already known cell cycle genes ( fig .
[ fig : m_n ] ) , and in the third type , no hits , n , we include the clusters that contain only one hit or are entirely composed of non - previously identified cell cycle genes ( fig .
[ fig : m_n ] ) .
it is worth mentioning that more cell cycle experiments have been done since spellman _
_ @xcite and new genes have been classified meanwhile as cell cycle regulated .
some of these newly reported cell cycle genes were obtained by cho _
@xcite , pramila _ et al . _
@xcite , rowicka _ et al .
_ @xcite and lichtenberg _ et al .
we analize our @xmath49 clusters taking now as hits , genes reported either by spellman _
@xcite or by one of the above mentioned studies . in this way
, we gained thirty additional hits in the spc clusters , while in spctf clusters we have fifty - two extra genes .
the results including all the aforementioned cell cycle studies are presented in figs .
[ fig : all_studies1][fig : m_n_all ] @xcite . most stable clusters .
hits are now taken as cell cycle genes reported by all studies .
gray bars correspond to the clusters obtained with the spc algorithm and black bars to the equivalent clusters in spctf.,height=211 ] in addition , we analyze the expression profiles of the genes conforming each cluster using the sceptrans tool @xcite , and we notice that all the genes grouped in the same cluster had the same expression pattern .
this gives us further confidence that our algorithm is grouping data correctly .
the expression profiles for a representative member of each cluster type are shown in fig .
[ fig : exp_pro ] .
we also find two clusters ( @xmath52 and @xmath49 ) that present an oscillating behaviour that is due to an artifact in the manner the microarray experiment was performed , see @xcite . in the supplementary information file
, we include the list of oscillating genes identified in @xcite and the number of these genes inside each of our first @xmath49 clusters .
we also include the expression profiles of these clusters as well as those of size @xmath53 and @xmath54 which contain hits with cell cycle genes identified by spellman _
these clusters have also similar expression profiles but were not further analyzed because of their low number of elements . in the case of gene annotation , it is important to have clusters of many elements to effectively assure that an unknown gene shares the biological function already assigned to the other genes in the same cluster .
+ + + the cc clusters are almost entirely composed of cell cycle regulated genes reported either by spellman _
@xcite or by other authors , besides , their expression patterns are similar , which leaves no doubt on their validity . for the m and
n clusters , we know that they are well grouped because their elements share the same expression patterns , but in order to select those of worth for further analysis ( for example in a laboratory experiment ) we analyze them through musa , motif finding using an unsupervised approach algorithm , that can be found at _
www.yeastract.com_. this program searches for the most common sequences ( motifs ) in the regulatory region of a set of genes , and compare them to the transcription factor binding sites already described in yeastract database @xcite .
results of this analysis are shown in table [ tabla2 ] , which includes the quorum or percentage of genes containing a motif in each cluster , and the alignment score , which quantifies the level of similarity between the encountered motif and the known transcription factor associated with it .
the clusters that probably would give us the best results would be those associated with cell cycle transcription factors with high percentages and scores .
we select in this way , the clusters @xmath55 , @xmath53 , @xmath56 , @xmath57 , @xmath58 and @xmath59 because they have percentages higher than @xmath60 and scores higher than @xmath61 . in order to validate the musa analysis
, we also constructed various clusters with sizes ranging from six to thirty - seven genes that were composed by genes selected at random from the original data . when analyzing these random clusters in the same way in musa , we obtain at most two cell cycle transcription factor coincidences .
* musa analysis *
large amounts of biological information are constantly obtained by throughput techniques and clustering algorithms have taken an important place in the unraveling of this information .
however , the clustering analyses offer a difficult challenge because any data set can be grouped in numerous ways , depending on the level of resolution asked for and the applied similarity measure . in this work ,
we propose the use of available biological information in order to strengthen the interaction between genes which share a transcription factor involved in any metabolic process , improving the similarity measure .
this information is introduced in the natural evolution of the spc algorithm , and in this way , we are able to enhance the creation and endurance of groups of possible coregulated genes . as the network spanned by the transcription factors information connects all genes , clustering directly _ a posteriori _ using only this information in the present case results into a single massive cluster ( see section iv of the supplementary information ) . however , by having the distance play an important weight in the interaction formula , the far - located clusters will not join , despite sharing transcription factors between their genes . with this in mind , we have modified the spc algorithm , and applied both the original and modified spctf algorithm to one of the three spellman _
@xcite data sets of the yeast cell cycle .
the expression profiles of the genes in all resulting clusters show a similar behavior , but we obtain larger clusters with spctf .
we classified them in three types , cc , m , and n , depending on the amount of cell cycle reported elements inside each cluster .
with spctf , the cc type clusters increase in size including more cell cycle genes , and for the m and n type clusters , we also looked for common sequences in its regulatory regions and selected various groups worth of further research in order to report possible new cell cycle genes . as expected , some of these clusters include already known cell cycle genes sharing a transcription factor , _ but more importantly , at the predictive level , they promote the inclusion of new genes with similar expression patterns_. it is also important to note that the modified algorithm can be applied to any data set , and the followed methodology leads to the selection of the potential gene subsets feasible to be experimentally investigated .
our work can serve as an example of how the inclusion of available biological information , such as transcription factors , and bioinformatic tools , such as musa , can lead to better and more confident results , aiding in the analysis of data coming from microarray experiments .
the authors thank drs . s. ahnert and g. sherlock for useful discussions and comments .
we also thank conacyt for providing support for two of the authors ( m.p.m.a . and j.c.n.m . ) and
also the referees for helpful remarks and information . this work was partly supported through the project sep - conacyt-2005 - 49039 .
p. t. monteiro , n. d. mendes , m. c. teixeira , s. dorey , s. tenreiro , n. p. mira , h. pais , a. p. francisco , a. m. carvalho , a. b. lourenco , i. sa - correia , a. l. oliveira , and a. t. freitas , nucleic acids res . * 36 * , d132 ( 2008 ) . | in this work , we modify the superparamagnetic clustering algorithm ( spc ) by adding an extra weight to the interaction formula that considers which genes are regulated by the same transcription factor . with this modified algorithm that we call spctf , we analyze spellman _ et al .
_ microarray data for cell cycle genes in yeast , and find clusters with a higher number of elements compared with those obtained with the spc algorithm .
some of the incorporated genes by using spcft were not detected at first by spellman _
et al .
_ but were later identified by other studies , whereas several genes still remain unclassified .
the clusters composed by unidentified genes were analyzed with musa , the motif finding using an unsupervised approach algorithm , and this allow us to select the clusters whose elements contain cell cycle transcription factor binding sites as clusters worth of further experimental studies because they would probably lead to new cell cycle genes .
finally , our idea of introducing available information about transcription factors to optimize the gene classification could be implemented for other distance - based clustering algorithms .
superparamagnetic clustering , similarity measure , microarrays , cell cycle genes , transcription factors .
+ paper - physa-3 20100901.tex physica a 389(24 ) , 5689 - 5697 ( 2010 ) + doi : 10.1016/j.physa.2010.09.006 |
one of the assumptions underpinning the conventional navier - stokes - fourier set of equations is that of local thermodynamic equilibrium . this assumption allows the representation of thermodynamic variables ( e.g. temperature , density , pressure ) as locally constant at a given time and position , and the use of equations of state .
the assumption that microscopic relaxation processes are not of concern is , however , inadequate in flows where the microscopic relaxation time is comparable to the characteristic time of evolution of the macroscopic field variables . in the kinetic theory of dilute gases ,
such flows are identified with high knudsen numbers ( conventionally defined as a ratio of the average time between molecule / molecule collisions to a macroscopic characteristic time of the flow , however see @xcite ) .
experimental observations of sound wave propagation at high knudsen number challenge many continuum hydrodynamics and kinetic theory models @xcite ; it is well - known that the navier - stokes - fourier model fails to predict sound wave propagation at high knudsen number .
another problem arises in the so - called `` heat conduction paradox '' , according to which an unphysical infinite speed of thermal wave propagation is predicted by the energy equation closed with fourier s law .
generally , techniques for investigating gas flows in which the navier - stokes - fourier model is inadequate are based on approximate solutions to the boltzmann dilute gas kinetic equation , for which a wide number of mathematical methods are found in the literature @xcite .
regarding the specific problem of predicting sound wave propagation in monatomic gases in the high knudsen number regime , many of these boltzmann based approximations fail , as does navier - stokes - fourier @xcite .
while a few have shown some agreement with experiments @xcite , detailed analysis makes any conclusion far from clear - cut @xcite .
for example , if the experimental set - up is configured to measure propagations of plane harmonic waves @xcite , boltzmann kinetic models predict unconventional pressure fields , even though the phase speeds and damping coefficients do agree with the experimental data @xcite .
recently developed continuum models also show discrepancies in these predictions , particularly in the damping @xcite .
the unphysical predictions of the conventional navier - stokes - fourier model have been investigated in terms of the `` heat conduction paradox '' .
early investigations criticized the expression of fourier s law , suggesting instead that the heat flux expression should be transformed from the parabolic form of the heat conduction equation to a simple hyperbolic equation with a finite speed of propagation . while the original demonstration by cattaneo @xcite has a flaw @xcite , a cattaneo - vermot heat flux has been formalized more elegantly using fading memory theory ( which essentially aims to remove the local equilibrium assumption ) .
variants and generalizations have been proposed , and compatibility with the second law of thermodynamics has been assessed @xcite
. however , these investigations concentrate on modifications to the simple heat conduction equation ; they are not , to our knowledge , developed within the framework of complete fluid dynamic equations and a full dispersion analysis . in this paper
we investigate hydrodynamic models in which the assumptions limiting the application of the conventional navier - stokes - fourier model are clearly released ; this is therefore outside the framework of pure approximation solutions to the boltzmann kinetic equation . in previous work , we proposed releasing the local equilibrium assumption by including the spatial distributions of molecules within the kinetic description @xcite .
while our description was motivated by an unusual volume diffusion claimed by brenner @xcite , it has been recently pointed out that the original brenner modification does not predict sound wave speeds correctly @xcite .
here we show that our volume - based hydrodynamic model can reproduce the experimental sound wave propagation data from ref .
@xcite with excellent agreement .
moreover , our model offers a more reliable explanation of the experiments , which were designed to range up to the free molecular regime in which there are no collisions between molecules and therefore the definition of sound as a pressure wave becomes problematic .
this paper starts with a summary of our volume model that incorporates effects from microscopic spatial distributions of the gaseous molecules .
subsequently , a linear stability analysis of the model equations is performed , and the predicted dispersion and damping compared with experiments .
the traditional single particle distribution function used in the boltzmann kinetic equation for a monatomic gas attributes no particular importance to the spatial arrangements of molecules .
an average number of molecules is associated with a position @xmath0 and a velocity @xmath1 . in order to account for microscopic spatial fluctuations , due to non - uniformity in molecular spatial configurations ,
we have considered within the set of microscopic variables the microscopic free volume , @xmath2 , around each gaseous molecule .
a single particle distribution function @xmath3 is then defined to describe the probability that a molecule at a given time @xmath4 is located in the vicinity of position @xmath0 , has its velocity in the vicinity of @xmath1 , and has around it a microscopic free space given by the additional variable @xmath2 . a boltzmann - like kinetic equation for @xmath5
is then derived as @xcite : @xmath6 in which the term on the right - hand - side is the hard sphere molecule collision integral ; @xmath7 and @xmath8 refer to post - collision molecules , @xmath9 and @xmath10 refer to pre - collision molecules , @xmath11 is the molecule relative velocity , @xmath12 the collision differential cross section , @xmath13 an element of solid angle . on the left - hand - side appears a new term involving @xmath14 , which arises primarily from the introduction of the new variable @xmath2 into the distribution function . in the derivation of equation ( [ eq.boltzmann.kok ] ) , molecular exchanges of momentum through interactions have been assumed to be independent of their spatial configurations .
three contributions to the time variations of @xmath15 are seen within equation ( [ eq.boltzmann.kok ] ) .
molecular free - stream motions are given by the second term on the left - hand - side .
the third term on the left - hand - side arises from effects of molecular interactions on their spatial distributions . finally , the collision integral is the traditional momentum exchange between molecules that provides changes in molecular velocities .
these latter two terms infer that the real molecular potential interactions are represented in this kinetic model by two separate actions : intermolecular force effects on spatial distributions , and collisional effects on molecular velocities . as @xmath16
is defined as a probability density function , we have a normalization factor , @xmath17 the mean value , @xmath18 , of a gas property @xmath19 is then defined by , @xmath20 the local average of @xmath2 is therefore the local mean free volume @xmath21 around each gaseous molecule , i.e. @xmath22 from this mean value of the volume around a molecule , we define the mass - density in the vicinity of position @xmath0 through : @xmath23 where @xmath24 is the molecular mass .
two mean velocities are defined using two different weighting values : the local mean mass - velocity , @xmath25 , is given through @xmath26 and a local mean volume - velocity , @xmath27 , by using the microscopic free volume as the weighting , @xmath28 the two definitions @xmath29 and @xmath30 coincide if @xmath2 is a constant , i.e. in a homogeneous medium where density is constant throughout .
it can be shown that the difference between these two velocities , @xmath31 , behaves like a mass - density diffusion @xcite .
hydrodynamic equations are derived as conservation equations obtained from the kinetic equation , accounting for a reclassification of convective / diffusive fluxes required by the appearance of the two different velocities .
the set of equations is obtained @xcite : continuity : : @xmath32 mass - density : : @xmath33 + a_nw , \end{aligned}\ ] ] momentum : : @xmath34 energy : : @xmath35 & = - \nabla \cdot a_n \left [ \left(\mathbf{p ' } - \frac{1}{\bar{v}^2}\mathbf{j}_v \mathbf{j}_v \right)\cdot u_m\right ] \\
\nonumber & - \nabla \cdot a_n \left[\mathbf{q ' } + \frac{1}{\bar{v}}\mathbf{p ' } \cdot \mathbf{j}_v + \frac{1}{\bar{v}}\left ( e'_{in } - \frac{1}{\bar{v}^2}\mathbf{j}_v^2\right ) \mathbf{j}_v\ \right ] \
.\end{aligned}\ ] ] where we denote the material derivative @xmath36 .
the flow variables are : the probability density @xmath37 ( which is , however , not a physical property ) , the mass - density @xmath38 , the mass - velocity @xmath30 , and the internal energy @xmath39 .
following , provisionally , the classical phenomenological fick s law for a diffusive flux , the model may be closed by the constitutive relations : @xmath40 in which we have defined @xmath41 with @xmath42 being the kinetic temperature , or @xmath43 with @xmath44 being the kinetic pressure , and @xmath45 .
the coefficients @xmath46 , @xmath47 , @xmath48 and @xmath49 are , respectively , dynamic viscosity , heat conductivity , bulk viscosity , and the mass - density diffusion coefficient .
as the kinetic pressure @xmath44 is defined by the trace of the pressure tensor we also have @xmath50 .
previous volume diffusion hydrodynamic models have been based on separating the mean velocity in the conventional mass conservation equation ( continuity equation ) , from the mean velocity in the navier - stokes momentum equation via newton s viscosity law @xcite .
this has proven controversial @xcite problems in differentiating the mass - flux from the momentum density , and in conserving angular momentum when the velocity on the left - hand - side of the navier - stokes equation is substituted for , have been raised . in our approach , however , a mass flux is given by @xmath51 from the mass - density equation ( [ densitynewhydro ] ) , and involves the same velocity , @xmath52 , as in newton s viscosity law ( equation [ fluxes - new ] ) .
meanwhile , the velocity on the left - hand - side of the new momentum equation ( [ momentumnewhydro ] ) remains the conventional mass velocity @xmath30 ( following newton s second law ) .
consequently the two flaws mentioned in connection with volume - based hydrodynamics in reference @xcite are not present in our set of equations ( [ massnewhydro])([fluxes - new - end ] ) .
a consequence of our localized microscopic volume description is the appearance of @xmath14 , the time rate of change of microscopic volume .
although this term could be proposed using details of the interactions between particles , here we instead test a phenomenological expansion of @xmath53 as a function of the fluid macroscopic thermodynamic variables .
first we relate variations of the microscopic @xmath2 to variation of its macroscopic average @xmath54 , through a relaxation approximation : @xmath55 the derivative @xmath56 refers to the time rate of change of microscopic properties while @xmath57 refers to the time rate of change of macroscopic properties , with @xmath58 a relaxation time . expanding @xmath59 as a function of thermodynamic variables we have : @xmath60 where @xmath61 , @xmath62 , @xmath63 , @xmath64 are the gas expansion and compressibility coefficients given by , @xmath65 and @xmath66 in our description local thermodynamic equilibrium is not required
. relations @xmath41 and @xmath43 define the temperature and pressure ( following their classical definitions in kinetic theory ) , therefore there is a reciprocal relation between temperature and pressure , @xmath67 , by construction without further assumption . if the perfect gas ( equilibrium ) equation of state is enforced , and we confuse @xmath68 with @xmath69 in equation ( [ expres_w ] ) , then the gas expansion and compressibility coefficients in equations ( [ expres - alpha - chi ] ) are the ideal gas coefficients , i.e. @xmath70 and @xmath71 , and the second order contributions vanish from equation ( [ expres_w ] ) . but as we are not restricting ourselves to local thermodynamic equilibrium , a departure from these ideal coefficients may be expected .
now we turn to investigate sound dispersion using both the first and the second order approximations to @xmath14 given in equation ( [ expres_w ] ) .
we consider our hydrodynamic model in a one - dimensional flow configuration .
an equilibrium ground state is defined by the flow variables @xmath72 , @xmath73 , @xmath74 , @xmath75 , u@xmath76 , with @xmath77 the specific gas constant .
then a perturbation from this ground state is introduced as follows : @xmath78 where the asterisked variables represent dimensionless quantities .
the perturbation of the volume velocity is specified through the relationship @xmath79 .
linearizing @xmath67 gives @xmath80 .
the dimensionless space and time variables are given by , @xmath81 with @xmath82 .
the dimensionless linearized equations , including the general expression for @xmath14 in equation ( [ expres_w ] ) , can therefore be written : continuity : : @xmath83 mass - density : : @xmath84 momentum : : @xmath85 energy : : @xmath86 where the different dimensionless transport coefficients are given through : @xmath87 and @xmath88 note that the dimensionless transport coefficients in equations ( [ transpo - coeff - dimens ] ) follow from the dimensionless form of the hydrodynamic set of equations .
instead of using these dimensionless coefficients , however , it may be more convenient to use conventional parameters , i.e. the knudsen number @xmath89 , the prandtl number @xmath90 , and an additional parameter @xmath91 that involves the mass - density diffusivity .
these are given by ( denoting @xmath92 ) : @xmath93 we assume the disturbances @xmath94 , @xmath95 , @xmath96 and @xmath97 to be wave functions of the form : @xmath98 , \label{harmonic - wrong}\ ] ] where @xmath99 is the complex wave frequency , @xmath100 is the complex wave number , and @xmath101 is the complex amplitude , so that : @xmath102 the linearized hydrodynamic set of equations then yields the homogeneous system , @xmath103 where @xmath104 the corresponding dispersion relation , obtained when the determinant of @xmath105 is zero , is @xmath106\times & \\
\nonumber \left[-\gamma^ * \omega^2+i \left(1-\chi^*\right ) \omega+\frac{k^2}{s_c}\right]-\left[\left(\beta^*-\gamma^*\right ) \omega^2+i \left(\alpha^ * -\chi^*\right ) w\right ] \times & \\
\left[-\frac{4 i \omega
k_n k^4}{3 s_c}-\frac{5 k^4}{3 s_c}+\frac{5 \omega^2 k^2}{3 s_c}\right ] & = 0 .
\end{aligned}\ ] ] when analyzing the dispersion and stability characteristics of our model , we compare our results for sound propagation in argon gas with experimental data from reference @xcite . choosing the harmonic wave expression ( [ harmonic - wrong ] )
is in line with previous analysis of this problem , and the dimensionless phase speed @xmath107 , and dimensionless spatial damping @xmath108 , are then commonly defined by @xcite : @xmath109}{\omega } , \ \
\lambda_l = - \sqrt{\frac{5}{3}}\frac{im[k]}{\omega } \ .\end{aligned}\ ] ] setting the knudsen number @xmath89 , defined in equation ( [ dimensionless - number ] ) , to @xmath110 makes our analysis agree with that of greenspan @xcite , in which variations of frequency @xmath99 are interpreted as variations of knudsen number ( the limitations of this particular interpretation are outlined in the appendix to this present paper ) .
although more recent experimental data with a different analysis exists , we choose this approach first in order to make comparisons with previously published works @xcite .
we also note here that a solution to a dispersion relation such as equation ( [ dispersion ] ) consists of various discontinuous solutions generating a number of modes ; one of these is expected to correspond to the sound mode . in this paper
, we include in our results figures all modes , for the sake of a complete analysis .
linear stability criteria are as follows @xcite : for the set of equations to be time stable , @xmath111 as a root of the dispersion relation ( [ dispersion ] ) should satisfy @xmath112\geq 0 $ ] for all @xmath100 real . on the other hand , the set of equations will be stable in space if @xmath113 as a root of the dispersion relation satisfies @xmath114\times re[k(\omega ) ] < 0 $ ] for all @xmath115 .
first we set @xmath116 , that is , @xmath14 is approximated only by the first order terms in equation ( [ expres_w ] ) .
for @xmath117 the dispersion and stability characteristics of our model correspond to those of the navier - stokes - fourier model .
the equations are also stable in both time and space .
figure [ navier ] shows both the inverse phase speed and the damping as a function of inverse frequency ( i.e. inverse knudsen number ) , compared with experimental data @xcite .
navier - stokes - fourier has only two modes : one mode fits the phase speed and damping measurements at low knudsen number , but has an infinite speed of propagation for high knudsen number .
the second mode shows an infinite inverse phase speed at low knudsen number , and is interpreted as the heat mode @xcite .
+ departures from these predictions are expected for our volume - based hydrodynamic model when @xmath118 .
we find that the model is stable , in the case of a first order approximation to @xmath14 , if @xmath119 and @xmath120 are both simultaneously smaller than one , or @xmath121 and @xmath122 , approximately ; this is illustrated in figure [ stab - navier - corrected ] .
comparison of the dispersion with experiments shows globally the same results as in the navier - stokes - fourier case .
but , as seen in figure [ navier - corrected ] where we have @xmath123 , @xmath124 and @xmath125 , the agreement with the low frequency regime is improved , particularly in the damping coefficient . both the phase speed and the damping are adequately predicted up to @xmath126 , whereas the damping was predicted only up to @xmath127 by navier - stokes - fourier alone ( figure [ damp - navier ] ) .
figure [ navier - corrected ] also shows that there are now three modes , two of which display transient diffusion behaviour ( i.e.high damping in low frequency regimes ) .
while one of these should be considered as the heat mode , as previously , the other should be attributed to transient mass - density diffusion , as introduced by our new volume - based description ( in addition to the heat diffusion ) .
this new mode is the most affected by the mass - density diffusivity , i.e. , by @xmath91 .
the high frequency regime is still incorrectly predicted by the sound mode , as in the case of navier - stokes - fourier .
later we will see that the infinite speed of propagation and zero damping in the high frequency regime can all be removed with the inclusion of the new mass - density mode .
+ + now we set , @xmath128 , that is , @xmath14 is given by an expression with only the second order terms of equation ( [ expres_w ] ) . in this case , we observe that the set of volume - based equations has a wider range of stability , provided @xmath129 approximately ( see figure [ stab - kokou - second ] ) .
figure [ kokou - second ] shows that the phase speed prediction of one of the modes now agrees perfectly with experiment , in both the low and the high frequency regimes .
this mode actually corresponds to the pressure mode , and it merges into the new mass - density mode in the high frequency regime . for comparison , in figure [ comparedother ] this physical mode is plotted with the experimental data and results from two recent continuum models derived as approximation solutions to the boltzmann equation @xcite .
we observe that our volume model is competitive with the best of these two models .
our new model has the best damping coefficient predictions in the low knudsen number regime , and we note an unphysical negative damping coefficient predicted by the second order model of spiegel and thiffeault @xcite .
+ + + in our investigations , our choice of the values of different coefficients in the volume model has been primarily motivated by finding the best agreement with the experimental data
. however , coefficient @xmath91 , set to @xmath130 for figure [ navier - corrected ] , agrees with an interpretation of @xmath91 as a schmidt number with a value of @xmath131 for monatomic hard sphere molecular gases ; a value of @xmath132 has been used for the dispersion analysis in reference @xcite .
while the stability depends on the expression of @xmath14 , our volume - based set of equations seems to remain stable for whatever value the schmidt number is set to , i.e. , whatever the mass - density diffusivity .
the dimensionless expansion and compressibility coefficients we obtained depart from their ( equilibrium state ) ideal gas values of @xmath110 .
these departures from ideality may be attributable to real gas effects now incorporated in our volume - based description .
similar results to those presented in our figures are also obtained with other combinations of the various coefficients .
for example , @xmath133 , @xmath134 and @xmath135 give the same results as in figure [ navier - corrected ] .
this recalls experimental reports that different gases can produce similar results @xcite .
in any case , the various coefficients in our volume model leave room to incorporate the various properties of the gas under investigation . in figures [ damp - navier ] , [ damp - navier - corrected ] and [ damp - kokou ] , the predicted damping coefficient tends to zero as the knudsen number becomes large .
this is a very common result when using continuum models , as seen on figure [ comparedother ] .
problems have also been pointed out in comparisons with experiments in this regime @xcite .
therefore , researchers have argued on the basis of spectral analysis that continuum models based on a finite set of partial differential equations can not capture this branch of the graph @xcite . in any case , interpreting sound waves in terms of pressure waves and momentum exchanges between ( only ) molecules during collisions should be expected to lead to vanishing damping as intermolecular collisions are no longer the dominant phenomena in the very high knudsen number regime @xcite .
we now consider earlier comments by some investigators @xcite who , analyzing the experimental set - up , suggested that a model to predict this sound dispersion must have a knudsen number expression and a dimensional analysis that reflects the distinction between the molecule / molecule collision - dominated regime and the molecule / surface collision - dominated regime . in the experimental set - up
the gas was placed between source and receiver then disturbed by a plane harmonic sound wave with a fixed frequency at the source @xcite .
the primary variable parameter in the experiments was the distance between the source and the receiver .
at very low pressures , the molecule / molecule collisions that predominate in a high pressure ( or continuum ) regime , become negligible , and molecular collisions with surfaces dominate . in this situation ,
the microscopic collision length scale becomes the distance traveled by molecules to reach the surfaces no longer the mean free path that is the length scale in the continuum regime .
accordingly , schotter @xcite , who also reported similar data to greenspan , meyer and sessler , presents a different dimensional analysis , introducing two different microscopic times leading to two different knudsen number expressions .
the first of these corresponds to a pressure - based intermolecular collision time , and is the same definition as in references @xcite .
the second microscopic time is independent of molecule / molecule momentum transfers and instead characterizes the frequency of collisions with the surfaces .
as we show explicitly in the appendix , greenspan s dimensionless quantities in equation ( [ damp - speed - expre ] ) , and the accompanying interpretation of frequency as a ( conventional ) knudsen number , are founded on molecule / molecule collisions and so become inappropriate at high knudsen number where these types of collisions are no longer the principal momentum transfer mechanism ( see also reference @xcite ) . a dimensional analysis using the separation distance between the surfaces leads to a different expression for the dimensional damping coefficient in a low pressure gas , which is also , conversely , invalid for high pressure cases ( i.e. at low conventional knudsen number ) .
this second expression may also be derived using the following observation . in section [ sec - linearequation ]
we performed a dimensional analysis , and introduced equation ( [ harmonic - wrong ] ) which assumes the harmonic wave form .
as the set of partial differential equations is linearized and dimensionless , characteristic time and length scales have therefore been introduced before equation ( [ harmonic - wrong ] ) . a better way of expressing the harmonic wave is in a completely dimensionless form , i.e. , @xmath136 , \label{harmonic - correct}\ ] ] where @xmath137 and @xmath138 are , respectively , the dimensionless complex wave frequency and dimensionless wave number .
moreover , @xmath139 and @xmath140 , with @xmath141 and @xmath142 the characteristic time and length previously introduced in equation ( [ micro - timedefo ] ) .
the constant coefficient @xmath143 , from the adiabatic exponent of a monatomic gas , could be simply incorporated in the definition of the reference speed and is not here the main issue . the dimensionless phase speed and dimensionless spatial damping coefficient
are therefore : @xmath144}{\omega^ * } , \ \
\lambda_h = - \sqrt{\frac{5}{3}}im[k^ * ] \ , \end{aligned}\ ] ] and we observe that while the dimensionless phase speed remains the same as previously , the dimensionless damping coefficient is different ( see equation [ damp - speed - expre ] ) : it does not contain the frequency
. described by a second order approximation , and using the definitions in equation ( [ damp - speed - expre - correct ] ) ; @xmath145 , @xmath146 , @xmath147 .
note the agreement with one of the modes at high knudsen numbers.,scaledwidth=80.0%,scaledwidth=40.0% ] in figure [ paradoxical_damp ] we plot the dimensionless damping coefficient by our new hydrodynamic model , but using the redefined expressions in equation ( [ damp - speed - expre - correct ] ) ( and using same coefficients @xmath91 , @xmath148 and @xmath149 as in figure [ kokou - second ] ) .
it is seen that our model reproduces the high frequency branch , with the correct asymptotic value of the damping .
in addition , this is represented by the new mass - density mode , not the classical pressure mode which instead diverges .
broadly , this curve catches the shape and the shallow maximum around @xmath150 .
the agreement is not so good by @xmath126 , and becomes somewhat inaccurate for low knudsen numbers , as expected . in summary , expressions ( [ damp - speed - expre ] ) and ( [ damp - speed - expre - correct ] ) are each compatible with different knudsen number regimes and are both required for a proper interpretation of the experimental results .
our volume - based hydrodynamic model has been shown , therefore , to predict both the low and the high frequency branch of the damping coefficient well , while the inverse phase speed is always well - predicted . in his experiments ,
schotter @xcite reported plane standing waves for all knudsen numbers . because of difficulties surrounding the predictions of the high knudsen number branch , other researchers assumed , however , that a plane wave analysis could not capture this regime @xcite . in our analysis ,
mass - density and pressure fields are plane harmonic and therefore agree also with schotter s experimental observation .
we also confirm the unusual ( i.e. non - pressure - wave ) characteristics of sound waves in this regime because our good predictions here are provided by our model s mass - density diffusion terms .
this is illustrated in figure [ dampingkokouall ] , where the two different modes fitting the experimental damping data in the low and the high frequency regimes are both plotted .
finally , even with the modified definitions of equation ( [ damp - speed - expre - correct ] ) , the navier - stokes - fourier model gives at @xmath151 a value of the damping which is @xmath152 times the experimental value of approximately @xmath153 .
so the conventional model still provides incorrect predictions .
predicting sound wave phase speed and damping is a challenge both for kinetic models derived from the boltzmann dilute gas equation and for continuum fluid hydrodynamics @xcite .
the few kinetic models @xcite that agree with the experimental data over the entire range of knudsen number suffer three major criticisms .
first , questions often arise about the compatibility of kinetic boundary value problems with experimental measurement @xcite .
second , the kinetic models predict non - standard pressure fields @xcite ; in contrast , experiments have been based on harmonic pressure waves , and indicate a plane standing wave existing in the gas medium at all knudsen numbers during measurement .
third , the different mechanisms of momentum transfer in the high pressure and the low pressure cases are not always compatible with the kinetic model predictions @xcite .
a final issue , often raised with continuum fluid models beyond navier - stokes - fourier , is the appearance of a large number of modes so it is not always easy to identify the mode that should describe the sound wave .
our figures [ kokou - second ] and [ paradoxical_damp ] show that the continuum - based model considered in this paper reproduces the experiments over the range of knudsen number without the difficulties listed above . in these figures
there are only three distinct modes to be associated with pressure , temperature and mass - density in a given regime . in our understanding ,
pressure and mass - density disturbances are distinct plane harmonic waves that dominate in different knudsen number regimes ( see figure [ dampingkokouall ] ) .
the existence of a mass - density wave explains the plane standing wave observed in experiments in the high knudsen number regime ; this mode is non - existent in conventional fluid dynamic equations as there is no explicit mass - density diffusion ( or mass - density wave propagation ) .
the agreement between our theoretical damping results and experiment can be fully explained in terms of two mean - free - paths inherent in the experimental set - up ; one mean free path is founded on the standard kinetic pressure and molecular collisions , and the other founded on the separation distances of the solid surfaces .
the latter also underlines the fundamental basis of our new approach itself : the variation of the surface position is easily associated with variation of the volume between molecules .
our prediction of the high knudsen number regime is possible only if we adopt the second - order expression for @xmath14 given in equation ( [ expres_w ] ) .
this shows that this regime is best described by microscopic structure evolutions , and not macroscopic average thermodynamic property evolutions ; therefore there is no localized thermodynamic equilibrium in this case .
indeed , in equation ( [ approx - property - fun ] ) the time rate of change of the microscopic volume @xmath2 is represented by the sum of the time rate of change of the average value @xmath54 and the change in its random component , which is approximated using a relaxation time .
consequently , the second - order terms involved in equation ( [ expres_w ] ) can be considered expressions of the random component of the microscopic volume evolutions .
( a representation of microscopic structure , as in equation ( [ approx - property - fun ] ) , is common in `` fading memory '' concepts , where it is given generally as a convolution function @xcite . )
the starting point of our volume - based hydrodynamic model is the representation of the fluid mass - density within conventional continuum fluid mechanics and kinetic theory @xcite . in this paper
, we have seen that a volume - modified hydrodynamic model can achieve surprisingly good results for sound wave dispersion in monatomic gases .
this problematic gas flow in the non - continuum regime has previously been classified as non - predictable using a continuum - based description .
moreover , our volume - based hydrodynamics offers a more plausible interpretation of the experimental data than some previous kinetic results .
we therefore propose the volume - based model for further investigations .
first , more sophisticated constructions of the new volume variation terms involved in the description are required , as results suggest some sensitivities to their formulation .
second , further application should be made to other flows and heat transfer configurations where the classical continuum models become inadequate .
for example , investigating heat transfer in the transition regime , where the dependency of heat conductivity on the knudsen number or pressure , and the definition of heat flux , are still unresolved problems @xcite .
this work is funded in the uk by the engineering and physical sciences research council under grant ep / d007488/1 , and through a royal society of edinburgh / scottish government support research fellowship for jmr .
the authors would like to thank the referees of this paper for their helpful comments .
s. k. dadzie , j. m. reese , the concept of mass - density in classical thermodynamics and the boltzmann kinetic equation for dilute gases , in : rarefied gas dynamics : 23rd international symposium , vol . 1084 , 2008 , p. 117
[ [ analysis - of - greenspans - interpretation - of - knudsen - number - variations ] ] analysis of greenspan s interpretation of knudsen number variations ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ this is a boundary value problem , with @xmath154 positive real , and @xmath155 a complex number . a plane harmonic wave @xmath156 is written with dimensional variables as @xmath157 .
\label{app - harmonic1}\ ] ] we seek dimensionless expressions for the phase speed and damping . first , equation ( [ app - harmonic1 ] ) is rewritten , @xmath158 \exp\left[\left(\frac{k_i}{\omega}\right ) \omega x\right ] .
\label{app - harmonic2}\ ] ] the experimental set - up infers a fixed frequency , @xmath159 @xcite .
suppose that the gas has well - defined microscopic time and length scales , @xmath141 and @xmath142 , respectively , which therefore specify a microscopic speed @xmath160 .
we may then define dimensionless frequency , time and length as @xmath161 using these definitions , equation ( [ app - harmonic2 ] ) becomes , @xmath162 \exp\left[c_0\frac{k_i}{\omega } \omega^*\omega_e \tau x^*\right ] .
\label{app - harmonic3}\ ] ] away from any gas / surface interaction , the mean free time describing the average collision time between two molecules is well - defined .
we may therefore choose @xmath141 to be the time between successive molecular collisions .
in such a case , and with @xmath163 defining the flow macroscopic time scale , we have a knudsen number @xmath164 .
subsequently , equation ( [ app - harmonic3 ] ) yields , @xmath165 \exp\left[c_0\frac{k_i}{\omega } \omega^*kn x^*\right ] .
\label{app - harmonic4}\ ] ] we therefore have a dimensionless inverse speed @xmath166 and a dimensionless damping coefficient @xmath167
. meanwhile , the dimensionless frequency is a product : @xmath168 .
this means that for a fixed value of @xmath169 , the knudsen number is a simple scaling factor for the dimensionless frequency .
conversely , a fixed value of the dimensionless frequency is a simple scaling factor for the knudsen number .
consequently , and for this particular configuration , one may absorb the factor @xmath169 into @xmath137 and interpret the variation of their product as either knudsen number or dimensionless frequency variations
. however , this description relies on the definition of the microscopic time @xmath141 as the time between molecule / molecule collisions . if this microscopic time is physically undefined , or becomes large , then equation ( [ app - harmonic4 ] ) and the interpretation that follows it becomes invalid because the product @xmath168 is indeterminate
this is the case when the gas is confined between two surfaces so that collisions between molecules are no longer the most important mechanism of momentum transfer from one surface to the other , and instead the interactions of the molecules directly with the two surfaces ( the source and receiver in the experiments ) is . in greenspan
s work , which has been followed by several authors , the non - dimensionalisation starts with a reference speed , denoted @xmath170 , which in our notation corresponds to @xmath171 , assuming an approximation of the dispersion at high pressure .
then the intermolecular collision mean time @xmath141 is determined assuming maxwell molecules . the dimensionless sound speed and damping
are given as they appear through equation ( [ app - harmonic4 ] ) while the inverse of the product @xmath168 is referred to as `` reynold s number '' .
in any case , one can see easily from the expression @xmath172 that for all theories predicting a finite value of the damping this dimensionless expression should give zero damping for @xmath99 tending to infinity .
so , the expression , at first glance , is not even a well - indicated form to compare between different theoretical results in this field .
a different analysis is therefore required .
returning to equation ( [ app - harmonic1 ] ) , for high knudsen numbers let us assume that the separation distance between the two surfaces , @xmath142 , is the relevant microscopic parameter . with a @xmath160 that may be the thermal speed ( or any other characteristic molecular speed ) , the average time spent travelling between the surfaces is now associated with @xmath141 @xcite .
as there are , on average , no intermolecular collisions in that period we expect the wave propagation to become independent of the conventional knudsen number beyond a certain limit .
equation ( [ app - harmonic1 ] ) is then written , @xmath173 , \label{app - harmonic12}\ ] ] which implies @xmath139 , @xmath174 , and the dimensionless sound speed and damping are given , respectively , by @xmath175 and @xmath176 , which are the expressions we defined in equation ( [ harmonic - correct ] ) ( allowing for the constant coefficient @xmath143 ) . moreover , this dimensionless phase speed and damping are independent of the dimensional frequency @xmath99 and so independent of @xmath163 . although our corrected dimensional analysis seems to work with the data in reference @xcite , further verifications with other experiments using reliable dimensionless parameters are necessary .
it is also worth noting that the failure of greenspan s analysis at high frequencies means that a high conventional knudsen number does not necessarily mean a high frequency , and _
vice versa_. in figure [ paradoxical_damp ] , @xmath99 is strictly speaking referring to a separation - distance - based knudsen number , not the real dimensional frequency as we have shown through equation ( [ app - harmonic12 ] ) .
we have not compared our theoretical results with the more recent experimental data by schotter @xcite .
this is because , while schotter differentiated between two microscopic time scales , he defined the dimensionless parameters as in greenspan s analysis , i.e. , a dimensionless damping coefficient that depends on the frequency over the full regime .
he reported different plots for different separation distances . | we investigate sound wave propagation in a monatomic gas using a volume - based hydrodynamic model . in reference
@xcite , a microscopic volume - based kinetic approach was proposed by analyzing molecular spatial distributions ; this led to a set of hydrodynamic equations incorporating a mass - density diffusion component . here
we find that these new mass - density diffusive flux and volume terms mean that our hydrodynamic model , uniquely , reproduces sound wave phase speed and damping measurements with excellent agreement over the full range of knudsen number . in the high knudsen number ( high frequency )
regime , our volume - based model predictions agree with the plane standing waves observed in the experiments , which existing kinetic and continuum models have great difficulty in capturing . in that regime ,
our results indicate that the `` sound waves '' presumed in the experiments may be better thought of as `` mass - density waves '' , rather than the pressure waves of the continuum regime . |
a quantum finite automaton ( qfa ) is a model for a quantum computer with a finite memory .
qfas can recognize the same languages as classical finite automata but they can be exponentially more space efficient than their classical counterparts @xcite . to recognize an arbitrary regular language , qfas need to be able to perform general measurements after reading every input symbol , as in @xcite .
if we restrict qfas to unitary evolution and one measurement at the end of computation ( which might be easier to implement experimentally ) , their power decreases considerably .
namely @xcite , they can only recognize the languages recognized by permutation automata , a classical model in which the transitions between the states have to be fully reversible .
similar decreases of the computational power have been observed in several other contexts .
quantum error correction is possible if we have a supply of quantum bits initialized to @xmath1 at any moment of computation ( see chapter 10 of @xcite ) . yet
, if the number of quantum bits is fixed and it is not allowed to re - initialize them by measurements , error correction becomes difficult @xcite .
simulating a probabilistic turing machine by a quantum turing machine is trivial if we allow to measure and reinitialize qubits but quite difficult if the number of qubits is fixed and they can not be reinitialized @xcite . thus , the availability of measurements is very important for quantum automata . what happens if the measurements are allowed but restricted ?
how can we use the measurements of a restricted form to enhance the abilities of quantum automata ?
can quantum effects be used to recognize languages that are not recognizable by classical automata with the same reversibility requirements ? in this paper , we look at those questions for measure - many " qfa model by kondacs and watrous @xcite .
this model allows intermediate measurements during the computation but these measurements have to be of a restricted type .
more specifically , they can have 3 outcomes : `` accept '' , `` reject '' , `` do nt halt '' and if one gets `` accept '' or `` reject '' , the computation ends and this is the result of computation .
the reason for allowing measurements of this type was that the states of a qfa then have a simple description of the form @xmath2 where @xmath3 is the probability that the qfa has accepted , @xmath4 is the probability that the qfa has rejected and @xmath5 is the remaining state if the automaton has not accepted or rejected .
allowing more general measurements would make the remaining state a mixed state @xmath6 instead of a pure state @xmath5 . having a mixed state as the current state of a qfa is very reasonable physically but the mathematical apparatus for handling pure states is simpler than one for mixed states . for this model
, it is known that @xcite * any language recognizable by a qfa with a probability @xmath7 , @xmath8 is recognizable by a reversible finite automaton ( rfa ) . *
the language @xmath9 can be recognized with probability @xmath10 but can not be recognized by an rfa .
thus , the quantum automata in this model have an advantage over their classical counterparts ( rfas ) with the same reversibility requirements but this advantage only allows to recognize languages with probabilities at most 7/9 , not @xmath11 with arbitrary @xmath8 .
this is a quite unusual property because , in almost any other computational model , the accepting probability can be increased by repeating the computation in parallel . as we see , this is not the case for qfas . in this paper , we develop a method for determining the maximum probability with which a qfa can recognize a given language .
our method is based on the quantum counterpart of classification of states of a markov chain into ergodic and transient states @xcite .
we use this classification of states to transform the problem of determining the maximum accepting probability of a qfa into a quadratic optimization problem .
then , we solve this problem ( analytically in simpler cases , by computer in more difficult cases ) .
compared to previous work , our new method has two advantages .
first , it gives a systematic way of calculating the maximum accepting probabilities .
second , solving the optimization problems usually gives the maximum probability exactly .
most of previous work @xcite used approaches depending on the language and required two different methods : one for bounding the probability from below , another for bounding it from above . often , using two different approaches gave an upper and a lower bound with a gap between them ( like @xmath12 vs. @xmath7 mentioned above ) .
with the new approach , we are able to close those gaps .
we use our method to calculate the maximum accepting probabilities for a variety of languages ( and classes of languages ) .
first , we construct a quadratic optimization problem for the maximum accepting probability by a qfa of a language that is not recognizable by an rfa .
solving the problem gives the probability @xmath13 .
this probability can be achieved for the language @xmath14 in the two - letter alphabet @xmath15 but no language that is no recognizable by a rfa can be recognized with a higher probability .
this improves the @xmath7 result of @xcite .
this result can be phrased in a more general way .
namely , we can find the property of a language which makes it impossible to recognize the language by an rfa .
this property can be nicely stated in the form of the minimal deterministic automaton containing a fragment of a certain form .
we call such a fragment a non - reversible construction " .
it turns out that there are many different `` non - reversible constructions '' and they have different influence on the accepting probability .
the one contained in the @xmath14 language makes the language not recognizable by an rfa but the language is still recognizable by a qfa with probability @xmath0 .
in contrast , some constructions analyzed in @xcite make the language not recognizable with probability @xmath16 for any @xmath8 . in the rest of this paper , we look at different non - reversible constructions " and their effects on the accepting probabilities of qfas .
we consider three constructions : `` two cycles in a row '' , `` @xmath17 cycles in parallel '' and a variant of the @xmath18 construction .
the best probabilities with which one can recognize languages containing these constructions are @xmath19 , @xmath20 and @xmath21 , respectively . the solution of the optimization problem for `` two cycles in a row '' gives a new qfa for the language @xmath22 that recognizes it with probability @xmath19 , improving the result of @xcite . again
, using the solution of the optimization problem gives a better qfa that was previously missed because of disregarding some parameters .
we define the kondacs - watrous ( `` measure - many '' ) model of qfas @xcite . [ def1 ]
a qfa is a tuple @xmath23 where @xmath24 is a finite set of states , @xmath25 is an input alphabet , @xmath26 is a transition function ( explained below ) , @xmath27 is a starting state , and @xmath28 and @xmath29 are sets of accepting and rejecting states ( @xmath30 ) .
the states in @xmath31 and @xmath32 , are called _ halting states _ and the states in @xmath33 are called _ non halting states_. * states of @xmath34 . * the state of @xmath34 can be any superposition of states in @xmath24 ( i. e. , any linear combination of them with complex coefficients ) .
we use @xmath35 to denote the superposition consisting of state @xmath36 only .
@xmath37 denotes the linear space consisting of all superpositions , with @xmath38-distance on this linear space .
* endmarkers .
* let @xmath39 and @xmath40 be symbols that do not belong to @xmath41 .
we use @xmath39 and @xmath40 as the left and the right endmarker , respectively .
we call @xmath42 the _ working alphabet _ of @xmath34
. * transition function . *
the transition function @xmath26 is a mapping from @xmath43 to @xmath44 such that , for every @xmath45 , the function @xmath46 defined by @xmath47 is a unitary transformation ( a linear transformation on @xmath37 that preserves @xmath38 norm ) .
* computation . *
the computation of a qfa starts in the superposition @xmath48 .
then transformations corresponding to the left endmarker @xmath39 , the letters of the input word @xmath49 and the right endmarker @xmath40 are applied .
the transformation corresponding to @xmath50 consists of two steps .
first , @xmath51 is applied .
the new superposition @xmath52 is @xmath53 where @xmath54 is the superposition before this step .
then , @xmath52 is observed with respect to @xmath55 where @xmath56 , @xmath57 , @xmath58 .
it means that if the system s state before the measurement was @xmath59 then the measurement accepts @xmath60 with probability @xmath61 , rejects with probability @xmath62 and continues the computation ( applies transformations corresponding to next letters ) with probability @xmath63 with the system having the ( normalized ) state @xmath64 where @xmath65 .
we regard these two transformations as reading a letter @xmath66
. * notation .
* we use @xmath67 to denote the transformation consisting of @xmath68 followed by projection to @xmath69 .
this is the transformation mapping @xmath54 to the non - halting part of @xmath70 .
we use @xmath71 to denote the product of transformations @xmath72 , where @xmath73 is the @xmath74-th letter of the word @xmath75 .
we also use @xmath76 to denote the ( unnormalized ) non - halting part of qfa s state after reading the left endmarker @xmath39 and the word @xmath77 . from the notation it follows that @xmath78 .
* recognition of languages .
* we will say that an automaton recognizes a language @xmath79 with probability @xmath80 @xmath81 if it accepts any word @xmath82 with probability @xmath83 and rejects any word @xmath84 with probability @xmath83 . for classical markov chains
, one can classify the states of a markov chain into _ ergodic _ sets and _ transient _ sets @xcite .
if the markov chain is in an ergodic set , it never leaves it .
if it is in a transient set , it leaves it with probability @xmath11 for an arbitrary @xmath8 after sufficiently many steps .
a quantum counterpart of a markov chain is a quantum system to which we repeatedly apply a transformation that depends on the current state of the system but does not depend on previous states
. in particular , it can be a qfa that repeatedly reads the same word @xmath49 .
then , the state after reading @xmath49 @xmath85 times depends on the state after reading @xmath49 @xmath17 times but not on any of the states before that .
the next lemma gives the classification of states for such qfas .
[ lemmaaf ] @xcite let @xmath86 .
there are subspaces @xmath87 , @xmath88 such that @xmath89 and 1 . if @xmath90 , then @xmath91 and @xmath92 , 2 . if @xmath93 , then @xmath94 when @xmath95 . instead of ergodic and transient sets
, we have subspaces @xmath87 and @xmath88 .
the subspace @xmath87 is a counterpart of an ergodic set : if the quantum process defined by repeated reading of @xmath49 is in a state @xmath90 , it stays in @xmath87 .
@xmath88 is a counterpart of a transient set : if the state is @xmath93 , @xmath88 is left ( for an accepting or rejecting state ) with probability arbitrarily close to 1 after sufficiently many @xmath49 s . in some of proofs
we also use a generalization of lemma [ lemmaaf ] to the case of two ( or more ) words @xmath49 and @xmath96 : [ akvlemma ] @xcite let @xmath97 .
there are subspaces @xmath87 , @xmath88 such that @xmath89 and 1 . if @xmath90 , then @xmath91 and @xmath98 and @xmath99 and @xmath100 , 2 . if @xmath93 , then for any @xmath101 , there exists @xmath102 such that @xmath103
. we also use a lemma from @xcite .
[ bvlemma ] @xcite if @xmath54 and @xmath104 are two quantum states and @xmath105 then the total variational distance between probability distributions generated by the same measurement on @xmath54 and @xmath104 is at most but it can be improved to @xmath106 . ]
ambainis and freivalds @xcite characterized the languages recognized by rfas as follows .
@xcite [ aftheorem ] let @xmath79 be a language and @xmath34 be its minimal automaton .
@xmath79 is recognizable by a rfa if and only if there is no @xmath107 such that 1 .
@xmath108 , 2 .
if @xmath34 starts in the state @xmath109 and reads @xmath49 , it passes to @xmath110 , 3 . if @xmath34 starts in the state @xmath110 and reads @xmath49 , it passes to @xmath110 , and 4
@xmath110 is neither `` all - accepting '' state , nor `` all - rejecting '' state , an rfa is a special case of a qfa that outputs the correct answer with probability 1 .
thus , any language that does not contain the construction of theorem [ aftheorem ] can be recognized by a qfa that always outputs the correct answer .
ambainis and freivalds @xcite also showed the reverse of this : any language @xmath79 with the minimal automaton containing the construction of theorem [ aftheorem ] can not be recognized by a qfa with probability @xmath7 .
we consider the question : what is the maximum probability of correct answer than can be achieved by a qfa for a language that can not be recognized by an rfa ?
the answer is : [ t1 ] let @xmath79 be a language and @xmath34 be its minimal automaton . 1 .
if @xmath34 contains the construction of theorem [ aftheorem ] , @xmath79 can not be recognized by a 1-way qfa with probability more than @xmath111 .
2 . there is a language @xmath79 with the minimal automaton @xmath34 containing the construction of theorem [ aftheorem ] that can be recognized by a qfa with probability @xmath111 . _ proof . _
we consider the following optimization problem .
* optimization problem 1 . *
find the maximum @xmath80 such that there is a finite dimensional vector space @xmath112 , subspaces @xmath113 , @xmath114 such that @xmath115 , vectors @xmath116 , @xmath117 such that @xmath118 and @xmath119 and probabilities @xmath120 , @xmath121 such that @xmath122 and 1 .
@xmath123 , 2 .
@xmath124 , 3 .
@xmath125 .
we sketch the relation between a qfa recognizing @xmath79 and this optimization problem .
let @xmath24 be a qfa recognizing @xmath79 .
let @xmath126 be the minimum probability of the correct answer for @xmath24 , over all words .
we use @xmath24 to construct an instance of the optimization problem above with @xmath127 .
namely , we look at @xmath24 reading an infinite ( or very long finite ) sequence of letters @xmath49 . by lemma [ lemmaaf ]
, we can decompose the starting state @xmath54 into 2 parts @xmath128 and @xmath129 .
define @xmath130 and @xmath131 .
let @xmath120 and @xmath121 be the probabilities of getting into an accepting ( for @xmath120 ) or rejecting ( for @xmath121 ) state while reading an infinite sequence of @xmath49 s starting from the state @xmath117 .
the second part of lemma [ lemmaaf ] implies that @xmath122 .
since @xmath109 and @xmath110 are different states of the minimal automaton @xmath34 , there is a word @xmath96 that is accepted in one of them but not in the other . without loss of generality , we assume that @xmath96 is accepted if @xmath34 is started in @xmath109 but not if @xmath34 is started in @xmath110 . also , since @xmath110 is not an `` all - accepting '' state , there must be a word @xmath132 that is rejected if @xmath34 is started in the state @xmath110 .
we choose @xmath113 and @xmath114 so that the square of the projection @xmath133 ( @xmath134 ) of a vector @xmath135 on @xmath113 ( @xmath114 ) is equal to the accepting ( rejecting ) probability of @xmath24 if we run @xmath24 on the starting state @xmath135 and input @xmath96 and the right endmarker @xmath40 .
finally , we set @xmath80 equal to the @xmath136 of the set consisting of the probabilities of correct answer of @xmath24 on the words @xmath96 and @xmath137 , @xmath138 for all @xmath139 .
then , condition 1 of the optimization problem , @xmath123 is true because the word @xmath96 must be accepted and the accepting probability for it is exactly the square of the projection of the starting state ( @xmath140 ) to @xmath133 . condition 2 follows from running @xmath24 on a word @xmath137 for some large @xmath74 . by lemma [ lemmaaf ] , if @xmath141 for some @xmath17 , @xmath142 .
also , @xmath116 , @xmath143 , @xmath144 , @xmath145 is an infinite sequence in a finite - dimensional space .
therefore , it has a limit point and there are @xmath146 , @xmath147 such that @xmath148 we have @xmath149 since @xmath99 for @xmath90 , @xmath150 and we have @xmath151 thus , reading @xmath152 has the following effect : 1 .
@xmath116 gets mapped to a state that is at most @xmath153-away ( in @xmath38 norm ) from @xmath116 , 2 .
@xmath117 gets mapped to an accepting / rejecting state and most @xmath153 fraction of it stays on the non - halting states . together
, these two requirements mean that the state of @xmath24 after reading @xmath152 is at most @xmath154-away from @xmath116 .
also , the probabilities of @xmath24 accepting and rejecting while reading @xmath152 differ from @xmath120 and @xmath121 by at most @xmath153 .
let @xmath155 be the probability of @xmath24 rejecting @xmath137 .
since reading @xmath96 in @xmath110 leads to a rejection , @xmath137 must be rejected and @xmath156 .
the probability @xmath155 consists of two parts : the probability of rejection during @xmath152 and the probability of rejection during @xmath96 .
the first part differs from @xmath121 by at most @xmath153 , the second part differs from @xmath157 by at most @xmath158 ( because the state of @xmath24 when starting to read @xmath96 differs from @xmath116 by at most @xmath154 and , by lemma [ bvlemma ] , the accepting probabilities differ by at most twice that ) .
therefore , @xmath159 since @xmath156 , this implies @xmath160 . by appropriately choosing @xmath74
, we can make this true for any @xmath8 .
therefore , we have @xmath161 which is condition 2 . condition 3 is true by considering @xmath138 .
this word must be accepted with probability @xmath80 .
therefore , for any @xmath74 , @xmath24 can only reject during @xmath152 with probability @xmath162 and @xmath125 .
this shows that no qfa can achieve a probability of correct answer more than the solution of optimization problem 1 .
it remains to solve this problem .
* solving optimization problem 1 . *
the key idea is to show that it is enough to consider 2-dimensional instances of the problem . since @xmath118 , the vectors @xmath163 form a right - angled triangle
this means that @xmath164 , @xmath165 where @xmath166 is the angle between @xmath116 and @xmath140 .
let @xmath167 and @xmath168 be the normalized versions of @xmath116 and @xmath117 : @xmath169 , @xmath170 .
then , @xmath171 and @xmath172 .
consider the two - dimensional subspace spanned by @xmath173 and @xmath174 .
since the accepting and the rejecting subspaces @xmath113 and @xmath114 are orthogonal , @xmath173 and @xmath174 are orthogonal .
therefore , the vectors @xmath175 and @xmath176 form an orthonormal basis .
we write the vectors @xmath167 , @xmath116 and @xmath140 in this basis .
the vector @xmath167 is @xmath177 where @xmath178 is the angle between @xmath167 and @xmath179 .
the vector @xmath171 is equal to @xmath180 .
next , we look at the vector @xmath140 .
we fix @xmath178 , @xmath166 and @xmath116 and try to find the @xmath117 which maximizes @xmath80 for the fixed @xmath178 , @xmath166 and @xmath116 . the only place where @xmath117 appears in the optimization problem 1
is @xmath181 on the left hand side of condition 1 .
therefore , we should find @xmath117 that maximizes @xmath181 .
we have two cases : 1 .
+ the angle between @xmath140 and @xmath179 is at least @xmath183 ( because the angle between @xmath116 and @xmath179 is @xmath178 and the angle between @xmath140 and @xmath116 is @xmath166 ) .
therefore , the projection of @xmath140 to @xmath179 is at most @xmath184 .
since @xmath185 is a part of the rejecting subspace @xmath114 , this means that @xmath186 .
the maximum @xmath187 is achieved if we put @xmath140 in the plane spanned by @xmath179 and @xmath185 : @xmath188 . + next
, we can rewrite condition 3 of the optimization problem as @xmath189 . then
, conditions 1 - 3 together mean that @xmath190 to solve the optimization problem , we have to maximize ( [ eq1 ] ) subject to the conditions of the problem . from the expressions for @xmath116 and @xmath140 above
, it follows that ( [ eq1 ] ) is equal to @xmath191 first , we maximize @xmath192 .
the first term is increasing in @xmath121 , the second is decreasing .
therefore , the maximum is achieved when both become equal which happens when @xmath193 .
then , both @xmath194 and @xmath195 are @xmath196 .
now , we have to maximize @xmath197 we first fix @xmath183 and try to optimize the second term .
since @xmath198 ( a standard trigonometric identity ) , it is maximized when @xmath199 and @xmath200 .
then , @xmath201 and ( [ eq3 ] ) becomes @xmath202 the first term is increasing in @xmath178 , the second is decreasing .
the maximum is achieved when @xmath203 the left hand side of ( [ eq5 ] ) is equal to @xmath204 .
therefore , if we denote @xmath205 by @xmath96 , ( [ eq5 ] ) becomes a quadratic equation in @xmath96 : @xmath206 solving this equation gives @xmath207 and @xmath208 .
2 . @xmath209 .
+ we consider @xmath210 . since the minimum of two quantities is at most their average , this is at most @xmath211 since @xmath209 , we have @xmath212 and ( [ eq6 ] ) is at most @xmath213 .
this is maximized by @xmath214 .
then , we get @xmath215 which is less than @xmath216 which we got in the first case .
this proves the first part of the theorem .
* construction of a qfa . *
this part is proven by taking the solution of optimization problem 1 and using it to construct a qfa for the language @xmath18 in a two - letter alphabet @xmath15 .
the state @xmath109 is just the starting state of the minimal automaton , @xmath110 is the state to which it gets after reading @xmath66 , @xmath217 , @xmath96 is the empty word and @xmath218 .
let @xmath178 be the solution of ( [ eq5 ] ) .
then , @xmath219 , @xmath220 , @xmath221 , @xmath222 and @xmath223 .
@xmath224 is the probability of correct answer for our qfa described below .
the qfa @xmath34 has 5 states : @xmath225 , @xmath226 and @xmath227 .
@xmath228 , @xmath229 .
the initial state is @xmath230 .
the transition function is @xmath231 @xmath232 @xmath233 to recognize @xmath79 , @xmath34 must accept all words of the form @xmath234 for @xmath235 and reject the empty word and any word that contains the letter @xmath236 . 1 . the empty word .
+ the only tranformation applied to the starting state is @xmath237 .
therefore , the final superposition is @xmath238 the amplitude of @xmath239 in the final superposition is @xmath240 and the word is rejected with a probability @xmath241 .
2 . @xmath234 for @xmath235 .
+ first , @xmath68 maps the @xmath242 component to @xmath243 the probability of accepting at this point is @xmath244 .
the other component of the superposition , @xmath245 stays unchanged until @xmath237 maps it to @xmath246 the probability of accepting at this point is @xmath247 .
the total probability of accepting is @xmath248 by equation ( [ eq6 ] ) , this is equal to @xmath224 .
3 . a word containing at least one @xmath236 .
+ if @xmath236 is the first letter of the word , the entire superposition is mapped to rejecting states and the word is rejected with probability 1 . otherwise , the first letter is @xmath66 , it maps @xmath249 to @xmath250 .
the probability of accepting at this point is @xmath251 . by equation ( [ eq6 ] ) , this is the same as @xmath252 .
after that , the remaining component ( @xmath245 ) is not changed by next @xmath66s and mapped to a rejecting state by the first @xmath236 . therefore , the total probability of accepting is also @xmath252 and the correct answer ( rejection ) is given with a probability @xmath224 .
we now look at fragments of the minimal automaton that imply that a language can not be recognized with probability more than @xmath80 , for some @xmath80 .
we call such fragments non - reversible constructions " .
the simplest such construction is the one of theorem [ aftheorem ] . in this section
, we present 3 other
non - reversible constructions " that imply that a language can be recognized with probability at most @xmath21 , @xmath19 and @xmath20 .
this shows that different constructions are non - reversible " to different extent .
comparing these 4 `` non - reversible '' constructions helps to understand what makes one of them harder for qfa ( i.e. , recognizable with worse probability of correct answer ) the first construction comes from the language @xmath22 considered in ambainis and freivalds @xcite .
this language was the first example of a language that can be recognized by a qfa with some probability ( 0.6822 ... ) but not with another ( @xmath7 ) .
we find the `` non - reversible '' construction for this language and construct the qfa with the best possible accepting probability .
[ tf4 ] let @xmath79 be a language and @xmath34 its minimal automaton . 1 .
if @xmath34 contains states @xmath109 , @xmath110 and @xmath253 such that , for some words @xmath49 and @xmath96 , 1
. if @xmath34 reads @xmath49 in the state @xmath109 , it passes to @xmath109 , 2 .
if @xmath34 reads @xmath96 in the state @xmath109 , it passes to @xmath110 , 3 .
if @xmath34 reads @xmath96 in the state @xmath110 , it passes to @xmath110 , 4 .
if @xmath34 reads @xmath49 in the state @xmath110 , it passes to @xmath253 , 5 .
if @xmath34 reads @xmath49 in the state @xmath253 , it passes to @xmath253 + then @xmath79 can not be recognized by a qfa with probability more than @xmath19 .
the language @xmath254 ( the minimal automaton of which contains the construction above ) can be recognized by a qfa with probability @xmath19 .
_ by a reduction to the following optimization problem .
* optimization problem 2 . *
find the maximum @xmath80 such that there is a finite - dimensional space @xmath255 , subspaces @xmath113 , @xmath114 such that @xmath256 , vectors @xmath116 , @xmath117 and @xmath257 and probabilities @xmath258 , @xmath259 , @xmath260 , @xmath261 such that 1 .
@xmath262 , 2 .
@xmath263 , 3 .
@xmath264 , 4 .
5 . @xmath266 ; 6 .
@xmath267 ; 7 . @xmath268 ; 8 .
@xmath269 ; 9 .
@xmath270 .
we use a theorem from @xcite .
[ t13 ] let @xmath79 be a language and @xmath34 be its minimal automaton .
assume that there is a word @xmath49 such that @xmath34 contains states @xmath109 , @xmath110 satisfying : 1 .
@xmath108 , 2 .
if @xmath34 starts in the state @xmath109 and reads @xmath49 , it passes to @xmath110 , 3 .
if @xmath34 starts in the state @xmath110 and reads @xmath49 , it passes to @xmath110 , and 4 . there is a word @xmath96 such that if m starts in @xmath110 and reads y , it passes to @xmath109 , then @xmath79 can not be recognized by any 1-way quantum finite automaton .
let @xmath24 be a qfa recognizing @xmath79 .
let @xmath271 be state where the minimal automaton @xmath34 goes if it reads @xmath96 in the state @xmath253 . in case when @xmath272 we get the forbidden construction of theorem [ t13 ] . in case when @xmath273 states @xmath110 and @xmath271 are different states of the minimal automaton @xmath34 .
therefore , there is a word @xmath132 that is accepted in one of them but not in the other . without loss of generality , we assume that @xmath96 is accepted if @xmath34 is started in @xmath110 but not if @xmath34 is started in @xmath271 .
we choose @xmath113 so that the square of the projection @xmath133 of a vector @xmath135 on @xmath113 is equal to the accepting probability of @xmath24 if we run @xmath24 on the starting state @xmath135 and input @xmath274 and the right endmarker @xmath40 .
we use lemma [ lemmaaf ] .
let @xmath275 be @xmath87 and @xmath276 be @xmath88 for word @xmath49 and let @xmath277 be @xmath278 and @xmath279 be @xmath278 for word @xmath96 .
without loss of generality we can assume that @xmath109 is a starting state of @xmath34 .
let @xmath280 be the starting superposition for @xmath24 .
we can also assume that reading @xmath49 in this state does not decrease the norm of this superposition .
we divide @xmath280 into three parts : @xmath116 , @xmath117 and @xmath257 so that @xmath281 and @xmath282 , @xmath283 and @xmath284 . due to @xmath285 is the starting superposition we have @xmath286(condition 1 ) . since @xmath287
we get that @xmath264(condition 3 ) due to @xmath284 .
similarly @xmath265(condition 4 ) and @xmath263(condition 2 ) .
it is easy to get that @xmath288(condition 7 ) because reading @xmath274 in the state @xmath109 leads to accepting state .
let @xmath258(@xmath259 ) be the accepting(rejecting ) probability while reading an infinite sequence of letters @xmath96 in the state @xmath285 .
then @xmath289(condition 5 ) due to @xmath281 and @xmath282 .
let @xmath260(@xmath261 ) be the accepting(rejecting ) probability while reading an infinite sequence of letters @xmath49 in the state @xmath140 .
then @xmath290(condition 6 ) due to @xmath291 and @xmath284 .
we find an integer @xmath74 such that after reading @xmath292 the norm of @xmath293 is at most some fixed @xmath101 . now similarly to theorem [ t1 ] we can get condition 8 : @xmath294 .
let @xmath295 , @xmath296 , @xmath297 .
we find an integer @xmath298 such that after reading @xmath299 the norm of @xmath300 is at most @xmath153 . since @xmath301 then @xmath302 .
therefore , @xmath303 .
then @xmath304 due to previous inequalities .
now similarly to theorem [ t1 ] we can get condition 9 : @xmath305 . +
we have constructed our second optimization problem .
we solve the problem by computer . using this solution
we can easily construct corresponding quantum automaton .
= 2.5 in [ tf5 ] let @xmath306
let @xmath79 be a language .
if there are words @xmath307 such that its minimal automaton @xmath34 contains states @xmath308 satisfying : 1 . if m starts in the state @xmath309 and reads @xmath310 , it passes to @xmath311 , 2 .
if m starts in the state @xmath312 and reads @xmath313 , it passes to @xmath311 , 3 . for each @xmath74 the state @xmath311 is not `` all - rejecting '' state , + then @xmath79 can not be recognized by a qfa with probability greater than @xmath314 .
2 . there is a language such that its minimal deterministic automaton contains this construction and the language can be recognized by a qfa with probability @xmath314 . for @xmath315 ,
a related construction was considered in @xcite .
there is a subtle difference between the two constructions ( the one considered here for @xmath315 and the one in @xcite ) .
the `` non - reversible construction '' in @xcite requires the sets of words accepted from @xmath109 and @xmath110 to be incomparable .
this extra requirement makes it much harder : no qfa can recognize a language with the `` non - reversible construction '' of @xcite even with the probability @xmath16 . _ proof .
_ * impossibility result .
* this is the only proof in this paper that does not use a reduction to an optimization problem .
instead , we use a variant of the classification of states ( lemma [ akvlemma ] ) directly .
we only consider the case when the sets of words accepted from @xmath311 and @xmath316 are not incomparable .
( the other case follows from the impossibility result in @xcite . )
let @xmath317 be the set of words accepted from @xmath312 .
this means that for each @xmath318 we have either @xmath319 or @xmath320 . without loss of generality
we can assume that @xmath321 .
now we can choose @xmath17 words @xmath322 such that @xmath323 and @xmath324 .
the word @xmath325 exists due to the condition _
( c)_. we use a generalization of lemma [ akvlemma ] .
[ lemma2 ] let @xmath326 .
there are subspaces @xmath87 , @xmath88 such that @xmath89 and 1 . if @xmath90 , then @xmath327 @xmath328 @xmath329 and @xmath330 @xmath328 @xmath331 2 . if @xmath93 , then for any @xmath101 , there exists a word @xmath332 such that @xmath103 .
the proof is similar to lemma [ akvlemma ] .
let @xmath79 be a language such that its minimal automaton @xmath34 contains the non reversible construction from theorem [ tf5 ] and @xmath333 be a qfa .
let @xmath80 be the accepting probability of @xmath333 .
we show that @xmath334 .
let @xmath75 be a word such that after reading it @xmath34 is in the state @xmath309 .
let @xmath335 , @xmath336 , @xmath337 .
we find a word @xmath338 such that after reading @xmath339 the norm of @xmath340 is at most some fixed @xmath8 .
( such word exists due to lemma [ lemma2 ] . )
we also find words @xmath341 such that @xmath342 , @xmath145 , @xmath343 . because of unitarity of @xmath344 , @xmath145 , @xmath345 on @xmath87 ( part ( i ) of lemma [ lemma2 ] ) , there exist integers @xmath346 such that @xmath347 , @xmath328 @xmath348 .
let @xmath349 be the probability of @xmath333 accepting while reading @xmath350 .
let @xmath351 be the probabilities of accepting while reading @xmath352 with a starting state @xmath76 and and @xmath353 be the probabilities of accepting while reading @xmath354 with a starting state @xmath355 .
let us consider @xmath356 words : + @xmath357 + @xmath358 + @xmath359 + @xmath360 + @xmath328 + @xmath361 + @xmath362 + @xmath363 @xmath333 accepts @xmath364 with probability at least @xmath365 and at most @xmath366 .
* proof . *
the probability of accepting while reading @xmath350 is @xmath349 .
after that , @xmath333 is in the state @xmath76 and reading @xmath367 in this state causes it to accept with probability @xmath120 .
the remaining state is @xmath368 .
if it was @xmath355 , the probability of accepting while reading the rest of the word ( @xmath369 ) would be exactly @xmath370 .
it is not quite @xmath355 but it is close to @xmath355 .
namely , we have @xmath371 by lemma [ bvlemma ] , this means that the probability of accepting during @xmath369 is between @xmath372 and @xmath373 .
this lemma implies that @xmath374 because of @xmath375 .
similarly , @xmath376 because of @xmath377 . finally , we have @xmath356 inequalities : + @xmath378 + @xmath379 + @xmath380 + @xmath381 + @xmath328 + @xmath382 + @xmath383 + @xmath384 + by adding up these inequalities we get @xmath385 .
we can notice that @xmath386 .
( this is due to the facts that @xmath387 , @xmath388 and @xmath389 . ) hence , @xmath390 . since
such @xmath356 words can be constructed for arbitrarily small @xmath153 , this means that @xmath333 does not recognize @xmath79 with probability greater than @xmath314 .
* constructing a quantum automaton . * we consider a language @xmath391 in the alphabet @xmath392 such that its minimal automaton has accepting states @xmath308 and rejecting state @xmath226 and the transition function @xmath393 is defined as follows : @xmath394 , @xmath395 , @xmath396 , @xmath397 , @xmath398 , @xmath399 , @xmath400 .
it can be checked that this automaton contains the non reversible construction from theorem 4 .
hence , this language can not be recognized by a qfa with probability greater than @xmath314 .
next , we construct a qfa @xmath333 that accepts this language with such probability .
the automaton has @xmath401 states : @xmath402 , @xmath403 , @xmath404 @xmath405 .
@xmath406 , @xmath407 .
the initial state is @xmath408 the transition function is @xmath409 @xmath410 @xmath411 1 . the empty word .
+ the only tranformation applied to the starting state is @xmath237 .
therefore , the final superposition is @xmath412 and the word is accepted with probability 1 . 2 .
the word starts with @xmath413 . + reading @xmath413 maps @xmath414 to @xmath415 .
therefore , this word is accepted with probability at least @xmath416 .
word is in form @xmath417 .
the superposition after reading @xmath418 is @xmath419 at this moment @xmath333 accepts with probability @xmath420 and rejects with probability @xmath421 .
the computation continues in the superposition @xmath422 clearly , that reading of all remaining letters does not change this superposition . since @xmath423 maps each @xmath424 to an accepting state then @xmath333 rejects this word with probability at most @xmath425 4 .
word @xmath49 starts with @xmath426 . before reading @xmath427
the superposition is @xmath422 + _ case 1 .
_ @xmath428 .
@xmath429 + since @xmath428 then reading @xmath427 maps at least @xmath430 states of @xmath431 to rejecting states .
this means that @xmath333 rejects with probability at least @xmath432 + _ case 2 .
_ @xmath433 .
@xmath434 since @xmath433 then reading @xmath427 maps at least @xmath435 states of @xmath431 to accepting states .
this means that @xmath333 accepts with probability at least @xmath436 [ tf2 ] let @xmath79 be a language . 1 . if there are words @xmath49 ,
@xmath325 , @xmath437 such that its minimal automaton @xmath34 contains states @xmath109 and @xmath110 satisfying : 1 . if m starts in the state @xmath109 and reads @xmath49 , it passes to @xmath110 , 2 .
if m starts in the state @xmath110 and reads @xmath49 , it passes to @xmath110 , 3 .
if m starts in the state @xmath109 and reads @xmath325 , it passes to an accepting state , 4 . if m starts in the state @xmath109 and reads @xmath437 , it passes to a rejecting state , 5 . if m starts in the state @xmath110 and reads @xmath325 , it passes to a rejecting state , 6 .
if m starts in the state @xmath110 and reads @xmath437 , it passes to an accepting state .
+ then @xmath79 can not be recognized by a qfa with probability greater than @xmath438 .
2 . there is a language @xmath79 with the minimum automaton containing this construction that can be recognized with probability @xmath438 .
_ * impossibility result .
* the construction of optimization problem is similar to the construction of optimization problem 1 .
for this reason , we omit it and just give the optimization problem and show how to solve it .
* optimization problem 3 .
* find the maximum @xmath80 such that there is a finite dimensional vector space @xmath112 , subspaces @xmath113 , @xmath114 ( unlike in previous optimization problems , @xmath113 and @xmath114 do not have to be orthogonal ) and vectors @xmath116 , @xmath117 such that @xmath118 and @xmath439 and probabilities @xmath120 , @xmath121 such that @xmath122 and 1 .
@xmath123 , 2 .
@xmath440 , 3 .
@xmath441 , 4 .
@xmath442 .
* solving optimization problem 3 . * without loss of generality we can assume that @xmath443 .
then these four inequalities can be replaced with only three inequalities 1 .
@xmath123 , 2 .
@xmath441 , 3 .
@xmath444 . clearly that @xmath80 is maximized by @xmath445 .
therefore , we have 1 .
@xmath123 , 2 .
@xmath446 .
next we show that it is enough to consider only instances of small dimension .
we denote @xmath447 as @xmath448 .
first , we restrict @xmath113 to the subspace @xmath449 generated by projections of @xmath116 and @xmath117 to @xmath113 .
this subspace is at most 2-dimensional .
similarly , we restrict @xmath448 to the subspace @xmath450 generated by projections of @xmath116 and @xmath117 to @xmath448 .
the lengths of all projections are still the same .
we fix an orthonormal basis for @xmath112 so that @xmath451 and @xmath452 are both parallel to some basis vectors .
then , @xmath453 and @xmath454 where the first two coordinates correspond to basis vectors of @xmath449 and the last two coordinates correspond to basis vectors of @xmath450 .
we can assume that @xmath455 and @xmath456 are both non - negative .
( otherwise , just invert the direction of one of basis vectors . )
let @xmath457 .
then , there is @xmath458 $ ] such that @xmath459 , @xmath460 .
let @xmath461 .
then , @xmath462 , @xmath463 because @xmath464 due to @xmath118 .
if @xmath465 , we can change @xmath466 and @xmath467 to @xmath468 and @xmath469 where @xmath470 and this only increases @xmath471 .
hence , we can assume that @xmath472 .
we denote @xmath473 .
then , @xmath474 , @xmath475 .
let @xmath476 .
then , @xmath477 and @xmath478 for some @xmath479 $ ] and @xmath480 .
this gives 1 .
@xmath481 , 2 .
@xmath482 . then after some calculations we get 1 .
@xmath483 , 2 .
@xmath484 .
if we fix @xmath485 and vary @xmath166 , then @xmath486 ( and , hence , @xmath487 ) is maximized by @xmath488 .
this means that we can assume @xmath488 and we have 1 .
@xmath489 , 2 .
@xmath490 .
if we consider @xmath491 then @xmath492 .
this means that we are only interested in @xmath493 .
let @xmath494 and @xmath495 .
if we fix @xmath178 and vary @xmath496 , then @xmath497 and @xmath498 are linear functions in @xmath496 and @xmath499 .
we consider two cases .
+ + _ case 1 .
_ @xmath500 .
( this gives @xmath501 for each @xmath496 .
therefore , in this case we only need to maximize the function @xmath498 . )
+ this means that @xmath502 @xmath503 @xmath504 @xmath505 @xmath506 @xmath507 @xmath508 so that @xmath493 , we have @xmath509 this means that @xmath510 $ ] . since @xmath511 , @xmath498 is maximized by @xmath512 and @xmath513 .
this gives @xmath80 equal to @xmath514 .
+ _ case 2 .
_ @xmath515 .
( this is equivalent to @xmath516 $ ] . )
+ this means that @xmath80 is maximized by @xmath517 .
therefore , 1 .
@xmath518 , 2 .
@xmath519 .
let @xmath96 be @xmath520 .
then @xmath521 $ ] and @xmath522 . therefore , 1 .
@xmath523 , 2 .
@xmath524 .
now we express @xmath80 using only @xmath96 .
we get @xmath525 .
finally , if we vary @xmath96 through the interval @xmath526 $ ] , then @xmath80 is maximized by @xmath527 .
this gives @xmath80 equal to @xmath514 .
* construction of a qfa . *
we consider the two letter alphabet @xmath528 .
the language @xmath79 is the union of the empty word and @xmath529 . clearly that the minimal deterministic automaton of @xmath79 contains the non reversible construction from theorem 5 ( just take @xmath66 as @xmath49 , the empty word as @xmath325 and @xmath236 as @xmath437 ) .
next , we describe a qfa @xmath34 accepting this language .
let @xmath178 be the solution of @xmath530 in the interval @xmath531 $ ] .
it can be checked that @xmath532 , @xmath533 , @xmath534 , @xmath535 .
the automaton has 4 states : @xmath225 and @xmath536 .
@xmath228 , @xmath537 .
the initial state is @xmath538 .
the transition function is @xmath539 @xmath540 @xmath541 @xmath542 1 .
the empty word .
+ the only tranformation applied to the starting state is @xmath237 .
therefore , the final superposition is @xmath543 and the word is accepted with probability @xmath532 .
2 . @xmath544 .
+ after reading @xmath236 the superposition is @xmath545 and word is rejected with probability @xmath532 .
3 . @xmath18 .
+ after reading the first @xmath66 the superposition becomes @xmath546 at this moment @xmath34 accepts with probability @xmath547 and rejects with probability @xmath548 .
the computation continues in the superposition @xmath549 it is easy to see that reading all of remaining letters does not change this superposition .
+ therefore , the final superposition ( after reading @xmath40 ) is @xmath550 this means that @xmath34 rejects with probability @xmath551 .
4 . @xmath529 .
+ before reading the first @xmath236 the superposition is @xmath552 and reading this @xmath236 changes this superposition to @xmath553 this means that @xmath34 accepts with probability @xmath554
quantum finite automata ( qfa ) can recognize all regular languages if arbitrary intermediate measurements are allowed . if they are restricted to be unitary , the computational power drops dramatically , to languages recognizable by permutation automata @xcite . in this paper
, we studied an intermediate case in which measurements are allowed but restricted to `` accept - reject - continue '' form ( as in @xcite ) .
quantum automata of this type can recognize several languages not recognizable by the corresponding classical model ( reversible finite automata ) . in all of those cases ,
those languages can not be recognized with probability 1 or @xmath11 , but can be recognized with some fixed probability @xmath555 .
this is an unusual feature of this model because , in most other computational models a probability of correct answer @xmath555 can be easily amplified to @xmath11 for arbitrary @xmath8 . in this paper
, we study maximal probabilities of correct answer achievable for several languages .
those probabilities are related to forbidden constructions " in the minimal automaton .
forbidden construction " being present in the minimal automaton implies that the language can not be recognized with a probability higher than a certain @xmath555 .
the basic construction is one cycle " in figure [ f1 ] .
composing it with itself sequentially ( figure [ f4 ] ) or in parallel ( figure [ f5 ] ) gives forbidden constructions " with a smaller probability @xmath80 .
the achievable probability also depends on whether the sets of words accepted from the different states of the construction are subsets of one another ( as in figure [ f1 ] ) or incomparable ( as in figure [ f2 ] ) .
the constructions with incomparable sets usually imply smaller probabilities @xmath80 .
the accepting probabilities @xmath80 quantify the degree of non - reversibility present in the forbidden construction " .
lower probability @xmath80 means that the language is more difficult for qfa and thus , the construction " has higher degree of non - reversibility . in our paper
, we gave a method for calculating this probability and used it to calculate the probabilities @xmath80 for several constructions " .
the method should apply to a wide class of constructions but solving the optimization problems can become difficult if the construction contains more states ( as for language @xmath556 studied in @xcite ) . in this case
, it would be good to have methods for calculating the accepting probabilities approximately .
a more general problem suggested by this work is : how do we quantify non - reversibility ? accepting probabilities of qfas provide one way of comparing the degree of non - reversibility in different `` constructions '' .
what are the other ways of quantifying it ? and what are the other settings in which similar questions can be studied ?
john watrous .
space - bounded quantum complexity .
_ journal of computer and system sciences _ , 59:281 - 326 , 1999 .
( preliminary version in proceedings of complexity98 , under the title `` relationships between quantum and classical space - bounded complexity classes '' . ) | one of the properties of the kondacs - watrous model of quantum finite automata ( qfa ) is that the probability of the correct answer for a qfa can not be amplified arbitrarily . in this paper
, we determine the maximum probabilities achieved by qfas for several languages .
in particular , we show that any language that is not recognized by an rfa ( reversible finite automaton ) can be recognized by a qfa with probability at most @xmath0 . quantum computation , finite automata , quantum measurement . |
in past decades , there have been tremendous efforts on designing high - order accurate numerical schemes for compressible fluid flows and great success has been achieved .
high - order accurate numerical schemes were pioneered by lax and wendroff @xcite , and extended into the version of high resolution methods by kolgan @xcite , boris @xcite , van leer @xcite , harten @xcite et al , and other higher order versions , such as essentially non - oscillatory ( eno ) @xcite , weighted essentially non - oscillatory ( weno ) @xcite , discontinuous galerkin ( dg ) @xcite methods etc . in the past decades
, the evaluation of the performance of numerical scheme was mostly based on the test cases with strong shocks for capturing sharp shock transition , such as the blast wave interaction , the forward step - facing flows , and the double mach reflection @xcite .
now it is not a problem at all for shock capturing scheme to get stable sharp shock transition .
however , with the further development of higher order numerical methods and practical demands ( such as turbulent flow simulations ) , more challenging test problems for capturing multiple wave structure are expected to be used . for testing higher - order schemes , the setting of these cases should be sufficiently simple and easy for coding , and avoid the possible pollution from the boundary condition and curvilinear meshes . to introduce a few tests which can be truthfully used to evaluate the performance of higher - order scheme
is the motivation for the current paper .
our selected examples include the following : one - dimensional cases , two - dimensional riemann problems , and the conservation law with source terms .
for the one - dimensional problems , the first case is a highly oscillatory shock - turbulence interaction problem , which is the extension of shu - osher problem by titarev and toro @xcite with much more severe oscillations , and the second one is a large density ratio problem with a very strong rarefaction wave in the solution @xcite , which is used to test how a numerical scheme capture strong waves .
for the two - dimensional cases , four groups are tested .
( i ) hurricane - like solutions @xcite , which are highly nontrivial two - dimensional time - dependent solutions with one - point vacuum in the center and rotational velocity field .
it is proposed to test the preservation of positivity and symmetry of the numerical scheme .
( ii ) the interaction of planar contact discontinuities for different mach numbers .
the multidimensional contact discontinuities are the composite of entropy waves and vortex sheets .
the simulation of such cases have difficulties due to the strong shear effects .
since the large mach number limits for these cases have explicit solutions @xcite , they are proposed here in order to check the ability of the current scheme for capturing wave structures of various scales and the asymptotic property .
( iii ) interaction of planar rarefaction waves with the transition from continuous fluid flows to the presence of shocks .
( iv ) further interaction of planar shocks showing the mach reflection phenomenon .
these two - dimensional problems fall into the category of two - dimensional riemann problems proposed in @xcite .
the two - dimensional riemann problems reveal almost all substantial wave patterns of shock reflections , spiral formations , vortex - shock interactions and so on , through the simple classification of initial data .
the rich wave configurations conjectured in @xcite have been confirmed numerically by several subsequent works @xcite .
since the formulation of these problems are extremely simple , there is no need of complicated numerical boundary treatment and they are suitable as benchmark tests . the case for the conservation law with source term
is also proposed . in order to provide reference solutions for all these test cases .
a gas - kinetic scheme will be used to calculate the solutions in this paper .
recently , based on the time - dependent flux function of the generalized riemann problem ( grp ) solver @xcite , a two - stage fourth - order time - accurate discretization was developed for lax - wendroff type flow solvers , particularly applied for the hyperbolic conservation laws @xcite .
the reason for the success of a two - stage l - w type time stepping method in achieving a fourth - order temporal accuracy is solely due to the use of both flux function and its temporal derivative . in terms of the gas evolution model , the gas - kinetic scheme provides a temporal accurate flux function as well , even though it depends on time through a much more complicated relaxation process from the kinetic to the hydrodynamic scale physics than the time - dependent flux function of grp . based on this time - stepping method and the second - order gas - kinetic solver @xcite ,
a fourth - order gas - kinetic scheme was constructed for the euler and navier - stokes equations @xcite . in comparison with the formal one - stage time - stepping third - order gas - kinetic solver @xcite ,
the fourth - order scheme not only reduces the complexity of the flux function , but also improves the accuracy of the scheme , even though the third - order and fourth - order schemes take similar computation cost .
the robustness of the fourth - order gas - kinetic scheme is as good as the second - order one .
numerical tests show that the fourth - order scheme not only has the expected order of accuracy for the smooth flows , but also has favorable shock capturing property for the discontinuous solutions .
this paper is organized as follows . in section 2
, we will briefly review the fourth - order gas - kinetic scheme . in section 3 ,
we select several groups of problems to show the performance of the scheme .
the final conclusion is made in the last section .
in this section , we will briefly review our recently developed two - stage fourth - order gas - kinetic scheme .
this scheme is developed in the framework of finite volume scheme , and it contains three standard ingredients : spatial data reconstruction , two - stage time stepping discretization , and second - order gas - kinetic flux function .
the spatial reconstruction for the gas - kinetic scheme contains two parts , i.e. initial data reconstruction and reconstruction for equilibrium . in this paper , the fifth - order weno method @xcite is used for the initial data reconstruction .
assume that @xmath0 are the macroscopic flow variables that need to be reconstructed .
@xmath1 are the cell averaged values , and @xmath2 are the two values obtained by the reconstruction at two ends of the @xmath3-th cell . the fifth - order weno reconstruction is given as follows @xmath4 where all quantities involved are taken as @xmath5 and @xmath6 are the nonlinear weights .
the most widely used is the weno - js non - linear weights @xcite , which can be written as follows @xmath7 where @xmath8 and @xmath9 is the smooth indicator , and the basic idea for its construction can be found in @xcite . in order to achieve a better performance of the weno scheme near smooth extrema
, weno - z @xcite and weno - z+ @xcite reconstruction were developed .
the only difference is the nonlinear weights .
the nonlinear weights for the weno - z method is written as @xmath10,\end{aligned}\ ] ] and the nonlinear weights for the weno - z+ method is @xmath11,\end{aligned}\ ] ] where @xmath9 is the same local smoothness indicator as in @xcite , @xmath12 is used for the fifth - order reconstruction , and @xmath13 is a parameter for fine - tuning the size of the weight of less smooth stencils . in the numerical tests , without special statement ,
weno - js method will be used for initial data reconstruction .
after the initial date reconstruction , the reconstruction of equilibrium part is presented .
for the cell interface @xmath14 , the reconstructed variables at both sides of the cell interface are denoted as @xmath15 . according to the compatibility condition , which will be given later
, the macroscopic variables at the cell interface is obtained and denoted as @xmath16 .
the conservative variables around the cell interface can be expanded as @xmath17 with the following conditions , @xmath18 the derivatives are given by @xmath19/\delta x.\end{aligned}\ ] ] the two - stage fourth - order temporal discretization was developed for lax - wendroff flow solvers , and was originally applied for the generalized riemann problem solver ( grp ) @xcite for hyperbolic equations . in @xcite ,
multi - stage multi - derivative time stepping methods were proposed and developed as well under different framework of flux evaluation .
consider the following time - dependent equation @xmath20 with the initial condition at @xmath21 , i.e. , @xmath22 where @xmath23 is an operator for spatial derivative of flux .
the time derivatives are obtained using the cauchy - kovalevskaya method , @xmath24 introducing an intermediate state at @xmath25 , @xmath26 the corresponding time derivatives are obtained as well for the intermediate stage state , @xmath27 then , a fourth - order temporal accuracy solution of @xmath28 at @xmath29 can be provided by the following equation @xmath30 the details of this analysis can be found in @xcite .
thus , a fourth - order temporal accuracy can be achieved by the two - stage discretization eq . and
eq .. consider the following conservation laws @xmath31 the semi - discrete form of a finite volume scheme can be written as @xmath32 where @xmath33 are the cell averaged conservative variables , @xmath34 are the fluxes at the cell interface @xmath35 , and @xmath36 is the cell size . with the temporal derivatives of the flux ,
the two - stage fourth - order scheme can be developed @xcite .
similarly , for the conservation laws with source terms @xmath37 the corresponding operator can be denoted as @xmath38 the two - stage fourth - order temporal discretization can be directly extended for conservation laws with source terms .
the two - dimensional bgk equation @xcite can be written as @xmath39 where @xmath40 is the gas distribution function , @xmath41 is the corresponding equilibrium state , and @xmath42 is the collision time .
the collision term satisfies the compatibility condition @xmath43 where @xmath44 , @xmath45 , @xmath46 is number of internal freedom , i.e. @xmath47 for two - dimensional flows , and @xmath48 is the specific heat ratio . to update the flow variables in the finite volume framework , the integral solution of bgk equation eq .
is used to construct the gas distribution function at a cell interface , which can be written as @xmath49 where @xmath50 is the location of the cell interface , @xmath51 and @xmath52 are the trajectory of particles , @xmath53 is the initial gas distribution function , and @xmath41 is the corresponding equilibrium state .
the time dependent integral solution at the cell interface @xmath14 can be expressed as @xmath54h(u)\nonumber\\ + & e^{-t/\tau}g_l[1-(\tau+t)(a_{1l}u+a_{2l}v)-\tau a_l)](1-h(u)).\end{aligned}\ ] ] based on the spatial reconstruction of macroscopic flow variables , which is presented before , the conservative variables @xmath55 and @xmath56 on the left and right hand sides of a cell interface , and the corresponding equilibrium states @xmath57 and @xmath58 , can be determined .
the conservative variables @xmath59 and the equilibrium state @xmath60 at the cell interface can be determined according to the compatibility condition eq . as follows @xmath61 the coefficients related to the spatial derivatives and time derivative @xmath62 and @xmath63 in gas distribution function eq . can be determined according to the spatial derivatives and compatibility condition .
more details of the gas - kinetic scheme can be found in @xcite . as mentioned in the section before , in order to utilize the two - stage temporal discretization , the temporal derivatives of the flux function need to be determined .
while in order to obtain the temporal derivatives at @xmath21 and @xmath64 with the correct physics , the flux function should be approximated as a linear function of time within the time interval .
let s first introduce the following notation , @xmath65 in the time interval @xmath66 $ ] , the flux is expanded as the following linear form @xmath67 the coefficients @xmath68 and @xmath69 can be fully determined as follows @xmath70 similarly , @xmath71 for the intermediate stage @xmath72 can be constructed as well .
thus , we have completed the review of three building block of fourth - order gas - kinetic scheme , i.e. spatial data reconstruction , two - stage temporal discretization , and second - order gas - kinetic flux solver .
more details about the implementation of the fourth - order gas - kinetic scheme can be found in @xcite .
in this section , we mainly focus on the numerical solutions of the compressible euler equations . based on the chapman - enskog expansion ,
the macroscopic equations can be derived from the bgk equation eq .. if the zeroth order truncation is taken , i.e. @xmath73 , the following two - dimensional euler equations can be obtained @xmath74 where @xmath75 . in this section
, a number of benchmark tests will be selected to test the performance of the fourth - order gas - kinetic scheme . the density distribution and local enlargement of titarev - toro problem at @xmath76 for weno - z+ weights with @xmath77 and @xmath36.,title="fig:",scaledwidth=47.5% ] the density distribution and local enlargement of titarev - toro problem at @xmath76 for weno - z+ weights with @xmath77 and @xmath36.,title="fig:",scaledwidth=47.5% ] the density distribution and local enlargement of titarev - toro problem at @xmath76 for weno - z+ weights with @xmath77 and @xmath36.,title="fig:",scaledwidth=47.5% ] the density distribution and local enlargement of titarev - toro problem at @xmath76 for weno - z+ weights with @xmath77 and @xmath36.,title="fig:",scaledwidth=47.5% ] for the inviscid flow , the collision time @xmath42 takes @xmath78 where @xmath79 , @xmath80 , @xmath81 and @xmath82 denote the pressure on the left and right sides of the cell interface . the reason for including artificial dissipation through the additional term in the particle collision time is to enlarge the kinetic scale physics in the discontinuous region for the construction of a numerical shock structure through the particle free transport and inadequate particle collision for keeping the non - equilibrium property in the shock region . in all simulations , the courant number @xmath83 . without special statement , the specific heat ratio @xmath84 .
the first one - dimensional problem is the extension of the shu - osher problem given by titarev and toro @xcite to test a severely oscillatory wave interacting with shock .
it aims to test the ability of high - order numerical scheme to capture the extremely high frequency waves .
the initial condition for this cases is given as follows , @xmath85 in the computation , @xmath86 uniform cells are used , and the density distributions at @xmath87 are presented .
weno - js , weno - z and weno - z+ methods are used to test the performance of different nonlinear weights in the weno reconstruction .
the density distribution and local enlargement of titarev - toro problem with different weights are given in fig.[toro-1 ] , where @xmath88 in weno - z+ methods .
similar to the results in @xcite , with a proper choice of the parameter @xmath13 , weno - z+ method performs much better than weno - js and weno - z methods . in order to check the role of @xmath13 in weno - z+ method ,
the density distributions are presented in fig.[toro-2 ] , for the cases with @xmath89 and @xmath36 .
it shows that the numerical performance is very sensitive withe choice of @xmath13 , which is also observed in @xcite .
however , the current scheme tends to provide better results than those in @xcite in all cases with @xmath90 , and @xmath36 due to the proper dynamics of the gas - kinetic flux solver .
large density ratio problem with density variation from @xmath86 to @xmath91 .
, title="fig:",scaledwidth=47.5% ] large density ratio problem with density variation from @xmath86 to @xmath91 . ,
title="fig:",scaledwidth=47.5% ] large density ratio problem with density variation from @xmath86 to @xmath91 .
, title="fig:",scaledwidth=47.5% ] large density ratio problem with density variation from @xmath86 to @xmath91 .
, title="fig:",scaledwidth=47.5% ] the second example is the large density ratio problem with a very strong rarefaction wave @xcite .
it is proposed to test the ability of a scheme for capturing strong waves .
the initial data @xmath92 it is a sod - type problem , and the solution consists of a very strong rarefaction wave with the density variation from @xmath93 to @xmath94 , a contact discontinuity jumping from @xmath94 to @xmath95 , and a shock wave jumping from @xmath91 to @xmath95 .
if uniform grid points are used , the jump of neighboring density values inside the rarefaction wave is about @xmath94 times the density jump in the shock layer .
hence the strong rarefaction wave is in a highly non - equilibrium state .
it was checked in @xcite that many high order numerical methods have to use very fine grid points to capture the rarefaction wave , the location of the shock and the contact discontinuity well . in this case , we start with the exact solution at time @xmath96 as an initial condition , and the output time for the solution is @xmath97 , with @xmath94 and @xmath98 uniform mesh points . the density and velocity distributions are shown in fig.[large - density-1 ] .
if we start with the initial data at @xmath99 , the numerical solution does not display so well with the same grid points although it is still acceptable .
however , as the rarefaction wave is not so strong , the scheme works well , which is verified through another example with the density and pressure jumps from @xmath93 to @xmath86 .
the density and velocity distributions are shown in fig .
[ large - density-2 ] . the grp simulation works better for such a case because the exact grp solver is used for the euler equations @xcite , especially at the initial time for the solution starting from the singular point .
this example shows that the flux is extremely important in the capturing of strong waves , refuting some misleading statement that the construction of numerical flux function is not so essential in the computation of fluid flows " . in this subsection ,
five groups of two - dimensional benchmark problems are presented to illustrate the performance of the current scheme .
hurricane - like solution : the critical case.,title="fig:",scaledwidth=47.5% ] hurricane - like solution : the critical case.,title="fig:",scaledwidth=47.5% ] + hurricane - like solution : the critical case.,title="fig:",scaledwidth=47.5% ] hurricane - like solution : the critical case.,title="fig:",scaledwidth=47.5% ] the first group of two - dimensional time - dependent solutions are the hurricane - like flow field , whose solution has one - point vacuum in the center with rotational velocity field .
the initial condition are given as @xmath100 where @xmath101 , @xmath102 is the initial entropy , @xmath103 . in this case
, the initial velocity distribution has a nontrivial transversal component , which makes the flow rotational .
the solutions are classified into three types @xcite according to the initial mach number @xmath104 , where @xmath105 is the sound speed .
_ 1 . critical rotation with @xmath106 . _ for this case , we have an exact solution with explicit formula .
this solution consists of two parts : a far field solution and a near - field solution .
the former far field solution is defined for @xmath107 , @xmath108 , @xmath109 and the near - field solution is defined for @xmath110 @xmath111 the curl of the velocity in the near - field is @xmath112 and the solution has one - point vacuum at the origin @xmath113 .
this is a typical hurricane - like solution that behaves highly singular , particularly near the origin @xmath113 .
there are two issues here challenging the numerical schemes : one is the presence of the vacuum state which examines whether a high order scheme can keep the positivity preserving property ; the other is the rotational velocity field that tests whether a numerical scheme can preserve the symmetry . in our computation
we take the data @xmath114 and @xmath115 .
numerical results are presented in fig.[rotation - c ] with the distribution of density and velocity field .
this case shows the robustness of the current scheme .
the positivity and symmetry are all preserved well .
high - speed rotation with @xmath116 .
_ for this case , the density goes faster to the vacuum and the fluid rotates more severe . in the computation @xmath117 and @xmath118 .
the numerical result is shown in fig .
[ high - low - rotation ] .
because of its high rotation speed , this case is more tough than the first one , and it also validates the robustness of the scheme . _ 3 .
low - speed rotation with @xmath119 .
_ this case is milder than the other cases above .
there is no vacuum in the solution , but it is still rotational with a low speed . in the computation @xmath120 and @xmath115 .
the numerical solution is presented in fig .
[ high - low - rotation ] .
we observe the symmetry of the flow structure is preserved well .
hurricane - like solution : the cases of high speed rotation and low speed rotation.,title="fig:",scaledwidth=47.5% ] hurricane - like solution : the cases of high speed rotation and low speed rotation.,title="fig:",scaledwidth=47.5% ] + hurricane - like solution : the cases of high speed rotation and low speed rotation.,title="fig:",scaledwidth=47.5% ] hurricane - like solution : the cases of high speed rotation and low speed rotation.,title="fig:",scaledwidth=47.5% ] the second group is about the interactions of planar contact discontinuities , whose solutions in the large mach number limit are singular and contain either vacuum or singular shocks or delta - shocks @xcite .
two typical cases related to the large mach number limit are provided in this group . as is well - known , the compressible fluid
flows become incompressible for the low mach number limit @xcite . on the other hand
, the large mach number limit leads to the pressureless model @xcite @xmath121 this system can be also regarded as the zero moment closure of the boltzmann - type equations @xcite or describe the single transport effect of mass , momentum .
the last equation of is decoupled from the first three equations and thus it is sufficient to consider the solution @xmath122 of the first three equations in eq .. the riemann - type initial conditions for eq .
are given as follows @xmath123 the negative contact discontinuity and positive contact discontinuity , which connect the @xmath124 and @xmath125 areas , are denoted as @xmath126 and @xmath127 respectively @xmath128 where @xmath129 are the normal velocity and @xmath130 are the tangential velocity .
two types of interaction of planar contact discontinuities are considered as follows .
the solution of the pressureless equations eq . with the data .
this figure is displayed in the self - similarity @xmath131plane .
the notation @xmath132 denotes the coordinate @xmath133 , @xmath134 .
, scaledwidth=45.0% ] density distribution and the local enlargement for the interaction of vortex sheets with same signs , where @xmath135.,title="fig:",scaledwidth=45.0% ] density distribution and the local enlargement for the interaction of vortex sheets with same signs , where @xmath135.,title="fig:",scaledwidth=45.0% ] density distributions for the interaction of vortex sheets with same signs , where @xmath136 and @xmath137.,title="fig:",scaledwidth=45.0% ] density distributions for the interaction of vortex sheets with same signs , where @xmath136 and @xmath137.,title="fig:",scaledwidth=45.0% ] density distributions for the interaction of vortex sheets with same signs , where @xmath136 and @xmath137.,title="fig:",scaledwidth=45.0% ] density distributions for the interaction of vortex sheets with same signs , where @xmath136 and @xmath137.,title="fig:",scaledwidth=45.0% ] _ 1 .
interaction of planar contact discontinuity with same sign vortex sheets _ @xmath138 .
the vacuum solutions will be provided by the large mach number limit@xcite .
for such a case , the initial data satisfies @xmath139 and in the computation , the initial data takes @xmath140 this initial pressure distribution @xmath141 is uniform and the density distribution could be arbitrary .
four planar contact discontinuities @xmath142 separate neighboring states and support the same sign vortex sheets . on the contact discontinuity @xmath143 , the density undergoes a jump , and
the vorticity is a singular measure @xmath144 where @xmath145 is the standard dirac function ( measure ) .
therefore this contact discontinuity is the composite of an entropy wave and a vortex sheet .
the same applies for @xmath146 , @xmath147 and @xmath148 . their instantaneous interaction results in a complex wave pattern .
the limiting system eq . has an explicit formula that consists of four constant states in @xmath149 and a vacuum state inside a pyramid with edges @xmath133 , i.e. @xmath150 the solution is schematically described in fig.[fig : pless ] .
we use the fourth order gas - kinetic scheme to look into the asymptotic process for this problem by letting the pressure to become smaller and smaller .
the initial pressure is taken to be @xmath151 and @xmath137 , respectively .
the density distributions are displayed in fig.[4j-1 ] and fig.[4j-2 ] at @xmath152 .
the uniform mesh with @xmath153 are used for @xmath154 and @xmath155 .
it is observed that as @xmath135 ( the mach number @xmath156 is relatively large ) , the numerical solution displays more small scale structures . with the increase of the initial mach number @xmath156
, the complicated flow structure disappears and the uniform mesh with @xmath157 are used for @xmath158 .
the solution becomes much closer to that given in fig.[fig : pless ] .
the delta - shock solution of the pressureless equations due to the interaction of vortex sheets .
this figure is displayed in the self - similarity @xmath131plane .
the notation @xmath132 denotes the coordinate @xmath133 , @xmath134 .
the dashed line denotes the support of delta - shocks.,width=302 ] _ 2 .
interaction of planar contact discontinuities with vortex sheets with different signs _ @xmath159 .
for such a case , the delta - shock solutions will be provided by the large mach number limit @xcite .
the initial data is so designed as to satisfy @xmath160 the initial pressure distribution is also uniform and the density distribution is arbitrary .
this case is different from the first case in this group .
the four planar contact discontinuities support vortex sheets of different signs and their interaction produces totally different flow patterns . with such initial data , the solution of eq . has a singular solution containing so - called delta - shocks , as shown in fig.[fig : delta - shock ] .
the solution formula is @xmath161 however , the solution becomes singular , particularly , the density takes a singular measure on the support @xmath162 in fig . [
fig : delta - shock ] @xmath163 where @xmath164 represents the support @xmath162 of the dirac measure in fig.[fig : delta - shock ] .
density distributions for the interaction of vortex sheets with different signs , with @xmath165 and @xmath166.,title="fig:",scaledwidth=45.0% ] density distributions for the interaction of vortex sheets with different signs , with @xmath165 and @xmath166.,title="fig:",scaledwidth=45.0% ] density distributions for the interaction of vortex sheets with different signs , with @xmath165 and @xmath166.,title="fig:",scaledwidth=45.0% ] density distributions for the interaction of vortex sheets with different signs , with @xmath165 and @xmath166.,title="fig:",scaledwidth=45.0% ] density distribution and the local enlargement for the interaction of vortex sheets with different signs , with @xmath167,title="fig:",scaledwidth=45.0% ] density distribution and the local enlargement for the interaction of vortex sheets with different signs , with @xmath167,title="fig:",scaledwidth=45.0% ] the numerical simulations are designed for the cases of different mach numbers .
the initial condition is given as follows @xmath168 the uniform mesh with @xmath153 is used .
the density distributions are presented in fig.[4j-3 ] with the initial pressure taken as @xmath165 and @xmath166 at @xmath169 , respectively . with the decrease of the pressure , or equivalently with the increase of initial mach number
, it is observed that more and more small scaled vortices are present in the solutions . for the case with @xmath167 , the density distribution is presented in fig.[4j-4 ] at @xmath170 .
the solution tends to be very close to the solution eq . and
this shows the capability of the current fourth order scheme to preserve the asymptotical stability for large mach number flow simulations .
interaction of planar rarefaction waves with initial condition eq .. ,title="fig:",scaledwidth=42.0% ] interaction of planar rarefaction waves with initial condition eq .. ,title="fig:",scaledwidth=42.0% ] interaction of planar rarefaction waves with initial condition eq .. ,title="fig:",scaledwidth=42.0% ] interaction of planar rarefaction waves with initial condition eq .. ,title="fig:",scaledwidth=42.0% ] in this group , we simulate the interaction of two - dimensional planar rarefaction waves , from which we observe the continuous transition from continuous flows to the presence of transonic shocks .
a global continuous solution is constructed in @xcite with a clear picture for the bi - symmetrical interaction only when the rarefaction waves involved are weak . in general
, such a bi - symmetric interaction may result in the presence of shocks , which never occurs in one - dimension since the interaction of one - dimensional rarefaction waves produces only continuous solutions @xcite .
the backward rarefaction wave and forward rarefaction wave , which connect the @xmath124 and @xmath125 areas , are denoted as @xmath171 and @xmath172 .
the four planar rarefaction waves @xmath173 with the following initial conditions are considered @xmath174 and @xmath175 the numerical results are presented in fig .
[ fig : interaction - rare1 ] - [ fig : interaction - rare2 ] , in which the uniform mesh with @xmath157 is used . for the solution with the initial condition
eq . , the planar rarefaction waves are strong and shocks are in the interior domain , for which there was a thorough study on how those shocks occur and on what the criterion is @xcite . on the contrary , if the planar rarefaction waves are relatively weak , e.g. with the initial data eq . , the global solution is continuous @xcite , but the density in the interior domain is quite low , as shown in fig.[fig : interaction - rare2 ] . so the simulation on the interaction of planar rarefaction waves is also multi - scaled and the numerical results in fig . [
fig : interaction - rare1 ] - [ fig : interaction - rare2 ] are fully consistent with the theoretical analysis in @xcite . the fourth group deals with the interaction of shocks , in which we observe the ability of the current scheme capturing the solution structures of small scales .
the study of shock wave interactions is always one of central topics in gas dynamics and related fields .
the benchmark problems of step - facing shock interaction and the oblique shock reflection presented in @xcite were carefully tested with three categories of numerical schemes . since then , almost all new designed schemes take those problems to display their numerical performance , which can be seen from the huge number of the google citations .
one example from a case of the two - dimensional riemann problems in @xcite with four planar shocks @xmath176 are tested , where the backward rarefaction wave and forward rarefaction wave , which connect the @xmath124 and @xmath125 areas , are denoted as @xmath177 and @xmath178 . to obtain the detailed flow structure with less computational mesh points , the initial condition
are given as follows @xmath179 for the shock wave interactions.,title="fig:",scaledwidth=45.0% ] for the shock wave interactions.,title="fig:",scaledwidth=45.0% ] the numerical solution is displayed in figure [ fig : shock - interaction ] , where the uniform mesh with @xmath180 is used .
this case is just the mathematical formation of the reflection problem of oblique shocks in @xcite and the symmetric line @xmath181 can be regarded as the rigid wall .
such a formulation can avoid the complexity of numerical boundary conditions and make the simulation simple so that any numerical scheme can perform it easily without worry about numerical treatment of boundary conditions .
the gas - kinetic scheme with second - order and fourth - order temporal accuracy are tested using the same fifth - order accurate weno spatial data reconstruction at each time level to simulate the wave patters resulting from the interaction of shocks .
it is evident that the second order scheme is much more dissipative than the four order accurate one .
the small scaled vortices are resolved sharply using the fourth - order gas - kinetic scheme , which hints that it may be suitable for the flow with small structure , such as turbulent flows @xcite .
rayleigh - taylor instability : density distribution with the mesh size @xmath182 at @xmath183 and @xmath184.,title="fig:",scaledwidth=23.0% ] rayleigh - taylor instability : density distribution with the mesh size @xmath182 at @xmath183 and @xmath184.,title="fig:",scaledwidth=23.0% ] rayleigh - taylor instability : density distribution with the mesh size @xmath182 at @xmath183 and @xmath184.,title="fig:",scaledwidth=23.0% ] rayleigh - taylor instability : density distribution with the mesh size @xmath182 at @xmath183 and @xmath184.,title="fig:",scaledwidth=23.0% ] rayleigh - taylor instability : density distribution with the mesh size @xmath185 at @xmath183 and @xmath184.,title="fig:",scaledwidth=23.0% ] rayleigh - taylor instability : density distribution with the mesh size @xmath185 at @xmath183 and @xmath184.,title="fig:",scaledwidth=23.0% ] rayleigh - taylor instability : density distribution with the mesh size @xmath185 at @xmath183 and @xmath184.,title="fig:",scaledwidth=23.0% ] rayleigh - taylor instability : density distribution with the mesh size @xmath185 at @xmath183 and @xmath184.,title="fig:",scaledwidth=23.0% ] the last group is the rayleigh - taylor instability to test the performance of the two - stage temporal discretization for the conservation laws with source terms .
rayleigh - taylor instability happens on an interface between fluids with different densities when an acceleration is directed from the heavy fluid to the light fluid .
the instability has a fingering nature bubbles of light fluid rising into the ambient heavy fluid and spikes of heavy fluid falling into the light fluid .
the initial condition of this problem @xcite is given as follows @xmath186 where @xmath187 is the sound speed and @xmath188 .
the computational domain is @xmath189\times[0,1]$ ] .
the reflective boundary conditions are imposed for the left and right boundaries ; at the top boundary , the flow values are set as @xmath190 , and at the bottom boundary , they are @xmath191 .
the source terms on the right side of the governing equations are @xmath192 . to achieve the temporal accuracy , @xmath193 is given by the cell averaged value @xmath194 and @xmath195 is given by the governing equation , respectively .
the uniform meshes with @xmath182 and @xmath196 are used in the computation .
the density distributions at @xmath183 and @xmath184 are presented in fig.[rayleigh - taylor1 ] and fig.[rayleigh - taylor2 ] . with the mesh refinement , the flow structures for the complicated flows are observed .
it hints that current scheme may be suitable for the flow with interface instabilities as well .
in this paper , we select several one - dimensional and two - dimensional challenging problems which can be used to check the performance of higher order numerical schemes .
the scheme for providing reference solutions is our recently developed two - stage four order accurate gks scheme .
these cases can be used to test the performance of other higher - order schemes as well , which have been intensively constructed in recent years .
the test problems are * one - dimensional problems 1 .
titarev - toro s highly oscillatory shock - entropy wave interaction ; 2 .
large density ratio problem with a very strong rarefaction wave ; * two - dimensional riemann problems 1 .
hurrican - like solutions with one - point vacuum and rotational velocity field ; 2 .
interaction of planar contact discontinuities , with the involvement of entropy wave and vortex sheets ; 3 .
interaction of planar rarefaction waves with the transition from continuous flows to the presence of shocks ; 4 .
interaction of planar ( oblique ) shocks . * conservation law with source terms 1 . rayleigh - taylor instability the construction of accurate and robust higher - order numerical schemes for the euler equations is related to many numerical and physical modeling issues . based on the exact riemann solution or other simplified approximate riemann solvers ,
the traditional higher - order approaches mostly concentrate on the different frameworks , such as dg , finite different , and finite volume , with all kind of underlying data reconstructions or limiters .
the higher - order dynamics under the flux function has no attracted much attention .
maybe this is one of the reasons for the stagnation on the development of higher - order schemes in recent years . along the framework of multiple derivatives for the improvement of time accuracy of the scheme
, it has a higher requirement on the accuracy of the flux modeling than the traditional higher - order methods , because the scheme depends not only on the flux , but also on the time derivative of the flux function .
most test cases in this paper are for the time accurate solutions , which require the close coupling of the space and time evolution .
these test cases can fully differentiate the performance of different kinds of higher - order schemes .
their tests and the analysis of the numerical solutions can guide the further development of higher - order schemes on a physically and mathematically consistent way .
the work of j. li is supported by nsfc ( 11371063 , 91130021 ) , the doctoral program from the education ministry of china ( 20130003110004 ) and the science challenge program in china academy of engineering physics .
the research of k. xu is supported by hong kong research grant council ( 620813 , 16211014 , 16207715 ) and hkust research fund ( provost13sc01 , irs15sc29 , sbi14sc11 ) .
j. glimm , x. ji , j. li , x. li , p. zhang , t. zhang and y. zheng , transonic shock formation in a rarefaction riemann problem for the 2d compressible euler equations , siam journal on applied mathematics , 69 ( 2008 ) 720 - 742 .
v. p. kolgan , application of the principle of minimum values of the derivative to the construction of finite - difference schemes for calculating discontinuous solutions of gas dynamics , scientific notes of tsagi , 3 ( 1972 ) 68 - 77 .
kreiss , j. lorenz , initial - boundary value problems and the navier - stokes equations .
classics in applied mathematics , 47 .
society for industrial and applied mathematics ( siam ) , philadelphia , ( 2004 ) .
w. e , y. g. rykov , y. g. sinai , generalized variational principles , global weak solutions and behavior with random initial data for systems of conservation laws arising in adhesion particle dynamics .
phys . 177 ( 1996 ) 349 - 380 . | there have been great efforts on the development of higher - order numerical schemes for compressible euler equations .
the traditional tests mostly targeting on the strong shock interactions alone may not be adequate to test the performance of higher - order schemes .
this study will introduce a few test cases with a wide range of wave structures for testing higher - order schemes . as reference solutions
, all test cases will be calculated by our recently developed two - stage fourth - order gas - kinetic scheme ( gks ) .
all examples are selected so that the numerical settings are very simple and any high order accurate scheme can be straightly used for these test cases , and compare their performance with the gks solutions .
the examples include highly oscillatory solutions and the large density ratio problem in one dimensional case ; hurricane - like solutions , interactions of planar contact discontinuities ( the composite of entropy wave and vortex sheets ) sheets with large mach number asymptotic , interaction of planar rarefaction waves with transition from continuous flows to the presence of shocks , and other types of interactions of two - dimensional planar waves .
the numerical results from the fourth - order gas - kinetic scheme provide reference solutions only .
these benchmark test cases will help cfd developers to validate and further develop their schemes to a higher level of accuracy and robustness .
euler equations , two - dimensional riemann problems , fourth - order gas - kinetic scheme , wave interactions . |
considerable observational evidence has built up over the past few years that a substantial fraction of the massive galaxies around us today were already massive at very early epochs .
this evidence comes primarily from three sources : * studies of local massive elliptical galaxies indicate that the stars in the most massive galaxies generally formed very early and over very short time intervals @xcite .
stars in less massive spheroids formed , on average , later and over longer time spans . *
massive galaxies in clusters show little evidence for significant evolution up to at least redshift @xmath1 ( _ e.g. , _ * ? ? ?
* ; * ? ? ?
* direct observations of massive galaxies at redshifts @xmath2 that are dominated by already old stellar populations show that significant numbers of massive galaxies were in place at even earlier epochs ( _ e.g. , _ * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
although the existence of massive galaxies at high redshifts is now well documented , there have been only a few high - resolution studies of their morphologies ( e.g. , @xcite ) .
morphologies are important , because they may well retain signs of the formation history of the galaxies .
this is particularly true for galaxies that show little or no recent star formation , so that we are able to observe relatively clean examples of the stellar population that formed earliest and that comprises the bulk of the mass of the galaxy . in this paper
, we present deep _ hubble space telescope _
( _ hst _ ) nicmos imaging of two galaxies with virtually pure old stellar populations at @xmath0 . in 2 ,
we briefly recount how these galaxies were selected . in 3 , we describe the observations and reduction procedures . in 4 and 5 ,
we analyze model fits to the images to determine morphologies , and in 6 we discuss the implications of our conclusions .
we assume a flat cosmology with @xmath3 km s@xmath4 mpc@xmath4 and @xmath5 .
our procedure for selecting galaxies with old stellar populations is described in some detail in @xcite ; here we give a brief synopsis .
we observe fields of radio sources in certain specific redshift ranges , selecting galaxies with photometric redshifts consistent with that of the radio source .
radio sources generally serve as beacons for some of the more overdense regions in the early universe .
furthermore , the specific redshift ranges selected are chosen to optimize discrimination with standard filter passbands between old stellar populations and highly reddened star - forming galaxies .
one of these redshift ranges is @xmath6 , for which the 4000 break , strong in old stellar populations , falls between the @xmath7 and @xmath8 bands .
we have used the @xcite ( bc03 ) spectral synthesis models , and , more recently , preliminary versions of the @xcite ( cb07 ) models , to evaluate and optimize our photometric selection of old stellar populations at various redshifts .
the preliminary cb07 models include more realistic prescriptions for thermally pulsing asymptotic - giant - branch stars ( @xcite ; see also @xcite ) .
although at low redshifts ( and for some seds at high redshifts ) the new models can significantly lower the masses estimated from @xmath9-band photometry , at the redshifts we are considering here for nearly pure old stellar populations , the masses ( and ages ) change hardly at all . the main effect of using the newer models is to reduce the amount of reddening required to obtain a good fit . if a stellar population were to have an age of 2 gyr at @xmath10 ( corresponding to all of the stars forming at @xmath11 ) , its observed colors would be @xmath12 and @xmath13 .
we use a photometric sieve procedure to optimize the selection with respect to available observing time , first obtaining relatively short @xmath7 and @xmath14 integrations ( typically 5 @xmath15 at @xmath16 and @xmath17 @xmath15 at @xmath18 ) .
if any objects with @xmath19 are found , we then obtain @xmath8 and deeper @xmath7 imaging .
finally , for fields with objects matching the expected spectral - energy distributions of an old stellar population at the redshift of the radio source , we attempt to obtain deep imaging at shorter wavelengths ( usually either @xmath20 or @xmath21 ) to set constraints on any residual star formation .
among the galaxies found by this technique are one each in the fields of the radio galaxy 4c23.56 @xcite and the quasar 4c05.84 .
we refer to these galaxies as 4c23.56er1 and 4c05.84er1 ; they are both luminous objects , and they have stellar populations that appear to be overwhelmingly dominated by old stars .
we obtained most of the near - ir observations ( @xmath7 , @xmath8 , and @xmath14 ) with the cisco ir camera @xcite on the 8.2 m subaru telescope @xcite in observing runs on 2000 november 8 ( ut ) , 2001 august 5 and 6 , and 2002 may 30june 1 .
the images have a scale of 0105 pixel@xmath4 and a field of @xmath22 .
in addition , we carried out deep @xmath20-band imaging of both fields with the echelle spectrograph and imager ( esi ; @xcite ) on the keck ii telescope on 2002 august 7 .
both the ir and optical imaging were reduced according to standard procedures using our own iraf scripts .
the calibrations used observations of ukirt faint standards @xcite for the ir photometry and landolt fields @xcite for the @xmath20-band imaging .
we also observed 4c23.56er1 at @xmath14 with the subaru 36-element curvature - sensing adaptive optics ( ao ) system @xcite and the infrared camera and spectrograph ( ircs ; @xcite ) on 2002 august 17 .
these results were reported by @xcite , but we will refer to them again in this paper .
we used ircs without the ao system , but with excellent natural seeing ( final images have fwhm of 035 ) to obtain a very deep image of the 4c05.84 field in the @xmath9 filter on 2004 august 1 .
finally , we obtained @xmath7-band imaging of the 4c05.84 field with nirc2 and the keck ii laser - guide - star adaptive - optics system on 2007 august 21 .
the nicmos observations used the nic2 camera ( 0075 pixel@xmath4 ) and the f110w and f160w filters .
they were obtained on ut 2005 january 3 ( 4c05.84er1 , f160w , total exposure 5376 s ) , 2005 january 4 ( 4c23.56er1 , f160w , total exposure 8192 s ) , 2005 january 8 ( 4c05.84er1 , f110w , total exposure 8448 s ) , and 2005 may 16 ( 4c23.56er1 , f110w , total exposure 11264 s ) as part of _ hst _ program 10418 . after doing a first - pass combination of the images to get a rough idea of the quality of the data
, we went back to the _ calnica _ processed images and corrected these for bias offsets and inverse flatfield effects using the stsdas _ pedsky _ task .
most of the f110w images were obtained in orbits impacted by passages through the south atlantic anomaly ( saa ) and needed special processing .
we used the idl routine saa_clean.pro @xcite to generate an image of the persistence from the routinely taken post - saa dark images and subtract it from the science images .
finally , for these images , _ pedsky _ was run again to remove any residual bias pedestals .
we then generated a bad - pixel mask from the data quality file , adding an additional mask for the coronographic occulter , which produces background that is detected in both filters .
most of the cosmic rays were removed with the contributed iraf procedure _ lacos_im _ @xcite . at this point
the images from the individual dither positions were combined onto a subsampled grid with the stsdas _ drizzle _ task to produce the final image . to choose the optimum _ drizzle _ parameters for our purposes
, we performed a series of tests with artificial psfs generated by tiny tim .
we ended up choosing a drop size of 0.7 and a subsampling factor of 2 .
the final combined images had fwhm of 0133 for the f160w images and
0115 for the 4c23.56er1 f110w image .
we were unfortunately unable to produce a useful image of 4c05.84er1 in the f110w band because of a combination of the object s low surface brightness at that wavelength and residual effects on the detector of the previous saa passage .
the final @xmath23 surface brightness limits ( in the vega system ) were @xmath24 and @xmath25 for 4c23.56er1 , and @xmath26 for 4c05.84er1 .
the drizzling process inevitably introduces some level of correlation between adjacent pixels .
where we needed to estimate absolute errors ( such as for our radial - surface - brightness plots ) , we made a statistical correction to the error determinations , following the prescriptions of @xcite although we obtained images of stars for point - spread - function ( psf ) determination at the ends of some of the orbits , psfs modeled from tinytim were quite consistent with the stellar profiles .
subtracting tinytim models from the observed stars gave residuals that were less than 1.5% ( rms ) of the peak over the central fwhm region ( and much lower outside this region ) , with maximum pixel deviations of 3% .
the tinytim profiles also have the advantage that they can be generated on a subsampled grid to minimize interpolation errors in matching the profiles to the undersampled nicmos2 images .
we accordingly used subsampled tinytim model profiles in our analysis .
in the field of the @xmath27 radio galaxy 4c23.56 , our photometric selection procedure picked out a galaxy that had previously been noted as a very red object by @xcite .
as mentioned above , we have already reported on our subaru ao / ircs imaging of 4c23.56er1(@xcite ; there , the galaxy is referred to as 4c23.56kc68 ) .
the main conclusions of that paper were that ( 1 ) the galaxy indeed has a redshift close to that of the radio galaxy , ( 2 ) the best fit to the photometry is an old ( 23 gyr ) stellar population with little reddening ( @xmath28 of @xcite extinction ) , and that ( 3 ) the morphology of this massive , old galaxy looked surprisingly disklike , with a projected axial ratio of 0.3 and a srsic index of 1.5 . because the details of the previous observations are available in that paper , and in a brief follow - up report @xcite , we restrict our discussion here to a comparison of the nicmos imaging with the previous ao imaging .
the nicmos f110w and f160w images are shown in fig .
[ 4c23mos ] , along with the best - fit srsic models ( convolved with the psf ) , the residuals from the subtraction of the models from the data , and , again , the best - fit models ( but _ without _ convolution with the psf ) .
we determine total magnitudes for 4c23.56er1 from the srsic models , finding ( on the vega system ) @xmath29 and @xmath30 after correction for galactic reddening @xcite .
the quoted uncertainties include only sky noise and uncertainty in the sky level ; they do not include any deviations between the models and the data ( which are in any case quite small over the region of good s / n for the data ) or other potential systematic effects .
we discuss the f160w image first , since it has a much higher s / n ratio than does the f110w image .
figure [ 4c23hrsb ] shows the radial - surface - brightness profile for the f160w image of 4c23.56er1 , along with the best - fit @xmath31-law , exponential , and srsic profiles , determined using galfit @xcite . among these
, the srsic profile clearly gives the best fit , as expected , because of the extra degree of freedom in the model .
the srsic profile has an index @xmath32 , an effective radius @xmath33 , and axial ratio @xmath34 . the uncertainties in @xmath35 and @xmath36 have been estimated by re - running the models with the sky level set @xmath37 above and below its median value . from our subaru ao imaging in the @xmath14 band @xcite , we had obtained @xmath38 , @xmath39 , @xmath40 , so the two independent profiles in different bands are in remarkably good agreement .
although the ao imaging had slightly better fwhm , and the two datasets had similar s /
n near the center of the galaxy , the nicmos2 data extends farther in semi - major axis because of its lower sky background .
we show a comparison of the two profiles in the region of overlap in fig .
[ 4c23drsbp ] .
both the @xmath31-law and exponential profiles fit the observed profile poorly . adding a small ( @xmath41 ) , weak ( 14% of total light ) bulge component to an exponential profile with an @xmath42
gives a fit that is as good as that of the srsic profile within a semi - major axis of 06 but somewhat worse beyond this radius . the f110w image shown in fig .
[ 4c23mos ] , which samples the morphology shortward of the 4000 break ( assuming that 4c23.56er1 has the same redshift as 4c23.56 itself ) , superficially has an even more `` disky '' appearance than does the f160w image .
this is partly due to the sharper psf at this wavelength : notice that the best - fit srsic models without psf convolution look much more similar than do the models with psf convolution .
nevertheless , there may be a detectable difference in morphology in the two bands .
the f110w srsic model has an index @xmath43 ( _ i.e. , _ essentially a pure exponential ) , an effective radius of @xmath44 , and @xmath45 .
the radial - surface - brightness profile is shown in fig .
[ 4c23jrsb ] , along with the best - fit srsic model and the f160w srsic model ( adjusted by a constant magnitude offset to approximately match the f110w points ) .
the observed differences are barely significant , given the uncertainties , but they seem to indicate a small color gradient , such that the outer parts of the galaxy are slightly bluer ( at least out to a semi - major axis of 07 , at which point the uncertainty in the sky background level becomes dominant ) .
this can not be a large effect because of the tight upper limits on the @xmath20 and @xmath21-band magnitudes ( see @xcite ) .
nevertheless , it does suggest a possible slight decrease in mean age and/or mean metallicity of the stellar population as one progresses from the center to the outskirts of the galaxy .
4c05.84er1 was found in the field of the @xmath46 quasar 4c05.84 ( fig . [ 4c05field ] ) .
the sed of 4c05.84er1 is shown in fig .
[ 4c05sed ] , including photometry from our spitzer irac images , which will be discussed elsewhere in more detail in the context of a larger sample of objects . while , for 4c23.56er1 , only upper limits at @xmath20 and @xmath21 bands have been obtained , for 4c05.84er1 we have detections at @xmath47 and @xmath48 , indicating the presence of some younger stars . in attempting to fit the observed sed ,
we have explored a range of exponentially decreasing star - forming models as well as instantaneous burst models ; we have also considered models with metallicities of solar , 0.4 solar , and 2.5 solar .
the formal best - fit model sed is at a redshift of 2.93 , substantially higher than that of 4c05.84 itself .
this model has a 0.4-solar - metallicity population with an age of 900 myr , an exponential time constant of 100 myr , and a reddening @xmath49 mag .
if we restrict ourselves to models with redshifts close to that of the quasar , we get a reasonable fit with a solar - metallicity model with a redshift of 2.40 , an age of 1.02 gyrs , an exponential time constant of 200 myr , and a reddening @xmath50 mag .
this model fits the @xmath21-band and irac 5.8 @xmath51 m photometry less well but the irac 7.9 @xmath51 m photometry slightly better . both of these models
are shown in fig .
[ 4c05sed ] .
we have no firm grounds for choosing one of these seds over the other ; but , given the uncertainties in the models and possible star - formation histories , we will accept for the remainder of this paper that the redshift closer to that of 4c05.84 itself is the correct one . in either case
, we are dealing with a massive galaxy comprising stars that mostly formed @xmath1 gyr before the observed epoch .
we show our nicmos2 f160w image of 4c05.84er1 in fig .
[ 4c05mos ] , along with our best - fit galfit model .
we have tried a series of models , including , again , @xmath31-law , exponential , and srsic .
radial - surface - brightness profiles of these are shown in the left panel of fig .
[ 4c05hrsb ] . in this case
, even the srsic profile is not a particularly good fit .
we get a significantly better fit with a two - component model incorporating a small @xmath31-law bulge comprising @xmath52% of the light and an exponential disk accounting for the rest .
this model is compared with the best srsic profile fit in the right panel of fig .
[ 4c05hrsb ] .
for the two - component model , the disk component has an effective radius @xmath53 , and the bulge component has @xmath54 .
this best - fit model gives a total magnitude ( on the vega system ) @xmath55 .
there is some evidence from our @xmath20 and @xmath21-band imaging and our recent keck ao imaging in the @xmath7 band that the bulge component virtually disappears at these shorter wavelengths , indicating that the two morphological components have different stellar populations .
such a result that would not be surprising .
the sed shown in fig .
[ 4c05sed ] would then be the linear combination of the two seds , with the bulge likely being a nearly pure old population and with younger stars being confined to the disk component .
we stress , however , that even the disk component must be dominated by old ( @xmath56 few hundred myr ) stars , with little very recent star formation .
we have experimented with a range of combinations of seds at the quasar redshift , but none with simple star - formation histories ( instantaneous bursts or exponentially decaying bursts ) gave a significantly better fit than did the single - population seds shown in fig .
[ 4c05sed ] .
we will explore this possibility in more detail elsewhere .
table [ tab1 ] summarizes the parameters for the two galaxies .
the morphologies of both 4c23.56er1 and 4c05.84er1 appear to be dominated by disks of old stars .
however , the disks are quite different in scale .
4c05.84er1 , at least , also appears to have a small bulge comprising about 1/3 of the total light in the f160w filter ( @xmath57 , rest frame ) .
we can not exclude the possibility that 4c23.56er1 also has a weak bulge , with up to @xmath58% of the total light in the f160w filter ; indeed , if the slight apparent difference in morphology between the f160w and f110w images is real , such a difference would seem to favor this possibility .
but it is the presence of massive , old disks that continues to give the strongest constraint on formation mechanisms .
such disks also have been seen at redshifts @xmath59 , where normal ellipticals with @xmath31-law profiles are also found @xcite .
it is difficult to imagine that these massive disks could have formed via any process other than the dissipative collapse of a large cloud of gas .
such disks are also unlikely to have survived major merging events , although the bulge component in 4c05.84er1 may testify to either some level of minor merging activity or bulge building via disk instabilities . for galaxies at @xmath0 ,
the evidence for a dominant old stellar population depends on the inflection in the sed shortward of the @xmath8 band , and establishing this inflection with optical / near - ir photometry depends on the relatively short baseline from the @xmath8 to the @xmath9 band .
furthermore , at the present epoch , essentially all strongly disk - dominated galaxies show evidence for continued star formation .
it is therefore not too surprising that claims of passive disks at high redshift should be doubted ( _ e.g. , _ * ? ? ?
however , as fig .
[ 4c05sed ] shows , _ spitzer _ irac data is entirely consistent with the sed of a moderately old stellar population , and no plausible sed incorporating very recent star formation combined with dust would fit the observed photometry .
we have recently also obtained irac imaging of the field of 4c23.56 , and our analysis of these data shows that the irac photometry falls squarely on our best - fit solar - metallicity bc03 model determined from the optical / near - ir photometry alone : an instantaneous burst with an age of 2.6 gyr and an extinction @xmath60 mag @xcite . using the more recent preliminary cb07 models , with their improved treatment of agb stars ,
we obtain a stellar population age of 2.8 gyr with @xmath61 .
again , no plausible model with significant star formation and reddening would fit these data .
l c c c c 4c23.56er1 & f110w & @xmath62 & @xmath44 & @xmath63 + 4c23.56er1 & f160w & @xmath64 & @xmath65 & @xmath66 + & & 1.00 & @xmath67 & @xmath68 + [ 0pt]4c05.84er1 & [ 0pt]f160w & 4.00 & @xmath69 & @xmath70 + [ tab1 ] masses for these galaxies can be estimated from the model fits .
assuming solar metallicities and a @xcite initial mass function , we obtain a mass of @xmath71 for 4c23.56er1 and @xmath72 for 4c05.84er1(assuming the model at @xmath73 with @xmath74 ) .
while the stellar - population age of 4c05.84er1 indicates that the last major star - formation episode occurred at @xmath75 , when the universe was @xmath76 gyr old , 4c23.56er1 has a stellar - population age that is formally slightly greater than the age of the universe at @xmath27 . clearly , the likely errors in the age determination and the usual caveats regarding the age - metallicity degeneracy mitigate any implied paradox
nevertheless , this massive galaxy must have formed at a very high redshift .
models with [ fe / h ] @xmath77 give an age of 1.9 gyr , but with a significantly worse fit .
it therefore seems likely that galaxy formation models will have to allow for the presence of early - forming massive disks .
this means that , at least in some dense regions , it has been possible to form @xmath78 @xmath79 of stars within a relatively short time via dissipative collapse and without the aid of major mergers . while our selection criteria have ensured that the galaxies we have discussed here comprise essentially pure old stellar populations , they may well be representative of many massive galaxies at high redshift , most of which would not be in our sample if they retained even tiny amounts of residual star formation or if they had had any significant star formation within a few hundred myr prior to the epoch at which we observe them .
4c05.84er1 has a luminosity and an effective radius that are similar to those of many local galaxies .
our best - fitting srsic model has @xmath80 kpc . for comparison , for galaxies of similar mass from the sloan survey with srsic @xmath81 , @xcite find @xmath82 kpc .
this galaxy could become , with passive evolution and perhaps a few minor mergers to increase the bulge - to - disk ratio somewhat , a typical s0 galaxy at the present epoch . on the other hand
, we do not see galaxies like 4c23.56er1 at the present epoch . by the prescription of @xcite , a low - srsic - index galaxy with the mass of 4c23.56er1 would have @xmath83 kpc , but 4c23.56er1 actually has @xmath84 kpc .
this means that the stellar mass surface density is much higher than for local galaxies , a result that has also been found for other samples of distant red galaxies ( _ e.g. , _ * ? ? ?
* ; * ? ? ?
it would seem that the only likely path for such galaxies to evolve to objects consistent with the local population of galaxies is through dissipationless mergers .
there is recent evidence that the most massive galaxies in the local universe are likely the result of dry mergers of galaxies with stars that are already old and with very little gas ( e.g. , * ? ? ?
* ) . with the constraint
that these merging components must themselves mostly be fairly massive ( to avoid a large dispersion and flattening in the observed color
magnitude relation for present - day massive galaxies , _
@xcite ) , it seems possible that these early massive disks may well be among the sources for the old stars that today are found in the most massive elliptical galaxies .
we thank s. charlot and g. bruzual for providing us with preliminary versions of their new spectral synthesis models prior to publication .
we also thank the anonymous referee for a detailed reading of the paper and a number of specific suggestions that helped us improve it .
support for _ hst _ program no .
10418 was provided by nasa through a grant from the space telescope science institute , which is operated by the association of universities for research in astronomy , inc .
, under nasa contract nas 5 - 26555 . this research has also been partially supported by nsf grant ast03 - 07335 .
it made use of the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration .
abraham , r. g. et al .
2007 , , submitted [ astro - ph/0701779 ] bergeron , l. e. , & dickinson , m. 2003 , nicmos instrument science report 2003 - 010 [ http://www.stsci.edu/hst/nicmos/ documents / isrs / isr_2003 - 10.pdf ] bernardi , m. , hyde , j. b. , sheth , r. k. , miller , c. j. , & nichol , r. c. 2007 , , 133 , 1741 bower , g. , kodama , t. , & terlevich , a. 1998 , , 299 , 1193 bruzual , g. , & charlot , s. 2003 , , 344 , 1000 calzetti , d. , armus , l. , bohlin , r. c. , kinney , a. l. , kornneef , j. , & storchi - bergmann , t. 2000 , , 533 , 682 cimatti , a. , et al . 2004 ,
nature , 430 , 184 chabrier , g. 2003 , , 115 , 763 charlot , s. , & bruzual , g. 2007 , in preparation daddi , e. , et al .
2005 , , 626 , 680 de propris , r. , stanford , s. a. , eisenhardt , p. r. , holden , b. , & rosati , p. 2007
, , in press [ astro - ph/0702050 ] fruchter , a. s. , & hook , r. n. 2002 , , 114 , 144 fu , h. , stockton , a. , & liu , m. 2005 , , 632 , 831 hawarden , t. g. , leggett , s. k. , letawsky , m. b. , ballantyne , d. r. , & casali , m. m. 2001 , , 325 , 563 iye , m. et al . 2004 , , 56 , 381 iye , m. et al .
2003 , , 590 , 770 knopp , g. p. , & chambers , k. c. 1996 , , 109 , 367 kobayashi , n. , et al .
2000 , in proc .
spie 4008 : optical and ir telescope instrumentation and detectors , ed .
m. iye & a. f. moorwood , p. 1056 kriek , m. , et al .
2006 , , 649 , l71 labb , i. et al .
2005 , , 624 , l81 landolt , a. u. 1992 , , 104 , 340 leggett , s. k. et al .
2006 , , 373 , 781 mccarthy , p.
2004 , , 614 , l9 mcgrath , e. j. , stockton , a. , canalzio , g. , iye , m. , & maihara , t. 2007 , , submitted maraston , c. 2005 , , 362 , 799 marigo , p. , & girardi , l. 2007 , , 469 , 239 motohara , k. , et al .
2002 , , 54 , 315 nelan , j. e. , smith , r. j. , hudson , m. j. , wegner , g. a. , lucey , j. r. , moore , s. a. w. , quinney , s. j. , & suntzeff , n. b. 2005 , , 632 , 137 papovich , c. et al .
2006 , , 640 , 92 peebles , p. j. e. 2002 , asp conf .
ser . , 283 , 351 peng , c. y. , ho , l. c. , impey , c. d. , & rix , h .- w . 2002 , , 124 , 266 pierini , d. , maraston , c. , gordon , k. d. , & witt , a. n. 2005 , , 363 , 131 reddy , n. a. , steidel , c. c. , fadda , d. , yan , l. , pettini , m. , shapley , a. e. , erb , d. k. , & adelberger , k. l. 2006 , , 644 , 792 renzini , a. 2006 , , 44 , 141 scarlata , c. et al .
2007 , , in press [ astro - ph/0701746 ] schlegel , d. j. , finkbeiner , d. p. , & davis , m. 1998 , , 500 , 525 sheinis , a. i. , bolte , m. , epps , h. w. , kibrick , r. i. , miller , j. s. , radovan , m. v. , bigelow , b. c. , & sutin , b. m. 2002 , , 114 , 851 shen , s. , mo , h. j. , white , s. d. m. , blanton , m. r. , kauffmann , g. , voges , w. , brinkmann , j. , & csabai , i. 2003 , , 343 , 978 stockton , a. , canalizo , g. , & maihara , t. 2004 , , 605 , 37 stockton , a. , & mcgrath , e. 2007 , asp conf ser , in press [ astro - ph/0702130 ] stockton , a. , mcgrath , e. , & canalizo , g. 2006 , , 650 , 706 takami , h. , et al .
2004 , , 56 , 225 thomas , d. , maraston , c. , bender , r. , & mendez de oliveira , c. 2005 , , 621 , 673 toft , s. , et al .
2007 , , in press [ arxiv:0707.4484 ] trujillo , i. et al . 2006 , , 650 , 18 van dokkum , p. g. 2001 , , 113 , 1420 van dokkum , p. g. , et al .
2004 , , 611 , 703 yan , l. , & thompson , d. 2003 , , 586 , 765 yan , l. , thompson , d. , & soifer , b. t. 2004 , , 127 , 1274 zirm , a. w. , et al .
2007 , , 656 , 66 | we present the results of nicmos imaging of two massive galaxies photometrically selected to have old stellar populations at @xmath0 .
both galaxies are dominated by apparent disks of old stars , although one of them also has a small bulge comprising about 1/3 of the light at rest - frame 4800 .
the presence of massive disks of old stars at high redshift means that at least some massive galaxies in the early universe have formed directly from the dissipative collapse of a large mass of gas .
the stars formed in disks like these may have made significant contributions to the stellar populations of massive spheroids at the present epoch . |
the heavy - ion collision can be divided into several successive processes including formation of quark - gluon plasma ( qgp ) and hadronization , etc .
some of such processes are believed to exist shortly after the big bang and explain the interior of compact stellar objects . on the other hand ,
the experimental devices are exclusively designed to detect hadrons , leptons and electromagnetically interacting particles .
thus , quarks and gluons ca nt directly be detected and therefore properties such as viscosity of qgp still represent a great challenge for the particle scientists . the nonnegligible elliptic flow measured at the relativistic heavy - ion collider ( rhic ) and recently at the large hadron collider ( lhc ) is a striking observation referring to the fact that the viscosity property turns to be measurable in the heavy - ion collision experiments , on one hand and the hydrodynamic response of produced matter to the initial geometry leads to such particle production asymmetry , on the other hand .
most of the experimental signals revealing the qgp formation and the elliptic flow , for instance , strengthen with increasing the collision centrality .
furthermore , the peripheral collisions introduce remarkable effects , as well .
the local imbalance in the momentum that carried by the colliding nucleons generates a nonvanishing local angular - momentum @xcite and the relativistic motion of the spectator nucleons generates currents of net positive charge moving off - center ( opposite to each other ) . both types of nuclear collisions generate a huge magnetic field @xcite , @xmath0 . in quantum electrodynamics ( qed ) , phenomena such as magnetic catalysis in which the magnetic field dynamically generates masses , and meissner effect where the magnetic field changes the order of the phase transition in type - i superconductor from second to first are examples on the possible influence of such large magnetic field on the phase - transitions in quantum choromodynamics ( qcd ) and/or the properties of the hadronic and partonic matter .
influence of this magnetic field on the chiral phase - transition has been examined by using the linear - sigma model @xcite . in recent works ,
we have implemented the polyakov linear - sigma model ( plsm ) in characterizing various qcd phenomena and quantities .
thermodynamic quantities including normalized and non - normalized higher - order moments of the particle multiplicity have been investigated @xcite .
the chiral phase - structure of ( pseudo)-scalar and ( axial)-vector meson masses in thermal and dense medium has been determined @xcite . also , it was reported on some properties of qgp in nonzero magnetic field @xcite .
let us assume that the generated magnetic field ( @xmath1 ) is directed along @xmath2-axis . from magnetic catalysis @xcite and by using landau quantization , the magnetic field modifies the quarks dispersion relations , @xmath3^{1/2 } , \label{eq : moddisp}\end{aligned}\ ] ] where @xmath4 runs over all quarks flavors , @xmath5 is the quantized landau number , @xmath6 is related to the spin quantum number and @xmath7 ; @xmath8 . from dimensional reduction of the magnetic catalysis property @xcite ,
@xmath9 where @xmath10 stands for degenerate landau levels and @xmath11 is replaced by the quantum number @xmath12 in mean field approximation , the plsm grand potential is given as @xcite , @xmath13 the mesonic potential @xmath14 is assumed to vanish at high temperatures @xmath15 .
the purely mesonic potential is given in ref .
@xcite . assuming landau quantization and from magnetic catalysis ,
the quarks and antiquark contribution to plsm potential becomes @xcite @xmath16 \right .
\nonumber \\ & & \hspace*{3.7 mm } \left.+\ln \left [ 1 + 3\left(\phi^*+\phi e^{-\frac{e_{b , f } + \mu _ f}{t}}\right)\ ; e^{-\frac{e_{b , f } + \mu _
f}{t}}+e^{-3 \frac{e_{b , f } + \mu _ f}{t}}\right ] \right\},\hspace*{5 mm } \label{new - qqpotio}\end{aligned}\ ] ] where @xmath17 is given in eq .
( [ eq : moddisp ] ) and @xmath18 refers to the chemical potential of @xmath4-quark . in vanishing magnetic field , the quarks potential reads @xmath19 \right .
\nonumber \\
& & \hspace*{34.mm } \left .
+ \ln \left [ 1 + 3\left(\phi^*+\phi e^{-\frac{e_f+\mu _ f}{t}}\right)\times e^{-\frac{e_f+\mu
_ f}{t}}+e^{-3 \frac{e_f+\mu _ f}{t}}\right ] \right\ } , \hspace*{7 mm } \label{thermalomg}\end{aligned}\ ] ]
from green - kubo formula , the bulk viscosity can be related to the correlation functions of trace of the energy - momentum tensor .
@xmath20\right\rangle e^{i \omega t } , \end{aligned}\ ] ] where @xmath21 is the frequency of quark and gluon vibration . for a narrow frequency region , @xmath22 . from the relaxation time approximation
, the deviation of energy - momentum tensor from its local equilibrium which is corresponding to the difference between the distribution function near and at equilibrium , @xmath23 , can be used to determine bulk and shear viscosities , @xcite @xmath24 ^ 2 \ , f_{f } ( t , \mu ) \left[1 + f_{f } ( t , \mu ) \right ] , \hspace*{5 mm } \label{eq : gkbulk}\\ % \eta ( t,\mu , eb ) & = & \frac{1}{5\ , t } \ , \sum_f \frac{|q_f|b}{2\pi } \sum_\nu \int \frac{dp_z}{2\pi } \left(2-\delta_{0\nu}\right ) \ , \frac{|\vec{p}|^4 \tau_f}{e_{b , f}^2 } \ ; f_{f } ( t , \mu ) \left[1 + f_{f } ( t , \mu ) \right ] .
\label{eq : gkshear}\end{aligned}\ ] ] the relaxation time ( @xmath25 ) can be related to the quark decay constant ( @xmath26 ) . in doing this
, two models can be implemented : * the dynamic quasi - particle model ( dqpm ) which describes well the phenomenology of interacting massless quarks and gluons as non - interacting massive quasi - particles @xcite .
the flavor - blind reaction rates for quarks and gluons are inversely dependent on the particle width , @xmath27 and @xmath28 , respectively . * the transport theory with boltzmann master equation which can be used to calculate the momentum loss and the relaxation time , as well @xcite . to this end ,
screening of long - range quark - quark interactions is assumed for the process of interpenetrating quark plasmas .
for a spatially uniform quark plasma flowing with respect to @xmath29 plasmas with another type of flavors , the relative flow velocity is assumed to relax in time due to the collisions @xcite . in partonic and hadronic phases , the relaxation time , respectively , scales with the temperature @xmath30 $ ] for @xmath31 and @xmath32 for @xmath33 . in vanishing magnetic field and at zero chemical potential , the temperature dependence of bulk viscosity ( @xmath34 ) normalized to thermal entropy ( @xmath35 )
is presented in left - hand panel of fig .
[ viscosityb0 ] .
the ratio of shear viscosity to thermal entropy ( @xmath36 ) is given in right - hand panel . at temperatures @xmath37
, @xmath38 has a good agreement with the lattice calculations @xcite .
lsm @xcite , dqpm and njl models @xcite are also good with the lattice .
the sharp raise of bulk viscosity slightly above the phase transition ( as the entropy is assumed to remain constant ) indicates instability in the hydrodynamic flow of the plasma as confirmed in rhic observables @xcite .
the agreement between plsm and lattice qcd calculations for @xmath36 is excellent above @xmath39 .
the left - hand panel of fig .
[ viscosityb ] shows the results of @xmath40 vs. temperature in finite magnetic field .
again the results at @xmath41gev@xmath42 are depicted as reference ( solid curve ) .
the results at @xmath43gev@xmath42 ( solid ) , @xmath44gev@xmath42 ( dotted ) and @xmath45gev@xmath42 ( dot - dashed ) show a remarkable dependence on temperature .
for instance , the finite magnetic field seems to enhance the appearance of the characterizing peak at @xmath39 .
the peaks are related to minima . at very low temperature ,
all curves seem to bundle together to large @xmath40-values . in right - hand panel of fig .
[ viscosityb ] , the ratio @xmath36 is depicted as function of temperature at different values of @xmath46 .
the solid curve represents the results at a vanishing magnetic field , while @xmath47 , @xmath48 and @xmath49gev@xmath42 are given as dotted , dot - dashed curves , receptively .
kss limit is represented by dashed line . at low temperatures
, @xmath36 becomes very large .
a peak positioned at @xmath39 starts to appear when the magnetic field increases while the peak s height is not sensitive to @xmath1 .
the remarkable effect around @xmath39 refers to instability in the hydrodynamic flow of quarks and hadrons . at extremely high temperatures ,
the coupling becomes weak and the hadrons are entirely liberated into quarks and gluons .
thus , it is likely that the deconfinement phase - transition is completed . at temperatures @xmath37 the values of @xmath36 are not sensitive to the change in @xmath50 . the left - hand panel of fig .
[ fig : ebdepence ] shows the magnetic susceptibility ( @xmath51 ) , which is determined from the second derivative of plsm free energy with respect to @xmath1 , as function of temperature at @xmath44gev@xmath42 and a vanishing baryon chemical potential . the magnetic susceptibility is a dimensionless quantity referring to the degree of magnetization as a response to the non - zero magnetic field .
the results from plsm are compared with various lattice simulations ( symbols ) and with the hrg model .
it worthwhile to notice the negative slope of @xmath51 ( inside - box in middle panel ) .
this signals that the qcd matter is diamagnetic . at high temperatures ,
i.e. restoring broken chiral symmetry , we observe a transition between dia- and para - magnetism .
it is likely that the qcd matter is para - magnetic is very high temperatures . in right - hand panel , influences of the magnetic field strength on the chiral phase - diagram
are analysed and compared with recent lattice qcd results ( circles with errorbars ) @xcite .
the vertical bands indicate the value of the magnetic fields at rhic ( @xmath52 ) and lhc energies ( @xmath53 ) .
there is a small suppression in the chiral condensates , i.e. magnetic catalysis . in plsm
we implement two methods in order to determine the quasi - critical temperature ( dotted curve ) .
the first one implements the higher - order moments of quark multiplicity , i.e. normalized quark susceptibility @xmath54 ( solid curve ) .
the second one uses the intersection of light - quark chiral condensate @xmath55with the deconfinement order - parameters ( dashed curve ) .
we find that the critical temperature decreases with increasing @xmath1 , i.e. inverse magnetic catalysis .
an excellent agreement is apparently achieved between our calculations and the lattice qcd , especially at small magnetic field @xmath56 gev@xmath42 ( second method ) .
the first method agrees well with the lattice data at a wider range of magnetic fields @xmath57 gev@xmath42 .
again , the first method apparently overestimates lattice data at low temperature , while the second method slightly underestimates these at high temperature .
the magnetic field which is likely generated in peripheral heavy - ion collisions due to local imbalance in momentum carried out by the participants and off - center relativistic motion of spectator s electric charges seems to influence the transport properties ( bulk and shear viscosity ) in both hadron and quarks phases .
also , the quark - hadron phase transitions are remarkable influenced by such large magnetic field . in vanishing magnetic field , we find that increasing @xmath58 decreases @xmath40 . but at high temperatures , a small increase can be observed . in finite magnetic filed , there is a remarkable dependence on temperature ; appearance of characterizing peaks at @xmath39 ( enhanced with increasing @xmath1 ) and very large values of @xmath40 observed at low temperatures . for @xmath36 ,
the peaks around @xmath39 gradually appear with increasing @xmath1 but their heights seem not being sensitive to @xmath1 . at temperatures
@xmath37 , @xmath36 is also not sensitive to the change in @xmath50 .
the magnetic susceptibility @xmath51 which measures the degree of magnetization as a response of finite magnetic field seems to raise with the temperature .
both plsm and lattice qcd calculations are in good agreement with each other .
it is worthwhile to highlight the slightly negative @xmath51 takes place at low temperature .
this refers to diamagnetic properties of the qcd matter .
switching to high temperatures , i.e. restoring broken chiral symmetry , a transition between dia- and para - magnetic properties is observed . at very high temperatures ,
the qcd matter is likely para - magnetic . in nonzero magnetic field ,
we have implemented two methods in order to determine the critical temperatures with changing @xmath1 .
magnetic catalysis is clearly observed in qcd matter .
the agreement with the lattice calculations is fairly good .
this work is partly supported by the world laboratory for cosmology and particle physics ( wlcapp ) , http://wlcapp.net/
99 a. tawfik and n. magdy , _ thermodynamics and higher order moments in su(3 ) linear -model with gluonic quasi - particles _ , j. phys .
* g42 * , 015004 ( 2015 ) .
a. tawfik and a. diab , _
polyakov su(3 ) extended linear @xmath6-model : sixteen mesonic states in chiral phase - structure _ , phys . rev . * c91 * , 015204 ( 2015 ) .
a. tawfik and n. magdy , phys . rev . * c90 * , 015204 ( 2014 ) .
h. b. meyer , phys .
d * 76 * , 101701 ( 2007 ) ; + h. b. meyer , phys .
lett . * 100 * , 162001 ( 2008 ) .
s. sakai and a. nakamura , pos * lat2007 * , 221 ( 2007 ) .
a. nakamura and s. sakai , phys .
lett . * 94 * , 072305 ( 2005 ) .
f. karsch , d. kharzeev and k. tuchin , phys . lett .
b * 663 * , 217 ( 2008 ) .
r. marty , e. bratkovskaya , w. cassing , j. aichelin and h. berrehrah , phys .
c * 88 * , 045204 ( 2013 ) .
h. mao , j. jin and m. huang , _ phase diagram and thermodynamics of the polyakov linear - sigma model with three quark flavors _ , j. phys .
g * 37 * , 035001 ( 2010 ) .
p. k. kovtun , d. t. son and a. o. starinets , phys .
* 94 * , 111601 ( 2005 ) . | in peripheral heavy - ion collisions , localized , short - lived an extremely huge magnetic field can be generated .
its possible influences on the quark - hadron phase transition(s ) and the transport properties of the hadronic and partonic matter shall be analysed from the polyakov linear - sigma model .
our calculations are compared with recent lattice qcd calculations . |
the paradigm of cosmic inflation @xcite is one of the main pillars of modern cosmology .
not only does inflation account for the vast size of the observable universe and its high degree of homogeneity and isotropy on cosmological scales ; it also seeds the post - inflationary formation of structure on galactic scales . in this sense
, inflation is a key aspect of our cosmic past and part of the reason why our universe is capable of harboring life . from the perspective of particle physics ,
the origin of inflation is , however , rather unclear .
after decades of model building , there exists a plethora of inflation models in the literature @xcite . but
a consensus about how to embed inflation into particle physics is out of sight . in this situation
, it seems appropriate to take a step back and ask ourselves what avenues have been left unexplored so far . in particular , we should question our dearly cherished prejudices and re - examine whether inflation might not be connected to other high - energy phenomena which , up to now , have been taken to be mostly unrelated to inflation . as we are going to demonstrate in this paper ,
an important example in this respect might be the interplay between inflation and the spontaneous breaking of supersymmetry ( susy ) . in recent years
, the picture of supersymmetry as a solution to the hierarchy problem has become increasingly challenged by the experimental data .
the null results of susy searches at the large hadron collider ( lhc ) @xcite and the rather large standard model ( sm ) higgs boson mass of a @xmath4 @xcite indicate that supersymmetry , if it exists in nature , must be broken at a high scale @xcite . based on this observation
, one could feel tempted to give up on supersymmetry as an extension of the standard model altogether .
but this would not do justice to supersymmetry s other virtues . provided
that supersymmetry is broken at a high scale @xcite , such as in the minimal framework of pure gravity mediation ( pgm ) @xcite , it may no longer be responsible for stabilizing the electroweak scale .
but in this case , supersymmetry is still capable of providing a viable candidate for dark matter @xcite , ensuring the unification of the sm gauge couplings @xcite and setting the stage for a uv completion of the standard model in the context of string theory .
in addition , high - scale supersymmetry does not suffer from a number of phenomenological problems that low - scale realizations of supersymmetry breaking are plagued with .
a high susy breaking scale does away with the cosmological gravitino problem @xcite and reduces the tension with constraints on flavor - changing neutral currents and @xmath5 violation @xcite .
moreover , in pgm , the susy - breaking ( or `` polonyi '' ) field is required to be a non - singlet @xcite , which solves the cosmological polonyi problem @xcite . in this paper , we will now concentrate our attention to yet another intriguing feature of supersymmetry which comes into reach , once we let go of the notion that supersymmetry s main purpose is to solve the hierarchy problem in the standard model .
the spontaneous breaking of supersymmetry at a scale @xmath6 results in a nonzero contribution to the total vacuum energy density , @xmath7 .
if we allow @xmath6 to take values as large as , say , the unification scale , @xmath8 , this susy - breaking vacuum energy density might , in fact , be the origin of the inflationary phase in the early universe !
such a connection between inflation and supersymmetry breaking not only appears economical , but also very natural .
first of all , supersymmetry tends to render inflation technically more natural , independent of the scale at which it is broken .
thanks to the susy nonrenormalization theorem @xcite , the superpotential @xmath9 in supersymmetric models of inflation does not receive any radiative corrections in perturbation theory .
this represents an important advantage in preserving the required flatness of the inflaton potential .
besides , all remaining radiative corrections ( which can be collected in an effective khler potential @xmath10 to leading order @xcite ) scale with the soft susy - breaking mass scale @xcite and are , thus , under theoretical control .
supersymmetry , therefore , has the ability to stabilize the inflaton potential against radiative corrections ; and it is , thus , conceivable that supersymmetry s actual importance may lie in the fact that it is capable of taming the hierarchy among different mass scales in the inflaton sector rather than in the standard model .
second of all , the spontaneous breaking of global supersymmetry via nonvanishing f - terms , i.e. , via the oraifeartaigh mechanism @xcite , always results in a pseudoflat direction in the scalar potential @xcite .
together with the constant vacuum energy density @xmath7 , such a flat potential for a scalar field is exactly one of the crucial requirements for the successful realization of an inflationary stage in the early universe . in principle
, the necessary ingredients for inflation are , therefore , already intrinsic features of every oraifeartaigh model .
inflation may be driven by the susy - breaking vacuum energy density @xmath7 and the inflaton field may be identified with the pseudoflat direction in the scalar potential .
the main obstacle in implementing this idea in realistic models is gravity . here , the crucial point is that the vanishingly small value of the cosmological constant ( cc ) tells us that we live in a near - minkowski vacuum with an almost zero total vacuum energy density , @xmath11 .
note that , as pointed out by weinberg , this not a mere observation , but a necessary condition for a sufficient amount of structure formation in our universe , so that it can support life @xcite . in the context of supergravity ( sugra )
@xcite , the fact that @xmath11 means that the susy - breaking vacuum energy density @xmath7 must be balanced , with very high precision , by a nonvanishing vacuum expectation value ( vev ) of the superpotential , @xmath12 , @xmath13 \frac{\left<\left|w\right|\right>^2}{m_{\rm pl}^2 } \simeq 0 \ , , \quad \left<\left|f\right|\right > = \lambda_{\rm susy}^2 \ , , \label{eq : min}\end{aligned}\ ] ] where @xmath14 denotes the reduced planck mass . if the susy breaking scale @xmath6 is indeed of @xmath15 , the requirement of a zero cc results in a huge vev of the superpotential , which , in turn , generates dangerously large sugra corrections to the scalar potential .
these corrections then easily spoil the flatness of the potential and render inflation impossible @xcite .
the most attractive and , in fact , only way out of this problem is @xmath0 symmetry . under @xmath0 symmetry ,
the superpotential carries charge @xmath16 , so that @xmath17 , as long as @xmath0 symmetry is preserved @xcite . in other words , by imposing @xmath0 symmetry , we promote @xmath12 to the order parameter of spontaneous @xmath0 symmetry breaking , in analogy to @xmath18 which acts as the order parameter of spontaneous supersymmetry breaking . in the true vacuum ,
@xmath0 symmetry must be broken , so that @xmath19 , in order to satisfy the condition in eq . .
but this does not necessarily mean that @xmath12 must be nonzero during the entire cosmological evolution .
it is conceivable that , at early times , @xmath0 symmetry is , in fact , a good symmetry , so that @xmath20 , thereby `` switching off '' the most dangerous sugra corrections to the polonyi potential .
the main intention of our paper now is to present a minimal dynamical model , in which this is indeed the case , so that inflation driven by the susy - breaking vacuum energy density , i.e. , _
polonyi inflation _ , becomes a viable option .
let us now outline our general philosophy in , to some extent , simplified terms .
we suppose that , from the perspective of the low - energy effective theory , the breaking of supersymmetry and @xmath0 symmetry are two distinct dynamical processes , taking place in two different hidden sectors at two different times , @xmath21 and @xmath22 .
in particular , we assume that the parameters @xmath21 and @xmath22 may be sampled in the uv theory , so that in some patches of the early universe ( or in the landscape of string vacua @xcite for that purpose ) @xmath23 and in other patches @xmath24 .
note that this differentiation equally includes the case of a maximally symmetric initial state , @xmath25 , as well as the possibility that either supersymmetry or @xmath0 symmetry is already broken from the very beginning , @xmath26 .
our crucial observation is that regions in space where @xmath0 symmetry is broken before supersymmetry correspond to bubbles of anti - de sitter ( ads ) space with an ads radius @xmath27 equal to the inverse of the gravitino mass @xmath28 , while regions in space where supersymmetry is broken before @xmath0 symmetry correspond to bubbles of de sitter ( ds ) space with a ds radius @xmath29 equal to the inverse of the hubble parameter @xmath30 , @xmath31 \left<\left|w\right|\right>/m_{\rm pl}^2 \ , , \label{eq : bubbles } \\
\nonumber t_{\rm susy } < t_r \quad\rightarrow\quad & \textrm{\phantom{a}ds bubbles with radius } r_{\rm ds } = h^{-1 } \ , , & h & = \left<\left|f\right|\right>/\sqrt{3}/m_{\rm pl } \,.\end{aligned}\ ] ] as long as supersymmetry is unbroken , the ads bubbles represent open friedmann - lematre - robertson - walker ( flrw ) universes with a negative cc and an oscillatory scale factor .
the ds bubbles , on the other hand , turn , as long as @xmath0 symmetry is unbroken , into ( asymptotically ) flat flrw universes with a positive cc and an exponentially growing scale factor @xcite .
this is to say that the ds bubbles experience inflation , while the ads bubbles remain limited in their spatial extent .
furthermore , if we suppose that there is no other source of inflation present in the theory , this means that the ads bubbles will never develop into habitable universes .
intelligent observers and humans can , therefore , only life in regions where , initially , @xmath0 symmetry remains unbroken up to a certain high energy scale .
this applies , in particular , to our own observable universe . under the above assumptions , the inflationary period in the early history of our universe must have been a consequence of spontaneous supersymmetry breaking and @xmath0 symmetry
must have been broken only at late times , i.e. , after a sufficient amount of inflation .
put differently , we can say that supersymmetry breaking and weinberg s argument regarding the size of the cc _ postdict _ a period of inflation and late - time @xmath0 symmetry breaking in our cosmic past .
this conclusion also sheds new light on the role of the cc itself .
the fine - tuning of the cc is now a dynamical process that takes place only after inflation . in order to obtain zero cc
, we have to require that the gravitino mass generated during @xmath0 symmetry breaking matches the inflationary hubble rate , and @xmath32 in our scenario is conceptually quite different from other inflation models , such as , e.g. , the one in @xcite , where @xmath33 on purely phenomenological grounds . in our case
, @xmath33 is a property of the low - energy effective lagrangian of our universe that ensures that we live in a vacuum with an almost zero cc . ]
@xmath34 if @xmath0 symmetry breaking results in a gravitino mass smaller than @xmath32 , inflation never ends ; if it `` overshoots '' and the gravitino mass is eventually larger than @xmath32 , our universe becomes ads .
we , thus , recognize the requirement that inflation must terminate at one point or another as part of the reason why the cc in our universe is fine - tuned .
a certain amount of fine - tuning during late - time @xmath0 symmetry breaking is inevitable , as inflation would otherwise not exit into a near - minkowski vacuum .
this situation needs to be contrasted with standard scenarios of inflation , such as chaotic @xcite or hybrid @xcite inflation , where the vacuum energy density driving inflation is neither related to @xmath0 symmetry breaking nor to low - energy supersymmetry breaking .
these scenarios require an independent reason for the vanishing of the inflationary vacuum energy density ( e.g. , a tuning of the inflaton potential or some kind of waterfall transition ) , whereas in our case this reason is already inherent to the fine - tuning of the cc in the course of spontaneous @xmath0 symmetry breaking at the end of inflation .
the above reasoning is just a rough sketch . to construct a realistic model of polonyi inflation
, we need to be more specific .
this pertains , first of all , to the kind of @xmath0 symmetry that we have in mind .
naively , our first attempt might be to protect @xmath12 by means of a global @xmath35 symmetry . on general grounds ,
quantum gravity is , however , expected to explicitly break all global symmetries ( see @xcite and references therein ) , so that a global @xmath35 does not appear to be a viable possibility .
meanwhile , gauging a continuous @xmath0 symmetry is a subtle issue that easily results in conflicts with anomaly constraints at low energies ( see @xcite for a recent discussion ) .
this leaves us with a _
discrete _ gauged ( i.e. , anomaly - free @xcite ) @xmath0 symmetry , @xmath36 , as a unique choice to ensure the vanishing of the superpotential at early times .
interestingly enough , such a discrete @xmath0 symmetry readily comes with a number of other advantages in the context of susy phenomenology : ( i ) a discrete @xmath0 symmetry prevents too rapid proton decay via perilous dimension-5 operators @xcite ; ( ii ) it may give rise to an accidental approximate global peccei - quinn symmetry and , thus , help in solving the strong @xmath5 problem @xcite ; ( iii ) and it may account for the approximate global continuous @xmath0 symmetry which is required to realize stable @xcite or meta - stable @xcite susy - breaking vacua in a large class of models of dynamical supersymmetry breaking ( dsb ) .
moreover , if we restrict ourselves to the special case of a @xmath37 symmetry , @xmath0 symmetry can also help us in solving the @xmath38 problem @xcite in the minimal supersymmetric standard model ( mssm ) .
any discrete @xmath0 symmetry suppresses the bilinear higgs mass term ( i.e. , the @xmath38 term ) in the superpotential .
but only in the case of a @xmath37 symmetry , we are , in addition , allowed to include a higgs bilinear term in the khler potential , @xmath39 ( see @xcite and references therein for an extended discussion of this point ) .
we are then able to generate the @xmath38 term in the course of @xmath0 symmetry breaking @xcite , which directly relates the @xmath38 parameter to the gravitino mass .
problem is not to be confused with the giudice - masiero mechanism @xcite , which relates the generation of the @xmath38 term to the spontaneous breaking of supersymmetry rather than to @xmath0 symmetry breaking . ]
it is for this reason that we will assume a discrete @xmath37 symmetry in the following . for our purposes
, it will not be necessary to specify the origin of this @xmath37 symmetry .
but it is interesting to note that orbifold compactifications of the heterotic string have the ability to yield discrete @xmath0 symmetries in the low - energy effective theory . in this case , the discrete @xmath0 symmetry at low energies is nothing but a remnant of the higher - dimensional lorentz symmetry that survives the compactification of the internal space ( for early as well as more recent work on this topic , see @xcite and @xcite , respectively ) .
we also mention that the special case of a discrete @xmath37 symmetry has received particular attention in the context of orbifold compactifications in recent years @xcite .
the assumption of a discrete @xmath37 symmetry in the low - energy effective theory is , therefore , well motivated and stands theoretically on a sound footing .
second of all , inflation is more than just a pure ds phase .
it represents a stage of quasi - ds expansion , in the course of which the hubble parameter slowly varies .
we , thus , need to specify the dynamics of supersymmetry breaking more precisely and check whether the corresponding polonyi potential is , in fact , suitable for slow - roll inflation . here
, we shall work within the framework of dynamical supersymmetry breaking @xcite , in which supersymmetry is assumed to be broken by the dynamics of a strongly coupled susy gauge theory .
this gives us the advantage that the susy breaking scale @xmath6 is generated dynamically via dimensional transmutation .
as far as supersymmetry breaking is concerned , we will , therefore , not have to rely on any dimensionful input parameters .
instead , @xmath6 and , thus , the energy scale of inflation , @xmath40 , will be controlled by the dynamical scale @xmath41 of the susy - breaking hidden sector , @xmath42 in this sense , our inflation model should be regarded as a variant of _ dynamical inflation _
@xcite , as we assume the energy scale of inflation to be generated by strong dynamics .
similarly , our model is closely related to _ natural inflation _
@xcite , which treats the inflaton as an axion - like field that is likewise subject to a scalar potential generated by nonperturbative dynamics ( for recent dynamical implementations of natural inflation in field theory , see @xcite ) , as well as to _
modulus inflation _
@xcite , where one identifies the inflaton as a ( composite ) modulus in the effective low - energy regime of strongly coupled gauge theories . in our case ,
the role of the inflaton is played by the scalar component of the chiral polonyi field @xmath43 , which breaks supersymmetry via its nonzero @xmath44-term . in global supersymmetry and at the classical level ,
the scalar polonyi potential is exactly flat and , thus , an ideal starting point for the realization of inflation . at the quantum level and in supergravity , the polonyi potential
, however , receives corrections , which may or may not spoil the flatness of the potential . here , the sugra corrections lead , in particular , to the notorious @xmath45 problem @xcite , which typically requires a parameter fine - tuning at the level of @xmath46 or so . the quantum and gravity corrections to the effective scalar potential scale with the coupling strengths of the yukawa interactions between the polonyi field and matter fields , @xmath47 , as well as with the coefficients of the higher - dimensional polonyi terms in the effective khler potential , @xmath48 , respectively . in order to assess the prospects of successful polonyi inflation , one , therefore , has to study the viability of inflation as a function of the parameters @xmath47 and @xmath48 and identify those parameter ranges that lead to consistency with the observational data on the cosmic microwave background ( cmb ) @xcite . to this end
, we will present in this paper a _
minimal _ realization of the idea of polonyi inflation .
our model is based on two strongly coupled @xmath49 hidden gauge sectors ( featuring two quark / antiquark pairs each ) , which we take to be responsible for the dynamical breaking of supersymmetry and @xmath0 symmetry , respectively .
we supplement both sectors with an appropriate number of gauge singlet fields , so that the susy - breaking sector becomes identical to the simplest version of the iyit dsb model @xcite , while the @xmath0-symmetry breaking sector turns into a strongly coupled sqcd theory with a quantum mechanically deformed moduli space and field - dependent quark masses @xcite .
more precisely , we assume the quark masses in the @xmath0 symmetry - breaking sector to be controlled by the vev of a singlet , which we call @xmath50 .
inflation is then driven by the susy - breaking vacuum energy density in the iyit sector , which results in an inflaton potential equivalent to that of supersymmetric f - term hybrid inflation ( fhi ) @xcite , including corrections from supergravity @xcite as well as from higher - dimensional terms in the tree - level khler potential @xcite .
this is reminiscent of the inflation models presented in @xcite and @xcite .
the model in @xcite , however , corresponds to a reduced version of the iyit model with less singlets , which leads to the vanishing of the inflationary vacuum energy at the end of inflation .
contrary to our approach , it , thus , does not establish a connection between inflation and supersymmetry breaking .
meanwhile , the model in @xcite identifies a different singlet field , other than the polonyi field , as the inflaton .
this allows the author of @xcite to separate the scales of inflation and supersymmetry breaking by imposing a hierarchy among the yukawa couplings of the inflaton and the susy - breaking field .
we , on the other hand , will show how to implement inflation into the full iyit model , sticking to the notion that the best motivated inflaton candidate in the iyit model is still the polonyi field itself .
an important consequence of this approach is that , in our case , f - term hybrid inflation does _ not _ end in a waterfall transition in the inflationary sector .
instead , we simply retain the inflationary vacuum energy density at low energies , which then continues to act as the vacuum energy density associated with the spontaneous breaking of supersymmetry .
this also means that our `` waterfall transition - free '' scenario of f - term hybrid inflation does not suffer from the usual production of topological defects , such as cosmic strings in the case of a @xmath51 waterfall transition , which would otherwise exert some serious phenomenological pressure on our model ( see , e.g. , @xcite ) .
we assume that the susy - breaking sector and the @xmath0-symmetry breaking sector only communicate with each other via interactions in the khler potential ( i.e , not via interactions in the superpotential ) .
this is sufficient to stabilize the scalar field @xmath50 during inflation at @xmath52 by means of its hubble - induced mass . during inflation , the ( fermionic ) quarks in the @xmath0-symmetry breaking sector are
, therefore , massless and the discrete @xmath37 symmetry remains unbroken in this sector .
as anticipated , the superpotential then lacks a constant term during inflation , which relieves the inflationary dynamics from the most dangerous gravitational corrections in supergravity .
in particular , the inflaton potential is free of the notorious `` tadpole term '' linear in the inflaton field @xcite . towards the end of inflation , the hubble - induced mass of the scalar field @xmath50 decreases .
adding an appropriately chosen superpotential for the field @xmath50 , we can use this fact to trigger a waterfall transition _ in the @xmath0 symmetry - breaking sector _ at small inflaton field values .
the field @xmath50 then acquires a large vev and the quarks in the @xmath0 symmetry - breaking sector become very massive .
consequently , the @xmath0 symmetry - breaking sector turns into a pure super - yang - mills ( sym ) theory and @xmath0 symmetry becomes spontaneously broken via gaugino condensation @xcite .
this external waterfall transition is associated with the breaking of a @xmath53 parity , which is , however , only an approximate symmetry .
for this reason , we do not have to fear the production of topological defects ( i.e. , domain walls ) during the waterfall transition in the @xmath0 symmetry - breaking sector .
after these introductory remarks , we are now in the position to present our analysis .
the remainder of this paper is organized as follows . in the next section
, we will show how the iyit dsb model may give rise to dynamical f - term hybrid inflation . here , we will first argue why the original polonyi model @xcite of supersymmetry breaking is _ not _ sufficient for a successful realization of polonyi inflation , even if we assume zero constant in the superpotential during inflation .
we , therefore , conclude that we only have a chance of successfully realizing polonyi inflation in the presence of radiative corrections which leads us to consider the iyit model as a possible uv completion of the original polonyi model .
we then derive the scalar potential of f - term hybrid inflation in the iyit model and discuss its embedding into supergravity . as an interesting aside , we demonstrate that polonyi inflation is incompatible with the concept of an approximate shift symmetry in the inflaton direction @xcite . instead , it turns out that polonyi inflation requires a near - canonical khler potential . in sec .
[ sec : rbreaking ] , we expand on our mechanism of late - time @xmath0 symmetry breaking , showing how a small inflaton field value may trigger gaugino condensation in a separate hidden sector .
related to that , we comment on the backreaction of the @xmath0 symmetry - breaking sector on the inflationary dynamics and discuss how the two sectors of supersymmetry and @xmath0 symmetry breaking have to conspire to yield a zero cc in the true vacuum after inflation . in sec .
[ sec : pheno ] , we turn to the phenomenological implications of our scenario . here , we identify the viable region in parameter space that leads to agreement with the latest planck data on the inflationary cmb observables @xcite . as we are able to show , a scalar spectral index of @xmath54 can be easily achieved for an @xmath55 yukawa coupling , @xmath56 , and a slightly suppressed coefficient in the noncanonical khler potential , @xmath57 .
the amplitude of the scalar power spectrum , @xmath58 , fixes the dynamical scale of the susy - breaking sector to a value close to the unification scale , @xmath59 , suggesting that our setup may eventually be part of a grand unified theory ( gut ) . as a characteristic feature of polonyi inflation
, we highlight the fact that the relation between the gravitino mass and the inflationary hubble rate in eq .
directly translates into a one - to - one correspondence between the gravitino mass @xmath28 and the tensor - to - scalar ratio @xmath60 , @xmath61 at this point , it is interesting to note that the observed value of the scalar spectral amplitude , @xmath62 , might be the result of anthropic selection @xcite .
together with the paradigm of slow - roll inflation ( which implies @xmath63 ) , the anthropic value of @xmath58 could , therefore , explain why the soft susy mass scale is so much higher than the electroweak or tev scale .
moreover , we point out in sec . [
sec : pheno ] that , after inflation , the polonyi field mostly decays into gravitinos .
polonyi inflation is , thus , followed by a phase of gravitino domination @xcite , which leads to reheating around temperatures of @xmath64 .
this paves the way for thermal wino dark matter ( dm ) as well as thermal leptogenesis @xcite enhanced by resonance effects @xcite .
appendix [ app : potential ] contains some technical details regarding the derivation of the effective inflaton potential .
in particular , we show how the one - loop corrections to the effective potential may be obtained , to leading order , from an effective khler potential . in appendix
[ app : viability ] , we explain in more detail why polonyi inflation in the iyit model does not work , if the superpotential already contains a constant term from the very beginning .
this completes our argument that , in the context of our minimal model , successful polonyi inflation requires ( i ) radiative corrections , ( ii ) a near - canonical khler potential , ( iii ) as well as late - time @xmath0 symmetry breaking . among all possible choices regarding ( i ) the type of interactions that the polonyi field participates in , ( ii ) the shape of the khler potential , and ( iii ) the chronology of supersymmetry and @xmath0 symmetry breaking , this leave one unique possibility for how to realize polonyi inflation .
the polonyi model @xcite is the simplest oraifeartaigh model of supersymmetry breaking via a nonvanishing f - term .
its superpotential consists of a tadpole term and a constant , @xmath65 here , @xmath43 denotes the chiral polonyi field , @xmath38 is the scale of supersymmetry breaking and @xmath66 is a constant that breaks @xmath0 symmetry and which determines the potential energy density in the ground state . for a canonical khler potential and fine - tuning @xmath66 , so that it takes the particular value @xmath67 , this model has a minkowski vacuum at @xmath68 , in which supersymmetry is broken by the polonyi f - term , @xmath69 .
it has been known for a long time , that the scalar potential for the polonyi field around this vacuum is unfortunately too steep to support slow - roll inflation @xcite . in appendix
[ app : viability ] , we review and extend this argument , showing for various choices of the khler potential , that , with @xmath70 from the very beginning , the polonyi model does not give rise to inflation .
we consider , in particular , a canonical khler potential supplemented by higher - dimensional corrections as well as khler potentials featuring an approximate shift symmetry either along the real or the imaginary axis in the complex plane . in none of the cases under consideration inflation
is viable either because we fail to satisfy the slow - roll conditions or because the scalar potential does not exhibit a global minkowski vacuum in the first place .
this immediately raises the question whether inflation might perhaps become possible in the polonyi model , if we impose a discrete @xmath0 symmetry at high energies , so that @xmath71 initially .
let us address this question for a canonical khler potential supplemented by a higher - dimensional correction , , we will discuss the same question for an approximately shift - symmetric khler potential . in this case
, inflation turns out be unfeasible because , with the superpotential being given as @xmath72 , the @xmath73\left|w\right|^2/m_{\rm pl}^2 $ ] sugra term in the scalar potential induces a tachyonic mass for the polonyi field .
this results in a global ads minimum .
[ fn : shift ] ] @xmath74 here , we assume that the khler potential is always dominated by the canonical term , @xmath75 , also at field values above the planck scale .
an exhaustive study of arbitrary choices for the khler potential is beyond the scope of this paper . under this assumption
, the scalar potential always picks up a sugra correction , @xmath76 , which spoils the flatness of the potential at super - planckian field values . for this reason
, we only have a chance of realizing slow - roll inflation at field values below the planck scale . for the khler potential in eq .
, the scalar potential in supergravity then takes the following form , @xmath77 \ , , \quad v_0 = \mu^4 \ , , \label{eq : vpolonyi}\end{aligned}\ ] ] where the real scalar field @xmath78 denotes the _ canonically normalized _ radial component of the complex polonyi scalar @xmath79 contained in @xmath43 ( see sec .
[ subsec : sugra ] ) and where we have introduced @xmath80 as the susy - breaking vacuum energy density at @xmath81 . from the form of the scalar potential in eq .
, it is evident that , even with @xmath66 being set to zero , the polonyi model fails to yield successful inflation .
for instance , if we choose @xmath48 to be negative , the field @xmath78 is driven towards the origin by a positive mass squared , similarly as in chaotic inflation @xcite .
inflation may then take place at small field values close to the origin but not in accord with the observational data . to see this ,
consider the slow - roll parameters @xmath82 and @xmath45 , @xmath83 independent of the sign of @xmath48 , we have @xmath84 . for negative @xmath48 ,
the slow - roll parameter @xmath45 is , therefore , bound to be positive , while the slow - roll parameter @xmath82 turns out to be negligibly small during inflation , @xmath85 .
according to the slow - roll formula for the scalar spectral index , @xmath86 , we will then always obtain a blue - tilted scalar spectrum ( @xmath87 ) . in view of the latest best - fit value for @xmath88 reported by the planck collaboration ,
@xmath89 @xcite , such a spectrum is clearly ruled out by the observational data . on the other hand ,
if we choose @xmath48 to be positive , the polonyi field acquires a tachyonic mass around the origin and inflation proceeds from small to large field values , similarly as in new inflation @xcite .
it is then hard to imagine how the generation of the constant @xmath66 in the superpotential should be triggered , after a sufficient amount of inflation , at field values close or even above the planck scale .
but more than that , even if we assume that this problem could somehow be solved , the scalar potential in eq . still does not lead to an acceptable phenomenology . in new inflation ,
the scalar spectral index turns out to be bounded from above , @xmath90
@xcite , which deviates from the observed value by at least @xmath91 .
therefore , also for @xmath92 , we fail to reach consistency with the observational data . in summary
, we conclude that the _ bare polonyi model _ based on the superpotential in eq .
and for reasonable anstze regarding the shape of the khler potential does not allow for a successful realization of slow - roll inflation .
in respect of this null result , two comments are in order : ( i ) we emphasize that our analysis of the polonyi khler potential in this paper does _ not _ mount up to a general no - go theorem . in the most general case ,
the khler potential for the polonyi field is given by an arbitrary function @xmath44 of the field @xmath43 and its conjugate , @xmath93 . in absence of any other scale , it is , in particular , clear that @xmath94 can be the only relevant scale in the khler potential .
it may then well be that certain fine - tuned functions @xmath44 do allow for successful inflation in the polonyi model , after all ( see , e.g. , @xcite for a discussion of fine - tuned khler potentials in the context of sugra models of inflation ) .
in the following , we will , however , ignore the possibility of such a biasedly chosen khler potential and focus on the usual suspects : khler potentials that are either near - canonical or approximately shift - symmetric .
( ii ) the fact that we are unable to realize successful inflation in the bare polonyi model is not a serious problem , as the polonyi model is not expected to be a fundamental description of spontaneous supersymmetry breaking , anyway .
it should rather be seen as the effective theory resulting from some uv dynamics that provide a dynamical explanation for the origin of the parameters @xmath38 and @xmath66 in eq . .
from this perspective , it is then more likely than not that the polonyi field is not only subject to its gravitational self - interaction , but that it also participates in yukawa interactions with heavy matter fields in the uv theory . in the corresponding _ effective polonyi model _ at low energies , these matter fields are integrated out , so that they no longer appear in the superpotential .
but their couplings to the polonyi field still yield radiative corrections to the scalar potential in eq . , which affect the inflationary dynamics .
in such a modified setup , i.e. , in the original polonyi model supplemented by radiative corrections , successful polonyi inflation may , therefore , very well be an option . in the following
, we will construct a minimal extension of the polonyi model where this is indeed the case .
we shall consider a minimal uv completion of the polonyi model consisting of two strongly coupled @xmath49 sectors that account for the dynamical origin of the parameters @xmath38 and @xmath66 , respectively and demonstrate that , in the presence of radiative corrections , successful polonyi inflation is indeed feasible for a natural choice of parameters values .
one of the simplest ways to generate the susy breaking scale @xmath38 in eq . is to identify the polonyi field as part of the iyit model the simplest vector - like model of dynamical supersymmetry breaking @xcite . in its most general formulation ,
the iyit model is based on a strongly coupled @xmath95 gauge theory featuring @xmath96 pairs of `` quark fields '' @xmath97 that transform in the fundamental representation of @xmath95 .
the gauge dynamics of this model are associated with a dynamical scale @xmath41 , which denotes the energy scale at which the @xmath95 gauge coupling formally diverges .
the low - energy effective theory below the dynamical scale exhibits a quantum moduli space of degenerate supersymmetric vacua , which is spanned by @xmath98 gauge - invariant composite flat directions ( or `` meson fields '' ) @xmath99 , @xmath100 here , the parameter @xmath45 is a dimensionless numerical factor , which ensures the canonical normalization of the meson fields @xmath99 at low energies . _
naive dimensionful analysis _ @xcite leads us to expect that @xmath45 should be of @xmath101 and , for definiteness , we will , therefore , simply set @xmath102 in the following .
the quantum moduli space of the iyit model is subject to the following constraint pertaining to the meson vevs , @xmath103 which represents the quantum mechanically deformed version of the classical constraint @xmath104 @xcite . in order to break supersymmetry in the iyit model
, one has to lift the flat directions @xmath99 , so that eq . is longer compatible with a vanishing vacuum energy density .
this is readily done by coupling the quark pairs @xmath105 to a corresponding number of @xmath95 singlet fields , @xmath106 , in the tree - level superpotential , @xmath107 where @xmath108 denotes a matrix of yukawa couplings with at most @xmath109 independent eigenvalues .
these yukawa couplings induce dirac mass terms for the meson and singlet fields at low energies , @xmath110 so that the singlet f - term conditions , @xmath111 , are incompatible with the deformed moduli constraint , @xmath112 .
supersymmetry is then broken la oraifeartaigh via nonvanishing f - terms . in global supersymmetry and for all yukawa couplings in eq .
being equal , @xmath113 , the anomaly - free global flavor symmetry @xmath114 of the iyit model is given as follows , @xmath115 here , the discrete @xmath116 symmetry is the anomaly - free subgroup of the anomalous @xmath51 that is contained in the full flavor symmetry at the classical level , @xmath117 . under the @xmath116 symmetry ,
all quarks carry charge @xmath118 , while the singlet fields carry charge @xmath119 . meanwhile , the presence of the global continuous @xmath0 symmetry is characteristic for a large class of dsb models @xcite . under @xmath35 ,
the quark and singlet fields carry charges @xmath120 and @xmath16 , respectively . in the susy - breaking vacuum of the iyit model , where @xmath121 and @xmath122 , @xmath0 symmetry , therefore , remains unbroken . as a global symmetry ,
the continuous @xmath0 symmetry of the iyit model is , of course , only an approximate symmetry , which we expect to be broken by quantum gravitational effects @xcite .
on the other hand , recall that we assume the discrete subgroup @xmath123 to be gauged ( see our discussion in sec .
[ subsec : ingredients ] ) .
this protects the quality of the @xmath35 symmetry ; and it is reasonable to assume that all gravity - induced @xmath35-breaking effects in the iyit sector are suppressed . in the following , we will , therefore , stick to the effective superpotential in eq . and neglect the possibility of small @xmath0 symmetry - breaking corrections .
from now on , let us restrict ourselves to the simplest version of the iyit model : a strongly coupled @xmath124 gauge theory featuring four matter fields @xmath97 and six singlet fields @xmath106 .
the non - abelian flavor symmetry @xmath125 then corresponds to a global @xmath126 , under which @xmath99 and @xmath106 transform as six - dimensional antisymmetric rank-2 tensor representations . here ,
note that @xmath126 is the double cover of @xmath127 .
this allows us to rewrite the meson and singlet fields as vector representations of @xmath127 , @xmath128 as we will see shortly , it will turn out to be convenient to study the susy - breaking dynamics of the iyit sector in terms of this `` @xmath127 language '' rather than in terms of the original `` @xmath126 language '' .
trading the mesons @xmath99 for the new fields @xmath129 , the pfaffian constraint in eq .
can be written as , @xmath130 which defines a sphere in the six - dimensional space spanned by the six meson coordinates @xmath129 .
an elegant way to enforce this constraint is to directly incorporate it into the effective superpotential @xcite , @xmath131 here , the field @xmath132 denotes a lagrange multiplier , the corresponding f - term condition of which , @xmath133 , is nothing but a reformulation of the moduli constraint in eq . .
the physical status of the field @xmath132 is unfortunately rather unclear and depends on whether ( uncalculable ) strong - coupling effects induce a kinetic term for @xmath132 or not .
if the field @xmath132 should be physical , it might represent a dynamical glueball field , @xmath134 , and the dimensionless coupling constant @xmath135 in eq . would be expected to take some value of @xmath55 . in that case , the pfaffian constraint in eq .
would be satisfied only approximately , depending on the competition between the different f - term conditions that enter into the determination of the true ground state .
if , on the other hand , @xmath132 should be unphysical , we would have to treat it as a mere auxiliary field . in that case
, the pfaffian constraint should be satisfied exactly , which would require us to eventually take the limit @xmath136 in our analysis . in the following , we will suppose that the lagrange multiplier field @xmath132 is , indeed , a physical ( glueball ) field and set @xmath137 for definiteness . in terms of the fields
@xmath129 and @xmath138 , the effective tree - level superpotential takes the following form , @xmath139 where we assume , w.l.o.g . , the yukawa couplings @xmath140 to be ordered by size , @xmath141 for all @xmath142 . note that we will refer to the smallest yukawa coupling , @xmath143 , simply as @xmath47 in the following , @xmath144 . for generic values of the six yukawa couplings
@xmath140 , the non - abelian flavor symmetry is completely broken , @xmath145 the total effective superpotential is given by the sum of @xmath146 in eq . and
@xmath147 in eq . ,
@xmath148 we mention once more that , as a consequence of the nonrenormalization theorem , the effective superpotential does not receive radiative corrections in perturbation theory @xcite . given this form of the total superpotential and assuming a quadratic khler potential for all meson and singlet fields , the global minimum of the resulting f - term scalar potential is located at @xmath149 where @xmath150 is undetermined at tree level .
the sign ambiguity is a consequence of the @xmath151 flavor symmetry ( see eq . ) .
moreover , the parameter @xmath152 measures how well the deformed moduli constraint is satisfied , @xmath153 for perturbative values of the yukawa coupling @xmath47 , i.e. , for @xmath154 or smaller , we have @xmath155 , which tells us that the deformed moduli constraint is fulfilled almost exactly . on the other hand , for nonperturbative values of @xmath47 ,
i.e. , for @xmath47 as large as @xmath156 , the parameter @xmath152 becomes of @xmath157 , indicating that the constraint function @xmath158 significantly deviates from zero . to put this result into perspective
, we must remember that , for nonperturbative values of the yukawa coupling , uncalculable corrections to the effective khler potential due to strong - coupling effects become important .
in fact , as pointed out by chacko et al . , these nonperturbative corrections are only negligible as long as @xmath159 @xcite .
we can , therefore , trust our above analysis , based on a canonical khler potential , only as long as @xmath47 remains in the perturbative regime .
for this reason , we will , from now on , only consider @xmath47 values at most as large as @xmath160 , so that we may always maintain a hierarchy among @xmath47 and @xmath45 ( i.e. , so that @xmath161 ) .
for the parameter @xmath152 , this then means that it can take values at most as large as @xmath162 .
this translates into the statement that the moduli constraint is always satisfied in our analysis up to a deviation of at most @xmath163 , @xmath164 .
in passing , we also mention that the scalar potential exhibits a saddle point at the origin in field space as well as two saddle points along each direction @xmath165 in moduli space . here
, the loci of the saddle points away from the origin have the same functional form as @xmath166 and @xmath167 in eq .
, the only difference being that @xmath47 in eq . needs to be exchanged with the respective yukawa coupling @xmath168 .
the low - energy vacuum along the @xmath169 axis , therefore , not only marks the global ( and only local ) minimum of the scalar potential , it is also the stationary point at which the deformed moduli constraint is fulfilled best . in the true vacuum , supersymmetry is broken by the nonvanishing f - terms of the fields @xmath170 and @xmath132 , @xmath171 the f - term of the meson field @xmath169 , on the other hand , ( which seems to be nonzero at first sight ) cancels , @xmath172 in order to identify the mass eigenstates around the true vacuum , we now shift the field @xmath169 by its vev , @xmath173 and rotate the susy - breaking fields @xmath170 and @xmath132 by their mixing angle @xmath174 , @xmath175 after these field transformations , the effective superpotential for @xmath43 , @xmath176 , @xmath177 , @xmath178 , and @xmath165 reads as follows , @xmath179 \ , , \label{eq : weff}\end{aligned}\ ] ] which is ( apart from the missing constant term @xmath66 ) exactly of the form anticipated at the end of sec .
[ subsec : original ] .
first of all , note that the first term on the right - hand side of eq . , @xmath180 , is nothing but the dynamical realization of the susy - breaking tadpole term in eq . within the iyit model .
the field @xmath43 is , thus , to be identified as the chiral polonyi field which breaks supersymmetry via its nonzero f - term , @xmath181 where the parameter @xmath38 denotes again the susy breaking scale . in this sense ,
eqs . and
show that the iyit model serves , indeed , as a viable uv completion for at least half the polonyi model : the susy - breaking dynamics of the iyit model manage to provide a dynamical explanation for the susy breaking scale @xmath38 .
however , as the iyit model preserves @xmath0 symmetry in its ground state , it is not capable of accounting for the origin of the @xmath0 symmetry - breaking constant @xmath66 in the polonyi superpotential .
second of all , the effective superpotential in eq . contains ( also as envisaged at the end of sec .
[ subsec : original ] ) yukawa couplings between the polonyi field @xmath43 and a number of massive matter fields , @xmath177 and @xmath165 . here ,
the meson masses , @xmath182 and @xmath183 , follow from dirac mass terms together with the singlet fields @xmath176 and @xmath178 , @xmath184 where we have introduced the flavor - independent mass scale @xmath185 as well as the respective ratios @xmath186 and @xmath187 between this scale @xmath185 and the masses @xmath182 and @xmath183 . from the fact that @xmath188 ( see eq .
below ) , it immediately follows that the scale @xmath185 represents the amount of susy - breaking mass splitting within the respective meson and singlet multiplets that is induced by the tadpole term in eq . ,
see appendix [ subsec : massspectrum ] for details . given the definition of @xmath152 in eq . and
recalling that we assume @xmath47 to be the smallest among all yukawa couplings , @xmath189 , we also find that the ratios @xmath186 and @xmath187 are bounded from above , @xmath190 so that the susy - breaking mass splitting @xmath185 never exceeds the supersymmetric dirac masses @xmath191 .
moreover , for the parameter range of interest , @xmath192 and @xmath193 , the ratio @xmath186 always turns out to be the smallest , @xmath194 , which leads to the interesting ( and to some extent counter - intuitive ) result that the zeroth flavor , i.e. , the flavor with the smallest yukawa coupling , ends up being stabilized the most .
in addition to that , the @xmath177 flavor is also singled out by the fact that its dirac mass partner is none of the original singlet fields @xmath138 , but the linear combination @xmath176 that we introduced in eq . and which , for small values of @xmath152 , mostly consists of the lagrange multiplier field @xmath132 .
this also explains why , for large @xmath135 ( and , hence , small @xmath152 ) , the @xmath177 mass @xmath182 diverges . in this limit
, the pfaffian constraint is fulfilled exactly , which results in the decoupling of @xmath132 and removes one meson multiplet ( i.e. , @xmath177 ) from the spectrum .
next to the polonyi field @xmath43 , also the `` stabilizer field '' @xmath176 couples to the meson fields @xmath177 and @xmath165 . here , the strengths of the respective yukawa couplings , @xmath195 and @xmath196 , are given by @xmath197 and the mixing angle @xmath174 , @xmath198 just like the mass @xmath182 , the yukawa coupling @xmath196 diverges in the limit @xmath199 .
this is a trivial consequence of its proportionality to @xmath197 . as for the polonyi coupling @xmath195 , this divergence is , however , canceled out by the @xmath200 factor .
in contrast to @xmath196 , the coupling @xmath195 , therefore , always remains finite . in the limit @xmath199
, it reproduces , in particular , the yukawa coupling of the singlet @xmath170 at energies above the dynamical scale , @xmath201 finally , we mention that the four new parameters @xmath38 , @xmath182 , @xmath195 , and @xmath196 introduced in eq . are not linearly independent .
in fact , they must be dependent , as they can all be expressed in terms of the three old parameters @xmath47 , @xmath41 , and @xmath135 . by making use of eqs . , , and
, one easily convinces oneself that @xmath202 in order to see how the superpotential in eq .
may give rise to polonyi inflation , it is instructive to forget about the yukawa couplings of the field @xmath176 for a moment and to rewrite eq .
as follows , @xmath203 + m_0\ , \sigma\ , \xi^0 + m_n\ , s_n x^n + \cdots \ , , \label{eq : weffv}\end{aligned}\ ] ] where the ellipsis stands for the yukawa couplings involving the stabilizer field @xmath176 and where we have introduced the mass scale @xmath204 , @xmath205 remarkably enough , the first part of the superpotential in eq .
has the same form as the superpotential of supersymmetric f - term hybrid inflation @xcite based on @xmath127 . in the context of this interpretation , the polonyi field plays the role of the chiral inflaton singlet , while the meson fields @xmath177 and @xmath165 act as a multiplet of fhi waterfall fields that transform in the vector representation of @xmath127 .
the mass scale @xmath204 is then to be identified as the energy scale of the waterfall transition at the end of inflation , while the mass splitting @xmath185 should be understood as the tachyonic mass of the fhi waterfall fields at @xmath206 .
we note that it is this picture that the authors of @xcite arrive at . in their model ,
no other singlets except for the polonyi field @xmath43 are introduced .
the meson fields @xmath177 and @xmath165 , therefore , lack their dirac mass partners , so that the resulting effective superpotential is exactly identical to the one of f - term hybrid inflation .
this allows the authors of @xcite to realize f - term hybrid inflation at large inflaton field values , @xmath207 , where the flatness of the inflaton potential is lifted by logarithmic loop corrections . in our case ,
the situation at large field values is quite similar , as we will discuss shortly ; but at small field values , it is drastically different .
the second part of the superpotential in eq . introduces explicit dirac mass terms for the `` would - be waterfall fields '' @xmath177 and @xmath165 that are absent in standard f - term hybrid inflation ( as well as in the model in @xcite ) . accounting for the presence of these dirac masses in the low - energy effective theory ,
the tachyonic waterfall mass @xmath185 is always compensated ( see eq . ) , so that none of the meson fields ever becomes destabilized .
because of that , the total superpotential in eq
. fails to give rise to a waterfall transition and , even in the low - energy vacuum , we retain the vacuum energy density resulting from the nonzero polonyi f - term , @xmath208 . as anticipated in sec .
[ subsec : setup ] , this is a characteristic feature of our construction , in which we intend to use one and the same vacuum energy density for driving inflation and breaking supersymmetry .
moreover , independent of the symmetry group under which the meson fields transform , the absence of the waterfall transition automatically implies that the end of inflation is not accompanied by the production of topological defects .
this may be regarded as a significant phenomenological advantage of our scenario over standard f - term hybrid inflation . to the best of our knowledge , the superpotential in eq . has not been considered as the dynamical origin of inflation , so far . here
, part of the reason certainly is that successful inflation based on eq .
is bound to require a rather high susy breaking scale @xmath38 . as explained in the introduction , in supergravity ,
this necessitates a large constant in the superpotential to cancel the cc in the true vacuum , which then spoils slow - roll inflation ( see sec .
[ subsec : origin ] ) . in the rest of this paper
, we will , however , show that the superpotential in eq
. can yield successful polonyi inflation , after all , if we generate the constant in the superpotential only towards the end of inflation . to this end
, we shall now examine the effective one - loop potential for the complex polonyi scalar @xmath209 in global supersymmetry more closely . in the next sections
, we will then turn to the embedding of the iyit model into supergravity ( see sec .
[ subsec : sugra ] ) as well as to the generation of the constant term in a separate hidden sector ( see sec . [
sec : rbreaking ] ) .
the yukawa interactions between the polonyi field @xmath43 and the meson fields @xmath177 and @xmath165 in eq . lead to radiative corrections to the polonyi potential , @xmath210 , that may be calculated according to the coleman - weinberg ( cw ) formula for the effective one - loop potential @xcite .
the details of our calculation may be found in appendix [ app : potential ] ; in the following , we will merely summarize our results .
generally speaking , the effective potential may be divided into two regimes : ( i ) at large field values , all of the meson fields acquire a large inflaton - dependent majorana mass @xmath211 . for @xmath212 , the supersymmetric dirac masses
@xmath191 in eq . are , hence , negligible and the effective potential takes the usual logarithmic form as in standard f - term hybrid inflation , @xmath213 .
( ii ) on the other hand , at small polonyi field values , such that @xmath214 , the dirac masses @xmath191 become more relevant . integrating out the `` heavy fields '' then leads to a quadratic polonyi potential around the origin , @xmath215 .
we note that , as has been shown for the first time in @xcite , the effective potential around the origin has _ positive _ curvature .
the low - energy vacuum at @xmath216 is , therefore , indeed stable .
moreover , we find that , at large as well as at small field values , the effective potential scales with the soft susy - breaking mass scale @xmath185 , i.e. , the smallest mass scale in our model ( see eq . ) .
this illustrates how supersymmetry succeeds in protecting the inflaton potential from picking up too large radiative corrections @xcite . to quantify the above statements , it is convenient to introduce the following mass ratios , @xmath217 for large values of @xmath218 , we then obtain a logarithmic one - loop potential , while for small values of @xmath218 , the one - loop corrections take a quadratic form
this behavior can be captured by studying the effective potential as a function of a single order parameter @xmath219 , the geometric mean of all ratios @xmath218 , @xmath220 here , @xmath221 counts the number of meson fields in the iyit sector , while @xmath222 stands for the geometric mean of all explicit mass parameters in eq . .
in this sense , @xmath222 denotes the `` supersymmetric mass scale '' of the iyit sector , i.e. , a characteristic value for the nonperturbatively generated dirac masses in the low - energy effective theory . in the following
, we will set @xmath222 to the dynamical scale @xmath41 , for definiteness , @xmath223 this mainly serves the purpose to account , in an effective way , for heavy composite states with masses around the dynamical scale @xmath41 that we expect to be present , but which we can unfortunately not explicitly describe in terms of our perturbative language at low energies .
_ formally _ , we can always set @xmath222 to @xmath41 in our calculation by choosing an appropriate value for the effective heavy - flavor yukawa coupling @xmath224 , @xmath225 by fixing @xmath224 at a nonperturbative value , we are , therefore , able to enforce our designated value for @xmath222 , @xmath226 the value of @xmath219 indicates whether the dirac masses @xmath191 are negligible or not and , thus , decides whether we are in the logarithmic or the quadratic part of the effective potential .
the transition between both regimes takes place at field values close to what we shall refer to as the critical field value @xmath227 , @xmath228 this implies that the order parameter @xmath219 can also be regarded as the ratio of the actual and the critical field value , @xmath229 . far away from the critical field value ,
i.e. , at @xmath230 and @xmath231 , we now find the following expressions for the effective potential ( see appendix [ app : potential ] for details ) , @xmath232 with @xmath233 denoting the effective one - loop mass of the polonyi field around the origin , ( and , hence , nonzero @xmath186 ) , which means that we do not necessarily enforce the moduli constraint exactly .
the calculation in @xcite , on the other hand , is based on the assumption that the moduli constraint _ is _ fulfilled exactly , so that @xmath234 . for a recent derivation and discussion of the effective polonyi mass in @xmath126 language ,
see @xcite .
[ fn : meff ] ] @xmath235 here , @xmath236 counts the effective number of mesons that contribute to the effective polonyi mass .
the full functional form of @xmath236 is a sum of complicated loop factors @xmath237 . to good approximation
, these loop functions , however , happen to coincide with the mass ratios @xmath238 squared , @xmath239 .
we can write the result in eq . more compactly , if we make use of the following two potential energy scales , @xmath240 the effective potential far away from the critical field value @xmath227 then takes the following form ( see fig . [
fig : effpot ] ) , @xmath241 , as a function of the order parameter @xmath242 for @xmath243 , see eq . .
the potential energy scales @xmath244 and @xmath245 are given in eq . .
for values @xmath246 , we can only have limited trust in our perturbative result because of potentially important strong - coupling effects at energies close to the dynamical scale @xmath41.,scaledwidth=65.0% ] the crucial question which we need to answer in the following is : can we use either the low - energy or the high - energy part of this effective potential to realize successful polonyi inflation ? let us first investigate whether inflation might occur in the quadratic part close to the origin . as we know from standard chaotic inflation @xcite , the effective inflaton mass then needs to take a value of @xmath247 to ensure the correct normalization of the scalar power spectrum .
this requires the coupling @xmath47 to take a value at least as large @xmath248 , since otherwise the dynamical scale @xmath41 would have to be super - planckian , @xmath249 at the same time , we know that chaotic inflation requires a large super - planckian field excursion to yield a sufficient number of @xmath250-folds .
the scalar perturbations probed in cmb observations , e.g. , cross outside the hubble horizon at a field value @xmath251 .
however , in the context of our susy breaking model , this large field range does not `` fit '' into the low - energy part of the effective potential .
this follows from the fact that , for @xmath248 and @xmath252 , the critical field value @xmath227 only becomes as large as @xmath253 , @xmath254 therefore , to raise @xmath227 , so as to make the field range required for chaotic inflation fit into the quadratic part of the effective potential , @xmath255 , we would have to go to smaller values of @xmath47 .
but then , we are either forced to push @xmath41 beyond the planck scale or we fail to reproduce the correct scalar spectral amplitude .
this eliminates the possibility of polonyi inflation in the low - energy part of the effective potential , which is why we will focus on inflation in the logarithmic part of the effective potential from now on . before continuing , we , however ,
point out that inflation close to the origin might become possible , after all , if we relax our assumptions .
that is , if we allowed for values of the dynamical scale as large as , say , @xmath256 , we would , in fact , be able to raise @xmath227 above @xmath257 . if we then trusted the full effective potential also at field values close to @xmath227 ( see fig .
[ fig : effpot ] ) , inflation in the transitioning regime between the quadratic and the logarithmic part of the effective potential might become feasible .
such a scenario would promise to interpolate between the predictions of chaotic inflation and f - term hybrid inflation , so that we would expect it to result in interesting predictions for the tensor - to - scalar ratio , @xmath258 .
because of the uncertainties involved in such a scenario , we , however , do not pursue this idea any further in this paper and leave a more detailed study for future work . in closing
, we remark that a similar model of _ subcritical hybrid inflation _ , based on a dynamically generated d - term @xcite , may be found in @xcite .
this model illustrates how to realize chaotic inflation after the waterfall transition of _ d - term _
hybrid inflation .
let us now turn to the possibility of inflation in the logarithmic part of the effective potential .
as we will show , in this part of the potential , successful polonyi inflation is indeed feasible . to be on the safe side
, we will limit our analysis in the following to field values that are larger than the critical field value by at least half an order of magnitude , @xmath259 .
we do so because the effective polonyi potential may receive nonperturbative corrections around @xmath260 that we do not have under control .
in fact , around the critical field value , the inflaton - dependent mass @xmath261 drops below the dynamical scale @xmath41 ( see eq . ) .
this triggers the iyit sector to transition from the high - energy quark - gluon regime into the low - energy meson regime . during this
_ mesonic phase transition _ , the iyit quarks become confined in the composite mesons and the strongly coupled @xmath49 gauge group becomes completely broken by the nonzero squark vevs @xmath262 and @xmath263 that contribute to the meson vev @xmath166 ( see eqs . and ) , follows from the fact that the linear combination @xmath264 vanishes in the vacuum , @xmath265 . ]
@xmath266 thanks to the fact that the @xmath49 symmetry is spontaneously broken down to `` nothing '' , no topological defects are formed during the confining phase transition @xcite . we emphasize that this is an important phenomenological feature of the iyit model based on @xmath49 , as it allows for the particular breaking pattern @xmath267 . in summary ,
our scenario of polonyi inflation , therefore , crucially differs from ordinary f - term hybrid inflation in the following respect : while ordinary f - term hybrid inflation ends in a waterfall transition in the course of which the inflationary vacuum energy density is `` eaten up '' by the fhi waterfall fields and which potentially leads to the production of troublesome topological defects our scenario of polonyi inflation undergoes a confining quark - meson phase transition that _ conserves _ the inflationary vacuum energy density and that does _ not _ lead to the production of topological defects . in supergravity ,
the flatness of the tree - level polonyi potential in global supersymmetry , @xmath268 , is not only lifted by the radiative corrections in eq . , but also by gravitational corrections ( see our discussion at the end of sec .
[ subsec : ingredients ] ) . in our case , these sugra corrections turn out to be rather mild for basically two reasons : ( i ) since the superpotential in eq . only contains terms linear in @xmath43 , the tree - level sugra mass of the polonyi field accidentally cancels , as long as we assume a canonical khler potential . as far as the embedding into supergravity
is concerned , this represents an important advantage of f - term hybrid inflation ( and of our model ) over alternative models that do feature higher powers of the inflaton field in the superpotential .
( ii ) since we intend to realize polonyi inflation in the logarithmic part of the effective potential , we will consistently work with sub - planckian field values .
our scenario of polonyi inflation will , hence , turn out to be a _ small - field _ model of inflation .
accordingly , the sugra corrections in our model are bound to be less significant than in alternative _ large - field _ models of inflation .
the total scalar potential for the complex polonyi field @xmath269 in supergravity now reads , is not necessarily canonically normalized , which we indicate by placing a tilde on top of the symbol @xmath270 . ]
@xmath271 with @xmath272 being given in eq . and where @xmath273 denotes the tree - level f - term potential in supergravity ,
@xmath274 \frac{\left|w\right|^2}{m_{\rm pl}^2 } \ , .
\label{eq : vf}\end{aligned}\ ] ] here , @xmath275 denotes the norm of the generalized f - term vector @xmath276 induced by the khler metric @xmath277 , @xmath278 the individual components of the the f - term vector @xmath276 are proportional to the conjugate the of khler - covariant derivatives of the superpotential , @xmath279 , multiplied by the inverse of the khler metric , @xmath280 , @xmath281 \left(d_j w\right)^ * \ , , \quad k_{\:\bar{\jmath}}^i = \left(k^{-1}\right)_{\:\bar{\jmath}}^i\ , , \quad d_i w = \frac{\partial w}{\partial \phi^i } + \frac{w}{m_{\rm pl}^2}\frac{\partial k}{\partial \phi^i } \ , .
\label{eq : fivector}\end{aligned}\ ] ] after integrating out the heavy fields @xmath177 , @xmath165 , @xmath176 , and @xmath178 in eq .
, the effective superpotential of the iyit sector reduces to the susy - breaking tadpole term for the polonyi field ( see eq . ) , @xmath282 meanwhile , the tree - level khler potential @xmath10 is not unambiguously defined . at first sight , there are several well motivated choices for @xmath10 that come into question .
for instance , we might think that an approximately shift - symmetric khler potential could help in keeping the sugra corrections small , , and the radiative corrections in the scalar potential ( see eq . ) .
for this reason , we expect that , also in the tree - level khler potential , the shift symmetry should be explicitly broken .
] @xmath283 depending on the sign choice , the real - valued ( not canonically normalized ) inflaton field @xmath284 is then identified either as the real ( @xmath285 ) or the imaginary ( @xmath286 ) part of the complex scalar @xmath269 .
likewise , the real - valued ( not canonically normalized ) scalar inflaton partner @xmath287 ( i.e. , the `` sinflaton '' ) corresponds to the imaginary ( @xmath285 ) or the real ( @xmath286 ) part of the complex scalar @xmath288 . as long as we do not assume an exact shift symmetry in the khler potential ( i.e. , @xmath289 ) , @xmath284 and @xmath287 need to be canonically normalized , @xmath290 unlike in the case of a canonical khler potential , the sugra f - term potential @xmath273 now contains nonzero tree - level masses for the inflaton and its scalar partner . at the origin in field space
, we find @xmath291 for @xmath292 , the inflaton direction is , hence , tachyonic ! if we forget about the radiative corrections @xmath272 for a moment , this immediately implies that polonyi inflation based on an approximately shift - symmetric khler potential is impossible which completes our argument in sec .
[ subsec : original ] ( see footnote [ fn : shift ] ) . on the other hand ,
taking the radiative corrections into account , we may hope that the effective inflaton mass @xmath233 around the origin could possibly compensate the tachyonic tree - level mass @xmath293 and , thus , stabilize the vacuum at @xmath81 .
it is clear that this will not be easy , since @xmath293 in eq .
is typically very large , @xmath294 in order to sufficiently suppress it , we need to go to smaller values of the dynamical scale @xmath41 . requiring that the total inflaton mass at the origin be non - tachyonic , @xmath295
, we find ( see eq . ) , @xmath296 for such small values of the dynamical scale , the critical field value @xmath227 is clearly sub - planckian ( see eq . ) . as discussed in the previous section , this eliminates the possibility of realizing inflaton in the quadratic part of the potential close to the origin .
our only remaining hope , therefore , is to realize inflation at field values @xmath297 . to this end , it is important that the total scalar potential does not exhibit a local minimum above the critical field value otherwise the inflaton will never reach the origin and inflation would never end .
but as it turns out , exactly such a local minimum is always present .
the point is that the quadratic part of the effective potential can only compensate the tachyonic tree - level mass up to field values of @xmath298 . at the same time
, the gradient of the tree - level potential only turns positive at very large field values , @xmath299 , where @xmath300 denotes the position of a global ads vacuum . as long as @xmath301 ,
the gradient of the total inflaton potential is , therefore , guaranteed to turn negative in between @xmath227 and @xmath300which indicates the presence of a local minimum . to evaluate the relation between @xmath227 and @xmath300 more precisely ,
consider the tree - level inflaton potential for @xmath302 , @xmath303 \left(v_0 + \frac{1}{2}\ , m_\varphi^2\ , \varphi^2 + \frac{\lambda_\varphi}{4!}\ , \varphi^4\right ) \,,\end{aligned}\ ] ] with the three parameters @xmath304 , @xmath305 , and @xmath306 being given as follows , @xmath307 the local minimum in the sugra tree - level potential , @xmath300 , is then located at @xmath308^{1/2 } m_{\rm pl } \
, , \quad a = -\frac{\epsilon}{1+\epsilon}\frac{6}{\lambda_\varphi}\frac{m_\varphi^2}{m_{\rm pl}^2 } \,,\end{aligned}\ ] ] so that the ads vacuum energy density takes the following value , @xmath309 \left(\frac{3-\epsilon}{2\left|\epsilon\right|}-2\right ) v_0 \,.\end{aligned}\ ] ] given the above expression for @xmath300 , we find that , in the entire parameter space of interest ( @xmath310 and @xmath311 ) , @xmath227 and @xmath300 are always separated by at least one order of magnitude , @xmath312 . in the shift - symmetric limit , @xmath313 , the ads vacuum disappears , in particular , altogether , @xmath314 , so that the total inflaton potential becomes unbounded from below . in summary , this shows that the effective potential manages to compensate the tachyonic tree - level mass only for too small a field range .
it only `` bends around '' the tachyonic part of the tree - level potential up to @xmath315 , while it should actually do this up to field values that are at least ten times as large , @xmath316 .
this finally completes our argument that polonyi inflation is incompatible with an approximate shift symmetry in the khler potential , independent of whether we account for the presence of radiative corrections or not .
this result leaves a near - canonical tree - level khler potential as the best possible choice , @xmath317 of course , simply assuming a canonical khler potential , @xmath318 , would represent an even more minimal choice . in that case
, the polonyi field would not `` equip itself '' with a hubble - induced mass ( see our discussion at the beginning of this section ) and we would not have to fear any dangerous sugra corrections .
but such a scenario is rather unlikely .
terms featuring higher powers of @xmath319 are not forbidden by any symmetry and , hence , we can not exclude their presence in the khler potential .
quite the contrary , we rather expect that they are the unavoidable consequence of radiative corrections in quantum gravity that manifest themselves as planck - suppressed operators in the low - energy effective theory .
the only open question regarding the higher - dimensional terms in the khler potential pertains to the size of their coefficients .
the leading correction to the canonical khler potential , e.g. , results in an inflaton mass squared of @xmath320 , which shifts the slow - roll parameter @xmath45 by @xmath321 . therefore , if @xmath48 is of @xmath55 , as one might naively expect , we would encounter the @xmath45 problem @xcite .
this tells us that the coefficient @xmath48 needs to be suppressed , @xmath322 , which introduces a parameter fine - tuning at the level of @xmath46 . in the following , we will not speculate what the physical reason for this fine - tuning might be and simply assume that @xmath48 is sufficiently small .
we merely remark that there is a number of models in the literature that attempt to explain why @xmath48 ends up being suppressed .
for instance , a small coefficient @xmath48 may be the result of a sub - planckian cut - off scale , @xmath323 , in the sector giving rise to the higher - dimensional term in eq .
@xcite ( see @xcite for a realization of this idea in string theory ) ; or @xmath48 may be small because it is , for one reason or another , suppressed by nda factors of @xmath101 @xcite .
having defined the khler potential , we now need to canonically normalize the complex polonyi field @xmath79 . here
, we will restrict ourselves to normalizing the radial component @xmath284 only , as the complex phase @xmath324 will be irrelevant during inflation , anyway .
the canonically normalized field @xmath78 is then given as @xmath325 \,.\end{aligned}\ ] ] this relation is readily inverted , which provides us with an expression for @xmath284 in terms of @xmath78 , @xmath326 \,.\end{aligned}\ ] ] we are now in the position to write down the full tree - level sugra potential @xmath273 for @xmath78 and @xmath324 .
in doing so , we shall first use the full polonyi superpotential in eq . for illustrative purposes ( i.e. , we will add a constant @xmath66 to the effective superpotential , although @xmath71 in the iyit model ) .
. then yields @xmath327 where the coefficients @xmath328 , @xmath329 , @xmath330 , @xmath331 , and @xmath332 are given as follows , , only contributes to the coefficient of the quartic inflaton term , @xmath333 .
this effect is negligibly small , which is why we will set @xmath334 in the following . ]
@xmath335 this form of the scalar potential illustrates the impact of the constant term in the superpotential , @xmath336 .
not only does @xmath66 induce a dependence on the phase @xmath324 of the complex inflaton field ( thereby introducing odd powers of the radial component @xmath78 in the scalar potential ) , it also results in a large tachyonic tree - level mass . in appendix
[ subsec : canonical ] , we show that this potential is always too steep for slow - roll inflation .
moreover , we note that the potential in eq . is similar to the tree - level potential of f - term hybrid inflation .
yields the sugra fhi tree - level potential up to @xmath337 , if we make the following replacements , @xmath338 where the constant @xmath66 is to be understood as a measure for the gravitino mass , @xmath339 . in f - term
hybrid inflation , the functional dependence of the scalar potential on the phase of the complex inflaton field is , therefore , exactly the same as in the full polonyi model ( see @xcite for the first analysis that properly treats f - term hybrid inflation as a two - field model in the complex plane ) .
by contrast , the soft inflaton mass is no longer tachyonic in f - term hybrid inflation , but identical to the gravitino mass .
meanwhile , the change in the coefficient @xmath332 is suppressed by @xmath340 and , thus , irrelevant .
furthermore , it is worthwhile pointing out that f - term hybrid inflation is _ not _ compatible with the idea of a gravitino mass as large as the inflationary hubble rate . instead , @xmath28 and
@xmath32 must always maintain a hierarchy of at least three orders of magnitude , @xmath341 , since otherwise slow - roll inflation becomes spoiled by too large sugra corrections @xcite .
this is in contrast to our scenario of polonyi inflation , where @xmath28 and @xmath32 are related to each other at a fundamental level and , in fact , equal to each other ( see eq . ) .
let us now turn to the scalar potential of @xmath342 inflationary scenario . for our purposes , the constant
@xmath66 needs to be zero during inflation .
the total inflaton potential for polonyi inflation , therefore , follows from the sum of the logarithmic one - loop potential in eq . and
@xmath273 in eq . evaluated for @xmath71 , @xmath343
this inflaton potential is identical to the potential of f - term hybrid inflation ( including corrections from supergravity and a noncanonical khler potential ) in the limit of a vanishing gravitino mass @xcite .
in the previous section , we have seen what it takes to identify the susy - breaking polonyi field @xmath43 as the chiral inflaton field .
we considered three different types of radiative corrections to the scalar polonyi potential ( none , quadratic corrections , logarithmic corrections ) , two different shapes of the tree - level khler potential ( approximately shift - symmetric and near - canonical ) , and two different choices for the constant in the superpotential ( @xmath71 and @xmath344)and we found only one possible combination of `` model ingredients '' that could not be ruled out immediately : the polonyi model supplemented by _
logarithmic _ radiative corrections in combination with a _ near - canonical _ khler potential and @xmath345 during inflation . in the following , we will focus on this scenario and and
present a dynamical model that illustrates how the constant in the superpotential , @xmath346 , may be generated at the end of inflation .
as we will see , the constant @xmath66 then represents ( not the only one , but ) the dominant contribution to the vev of the superpotential in the low - energy vacuum , @xmath347 .
consequently , @xmath66 turns out to be the main source of spontaneous @xmath0 symmetry breaking at low energies .
this is the reason why , throughout the entire paper , we refer to _ the generation of the constant term in the superpotential _
after inflation also as the _ spontaneous breaking of @xmath0 symmetry at late times_. technically speaking , this is not quite correct , as , globally , @xmath0 symmetry is already spontaneously broken during inflation by the vev of the polonyi field , @xmath348 this is because @xmath43 carries @xmath0 charge @xmath16 .
what we actually mean by speaking of _ late - time @xmath0 symmetry breaking _ is , therefore , not the spontaneous breaking of @xmath0 symmetry from a global perspective , but only the spontaneous breaking of @xmath0 symmetry in the sector responsible for the generation of the constant term @xmath66 .
from the perspective of the low - energy effective theory , the process of late - time @xmath0 symmetry breaking then amounts to the ( explicit ) breaking of @xmath0 symmetry at the level of the lagrangian by means of a constant in the superpotential , as opposed to the ( spontaneous ) breaking of @xmath0 symmetry at the level of a scalar field vev during inflation . here
, it is important to note that the latter kind of @xmath0 symmetry breaking mostly vanishes in the true vacuum after inflation , whereas the former kind of @xmath0 symmetry also remains at low energies .
likewise , we shall refer to the sector responsible for the generation of the constant @xmath66 ( i.e. , the sector responsible for @xmath0 symmetry breaking at low energies ) as the _ @xmath0 symmetry - breaking sector _ for short .
we also point out that the temporary breaking of @xmath0 symmetry by the vev of the inflaton field is , in fact , mandatory from a phenomenological point of view .
otherwise , the spontaneous breaking of @xmath0 symmetry during the generation of the constant @xmath66 would result in the production of domain walls at the end of inflation , which would render our scenario phenomenologically unviable @xcite .
symmetry ( in the context of some _ other _ model ) , @xmath0 symmetry would need to be broken in a separate sector way before the end of inflation , to make sure that all domain walls are sufficiently diluted .
such an alternative scenario provides a different explanation for why supersymmetry is necessarily broken at a high scale @xcite . ] similarly to the polonyi vev , also the constant @xmath66 breaks the @xmath37 symmetry down to a @xmath349 parity , @xmath350 so that , during inflation as well as in the low - energy vacuum after inflation , the only discrete symmetry that remains globally unbroken corresponds to an exact @xmath349 .
interestingly enough , this parity is suited to be identified as the @xmath0 parity of the mssm , @xmath351 @xcite . here , note that @xmath0 parity is not a proper @xmath0 symmetry in the actual sense , as it is equivalent to matter parity , @xmath352 , which is a non-@xmath0 symmetry @xcite .
our model , therefore , automatically offers a possible explanation for the origin of @xmath0 parity in the mssm .
in contrast to other models , it does not depend on extra continuous symmetries , such as a global or local @xmath353 , that would leave behind @xmath354 as a discrete remnant subgroup after spontaneous symmetry breaking .
given the above sources of @xmath0 symmetry breaking during and after inflation , the physical gravitino mass @xmath28 varies as a function of time . in general , @xmath28 is controlled by the vev of the superpotential , @xmath355 \frac{\left < w\right>}{m_{\rm pl}^2 } \ , , \quad w = w_{\rm eff } + w \ , .
\label{eq : m32}\end{aligned}\ ] ] during inflation , the constant @xmath66 vanishes and , thus , @xmath28 turns out to be roughly proportional to the polonyi field value .
meanwhile , after inflation , @xmath28 is dominated by the contribution from the constant @xmath66 . in order to tune the cc to zero
, we have to require that @xmath28 eventually takes the following value , @xmath356 this condition can always be satisfied by fine - tuning the constant @xmath66 to the particular value @xmath357 , @xmath358 \frac{m_{\rm pl } } { \sqrt{3}}\left<\left|f\right|\right>_{w = w_0}-\left < w_{\rm eff}\right > \ , .
\label{eq : w0}\end{aligned}\ ] ] here , it is important to note that the susy - breaking f - term @xmath275 also depends on the constant @xmath66 via the covariant derivative of the superpotential ( see eq . ) .
for this reason , the above relation is , in fact , only an implicit definition of @xmath357 , which , _ in principle _ , still needs to be solved for @xmath357 . _ in practice _ , the dependence of @xmath275 on @xmath66 is , however , negligible in most cases , so that the right - hand side of eq .
readily yields the required value for @xmath66 . in the following
, we shall now show how the constant @xmath66 may be generated after inflation ( see secs .
[ subsec : rbreakingsector ] and [ subsec : trigger ] ) and discuss under which conditions it may be successfully matched with the desired value @xmath357 , so that the cc vanishes at low energies ( see sec .
[ subsec : backreaction ] ) .
the simplest way to break @xmath0 symmetry via strong dynamics is to make use of dynamical gaugino condensation in a pure sym theory @xcite .
for instance , in a strongly coupled pure sym theory based on @xmath359 and associated with a dynamical scale @xmath360 , gaugino condensation results in an @xmath0 symmetry - breaking constant @xmath66 of @xmath361 .
let us now illustrate how we can use this property for our purposes .
our starting point is supersymmetric quantum chromodynamics ( sqcd ) : a strongly coupled @xmath359 gauge theory with @xmath362 colors and @xmath363 flavors , where each flavor consists of a quark / antiquark pair @xmath364 .
we will constrain the viable values for @xmath362 and @xmath365 shortly . for the moment , however , let us stay as general as possible and leave the concrete values of @xmath362 and @xmath365 unspecified .
we assume that all quark / antiquark pairs are _ a priori _ massless , i.e. , we do _ not _ introduce any bare mass terms of the form @xmath366 in the tree - level superpotential . instead , we suppose that all of the @xmath365 flavors share a yukawa coupling with some @xmath359 singlet field @xmath50 , which , thus , results in _ field - dependent _ masses for all flavors , @xmath367 here , the coefficients @xmath368 denote dimensionless yukawa couplings of @xmath157 . by varying the vev of the singlet field @xmath50 , we are then able to control the quark mass matrix , @xmath369 , in our sqcd theory , @xmath370 as long as @xmath50
is stabilized at the origin , @xmath52 , all quarks are massless and the sqcd sector remains what it is : an @xmath359 gauge theory with @xmath365 dynamical flavors , described by a quantum moduli space at low energies @xcite . ,
these stabilize the runaway directions on the quantum moduli space induced by the nonperturbative ads superpotential @xcite . ] on the other hand , once @xmath50 acquires a large vev , all flavors become heavy and can be integrated out .
this mass deformation transforms the sqcd sector into a pure sym theory , in which @xmath0 symmetry is spontaneously broken : first , @xmath0 symmetry is broken down to @xmath371 due to @xmath359 instantons .
then , the discrete @xmath371 symmetry is broken further down to @xmath349 via gaugino condensation .
this results in the eagerly anticipated constant term @xmath66 in the effective superpotential , @xmath372 here , @xmath373 is a dimensionless `` fudge factor '' that encompasses all numerical factors entering into the expression for @xmath66 except for @xmath362 and @xmath374 .
based on nda , we again expect @xmath373 to be of @xmath101 , which is why we will set @xmath375 in the following , for definiteness .
meanwhile , @xmath376 denotes the effective dynamical scale of the sym theory after integrating out all heavy flavors .
we obtain an expression for @xmath376 by matching the running of the @xmath359 gauge coupling constant @xmath377 at the respective heavy - quark mass thresholds , @xmath378 where @xmath379 denotes the dynamical scale of the original high - energy theory and where @xmath380 represents the effective quark mass scale , i.e. , the geometric mean of all quark mass eigenvalues @xmath381 , @xmath382 in the following , we will set @xmath383 ( the geometric mean of all yukawa couplings @xmath368 ) to @xmath384 , for simplicity . for given values of @xmath362 and @xmath365 , the constant @xmath66 in eq . then ends up being a function of @xmath385 and @xmath379 , @xmath386 here , @xmath387 must be larger than @xmath379 to ensure that the heavy quark flavors can be integrated out perturbatively in the high - energy theory .
our result for @xmath66 in eq .
is , therefore , bounded from above , @xmath388 the dynamical scale of the high - energy theory , @xmath379 , is generated via dimensional transmutation and , hence , solely depends on the value of the @xmath359 gauge coupling constant @xmath377 at the planck scale , @xmath389 \ , , \quad b = 3\ , n_c - n_f \ , .
\label{eq : lambdatilde}\end{aligned}\ ] ] we assume the @xmath359 beta function coefficient @xmath390 to be positive , which means that @xmath379 can basically take any desired value below the planck scale . for a given vev @xmath385 , the constant @xmath66 can then be tuned simply by varying the value of the gauge coupling constant @xmath377 at the uv cutoff scale . from this perspective
, the fine - tuning of the cc in the true vacuum turns into an issue that pertains to the uv boundary conditions of the low - energy effective theory .
it is no longer a problem within the low - energy effective theory itself .
how can we now apply these results in the context of polonyi inflation ?
the crucial idea is to relate the value of the vev @xmath385 to the inflationary dynamics .
we must make sure that the singlet @xmath50 is stabilized at @xmath52 during inflation and that it becomes destabilized only towards the end of inflation . in other words , we need to achieve a situation in which small inflaton field values trigger a mass deformation of the sqcd sector , @xmath391 , so that this sector turns into a pure sym theory and spontaneously breaks @xmath0 symmetry via gaugino condensation .
this approach shares some similarities with the models presented in @xcite , in which small inflaton field values trigger the mass generation in a separate iss sector @xcite , which then spontaneously breaks supersymmetry .
moreover , our scenario of @xmath0 symmetry breaking should be compared with the approaches in @xcite and @xcite . in @xcite
, inflation ends in a supersymmetric , but @xmath0 symmetry - breaking ground state , while in @xcite , @xmath0 symmetry is broken by a `` field - dependent constant '' term in the superpotential that only becomes large at the end of inflation .
we can realize the scenario described in the previous section by introducing the following superpotential , @xmath392 here , @xmath393 is a singlet field that carries @xmath0 charge @xmath16 ; @xmath394 and @xmath174 are dimensionless coupling constant ; and @xmath395 is a mass scale .
it is tempting to speculate that also the scale @xmath395 might be dynamically generated in a strongly coupled hidden sector , similar to the scale @xmath204 in eq . .
but for our purposes , it will not be necessary to specify its origin .
instead , we will treat it as a hard dimensionful input scale , the only one in our model . before we explain in more detail how @xmath396 allows us to give a vev to the field @xmath50 at the end of inflation ,
let us examine for which values of @xmath362 and @xmath365 the superpotential in eq
. is consistent with the ( approximate ) global symmetries of the theory in the first place . promoting these global symmetries to gauge symmetries ( and/or assuming that they are at least sufficiently protected by other gauge symmetries )
, we will then be able to narrow down the viable choices for @xmath362 and @xmath365 .
the superpotential in eq .
is invariant under an anomaly - free global @xmath397 flavor symmetry , under which the fields @xmath398 , @xmath399 , @xmath50 , and @xmath393 are charged as follows , [ cols="^,^,^,^,^",options="header " , ] as mentioned earlier , in the pure sym theory after integrating out the heavy quarks , the continuous @xmath35 is broken to a discrete @xmath400 symmetry by @xmath359 instantons .
this immediately suggests to set the number of colors @xmath362 to @xmath16 , so that the global @xmath400 symmetry of the pure sym theory can be identified with our gauged @xmath37 symmetry .
, that relates @xmath0 symmetry transformations in the one sector to @xmath0 symmetry transformations in the other sector . besides that , we assume the coefficients of these mixing operators to be negligibly small for all practical purposes , @xmath401 .
this might , e.g. , be achieved if the cross terms in the khler potential are suppressed by ( several powers of ) a large cut - off scale . ] according to the above table , the operator @xmath402 in eq .
then only complies with the exact @xmath403 symmetry , if we choose the number of flavors @xmath365 to be either @xmath118 or @xmath16 , @xmath404_r = 2 + \frac{4n_c}{n_f } = 2 \textrm { mod } 2n_c \quad \rightarrow \quad n_f = 1,2 \,.\end{aligned}\ ] ] on second sight , @xmath405 is , however , not a viable option . in that case , the @xmath0 charge of the field @xmath50 vanishes , @xmath406_r = 0 $ ] , so that we can not prevent the unwanted operator @xmath407 from popping up in the superpotential .
also the approximate discrete @xmath408 flavor symmetry can not help us in this situation , since the singlet field @xmath50 is required to transform even , @xmath406_{z_2 } = 0 $ ] , under this parity . at the same time , the singlet field @xmath393 must also transform even , @xmath409_{z_2 } = 0 $ ] , to allow for the presence of the @xmath393 tadpole term in eq . . the discrete @xmath53 flavor symmetry ,
therefore , does not allow us to eliminate the operator @xmath407 , as long as we want to keep the @xmath393 tadpole term .
this leaves us with the following unique choice for @xmath362 and @xmath365 , @xmath410 which results in exactly the same matter and gauge field content as in the iyit sector . for @xmath411
, the singlet field @xmath50 carries @xmath0 charge @xmath16 . from the perspective of @xmath0 symmetry , the fields @xmath50 and @xmath393 are , hence , indistinguishable from each other , so that , next to the three wanted operators @xmath393 , @xmath402 , and @xmath412 , also the three unwanted operators @xmath50 , @xmath413 , and @xmath414 are allowed to appear in the superpotential .
we are , therefore , led to impose the @xmath415 flavor symmetry as an approximate symmetry which we assume to be protected by some exact gauge symmetry in the uv . of the pure sym theory with our anomaly - free @xmath37 ,
i.e. , if we allowed for @xmath416 ( while keeping @xmath417 fixed ) , we would not be forced to impose any _ additional _ approximate flavor symmetry . in that case
, the @xmath400 alone would suffice to restrict the set of allowed operators to @xmath393 and @xmath402 ( meaning that the @xmath412 term would be suppressed ) . such a scenario would , however , not present an advantage over the @xmath418 case . either way
, we have to assume a sufficiently intact global symmetry . while for @xmath418 the @xmath151 needs to be imposed as an approximate symmetry , the @xmath400 itself would need to be play the role of an approximate symmetry for @xmath416 .
we will , therefore , ignore this possibility in the following . ]
this approximate @xmath151 acts on the fields @xmath50 and @xmath393 again as a @xmath53 parity the crucial difference to the @xmath405 case being that , now , @xmath50 transforms odd , @xmath406_{z_2 } = 1 $ ] , while @xmath393 still transforms even , @xmath409_{z_2 } = 0 $ ] . by virtue of this @xmath53 parity , the coefficients of the unwanted operators @xmath50 , @xmath413 , and @xmath414 then end up being suppressed .
moreover , we emphasize the importance of the fact that the @xmath151 symmetry is only an approximate symmetry .
if it was an exact symmetry , its spontaneous breaking at the end of inflation would result in the formation of dangerous domain walls . as the @xmath151 symmetry is , however , bound to be explicitly broken ( e.g. , by the suppressed @xmath53-odd operators @xmath50 , @xmath413 , and @xmath414 ) , we are safe from running into this problem
. given the superpotential in eq .
, let us now demonstrate how it enables us to trigger a mass deformation of the @xmath0 symmetry - breaking sector towards the end of inflation .
the first thing to note is that the superpotential in eq .
results in two tachyonic mass eigenstates in global supersymmetry , @xmath419 where @xmath420 and @xmath421 denote the real scalar degrees of freedom ( dofs ) contained in @xmath50 and @xmath393 , respectively .
these masses , however , receive corrections in supergravity that may render them non - tachyonic , @xmath422 here , the corrections proportional to @xmath423 are nothing but the usual hubble - induced masses that scalar fields typically acquire during inflation . apart from that , we collect all further corrections that explicitly depend on @xmath78 in the field - dependent mass contributions @xmath424 and @xmath425 . just like the hubble - induced masses , these corrections result from the f - term tree - level potential in supergravity . as an important aside , we point out that the hubble - induced masses in eq . are , in fact , sensitive to higher - dimensional operators in the khler potential between the fields @xmath50 and @xmath393 and the polonyi field , @xmath426 where @xmath427 and @xmath428 denote two unknown functions of @xmath43 and @xmath429 and where @xmath430 and @xmath431 represent appropriate cut - off scales .
note that such couplings between @xmath50 , @xmath393 , and @xmath43 in the khler potential are the only relevant means of communication between the susy - breaking sector and the @xmath0 symmetry - breaking sector in our model .
for instance , we could imagine that both sectors only communicate with each other via gravitational interactions .
in that case , the scales @xmath430 and @xmath431 should be identified as the planck scale . alternatively
, we may assume that the @xmath0 symmetry - breaking sector couples to the susy - breaking sector via the exchange of heavy messenger particles of mass @xmath432 .
then , @xmath433 would have to be identified as a typical mass scale of the messenger sector , @xmath434 . to illustrate the effect of the khler potential in eq .
, let us suppose that the leading - order contributions to @xmath427 and @xmath428 take the following form , @xmath435 such `` mixing terms '' shift the the hubble - induced masses in eq . .
consider , e.g. , the mass of @xmath436 , @xmath437 where the ellipsis stands for further , subdominant corrections . @xmath438 and
@xmath439 are modified in a similar way . for negative coefficients
@xmath440 and @xmath441 , these corrections may help establish a strong hierarchy between the masses @xmath442 and @xmath443 on the one hand and the inflationary hubble rate @xmath32 on the other hand , @xmath444 so as to make sure that , during inflation , all of the scalar fields @xmath420 and @xmath421 are sufficiently stabilized . at the same time
, it is clear that the khler potential in eq .
significantly complicates the analysis of the @xmath0 symmetry - breaking sector . for presentational purposes
, we will , therefore , neglect the effect of @xmath445 in the following .
we have checked explicitly that , accounting for the exemplary khler potential in eq .
, all results remain quantitatively the same the only difference being the hierarchy in eq . . in this sense , all results in this and the following section should be regarded as approximate results that convey the general flavor of our mechanism of late - time @xmath0 symmetry breaking . an exact numerical study including a more systematic treatment of the functions @xmath427 and @xmath428 in eq .
is left for future work . for our purposes , it is merely important to remember that the masses of the fields @xmath420 and @xmath421 during inflation may be further increased by adding higher - dimensional terms to the khler potential .
the idea now is to choose the coupling constants @xmath394 and @xmath174 such that the scalar dof @xmath436 is stabilized during inflation , turning tachyonically unstable only at a certain ( small ) inflaton field value @xmath446 , @xmath447 to determine the required values for @xmath394 and @xmath174 in order to implement such a scenario , we first note that the field - dependent mass @xmath448 in eq
. vanishes at @xmath81 for the following choice of @xmath394 , @xmath449 in other words , for @xmath450 , the hubble - induced mass compensates the tachyonic mass , @xmath451 , for all field values during inflation , i.e. , as long as @xmath452 , and only at the origin the scalar @xmath436 becomes massless . at the same time
, we need to make sure that the scalar dofs contained in @xmath393 are stabilized at all times .
this is achieved by requiring that their respective masses in global supersymmetry , @xmath453 , are always outweighed by the hubble - induced mass , even at the origin , where the hubble - induced mass is smallest , @xmath454 as long as this condition is satisfied , the field @xmath393 is stabilized at a nonzero vev , vanishes during and after inflation . in supergravity ,
the scalar potential , however , contains a tadpole term linear in @xmath455 ( due to the nonzero f - term of the field @xmath393 ) that displaces @xmath456 from the origin . ]
@xmath457 which induces a positive majorana mass for the field @xmath50 via the waterfall superpotential in eq . ,
@xmath458 this field - dependent mass is contained in @xmath424 in eq . and , similarly as in ordinary f - term hybrid inflation , it helps to stabilize the field @xmath50 during inflation . only at the end of inflation , @xmath424 and , in particular , the mass term in eq .
vanish , so that @xmath50 has a chance of becoming unstable .
let us now study the inflaton - dependent mass corrections in eq . more closely .
in doing so , it turns out to be convenient to use @xmath459 and @xmath460 as reference values for @xmath394 and @xmath174 . to this end , we define @xmath461 the field - dependent shift @xmath462 in the mass of the scalar dof @xmath436 can then be written as follows ,
@xmath463 \frac{v_0}{m_{\rm pl}^2 } \frac{\varphi^2}{2m_{\rm pl}^2 } + \mathcal{o}\left(\varphi^4\right ) \,.\end{aligned}\ ] ] the parameter @xmath464 is supposed to be chosen , such that the total mass @xmath448 vanishes at @xmath465 , @xmath466 for @xmath467 , this condition yields a quadratic equation for @xmath468 , which has two independent solutions , @xmath469 here , the first solution is close to unity , @xmath470 , so that @xmath471 . for @xmath472 ,
the hubble - induced mass @xmath473 cancels the tachyonic mass , @xmath451 , to good precision and the field - dependent correction @xmath474 nearly vanishes on its own .
the second solution is , by contrast , much larger , @xmath475 . in that case
, the individual contributions to @xmath476 do not vanish independently .
instead , all contributions are large , have different signs , and simply happen to cancel each other at @xmath465 .
as we will see further below , the solution @xmath477 will turn out to be unviable for phenomenological reasons . for now , we will , however , remain impartial and treat both solutions on the same footing . combining our results in eqs .
, , , and , we eventually obtain our final expressions for the parameter @xmath394 .
again , we find two solutions , @xmath478 these expressions mark one of the main results in this paper : provided that the parameter @xmath394 in the superpotential @xmath396 in eq
. takes either of these two values , the scalar field @xmath436 becomes unstable , once the inflaton field @xmath78 reaches the value @xmath446 at the end of inflation or after inflation .
this demonstrates how varying the value of the parameter @xmath394 puts us in the position to control the time at which @xmath0 symmetry is spontaneously broken via gaugino condensation at the end of inflation . for small values of @xmath174 and @xmath395 ( such that @xmath479 and @xmath480 ) , the solutions for @xmath394 in eq .
reduce to more compact expressions , @xmath481 as evident from eq .
, the first solution , @xmath482 , depends only weakly on the field value @xmath446 .
in fact , for given values of all other free parameters ( @xmath38 , @xmath47 , @xmath395 , and @xmath174 ) , matching @xmath394 with @xmath483 for a reasonable value of @xmath446 requires some amount of fine - tuning .
to first approximation , all values @xmath483 correspond to the reference value @xmath459 , their variation with @xmath446 being a subdominant effect , @xmath484 in the context of our model , this is , however , not a problem . recall that the ultimate purpose of the superpotential @xmath396 in eq .
is to help us in addressing the cc problem . from that point of view
, it is expected that some ( if not all ) of the parameters involved in determining the final value of the cc end up being subject to some amount of fine - tuning .
in addition , we emphasize that the value of @xmath394 may have an important impact on the time when inflation ends . in regions of space ( or the `` string landscape '' ) where @xmath394 takes too small a value , gaugino condensation never occurs , i.e. , inflation never ends . on the other hand , in regions of space where @xmath394 takes too large a value , the constant in the superpotential , @xmath346 ,
is already generated at early times .
inflation then only lasts for a few @xmath250-folds or does not take place at all .
universes in which @xmath394 matches @xmath483 for a reasonable value of @xmath446 , i.e. , universes in which inflation lasts for a long but finite time interval , therefore , eventually occupy the largest spatial volume among all post - inflationary universes .
this might explain why , in habitable universes , @xmath394 appears to be fine - tuned . finally , we evaluate the total @xmath436 mass @xmath448 in eq . for the two solutions for @xmath394 in eq .
, @xmath485 \frac{\varphi^2-\varphi_0 ^ 2}{m_{\rm pl}^2 } + \mathcal{o}\left(\varphi^4\right ) \ , , \label{eq : mpm2}\\ \nonumber \alpha \rightarrow \alpha_2 \quad\rightarrow\quad m_{p^-}^2\left(\varphi\right ) & = \left[2\left(1+b\right)^2\vphantom{\frac{3(-b)}{4\left(1+b\right)^2 } } \frac{v_0}{\varphi_0 ^ 2 } + \mathcal{o}\left(m_{\rm pl}^{-2}\right)\right ] \frac{\varphi^2-\varphi_0 ^ 2}{\varphi_0 ^ 2 } + \mathcal{o}\left(\varphi^4\right ) \,,\end{aligned}\ ] ] which nicely illustrates how the scalar dof @xmath436 turns tachyonic , once the inflaton field @xmath78 reaches the value @xmath446 . in this sense ,
polonyi inflation does end in a `` waterfall transition '' , after all .
but , instead of taking place in the inflaton sector itself , this transition occurs in separate hidden sector .
the inflaton and waterfall sectors are , therefore , separated in our model , which allows us to bring together two phenomena that would otherwise mutually exclude each other : ( i ) inflation in a scalar potential equivalent to that of ordinary f - term hybrid inflation and ( ii ) supersymmetry breaking at a very high scale . for a discussion of the tension between these two phenomena and a possible resolution in the context of string theory ,
see @xcite .
we note that , in contrast to @xcite , our scenario represents a fully field - theoretic construction . in closing
, we also mention that , adding the particular khler potential in eq . , the expressions for @xmath486 in eq . remain quantitatively the same .
the only effect of @xmath445 is that it increases the overall magnitude of the terms in square brackets in eq . .
we have explicitly checked numerically that , if we choose @xmath487 , @xmath488 , and @xmath489 in eq .
, our first solution for @xmath486 exceeds the inflationary hubble rate for all times during inflation , @xmath490 . at the same time
, for field values slightly below @xmath491 , the absolute value of @xmath486 is again much larger than the hubble rate , @xmath492 .
this ensures that the scalar field @xmath436 quickly reaches the minimum of the scalar potential .
meanwhile , all other scalars always have masses that are large compared to @xmath32 , so that they are safely stabilized at the origin .
a possible interpretation of these numerical findings is to suppose the existence of an other strongly coupled hidden gauge sector around the scale @xmath488 .
if the singlet @xmath50 is the only field that is strongly coupled to this new dynamical sector ( meaning that @xmath393 and @xmath43 are only weakly coupled ) , simple arguments from nda automatically result in a khler potential of just the desired form . , the kinetic term for the field @xmath50 becomes singular at @xmath493 , as long as we only assume the khler potential in eq . .
for this reason , we expect additional higher - dimensional operators in @xmath427 to become important around @xmath493 , which regulate the kinetic term for @xmath50 ( see eq . ) . here , one simple solution is to assume that @xmath427 as a function of @xmath78 simply saturates at a maximal negative value , @xmath494 , around field values of @xmath495 , so that the prefactor of the kinetic term for @xmath50 always remains positive , @xmath496 . at large field values , @xmath497 , the field @xmath50
is then stabilized by the field - dependent mass in eq . , while for @xmath498 , it is stabilized by the terms in eq . . ] in particular , such new strong dynamics do not induce a dangerous @xmath499 term in the khler potential ( which would otherwise spoil slow - roll inflation ) , as long as the polonyi field is only weakly coupled to the new sector .
however , we emphasize that this picture of a new gauge sector around @xmath488 is just a simple example to illustrate what additional dynamics could possibly lead to a cut - off scale of @xmath500 in the @xmath501 operator in the noncanonical khler potential @xmath445 . from the perspective of our model , the main consequence of @xmath445 in eq .
is that it provides us with the ability to increase the absolute value as well as the gradient of @xmath486 as a function of @xmath78 .
apart from that , our simplified analysis based on @xmath502 captures all relevant aspects of our mechanism of @xmath0 symmetry breaking . for this reason
, we leave a more detailed study of @xmath445 and its uv origin for future work .
the `` waterfall superpotential '' @xmath396 in eq .
not only triggers a mass deformation in the @xmath0 symmetry - breaking sector , it also results in planck - suppressed corrections to the inflaton potential in eq . .
during inflation , i.e. , as long as @xmath71 , the scalar polonyi potential now reads @xmath503 where the coefficients @xmath504 , @xmath505 , and @xmath506 correspond to the coefficients @xmath328 , @xmath330 , and @xmath332 in eq
. evaluated at @xmath507 and multiplied by correction factors that stem from the @xmath0 symmetry - breaking sector , @xmath508 \left.c_4\right|_{w=0 } \,.\end{aligned}\ ] ] here , we have introduced the parameter @xmath509 to measure the relative importance of these new corrections , @xmath510 which is nothing but the ratio of the @xmath393 and @xmath43 f - terms during inflation , @xmath511 . to make sure that the @xmath0 symmetry - breaking sector does not disturb the inflationary dynamics , we have to require that @xmath509 is sufficiently small . in view of the expression for @xmath505 in eq .
, we have to demand in particular that @xmath512 since otherwise the inflaton field will receive too large a hubble - induced mass , @xmath513 .
this constraint then translates into upper bounds on the mass scale @xmath395 in the @xmath0 symmetry - breaking sector , @xmath514 the scale @xmath395 is a free parameter , which suggests that we should actually always be able to satisfy these bounds . at the end of this section , we will , however , see that the upper bound @xmath515 turns out to be too restrictive , so that only our first solution , @xmath482 , remains phenomenologically viable .
to get there , we first need to examine the true vacuum after the end of inflation as well as the tuning of the cc in more detail . in global supersymmetry , the vevs of the fields @xmath43 , @xmath393 , and @xmath50
are found to be @xmath516 the fluctuations around the true vacuum are , therefore , described by the following superpotential , @xmath517 where @xmath518 denotes the singlet field @xmath50 shifted by its vev in global supersymmetry , @xmath519 .
eq . tells us that , from the perspective of global supersymmetry , the f - term of the field @xmath393 completely vanishes in the true vacuum , so that the polonyi field @xmath43 remains as the only susy - breaking field .
in addition , we find that the fields @xmath393 and @xmath50 now share a common supersymmetric dirac mass . in global supersymmetry ,
the scalar mass eigenvalues at low energies are consequently given as follows , @xmath520 where @xmath270 , @xmath521 , and @xmath522 stand for the complex scalars contained in the chiral fields @xmath43 , @xmath393 , and @xmath518 , respectively . in order to find the sugra corrections to the vevs in eq . and
the masses in eq . , we expand the full sugra scalar potential up to second order in the fluctuations @xmath270 , @xmath521 , and @xmath522 , @xmath523 here , the coefficients of the linear terms ( @xmath524 , @xmath525 , @xmath526 ) , the masses around the true vacuum ( @xmath527 , @xmath528 , @xmath529 ) as well as the mass mixing parameters ( @xmath530 , @xmath531 , @xmath532 ) take , to leading order , the following form , @xmath533 given that @xmath534 ( see eqs . and ) , the mass mixing among @xmath270 , @xmath521 , and @xmath522 turns out be negligible . setting the off - diagonal masses @xmath530 , @xmath531 , and @xmath532 in eq . to zero , we then arrive at the following simple expressions for the vevs of the fields @xmath43 , @xmath393 , and @xmath50 in supergravity , @xmath535 which results in the following vacuum energy density , @xmath536 where the inflaton term , @xmath537 , clearly constitutes the largest correction to the leading @xmath328 term . in order to make the cc @xmath538 vanish
, we have to fine - tune the constant @xmath66 , so that it takes the following value , @xmath539 here , @xmath540 denotes the `` asymptotic gravitino mass '' in the limit @xmath541 .
the physical gravitino mass , on the other hand , follows from plugging our results for @xmath357 as well as for the field vevs into eq . ,
@xmath542 \left(\frac{1-\gamma}{\sqrt{3}}+\frac{\left<\phi\right>}{m_{\rm pl}}\right ) \frac{\mu^2}{m_{\rm pl } } \ , , \quad \left<\phi\right > \simeq \frac{\sqrt{3}\,\gamma}{1-\gamma+\gamma^2}\ , m_{\rm pl } \approx \frac{4}{3}\sqrt{3}\,\gamma m_{\rm pl } \ , .
\label{eq : m32phi}\end{aligned}\ ] ] as evident from eq .
, the parameter @xmath543 controls the size of the polonyi vev in the true vacuum .
in fact , it is a convenient measure for the relation between the individual contributions to the `` asymptotic inflaton mass '' @xmath544 in the limit @xmath541 . in that limit
, we have @xmath545 , @xmath546 , and @xmath547 , with the asymptotic inflaton mass @xmath544 being given as follows , @xmath548 here , @xmath549 denotes the effective inflaton mass close to the origin for @xmath507 , whereas @xmath550 represents the correction to @xmath544 in the true vacuum , i.e. , the correction appearing after the generation of the constant @xmath66 . making use of these definitions ,
we recognize that @xmath543 parametrizes nothing else than the following ratio , @xmath551 in order to stabilize the polonyi field at a sub - planckian field value after the end of inflation , @xmath552 , we have to require that the additional sugra correction generated in the course of late - time @xmath0 symmetry breaking , @xmath550 , always remains smaller than the effective one - loop mass @xmath233 , @xmath553 we remark that this requirement is a consequence of the fact that @xmath233 and @xmath540 are controlled by the same dynamics , i.e. , by the dynamical scale @xmath41 and the yukawa coupling @xmath47 .
these two masses are , therefore , potentially of the same order of magnitude , in which case the sugra correction @xmath550 threatens to destabilize the minimum of the effective potential in global supersymmetry .
for this reason , we have to explicitly impose the requirement @xmath554 as an extra condition . once this condition is satisfied , our results in eqs . , , and simplify considerably . for small values of @xmath543 , we find @xmath555 in combination with eqs . and , these results lead to simple expressions for @xmath556 , @xmath456 , and @xmath385 that are valid for @xmath557 . neglecting all effects of @xmath558 and @xmath559 wherever possible , we obtain @xmath560 these expressions are the sugra counterpart to the vevs in eqs . and represent our final results for @xmath556 , @xmath456 , and @xmath385 . in this vacuum , also the fields @xmath393 and @xmath50 have nonzero f - terms .
these f - terms are suppressed compared to the polonyi f - term by a factor of @xmath561 and are , hence , negligible .
the masses @xmath540 and @xmath233 scale as follows with the yukawa coupling @xmath47 and the dynamical scale @xmath41 , @xmath562 for a fixed value of @xmath41 , the requirement that @xmath233 must exceed @xmath540 then results in a lower bound on @xmath47 , @xmath563 but this is not the end of the story .
we must also make sure that the final polonyi vev , @xmath564 , lies below the critical field value @xmath227 , so as to stay in the quadratic part of the effective potential close to the origin , @xmath565 otherwise , the mass term in the effective potential disappears altogether and the polonyi field rolls back to field values of the order of the planck scale .
this requirement yields an even stronger bound on @xmath47 , @xmath566 together with the requirement of perturbativity , this limits the range of viable @xmath47 values to @xmath567 .
having determined the position of the true vacuum after inflation in field space , we are now finally in the position to link the @xmath0 symmetry - breaking sector to the iyit sector . in order to tune the cc in the true vacuum to zero ,
the constant @xmath66 generated in the @xmath0 symmetry - breaking sector ( see eq . ) needs to be matched with the value @xmath357 dictated by the iyit sector ( see eq . ) .
for @xmath411 , we have @xmath568 which results in a condition on the dynamical scale @xmath569 in the @xmath0 symmetry breaking sector , @xmath570 this solution for @xmath379 is only consistent as long as it is smaller than the heavy - quark mass scale @xmath571 , i.e. , as long as the gaugino condensation scale is smaller than heavy - quark mass scale , @xmath572 ( see eq . ) .
this requirement can be reformulated as an upper bound on the susy - breaking scale @xmath38 , @xmath573 in supergravity , @xmath385 follows from eq . .
depending on our choice for @xmath394 , we find two solutions , @xmath574 together with the limits on the scale @xmath395 in eq .
, these relations imply upper bounds on @xmath41 , @xmath575 once this condition is satisfied , there always exists a sufficiently small value of @xmath569 that allows us to tune the cc to zero ( so that @xmath576 , while at the same time @xmath577 ) .
we , however , note that the upper bound on @xmath41 in the case of the second solution is too severe .
for such a small dynamical scale , the vacuum energy density during inflation does not suffice to account for the scalar spectral amplitude @xmath58 .
the requirement of successful polonyi inflation , therefore , singles out the fine - tuned first solution @xmath578 .
we now have everything at our disposal that we need to study the phenomenological implications of our model . as argued in sec .
[ subsec : sugra ] , the scalar potential driving inflation takes the following form ( see eq . ) , @xmath579 where @xmath80 and @xmath580 are given in eqs . and , respectively , and where @xmath581 and @xmath306 follow from eq .
, @xmath582 this scalar potential is identical to the inflaton potential of f - term hybrid inflation ( including corrections from supergravity and a noncanonical khler potential ) in the limit of a vanishing gravitino mass .
its implications for the inflationary cmb observables have been studied for the first time in @xcite . in the following , we shall review the aspects of this inflationary scenario that are most relevant for our purposes and summarize the predictions for the inflationary cmb observables in our notation .
first of all , we note that the potential in eq .
has an inflection point at the following field value , @xmath583^{1/2 } \
, , \quad v''\left(\varphi_{\rm flex}\right ) = 0 \ , .
\label{eq : phiflex}\end{aligned}\ ] ] _ a priori _ , the sign of the potential gradient at @xmath584 is undetermined . in particular , the inflection point at @xmath584 turns into a saddle point , @xmath585 , once the following relation is satisfied , @xmath586 this condition can be fulfilled by setting the coefficient @xmath48 to a particular critical value @xmath587 , @xmath588 which solely depends on the size of the yukawa coupling @xmath47 . for @xmath47 values of @xmath55 ,
the critical coefficient @xmath587 takes values of @xmath589 .
once we have @xmath590 , the potential gradient at @xmath584 is negative , @xmath591 , whereas , for @xmath592 , the potential gradient at @xmath584 is positive , @xmath593 . for @xmath590 , the scalar potential ,
therefore , exhibits a local minimum ( @xmath286 ) as well as a local maximum ( @xmath285 ) in the vicinity of @xmath594 , @xmath595^{1/2 } \
, , \quad \varphi_{\rm max } \leq \varphi_{\rm flex } \leq \varphi_{\rm min } \,.\end{aligned}\ ] ] note that these two field values become identical , once the condition in eq .
is satisfied , @xmath596 the presence of two local extrema in the scalar potential may be regarded as a disadvantage , as it requires the initial inflaton field value , @xmath597 , to be smaller than @xmath598 .
for one thing , this requirement may necessitate some amount of fine - tuning , @xmath599 , so as to achieve a sufficient number of @xmath250-folds of inflation .
for another thing , the constraint @xmath600 is not compatible with the notion that inflation is expected to descend from a `` planck epoch '' , during which the inflaton field , the potential energy density and all other relevant quantities take values of the order of the planck scale .
for these reasons , we consider the option @xmath601 less likely than the alternative possibility @xmath602 .
we can take this argument even one step further by making the explicit assumption that , for one reason or another , @xmath597 is necessarily larger than the planck scale , @xmath603 . under this assumption ,
@xmath48 _ must _ be smaller than @xmath587 , as the inflaton would otherwise get trapped in the false vacuum at @xmath604 . from this perspective , the critical coefficient @xmath587 plays the role of an upper bound on @xmath48 , which decides whether the inflaton field has a chance of reaching the true vacuum or not . in fig .
[ fig : inflation ] , we plot the scalar potential in eq . for an @xmath48 value that is slightly smaller than the critical value @xmath587 . in this case , the scalar potential features a flat plateau around @xmath605 , which gives rise to successful inflation in accord with the observational data . ) . here
, the values for the free parameters in the inflaton sector have been chosen as in the benchmark scenario discussed in sec .
[ subsec : bench ] : @xmath606 , @xmath607 , and @xmath608.,scaledwidth=67.0% ] last but not least , we note that the scalar potential in eq . only allows for slow - roll inflation within a limited range of field values .
the slow - roll conditions , @xmath609 and @xmath610 ( see eq . ) , are only satisfied in the interval @xmath611 , where the two boundary values @xmath612 are given as follows , , as @xmath613 depends on @xmath614 . in practice , this dependence is , however , very weak , @xmath615 , so that the right - hand side of eq .
readily yields the desired values of @xmath612 . ]
@xmath616^{1/2 } \
, , \quad m_\pm = \left(m_\varphi^2 \pm \left|\eta_{\rm max}\right|\frac{v\left(\varphi_{\rm sr}^{\pm}\right ) } { m_{\rm pl}^2}\right)^{1/2 } \ , .
\label{eq : phisr}\end{aligned}\ ] ] here , @xmath617 denotes the maximally allowed absolute value of the slow - roll parameter @xmath45 .
the slow - roll parameter @xmath82 is , by comparison , always subdominant during inflation , @xmath85 . for definiteness
, we will use @xmath618 in the following .
finally , we also mention that , if we formally take the limit @xmath619 in eq . , the expression for @xmath620 in eq .
reduces to our result for @xmath594 in eq . .
for the parameter region of interest , @xmath567 ( see eq . ) , the field value @xmath621 is always larger than the critical value field @xmath227 associated with the quark - meson phase transition in the iyit model .
inflation , therefore , always ends , once @xmath78 reaches @xmath621 .
that is , inflation ends because the slow - roll conditions become violated , and not because of a sudden phase transition ( such as the one in the @xmath0 symmetry - breaking sector ) triggered by a small inflaton field value .
correspondingly , the number of @xmath250-folds of slow - roll inflation , @xmath622 , in dependence of the inflaton field value is always given by the following integral , @xmath623 where @xmath622 is defined such that it counts the remaining number of @xmath250-folds until the end of inflation . in the following , we will work in the approximation that the number of @xmath250-folds in between the end of slow - roll inflation and the onset of reheating is negligible .
this is to say that we assume all scalar fields to settle at their respective vevs sufficiently fast , once slow - roll inflation has ended . solving the integral in eq . and
inverting the relation @xmath624 then provides us with the inflaton field value as a function of the number of @xmath250-folds , @xmath625 . for the purposes of this paper , we perform these steps numerically .
the function @xmath626 is exactly what we need to determine our predictions for the cmb observables ( the scalar spectral amplitude @xmath58 , the scalar spectral index @xmath88 as well as the tensor - to - scalar ratio @xmath60 ) , @xmath627 here , the asterisk indicates that all quantities are to be evaluated at @xmath628 , where @xmath629 denotes the required number of @xmath250-folds during slow - roll inflation . according to the latest results of the planck collaboration , the @xmath630 confidence intervals for these observables
are given as follows @xcite , @xmath631 the scalar potential in eq .
is basically controlled by three free parameters : the dynamical scale @xmath41 , the yukawa coupling @xmath47 , as well as the coefficient in the higher - dimensional khler potential , @xmath48 .
this parametric freedom always allows us to choose @xmath41 , such that we succeed in reproducing the correct normalization of the scalar power spectrum , @xmath632 . after fixing @xmath41 in this way
, we then have to deal with two free parameters@xmath47 and @xmath48 ( see also the discussion at the end of sec .
[ subsec : ingredients])which leaves us with the task of studying the predictions for the inflationary cmb observables as functions of @xmath47 and @xmath48 . and the coefficient in the noncanonical khler potential , @xmath48 . here
, the @xmath48 values on the vertical axis are normalized by the critical values @xmath587 , see eq . .
for @xmath633 , the vev of the polonyi field begins to exceed the critical field value @xmath227 , see eq . , whereas for @xmath634 , the polonyi field begins to take super - planckian values during inflation ( this follows from eq . ) .
the black contours respectively indicate the values of @xmath41 * ( left panel ) * and @xmath28 * ( right panel ) * required to obtain the correct scalar spectral amplitude , @xmath632.,title="fig:",scaledwidth=48.0% ] and the coefficient in the noncanonical khler potential , @xmath48 . here , the @xmath48 values on the vertical axis are normalized by the critical values @xmath587 , see eq . .
for @xmath633 , the vev of the polonyi field begins to exceed the critical field value @xmath227 , see eq . , whereas for @xmath634 , the polonyi field begins to take super - planckian values during inflation ( this follows from eq . ) .
the black contours respectively indicate the values of @xmath41
* ( left panel ) * and @xmath28 * ( right panel ) * required to obtain the correct scalar spectral amplitude , @xmath632.,title="fig:",scaledwidth=48.0% ] the outcome of our analysis is shown in fig . [
fig : scan ] . both panels of fig .
[ fig : scan ] indicate where in parameter space our prediction for @xmath88 falls into the observed @xmath635 range .
interestingly enough , we are able to reproduce the observed value @xmath636 only in the `` inflection - point regime '' ( @xmath602 ) , in which the scalar potential does not exhibit any local extrema close to the inflection point at @xmath584 . in the `` hill - top regime '' ( @xmath601 ) ,
on the other hand , the scalar spectral index always turns out to be small , @xmath637 . if we adopt the notion that the critical coefficient @xmath587 should , in fact , be regarded as an upper bound on @xmath48 ( because the inflaton would otherwise get trapped in a false vacuum )
, this result entails the following interesting observation : maybe the coefficient @xmath48 has , for one reason or another , the tendency to saturate its upper bound from below , @xmath638 . the coefficient
@xmath48 might , e.g. , be intrinsically of @xmath55 ; and it only ends up being slightly suppressed , @xmath57 , because it is bounded from above .
alternatively , @xmath48 might tend to be close to @xmath587 , because such a parameter choice leads to a particularly flat potential and , hence , to a particularly long period of inflation ( i.e. , to a large spatial volume in which @xmath638 ) .
but anyhow , no matter what the origin of this fine - tuning is , the point is : once we suppose that there is a physical reason that singles out @xmath48 values close to @xmath587 , we are automatically led to values of @xmath88 close to the observed value . or put differently , it is an intriguing coincidence that the observed value of @xmath88 is reproduced just in the vicinity of the only special point on the @xmath48 axis .
one could have expected that successful inflation would require @xmath639 .
but then it would not have been guaranteed that the resulting values of @xmath88 would end up being close to @xmath636 .
another interesting coincidence pertains to the required value of the dynamical scale @xmath41 . as evident from the left panel of fig .
[ fig : scan ] , the scale @xmath41 needs to take values very close to the scale of grand unification , @xmath640 this suggest the fascinating possibility that the scales of supersymmetry breaking , inflation , and grand unification might , in fact , all be unified , @xmath641 a trivial condition for this kind of `` triple unification '' is that the energy scales of inflation and supersymmetry breaking are identical to each other .
the unification of the three scales @xmath6 , @xmath40 , and @xmath642 can , thus , only be realized in the context of polonyi inflation and in no other inflation model . in this paper , we have not made any attempt to embed our model into a larger gut framework .
more work on the possible connection between polonyi inflation and grand unification is , therefore , certainly needed . as evident from the right panel of fig .
[ fig : scan ] , the gravitino turns out to be superheavy in our model , @xmath643 as already pointed out in the introduction ( see eq . ) , the gravitino mass is directly related to our prediction for the tensor - to - scalar ratio @xmath60 ( this relation readily follows from @xmath644 in our model ) , @xmath645 the experimental confirmation of this relation is certainly challenging but _ if _ it should be accomplished , it would represent an unequivocal smoking - gun signal for polonyi inflation , i.e. , for a close link between the dynamics of supersymmetry breaking and inflation . in the minimal framework of pure gravity mediation , gravitino masses of @xmath646 typically result in mssm gaugino masses
@xmath647 of @xmath648 , @xmath649 where the suppression by the loop factor @xmath650 is a consequence of anomaly mediation @xcite . here , the wino @xmath651 typically ends up being the lightest gaugino .
at tree level , the gaugino masses vanish in our model . to see this , recall that the polonyi field @xmath43 carries @xmath0 charge @xmath16 .
couplings between the polonyi field and the sm gauge fields of the form @xmath652 are , thus , forbidden in the superpotential . after inflation , the polonyi field predominantly decays into gravitinos ( via the @xmath653 term in the effective khler potential @xcite ) , so that inflation is followed by a phase of gravitino domination in our scenario @xcite . the nonthermal wino production in gravitino decays then easily overcloses the universe @xcite .
to achieve an acceptable present - day wino abundance , we , therefore , have to assume that the wino mass is somehow suppressed .
fortunately , this is possible in the context of pure gravity mediation , where the gaugino masses in eq .
also receive threshold corrections from higgsino loops @xcite .
these loop corrections are potentially of the same order of magnitude as the anomaly - mediated masses .
the different contributions to the wino mass can , in particular , cancel , so that the mass @xmath654 is reduced down to , which would even require some extra mass contributions .
a split - susy spectrum may , e.g. , be achieved , if supersymmetry breaking is a pure sugra effect @xcite .
it is an interesting open question how to realize a split - susy spectrum in the context of dsb models la iyit .
we leave this question for future work . for now , we simply assume that the contribution to @xmath654 from anomaly mediation ends up being canceled by another mass contribution of equal magnitude and opposite sign . ]
@xmath655 we shall assume that this is case for anthropic reasons . in case the wino mass
is in the tev range , the wino reaches thermal equilibrium after reheating and eventually represents an excellent candidate for dark matter in the form of _ weakly interacting massive particles _ or wimps .
larger wino masses , on the other hand , result in an overabundance of dark matter , thereby making our universe hostile for life .
the phase of gravitino domination after polonyi inflation ends once the gravitino begins to decay into the massless fields of the mssm . here
, the corresponding gravitino decay rate is given as @xcite @xmath656 the reheating temperature reached during gravitino decay , @xmath657 , then takes the following value , @xmath658 where @xmath659 denotes the effective number of relativistic dofs in the mssm at high temperatures .
contrary to many other inflation models , the reheating temperature after polonyi inflation is , therefore , not a free parameter , which is controlled by the unknown strength of the inflaton couplings to matter .
@xmath657 rather follows directly from the universal decay rate of the gravitino , which only depends on the gravitino mass . as the gravitino mass is , in turn , more or less fixed by the amplitude of the scalar power spectrum in polonyi inflation ( see fig . [
fig : scan ] ) , we arrive at the interesting result that polonyi inflation makes a definite prediction for the reheating temperature : @xmath660 . remarkably enough , this is not far away from the temperature that is needed for the successful realization of thermal leptogenesis @xcite . in its simplest form ,
thermal leptogenesis requires the reheating temperature to be at least of @xmath661 . but
resonance effects in the case of mildly degenerate heavy - neutrino mass spectrum easily increase the efficiency of thermal leptogenesis also at lower temperatures @xcite .
the generation of the baryon asymmetry of the universe after polonyi inflation may , thus , very well be accounted for by thermal leptogenesis . llll quantity & symbol & value [ gev ] & reference + + dynamical scale & @xmath41 & @xmath662 & eq . + effective supersymmetric mass scale & @xmath222 & @xmath662 & eq . + soft susy - breaking mass scale & @xmath185 & @xmath663 & eq .
+ susy breaking scale & @xmath38 & @xmath664 & eq .
+ energy scale of inflation & @xmath665 & @xmath664 & eq .
+ effective potential energy scale at large field values & @xmath245 & @xmath666 & eq . + effective potential energy scale at small field values & @xmath244 & @xmath667 & eq . + effective polonyi mass around the origin & @xmath233 & @xmath668 & eq . + gravitino mass & @xmath28 & @xmath669 & eq . + inflationary hubble rate & @xmath32 & @xmath669 & eq .
+ + vev of the field @xmath50 in the true vacuum@xmath670 & @xmath385 & @xmath671 & eq . + characteristic value for the scale @xmath395 & @xmath672 & @xmath673 & eq .
+ gaugino condensation scale & @xmath674 & @xmath675 & eq .
+ vev of the field @xmath393 in the true vacuum@xmath670 & @xmath456 & @xmath676 & eq . + dynamical scale@xmath670 & @xmath379 & @xmath677 & eq . + constant term in the superpotential & @xmath678 & @xmath679 & eq .
+ mass of the field @xmath50 in the true vacuum@xmath670 & @xmath529 & @xmath680 & eq .
+ mass of the field @xmath393 in the true vacuum@xmath670 & @xmath528 & @xmath680 & eq . + + beginning of slow - roll inflation & @xmath681 & @xmath682 & eq .
+ location of the inflection point & @xmath594 & @xmath683 & eq . + field value @xmath684 @xmath250-folds before the end of inflation & @xmath257 & @xmath685 & eq .
+ end of slow - roll inflation & @xmath621 & @xmath686 & eq . + critical field value & @xmath227 & @xmath679 & eq . + polonyi vev in the true vacuum & @xmath564 & @xmath687 & eq .
+ finally , we shall illustrate our findings by means of a concrete example . to this end
, we will now consider the following point in parameter space , which may be regarded as a representative benchmark scenario , @xmath688 for this choice of @xmath47 , the critical coefficient @xmath587 is given by @xmath689 ( see eq . ) .
our choice of @xmath48 is , therefore , fine - tuned in the sense that it deviates from @xmath587 only by half a per cent or so . for the parameter values in eq . , we then obtain the following predictions for the inflationary cmb observables , @xmath690 where the values for @xmath58 and @xmath88 coincide with the experimental best - fit values by construction .
next to the parameters @xmath41 , @xmath47 , and @xmath48 ( which represent the free parameters in the susy - breaking sector ) , we also need to specify the parameters in the @xmath0 symmetry - breaking sector , i.e. , the coefficients @xmath394 and @xmath174 as well as the energy scale @xmath395 ( see eq . ) . here
, the scale @xmath395 is required to take a value within a finite interval , @xmath691 we recall that the lower bound on @xmath395 is a consequence of the requirement that the heavy - quark mass scale in the @xmath0 symmetry - breaking sector should always exceed the dynamical scale , @xmath692 ( see eq . ) , while the upper bound on @xmath395 follows from the requirement that the dynamics of the @xmath0 symmetry - breaking sector should have no noticeable effect on inflation , @xmath693 ( see eq . ) . in the following
, we shall use the geometric mean of the two boundary values in eq . as a characteristic value for @xmath395 , @xmath694 for the vevs @xmath385 and @xmath456 as well as for the dynamical scale @xmath379 ,
we then obtain ( see eqs . and )
@xmath695 similarly , we find the following relations for the two reference values @xmath459 and @xmath460 ( see eqs . and ) , @xmath696 the actual value of @xmath394 is given by @xmath697 ( see eq . ) , where @xmath698 varies as a function of @xmath446 , i.e , the polonyi field value at the onset of gaugino condensation in the @xmath0 symmetry - breaking sector .
allowing for @xmath446 values in the range @xmath699 , the function @xmath698 roughly takes values between @xmath700 and @xmath701 . up to deviations of the order of one per cent or so
, the coefficient @xmath394 , therefore , coincides with the reference value @xmath459 .
meanwhile , the coefficient @xmath174 can be freely chosen , as long as we respect the constraint @xmath702 .
the parameter values in eqs . , , and now completely fix the numerical properties of our benchmark scenario . in tab . [ tab : benchmark ] , we give an overview of all the resulting numerical values for the various energy scales and field values of interest in our model . tab . [ tab : benchmark ] , thus , provides an example for a possible realization of our model , which illustrates how to achieve successful polonyi inflation as well as spontaneous @xmath0 symmetry breaking at late times in the context of strongly coupled gauge theories .
the large value of the sm higgs boson mass as well as the null result of susy searches at the lhc call for a paradigm shift in our expectations towards the role of supersymmetry in nature . from the perspective of string theory , dark matter , and grand unification ,
supersymmetry still represents a well - motivated extension of the standard model .
but in view of the recent experimental data , we now begin to realize that supersymmetry s main purpose may actually not lie in stabilizing the electroweak scale .
instead , it now appears more likely that supersymmetry is , in fact , spontaneously broken at a scale much higher than the electroweak scale .
this opens up a whole range of new phenomenological possibilities .
as we have demonstrated in this paper , the spontaneous breaking of supersymmetry at a high scale , @xmath703 , may , e.g. , offer a dynamical explanation for the occurrence of cosmic inflation in the early universe . we have dubbed the ensuing inflationary scenario _ polonyi inflation _ , as it identifies cosmic inflation as a natural by - product of spontaneous supersymmetry breaking in an effective polonyi model . generally speaking , the key idea of our proposal is to obtain inflation from the polonyi superpotential , @xmath704 with the polonyi superfield @xmath43 breaking supersymmetry via its nonvanishing f - term , @xmath705 , and acting as the chiral inflaton field at the same time . in the context of polonyi inflation ,
spontaneous supersymmetry breaking and cosmic inflation are nothing but two sides of the same coin : inflation is driven by the vacuum energy density associated with the spontaneous breaking of supersymmetry and the scalar inflaton is identified as the complex polonyi field , i.e. , the pseudoflat direction in the scalar potential of the susy - breaking sector .
this connection between supersymmetry breaking and inflation results in several characteristic parameter relations that cast a new light on well - known quantities .
in polonyi inflation , the hubble rate during inflation , e.g. , is equal to the gravitino mass at low energies , @xmath706 this implies that the amplitude of the primordial density fluctuations , @xmath707 , scales with @xmath28 , @xmath708 there are indications that the observed value of @xmath707 might be the result of anthropic selection @xcite .
given typical values of the tensor - to - scalar ratio @xmath60 in slow - roll inflation , the above relation then points towards very large values of the gravitino mass . in polonyi inflation ,
the ultimate reason why supersymmetry is broken at a high scale , therefore , consists in the necessity of reproducing the anthropic value of @xmath707 . the general concept of polonyi inflation , thus , not only yields an answer to the question _ `` why inflation ? ''
_ , it also explains why supersymmetry is necessarily broken at a very high scale .
moreover , we point out that @xmath709 is an interesting result in view of the stability of the sm higgs potential . recall that , solely within the standard model , the higgs quartic coupling ( most likely ) turns negative at energies around @xmath710 @xcite .
this instability can be remedied by supersymmetry
as long as the soft sparticle masses are at most of @xmath646 . in this sense ,
our result for @xmath28 turns out to saturate the upper bound on the gravitino mass implied by vacuum stability !
this may or may not be a coincidence . in this paper ,
we have constructed a minimal model that illustrates how the idea of polonyi inflation may be realized in the context of strongly coupled supersymmetric gauge theories .
our construction consists of two separate hidden sectors that respectively account for the dynamical origin of the parameters @xmath38 and @xmath66 in eq . . here
, one hidden sector is identical to the simplest version of the iyit model of dynamical supersymmetry breaking .
this sector contains the polonyi field @xmath43 and is responsible for the generation of the susy - breaking parameter @xmath38 via strong dynamics .
the other hidden sector features the same matter content as the susy - breaking sector , but contains less singlet fields .
the masses of the matter fields in this sector are controlled by the vev of a singlet field @xmath50 . at early times , @xmath385 vanishes . at this time , the constant in the superpotential , @xmath346 , is zero .
as we are able to show , the effective scalar potential for the polonyi field at this stage then takes the same form as the inflaton potential of ordinary f - term hybrid inflation in the limit of a vanishing gravitino mass .
it is this potential that gives rise to the actual period of polonyi inflation.albeit , in the true vacuum at low energies , @xmath28 is exceptionally large in our model . ] at small polonyi field values after the end of slow - roll inflation , the hubble - induced mass for the field @xmath50 decreases until , at a certain field value , @xmath50 becomes tachyonically unstable .
this instability results in a waterfall transition in the @xmath0 symmetry - breaking sector , such that @xmath711 .
the second hidden sector then turns into a pure sym theory , in which @xmath0 symmetry is broken via gaugino condensation . in consequence of that
, the constant term @xmath66 appears in the superpotential . here , the final value of @xmath66 needs to be fine - tuned , such that the cc in the true vacuum vanishes . to sum up
, we conclude that our effective polonyi model results in an inflationary scenario similar to ordinary f - term hybrid inflation , apart from a few key differences : ( i ) all corrections to the scalar potential proportional to @xmath28 are missing in our model .
( ii ) this means , in particular , that the scalar potential does not contain any odd powers of the inflaton field . because of that , the inflaton potential does not depend on the complex phase of the inflaton field , as it is usually the case in f - term hybrid inflation .
( iii ) moreover , in our scenario , the low - energy value of the gravitino mass,@xmath28 , is not bounded from above by the inflationary hubble rate .
in polonyi inflation , these two scales are , in fact , equal to each other , @xmath644 , whereas in f - term hybrid inflation we have to demand that @xmath712 , so as to ensure that the slow - roll conditions do not get violated .
( iv ) in polonyi inflation , the inflaton sector does not undergo a waterfall phase transition at the end of inflation . instead , the inflationary vacuum energy density continues to act as the vacuum energy density associated with the spontaneous breaking of supersymmetry after inflation . ( v ) meanwhile , small inflaton field values trigger a ( harmless ) waterfall phase transition in a separate hidden sector , in the course of which only an approximate global @xmath53 symmetry becomes broken .
in contrast to f - term hybrid inflation , polonyi inflation , therefore , does not suffer from the production of dangerous topological defects at the end of inflation .
in the present paper , we have only touched on the phenomenological implications of our model and more work is certainly needed . for one thing ,
the study of the inflationary phase may still be further refined .
as it turns out , inflation takes place at field values only shortly below the planck scale in our model .
inflaton terms of @xmath713 in the scalar potential may , therefore , have a noticeable effect on the predictions for the inflationary cmb observables .
we do not expect these terms to change our conclusions qualitatively .
but quantitatively , they may be relevant .
for another thing , the end of inflation and the subsequent reheating phase require a closer examination .
here , it would be interesting to study the implications of different interactions between the two hidden sectors in the khler potential more comprehensively . also , the actual dynamics of reheating deserve further investigation
. our model does not suffer from the usual polonyi problem , as the decay of the polonyi field is identified with the first stage of the reheating process itself .
but besides that , a dedicated study of reheating tracking the expansion of the universe during the transition to the radiation - dominated era as well as the oscillatory motion of all scalar fields is needed . here
, particular attention should be paid to the study of all possible decay modes of the polonyi field as well as of the gravitino . a better understanding of reheating
would then enable us to better assess the prospects of successful thermal leptogenesis .
finally , the phenomenological implications of our model at low energies need to be explored in more detail . as examples of observational signatures at low energies ,
we merely mention two interesting possibilities : ( i ) in our scenario , a wino lsp with an anthropically selected mass around 3 tev accounts for dark matter .
this wino can be searched for in direct and indirect detection experiments . in addition
, the almost mass - degenerate chargino may be seen in the form of macroscopic charged tracks in collider experiments .
all in all , our model predicts that the neutral and the charged wino are the only sparticles that should show up at low energies .
all other sparticles have masses of at least @xmath648 and are , thus , expected to be decoupled from low - energy phenomenology for the most part .
( ii ) moreover , the discrete @xmath37 symmetry that we use to forbid the constant term in the superpotential at early times predicts the existence of vector - like matter fields charged under the sm gauge group , which cancel the sm contributions to the @xmath37 gauge anomalies .
these vector - like matter fields may have masses within the reach of collider experiments , which would allow to probe our assumption of an anomaly - free @xmath0 symmetry at accessible energies .
last but not least , one should attempt to implement our idea of polonyi inflation into other dynamical models . it would be interesting to assess which other dsb models apart from the iyit model might serve as a uv completion of the inflationary dynamics .
in addition to that , a more systematic study of the polonyi khler potential beyond near - canonical and approximately shift - symmetric khler potentials would be desirable .
maybe there are particular ( exotic ) choices for the khler potential that render it unnecessary to supplement the tree - level polonyi model with radiative corrections , as we have done it in this paper .
likewise , our model of late - time @xmath0 symmetry breaking may be extended or supplemented by alternative means to generate the constant term in the superpotential at the end of inflation . as an alternative to our scenario , which employs the waterfall field @xmath50
, one might consider a direct coupling of the polonyi field to the gauge fields of a strongly coupled pure sym theory via the gauge - kinetic function , @xmath714 for appropriately chosen values of the parameters @xmath715 and @xmath716 and the integer @xmath717 , the sym theory is weakly coupled during inflation , returning to the strongly coupled regime only at the end of inflation .
this results in gaugino condensation and , hence , late - time @xmath0 symmetry breaking at @xmath718 .
we will give a more careful discussion of this mechanism in the context of polonyi inflation elsewhere .
furthermore , gaugino condensation in a pure sym theory is not the only way to break @xmath0 symmetry and other possibilities should be considered as well .
eventually , the model presented in this paper may only be regarded as a first step towards a new understanding of the close relation between inflation and spontaneous supersymmetry breaking .
the concept of polonyi inflation leads us into uncharted territory and we are convinced that , exploring these new avenues , we will encounter both big surprises and valuable insights . at the early stages of this project ,
one of us ( k.s . ) was still employed as a postdoctoral researcher at kavli ipmu .
after the completion of roughly half the project , k.s
. then transferred to mpik .
this work has been supported in part by grants - in - aid for scientific research from the ministry of education , science , sports , and culture ( mext ) , japan , no .
26104009 and no .
26287039 ( t.t.y . ) as well as by the world premier international research center initiative ( wpi ) , mext , japan ( k.s . and t.t.y . ) .
in this appendix , we give the details of our derivation of the effective one - loop potential , @xmath272 , for the polonyi field @xmath209 , i.e. , the pseudoflat scalar direction in the iyit dsb model , see eqs . , , , and in sec .
[ subsec : potential ] .
the starting point of our computation is the coleman - weinberg formula for the effective one - loop potential @xcite , which takes the following form in supersymmetric theories , @xmath719 \ , .
\label{eq : vcw}\end{aligned}\ ] ] here , @xmath720 is the renormalization scale and @xmath721 denotes a dimensionless constant that is introduced for notational convenience .
we note that @xmath720 and @xmath721 are only defined up to the following transformation , @xmath722 so that the constant @xmath721 can always be removed by a finite renormalization .
eq . now tells us that the potential @xmath272 is given as the supertrace ( str ) of a specific function of the mass matrix @xmath723 , which encompasses all bosonic and fermionic mass eigenvalues in the presence of a nonvanishing polonyi field background . in a first step ,
we , therefore , need to determine the full mass spectrum of the iyit sector .
apart from the polonyi field @xmath43 ( which is massless at tree level in global supersymmetry ) , the iyit sector of our model consists of six meson flavors , @xmath177 and @xmath165 , and six singlet fields , @xmath176 and @xmath178 .
the scalar and fermionic masses for these chiral multiplets follow from the effective tree - level superpotential in eq . .
for an arbitrary background value of the polonyi field @xmath78 , we find the following scalar mass eigenvalues , @xmath724^{1/2 } \ ,
, \label{eq : m2bos}\end{aligned}\ ] ] where @xmath725 refers to the zeroth flavor , i.e. , to the fields @xmath177 and @xmath176 , while @xmath726 refers to the @xmath717 other flavors , i.e. , to the fields @xmath165 and @xmath178 .
moreover , the discrete parameter @xmath727 distinguishes between the two different types of particles involved , i.e. , between mesons ( @xmath728 ) and singlets ( @xmath729 ) , while the parameter @xmath730 accounts for the soft mass splitting within each complex scalar , @xmath731 , in consequence of spontaneous supersymmetry breaking . in this respect , eq
. serves as an illustration of our statement in sec .
[ subsec : susy ] , where we note that the mass parameter @xmath185 plays the role of the soft susy - breaking mass scale in the iyit model , see the discussion below eq . .
by setting the soft mass @xmath185 to zero , we then readily obtain the mass eigenvalues of the fermionic components in the meson and singlet multiplets , @xmath732^{1/2 } \ , .
\label{eq : m2fer}\end{aligned}\ ] ] for @xmath733 , all scalar and fermionic masses are different from each other ( as long as none of the dirac masses @xmath191 are identical ) .
away from the origin , the meson and singlet fields , therefore , give rise to , in total , 24 real scalars and 12 majorana fermions . at the origin , @xmath734 and @xmath735 reduce to @xmath736 which illustrates that , for @xmath81 , half of the real scalars and all of the majorana fermions pair up to form complex scalars and dirac fermions , respectively . at @xmath81 , we then end up with 12 real and 6 complex scalars as well as with 6 dirac fermions . here , the complex scalars and the dirac fermions are , in particular , composed half of mesonic and half of singlet dofs . furthermore , we note that none of the masses @xmath734 and @xmath735 ever turns tachyonic , is that the soft susy - breaking mass scale @xmath185 does not exceed any of the supersymmetric dirac masses : @xmath737 for all flavors @xmath464 , which is always satisfied in our model , see eq . . ] which means that the meson and singlet fields always remain stabilized at the origin .
this should be compared with the situation in ordinary f - term hybrid inflation , where the masses of the fhi waterfall fields correspond to ( the nonzero values of ) @xmath734 and @xmath735 in the limit of zero dirac masses , @xmath738 in this case , the scalar mass eigenstates corresponding to @xmath739 are tachyonically unstable , once @xmath740 .
this instability is absent in our `` waterfall transition - free '' version of f - term hybrid inflation .
finally , before moving on and presenting our results for the effective polonyi potential , let us state approximate expressions for @xmath734 and @xmath735 that are valid at small and large values of the polonyi field , respectively , and which will become useful later on . in the low - energy regime close to the origin in field space ,
i.e. , at small values of the order parameter , @xmath741 ( see eq . ) , we find , @xmath742 which reduces to the expressions in eq . in the limit @xmath743 .
conversely , in the high - energy regime , at large values of @xmath219 , the masses @xmath734 and @xmath735 may be approximated as follows , @xmath744 in this regime , the meson masses ( @xmath728 ) asymptotically approach the usual expressions from f - term hybrid inflation , see eq .
, while the the singlet masses ( @xmath729 ) only acquire suppressed masses of @xmath745 . here , note that the small singlet masses in the large - field regime are the result of a seesaw mechanism of a sort : for a large majorana mass @xmath261 , the dirac masses @xmath191 become suppressed by a factor of @xmath746 .
this is exactly what happens in the usual seesaw mechanism , which explains the small masses of the sm neutrinos as the outcome of large majorana and small dirac neutrino masses @xcite .
we are now in the position to calculate the effective one - loop potential for the scalar polonyi field .
the full result for @xmath272 follows from eqs .
, , and and takes a rather complicated form .
for this reason , we refrain from explicitly writing down the full expression for @xmath272 , and merely refer to fig .
[ fig : effpot ] , where we plot the exact result for @xmath272 as a function of the order parameter @xmath219 . instead ,
let us now evaluate @xmath272 for specific field values of interest . for instance ,
at the origin , we find @xmath747 \,,\end{aligned}\ ] ] where @xmath221 counts the number of meson flavors in the iyit model , @xmath748 denotes the effective supersymmetric mass scale in the iyit sector ( see eqs . and ) , and where @xmath749 is a loop function that needs to be evaluated at the respective mass ratios @xmath238 ( see eq . ) , @xmath750 note that this result for @xmath751 can also be obtained by plugging the masses in eq . into the cw formula .
the function @xmath749 is nearly constant over its entire domain , @xmath752 , and thus , well approximated by @xmath753 .
this motivates the following choices for @xmath720 and @xmath721 , @xmath754 for these values of the renormalization scale @xmath720 and the constant @xmath721 , the tree - level vacuum energy density of the iyit model , @xmath208 , receives a small negative shift @xmath755 of @xmath756 , @xmath757 this correction to the vacuum energy density and , hence , the ratio @xmath758 , are bounded from above , @xmath759 here , we have used that @xmath749 takes its smallest value in the limit @xmath760 , where @xmath761 . for perturbative values of the yukawa coupling ,
@xmath192 , we can , therefore , safely neglect the radiative correction @xmath755 . having fixed @xmath720 and @xmath721 ,
let us now study the effective potential for extreme values of the order parameter , i.e. , for @xmath230 and @xmath231 .
below the critical field value , the inflaton - induced mass @xmath261 is smaller than the supersymmetric dirac masses @xmath191 . at energies below @xmath191 , the `` heavy '' meson flavors
can then be integrated out , which results in a quadratic potential for @xmath78 around the origin , @xmath762 where @xmath233 denotes the effective polonyi mass at one - loop level ( see also eq . and
footnote [ fn : meff ] ) , @xmath763 here , @xmath764 represents a loop function that acts as a normalized weight , @xmath765 , for each meson flavor , @xmath766 evaluated at the mass ratio @xmath238 , the function @xmath764 quantifies the contribution from the respective meson flavor to @xmath233 .
correspondingly , the sum over all factors @xmath237 yields the effective number of flavors , @xmath767 , that contribute to the effective polonyi mass .
moreover , we note that , over its entire domain , @xmath768 , the function @xmath764 is well approximated by @xmath239 .
this allows us to rewrite @xmath272 in the following way , @xmath769 which provides us with a couple of alternative expressions for @xmath272 .
for instance , the first line of eq .
makes explicit the dependence of the effective potential on the two mass scales @xmath185 and @xmath261 at small field values , @xmath215 .
it is interesting to note that this result for the effective potential can also be obtained by plugging the approximate expressions for the scalar and fermionic masses in eq . into the cw formula in eq . .
meanwhile , the third line of eq . illustrates how @xmath272 may be expressed as a function of the ratios @xmath770 ( see eq . ) . in this case
, the effective potential turns out to be given by the fourth power of the soft susy - breaking mass , @xmath771 , times a function of the susy - preserving mass parameters @xmath261 and @xmath191 , @xmath772 .
the expressions for @xmath272 in eq .
, therefore , comply with the expectation that , in supersymmetric theories , the effective potential ought to be proportional to the soft susy - breaking mass scale @xmath185 @xcite .
moreover , we can write @xmath272 in the small - field regime directly as a function of the order parameter , @xmath773 where we have introduced @xmath244 as the effective potential energy scale at small field values , @xmath774 finally , let us examine the effective polonyi potential at small field values in the limit of small supersymmetry breaking , i.e. , for a small soft susy - breaking mass @xmath185 in comparison to large susy - preserving dirac masses , @xmath775 .
this limit is best quantified in terms of the geometric mean of all mass ratios @xmath238 , which we will refer to as the hierarchy parameter @xmath521 in the following ( see eq . ) , @xmath776 which can take values between @xmath120 and @xmath118 ( see eq . ) . here
, @xmath777 corresponds the susy - preserving limit , while @xmath778 represents the case of maximal supersymmetry breaking . from eqs . and
, it immediately follows that the hierarchy parameter @xmath521 is , in fact , nothing but an alternative measure for the strength of the yukawa coupling of the polonyi field to the iyit matter fields , @xmath779 that is , for perturbative values of the yukawa coupling @xmath47 , supersymmetry is only mildly broken in the meson and singlet multiplets in the iyit sector , while in the strongly coupled limit , i.e. , for nonperturbative values of the yukawa coupling @xmath47 , the effect of supersymmetry breaking is maximal . in passing ,
we mention that the parameter @xmath521 also quantifies the ratio between the order parameter of our model , @xmath219 , and the order parameter of the waterfall transition in ordinary f - term hybrid inflation , @xmath780 , @xmath781 in the limit of maximal supersymmetry breaking , both order parameters , therefore , coincide with each other . in all other cases , we have @xmath782 .
this implies that , for @xmath783 , the mesonic phase transition in the iyit model ( see fig .
[ fig : effpot ] ) takes place around a critical field value , @xmath315 , that is parametrically larger than the critical field value associated with the waterfall transition in f - term hybrid inflation , @xmath784 .
let us now turn to the limit of small supersymmetry breaking .
for small values of @xmath521 , i.e. , for small mass ratios @xmath238 , the weight function @xmath764 in eq . can be approximated as follows , @xmath785 the effective potential @xmath272 and the potential energy scale @xmath244 at small field values then reduce to @xmath786 we will come back to these results in sec .
[ subsec : kahler ] .
next , we shall examine the effective polonyi potential for large values of the order parameter @xmath219 . in the large - field regime , @xmath787
, the iyit matter fields acquire a large field - dependent mass , such that their `` bare '' supersymmetric masses @xmath191 become irrelevant , @xmath788 . integrating out the matter fields
then results in the usual logarithmic effective potential known from ordinary f - term hybrid inflation , @xmath789 at large field values , the dependence of the effective potential on the mass scales @xmath185 and @xmath261 is , therefore , of the form @xmath213 . up to corrections of @xmath790 , we are able to reproduce this result by inserting the approximate masses in eq . into the cw formula in eq . .
moreover , we can read off the effective potential energy scale in the large - field regime , @xmath245 , from eq . , @xmath791 which is again of @xmath792 .
last but not least , we can rewrite the effective potential as a function of the mass ratios @xmath218 , which provides us with a large - field counterpart to the small - field result in eq . , @xmath793 at tree level ,
the f - term scalar potential is determined by two input functions , the superpotential @xmath9 and the khler potential @xmath10 . here
, the superpotential does not receive any radiative corrections in perturbation theory according to the susy nonrenormalization theorem @xcite .
the khler potential , on the other hand , _ is _ renormalized , which allows one to rewrite parts of the effective action in the form of an effective khler potential .
we shall now demonstrate how this applies to our results in the previous section .
first , we note that , schematically , the full result for @xmath272 in eq .
is of the following form , @xmath794 where the order parameter @xmath795 and the mass ratios @xmath796 denote the ratios of supersymmetric mass scales , while the hierarchy parameter @xmath797 quantifies the ratio between the soft susy - breaking mass scale @xmath185 and the effective supersymmetric mass scale @xmath222 .
the above expression for @xmath272 , thus , illustrates that the effective potential represents a power series in @xmath798 , the coefficients of which are functions of susy - preserving mass parameters . from our results in eqs . and
, we can read off the coefficient function @xmath799 belonging to the leading - order contribution to @xmath272 , @xmath800 as we are now going to show , this result for the function @xmath799 can be reproduced by supplementing the tree - level khler potential , @xmath801 , by an effective one - loop khler potential , @xmath802 , @xmath803 where @xmath801 and @xmath802 are respectively given as follows @xcite , @xmath804 \ , .
\label{eq : k1loop}\end{aligned}\ ] ] here , @xmath805 denotes the fermionic mass matrix , which is identical to the hessian of the superpotential , and @xmath806 is a numerical constant that we will determine shortly . in global supersymmetry , the effective khler potential in eq .
results in the following f - term scalar potential for the polonyi field @xmath209 , @xmath807 where the effective superpotential consists of the polonyi tadpole term , @xmath808 , such that @xmath809 it is important to note that the scalar potential in eq .
is only an approximation of the full effective potential . to see this ,
let us expand the inverse khler metric in powers of the loop factor @xmath810 , @xmath811 the leading contribution to @xmath812 then yields what we shall refer to as the _ truncated effective potential _
, @xmath813 this truncated effective potential coincides with the full effective potential up to corrections of @xmath814 , @xmath815 an explicit proof of this relation is given by intriligator , seiberg , and shih in appendix a.5 of @xcite . for an earlier discussion of the effective action in supersymmetric theories ,
see @xcite . from the relation in eq .
, it is evident that the truncated potential @xmath816 is only a good approximation of the actual ( true ) effective potential , as long as the effect of spontaneous supersymmetry breaking is small , i.e. , as long as @xmath817 . otherwise , the @xmath814 corrections in eq .
become important .
these radiative corrections while present in the full expression for @xmath272 in eq .
cannot be captured by the effective khler potential .
one example for such a correction is , e.g. , the shift in the tree - level vacuum energy density , @xmath755 ( see eq . ) , which is proportional to @xmath818 and which can not be explained in terms of an effective khler potential . instead , the shift in the vacuum energy density @xmath755 and all other @xmath814 corrections follow from an effective potential for the auxiliary component of the polonyi multiplet , @xmath819 .
this _ effective auxiliary field potential _ contains higher - order terms in @xmath820 and is only negligible , once the effect of supersymmetry breaking is nothing but a small correction to the otherwise supersymmetric dynamics @xcite .
fortunately , this is exactly the case in our model , since the soft susy - breaking mass scale @xmath185 is always the smallest mass parameter in the iyit sector ( see eq . ) .
let us now derive an explicit expression for the one - loop khler potential @xmath802 .
to do so , we first need to promote the fermionic masses in eq . to functions of the chiral polonyi superfield @xmath43 , @xmath821^{1/2 } \,.\end{aligned}\ ] ]
these masses then need to be inserted into the general expression for @xmath802 in eq . , @xmath822 \,.\end{aligned}\ ] ] note that this result for @xmath802 is independent of the susy - breaking mass scale @xmath185 . or put differently , the effective khler potential `` does not know anything about supersymmetry breaking '' .
it is , thus , clear that @xmath816 ( or @xmath812 for that purpose ) can approximate the full effective potential only up to @xmath823 .
we shall now study @xmath802 as a function of @xmath824 , the superfield analog of the order parameter @xmath219 , @xmath825 close to the origin , @xmath826 , the effective one - loop khler potential consists of the following terms , @xmath827 where the first term , @xmath828 , denotes a constant shift in the vev of khler potential that does not have any consequences in global supersymmetry and that is negligible in supergravity , @xmath829 , @xmath830 meanwhile , the second term in eq .
represents a shift in the canonical kinetic term of the polonyi field . in the following , we will choose the renormalization scale @xmath720 and the constant @xmath806 as follows , @xmath831 so that this term exactly vanishes ( i.e. , so that the polonyi field @xmath43 remains canonically normalized ) .
the first relevant correction to @xmath801 in the small - field regime , @xmath832 , is then of @xmath833 , @xmath834 this form of the khler potential nicely reflects the fact , at small field values , the radiative corrections to the polonyi potential follow from integrating out `` heavy '' meson fields with masses @xmath191 . in terms of the order parameter superfield @xmath824 , the one - loop khler potential in eq
. reads as follows , @xmath835 where the expression for the effective potential energy scale @xmath244 is the identical to the one that we obtained in sec .
[ subsec : cw ] in the limit of small supersymmetry breaking , see eq . . in this sense
, @xmath836 indeed reproduces the effective one - loop potential in the small - field regime up to corrections of @xmath814 , @xmath837 meanwhile , for large values of the polonyi field , @xmath838 , the one - loop khler potential reduces to @xmath839 which represents a logarithmic renormalization of the kinetic term . here , @xmath840 is a correction to the vev of the khler potential , which is meaningless in global supersymmetry and irrelevant in supergravity , @xmath841 again , we may rewrite the one - loop khler potential in terms of the order parameter superfield @xmath824 , @xmath842 which reproduces @xmath272 and the potential energy scale @xmath245 in the large - field regime , see eq .
, @xmath843 we , therefore , conclude that our results for @xmath836 and @xmath844 in eqs . and indeed suffice to describe the radiative corrections to the polonyi potential , as long as the effect of supersymmetry breaking remains small , @xmath845 . in the limit of large supersymmetry
breaking , @xmath846 , the full information on the radiative corrections is , however , only contained in the effective one - loop potential in eq . .
in the main text , we study inflation in an effective polonyi model that is based on the effective superpotential @xmath847 , a near - canonical khler potential and logarithmic radiative corrections . in particular , we assume that the constant term in the superpotential , @xmath346 , which allows one to tune the cc in the true vacuum to zero , is generated only after inflation . in this appendix
, we are now going to show that ( for all khler potentials of interest ) this assumption is , indeed , inevitable .
that is , in contrast to the main text , we are now going to assume that the constant @xmath66 takes its low - energy value already from the very beginning , @xmath576 .
we then study the prospects of realizing successful inflation in the polonyi model for different choices of the khler potential ( near - canonical and approximately shift - symmetric ) , finding that , for @xmath576 during inflation , polonyi inflation is always bound to fail .
either the polonyi potential turns out to be too steep to support slow - roll inflation or there does not even exist a global minkowski vacuum .
of course , our analysis does not represent a general no - go theorem , as some intricate choice for the khler potential may render inflation in the polonyi model viable , after all @xcite .
, may play the role of the inflaton @xcite .
this scenario requires a particular tuning of the khler potential , in order to stabilize the radial component of the complex polonyi field , @xmath848 , and eventually results in inflationary predictions that are equivalent to those of natural inflation @xcite . ] but for our purposes , as we intend to focus on _ simple _ forms of the khler potential , these results suffice to convince us that successful polonyi inflation is better off , if @xmath0 symmetry is broken at late times .
we first consider the full polonyi superpotential in combination with a near - canonical khler potential , @xmath849 apart from the subdominant higher - dimensional correction ( parametrized in terms of the parameter @xmath48 ) , this is nothing but the original polonyi model @xcite . for now , let us assume that this model does not receive any radiative corrections from yukawa couplings between @xmath43 and some ( heavy ) matter fields . given the above khler potential , the complex scalar contained in the chiral polonyi field , @xmath269 , @xmath850 is not canonically normalized . the polonyi phase , @xmath851 , is always stabilized at @xmath852 , so that we can neglect it in the following .
meanwhile , the properly normalized radial component @xmath78 is given as follows , @xmath853 \ , .
\label{eq : phit}\end{aligned}\ ] ] the vacuum energy density in the global minimum vanishes , once the constant @xmath66 is fine - tuned to @xmath854 for this value of @xmath66 , the real polonyi field @xmath78 takes the following value in the true vacuum , @xmath855 m_{\rm pl } \ , .
\label{eq : phican}\end{aligned}\ ] ] around this vacuum , the polonyi field has a tree - level mass of the order of the gravitino mass , @xmath856 m_{3/2}^2 \,,\end{aligned}\ ] ] where the gravitino mass follows from inserting eqs . and into the general formula in eq .
, @xmath857 \frac{\mu^2}{m_{\rm pl } } \,.\end{aligned}\ ] ] as it turns out , the scalar potential around the true vacuum is too steep to support slow - roll inflation .
this can be seen by inspecting the slow - roll parameters @xmath82 and @xmath45 in the vicinity of @xmath858 , @xmath859^{2 } + \frac{\sqrt{3}+1}{\sqrt{3 } } - \frac{5}{3}\,\epsilon + \mathcal{o}\left(\epsilon^2,\delta\right ) \ , , \label{eq : srparacan}\\ \nonumber \eta\left(\varphi\right ) & = 2 \left[\frac{1}{\delta\left(\varphi\right ) } + \frac{2\sqrt{3}-1}{\sqrt{2 } } - \frac{2\sqrt{3}-3}{2\sqrt{2 } } \,\epsilon\right]^{2 } + \frac{17}{2\sqrt{3 } } - 4 - \frac{2}{3}\left(6\sqrt{3}-5\right)\epsilon + \mathcal{o}\left(\epsilon^2,\delta\right ) \,,\end{aligned}\ ] ] where @xmath860 parametrizes the displacement from the true vacuum in units of the planck scale , @xmath861 the expressions in eq .
imply that we only have a chance of sufficiently suppressing @xmath82 and @xmath45 , if we displace the inflaton field by some amount of @xmath862 from its vev . in the case of such large field displacements ,
the scalar potential is , however , severely steepened by sugra corrections .
in fact , scanning over the entire polonyi field range and evaluating the slow - roll parameters @xmath82 and @xmath45 numerically , we find that nowhere in field space the scalar potential is sufficiently flat for slow - roll inflation , @xmath863 in view of this result , we then also have little hope that radiative corrections may improve the situation .
instead , we expect that radiative correction would , overall , rather increase the magnitude of the slow - roll parameters @xmath82 and @xmath45 even further .
we , thus , arrive at the same conclusion as the analysis in @xcite , namely that inflation in the original polonyi model , with @xmath576 during inflation , does not work .
next , we assume that the khler potential exhibits an approximate shift symmetry along the real axis , @xmath864 the complex scalar contained in the chiral polonyi field , @xmath269 , then decomposes as follows , @xmath865 where @xmath284 and @xmath287 denote the ( not canonically normalized ) inflaton field and its scalar partner ( i.e. , the `` sinflaton '' ) , respectively .
first , we note that , in the limit of an exact shift symmetry , this theory does not admit a minkowski vacuum . for @xmath866 , the only stationary point of the scalar potential
is located at @xmath867 so that the polonyi tadpole term `` eats up '' the constant @xmath66 . at this stationary point , we , therefore , have @xmath17 , which implies unbroken @xmath0 symmetry and , hence , zero gravitino mass .
but more importantly , this point in field space corresponds to a ds state , as the vacuum energy density does not vanish , @xmath868 . on top of that , it is not even stable , as the @xmath284 direction acquires a tachyonic mass , @xmath869 .
we can remedy this situation and distort the scalar potential by slightly breaking the shift symmetry in the @xmath284 direction , i.e. , by allowing for small nonzero values of @xmath48 .
this stabilizes the stationary point in eq . and allows for the possibility of a global minkowski vacuum .
in addition , a slightly broken shift symmetry in the khler potential appears to be more natural , anyway , as the superpotential already breaks the shift symmetry explicitly .
the canonically normalized fields @xmath78 and @xmath870 are then given as follows , @xmath290 for @xmath871 , the model exhibits a global minkowski vacuum at a super - planckian value of the inflaton field , @xmath872 m_{\rm pl } \ , , \quad \left<\sigma\right > = 0 \ , .
\label{eq : phishire}\end{aligned}\ ] ] we note that this solution is only consistent as long as @xmath48 is positive .
moreover , it becomes unphysical in the limit of an exact shift symmetry , i.e. , for @xmath313 .
this is also the reason why , in contrast to sec .
[ subsec : canonical ] , we now refrain from expanding our results in powers of the small parameter @xmath48 . in order to tune the vacuum energy density in the vacuum to zero ,
the constant @xmath66 needs to take the following value , @xmath873 \mu^2 m_{\rm pl } \
, , \label{eq : wshire}\end{aligned}\ ] ] which also diverges for @xmath313 . as we are able to read off from eqs . in , the relation between @xmath874 and @xmath66 in eq
now no longer applies .
thanks to the slightly broken shift symmetry , we now have @xmath875 , which results in a nonzero gravitino mass , @xmath876 \frac{\mu^2}{m_{\rm pl } } \,.\end{aligned}\ ] ] the field @xmath78 then acquires a non - tachyonic mass of the order of the gravitino mass , @xmath877 again , the scalar potential around the true vacuum is too steep for slow - roll inflation . in the vicinity of @xmath858 , the slow - roll parameters @xmath82 and @xmath45 ( as functions of @xmath878 )
are now given as , @xmath879 ^ 2 + \mathcal{o}\left(\epsilon\right ) \
\nonumber \eta\left(\varphi\right ) & = 2\left[\frac{1}{\delta\left(\varphi\right ) } + \sqrt{6 } -\epsilon^{1/2}\right]^2 - \sqrt{6}\left(\sqrt{6}-2\,\epsilon^{1/2}\right ) + \mathcal{o}\left(\epsilon\right ) \,.\end{aligned}\ ] ] a numerical scan of the full expressions for @xmath82 and @xmath45 over the entire field range reveals that both parameters are always of @xmath55 or larger and , hence , always too large for slow - roll inflation .
similarly as in the case of a near - canonical khler potential , we do not expect that radiative corrections could improve this situation .
we , therefore , conclude that , assuming an approximate shift symmetry along the real axis in the khler potential , inflation based on the full polonyi superpotential is unfortunately not an option .
finally , let us examine what happens if we replace the approximate shift symmetry along the real axis by an approximate shift symmetry along the imaginary axis .
as we will see shortly , in this case , we find a stationary point at @xmath880 , similarly as in sec .
[ subsec : sugra ] . for that reason
, our analysis in this section will resemble the discussion in sec .
[ subsec : sugra ] much more closely than our analysis in the previous section .
but , first things first .
now , we supplement the polonyi potential by a khler potential of the following form , @xmath881 the ( not canonically normalized ) inflaton field @xmath284 and the ( not canonically normalized ) sinflaton field @xmath287 then correspond to the imaginary and the real part of the complex scalar @xmath269 , respectively , @xmath882 as in the previous section , the canonically normalized fields @xmath78 and @xmath870 are related to @xmath284 and @xmath287 as follows , @xmath290 now we find a stationary point at a planckian value of the sinflaton field rather than the inflaton field , @xmath883 m_{\rm pl } \,.\end{aligned}\ ] ] the vacuum energy density at this stationary point vanishes , once @xmath66 takes the following value , @xmath884 \mu^2 m_{\rm pl } \,.\end{aligned}\ ] ] we then obtain for the gravitino mass as well as for the inflaton mass around this stationary point , @xmath885 \frac{\mu^2}{m_{\rm pl } } \ , , \quad m_\varphi^2 = \left[4 - 2\sqrt{3 } \left(\frac{2}{1+\epsilon}\right)^{3/2}\right ] m_{3/2}^2 \ , .
\label{eq : m32mphi}\end{aligned}\ ] ] none of the above quantities diverges in the limit of an exact shift symmetry .
in contrast to the previous section , the limit @xmath886 , therefore , does not take us out of the physical regime .
at the same time , independent of whether @xmath48 is exactly zero or just small , @xmath292 , we find that the inflaton direction in field space turns out to be tachyonic , @xmath887 .
the stationary point at @xmath81 , therefore , does not represent the global minimum of the scalar potential .
instead , the scalar potential is either unbounded from below ( for @xmath866 ) or it exhibits a global ads vacuum ( for @xmath871 ) at a large inflaton field value @xmath888 .
it is obvious that the former case does not represent a viable setting for inflation .
for this reason , we shall focus on the latter case in the following and argue that in this case , too , inflation can not be successfully realized . here , our argument will closely follow the discussion in sec .
[ subsec : sugra ] .
first , let us consider the scalar potential for the inflaton field @xmath78 with the sinflaton field @xmath870 being fixed at @xmath889 , @xmath890 \left(\frac{1}{2}m_\varphi^2\varphi^2 + \frac{\lambda_\varphi}{4!}\ , \varphi^4\right ) \ ,
, \quad \lambda_\varphi = \frac{48\,\epsilon^2}{\left(1+\epsilon\right)^3 } \frac{m_{3/2}^2}{m_{\rm pl}^2}\end{aligned}\ ] ] we assume that the shift symmetry the khler potential in eq .
is only slightly broken , @xmath891 , so that @xmath887 ; and we assume that the potential is not unbounded from below , @xmath871 , so that @xmath892 . under these assumptions , the inflaton potential exhibits a global ads vacuum at @xmath893^{1/2 } m_{\rm pl } \ , , \quad a = -\frac{\epsilon}{1+\epsilon}\frac{6}{\lambda_\varphi}\frac{m_\varphi^2}{m_{\rm pl}^2 } \,.\end{aligned}\ ] ] at this field value , the scalar potential takes the following value , @xmath894 \left[\left(1+a^2\right)^{1/2}-1\right ] \frac{4}{1+\epsilon } \ , m_{3/2}^2\ , m_{\rm pl}^2 \,,\end{aligned}\ ] ] which represents the vacuum energy density in the global ads minimum .
similarly as in sec .
[ subsec : sugra ] , we may hope that there might be a chance to lift this ads vacuum by radiative corrections .
but as it turns out , this attempt to stabilize the inflaton potential fails for the same reason as in sec .
[ subsec : sugra ] . within the parameter ranges of interest ,
@xmath311 and @xmath161 , we always find a hierarchy among the critical field value @xmath227 and the position of the ads vacuum , @xmath312 .
the quadratic radiative corrections close to the origin , therefore , fail to stabilize the inflaton potential all the way up to @xmath604 . in consequence of that
, we always end up with a local minimum in between @xmath227 and @xmath300 .
inflation taking place in the logarithmic part of the effective potential then always `` gets stuck '' and the inflaton has no chance of reaching the minkowski vacuum at the origin ( see also our discussion in sec .
[ subsec : sugra ] ) .
this completes our argument that inflation based on the polonyi superpotential @xmath895with the constant @xmath66 being set to its final value @xmath576 already from the very beginning neither works for a near - canonical nor for an approximately shift - symmetric khler potential .
it is this conclusion that leads us to resort to studying inflation in combination with late - time @xmath0 symmetry breaking in the main text . and
indeed , successful polonyi inflation turns out to be feasible , once @xmath66 vanishes during inflation .
a. h. guth , _ the inflationary universe : a possible solution to the horizon and flatness problems _ , phys .
d * 23 * , 347 ( 1981 ) .
a. d. linde , _ a new inflationary universe scenario : a possible solution of the horizon , flatness , homogeneity , isotropy and primordial monopole problems _ , phys .
b * 108 * , 389 ( 1982 ) ; a. albrecht and p. j. steinhardt , _
cosmology for grand unified theories with radiatively induced symmetry breaking _ , phys . rev .
lett . * 48 * , 1220 ( 1982 ) . for reviews ,
see : d. h. lyth and a. riotto , _ particle physics models of inflation and the cosmological density perturbation _ ,
rept . *
314 * , 1 ( 1999 ) , hep - ph/9807278 ; a. d. linde , _ particle physics and inflationary cosmology _ ,
contemp .
concepts phys .
* 5 * , 1 ( 1990 ) , hep - th/0503203 .
l. randall and s. d. thomas , _ solving the cosmological moduli problem with weak scale inflation _
b * 449 * , 229 ( 1995 ) , hep - ph/9407248 .
a. riotto , _ inflation and the nature of supersymmetry breaking _ ,
b * 515 * , 413 ( 1998 ) , hep - ph/9707330 .
k. i. izawa , _ supersymmetry - breaking models of inflation _ , prog . theor .
phys . * 99 * , 157 ( 1998 ) , hep - ph/9708315 .
w. buchmuller , l. covi and d. delepine , _ inflation and supersymmetry breaking _ , phys .
b * 491 * , 183 ( 2000 ) , hep - ph/0006168 .
s. chatrchyan _ et al .
_ [ cms collaboration ] , _ search for new physics in the multijet and missing transverse momentum final state in proton - proton collisions at @xmath896 _ , jhep * 1406 * ( 2014 ) 055 , arxiv:1402.4770 [ hep - ex ] ; g. aad _ et al . _ [ atlas collaboration ] , _
search for squarks and gluinos with the atlas detector in final states with jets and missing transverse momentum using @xmath897 proton proton collision data _ , jhep * 1409 * , 176 ( 2014 ) , arxiv:1405.7875 [ hep - ex ] . g. aad _ et al . _ [ atlas collaboration ] , _ observation of a new particle in the search for the standard model higgs boson with the atlas detector at the lhc _ , phys .
b * 716 * , 1 ( 2012 ) , arxiv:1207.7214 [ hep - ex ] ; s. chatrchyan _ et al . _
[ cms collaboration ] , _ observation of a new boson at a mass of 125 gev with the cms experiment at the lhc _ ,
b * 716 * , 30 ( 2012 ) , arxiv:1207.7235 [ hep - ex ] ; g. aad _ et al .
_ [ atlas and cms collaborations ] , _ combined measurement of the higgs boson mass in @xmath898 collisions at @xmath899 and 8 tev with the atlas and cms experiments _ , phys . rev .
lett . * 114 * , 191803 ( 2015 ) , arxiv:1503.07589 [ hep - ex ] . y. okada , m. yamaguchi and t. yanagida , _ upper bound of the lightest higgs boson mass in the minimal supersymmetric standard model _ , prog .
* 85 * , 1 ( 1991 ) ; _ renormalization group analysis on the higgs mass in the softly broken supersymmetric standard model _ ,
b * 262 * , 54 ( 1991 ) ; j. r. ellis , g. ridolfi and f. zwirner , _ radiative corrections to the masses of supersymmetric higgs bosons _
b * 257 * , 83 ( 1991 ) ; _ on radiative corrections to supersymmetric higgs boson masses and their implications for lep searches _ , phys .
b * 262 * , 477 ( 1991 ) ; h. e. haber and r. hempfling , _ can the mass of the lightest higgs boson of the minimal supersymmetric model be larger than @xmath900 ? _ , phys .
lett . * 66 * , 1815 ( 1991 ) .
g. f. giudice , m. a. luty , h. murayama and r. rattazzi , _ gaugino mass without singlets _ , jhep * 9812 * , 027 ( 1998 ) , hep - ph/9810442
. j. d. wells , _ implications of supersymmetry breaking with a little hierarchy between gauginos and scalars _ , hep - ph/0306127 ; _ pev - scale supersymmetry _ , phys .
d * 71 * , 015013 ( 2005 ) , hep - ph/0411041 . m. ibe , t. moroi and t. t. yanagida , _
possible signals of wino lsp at the large hadron collider _ , phys .
b * 644 * , 355 ( 2007 ) , hep - ph/0610277 ; m. ibe and t. t. yanagida , _ the lightest higgs boson mass in pure gravity mediation model _
b * 709 * , 374 ( 2012 ) , arxiv:1112.2462 [ hep - ph ] ; n. arkani - hamed , a. gupta , d. e. kaplan , n. weiner and t. zorawski , _ simply unnatural supersymmetry _ , arxiv:1212.6971 [ hep - ph ] .
m. ibe , s. matsumoto and t. t. yanagida , _ pure gravity mediation with @xmath901 @xmath902 _ , phys .
d * 85 * , 095011 ( 2012 ) , arxiv:1202.2253 [ hep - ph ] .
l. j. hall and y. nomura , _ spread supersymmetry _ , jhep * 1201 * , 082 ( 2012 ) , arxiv:1111.4519 [ hep - ph ] ; a. arvanitaki , n. craig , s. dimopoulos and g. villadoro , _ mini - split _ , jhep * 1302 * , 126 ( 2013 ) , arxiv:1210.0555 [ hep - ph ] . l. j. hall , y. nomura and s. shirai , _ spread supersymmetry with wino lsp : gluino and dark matter signals _ ,
jhep * 1301 * , 036 ( 2013 ) , arxiv:1210.2395 [ hep - ph ] .
m. ibe , s. matsumoto , s. shirai and t. t. yanagida , _
ams-02 positrons from decaying wino in the pure gravity mediation model _ ,
jhep * 1307 * , 063 ( 2013 ) , arxiv:1305.0084 [ hep - ph ] . j. l. evans , n. nagata and k. a. olive , _
@xmath903 grand unification in pure gravity mediation _ ,
d * 91 * , 055027 ( 2015 ) , arxiv:1502.00034 [ hep - ph ] .
h. pagels and j. r. primack , _ supersymmetry , cosmology and new tev physics _
* 48 * , 223 ( 1982 ) ; s. weinberg , _ cosmological constraints on the scale of supersymmetry breaking _
lett . * 48 * , 1303 ( 1982 ) ; m. y. khlopov and a. d. linde , _ is it easy to save the gravitino ? _ , phys .
b * 138 * , 265 ( 1984 ) ; j. r. ellis , d. v. nanopoulos and s. sarkar , _ the cosmology of decaying gravitinos _
b * 259 * , 175 ( 1985 ) ; m. kawasaki , k. kohri and t. moroi , _ hadronic decay of late - decaying particles and big - bang nucleosynthesis _ , phys .
b * 625 * , 7 ( 2005 ) , astro - ph/0402490 ; _ big - bang nucleosynthesis and hadronic decay of long - lived massive particles _
d * 71 * , 083502 ( 2005 ) , astro - ph/0408426 ; k. jedamzik , _ big bang nucleosynthesis constraints on hadronically and electromagnetically decaying relic neutral particles _ , phys .
d * 74 * , 103509 ( 2006 ) , hep - ph/0604251 ; m. kawasaki , k. kohri , t. moroi and a. yotsuyanagi , _ big - bang nucleosynthesis and gravitino _ , phys .
d * 78 * , 065011 ( 2008 ) , arxiv:0804.3745 [ hep - ph ] .
f. gabbiani , e. gabrielli , a. masiero and l. silvestrini , _ a complete analysis of fcnc and @xmath5 constraints in general susy extensions of the standard model _ , nucl .
b * 477 * , 321 ( 1996 ) , hep - ph/9604387 ; t. moroi and m. nagai , _ probing supersymmetric model with heavy sfermions using leptonic flavor and @xmath5 violations _ , phys .
b * 723 * , 107 ( 2013 ) , arxiv:1303.0668 [ hep - ph ] ; d. mckeen , m. pospelov and a. ritz , _ electric dipole moment signatures of pev - scale superpartners _ , phys .
d * 87 * , 113002 ( 2013 ) , arxiv:1303.1172 [ hep - ph ] ; w. altmannshofer , r. harnik and j. zupan , _ low energy probes of pev scale sfermions _ ,
jhep * 1311 * , 202 ( 2013 ) , arxiv:1308.3653 [ hep - ph ] ; m. baumgart , d. stolarski and t. zorawski , _ split supersymmetry radiates flavor _ , phys .
d * 90 * , 055001 ( 2014 ) , arxiv:1403.6118 [ hep - ph ] .
see , e.g. , the discussion and references in : k. s. babu , k. schmitz and t. t. yanagida , _ pure gravity mediation and spontaneous @xmath904@xmath285@xmath749 breaking from strong dynamics _
b * 905 * , 73 ( 2016 ) , arxiv:1507.04467 [ hep - ph ]
. g. d. coughlan , w. fischler , e. w. kolb , s. raby and g. g. ross , _ cosmological problems for the polonyi potential _
b * 131 * , 59 ( 1983 ) ; j. r. ellis , d. v. nanopoulos and m. quiros , _ on the axion , dilaton , polonyi , gravitino and shadow matter problems in supergravity and superstring models _
b * 174 * , 176 ( 1986 ) ; m. ibe , y. shinbara and t. t. yanagida , _ the polonyi problem and upper bound on inflation scale in supergravity _ , phys .
b * 639 * , 534 ( 2006 ) , hep - ph/0605252 . m. t. grisaru , w. siegel and m. rocek , _ improved methods for supergraphs _ , nucl .
b * 159 * , 429 ( 1979 ) ; n. seiberg , _ naturalness versus supersymmetric nonrenormalization theorems _ , phys .
b * 318 * , 469 ( 1993 ) , hep - ph/9309335 .
m. k. gaillard and v. jain , _ supergravity coupled to chiral matter at one loop _ , phys .
d * 49 * , 1951 ( 1994 ) , hep - th/9308090 ; e. d. stewart , _ flattening the inflaton s potential with quantum corrections _ , phys .
b * 391 * , 34 ( 1997 ) , hep - ph/9606241 ; _ flattening the inflaton s potential with quantum corrections 2 _ , phys .
d * 56 * , 2019 ( 1997 ) , hep - ph/9703232 .
j. r. ellis , d. v. nanopoulos , k. a. olive and k. tamvakis , _
cosmological inflation cries out for supersymmetry _ , phys .
b * 118 * , 335 ( 1982 ) .
l. oraifeartaigh , _ spontaneous symmetry breaking for chiral scalar superfields _ , nucl . phys .
b * 96 * , 331 ( 1975 ) .
s. ray , _ some properties of meta - stable supersymmetry - breaking vacua in wess - zumino models _ , phys .
b * 642 * , 137 ( 2006 ) , hep - th/0607172 ; d. shih , _
spontaneous @xmath0-symmetry breaking in oraifeartaigh models _
, jhep * 0802 * , 091 ( 2008 ) , hep - th/0703196
. s. weinberg , _ anthropic bound on the cosmological constant _ , phys .
lett . * 59 * , 2607 ( 1987 ) ; _ the cosmological constant problem _ , rev .
phys . * 61 * , 1 ( 1989 ) . for a review ,
see : h. p. nilles , _ supersymmetry , supergravity and particle physics _ , phys .
rept . * 110 * , 1 ( 1984 ) .
b. a. ovrut and p. j. steinhardt , _ supersymmetry and inflation : a new approach _ , phys .
b * 133 * , 161 ( 1983 ) .
m. dine and j. kehayias , _
discrete @xmath0 symmetries and low energy supersymmetry _ , phys .
d * 82 * , 055014 ( 2010 ) , arxiv:0909.1615 [ hep - ph ]
. l. susskind , _ the anthropic landscape of string theory _
, in _ `` universe or multiverse ? '' _ , 247266 , cambridge university press ( 2007 ) , edited by b. carr , hep - th/0302219 . s. m. carroll , _ lecture notes on general relativity _ , gr - qc/9712019 .
a. d. linde , _
chaotic inflation _ , phys .
b * 129 * , 177 ( 1983 ) ; _ chaotic inflating universe _ , jetp lett . *
38 * , 176 ( 1983 ) [ pisma zh .
fiz . * 38 * , 149 ( 1983 ) ] .
a. d. linde , _ axions in inflationary cosmology _ , phys .
b * 259 * , 38 ( 1991 ) ; _ hybrid inflation _ , phys .
d * 49 * , 748 ( 1994 ) , astro - ph/9307002 .
t. banks and n. seiberg , _ symmetries and strings in field theory and gravity _ , phys .
d * 83 * , 084019 ( 2011 ) , arxiv:1011.5120 [ hep - th ] . i. antoniadis and r. knoops , _
gauge @xmath0-symmetry and de sitter vacua in supergravity and string theory _ , nucl .
b * 886 * , 43 ( 2014 ) , arxiv:1403.1534 [ hep - th ] ; _ mssm soft terms from supergravity with gauged r - symmetry in de sitter vacuum _ , nucl .
b * 902 * , 69 ( 2016 ) , arxiv:1507.06924 [ hep - ph ] ; i. antoniadis , d. m. ghilencea and r. knoops , _ gauged r - symmetry and its anomalies in 4d @xmath905 supergravity and phenomenological implications _ , jhep * 1502 * , 166 ( 2015 ) , arxiv:1412.4807 [ hep - th ] . l. m. krauss and f. wilczek , _ discrete gauge symmetry in continuum theories _ , phys .
lett . * 62 * , 1221 ( 1989 ) ; l. e. ibanez and g. g. ross , _ discrete gauge symmetry anomalies _ , phys .
b * 260 * , 291 ( 1991 ) ; _ discrete gauge symmetries and the origin of baryon and lepton number conservation in supersymmetric versions of the standard model _ , nucl . phys .
b * 368 * , 3 ( 1992 ) ; t. banks and m. dine , _ note on discrete gauge anomalies _ , phys .
d * 45 * , 1424 ( 1992 ) , hep - th/9109045 ; l. e. ibanez , _ more about discrete gauge anomalies _ , nucl .
b * 398 * , 301 ( 1993 ) , hep - ph/9210211 .
n. sakai and t. yanagida , _ proton decay in a class of supersymmetric grand unified models _ , nucl .
b * 197 * , 533 ( 1982 ) ; s. weinberg , _ supersymmetry at ordinary energies .
1 . masses and conservation laws _ , phys .
d * 26 * , 287 ( 1982 ) .
k. harigaya , m. ibe , k. schmitz and t. t. yanagida , _ peccei - quinn symmetry from a gauged discrete @xmath0 symmetry _ , phys .
d * 88 * , 075022 ( 2013 ) , arxiv:1308.1227 [ hep - ph ] .
k. harigaya , m. ibe , k. schmitz and t. t. yanagida , _ peccei - quinn symmetry from dynamical supersymmetry breaking _ , phys .
d * 92 * , 075003 ( 2015 ) , arxiv:1505.07388 [ hep - ph ] .
a. e. nelson and n. seiberg , _ @xmath906 symmetry breaking versus supersymmetry breaking _ , nucl .
b * 416 * , 46 ( 1994 ) , hep - ph/9309299 .
k. a. intriligator , n. seiberg and d. shih , _ dynamical susy breaking in meta - stable vacua _
, jhep * 0604 * , 021 ( 2006 ) , hep - th/0602239 . j. e. kim and h. p. nilles , _ the mu problem and the strong cp problem _ , phys
. lett .
b * 138 * , 150 ( 1984 ) .
k. inoue , m. kawasaki , m. yamaguchi and t. yanagida , _ vanishing squark and slepton masses in a class of supergravity models _ ,
d * 45 * , 328 ( 1992 ) ; j. a. casas and c. munoz , _ a natural solution to the mu problem _ , phys .
b * 306 * , 288 ( 1993 ) , hep - ph/9302227 .
g. f. giudice and a. masiero , _ a natural solution to the mu problem in supergravity theories _ , phys .
b * 206 * , 480 ( 1988 ) .
a. font , l. e. ibanez , h. p. nilles and f. quevedo , _ on the concept of naturalness in string theories _ ,
b * 213 * , 274 ( 1988 ) ; t. kobayashi , s. raby and r. j. zhang , _ searching for realistic 4d string models with a pati - salam symmetry : orbifold grand unified theories from heterotic string compactification on a @xmath907 orbifold _ , nucl .
b * 704 * , 3 ( 2005 ) , hep - ph/0409098 .
n. g. cabo bizet , t. kobayashi , d. k. mayorga pena , s. l. parameswaran , m. schmitz and i. zavala , _ @xmath906-charge conservation and more in factorizable and non - factorizable orbifolds _ , jhep * 1305 * , 076 ( 2013 ) , arxiv:1301.2322 [ hep - th ] ; n. g. cabo bizet , t. kobayashi , d. k. mayorga pena , s. l. parameswaran , m. schmitz and i.zavala , _ discrete @xmath906-symmetries and anomaly universality in heterotic orbifolds _ , jhep * 1402 * , 098 ( 2014 ) , arxiv:1308.5669 [ hep - th ] ; h. p. nilles , s. ramos - snchez , m. ratz and p. k. s. vaudrevange , _ a note on discrete @xmath906 symmetries in @xmath907-ii orbifolds with wilson lines _ , phys .
b * 726 * , 876 ( 2013 ) , arxiv:1308.3435 [ hep - th ]
. h. m. lee , s. raby , m. ratz , g. g. ross , r. schieren , k. schmidt - hoberg and p. k. s. vaudre - vange , _ a unique @xmath908 symmetry for the mssm _ , phys .
b * 694 * , 491 ( 2011 ) , arxiv:1009.0905 [ hep - ph ] ; _ discrete @xmath906 symmetries for the mssm and its singlet extensions _ , nucl .
b * 850 * , 1 ( 2011 ) , arxiv:1102.3595 [ hep - ph ] ; r. kappl , b. petersen , s. raby , m. ratz , r. schieren and p. k. s. vaudrevange , _ string - derived mssm vacua with residual r symmetries _ , nucl .
b * 847 * , 325 ( 2011 ) , arxiv:1012.4574 [ hep - th ] .
i. affleck , m. dine and n. seiberg , _ dynamical supersymmetry breaking in supersymmetric qcd _ , nucl .
b * 241 * , 493 ( 1984 ) ; _ dynamical supersymmetry breaking in chiral theories _ , phys .
b * 137 * , 187 ( 1984 ) ; _ dynamical supersymmetry breaking in four - dimensions and its phenomenological implications _ , nucl .
b * 256 * , 557 ( 1985 ) .
s. dimopoulos , g. r. dvali and r. rattazzi , _ dynamical inflation and unification scale on quantum moduli spaces _ , phys .
b * 410 * , 119 ( 1997 ) , hep - ph/9705348 .
n. j. craig , _ iss - flation _ , jhep * 0802 * , 059 ( 2008 ) , arxiv:0801.2157 [ hep - th ] .
k. harigaya , m. ibe , k. schmitz and t. t. yanagida , _
chaotic inflation with a fractional power - law potential in strongly coupled gauge theories _ , phys .
b * 720 * , 125 ( 2013 ) , arxiv:1211.6241 [ hep - ph ] ; _ dynamical chaotic inflation in the light of bicep2 _ , phys .
b * 733 * , 283 ( 2014 ) , arxiv:1403.4536 [ hep - ph ] ; _ dynamical fractional chaotic inflation _ , phys .
d * 90 * , 123524 ( 2014 ) , arxiv:1407.3084 [ hep - ph ] .
k. freese , j. a. frieman and a. v. olinto , _ natural inflation with pseudo - nambu - goldstone bosons _ , phys .
lett . * 65 * , 3233 ( 1990 ) ; f. c. adams , j. r. bond , k. freese , j. a. frieman and a. v. olinto , _ natural inflation : particle physics models , power law spectra for large scale structure , and constraints from cobe _ , phys .
d * 47 * , 426 ( 1993 ) , hep - ph/9207245 .
k. harigaya and m. ibe , _ simple realization of inflaton potential on a riemann surface _ , phys
. lett .
b * 738 * , 301 ( 2014 ) , arxiv:1404.3511 [ hep - ph ] ; _ phase locked inflation effectively trans - planckian natural inflation _
, jhep * 1411 * , 147 ( 2014 ) , arxiv:1407.4893 [ hep - ph ] ; r. kallosh , a. linde and b. vercnocke , _ natural inflation in supergravity and beyond _ , phys . rev .
d * 90 * , 041303 ( 2014 ) , arxiv:1404.6244 [ hep - th ] .
m. dine , p. draper and a. monteux , _
monodromy inflation in susy qcd _
, jhep * 1407 * , 146 ( 2014 ) , arxiv:1405.0068 [ hep - th ] ; k. yonekura , _ notes on natural inflation _ ,
jcap * 1410 * , 054 ( 2014 ) , arxiv:1405.0734 [ hep - th ] ; k. harigaya , m. ibe and t. t. yanagida , _ @xmath906-symmetric axion / natural inflation in supergravity via deformed moduli dynamics _ , phys .
b * 739 * , 352 ( 2014 ) , arxiv:1409.0330 [ hep - ph ] . s. d. thomas , _
moduli inflation from dynamical supersymmetry breaking _ ,
b * 351 * , 424 ( 1995 ) , hep - th/9503113 ; w. h. kinney and a. riotto , _ dynamical supersymmetric inflation _ , astropart .
* 10 * , 387 ( 1999 ) , hep - ph/9704388 ; k. hamaguchi , k .- i .
izawa and h. nakajima , _
supersymmetric inflation of dynamical origin _ ,
b * 662 * , 208 ( 2008 ) , arxiv:0801.2204 [ hep - ph ] .
m. dine , w. fischler and d. nemeschansky , _ solution of the entropy crisis of supersymmetric theories _
b * 136 * , 169 ( 1984 ) ; g. d. coughlan , r. holman , p. ramond and g. g. ross , _ supersymmetry and the entropy crisis _ , phys .
b * 140 * , 44 ( 1984 ) .
p. a. r. ade _ et al .
_ [ planck collaboration ] , _ planck 2015 results .
cosmological parameters _ , arxiv:1502.01589 [ astro-ph.co ] ; _ planck 2015 results . xx .
constraints on inflation _ , arxiv:1502.02114 [ astro-ph.co ] . k. i. izawa and t. yanagida , _ dynamical supersymmetry breaking in vector - like gauge theories _ , prog
. theor .
phys . * 95 * , 829 ( 1996 ) , hep - th/9602180 ; k. a. intriligator and s. d. thomas , _ dynamical supersymmetry breaking on quantum moduli spaces _ ,
b * 473 * , 121 ( 1996 ) , hep - th/9603158 .
n. seiberg , _ exact results on the space of vacua of four - dimensional susy gauge theories _ , phys .
d * 49 * , 6857 ( 1994 ) , hep - th/9402044 .
e. j. copeland , a. r. liddle , d. h. lyth , e. d. stewart and d. wands , _ false vacuum inflation with einstein gravity _ ,
d * 49 * , 6410 ( 1994 ) , astro - ph/9401011 ; g. r. dvali , q. shafi and r. k. schaefer , _ large scale structure and supersymmetric inflation without fine tuning _ , phys . rev .
lett . * 73 * , 1886 ( 1994 ) , hep - ph/9406319 . c. panagiotakopoulos , _ blue perturbation spectra from hybrid inflation with canonical supergravity _ ,
d * 55 * , 7335 ( 1997 ) , hep - ph/9702433 ; a. d. linde and a. riotto , _
hybrid inflation in supergravity _ , phys .
d * 56 * , 1841 ( 1997 ) , hep - ph/9703209 .
t. asaka , k. hamaguchi , m. kawasaki and t. yanagida , _ leptogenesis in inflationary universe _ , phys .
d * 61 * , 083512 ( 2000 ) , hep - ph/9907559 .
m. bastero - gil , s. f. king and q. shafi , _ supersymmetric hybrid inflation with non - minimal khler potential _ , phys .
b * 651 * , 345 ( 2007 ) , hep - ph/0604198 .
w. buchmller , v. domcke , k. kamada and k. schmitz , _
hybrid inflation in the complex plane _ , jcap * 1407 * , 054 ( 2014 ) , arxiv:1404.1832 [ hep - ph ] .
k. nakayama , f. takahashi and t. t. yanagida , _ constraint on the gravitino mass in hybrid inflation _
, jcap * 1012 * , 010 ( 2010 ) , arxiv:1007.5152 [ hep - ph ]
. g. veneziano and s. yankielowicz , _ an effective lagrangian for the pure @xmath905 supersymmetric yang - mills theory _ , phys .
b * 113 * , 231 ( 1982 ) ; t. r. taylor , g. veneziano and s. yankielowicz , _ supersymmetric qcd and its massless limit : an effective lagrangian analysis _ , nucl .
b * 218 * , 493 ( 1983 ) .
j. polonyi , _ generalization of the massive scalar multiplet coupling to the supergravity _ , hungary central inst
. res .
( 1978 ) , kfki-1977 - 93 .
m. kawasaki , m. yamaguchi and t. yanagida , _ natural chaotic inflation in supergravity _ , phys .
lett . * 85 * , 3572 ( 2000 ) , hep - ph/0004243 .
m. tegmark and m. j. rees , _ why is the cosmic microwave background fluctuation level @xmath909 _ ?
, astrophys .
j. * 499 * , 526 ( 1998 ) , astro - ph/9709058 . k. s. jeong and f. takahashi , _ a gravitino - rich universe _
, jhep * 1301 * , 173 ( 2013 ) , arxiv:1210.4077 [ hep - ph ]
. m. fukugita and t. yanagida , _ baryogenesis without grand unification _ ,
b * 174 * , 45 ( 1986 ) .
a. pilaftsis , _ cp violation and baryogenesis due to heavy majorana neutrinos _ ,
d * 56 * , 5431 ( 1997 ) , hep - ph/9707235 ; a. pilaftsis and t. e. j. underwood , _ resonant leptogenesis _ , nucl .
b * 692 * , 303 ( 2004 ) , hep - ph/0309342 .
izawa and t. yanagida , _ natural new inflation in broken supergravity _ ,
b * 393 * , 331 ( 1997 ) , hep - ph/9608359 ; k. i. izawa , m. kawasaki and t. yanagida , _ dynamical tuning of the initial condition for new inflation in supergravity _ , phys .
b * 411 * , 249 ( 1997 ) , hep - ph/9707201 ; m. ibe , k .-
izawa , y. shinbara and t. t. yanagida , _ minimal supergravity , inflation , and all that _ , phys .
b * 637 * , 21 ( 2006 ) , hep - ph/0602192 .
k. nakayama and f. takahashi , _ pev - scale supersymmetry from new inflation _
, jcap * 1205 * , 035 ( 2012 ) , arxiv:1203.0323 [ hep - ph ] .
izawa and y. shinbara , _ supersymmetric tuned inflation _ , arxiv:0710.1141 [ hep - ph ] .
a. manohar and h. georgi , _
chiral quarks and the nonrelativistic quark model _ , nucl .
b * 234 * , 189 ( 1984 ) ; h. georgi and l. randall , _ flavor conserving @xmath5 violation in invisible axion models _
b * 276 * , 241 ( 1986 ) ; m. a. luty , _ naive dimensional analysis and supersymmetry _ , phys .
d * 57 * , 1531 ( 1998 ) , hep - ph/9706235 ; a. g. cohen , d. b. kaplan and a. e. nelson , _ counting @xmath910 in strongly coupled supersymmetry _ , phys .
b * 412 * , 301 ( 1997 ) , hep - ph/9706275 . v. domcke , k. schmitz and t. t. yanagida , _ dynamical d - terms in supergravity _ , nucl . phys .
b * 891 * , 230 ( 2015 ) , arxiv:1410.4641 [ hep - th ] .
z. chacko , m. a. luty and e. ponton , _ calculable dynamical supersymmetry breaking on deformed moduli spaces _ ,
jhep * 9812 * , 016 ( 1998 ) , hep - th/9810253 . s. r. coleman and e. j. weinberg , _ radiative corrections as the origin of spontaneous symmetry breaking _ , phys . rev .
d * 7 * , 1888 ( 1973 ) .
w. buchmuller , v. domcke and k. schmitz , _ the chaotic regime of d - term inflation _
, jcap * 1411 * , 006 ( 2014 ) , arxiv:1406.6300 [ hep - ph ] ; w. buchmuller and k. ishiwata , _ grand unification and subcritical hybrid inflation _ , phys .
d * 91 * , 081302 ( 2015 ) , arxiv:1412.3764 [ hep - ph ] .
a. vilenkin , _ cosmic strings and domain walls _ , phys .
rept . * 121 * , 263 ( 1985 ) . c. pallis and q. shafi ,
_ from hybrid to quadratic inflation with high - scale supersymmetry breaking _ , phys .
b * 736 * , 261 ( 2014 ) , arxiv:1405.7645 [ hep - ph ] .
t. watari and t. yanagida , _
@xmath911 supersymmetry in a hybrid inflation model _ ,
b * 499 * , 297 ( 2001 ) , hep - ph/0011389 .
t. watari and t. yanagida , _ gut phase transition and hybrid inflation _ , phys .
b * 589 * , 71 ( 2004 ) , hep - ph/0402125 .
m. dine , f. takahashi and t. t. yanagida , _
discrete @xmath0 symmetries and domain walls _ , jhep * 1007 * , 003 ( 2010 ) , arxiv:1005.3613 [ hep - th ] .
k. harigaya , m. ibe , k. schmitz and t. t. yanagida , _ cosmological selection of multi - tev supersymmetry _ , phys .
b * 749 * , 298 ( 2015 ) , arxiv:1506.00426 [ hep - ph ] . w. buchmuller and d. wyler , _
@xmath5 violation and r invariance in supersymmetric models of strong and electroweak interactions _ ,
b * 121 * , 321 ( 1983 ) ; s. p. martin , _ some simple criteria for gauged r - parity _ , phys . rev .
d * 46 * , 2769 ( 1992 ) , hep - ph/9207218 ; _ implications of supersymmetric models with natural r - parity conservation _ , phys .
d * 54 * , 2340 ( 1996 ) , hep - ph/9602349 .
i. affleck , m. dine and n. seiberg , _ supersymmetry breaking by instantons _ , phys .
lett . * 51 * , 1026 ( 1983 ) . c. a. savoy and a. sil ,
_ can inflation induce supersymmetry breaking in a metastable vacuum ? _ , phys .
b * 660 * , 236 ( 2008 ) , arxiv:0709.1923 [ hep - ph ] ; p. brax , c. a. savoy and a. sil , _ intermediate scale inflation and metastable supersymmetry breaking _ , phys .
b * 671 * , 374 ( 2009 ) , arxiv:0807.1569 [ hep - ph ] ; _ sqcd inflation & susy breaking _
, jhep * 0904 * , 092 ( 2009 ) , arxiv:0902.0972 [ hep - ph ] .
k. harigaya , m. ibe and t. t. yanagida , _ lower bound on the gravitino mass @xmath912 @xmath913 in @xmath906-symmetry breaking new inflation _ , phys .
d * 89 * , 055014 ( 2014 ) , arxiv:1311.1898 [ hep - ph ] ; m. kawasaki , m. yamada , t. t. yanagida and n. yokozaki , _ high - scale susy from an r - invariant new inflation in the landscape _ , arxiv:1512.04259 [ hep - ph ] .
m. kawasaki , n. kitajima , k. nakayama and t. t. yanagida , _ heavy gravitino in hybrid inflation _
, jcap * 1306 * , 037 ( 2013 ) , arxiv:1301.6281 [ hep - ph ] .
t. higaki , k. s. jeong and f. takahashi , _
hybrid inflation in high - scale supersymmetry _
, jhep * 1212 * , 111 ( 2012 ) , arxiv:1211.0994 [ hep - ph ] .
m. dine and d. macintire , _ supersymmetry , naturalness , and dynamical supersymmetry breaking _
d * 46 * , 2594 ( 1992 ) , hep - ph/9205227 ; l. randall and r. sundrum , _ out of this world supersymmetry breaking _ , nucl .
b * 557 * , 79 ( 1999 ) , hep - th/9810155 ; j. a. bagger , t. moroi and e. poppitz , _ anomaly mediation in supergravity theories _ , jhep * 0004 * , 009 ( 2000 ) , hep - th/9911029 ; p. binetruy , m. k. gaillard and b. d. nelson , _ one loop soft supersymmetry breaking terms in superstring effective theories _ , nucl .
b * 604 * , 32 ( 2001 ) , hep - ph/0011081 ; f. deramo , j. thaler and z. thomas , _ the two faces of anomaly mediation _ ,
jhep * 1206 * , 151 ( 2012 ) , arxiv:1202.1280 [ hep - ph ] ; k. harigaya and m. ibe , _ anomaly mediated gaugino mass and path - integral measure _
d * 90 * , 085028 ( 2014 ) , arxiv:1409.5029 [ hep - th ] .
m. kawasaki , f. takahashi and t. t. yanagida , _ gravitino overproduction in inflaton decay _ , phys .
b * 638 * , 8 ( 2006 ) , hep - ph/0603265 .
t. moroi and l. randall , _
wino cold dark matter from anomaly mediated susy breaking _ , nucl .
b * 570 * , 455 ( 2000 ) , hep - ph/9906527 ; w. buchmuller , v. domcke and k. schmitz , _ wimp dark matter from gravitino decays and leptogenesis _ , phys .
b * 713 * , 63 ( 2012 ) , arxiv:1203.0285 [ hep - ph ] ; t. moroi , m. nagai and m. takimoto , _ non - thermal production of wino dark matter via the decay of long - lived particles _
, jhep * 1307 * , 066 ( 2013 ) , arxiv:1303.0948 [ hep - ph ]
. n. arkani - hamed and s. dimopoulos , _ supersymmetric unification without low energy supersymmetry and signatures for fine - tuning at the lhc _ , jhep * 0506 * , 073 ( 2005 ) , hep - th/0405159 ; g. f. giudice and a. romanino , _ split supersymmetry _ , nucl .
b * 699 * , 65 ( 2004 ) , erratum - ibid . * 706 * , 65 ( 2005 ) , _ hep - ph/0406088 _ ; n. arkani - hamed , s. dimopoulos , g. f. giudice and a. romanino , _ aspects of split supersymmetry _ , nucl .
b * 709 * , 3 ( 2005 ) , hep - ph/0409232 .
izawa , t. kugo and t. t. yanagida , _ gravitational supersymmetry breaking _ , prog .
phys . * 125 * , 261 ( 2011 ) , arxiv:1008.4641 [ hep - ph ] .
g. degrassi , s. di vita , j. elias - miro , j. r. espinosa , g. f. giudice , g. isidori and a. strumia , _ higgs mass and vacuum stability in the standard model at nnlo _
, jhep * 1208 * , 098 ( 2012 ) , arxiv:1205.6497 [ hep - ph ] .
t. yanagida , _ horizontal symmetry and masses of neutrinos _ , in _
`` proceedings of the workshop on the unified theory and the baryon number in the universe '' _ , tsukuba , japan , feb . 13 - 14 ( 1979 ) , edited by o. sawada and a. sugamoto , kek report kek-79 - 18 , conf . proc .
c * 7902131 * , 95 ( 1979 ) ; prog .
phys . * 64 * , 1103 ( 1980 ) ; m. gell - mann , p. ramond and r. slansky , _ complex spinors and unified theories _ , in _ supergravity _ , north - holland , amsterdam ( 1979 ) , edited by d. z. freedom and p. van nieuwenhuizen , cern report print-80 - 0576 , conf .
c * 790927 * , 315 ( 1979 ) , arxiv:1306.4669 [ hep - th ] ; see also : p. minkowski , _
@xmath914 at a rate of one out of @xmath915 muon decays ? _ , phys .
b * 67 * , 421 ( 1977 ) .
m. t. grisaru , m. rocek and r. von unge , _ effective khler potentials _ , phys .
b * 383 * , 415 ( 1996 ) , hep - th/9605149 .
i. l. buchbinder , s. kuzenko and z. yarevskaya , _ supersymmetric effective potential : superfield approach _ , nucl .
b * 411 * , 665 ( 1994 ) .
s. v. ketov and t. terada , _ single - superfield helical - phase inflation _
b * 752 * ( 2016 ) 108 , arxiv:1509.00953 [ hep - th ] . | spontaneously broken supersymmetry ( susy ) and a vanishingly small cosmological constant imply that @xmath0 symmetry must be spontaneously broken at low energies . based on this observation , we suppose that , in the sector responsible for low - energy @xmath0 symmetry breaking , a discrete @xmath0 symmetry remains preserved at high energies and only becomes dynamically broken at relatively late times in the cosmological evolution , i.e. , after the dynamical breaking of susy .
prior to @xmath0 symmetry breaking , the universe is then bound to be in a quasi - de sitter phase which offers a dynamical explanation for the occurrence of cosmic inflation .
this scenario yields a new perspective on the interplay between susy breaking and inflation , which neatly fits into the paradigm of high - scale susy : inflation is driven by the susy - breaking vacuum energy density , while the chiral field responsible for susy breaking , the polonyi field , serves as the inflaton . because @xmath0 symmetry is broken only after inflation , slow - roll inflation is not spoiled by otherwise dangerous gravitational corrections in supergravity .
we illustrate our idea by means of a concrete example , in which both susy and @xmath0 symmetry are broken by strong gauge dynamics and in which late - time @xmath0 symmetry breaking is triggered by a small inflaton field value . in this model ,
the scales of inflation and susy breaking are unified ; the inflationary predictions are similar to those of f - term hybrid inflation in supergravity ; reheating proceeds via gravitino decay at temperatures consistent with thermal leptogenesis ; and the sparticle mass spectrum follows from pure gravity mediation .
dark matter consists of thermally produced winos with a mass in the tev range .
april 2016 ipmu 16 - 0051 2.5 cm * polonyi inflation * * dynamical supersymmetry breaking and late - time + symmetry breaking as the origin of cosmic inflation * kai schmitz@xmath1 and tsutomu t. yanagida@xmath2 + _ @xmath3 max planck institute for nuclear physics ( mpik ) , 69117 heidelberg , germany + @xmath2 kavli ipmu ( wpi ) , utias , the university of tokyo , kashiwa , chiba 277 - 8583 , japan _ |
disorder - averaged observables , e.g. , the density of states or conductivity , calculated perturbatively in disorder strength using action in dimensions @xmath3 contain ultravioletly - divergent contributions that require an appropriate renormalisation - group treatment . in this paper
we use the minimal - subtraction renormalisation - group scheme@xcite .
the respective integrals in this scheme are evaluated in lower @xmath35 dimensions ( @xmath36 ) , to ensure their ultraviolet convergence , making analytic continuation to higher dimensions ( @xmath37 ) in the end of the calculation .
also , as we show below , the infrared convergence of momentum integrals is ensured by using matsubara frequencies @xmath66 in place of real frequencies .
the renormalisation procedure consists in calculating perturbative corrections to the disorder - free particle propagator @xmath67 and the coupling @xmath41 in the lagrangian and adding counterterms @xmath42 to the lagrangian in order to cancel divergent ( in powers of @xmath43 ) contributions .
the renormalised quantities @xmath41 and @xmath40 can then be related to the `` bare '' @xmath46 and @xmath30 by comparing the initial lagrangian and the lagrangian expressed in the renormalised variables .
perturbative corrections to the propagator and disorder strength can be obtained straightforwardly using the lagrangian . for convenience
we utilise the conventional disorder - averaging diagrammatic technique@xcite , fig .
[ diagrelem ] .
the impurity line , fig [ diagrelem]a , is a tensor product of two operators @xmath68 in the pseudospin subspaces that correspond to the two ends of the impurity line .
hereinafter scalar expressions for impurity lines are implied to be multiplied by @xmath69 . when evaluating diagrams below we use the following values of momentum integrals in dimension @xmath0 : @xmath70 @xmath71 } & = \co(1 ) , \label{iia } \\
\int_{\bp,\bq}\frac{\bp\bq}{(\omega^2+p^2)(\omega^2+q^2)[\omega^2+(\bp+\bq)^2 ] } \label{iib } & = -\frac{1}{2}\left(\frac{c_{2-\varepsilon}}{\varepsilon}\omega^{-\varepsilon}\right)^2+\co(1 ) , \\ \int_{\bp,\bq}\frac{(\bp\hbsigma)(\bq\hbsigma)}{(\omega^2+p^2)(\omega^2+q^2)[\omega^2+(\bp+\bq)^2 ] } & = -\frac{1}{2}\left(\frac{c_{2-\varepsilon}}{\varepsilon}\omega^{-\varepsilon}\right)^2+\co(1 ) , \label{iic } \\ \int_{\bp,\bq}\frac{(\bp\hbsigma)(\bq\hbsigma)}{(\omega^2+p^2)^2[\omega^2+(\bp+\bq)^2 ] } & = -\left(\frac{c_{2-\varepsilon}}{\varepsilon}\omega^{-\varepsilon}\right)^2 \left(1-\frac{\varepsilon}{2}\right)+\co(1 ) , \label{iid } \\ \int_{\bp,\bq}\frac{(\hbsigma\bp)(\hbsigma\bq)}{(\omega^2+p^2)^2(\omega^2+q^2)[\omega^2+(\bp+\bq)^2 ] } & = \co(1 ) , \label{iie } \\ \int_{\bp,\bq}\frac{\bp\bq}{(\omega^2+p^2)^2(\omega^2+q^2)[\omega^2+(\bp+\bq)^2 ] } & = \co(1 ) , \label{iif}\end{aligned}\ ] ] where the coefficient @xmath72 is defined after eq . , and @xmath73 .
detailed calculations of integrals - are presented , e.g. , in ref . .
integrals and can be reduced to similar integrals using @xmath74 .
integral can be evaluated by introducing two feynman parametrisations@xcite : @xmath71 } = \int_\bq\frac{1}{\omega^2+q^2}\int_0 ^ 1du\int_\bp\frac{1}{[\omega^2+(1-u)p^2+u(\bp+\bq)^2]^2 } \nonumber\\ = \int_\bq\frac{1}{\omega^2+q^2}\int_0 ^ 1du \frac{c_d\gamma\left(2-\frac{d}{2}\right)\gamma\left(\frac{d}{2}\right)}{2\gamma(2 ) } [ \omega^2+u(1-u)q^2]^{\frac{d}{2}-2 } \overset{\varepsilon\ll1}{\approx } \frac{c_{2-\varepsilon}}{2 } \int_0 ^ 1 du\int_\bq\frac{1}{(\omega^2+q^2)\left[\omega^2+q^2u(1-u)\right]^{2-\frac{d}{2 } } } \nonumber\\ \approx\left(\frac{c_{2-\varepsilon}}{2}\right)^2\iint\limits_0 ^ 1du\,dt \int_\bq \frac{t^{1-\frac{d}{2}}}{\left[\omega^2+q^2tu(1-u)+(1-t)q^2\right]^{3-\frac{d}{2 } } } { \approx } \left(\frac{c_{2-\varepsilon}}{2}\right)^2\omega^{2 - 2\varepsilon}\iint\limits_0 ^ 1 dt\,du \frac{t^{1-\frac{d}{2}}}{[tu(1-u)+1-t]^\frac{d}{2 } } = \co(1)\end{aligned}\ ] ] integral can be reduced to the previous integrals by using that @xmath75 . in order to evaluate integrals
- we note that they are invariant under the interchange of @xmath76 and @xmath77 .
they can thus be reduced to the previous integrals by replacing @xmath78= \frac{1}{2}(\bp+\bq)^2-\frac{1}{2}p^2-\frac{1}{2}q^2 $ ] or @xmath79 .
one - loop renormalisation is mimicked by the diagrams in fig .
[ oneloop ] . in
what follows expressions in square brackets is our convention for the values of the respective diagrams .
diagram [ oneloop]a , the leading - order - in-@xmath41 self - energy of the particles , is independent of the incoming and outgoing momenta and can be evaluated as @xmath80=\varkappa\int_\bp(i\omega-\hbsigma\bp)^{-1 } \overset{\eqref{ia}}{=}-i\omega\varkappa\frac{c_{2-\varepsilon}}{\varepsilon}+\co(\varepsilon ) .
\label{omegaoneloop}\end{aligned}\ ] ] diagrams [ oneloop]b-[oneloop]e mimic the corrections to the disorder strength @xmath41 . because we study the long - wavelength behaviour of the system ( at finite frequency ) , these diagrams can be evaluated for zero incoming and outgoing momenta , integrating with respect to the intermediate momenta : @xmath81=[\ref{oneloop}\text{c}]=\varkappa^2\int_\bp(i\omega-\hbsigma\bp)^{-2 } \overset{\eqref{ic}}{=}\varkappa^2\frac{c_{2-\varepsilon}}{\varepsilon}\omega^{-\varepsilon}+\co(1 ) , \label{kappaoneloop1}\end{aligned}\ ] ] @xmath82+[\ref{oneloop}\text{e}]= \varkappa^2\int_\bp \frac{1}{i\omega-\hbsigma\bp}\otimes \frac{1}{i\omega-\hbsigma\bp } + \varkappa^2\int_\bp \frac{1}{i\omega-\hbsigma\bp}\otimes \frac{1}{i\omega+\hbsigma\bp } = -\varkappa^2\int_\bp\frac{2\omega^2}{(\omega^2+p^2)^2}=\co(1 ) .
\label{kappaoneloop2}\end{aligned}\ ] ] to cancel the divergent in @xmath43 corrections to the scale @xmath40 and to the disorder strength @xmath41 , given by eqs .
( [ omegaoneloop ] ) and ( [ kappaoneloop1])-([kappaoneloop2 ] ) respectively , we add to the lagrangian ( [ le ] ) the counterterm - lagrangian @xmath83 eqs . and describe the one - loop renormalisation of the system parameters .
we note , that there is no one - loop renormalisation of the particle velocity ( the coefficient before @xmath84 in the lagrangian ) . in order to take into account these one - loop corrections
when performing the two - loop renormalisation we introduce additional elements to the diagrammatic technique , fig .
[ counterelem ] .
the two - loop contribution to the particle self - energy is given by the diagrams in figs .
[ selfenergy ] and [ selfenergycounter ] .
selfenergy]a depends on the external momentum @xmath85 , while the other diagrams are momentum - independent . the momentum dependency of the two - loop self - energy leads to the renormalisation of the particle velocity , in addition to the energy scale @xmath40 . in order to obtain the two - loop corrections
to @xmath40 it is sufficient to evaluate the diagrams in figs . [ selfenergy ] and [ selfenergycounter ] for zero external momenta .
diagrams [ selfenergy]a and [ selfenergy]b for @xmath86 are given by ( in units @xmath87 ) @xmath88= -\int_{\bp,\bq}\frac{(i\omega+\hbsigma\bp)\left[i\omega+\hbsigma(\bp+\bq)\right](i\omega+\hbsigma\bq ) } { ( \omega^2+p^2)[\omega^2+(\bp+\bq)^2](\omega^2+q^2 ) } \nonumber\\ \overset{\eqref{iia}}{=}-i\omega\int_{\bp,\bq}\frac{(\bp\hbsigma)(\bq\hbsigma)+(\bp\hbsigma)(\bp\hbsigma+\bq\hbsigma ) + ( \bp\hbsigma+\bq\hbsigma)(\bq\hbsigma ) } { ( \omega^2+p^2)(\omega^2+q^2)[\omega^2+(\bp+\bq)^2 ] } + \co(1 ) \overset{\eqref{iic}}{= } -\frac{i\omega}{2}\left(\frac{c_{2-\varepsilon}}{\varepsilon}\omega^{-\varepsilon}\right)^2+\co(1 ) \\ [ \ref{selfenergy}\text{b}]\overset{\eqref{ia},\eqref{ic } } = -i\omega\left(\frac{c_{2-\varepsilon}}{\varepsilon}\omega^{-\varepsilon}\right)^2(1-\varepsilon)+\co(1)\end{aligned}\ ] ] diagrams of the second order in the disorder strength @xmath41 that contain the one - loop counterterms , fig .
[ counterelem ] , are shown in fig .
[ selfenergycounter ] .
.,scaledwidth=35.0% ] @xmath89=\varkappa\:\delta^{(1)}\varkappa\int_\bp(i\omega-\hbsigma\bp)^{-2 } \overset{\eqref{ic},\eqref{kappacounteroneloop}}{= } i\omega\varkappa^2\left(\frac{c_{2-\varepsilon}}{\varepsilon}\omega^{-\varepsilon}\right)^2(1-\varepsilon)+\co(1 ) \\ [ \ref{selfenergycounter}\text{b}]=2i\omega\varkappa^2 \left(\frac{c_{2-\varepsilon}}{\varepsilon}\omega^{-\varepsilon}\right)^2+\co(1).\end{aligned}\ ] ] the velocity renormalisation is determined by the linear - in-@xmath85 contribution to the self - energy . in order to obtain it ,
we expand to the linear order the @xmath85-dependent propagator in diagram [ selfenergy]a : @xmath90 the diagram , corresponding to the linear - in-@xmath85 contribution , is shown in fig .
[ velocityrenorm ] . in units
@xmath87 @xmath91=-\int_{\bp,\bq } \frac{1}{i\omega-\hbsigma\bp}\frac{1}{i\omega-\hbsigma(\bp+\bq)}\hbsigma\bk \frac{1}{i\omega-\hbsigma(\bp+\bq)}\frac{1}{i\omega-\hbsigma\bq } \nonumber\\ = -\int_{\bp,\bq}\frac{(i\omega+\hbsigma\bp)[i\omega+\hbsigma(\bp+\bq)](\hbsigma\bk)[i\omega+\hbsigma(\bp+\bq)](i\omega+\hbsigma\bq ) } { ( \omega^2+p^2)[\omega^2+(\bp+\bq)^2]^2(\omega^2+q^2 ) } \label{vrenormc1}\end{aligned}\ ] ] using that @xmath92 , eq .
[ vrenormc1 ] gives @xmath91=\int_{\bp,\bq}\frac{(i\omega+\hbsigma\bp)(\hbsigma\bk)(i\omega+\hbsigma\bq ) } { ( \omega^2+p^2)[\omega^2+(\bp+\bq)^2](\omega^2+q^2)}+i_1 \label{vrenormc15}\end{aligned}\ ] ] where @xmath93(i\omega+\hbsigma\bq ) } { ( \omega^2+p^2)[\omega^2+(\bp+\bq)^2]^2(\omega^2+q^2 ) } ( \bp+\bq)\bk \label{vrenormc2}\end{aligned}\ ] ] let us demonstrate that @xmath94 , eq .
( [ vrenormc2 ] ) , does not contain @xmath43 or @xmath95 singularities . replacing in the numerator @xmath96(i\omega+\hbsigma\bq )
\rightarrow p^2(\hbsigma\bq)+q^2(\hbsigma\bp)-2\omega^2\hbsigma(\bp+\bq)$ ] ( as the other contributions vanish ) and using gives @xmath97 ^ 2(\omega^2+q^2 ) } ( \bp+\bq)\bk+\co(1 ) \label{vrenormc3}\end{aligned}\ ] ] using that the integral in is invariant with respect to interchanging @xmath76 and @xmath77 and to changing signs of momentum components ( @xmath98 ) , we replace @xmath99\equiv\sum_{\alpha,\beta}\sigma_\alpha q_\alpha ( p+q)_\beta k_\beta\rightarrow\sum_\alpha \sigma_\alpha q_\alpha ( p+q)_\alpha k_\alpha \rightarrow\frac{1}{d}\bq(\bp+\bq)\cdot(\hbsigma\bk)$ ] and arrive at @xmath100 ^ 2(\omega^2+q^2)}\hbsigma\bk \nonumber\\ = -\frac{4}{d}\int_{\bp,\bq}\frac{\bq(\bp+\bq)}{(\omega^2+q^2)[\omega^2+(\bp+\bq)^2 ] } + \frac{4}{d}\omega^2\int_{\bp,\bq}\frac{\bq(\bq+\bp)}{(\omega^2+p^2)[\omega^2+(\bp+\bq)^2]^2(\omega^2+q^2 ) } \overset{\eqref{iif}}{=}\co(1 ) .
\label{vrenormc4}\end{aligned}\ ] ] eqs . , , and give @xmath91=\int_{\bp,\bq}\frac{(\hbsigma\bp)(\hbsigma\bk)(\hbsigma\bq ) } { ( \omega^2+p^2)(\omega^2+q^2)[\omega^2+(\bp+\bq)^2]}+\co(1 ) \nonumber\\ = \left(\frac{2}{d}-1\right)\int_{\bp,\bq}\frac{\bp\bq}{(\omega^2+p^2)(\omega^2+q^2)[\omega^2+(\bp+\bq)^2]}\hbsigma\bk+\co(1 ) \overset{\eqref{iib}}{=}-\frac{1}{4\varepsilon}(c_{2-\varepsilon}\omega^{-\varepsilon})^2\hbsigma\bk+\co(1).\end{aligned}\ ] ]
the two - loop renormalisation of the disorder strength @xmath41 corresponds to the diagrams in figs .
[ vertexfive][twoloopcounter ] ( we show only topologically inequivalent diagrams ) . in what immediately follows
we present a detailed calculation of each of these diagrams .
for simplicity all expressions for the diagrams are given in units @xmath101 .
@xmath102=\left[\int_\bp\frac{1}{(i\omega-\hbsigma\bp)^2}\right]^2 \overset{\eqref{ia}}{=}\left(c_{2-\varepsilon}\omega^{-\varepsilon}\right ) \left(\frac{1}{\varepsilon^2 } -\frac{2}{\varepsilon}\right)+\co(1)\end{aligned}\ ] ] @xmath103= \int_{\bp,\bq}\frac{1}{i\omega-\hbsigma\bp}\frac{1}{[i\omega-\hbsigma(\bp+\bq)]^2}\frac{1}{i\omega-\hbsigma\bq } = \int_{\bp,\bq}\frac{(i\omega+\hbsigma\bp)[i\omega+\hbsigma(\bp+\bq)]^2(i\omega+\hbsigma\bq ) } { ( \omega^2+p^2)[\omega^2+(\bp+\bq)^2]^2(\omega^2+q^2 ) } \nonumber\\ = \int_{\bp,\bq}\frac{(\hbsigma\bp)(\hbsigma\bq)(\bp+\bq)^2}{(\omega^2+p^2)[\omega^2+(\bp+\bq)^2]^2(\omega^2+q^2 ) } -\omega^2\int_{\bp,\bq } \frac{2(\hbsigma\bp)(\hbsigma\bp+\hbsigma\bq)+2(\hbsigma\bp+\hbsigma\bq)(\hbsigma\bq)+(\hbsigma\bp)(\hbsigma\bq ) } { ( \omega^2+p^2)[\omega^2+(\bp+\bq)^2]^2(\omega^2+q^2)}+\co(1 ) \nonumber\\ \overset{\eqref{iia}}{= } \int_{\bp,\bq}\frac{(\hbsigma\bp)(\hbsigma\bq)}{(\omega^2+p^2)[\omega^2+(\bp+\bq)^2](\omega^2+q^2 ) } -2\omega^2\int_{\bp,\bq}\frac{(\hbsigma\bp)(\hbsigma\bq)}{(\omega^2+p^2)[\omega^2+(\bp+\bq)^2]^2(\omega^2+q^2 ) } + \co(1 ) \nonumber\\ = \int_{\bp,\bq}\frac{(\hbsigma\bp)(\hbsigma\bq)}{(\omega^2+p^2)[\omega^2+(\bp+\bq)^2](\omega^2+q^2 ) } -\omega^2\int_{\bp,\bq}\frac{(\bp+\bq)^2-p^2-q^2 } { ( \omega^2+p^2)[\omega^2+(\bp+\bq)^2]^2(\omega^2+q^2)}+\co(1 ) \nonumber\\ \overset{\eqref{iic},\eqref{iib},\eqref{ib},\eqref{ic}}{= } -\frac{1}{2}\varkappa^3 ( c_{2-\varepsilon}\omega^{-\varepsilon})^2 \left(\frac{1}{\varepsilon^2}-\frac{2}{\varepsilon}\right)+\co(1),\end{aligned}\ ] ] @xmath104= \int_{\bp,\bq}\frac{1}{(i\omega-\hbsigma\bp)^2}\frac{1}{i\omega-\hbsigma(\bp+\bq)}\frac{1}{i\omega-\hbsigma\bq } \nonumber\\ \overset{\bp+\bq\rightarrow\bp,\bq\rightarrow-\bq}{= } \int_{\bp,\bq}\frac{[i\omega+\hbsigma(\bp+\bq)]^2}{[\omega^2+(\bp+\bq)^2 ] } \frac{i\omega+\hbsigma\bp}{\omega^2+p^2}\frac{i\omega-\hbsigma\bq}{\omega^2+q^2 } \nonumber\\ = -\int_{\bp,\bq}\frac{(\bp+\bq)^2(\hbsigma\bp)(\hbsigma\bq ) } { ( \omega^2+p^2)[\omega^2+(\bp+\bq)^2]^2(\omega^2+q^2 ) } \nonumber\\ -\omega^2
\int_{\bp,\bq}\frac{2(\hbsigma\bp+\hbsigma\bq)(\hbsigma\bp)+2(\hbsigma\bp+\hbsigma\bq)(\hbsigma\bq)-(\hbsigma\bp)(\hbsigma\bq ) } { ( \omega^2+p^2)[\omega^2+(\bp+\bq)^2]^2(\omega^2+q^2)}+\co(1 ) \nonumber\\ = -[\ref{vertexfive}b]+\co(1)= \frac{1}{2}\varkappa^3 ( c_{2-\varepsilon}\omega^{-\varepsilon})^2 \left(\frac{1}{\varepsilon^2}-\frac{2}{\varepsilon}\right)+\co(1)\end{aligned}\ ] ] @xmath105=\int_\bp\left(\frac{1}{i\omega-\hbsigma\bp}\right)^3 \int_\bq\frac{1}{i\omega-\hbsigma\bp}\overset{\eqref{ia},\eqref{id}}{= } = -\frac{1}{2\varepsilon}\,(c_{2-\varepsilon}\omega^{-\varepsilon})^2+\co(1)\end{aligned}\ ] ] @xmath106 = \int_\bp\frac{1}{(i\omega-\hbsigma\bp)^2}\otimes\int_\bq\frac{1}{(i\omega-\hbsigma\bq)^2 } \overset{\eqref{ic}}{=}\left(c_{2-\varepsilon}\omega^{-\varepsilon}\right)^2 \left(\frac{1}{\varepsilon^2}-\frac{2}{\varepsilon}\right)+\co(1)\end{aligned}\ ] ] @xmath107= \int_{\bp,\bq}\frac{1}{i\omega-\hbsigma\bp}\frac{1}{i\omega-\hbsigma(\bp+\bq)}\frac{1}{i\omega-\hbsigma\bq } \otimes\left(\frac{1}{i\omega-\hbsigma\bq}+\frac{1}{i\omega+\hbsigma\bq}\right ) \nonumber\\ = 2i\omega\int \frac{(i\omega+\hbsigma\bp)[i\omega+\hbsigma(\bp+\bq)](i\omega+\hbsigma\bq ) } { ( \omega^2+p^2)[\omega^2+(\bp+\bq)^2](\omega^2+q^2)^2 } = -2\omega^2\int_{\bp,\bq } \frac{(\hbsigma\bp)(\hbsigma\bq)+(\bp+\bq)^2}{(\omega^2+p^2)[\omega^2+(\bp+\bq)^2](\omega^2+q^2)^2 } + \co(1 ) \nonumber\\ = -\omega^2\int_{\bp,\bq } \frac{3(\bp+\bq)^2-p^2-q^2}{(\omega^2+p^2)[\omega^2+(\bp+\bq)^2](\omega^2+q^2)^2 } \nonumber\\ = -3\omega^2\int_{\bp,\bq}\frac{1}{(\omega^2+p^2)(\omega^2+q^2)^2 } + \omega^2\int_{\bp,\bq}\frac{1}{[\omega^2+(\bq+\bp)^2](\omega^2+q^2)^2}+\co(1 ) \nonumber\\ \overset{\eqref{ia},\eqref{ib}}{=}-\frac{1}{\varepsilon } \left(c_{2-\varepsilon}\omega^{-\varepsilon}\right)^2+\co(1)\end{aligned}\ ] ] @xmath108= \int_{\bp,\bq}\frac{1}{i\omega-\hbsigma\bp}\frac{1}{i\omega-\hbsigma(\bp+\bq)}\frac{1}{i\omega-\hbsigma\bp } \otimes\left(\frac{1}{i\omega-\hbsigma\bq}+\frac{1}{i\omega+\hbsigma\bq}\right ) \nonumber\\ = 2i\omega\int \frac{(i\omega+\hbsigma\bp)[i\omega+\hbsigma(\bp+\bq)](i\omega+\hbsigma\bp ) } { ( \omega^2+p^2)^2[\omega^2+(\bp+\bq)^2](\omega^2+q^2 ) } = -2\omega^2\int_{\bp,\bq } \frac{(\bp^2+\bq^2)-p^2-q^2}{(\omega^2+p^2)^2[\omega^2+(\bp+\bq)^2](\omega^2+q^2 ) } + \co(1 ) \nonumber\\ \overset{\eqref{iia}}{= } -2\omega^2\int_{\bp,\bq}\frac{1}{(\omega^2+p^2)^2(\omega^2+q^2 ) } + 2\omega^2\int_{\bp,\bq}\frac{1}{(\omega^2+p^2)^2[\omega^2+(\bp+\bq)^2]}+\co(1 ) = \co(1)\end{aligned}\ ] ] @xmath109= \int_{\bp,\bq}\frac{1}{i\omega-\hbsigma\bp}\frac{1}{i\omega-\hbsigma\bq}\frac{1}{i\omega-\hbsigma\bp } \otimes \left(\frac{1}{i\omega-\hbsigma\bp}+\frac{1}{i\omega+\hbsigma\bp}\right ) \nonumber\\ \overset{\eqref{ia}}{= } \left[-i\omega\frac{c_{2-\varepsilon}}{\varepsilon}\omega^{-\varepsilon}+\co(\varepsilon)\right]\cdot 2i\omega\int_\bp\frac{(2i\omega+\hbsigma\bp)^2}{(\omega^2+p^2)^3 } = 2\omega^2\left[\frac{c_{2-\varepsilon}}{\varepsilon}+\co(\varepsilon)\right]\cdot \left[\int_\bp\frac{1}{(\omega^2+p^2)^2}-2\omega^2\int_\bp\frac{1}{(\omega^2+p^2)^3}\right ] \nonumber\\ \overset{\eqref{ib},\eqref{ib}}{= } \frac{1}{\varepsilon}\cdot\co(\varepsilon)=\co(1)\end{aligned}\ ] ] the sum of the two diagrams in fig . [ vertexn ] is given by @xmath110=\int_{\bp,\bq}\frac{1}{i\omega-\hbsigma\bp}\frac{1}{i\omega-\hbsigma\bq } \otimes\frac{1}{i\omega+\hbsigma\bp}\frac{1}{i\omega+\hbsigma\bq } + \int_{\bp,\bq}\frac{1}{i\omega-\hbsigma\bp}\frac{1}{i\omega-\hbsigma\bq } \otimes\frac{1}{i\omega-\hbsigma\bq}\frac{1}{i\omega-\hbsigma\bp } \nonumber\\ = \int_{\bp,\bq}\frac{(i\omega+\hbsigma\bp)(i\omega+\hbsigma\bq)\otimes \left[-2\omega^2+(\hbsigma\bp)(\hbsigma\bq)+(\hbsigma\bq)(\hbsigma\bp)\right ] } { ( \omega^2+p^2)^2(\omega^2+q^2)^2 } = 2\int_{\bp,\bq}\frac{[-\omega^2+(\hbsigma\bp)(\hbsigma\bq)]\otimes ( -\omega^2+\bp\bq ) } { ( \omega^2+p^2)^2(\omega^2+q^2)^2 } \nonumber\\ = 2\int_{\bp,\bq}\frac{(\hbsigma\bp)(\hbsigma\bq)\otimes\bp\bq}{(\omega^2+p^2)^2(\omega^2+q^2)^2}+\co(1 ) \label{vertnintegralinterm}\end{aligned}\ ] ] using that in the tensor product @xmath111 only the terms with @xmath112 contribute to the integral ( [ vertnintegralinterm ] ) we replace @xmath113 : @xmath110=\frac{2}{d}\left[\int_\bp\frac{p^2}{(\omega^2+p^2)^2}\right]^2 = \frac{2}{d}\left[\int_\bp\frac{1}{\omega^2+p^2}-\omega^2\int_\bp\frac{1}{(\omega^2+p^2)^2}\right]^2 \nonumber\\ = \frac{2}{2-\varepsilon } \left[\frac{1}{\varepsilon}c_{2-\varepsilon}\omega^{-\varepsilon } -\frac{1}{2}c_{2-\varepsilon}\omega^{-\varepsilon}+\co(\varepsilon)\right]^2 = \left(\frac{1}{\varepsilon^2}-\frac{1}{2\varepsilon}\right ) \left(c_{2-\varepsilon}\omega^{-\varepsilon}\right)^2+\co(1).\end{aligned}\ ] ] @xmath114=\int_{\bp,\bq}\frac{1}{i\omega-\hbsigma\bp}\frac{1}{i\omega-\hbsigma\bq}\otimes \frac{1}{i\omega+\hbsigma\bp}\frac{1}{i\omega+\hbsigma(\bp-\bq)}+ \int_{\bp,\bq}\frac{1}{i\omega-\hbsigma\bp}\frac{1}{i\omega-\hbsigma\bq}\otimes \frac{1}{i\omega-\hbsigma(\bp-\bq)}\frac{1}{i\omega-\hbsigma\bp } \nonumber\\ = \int_{\bp,\bq } \frac{(i\omega+\hbsigma\bp)(i\omega+\hbsigma\bq)\otimes[-2\omega^2+(\hbsigma\bp)(\hbsigma\bp-\hbsigma\bq)+(\hbsigma\bp-\hbsigma\bq)(\hbsigma\bp ) ] } { ( \omega^2+p^2)^2(\omega^2+q^2)[\omega^2+(\bp-\bq)^2 ] } \nonumber\\ = 2\int_{\bp,\bq } \frac{(i\omega+\hbsigma\bp)(i\omega+\hbsigma\bq)\otimes[-\omega^2+\bp(\bp-\bq ) ] } { ( \omega^2+p^2)^2(\omega^2+q^2)[\omega^2+(\bp-\bq)^2 ] } = 2\int_{\bp,\bq}\frac{[-\omega^2+(\hbsigma\bp)(\hbsigma\bq)][-\omega^2+\bp(\bp-\bq ) ] } { ( \omega^2+p^2)^2(\omega^2+q^2)[\omega^2+(\bp-\bq)^2 ] } \nonumber\\ = 2\int_{\bp,\bq}\frac{(\hbsigma\bp)(\hbsigma\bq)\otimes\bp(\bp-\bq)}{(\omega^2+p^2)^2(\omega^2+q^2)[\omega^2+(\bp-\bq)^2 ] } -2\omega^2\int_{\bp,\bq}\frac{(\hbsigma\bp)(\hbsigma\bq)\otimes\holone}{(\omega^2+p^2)^2(\omega^2+q^2)[\omega^2+(\bp-\bq)^2 ] } \nonumber\\ -2\omega^2\int_{\bp,\bq}\frac{\holone\otimes\bp(\bp-\bq)}{(\omega^2+p^2)^2(\omega^2+q^2)[\omega^2+(\bp-\bq)^2 ] } + \co(1 ) \overset{\eqref{iie},\eqref{iif}}{= } \int_{\bp,\bq}\frac{(\hbsigma\bp)(\hbsigma\bq)\otimes[p^2+(\bp-\bq)^2-q^2 ] } { ( \omega^2+p^2)^2(\omega^2+q^2)[\omega^2+(\bp-\bq)^2]}+\co(1 ) \nonumber\\ = \int_{\bp,\bq}\frac{(\hbsigma\bp)(\hbsigma\bq)\otimes\holone}{(\omega^2+p^2)(\omega^2+q^2)[\omega^2+(\bp-\bq)^2 ] } -\int_{\bp,\bq}\frac{(\hbsigma\bp)(\hbsigma\bq)\otimes\holone}{(\omega^2+p^2)^2[\omega^2+(\bp-\bq)^2 ] } \nonumber\\ -\omega^2\int_{\bp,\bq } \frac{(\hbsigma\bp)(\hbsigma\bq)}{(\omega^2+p^2)^2(\omega^2+q^2)[\omega^2+(\bp-\bq)^2]}+\co(1 ) \nonumber\\ \overset{\eqref{iic},\eqref{iid}}{= } \frac{1}{2}\left(\frac{c_{2-\varepsilon}}{\varepsilon}\omega^{-\varepsilon}\right)^2 -\left(\frac{c_{2-\varepsilon}}{\varepsilon}\omega^{-\varepsilon}\right)^2 \left(1-\frac{\varepsilon}{2}\right)+\co(1 ) = -\frac{1}{2}(c_{2-\varepsilon}\omega^{-\varepsilon})^2 \left(\frac{1}{\varepsilon^2}+\frac{1}{\varepsilon}\right)+\co(1)\end{aligned}\ ] ] diagrams for the two - loop renormalisation that contain the one - loop counterterms , fig .
[ counterelem ] , are shown in fig .
[ twoloopcounter ] . diagram [ twoloopcounter]c can be evaluated straightforwardly using eq .
( [ i d ] ) , and the other diagrams similarly to the corresponding one - loop diagrams in fig . [ oneloop ] : @xmath115=[\ref{twoloopcounter}b]=-2\varkappa^3 ( c_{2-\varepsilon}\omega^{-\varepsilon})^2 \left(\frac{1}{\varepsilon^2}-\frac{1}{\varepsilon}\right)+\co(1),\end{aligned}\ ] ] @xmath116=\frac{1}{2}\varkappa^3(c_{2-\varepsilon}\omega^{-\varepsilon})^2/\varepsilon+\co(1),\end{aligned}\ ] ] @xmath117=2\varkappa^3(c_{2-\varepsilon}\omega^{-\varepsilon})^2/\varepsilon+\co(1),\end{aligned}\ ] ] @xmath118=\co(1).\end{aligned}\ ] ]
in the previous sections we calculated in two loops perturbative corrections to observable couplings @xmath40 , @xmath41 , and the quasiparticle velocity ( coefficient before @xmath119 ) that come from the action and contain divergent contributions @xmath120 and @xmath95 . in order to cancel these divergencies , following the minimal subtraction scheme@xcite
, we add to the action the counterterm lagrangian @xmath121\psi\ : d\br + \frac{1}{2}\delta\varkappa\int\left(\psi^\dagger\psi\right)^2d\br,\end{aligned}\ ] ] where @xmath122+[\ref{selfenergy}a]_{\bk=0}+[\ref{selfenergy}b ] + [ \ref{selfenergycounter}a]+[\ref{selfenergycounter}b ] = i\omega\left[-\frac{1}{\varepsilon}\varkappa c_{2-\varepsilon}\omega^{-\varepsilon } + \frac{3}{2}\frac{1}{\varepsilon^2 } \left(\varkappa c_{2-\varepsilon}\omega^{-\varepsilon}\right)^2 \right ] , \label{omegadivergent}\end{aligned}\ ] ] @xmath123 = \frac{1}{4\varepsilon}(\varkappa c_{2-\varepsilon}\omega^{-\varepsilon})^2\hbsigma\bk\end{aligned}\ ] ] @xmath124 - 2\times[\ref{vertexfive}a]-2\times[\ref{vertexfive}b ] -4\times[\ref{vertexfive}c]-4\times[\ref{vertexfive}d]-[\ref{vertexfive}e]-4\times[\ref{vertexk } ] -2\times[\ref{vertexl}]-2\times[\ref{vertexm}]-[\ref{vertexn}]-2\times[\ref{vertexo } ] \nonumber\\ -2\times[\ref{twoloopcounter}a]-2\times[\ref{twoloopcounter}b ] -4\times[\ref{twoloopcounter}c]-2\times[\ref{twoloopcounter}d]-2\times[\ref{twoloopcounter}e ] = \varkappa\left[-\frac{2}{\varepsilon}\varkappa c_{2-\varepsilon}\omega^{-\varepsilon } -\left(-\frac{4}{\varepsilon^2}+\frac{1}{2\varepsilon}\right)\left(\varkappa c_{2-\varepsilon}\omega^{-\varepsilon}\right)^2\right ] \label{kappadivergent}\end{aligned}\ ] ] where we kept only the terms divergent at @xmath53 and took into account the numbers of topologically equivalent diagrams . introducing dimensionless disorder strength , eq .
( [ dimensdisorder ] ) , the full lagrangian ( [ action1 ] ) of the system can be rewritten in terms of the observable variables @xmath41 and @xmath40 as @xmath125\psi \,d\br -\frac{\gamma\omega^\varepsilon}{4c_{2-\varepsilon } } \left(1-\frac{\gamma}{\varepsilon}-\frac{\gamma^2}{8\varepsilon } + \frac{\gamma^2}{\varepsilon^2}\right)\int(\psi^\dagger\psi)^2d\br .
\label{lagrangianfinal}\end{aligned}\ ] ] the renormalised observables @xmath126 , @xmath47 , and @xmath40 can be related to the `` bare '' @xmath127 , @xmath46 and @xmath30 by comparing the lagrangian with the `` bare '' one , eq .
( [ action0 ] ) : @xmath128 where @xmath129 describes the rescaling of the particle wavefunctons : @xmath130 . the energy scale @xmath40 sets the characteristic momentum of long - wavelength behaviour of the system .
in order to obtain the rg flow of the renormalised disorder strength @xmath47 as a function of @xmath40 for a given bare disorder strength @xmath46 , we require @xmath131 which , together with eq .
( [ kappaogamma ] ) , gives the rg equation . | disordered non - interacting systems in sufficiently high dimensions have been predicted to display a non - anderson disorder - driven transition that manifests itself in the critical behaviour of the density of states and other physical observables .
recently the critical properties of this transition have been extensively studied for the specific case of weyl semimetals by means of numerical and renormalisation - group approaches . despite this , the values of the critical exponents at such a transition in a weyl semimetal are currently under debate .
we present an independent calculation of the critical exponents using a two - loop renormalisation - group approach for weyl fermions in @xmath0 dimensions and resolve controversies currently existing in the literature .
= 1 it was proposed@xcite 30 years ago that three - dimensional ( 3d ) disordered systems with weyl and dirac quasiparticle dispersion can display an unconventional disorder - driven transition that lies in a non - anderson universality class .
in particular , in contrast with the anderson localisation transition , the density of states at this transition has been suggested@xcite to display a critical behaviour , with the scaling function proposed in ref . .
recently we have demonstrated@xcite that such transitions occur near nodes and band edges in _ all _ materials in sufficiently _ high dimensions _
@xmath1 and are not unique to dirac ( weyl ) systems . in systems that allow for localisation
the transition manifests itself also in the unusual behaviour of the mobility threshold@xcite . as the concept of high dimensions here
is defined relative to the quasiparticle dispersion , possible playgrounds include a number of systems in physical @xmath2 dimensions , with higher dimensions being accessible numerically .
for example , recently we have shown how this transition can be observed in 1d and 2d arrays of ultracold ions in optical or magnetic traps@xcite . because weyl semimetals ( wsms ) are currently one of the most well - known and experimentally accessible platforms@xcite for the observation of these high - dimensional disorder - driven phenomena , tremendous research efforts have been directed at studying the critical properties of the transition in a wsm .
nevertheless , the values of the critical exponents at the transition are currently under debate .
quantum criticality near the transition has been studied analytically for dirac and weyl particles in dimensions @xmath3 by means of perturbative renormalisation - group approaches^,,^ , large-@xmath4 ( large number of valleys or particle flavours ) analysis@xcite , and ( uncontrolled ) self - consistent born approximation@xcite ( scba ) , equivalent to a large-@xmath4 calculation in the limit @xmath5 .
usually weyl and dirac materials have only several nodes , @xmath6 , which makes large-@xmath4 approaches quantitatively inaccurate in all dimensions @xmath1 [ for detailed criticism of the scba vs. rg see refs . , (
although refs . and
are devoted to graphene , their arguments apply as well to dirac and weyl particles in all dimensions ) ] .
perturbative renormalisation - group ( rg ) analysis are controlled by the small parameter @xmath7 .
although such rg analysis for small @xmath8 is not guaranteed to be quantitatively accurate when analytically continued to 3d ( @xmath9 ) , already the one - loop results@xcite@xcite predict the correlation - length exponent @xmath10 and the dynamical exponent @xmath11 that lie within @xmath12 and several percent of the values obtained numerically in refs .
, , . also , numerical analysis of ref . ,
albeit carried out for 1d chiral systems , suggest that one - loop rg results accurately describe the critical properties of such type of transitions .
recently a value of @xmath13 very close to the one - loop result @xmath11 has been obtained numerically in ref . .
however , the same simulations found a value of @xmath14 very different from the one - loop prediction @xmath10 .
it has been argued in ref . that the one - loop result @xmath11 for the dynamical critical exponent is exact and holds in all orders of the renormalisation , in contradiction with the analytical two - loop calculations of ref .
predicting is defined as @xmath15 , while our convention in this paper is @xmath7 . ] @xmath16^{-1}\approx 1.14 $ ] and @xmath17 .
similar analytical two - loop rg calculations for graphene@xcite and for related the gross - neveu model@xcite , with the results for the latter being well established in the high - energy literature , yield beta - functions inconsistent with those of refs . and .
in this paper we present an independent calculation of the critical exponents for the transition in a weyl semimetal using a two - loop renormalisation group approach for weyl fermions in @xmath0 dimensions and resolve controversies currently existing in the literature .
we obtain beta functions consistent with those obtained for graphene in ref . and for gross - neveu model in refs .
, but in disagreement with the results of refs . and .
we find the critical exponents to be @xmath18^{-1 } , \label{nu } \\ z & = 1-\frac{\varepsilon}{2}-\frac{\varepsilon^2}{8}+\ldots .
\label{z}\end{aligned}\ ] ] for a 3d weyl semimetal ( @xmath19 ) eqs .
( [ nu ] ) and ( [ z ] ) give @xmath20 we emphasise , however , that higher - loop corrections for @xmath19 will lead to deviations from these values obtained from an rg calculation controlled by small @xmath8 .
the critical density of states in a wsm with short - range - correlated disorder can be described in the supersymmetric representation@xcite by a field theory with the action @xmath21\phi\ : d\br + \frac{1}{2}\varkappa_0\int(\phi^\dagger\phi)^2d\br , \label{action0}\end{aligned}\ ] ] where @xmath22 is a vector consisting of an anticommuting ( fermionic ) @xmath23 and commuting @xmath24 ( bosonic ) components , @xmath25 is the matsubara frequency in the upper half - plane , @xmath26 is a vector of matrices generating a @xmath1-dimensional clifford algebra ( @xmath27 ) , @xmath28 is the momentum operator , and @xmath29 is the strength of the short - range - correlated random disorder potential under consideration .
equivalent field theories can be derived also in keldysh@xcite and replica@xcite representations .
we emphasise , that we express the action ( [ action0 ] ) in terms of a positive matsubara frequency @xmath30 , in contrast with the conventional real - frequency representation@xcite , in order to regularise integrals in the renormalisation scheme used below .
also , frequency @xmath30 in eq .
plays the same role as the mass @xmath31 in the related gross - neveu model@xcite .
the density of states is determined by the retarded green s function @xmath32 that can be obtained from the action ( [ action0 ] ) using analytic continuation to real frequencies @xmath33 .
we note , that realistic materials always have an even number of weyl nodes , due to the fermion doubling theorem@xcite , and thus in general should be described by an action with an even number of weyl fermion flavours .
however , for sufficiently smooth disorder internodal scattering can be neglected , and the material is equivalent to an even number of copies of single - node wsms described by the action ( [ action0 ] ) . in dimensions
@xmath3 this field theory leads to ultraviolet divergencies in physical observables and requires an appropriate rg treatment .
we study the behaviour of the system at frequency @xmath30 and at long length scales , @xmath34 , following the _ minimal subtraction _
renormalisation scheme@xcite .
we use dimensional regularisation by computing observables in lower @xmath35 dimensions ( with small @xmath36 ) and then analytically continue renormalised observables to the higher dimensions of interest ( @xmath37 ) . the lagrangian in this scheme
is separated into the effective lagrangian @xmath38 of variables observable in the long - wave limit of interest and the counterterms : @xmath39 where the energy scale @xmath40 and the renormalised disorder strength @xmath41 are experimentally observable , and the counterterms @xmath42 cancel the divergent ( in the powers of @xmath43 ) contributions to physical observables that come from the lagrangian .
the strength of disorder can be conveniently characterised by the dimensionless parameter@xcite @xmath44 where @xmath45 . for a given `` bare '' disorder strength @xmath46 the renormalised dimensionless disorder strength @xmath47 and the characteristic energy @xmath40 of the long - wave behaviour of disorder - averaged observables
are related by the rg equation ( see appendix for a detailed derivation ) @xmath48 the dependence of @xmath40 on the frequency @xmath30 is described by the rg equation @xmath49 our two - loop rg equations and are consistent with the previous studies of weyl fermions in @xmath35 dimensions : in the framework of gross - neveu model in refs . and of graphene in ref . .
. shows that the dimensionless disorder strength grows or decreases depending on whether or not it exceeds a critical disorder strength @xmath50 the existence of such a repulsive fixed point signals a transition , discussed in the literature@xcite , between a weak - disorder phase and a strong - disorder phase . using that near the transition @xmath51 and eq .
we obtain the correlation - length critical exponent .
the divergence of @xmath52 in the @xmath53 limit reflects the fact that @xmath54 ( @xmath55 ) is the lower critical dimension for this transition .
our analysis predicts @xmath56 at the critical disorder strength @xmath57 .
. then gives the dynamical critical exponent .
we note that for small @xmath58 the correlation - length exponent @xmath52 , eq . , satisfies harris criterion ( chayes inequality)@xcite @xmath59 .
because the correlation length @xmath60 can be measured as a function of the energy @xmath30 for @xmath61 as well as a function of disorder strength @xmath62 for @xmath63 , a similar criterion can be applied heuristically to the exponent @xmath64 , yielding @xmath65 , consistent with our result eq . . in conclusion ,
we have presented a two - loop renormalisation - group analysis of the critical properties of the unconventional disorder - driven transition for weyl fermions above two dimensions and found the correlation - length and dynamical critical exponents , eqs . and .
the beta - functions that we obtain are consistent with those for graphene and gross - neveu models studied previously in the literature in the disordered problem under consideration is opposite to that of the respective models studied in high - energy literature for repulsively interacting fermions , which reflects in different signs of even - in-@xmath47 terms of beta - functions . ] .
_ acknowledgements .
_ we appreciate useful discussions with a.w.w .
ludwig .
our work was supported by the alexander von humboldt foundation through the feodor lynen research fellowship ( svs ) and by the nsf grants dmr-1001240 ( lr and svs ) , dmr-1205303(vg and svs ) , phy-1211914 ( vg and svs ) , and phy-1125844 ( svs ) .
lr also acknowledges support by the simons investigator award from the simons foundation .
_ note added . _ after posting this paper on arxiv , the results of refs . and , contradicting our conclusions here , have been withdrawn by their authors ; the claim of @xmath11 being the exact dynamical exponent has been removed@xcite by the authors of ref . , while the results of ref . for
the critical exponents have been retracted in erratum .
39 natexlab#1#1bibnamefont # 1#1bibfnamefont # 1#1citenamefont # 1#1url # 1`#1`urlprefix[2]#2 [ 2][]#2 , * * , ( ) . , * * , ( ) . , , , , * * , ( ) .
, , , * * , ( ) . , , , * * , ( ) . , , , , , * * , ( ) .
, , , , , , , , , , , * * , ( ) . , , , , , , , , , , ,
* * , ( ) . , , , , , , , , , , , * * , ( ) . , * * , ( ) . , * * , ( ) . , * * , ( ) . , * * , ( ) . , * * , ( ) . , * * , ( ) . , , , * * , ( ) . , , ( ) , . , , ( ) , .
, , , * * , ( ) . , , ( ) , .
, , , , * * , ( ) . , * * , ( ) . , * * , ( ) . , * * , ( ) . , , , , * * , ( ) . , * * , ( ) . , * * , ( ) . , _ _ ( , , ) . , _ _ ( , ) . , * * , ( ) . , * * , ( ) . , _ _ ( , ) . , , , , * * , ( ) . ,
* * , ( ) .
, , , , * * , ( ) . , * * , ( ) . , , , * * , ( ) .
( ) , . , , , _ _ ( , ) |
evolving random graphs have recently attracted attention , see e.g. refs @xcite and references therein .
this interest is mainly motivated by concrete problems related to the structure of communication or biological networks .
experimental data are now available in many contexts @xcite . in these examples ,
the asymmetry and the evolving nature of the networks are likely to be important ingredients for deciphering their statistical properties .
it is however far from obvious to find solvable cases that would possibly account for some relevant features of , say , the regulating network of a genome .
although biology has strongly influenced our interest in evolving networks , the model we solve is not based on realistic biological facts but it nevertheless incorporates asymmetry and chronological order .
understanding such simple evolving graphs may help understanding biological networks , at least by comparison and opposition .
we were initially motivated by the study of the yeast genetic regulatory network presented in ref.@xcite .
the authors studied in and out degree distributions and discovered a strong asymmetry : a single gene may participate to the regulation of many other genes the law for out - degrees seems to be large , but each genes is only regulated by a few other genes the law for in - degrees seems to have finite moments .
this is why we consider oriented evolving random graphs in the sequel .
a biological interpretation for the asymmetry is that the few promoter - repressor sites for each gene bind only to specific proteins , but that along the genome many promoter - repressor sites are homologous .
however , this does not predict the precise laws .
an understanding of the same features from a purely probabilistic viewpoint would be desirable as well .
the recent experimental studies dealt with global statistical properties of evolving graphs , i.e. when the evolving network is observed at some fixed time with the ages of different vertices and edges not taken into account .
there are simple experimental reasons for that : to keep track of the ages would in many cases dramatically reduce the statistics , and in other cases this information is even not available .
our second motivation is a better understanding of the local - in - time statistical properties of evolving networks .
this helps dating or assigning likely ages to different structures of the networks .
as we shall later see , the global analysis , which is like a time average , gives a distorted view of the real structure of the networks .
we shall present a detailed analysis of local - in - time features in our model .
the model we study is the natural evolving cousin of the famous erds - renyi random graphs @xcite .
starting from a single vertex at time @xmath0 , a new vertex is created at each time step so that at time @xmath1 , the size of the system , i.e. the number of vertices , is @xmath1 , and new oriented edges are created with specified probabilistic rules .
a tunable parameter @xmath2 ranging from @xmath3 to @xmath4 describes asymptotically the average number of incoming edges on a vertex .
precise definitions are given in the next section .
our main results are the following : from very simple rules , we see an asymmetry emerging .
the global in and out degree distributions are different .
we also compute the local profiles of in and out degree distributions , and comment on the differences .
we make a detailed global analysis for the structure and sizes of the connected components .
we use generating function methods to write down a differential equation that implies recursion relations for the distribution of component sizes , see eqs.([cdiff],[crecur ] ) .
a salient global feature of the model is a percolation phase transition at a critical value of the average connectivity . below this value
, no single component contains a finite fraction of the sites in the thermodynamic limit , i.e. in the large @xmath1 limit .
however , a slightly unusual situation occurs in that below the transition the system contains components whose sizes scale like a power of the total size of the graph , see eq.([eq : grosclu ] ) .
correspondingly , the probability distribution for component sizes has an algebraic queue , see eq.([asympk ] ) , and its number of finite moments jumps at specific values of the average connectivity . above the transition , this probability distribution becomes defective , but its decrease is exponential , see eq.([pklarge ] ) .
the transition is continuous .
close to the threshold , the fraction of sites in the giant component the percolation cluster has an essential singularity , see eq.([eq : pof ] ) .
we argue that this result is universal , with the meaning used in the study of critical phenomena .
the essential singularity at the percolation threshold had already been observed numerically by @xcite in a different model which we show to be in the same universality class as ours for the percolation transition , and computed analytically for another class of models in @xcite .
we then turn to the study of local - in - time profiles of connected components .
guided by a direct enumeration based on tree combinatorics , we show that they satisfy recursion relations , and we give the first few profiles ( isolated vertices , pairs , triples ) explicitly .
the profile of the giant component is given by a differential equation , from which we extract the singularity in the far past and the critical singularity in the present
see eqs([eq : rho_0],[eq : rho_1 ] ) .
in particular the giant component invades all the time slices of the graph above the transition .
one strange feature of profiles , which would deserve a good explanation , is that in several instances the formal parameter involved in generating functions for global quantities is simply traded for the relative age to obtain interesting local - in - time observables , see eqs.([eq : young],[ddiff ] ) .
we have compared our analytical results with numerical simulation whenever possible . while polishing this paper , we became aware of @xcite , whose goals overlap partly with ours . when they can be compared , the results agree
we construct evolving random graphs with the following rules : + ( i ) we consider a triangular array of independent random variables @xmath5 , @xmath6 , where @xmath5 takes value @xmath0 with probability @xmath7 $ ] and value @xmath3 with probability @xmath8 .
+ ( ii ) we start from the graph made of single vertex at initial time @xmath9 . at time @xmath1 , @xmath10 , a vertex with label @xmath1 is added together with the _ directed _ edges @xmath11 $ ] for which @xmath12 .
we shall often take the viewpoint that the ( biased ) coin tossings defining @xmath13 are done at time @xmath1 .
we shall assume that @xmath14 at large time @xmath1 , with @xmath2 a parameter which we shall identify as half the average connectivity .
this choice ensures the convergence of various distributions to stationary measures , most of them being independent of the precise values of the early probabilities . by construction
all edges arriving at a given vertex are simultaneously created at the instant of creation of this vertex . as a consequence ,
these graphs are not only oriented but chronologically oriented this is unrealistic from the biological viewpoint .
in this section we give the incoming and outgoing edge distributions .
+ let @xmath15 be the number of incoming edges at the vertex @xmath16 , and @xmath17 be the number of outgoing edges at this vertex at time @xmath1 .
+ let @xmath18 be the number of vertices with @xmath19 incoming edges , and @xmath20 be the number of vertices with @xmath19 outgoing edges at time @xmath1 .
we may look either at the edge distributions at a given vertex , or we may look at the edge distributions defined by gathering averaged histograms over whole graphs .
the former are specified by their generating functions , @xmath21 where @xmath22 denotes expectation value .
it may depend on the specified vertex labeled by @xmath16 .
the latter is defined by the generating functions , @xmath23 we remark that this global histogram distribution is the average of the local - in - time quantity @xmath24 . since at time @xmath1 the total number of vertices is @xmath1 , @xmath25 is properly normalized , @xmath26 , to define an averaged probability distribution function , independent of the vertices , for the incoming or outgoing edge variables @xmath27 : @xmath28 _ incoming vertices . _
the number of incoming edges @xmath29 at vertex @xmath30 asymptotically possesses a poisson distribution since @xmath31\simeq_{j\to\infty}\exp({\alpha}(z-1))\ ] ] the convergence of this distribution justifies our choice of asymptotic probabilities @xmath32 .
only the vertices whose ages @xmath16 scale with the age of the graph , i.e. with @xmath33 fixed , @xmath34 , give non trivial contributions at large time to the averaged histogram ( [ histo ] ) and @xmath35 this expression may also be retrieved by looking at the evolution equation of @xmath36 . indeed , consider adding the new vertex at time @xmath1 . since the edges are oriented from older to younger vertices , .
] we have @xmath37 from the second definition in eq.([histo ] ) .
this is equivalent to @xmath38 as @xmath39 at large time , the stationary limit is given by eq.([v- ] ) .
this yields a poissonian distribution with probabilities @xmath40 _ outgoing vertices .
_ at a given vertex @xmath30 , with @xmath33 fixed , the number of outgoing edges @xmath41 at vertex @xmath16 also have a poisson distribution at large time , @xmath42 \simeq_{t\to\infty } \exp(-{\alpha}\log\sigma(z-1))\ ] ] but with a parameter @xmath43 depending on the age of the vertex .
approximating at large time the sum over @xmath16 in eq.([histo ] ) by an integral over @xmath44 gives the histogram distribution : @xmath45 as for incoming vertices , this formula follows from the evolution equation for @xmath46 . indeed , since the numbers of outgoing edges
@xmath47 from vertex @xmath16 at time @xmath1 and @xmath48 differ by @xmath49 we have @xmath50 . from definition ( [ histo ] ) this gives @xmath51 where the first term is the contribution of the newly added vertex at time @xmath1 .
the stationary limit is given by eq.([v+ ] ) .
this is a geometric distribution , slightly larger than the poisson distribution , with probabilities @xmath52 _ mixed distribution_. let @xmath53 be the number of vertices with @xmath54 outgoing and @xmath55 incoming edges at time @xmath1 . as in eq.([histo ] ) , the generating function for the mixed histogram distribution is defined by @xmath56 by construction the outgoing and incoming edges variables @xmath57 are statistically independent for @xmath16 fixed , so that the last expectation values factorize . as above we may derive an evolution equation by evaluating the contribution of the newly added vertex at time @xmath1
this yields : @xmath58 its stationary limit is factorized : @xmath59 outgoing and incoming edges are statistically independent at large time .
in this section , we present the main relations governing the probability distributions of connected components of the graphs .
two vertices belong to the same connected component if they can be joined by a path made of edges , without any reference to orientation .
this definition ensures that the property of being in the same connected component is an equivalence relation , but does not ensure that two points in a connected component can be joined by an oriented path . to partly avoid repetitions , the term _ cluster _ is used as a synonymous for _ connected component _ in the sequel .
intuitively , the fact that the network is fragmented can be understood as follows : when a vertex @xmath60 is created , it has a finite probability to be isolated , and the probability that none of the vertices @xmath61 connects to vertex @xmath60 is @xmath62 which scales as @xmath63 .
this quantity remains finite as long as @xmath64 does .
this argument shows that there are isolated vertices in the system .
a small extension of the argument shows that there are also finite components and that young vertices are more likely to be in small components than old ones .
this will be made more rigourous in the study of profiles , see section [ sec : chropro ] .
let @xmath65 be the number of connected components with @xmath19 vertices at time @xmath1 and let @xmath66 be the generating function , @xmath67 by definition , @xmath68 is the number of components and @xmath69 the total number of vertices , @xmath70 at any finite time .
let us write an evolution equation for @xmath66 . at time @xmath71
, we add the vertex with label @xmath71 which may then be connected to @xmath72 connected components of size @xmath19 .
this creates a new component of size @xmath73 , but also removes @xmath72 components of size @xmath19 .
thus , at time @xmath71 we have : @xmath74 with @xmath75 the kronecker symbol .
alternatively , @xmath76 as is apparent from this formulation , the transition probability from a given @xmath66 to a given @xmath77 can be given in closed form . to be precise ,
the admissible @xmath66 s ( describing the accessible distributions of components at time @xmath1 ) are polynomials with integral non - negative coefficients , whose derivative at @xmath78 have value @xmath1 .
now suppose @xmath66 and @xmath77 are admissible . if the difference @xmath79 can not be written as @xmath80 for some set of nonnegative integers @xmath72 , the transition is forbidden .
if it can , then the @xmath72 s are uniquely defined and the transition probability is @xmath81 the meaning of this equation is simple . at time @xmath71 , the new vertex is added , and for each of the former @xmath1 points a ( biased ) coin is tossed to decide the value of the edge variables @xmath82 .
the tossings are independent with the same law , so the probability that the new point does not attach to a given component of size @xmath19 is @xmath83 , and distinct components are independent .
hence for each @xmath19 one makes @xmath65 independent bernoulli trials with failure probability @xmath83 , and the transition from @xmath84 requires exactly @xmath72 successes .
this shows that the graph evolution is a ( time inhomogeneous ) markov process on the space of components distributions , a fact that we shall use for the purpose of numerical simulations .
this explicit representation of the transition probability could be used to average equation ( [ ntbis ] ) .
alternatively , one can represent the number @xmath72 of components of size @xmath19 which are connected to the new vertex in terms of the edge variables @xmath5 as @xmath85=1}^{n_k(t)}[1 - \prod_{j\in[k]}(1-\ell_{j , t+1})]\ ] ] where @xmath86 $ ] runs over connected components of size @xmath19 . since the edge variables @xmath82 are statistically independent of the earlier edge variables , @xmath87 with @xmath88 , and therefore also independent of the @xmath65 s , we have for any @xmath89 , @xmath90^{n_k(t ) } \rangle } \label{ndist}\end{aligned}\ ] ] in particular , @xmath91 we can now take the average value of eq.([ntbis ] ) to get @xmath92^{n_k(t ) }
\rangle } \nonumber \\ & = & { \langle n_{t}(zq_{t+1 } ) \rangle } + z q_{t+1}^t { \langle \prod_{k\geq 1}[1+(q_{t+1}^{-k}-1)z^k]^{n_k(t ) } \rangle } \label{nmoyen}\end{aligned}\ ] ] assume , but this will be justified later using the asymptotic behavior of the @xmath93 s , that at large time and for fixed component size @xmath94 is self - averaging , meaning that @xmath95 with @xmath96 equals to its averaged value with probability one and with the remaining other @xmath97 terms random .
this in particular implies that for any finite size @xmath19 the number of connected components of size @xmath19 scales thermodynamically with the graph size .
if both sides of eq.([nmoyen ] ) are expended in powers of @xmath98 , a given degree involves only components of bounded ( @xmath1-independent ) size .
so , order by order in @xmath98 , self - averaging applies and we conclude that taking @xmath99 , the following is an accurate approximation for the last term in eq.([nmoyen ] ) at large times : @xmath100^{n_k(t ) } \simeq_{t\to \infty } \exp(-{\alpha}+{\alpha}\sum_{k\geq 1}kz^kc_k).\ ] ] the averaged evolution equation ( [ nmoyen ] ) then gives a deterministic differential equation @xmath101 for the generating function @xmath102 the function @xmath103 has a direct probabilistic interpretation .
indeed , @xmath104 is the fraction of points in clusters of size @xmath19 , or equivalently the probability that a randomly chosen vertex belongs to a connected component of size @xmath19
. by construction @xmath105 is a taylor series with positive coefficients .
the series @xmath106 is convergent because it counts the fraction of points in finite clusters which is @xmath107 . as a consequence
the radius of convergence of @xmath105 is at least @xmath0 : if we denote by @xmath108 the radius of convergence @xmath105 , we know that @xmath109 for @xmath110 .
we show in appendix a that analogous methods can be used to describe the connected components of the erds - renyi random graph model .
most of our subsequent analysis will be based on ( [ cdiff ] ) .
it recursively determines the @xmath96 s .
the first few are : @xmath111 more generally , the combinations @xmath112 are rational functions in @xmath2 .
they are more efficiently computed using a second order differential equation which follows from derivation of the logarithm of eq.([cdiff ] ) : @xmath113 this leads to @xmath114 a general feature of the recursion relations is that the rational function @xmath112 has no poles except possibly at the points @xmath115 .
eq.([nmoyen ] ) is not a closed formula for the probability distributions of the number of connected components .
but as shown in appendix b , one may improve slightly the above argument to obtain a closed system ( see eq.[mast ] ) which can be used to prove systematically that the variables @xmath96 are self - averaging , a task we perform for @xmath116 and @xmath117 in appendix b. the proof becomes more and more tedious for large @xmath19 .
but we are confident that the self averaging property is general because of the close relationship with the erds - renyi model presented in appendix a.
up to now , our analysis of connected components has always concentrated on finite components : the thermodynamic limit @xmath118 has been taken while keeping @xmath19 , the component size , arbitrary but fixed .
we have argued that in this regime , the number of components of size @xmath19 is proportional to @xmath1 .
this has lead to a satisfactory description of @xmath96 .
our arguments did not use any hypothesis on whether or not only finite components play a role in the thermodynamic limit .
however , as we observed already , the sum @xmath119 measures exactly the fraction of the sites that are either isolated or in components of size @xmath120 or @xmath121 or @xmath122 . to rephrase it more vaguely , @xmath119 counts the fraction of vertices that belong to finite connected components . if only clusters of finite size contribute to the thermodynamic limit , then @xmath123 . else , by standard physical arguments , a single giant component of size @xmath124 accounts for the deficit .
the giant component is also called the percolating cluster .
its relative size , which we denote by @xmath125 , is @xmath126 and @xmath127 .
left : the fraction of sites occupied by the giant component .
right : the variation of @xmath128 close to the transition . _
_ ] and @xmath127
. left : the fraction of sites occupied by the giant component .
right : the variation of @xmath128 close to the transition . _
_ ] to discriminate between the two situations , @xmath125 may be computed numerically by evaluating a large number of coefficients @xmath104 using the recursion relation ( [ crecur ] ) . as we shall see later
, the convergence of the series is slow for the whole relevant range of @xmath2 .
the result of such a partial summation @xmath129 is plotted in fig.[fig : perconum ] for @xmath130 .
it reveals a phase transition at a value @xmath131 between @xmath132 and @xmath133 , going from a regime where finite components contain all vertices to a regime where they do not .
below @xmath132 and above @xmath133 , the plots corresponding to different values of @xmath134 are hard to distinguish , but in the transition region , large values of @xmath19 make substantial contributions to @xmath135 .
the transition is also manifest on an analogous study of @xmath136 .
we shall show later that @xmath137 is the exact threshold in this model and that @xmath136 is discontinuous at the transition .
the growth of @xmath125 just above the threshold seems to start with many vanishing derivatives , in strong contrast with what happens in the erds - renyi random graph , for which the growth of the giant component is linear close to the transition .
this can be related to the following observations : as we have recalled in appendix a , in the erds - renyi model the components of size @xmath19 occupy a fraction @xmath138 . as a function of @xmath2 ,
this fraction has a single maximum at @xmath139 .
these values accumulate at @xmath140 , the well - known transition point for the standard random graph .
then for @xmath141 , the fraction of sites occupied by components of size @xmath19 decreases with a finite slope for all @xmath19 s , and so does the sum , so that the growth of the giant component is linear close to the transition .
in the model studied in this paper , the behavior of @xmath142 as a function of @xmath2 for generic @xmath19 is not so easy to get at .
however , a simple numerical analysis leads to the following picture : @xmath143 is a decreasing function of @xmath2 , but for @xmath144 , @xmath104 has a single maximum , at say @xmath145 .
this sequence starts with @xmath146 , @xmath147 @xmath148 , is maximum for @xmath149 with @xmath150 and then decreases very slowly ( @xmath151 , @xmath152 , @xmath153 ) , apparently getting closer and closer to @xmath137 . at the transition
, most @xmath142 s are still increasing , and close enough to the transition a finite but large number of them is still increasing .
so subtle compensation mechanisms can take place , leading possibly to the vanishing of ( infinitely ) many derivatives of @xmath125 at @xmath154 .
this is confirmed in the following subsections , which are also devoted to a more precise description of the distributions of finite and infinite clusters .
to summarize : @xmath155
we turn to the examination of the consequences of eq.([cdiff ] ) , the equation that determines the generating function for the number of clusters of given finite size . we know that @xmath108 , the radius of convergence of @xmath105 , is at least @xmath0 . to analyze the behavior of @xmath105 around @xmath78
, we define @xmath157 from eq.([cdiff ] ) @xmath158 satisfies @xmath159 below the transition , there is no percolating cluster so that @xmath160 or alternatively , @xmath161 this makes clear that the normalized positive numbers @xmath104 are the probabilities for a vertex to be in a connected component of size @xmath19 . for @xmath162
this translates into the boundary condition @xmath163 we first look for a formal solution @xmath164 of eq.([feq ] ) in the form of a taylor series in @xmath165 : @xmath166 the taylor coefficients are the moments of the measure @xmath104 : @xmath167 .
for example @xmath168 is the average proportion of vertices per component . as eq.([cd2 ] ) the differential equation eq.([cdiff ] ) may be turned into a second order differential equation for @xmath169 , @xmath170 eq.([cdiff ] ) with @xmath171 gives @xmath172 .
eq.([ydiff ] ) then allows us to recursively compute the @xmath173 s : @xmath174 with @xmath175 the square root singularity in the expression of @xmath168 indicates that the initial boundary condition ( [ mesp ] ) becomes pathological at @xmath154 and thus signals the percolation transition .
the recursion relation for the higher taylor coefficients shows that @xmath176 possesses a pole at @xmath177 .
it actually changes sign , from positive to negative , across the pole . as a function of @xmath178 , the @xmath176
s have a simple pole at @xmath179 , @xmath180 where the numerator does not vanish at @xmath181 .
since @xmath182 at least for @xmath183 , this simple pole codes for the asymptotic behavior of @xmath104 at large @xmath19 . indeed , recall that if @xmath184 as @xmath185 then @xmath186 diverges as @xmath187 from below with a simple pole at this value . thus assuming that we may extend the simple pole ( [ simplet ] ) to non - integers value of @xmath178 we learn that @xmath188 with @xmath189 this means in particular that the probability distribution @xmath104 ( which describes the proportion of the system occupied by clusters of size @xmath19 ) is large and only the moments @xmath176 for @xmath183 , exist .
the fact that the taylor coefficients @xmath176 , computed from the recursion relation ( [ mu < ] ) , can not consistently be interpreted as moments of a probability distribution for @xmath190 indicates that the expansion ( [ trunc ] ) is only up to @xmath191 terms .
indeed , the differential equation ( [ feq ] ) is compatible with an expansion of @xmath162 for @xmath192 , @xmath193 , of the form : @xmath194 with @xmath195 and @xmath196 . as a function of the complex variable @xmath98
, @xmath197 has thus a branch cut starting at @xmath78 : @xmath198 for @xmath179 an integer , this formula should becomes @xmath199 .
the cut implies the asymptotic behavior ( [ asympk ] ) for @xmath200 . at the transition but from below ,
i.e. @xmath201 , there are logarithmic correction to the scaling behavior ( [ asympk ] ) , and the branch cut equation is @xmath202 .
this ensures that the first moment is finite , and its value , computed from ( [ mu < ] ) , is @xmath203 the scaling law ( [ asympk ] ) may be linked to the typical growth rate of large clusters in the system . for concreteness ,
consider the component of vertex @xmath205 for any given @xmath205 .
for very large @xmath1 , the number of arrows emerging from vertex @xmath205 grows like @xmath206 .
then we infer that the size of the genealogical tree of @xmath205 will grow like @xmath207 ( under the hypothesis that the genealogical tree is indeed tree - like , a reasonable assumption for small @xmath2 ) .
this counting of descendants gives a crude lower bound for the size of the connected component of @xmath205 .
hence we expect that the system contains components whose sizes grow like a power of @xmath1 . to estimate this power , we argue as follows .
consider a given large cluster of size @xmath208 at time @xmath1 : \i ) @xmath209 is @xmath3 with probability @xmath210 and @xmath211 with probability @xmath212 times the probability that vertex @xmath71 connects to @xmath213 clusters of size @xmath214 apart from the large cluster ; \ii ) removing the given large cluster does not change the thermodynamical properties of the graph , so the probability that vertex @xmath71 connects to @xmath213 clusters of size @xmath214 apart from the large cluster is simply the probability that vertex @xmath71 connects to @xmath213 clusters of size @xmath214 .
hence for large @xmath1 , from eq.([nbis ] ) , @xmath215 ; \iii ) suppose we add @xmath216 new vertices with @xmath217 but @xmath218 . between time @xmath1 and time @xmath219
, many new clusters have be connected to the given large components so @xmath220 , but this has not changed the thermodynamical properties of the graph .
hence we can average the equation in i ) to get a deterministic equation @xmath221 this leads to @xmath222 we find that the growth rate of large clusters is universal . as expected , their growth exponent is larger than @xmath2 , the genealogical tree growth , because it takes not only descendants into account but the whole component .
the difference is maximum at the transition , where @xmath223 .
the fact that the same exponent , @xmath224 , governs the asymptotic behavior of @xmath104 at large @xmath19 and the size of large clusters for large @xmath1 can be understood directly as follows .
first consider one realization of the random graph for a given @xmath1 and suppose that there is a single component of maximal size , say @xmath225 .
observe that @xmath226 is by definition the number of points in components of size larger than @xmath227 .
this number is strictly larger than @xmath227 if @xmath228 , but vanishes if @xmath229 .
so @xmath225 is characterized by the identity @xmath230 from this we infer by taking the average that for large @xmath1 the relation @xmath231 gives a sensible characterization for @xmath232 , the order of magnitude of the size of large components in the graph .
we write @xmath233 and use that for @xmath234 , @xmath123 to write @xmath235 . for large @xmath232 ,
the asymptotics of the first sum is @xmath236 .
the second sum is made of finite size corrections .
if we assume that these that these are not too large , we conclude that @xmath237 , i.e. that @xmath238 .
in fact , experience from finite size scaling suggests that the two sums give contributions of the same order of magnitude .
the idea is that @xmath239 is the main contribution to @xmath240 not only when @xmath239 is of order @xmath1 , but even when simply @xmath241 so that self averaging remains valid .
this means that @xmath242 and @xmath240 become of the same order of magnitude only when @xmath19 is so large that @xmath243 .
so again we see that @xmath232 is characterized by @xmath244 , and we conjecture that @xmath245 is a scaling variable .
above the transition there is an giant component .
let @xmath125 be its relative size
. by definition , @xmath247 or alternatively @xmath248 this makes clear that the @xmath104 s and @xmath125 define the probability distribution of vertices among the clusters of different sizes , with @xmath125 the probability for a vertex to be in the percolating cluster .
the size of the giant component @xmath249 increases linearly with time : @xmath250 for @xmath1 large .
the slope may be evaluated as follows .
imagine adding a new vertex at time @xmath71 .
it is connected to the percolating cluster with probability @xmath251 so that @xmath252 where , as in eq.([nt ] ) , @xmath213 are the numbers of components of size @xmath214 connected to the new vertex . this can be rearranged as @xmath253 .
the quantity @xmath254 is self averaging , so using @xmath255 and @xmath256 at large time we infer @xmath257 which by use of eq.([nbis ] ) leads to @xmath258 as shown below the above equation is exact , but it does not determine @xmath125 as this requires knowing the moment @xmath168 . above the transition , @xmath162 still satisfies the differential equation ( [ feq ] ) but with a different boundary condition : @xmath259 with @xmath125 vanishing at the transition i.e. as @xmath260 .
this modifies the behavior of its moments .
so let us expand @xmath162 in taylor series around the origin : @xmath261 with @xmath262 .
eq.([feq ] ) implies eq.([pinf1 ] ) which is thus exact . the second order differential equation ( [ ydiff ] )
does not fix @xmath125 but determines recursively the @xmath173 s which depend parametrically on @xmath125 .
it remains to decipher what the behavior of the size of the giant component is , at least close to the transition .
this will follow from an analysis of the behavior of @xmath162 close to its singularities . as we are going to show
, @xmath162 possesses square root branch cut at a point @xmath263 but exponentially close to the origin .
this means that above the transition the singularity of the function @xmath197 is at @xmath264 , with @xmath265 depending on @xmath2 , a behavior which has to be compared with the cut at @xmath78 below the transition , eq.([cut1 ] ) .
more precisely , we show that @xmath266 with @xmath267 defined by @xmath268 to prove eq.([eq : sing ] ) we shall look at the behavior of @xmath158 in the neighborhood of three different points : at the origin @xmath269 where @xmath162 takes the boundary value ( [ fpinfty ] ) , at the branch point @xmath263 with @xmath270 , @xmath271 and , around the point @xmath272 specified by the condition @xmath273 .
let us first show @xmath274 and @xmath275 exist .
the function we are interested in is defined by @xmath276 , and satisfies the differential equation eq.([feq ] ) .
so @xmath277 and @xmath158 decreases from @xmath278 at @xmath279 to @xmath3 at a point @xmath274 which is non - negative since @xmath280 .
@xmath158 also satisfies the obvious inequalities @xmath281 for @xmath192 and @xmath282 .
so there is a single point @xmath283 , with @xmath284 , at which @xmath285 and @xmath286 has the sign of @xmath287 .
both @xmath274 and @xmath275 vanish at the transition , @xmath288 .
it turns out that most of these properties could be proved using only the differential equation .
more precisely , take any solution @xmath289 of ( [ feq ] ) with an initial condition at @xmath290 such that @xmath291 .
the function @xmath289 can be extended on the right as a positive function on a maximal interval @xmath292 .
then @xmath289 is strictly decreasing for @xmath293 , @xmath294 , @xmath295 , and @xmath296 if @xmath297 .
this discussion emphasizes the intrinsic role played by @xmath298 in this problem . before proving eq.([eq :
sing ] ) let us get intuition from a simpler , but more universal , version of the differential equation ( [ feq ] ) .
if we look at a region where @xmath158 is small and we know that such regions exist generically we can estimate @xmath299 by @xmath300 with @xmath301 approximating @xmath158 .
this leads to the simplified , but still non - linear , differential equation @xmath302 it turns out that this simpler equation can be solved in closed form . before doing that ,
let us further assume that @xmath165 is close to @xmath274 so that @xmath301 can be neglected compared to @xmath165 .
then we get @xmath303 , with solution @xmath304 leading to the announced square root singularity . but this does not give informations on the location of the branch point .
consider now the following functional of an arbitrary function @xmath301 of @xmath165 @xmath305 to fix conventions , we specify the function @xmath306 by demanding that it is continuous and takes value in @xmath307-\pi/2,\pi/2[$ ] .
the total derivative of this functional with respect to @xmath165 is @xmath308 it vanishes if @xmath301 is a solution of eq.([eq : univ ] ) , so we have indeed solved in closed form equation ( [ feq ] ) for small @xmath158 . as this is the domain we are interested in , it is tempting to argue that @xmath158 and @xmath301 should exhibit the same singular behaviour .
this turns out to be true , but there are some subtleties because the limit @xmath260 is singular : the size of the domain for which @xmath301 is a uniform approximation to @xmath158 shrinks to @xmath3 .
our strategy is to use the invariant ( [ eq : int ] ) for the approximate equation ( [ eq : univ ] ) to derive exact inequalities for @xmath158 .
let us observe that the functional ( [ eq : int ] ) is singular at points where @xmath309 where it has a jump of amplitude @xmath310 .
with this in mind , we define the functional @xmath311 in this definition , @xmath312 is a priori independent of @xmath2 , the value for which @xmath162 is considered .
the functional @xmath313 is a smooth function of @xmath165 on @xmath307-\infty,{\tau}_c[$ ] .
using the differential equation for @xmath158 , its total derivative with respect to @xmath165 is found to be @xmath314 if @xmath315 , the right - hand side is always negative , so @xmath316 is decreasing , and comparing its values at @xmath317 , at @xmath275 , at @xmath3 and at @xmath274 we find @xmath318 we know that @xmath319 , so taking a fixed @xmath320 , we can take the limit @xmath260 . at point @xmath165
, @xmath158 is analytic in @xmath321 , so @xmath322 has a finite limit , and we get exponentially small upper - bounds for @xmath323 and @xmath274 . for instance , @xmath324 is bounded above when @xmath260 . for @xmath325 , @xmath326 changes sign at a point @xmath327 solution of @xmath328 .
this point is unique as @xmath158 is decreasing and goes from @xmath278 at @xmath329 to @xmath3 at @xmath274 .
let us choose @xmath330 such that @xmath331 and such choices exist .
then @xmath326 goes from negative to positive if @xmath165 increases so that @xmath316 increases for @xmath165 varying from @xmath327 to @xmath274 .
this leads to lower bounds @xmath332 take @xmath333 , in such a way that @xmath334 vanishes exactly at @xmath275 .
then @xmath335 , and @xmath336 when @xmath260 . comparing the lower bounds with the upper bounds obtained above
, we see that @xmath337 to get a lower bound for @xmath338 we take @xmath339 such that @xmath334 vanishes at @xmath340 .
then @xmath341 and @xmath342 so we have a crude bound @xmath343 .
taking for instance @xmath344 , so that @xmath345 and @xmath346 as @xmath347 . reporting in @xmath348 yields @xmath349 .
thus , from ( [ eq : borninf ] ) , @xmath350 is bounded below when @xmath260 .
the above upper and lower bounds then imply @xmath351 we restate the physically most important result : up to an algebraic prefactor , the size of the percolation cluster close to percolation is @xmath352 in fact , we have obtained a better estimate .
we expect that @xmath353 as a finite limit for @xmath347 for a certain @xmath354 , or equivalently @xmath355 we have proved that if @xmath354 exists , @xmath356 . the asymptotic behavior of the probabilities
@xmath104 is not the same below and above the transition as @xmath105 does not have the same analytical properties on the two sides of the transition . since above the transition the branch point
is located at @xmath357 , the @xmath104 s now decrease exponentially .
more precisely , @xmath197 possesses a square root branch point at @xmath265 , @xmath358 so that @xmath359 behave as @xmath360 to be compared with eq.([asympk ] ) .
we give now a universality argument suggesting that @xmath361 and this is confirmed by solving numerically the differential equation ( [ feq ] ) .
our invocation of universality rests on eq.([pklarge ] ) , in which @xmath274 , or equivalently ( in the vicinity of the critical point ) @xmath362 , controls the exponential decrease of the @xmath104 s and plays the role of a mass gap .
this mass gap is exponentially small close to the transition , and by analogy we may argue that increasing @xmath267 is a marginally relevant perturbation of the percolating critical point . introducing the wilson - callan - symanzik beta function @xmath363 , we expect a relation @xmath364 comparison with our formula gives @xmath365 .
it is known from field theory that the coefficient @xmath366 , which dictates the exponent of the algebraic prefactor , is universal .
such universal features are controlled by the continuum limit . in our framework , the continuum region is reached at small @xmath165 , and small @xmath158 , and we expect that the continuum limit is governed by ( [ eq : univ ] )
. note that both @xmath274 and @xmath338 are exponentially small close to the transition , and @xmath162 remains small for @xmath165 between these points .
this a posteriori justifies looking at the approximate equation ( [ eq : univ ] ) .
consequently , as far as universal quantities are concerned , the inequalities in ( [ eq : bornsup ] ) can be replaced by equalities .
this leads to @xmath361 , or @xmath367
i.e. @xmath368 as announced .
as universality could suggest , eq([eq : univ ] ) turns out to describe the continuum limit for a larger class of evolving networks than just the specific one we are studying .
this is the case for the model studied in @xcite .
we refer to the original paper for the definitions .
it suffices to say that eq.([feq ] ) is replaced by @xmath369 in this equation , @xmath370 is a generating function , the coefficient of @xmath371 giving the fraction of points in components of finite @xmath19 , so @xmath372 is the fraction of sites occupied by the giant component .
the parameter @xmath373 is the average number of edges created at each time step , so this is is precisely the equivalent of our @xmath2 . to study @xmath374 close to @xmath269 ,
the approximation is @xmath375 , and we retrieve ( [ eq : univ ] ) , with @xmath2 replaced by @xmath376 .
the percolation threshold is @xmath377 ( as expected , percolation thresholds are not universal ) , and the size of the infinite component behaves like @xmath378 .
the prefactor @xmath379 compares quite well with the numerical value @xmath380 obtained in the original paper .
this fact was already noted in @xcite .
we have accumulated evidence that the percolation phase transition is very similar to the kosterlitz - thouless phase transition in the xy model .
for instance , the probabilities @xmath104 follow the scaling laws ( [ asympk ] ) below the transition with scaling exponents varying continuously with the parameter @xmath2 while they decrease exponentially above the transition .
the discussion of universality in the previous section suggests to look for scaling functions describing the neighborhood of the critical point .
for instance , from eq.([ydiff ] ) it follows that the moments @xmath173 , @xmath381 , diverge as @xmath382 with @xmath383 in particular , @xmath168 is discontinuous at the transition , jumping from @xmath384 to @xmath385 .
these scaling relations imply that the function @xmath386 has a finite limit , denoted by @xmath387 , as @xmath260 for any fixed @xmath388 , @xmath389 the differential equation ( [ feq ] ) at @xmath390 then reduces to @xmath391 .
with the boundary condition @xmath392 , this is integrated as : @xmath393 the left - hand side is very reminiscent of the limit @xmath394 of eq.([eq : int ] ) . as a consequence @xmath387 has a square root branch point at @xmath395 at which @xmath396 vanishes and @xmath397 for @xmath398 large
actually , a more precise computation based on the lagrange formula presented in appendix c yields the exact value of the scaling coefficients @xmath399 : @xmath400
one can make a direct counting of the average number of connected components in the random graph at time @xmath1 which are copies of a given finite labeled graph .
this will allow us to retrieve the results of section [ sec : clusters ] .
but this also leads to a detailed local - in - time description of the connected components which illustrates the consequences of the chronological memory of our model .
let @xmath401 be a labeled graph with vertices @xmath402 .
we let @xmath403 be the number of edges connecting vertex @xmath16 to a vertex with smaller label , so that @xmath404 is the number of edges of @xmath401 .
we look for the average number of increasing maps @xmath405 from @xmath406 $ ] to @xmath407 $ ] such that the vertices @xmath408 span a connected component of the random graph isomorphic to @xmath401 .
this number is the average number of connected components isomorphic to @xmath401 in the random graph . by the rules of construction of the random graph , the probability that the vertices @xmath409 with @xmath410 span a connected component of the random graph isomorphic to @xmath401 is @xmath411 with the convention @xmath412 . the average we look for is obtained by summing this expression over the @xmath413 s . for large @xmath1 , using the asymptotic behavior @xmath414 for large @xmath16
, the sum can be formally reinterpreted as a riemann sum , leading to a contribution @xmath415 in deriving this formula , we have not treated carefully the contribution of small values of the @xmath413 s . this is reflected in the fact that the integral can be divergent if some @xmath416 s are too large . however , the prefactor @xmath417 is the sign that in this case the sum over the @xmath413 s is nevertheless negligible in the large @xmath1 limit , as a more careful treatment would show .
the most salient feature of this formula is that only connected graphs with @xmath418 give a contribution proportional to @xmath1 , i.e. contribute to @xmath96 .
since @xmath19 is the number of vertices and @xmath398 the number of links , this relation characterizes trees as follows from the euler formula : only trees contribute to the thermodynamic limit . a given tree on @xmath19 vertices with incoming degrees @xmath416
, @xmath419 gives a contribution @xmath420 to @xmath96 .
observe that for a tree , @xmath421 for @xmath419 so that all integrals are well - defined and finite for real non - negative @xmath2 .
it is amusing to note also that the contribution of a single tree of size @xmath19 can contain poles in @xmath2 at values that are not in the list @xmath115 .
these poles have to cancel between different trees in the sum over trees of size @xmath19 , because we know that they are absent in @xmath96 .
but we have no simple explanation for this cancellation .
this explicit formula makes it easy to show that @xmath105 is ( complex ) analytic in @xmath2 in a neighborhood of @xmath3070,+\infty[$ ] for every @xmath98 such that @xmath422 .
indeed , we know that for @xmath4230,+\infty[$ ] , and @xmath422 the series @xmath424 is absolutely convergent . but @xmath142 is a sum of non - negative contributions , each tree giving a contribution of the form ( [ eq : tree ] ) .
now suppose that @xmath2 is complex with positive real part .
for each tree contribution and for fixed @xmath425 , the modulus of @xmath426 is maximal when @xmath427 and then the expression is real and positive .
this is because the statement is true for every factor . taking the sum over trees we infer that @xmath428 .
so the series @xmath424 is absolutely convergent if @xmath429 .
this equation defines , for fixed @xmath422 , a neighborhood of @xmath4230,+\infty[$ ] in which we have an absolutely convergent sum of analytic functions analytic of @xmath2 .
hence the sum is analytic in @xmath2 as claimed . to resum more explicitly the contribution of all trees of a given size , we need the generating function for labeled trees with given incoming degrees .
suppose more generally that we give a weight @xmath430 for each edge leaving vertex @xmath431 ( i.e. connecting @xmath431 to a @xmath432 ) and a weight @xmath433 for each edge entering vertex @xmath431 ( i.e. connecting @xmath431 to a @xmath434 ) .
the generating function @xmath435 for weighted trees on @xmath178 vertices factorizes nicely as @xmath436 this generalization of the famous caley tree formula labeled trees on @xmath178 vertices . ]
implies it immediately .
it seems to be little known , although it is implicit in the mathematical literature @xcite .
gilles schaeffer @xcite provided us with a clean proof using a refined version of one of the standard proofs of the caley tree formula , putting trees on @xmath19 vertices in one to one correspondence with applications from @xmath406 $ ] to @xmath406 $ ] fixing @xmath0 and @xmath19 , see e.g. @xcite .
this formula can be specialized to @xmath437 and @xmath438 for @xmath419 to give @xmath439 if one integrates only over a subset of the @xmath44 s , one gets marginal distributions .
for instance , for @xmath440 , we get that in the thermodynamic limit the fraction of sites with age close to @xmath441 that are isolated is @xmath442 . for @xmath443 , if we integrate over @xmath444 , we get that the fraction of sites with age close to @xmath441 that are the older vertex of a tree on two vertices is @xmath445 while if we integrate over @xmath446 , we get that the fraction of sites with age close to @xmath441 that are the younger vertex of a tree on two vertices is @xmath447 .
the sum , @xmath448 , gives the probability that a site with age close to @xmath441 belongs to a tree on @xmath120 vertices .
our explicit representation in terms of trees shows that in this model , and at least for questions concerning connected components , the thermodynamic limit applies not only to the full system , but also to slices of fixed relative age @xmath449 . in the next section we shall study @xmath44-dependent profiles . for small @xmath19 s
, we have done all the integrals and checked the agreement with the value of @xmath96 obtained by the recursion relation . but a general proof valid for all @xmath19 s is lacking . to illustrate the evolving nature of our model , we now determine the local - in - time distribution of the cluster sizes .
this means determining for any given age interval what is the proportion of vertices of these ages which belong to clusters of given size .
define @xmath450 to be the probability that vertex @xmath205 belongs to a component of size @xmath19 at time @xmath1 .
guided by the previous tree representation , we infer that in the thermodynamical limit @xmath451}\ , p_k(t , t')\simeq t\rho_k(\sigma)d\sigma$ ] , with @xmath452 a deterministic function . by construction , @xmath453 , the total fraction of points that belong to components of size @xmath19 .
the reasoning leading to ( [ cdiff ] ) can be generalized : one writes down a recursion relation for @xmath454 and then takes the average , a step justified by the explicit tree representation .
the event that vertex @xmath205 belongs to a component of size @xmath19 at time @xmath71 is the exclusive union of several events .
\i ) vertex @xmath205 belonged to a component of size @xmath19 at time @xmath1 and this component is not linked to the new vertex @xmath71 .
this has probability @xmath455 .
\ii ) vertex @xmath205 belonged to a component of size @xmath456 at time @xmath1 , this component is linked to the new vertex @xmath71 , and together with the other components linked to @xmath71 ( say @xmath457 components of size @xmath398 ) , it builds a component of size @xmath458 .
this has probability @xmath459 to perform the explicit sum over @xmath227 and the @xmath457 s , we introduce again generating functions and set @xmath460 .
this leads to @xmath461 in the large @xmath1 limit this complicated formula simplifies if we use again the hypothesis of self - averaging and asymptotic independence . defining @xmath462
this leads to @xmath463 together with the sum rule relating @xmath464 to @xmath465 , this fixes completely the profiles @xmath452 .
a relation between @xmath466 and @xmath105 is obtained by integrating ( [ profit ] ) for @xmath44 between @xmath3 and @xmath0 . using the defining equation for @xmath105 , eq.([cdiff ] ) , this leads to @xmath467 .
thus , we can summarize our knowledge on component time profiles with the four relations : @xmath468 expansion in powers of @xmath98 leads to recursion relations for the @xmath452 s .
they can be shown to be alternating polynomials of degree @xmath19 in @xmath469 with vanishing constant coefficient .
the first few polynomials are @xmath470 again , we have checked that for small @xmath19 s the values of @xmath471 as computed from iterated tree integrals and from the generating function coincide , but we have no general proof . as an application , let us look at the profile of vertices of relative age close to @xmath44 which are the youngest in components of size @xmath19 .
these are vertices that created , when they appeared , a component of size @xmath19 this happens with probability @xmath472 which was then left untouched for the rest of the evolution this happens with probability @xmath473 .
so the distribution of vertices that are the youngest in their connected ( finite ) component is @xmath474 in this expression , the formal parameter @xmath98 has been replaced by @xmath475 .
the giant component , when it exists , also has a well - defined profile to which we turn now .
we now determine the profile of the giant component .
so let @xmath476 be the fraction of vertices whose ages are between @xmath477 and @xmath478 which belong to the giant component . by definition @xmath479 .
we know from the tree representation that for finite clusters the thermodynamic limit applies not only on the full graph , but also on time slices . the giant component is the complement of finite components , so we expect that the density @xmath480 is self - averaging as is the size of the percolating cluster . to derive an equation fixing this density we look for the probability for a site of age @xmath481 not to be in the percolating cluster . on the one hand , by definition of the density
this probability is @xmath482 ) = 1-\rho_{\infty}(\sigma)\ ] ] on the other hand , this probability may be evaluated by demanding the vertex @xmath16 not to be connected to the older and younger vertices of the percolating cluster : @xmath482)= \prod_{k < j , k\in[k_\infty]}(1-\frac{{\alpha}}{j } ) \prod_{k > j , k\in[k_\infty]}(1-\frac{{\alpha}}{k})\ ] ] at large time , the first above product converges to @xmath483 and the second to @xmath484 .
so we get the relation : @xmath485 as @xmath486 is positive , @xmath487 is an increasing function of @xmath44 .
so @xmath480 is decreasing and has a right limit at @xmath3 .
unless @xmath488 ( i.e. @xmath489 ) , the integral has a logarithmic divergence , and henceforth @xmath490 .
more precisely , @xmath491 this means that the early vertices belong to the giant component with probability @xmath0 . on the other hand , by definition @xmath492 so taking @xmath493 in ( [ profil ] ) we get that @xmath494 this means that the late vertices always belong to the giant component with a non vanishing probability although this probability is exponentially small close to the threshold .
hence the giant component invades all time slices above the threshold , and the term percolation transition is appropriate even with this unusual interpretation .
our results are illustrated on fig.[fig : profinfin ] .
random graphs of size @xmath495 . from top to bottom , the values of @xmath2 are @xmath0 , @xmath496 and @xmath497.__,scaledwidth=70.0% ] to conclude this discussion , let us observe that @xmath486 can be expressed in terms of the profiles of finite components that we studied before : @xmath498 with @xmath499 defined in eq.([defrhoz ] ) . we end up with yet another equation constraining @xmath499 , @xmath500 if we expand @xmath501 in powers of @xmath502 as @xmath503 and define @xmath504 , we can check that @xmath505 in particular , by differentiation this implies that @xmath506 the resemblance with equations ( [ cdiff ] ) and ( [ crecur ] ) for the function @xmath105 is rather striking , but we have not been able to use this more deeply .
one difference is that ( [ cdiff ] ) determines @xmath116 from scratch , whereas ( [ ddiff ] ) does it in a two step process .
if @xmath507 is the solution of this equation satisfying @xmath508 , then @xmath509 , and then @xmath510 has to satisfy @xmath511 another difference is that the sequence @xmath512 alternates in sign .
again , the formal parameter @xmath98 receives a simple physical interpretation @xmath513 where @xmath44 labels the relative date of birth of vertices .
in this study , we have described detailed global and local - in - time features of evolving random graphs with uniform attachment rules . concerning global properties , we have shown that the model has a percolation phase transition at @xmath154 .
below the transition , the system contains clusters whose sizes scale like @xmath514 . above the transition , a single component , the giant component , grows steadily with time .
we have shown that , close to the threshold , the fraction of sites in the giant component has an essential singularity and behaves as @xmath515 .
the behaviors below and above the transitions are strongly reminiscent of the two dimensional @xmath516 model , so that our model can be interpreted as some algorithmic equivalent of it , but it seems unlikely that a direct connexion exists .
this analogy and further scaling properties we present call for an alternative renormalization group approach to the transition . by describing local - in - time profiles ,
we have shown that they offer a more accurate vision of the specificities of evolving graphs .
it would be desirable to generalize this approach by answering the following question : assume that a procedure to assign ages to the vertices of some evolving graphs has been given , what informations on the microscopic evolution rules of these graphs can be decoded from the knowledge of local - in - time statistics ?
aknowledgements : we thank gilles schaeffer for his clarifying help in tree generating functions and sergei dorogovtsev for his comments and interest .
we show how our arguments of section [ sec : clusters ] can be modified to describe the erds - renyi random graph model . in this model ,
one starts with @xmath178 points , and any two points are connected by an edge with probability @xmath517 ( so in this model all the points are equivalent ) . then a limit @xmath518 is taken .
this famous model describes a static graph , but it can also be rephrased as an evolving graph in the following way : set @xmath519 and suppose that points are added one by one , from @xmath0 to @xmath1 , each new point connecting to any previous one with probability @xmath520 . from this point of view , it can be seen that looking only at the first @xmath205 vertices ( @xmath521 ) amount to look at an erds - renyi random graph with a modified connectivity parameter @xmath522 . to get a recursion relation , we start from an erds - renyi random graph of size @xmath1 with connectivity parameter @xmath2 .
we add vertex @xmath71 and connect any older vertex to it with probability @xmath523 , so that the effective connectivity parameter for the graph on @xmath71 vertices is @xmath524 .
then the derivation proceeds as before , with the little proviso that in eq.([nmoyen ] ) , on the left - hand side @xmath525 has an effective connectivity parameter @xmath524 instead of @xmath2 .
so when we take the thermodynamic limit , nothing changes on the right - hand side of ( [ nmoyen ] ) , but an additional term @xmath526 contributes to the left - hand side . if @xmath527 denotes the analog of @xmath465 but for the erds - renyi random graph , then @xmath528 this little modification in the equation has drastic consequences .
the new equation has a single solution regular at @xmath529 , namely @xmath530 which is the well - known result .
we conclude that our self - averaging hypothesis is valid for the erds - renyi model .
this makes it more plausible that it works for our original model as well , a fact also confirmed by numerical simulations .
our goal is to prove the following formula and to investigate a few of its consequences .
we claim that @xmath531^{n_m(t ) } \rangle } \label{mast}\end{aligned}\ ] ] where the contour integral is around the origin .
it is a fokker - planck equation for the markov process formed by the @xmath65 s , which could be used to prove systematically that the variables @xmath96 are self - averaging , a task we perform for @xmath116 and @xmath117 at the end of this appendix .
we start from eq.([nt ] ) which may be rewritten as @xmath532 to compute the last term we insert the tautological identity @xmath533 in the r.h.s . to get @xmath534 to compute the r.h.s .
expectation value we use a contour integral representation of the kronecker symbol @xmath535 this yields @xmath536 the r.h.s .
can now be computed using eq.([ndist ] ) and gives eq.([mast ] ) .
this computation has a rather simple combinatorial reinterpretation .
we set @xmath537 and observe that going from time @xmath1 to @xmath71 , we add vertex @xmath71 and edges from the rest of the graph to @xmath71 .
suppose that the component of vertex @xmath71 has size say @xmath19 .
this component was build by `` eating '' some components of the graph at time @xmath1 .
a component of size @xmath16 is swallowed with probability @xmath538 and survives with probability @xmath539 .
on the other hand , if one expands @xmath540 in powers of @xmath541 , the term of degree @xmath542 enumerates all the possibilities to build the component of vertex @xmath71 with the correct probability .
defining @xmath543 , summation over @xmath19 gives @xmath544 as obtained previously .
the fokker - planck eq.([mast ] ) may also be formulated as a difference equation , similar to a discrete schrodinger equation .
let us for instance specify it for @xmath545 with @xmath546 , @xmath547 , a set of complex numbers and let @xmath548^{n_k(t ) } \rangle } \nonumber\end{aligned}\ ] ] this parametrization is similar to that used in matrix theory where @xmath549 may be thought of as the trace of the @xmath550 power of matrix whose eigenvalues are the @xmath546 s .
the contour integral in eq.([mast ] ) can then be explicitly evaluated by deforming the integration contour to pick the simple pole contributions located at the points @xmath551 .
this gives : @xmath552 eq.([mast ] ) or ( [ diffshro ] ) may be used to prove that the numbers @xmath96 are self - averaging .
let us choose for example two parameters @xmath553 and @xmath554 .
then , only the clusters of size @xmath0 give a non trivial contribution to @xmath555 at large time so that @xmath556 at large time , the difference equation ( [ diffshro ] ) then reduces to a differential equation for @xmath557 , @xmath558 it implies that @xmath559 is linear in @xmath98 at large time which means that @xmath116 is self - averaging and stationary at large time .
more precisely , integrating the above equation gives : @xmath560 in agreement with eq.([lesc ] ) .
similarly , to prove that @xmath117 possesses a finite self - averaging limit as @xmath561 we choose three parameters @xmath553 , @xmath562 so that only clusters of size two survive in @xmath555 at large time , @xmath563 eq.([diffshro ] ) then gives a first order differential equation for @xmath564 which implies that @xmath565 in agreement with eq.([lesc ] ) .
although we do not have a global argument this proof may clearly be extended to recursively prove self - averageness of any @xmath96 by choosing the parameters @xmath566 with @xmath567 .
here we present the proof of eq.([gscal ] ) . let us first recall lagrange formula . consider a variable @xmath568 defined by the implicit relation @xmath569 for some given analytic function @xmath301 .
the solution of this equation is supposed to be unique so that @xmath568 is function of @xmath570 . given another analytic function @xmath571 we look for the taylor series in @xmath570 of @xmath572 .
this composed function may be presented as a contour integral : @xmath573 expanding the integrated rational function in taylor series in @xmath570 gives lagrange formula : @xmath574 let @xmath575 .
eq.([eqgc ] ) translates into @xmath576 or @xmath577 for @xmath578 and @xmath579 .
we now apply lagrange formula with @xmath580 and @xmath581 .
this gives @xmath582 with @xmath583 and @xmath578 , this proves eq.([gscal ] ) . | we introduce a new oriented evolving graph model inspired by biological networks
. a node is added at each time step and is connected to the rest of the graph by random oriented edges emerging from older nodes .
this leads to a statistical asymmetry between incoming and outgoing edges .
we show that the model exhibits a percolation transition and discuss its universality . below the threshold ,
the distribution of component sizes decreases algebraically with a continuously varying exponent depending on the average connectivity .
we prove that the transition is of infinite order by deriving the exact asymptotic formula for the size of the giant component close to the threshold .
we also present a thorough analysis of aging properties .
we compute local - in - time profiles for the components of finite size and for the giant component , showing in particular that the giant component is always dense among the oldest nodes but invades only an exponentially small fraction of the young nodes close to the threshold .
michel bauer and denis bernard service de physique thorique de saclay ce saclay , 91191 gif sur yvette , france |
extended nonlinear systems with broken phase invariance ( e.g. , systems with only two possible phase values for a given state ) , are common in nature .
these systems may exhibit different types of patterns but , importantly , the broken phase invariance lies at the origin of the appearance , in particular , of domain walls ( dws ) which are the interfaces that appear at the boundaries between two spatial regions occupied by two different phase states @xcite . in nonlinear optics
there are several examples of spatially extended bistable systems that can show solutions for the emitted field with a given amplitude but opposite phases ( that is , phases differing by @xmath0 ) , such as degenerate optical parametric oscillators ( dopos ) or intracavity degenerate four wave mixing @xcite .
the interface which connects both solutions , the dw , can be of either one of two different types : on the one hand , there are ising walls in which the light intensity of the field at the core of the interface is zero and the phase changes abruptly from @xmath1 to @xmath2 ; on the other hand , there are bloch walls in which the light intensity never reaches zero and the change of phase is smooth across the dw @xcite .
in addition to this , ising walls are always static whereas bloch walls are usually moving fronts ( they are static only when the system is variational what is an uncommon situation for dissipative systems ) .
it is important to remark that bloch walls are chiral ( they are not identical to their mirror images ) as in the bloch wall the phase angle rotates continuously through @xmath0 and two directions of rotation are possible .
this fact has important dynamical consequences as bloch walls with opposite chirality move in opposite directions @xcite .
both ising and bloch walls have been found in nonlinear optical cavity experiments @xcite . when a control parameter is varied a bifurcation that changes the nature of the dw may happen .
this is the nonequilibrium ising bloch transition ( nibt ) that has been investigated theoretically in @xcite and has been repeatidly observed in liquid crystals ( see , e.g. , @xcite ) . in the context of nonlinear optical systems ,
the nibt has been predicted to occur in type i @xcite and type ii @xcite dopos , in intracavity type ii second harmonic generation @xcite , and in vectorial kerr cavities @xcite .
recently , we have reported the first observation of this phenomenon , the nibt , in an optical system , namely a photorefractive oscillator @xcite . moreover ,
our observation is rare in the sense that we observed a hysteretic nibt @xcite .
the aim of the present work is to study in detail the dynamics of the dws we reported in @xcite by means of the measurement of the different dws characteristics , namely intensity , phase gradient and velocity , establishing relations among them . in particular
, we consider whether the chirality parameter , which will be described later on , is appropriate for characterizing the dw .
our experimental setup , fig.1 , is a single - longitudinal mode photorefractive oscillator ( pro ) formed by a fabry perot resonator in a near self - imaging arrangement @xcite chosen in order to achieve a high fresnel number @xcite .
the nonlinear material , a @xmath3 crystal , is pumped by two counterpropagating laser beams of the same frequency . in this way a degenerate four
wave mixing process occurs within the cavity .
the degeneracy implies that the field exiting the nonlinear cavity is phase - locked and only two values of the phase ( differing by @xmath0 ) are allowed @xcite .
hence dw formation is allowed .
the system performance is ruled by different parameters such as detuning ( which is the difference between the frequency of the pump and the frequency of the cavity longitudinal mode in which emission occurs ) , gain , losses and diffraction .
all these parameters can be controlled up to some extent .
we choose as in @xcite cavity detuning as the control parameter as it can be finely tuned in an actively stabilized system @xcite .
regarding diffraction , the system is intentionally made quasi
one dimensional in the transverse dimension ( 1d system ) in order to avoid the influence of dw curvature in the observed dynamics : curvature induces a movement in the dw @xcite that contaminates that due to the nature of the dw ( i.e. , its ising or bloch character ) .
this is achieved by properly placing slits inside the nonlinear cavity ( d in fig .
1 ) , in particular , at the fourier planes ( fp in fig .
the width of the slits is adjusted to the size of the diffraction spot in these planes . in this way
beams with too large inclination ( such that their transverse wavevector falls outside the plane defined by the center line of the slit ) are not compatible with the diffraction constraints of the cavity .
this fourier filtering allows the use of finite width slits and still gets rid of most 2d effects .
it is also by using a diafragm that spatial frequencies belonging to other longitudinal modes than the one of interest are removed @xcite .
detuning , our control parameter , can be changed by means of a piezo - mirror . depending on detuning ,
different types of structures can be found @xcite but for our present purposes it suffices to say that dws exist in resonance or for positive cavity detuning ( i.e. , when the frequency of the pumping field is smaller than the frequency of the nearest cavity mode ) : at zero ( or small positive ) cavity detuning dws are static ( ising type ) , whilst they start moving ( bloch type ) if detuning is increased enough @xcite .
dws can form spontaneously from noise when the detuning value lies in the appropriate domain , as it was the case with the dws reported in @xcite . but
waiting for the appearance of dws from noise is not the most adequate strategy for their study for several reasons . on
the one hand one must wait until the formation of the dw occurs and it is very likely that not a single dw but some of them appear , which complicates the analysis because of dw interaction . on the other hand even when a single structure is formed from noise , there is the problem that its position on the transverse plane will not always be the same .
owing to these and other reasons it is most convenient to find a way for injecting , or writing , dws at will .
an injection technique to remove the initial structure and write a single dw has been devised @xcite .
it consists in injecting , for a short time ( the shutter in fig .
1 remains open for a few seconds ) , a laser beam into the photorefractive oscillator that is tilted with respect to the resonator axis .
this tilted beam has a phase profile in the transverse dimension @xmath4 given by @xmath5 , being @xmath6 a constant , @xmath7 the light wavelength , and @xmath8 the tilt angle with respect to the resonator axis .
now , this injected tilted beam forces a phase varition , along the transverse dimension , of the intracavity field phase , but given the phase sensitive nature of the degenerate four - wave mixing process @xcite , only two phase values ( say @xmath9 and @xmath0 , modulo @xmath10 ) can be amplified .
thus , the regions of the field whose phase lies in the domain @xmath11 -\pi/2,+\pi/2\right [ $ ] will be attracted towards the @xmath9 phase value , whilst those with a phase lying in the domain @xmath11 + \pi/2,+3\pi/2\right [ $ ] will be attracted towards the phase value @xmath0 . as an inmediate consequence ,
the appearance of a dw is forced as the points at which @xmath12 ( modulo @xmath10 ) .
these dws are dark because those phase values cannnot be amplified but also , and this is a more fundamental reason , because these points separate adjacent domains with opposite phases .
these points , the dws , are thus topological defects @xcite what makes them very robust as they can not be removed via continous changes . in fig .
2 we show several snapshots of the interferogram of the output field during the writing process . in fig .
2a the initial homogeneous field is shown , and the interferogram shows the homogeneity of the field phase .
then , in fig .
2b the tilted beam is injected .
while the injection is being applied , a blurred dw appears on the left , fig .
2c , and moves to the right , figs .
when the wall is located at the centre of the crystal , fig .
2f , the injection is blocked and the wall is generated , figs .
2 g and 2h . in this way
a single dw is written at the desired location along the transverse dimension .
notice that more than one dw can be written if larger angles in the writing beam are used .
in order to fully characterize the features of the different dws that are being observed , it is necessary to measure the complex field .
this is achieved by means of the interferometric technique already used in @xcite : a ccd video camera records the interference pattern between the near field ( i.e. , the intracavity field at the crystal plane ) and a homogeneous reference beam . from the interferometric pattern , like those of fig .
2 , a fourier transform technique allows the reconstruction of both the amplitude and phase of the optical field , see @xcite for full details .
some comments are in order regarding the way data must be extracted and processed .
an important feature that must be pointed out is that the spatial profile of the emitted field is not flat , as can be appreciated in fig .
3 . of course one can not expect a perfectly flat profile as the pumping fields are gaussian , but the interesting point is that the more positive cavity detuning is the less flat and less extended the output profile is , i.e. , there is a noticeable self focusing effect @xcite .
this is particularly relevant in the `` free '' transverse dimension ( @xmath4 direction in fig .
3 , which is the direction that is not limited by the intracavity diafragms ) .
this fact is at the origin of one effect that has to be taken into account when characterizing the dw dynamics : the variation of the intensity profile along the @xmath4 dimension is quite strong , stronger than for negative detuning , and no simple experimental arrangement can cope with this .
we shall come back to this point in the following section when studying the dynamics of bloch walls as @xmath4 is the direction along which bloch walls move .
apart from these nonuniformities along the @xmath4 direction , one has to take into account also the nonuniformities along the direction of the dw , i.e. , the @xmath13 direction , see fig .
the finite extension of the pump as well as the already commented self focusing effect make that the output intensity departs noticeably from a top hat profile .
this fact affects the nature of the dw as it can not be strictly uniform in the @xmath13 direction thus making necessary to discard the borders of the illuminated region in order to characterize dws properly .
let us consider , for example , the interferogram corresponding to a static wall shown in fig .
4a . for the reconstruction of the field amplitude and phase
first we must choose between taking data for a particular value of @xmath13 or making an average for different values of @xmath13 .
we choose the second option in order to smooth the effect of any local imperfection .
then we average the field obtained from the interferometric technique @xcite and obtain the average field @xmath14 .
but before we must decide if we consider all data or we take into account only the central part ( in the @xmath13 direction ) : if only the central part of the interferogram is considered for reconstructing the field amplitude and phase , one obtains the profile of an ising wall , fig .
4b , with null intensity at the center and a sharp phase jump .
contrarily , if one does not discard the borders and takes all data into account , one obtains a profile that does not correspond so clearly to an ising wall , as can be seen in fig .
4c ( the intensity does not reach zero and the phase jump is smoother ) .
of course this is nothing but the influence of a border effect that is due to the already commented nonuniformity of the fields along the @xmath13 direction , and indicates the way in which data must be treated because the ising character of the dw is clear from its dynamics : they are static . in the case of moving walls
the border effect is not so noticeable : fig .
5a shows a typical interferogram of a moving wall and the corresponding averages of the reconstructed field ( without borders in fig . 5b and with them in fig .
5c ) . in this case
the wall moves from right to left and it is interesting to notice that the field amplitude on the left side of the bloch wall is smaller than that of the right side .
this occurs always : the intensity is larger on the back side of the wall with respect the direction of movement , a feature that is reproduced with the simple model used in @xcite for reproducing a hysteretic nibt .
then the experimental procedure consists in injecting a single dw , recording the interferogram , extracting the information about intensity , phase gradient and spatial position of a single dw for different instants of time , and repeating this procedure for different values of the cavity detuning .
in this section we deal with the dynamics of dws in our particular non flat background profile .
6 shows the trajectories followed by both a static wall and a moving one that were created at the same spatial position ( the centre of the nonlinear crystal ) . as commented below , the static walls shows some slow dynamics .
the static wall is an ising wall , as its intensity and phase profiles for different intants of time , shown in fig .
7 , clearly show .
one can observe that after an initial transient in which the intensity does not reach zero and the phase jumps are relatively smooth , the intensity and phase profiles progresively approach those of an ising - type : after the transient , the change of phase through the interface is quite abrupt and the intensity at the center of the wall is very close to zero .
it must also be noticed that the amplitude of the field on the right and on the left of the dw are similar .
the initial movement can be easily attributed to the transiet behaviour that occurs until the dw reaches its final profile .
the remaining movement must be attributed to noise as the trajectory shows more a wandering behaviour than a systematic one .
the above completely characterizes static dws that are clearly identified as ising dws .
let us now consider moving walls . in fig .
8 we display their intensity and phase profiles , again for different time values .
the behaviour is very different from that of the static wall of fig .
7 : the moving dw moves continuously to the left , its velocity changes with time , and the intensity and phase also change during the motion .
we can conclude that it is a bloch wall ( smooth change of phase and nonzero intensity at the core of the wall ) .
the fact that the velocity is not constant can be attributed to the fact that for the different positions the dw is placed on a different background as the intensity decreases towards the borders of the illuminated region , fig .
3 . in other words
, the acceleration shown by the dw can be attributed to the existence of an intensity gradient in the field .
it is also to be noted that the intensity in the left side of the dw is smaller than that in the right side , and also that this effect increases as the wall approaches the border of the illuminated region .
we have checked that all these features are captured by a simple model for dw dynamics ( a ginzburg landau equation with broken phase symmetry , see @xcite ) when pump is spatially inhomogeneous .
looking at how intensity and phase change during the accelerated motion , one can qualitatively conclude that the larger the wall velocity is , the larger ( smaller ) the intensity ( the phase gradient ) at the centre of the wall is .
these facts make it difficult the determination of the dw velocity for a given set of parameters : our procedure has consisted in taking an average velocity in those cases in which it is more or less constant , discarding always the velocities corresponding to the regions located close to the border of the illuminated region ( where the acceleration is more obvious ) .
the measurements we have just commented correspond to a particular value of the cavity detuning . in order to see how a change in this parameter affects the moving dw
, we show in fig .
9 the intensity and phase profiles for three different values of the cavity detuning but corresponding to approximately the same position on the transverse direction , i.e. , to the same intensity background .
we observe again that for increasing detuning the velocity increases and , again , the more velocity the wall gets , the more intensity and less phase gradient it has . to sum up , we can conclude that there is experimental evidence of the connection between the dw velocity and the characteristics of the optical field at the centre of the interface : the smoother the phase jump is and the larger the intensity at the wall core is , the faster the dw is .
in the previous section we have shown how the intensity and phase gradient of the dw change with detuning and background conditions .
we pass now to a more quantitative study of the relation existing between the intensity at the core of the dw , @xmath15 , and the phase gradient , @xmath16 . in fig .
10 we represent , in a log log plot , @xmath16 versus @xmath15 for fixed parameters , the different points correponding to different time instants ( it is a parametric plot ) .
it is clear that a linear fit is well suited , which leads to the relation @xmath17 where the typical values for the slope and crossing point are @xmath18 and @xmath19 , respectively , and the coefficient of determination r squared has a value of @xmath20 .
a similarly good correlation is also found for other detuning values , although @xmath21 and @xmath22 change slightly with detuning _ _
( @xmath21 varies between @xmath23 and @xmath24 , and @xmath22 varies between @xmath25 and @xmath26 ) .
it is obvious that these fits do not allow saying that @xmath21 and @xmath22 are constant , but we think that their change with detuning could be attributed to a correponding change with detuning in the shape and intensity of the background ( due to the already commented self focusing effect ) . in any case
, this result invites to try to describe the dynamics in terms of only one magnitude of the field at the interface , either the phase gradient or the intensity or a combination of both .
as already commented , bloch walls are chiral structures as the complex field can rotate in the complex plane when passing from one side to the other of the dw with two different directions @xcite .
this has an important physical consequence , as bloch walls with different chiralities move in opposite directions @xcite . in @xcite
a chirality parameter defined as@xmath27 was used for characterizing the dw dynamics ( other definitions of the chirality can be proposed ) . for ising walls the chirality parameter is zero and the wall remains static , whilst for bloch walls the chirality is non zero and the wall moves with a velocity proportional to the value of @xmath28 . from our experimental results we can now justify the use of ( [ chi ] ) as a quantity that characterizes the dw dynamics . in fig .
11 a linear fit between the velocity and the chirality parameter of a dw , as it evolves with time , is shown for a particular value of the detuning .
it is worth stressing that the measurement of the velocities is not so reliable as the measurement of intensity and phase , for the reasons commented above , and for this reason the linear fit is not so good as that of fig .
we have also tried using other quantities as the chirality , the results in fig .
11 being the best ( e.g. , if one uses the intensity or the inverse of the phase gradient as chirality parameters , the correlations one obtains are @xmath29 and @xmath30 , respectively , which are considerably smaller than the @xmath31 obtained in fig .
there remains to explain how the transition from ising walls to bloch walls ( the nonequilibriium ising bloch transition , nibt ) occurs as cavity detuning is increased .
in fact this was the subject of ref .
we reproduce here a brief explanation of this for the sake of completeness . in fig .
12 a typical plot of the dw velocity as a function of cavity detuning is shown .
the rare aspect of this nibt is that it exhibits hysteresis : we start with a static wall at resonance and increase detuning until the dw starts moving , signaling an ising
bloch transition ( this is marked as ibt in fig .
12 ) . for larger detuning values ,
dws always move . then , from a large positive detuning , we inject a dw and check whether it moves or not , and repite this operation for decreasing detuning values until we obtain an static ( ising ) wall ( this is marked as bit in fig . 12 ) .
the interesting thing is that the two nibts occur at different detuning values . in other words
, there is a finite detuning range ( between bit and ibt ) where , for fixed detuning , we obtain either ising or bloch walls in subsequent injections .
notice that the bistability range an the velocity values in fig .
12 are smaller , roughly by a factor of 50% , than those reported in @xcite .
the reason is that the two cases correspond to two different experiments with different gains what obviously affects the dw dynamics quantitatively .
let us finally comment that a simple model that predicts a hysteretic nibt like the one we have just commented can be found in @xcite .
in this model one finds that the origin of the hysteresys lies on the existence of bistability in the homogeneous solutions of the system , a bistabillity that we have also found in the experiment .
we refer the interested reader to @xcite for further details .
in conclusion , we have reported a detailed experimental characterization of domain walls dynamics in a nonlinear optical cavity , namely a photorefractive oscillator working in a degenerate four wave mixing configuration .
an injection technique to remove any spontaneous initial structure and to write a single dw at any desired location has been shown .
the dynamics of the dws has been characterized in terms of the intensity and phase gradient of the optical field .
we have also shown the relation existing between these two quantities that leads naturally to the identification of the chirality parameter .
finally , the hysteretic nonequilibrium ising bloch transition , which turns out to be hysteretic in our case , has been shown .
this work has been financially supported by the spanish ministerio de ciencia y tecnologa and european union feder ( projects bfm2002 - 04369-c04 - 01 , fis2004 - 06947-c02 - 01 and fis2005 - 07931-c03 - 01 ) , and by the agncia valenciana de cincia i tecnologia of the valencian government ( project grupos03/117 ) .
was financially supported by the spanish ministerio de educacin , cultura y deporte ( grant sab2002 - 0240 ) .
we note that for negative cavity detuning ( where a stripe pattern is observed , see @xcite ) a wide output beam is obtained in the @xmath4 direction .
as detuning is made positive and increased the output beam becomes increasingly narrower in the `` free '' transverse dimension .
this observation is compatible with the observations in @xcite . *
1.- * scheme of the experimental setup .
m : cavity mirrors ; l : effective cavity length ; d : diaphragm located in the fourier plane ( fp ) that makes the system quasi-1d ; pm : piezo mirror for control of the cavity detuning ; ccd : cameras to take pictures from near field ; and l : lenses . * fig . 2.-
* experimental interferometric snapshots of the domain wall injection process .
a ) initial homogeneous field , b ) injection is applied , c)-e ) a domain wall appears and moves , f ) the injected wall is located at the centre of the crystal , g ) the injection beam is blocked , and h ) the domain wall is finally written .
time runs from top to bottom in steps of 5 .
the transverse dimension is 1.6 mm . *
fig . 4.- * experimental interferometric snapshot of a static domain wall ( a ) , reconstructed amplitude and phase neglecting the boundaries ( b ) , and reconstructed amplitude and phase taking into account the boundaries ( c ) .
the cavity detuning is zero . *
5.- * experimental interferometric snapshot of a static domain wall ( a ) , reconstructed amplitude and phase neglecting the boundaries ( b ) , and reconstructed amplitude and phase taking into account the boundaries ( c ) .
the cavity detuning is 12% of the free spectral range ( fsr ) .
the fsr of the cavity is 120mhz . *
* plot of the trajectories of a static ( full line ) and a moving ( dashed line ) domain wall that were generated at the same spatial position .
the detuning for the static and moving wall are zero and 20% of the free spectral range , respectively . *
fig . 7.- * reconstructed amplitudes ( top ) and phases ( bottom ) corresponding to a static domain wall at different instants ( the time interval between different profiles is 5s ) .
the different location of the phase gradients ( bottom ) is artificial and has been introduced for the sake of clarity ( this applies also to fig . 8) . *
fig . 9.- * reconstructed amplitudes and phases of domain walls located at the same spatial position but obtained for different detunings .
detuning increases from left to rigth ( in percentage of the free spectral range ) : a ) 0 % , b ) 10 % and c ) 20 % .
velocity increases as detuning does : a ) @xmath9 @xmath32 , b ) @xmath33 @xmath32 , and c ) @xmath34 @xmath32 . *
10.- * logarithm of the intensity at the core of the dw , @xmath15 , versus the logarithm of the phase gradient , @xmath16 , for different instants and fixed detuning ( 20% of the free spectral range ) .
squares correspond to experimental data and the straight line correspond to the fit . *
* relation between the velocity of a moving domain wall and the chirality parameter measured during its motion in a non flat background field .
squares correspond to the experimental data and straight lines correspond to the linear fit . | we report the experimental characterization of domain walls dynamics in a photorefractive resonator in a degenerate four wave mixing configuration .
we show how the non flat profile of the emitted field affects the velocity of domain walls as well as the variations of intensity and phase gradient during their motion .
we find a clear correlation between these two last quantities that allows the experimental determination of the chirality that governs the domain walls dynamics . |
models with an extra u(1 ) are among the simplest and most natural extensions of the standard model ( sm ) .
they enjoy both the top down and bottom up motivation .
in particular , additional u(1 ) s appear in many string constructions . from the low energy perspective , the coupling between an sm fermions @xmath5 and a massive gauge boson z@xmath0 @xcite @xmath6 where @xmath7 are some constants , represents one of the dimension-4 `` portals '' ( see e.g. @xcite ) connecting the observable world to the sm singlet sector .
this is particularly important in the context of dark matter models @xcite .
if dark matter is charged under the extra u(1 ) , the above coupling provides a dm annihilation channel into visible particles .
as long as the z@xmath0 has a tev scale mass and the couplings are not too small , this framework fits the wimp miracle paradigm @xcite .
recent lhc @xcite and direct dm detection constraints @xcite , however , put significant pressure on this idea since no traces of a z@xmath0 were found in either direct collider searches or dm scattering on nuclei . in this letter , we argue that these negative results may be due to the axial nature of the z@xmath0 and its stronger coupling to dark matter compared to @xmath8 above . in this case , which we call `` axial dark matter '' ( axdm ) , dm scattering on nuclei is spin dependent and weakly constrained .
the lhc has limited sensitivity to such a z@xmath0 due to the fact that it decays predominantly into dark matter , as in @xcite . to couple
universally to sm fermions , which distinguishes the model from the leptophobic scenarios ( see e.g. @xcite ) . ]
we thus find that all of the constraints can be satisfied , which adds some credibility to the wimp paradigm .
in what follows , we consider the possibility that z@xmath0 is purely axial , with the couplings @xmath9 here @xmath5 represents the standard model ( sm ) fermions , @xmath10 is a dirac fermion constituting dark matter and @xmath11 are the corresponding z@xmath0 couplings .
this lagrangian represents an effective low energy interaction after heavy particles have been integrated out and the vector boson kinetic terms have been diagonalized .
clearly , the microscopic theory can be made anomaly
free by assigning appropriate charges to fermions ( we do not exclude the possibility of having further heavy fermions coupled to z@xmath0 ) . one may ask how natural it is to have a pure axial vector interaction . in our opinion , this choice is quite natural given the fact that the photon interaction is purely vector and the axial case is just the other extreme . also , our considerations hold in the presence of a small vector component of z@xmath0 , which may be generated through kinetic mixing @xcite . to make our model as simple as possible , we will focus on the case of a universal coupling of z@xmath0 to the sm fermions , @xmath8 .
( this assumption can of course be easily relaxed by inserting the fermion dependent charges . )
we then find that cosmological and accelerator constraints require @xmath12 by a factor of @xmath13 to @xmath14 .
one would be hesitant to attribute such a hierarchy to the difference in the observable and hidden charges .
on the other hand , factors of this type can arise in the system of two u(1 ) s mixing with each other .
consider the general lagrangian describing two massive abelian gauge bosons , @xmath15 where @xmath16 couples only to the dark sector with coupling @xmath17 , while @xmath18 couples only to the visible sector with coupling @xmath19 .
the lighter mass eigenstate would be a mixture of @xmath16 and @xmath18 , which couples to both sectors .
the hierarchy ( [ hierarchy ] ) can then be recovered in various limits .
for example , it can result from @xmath20 . for
order one kinetic mixing , @xmath21 , the z@xmath0 is composed mostly of @xmath16 and @xmath22 another possibility is to attribute ( [ hierarchy ] ) to the hierarchy in the couplings , @xmath23 . for a small kinetic mixing @xmath24 and large mass
mixing @xmath25 , the z@xmath0 is a mixture of @xmath16 and @xmath18 with @xmath26 note that for @xmath27 , the mixing is nearly maximal and the second mass eigenstate becomes heavy . in what follows
, we will be agnostic as to the origin of the hierarchy ( [ hierarchy ] ) and will treat the two couplings as free parameters .
in this section , we provide a list of cosmological and accelerator constraints on the model . these set bounds on the two couplings @xmath28 and the z@xmath0 mass @xmath29 . in order to understand their qualitative behaviour and compatibility ,
we provide simple analytic approximations for the observables . * planck / wmap and dm annihilation . *
suppose that dm is produced thermally , as in the traditional wimp scenario .
the main dark matter annihilation mechanism is the @xmath30-channel annihilation into sm fermion pairs . although we will use the exact tree
level result in our numerical analysis , it is instructive to consider the heavy z@xmath0 , @xmath31 , and zero dm velocity limit .
, while for a heavier z@xmath0 velocity dependent contributions are equally important .
we choose the limit @xmath32 for transparency of our discussion , while using the full result in our numerical analysis . ] in this case , the cross section takes on a particularly simple form , @xmath33 where @xmath34 is the number of colors for quarks and 1 for leptons .
we see that , for light final state fermions , the cross section is suppressed .
the origin of the @xmath35 factor can be understood from ( conserved ) c - parity considerations .
the c - parity of the initial state must be + 1 to match that of z@xmath0 . since for a fermion antifermion pair
it is given by @xmath36 with @xmath37 and @xmath30 being the angular momentum and spin quantum numbers , the @xmath30-wave initial state ( @xmath38 ) must then have @xmath39 . on the other hand , the helicities of the relativistic final state fermions
add up to 1 .
hence , a spin flip is required leading to the @xmath40 dependence .
note however that , for heavy fermions like the top quark , this factor does not lead to significant suppression of the amplitude .
suppose that dm is sufficiently heavy such that its pair annihilation into top quarks is allowed .
then for @xmath41 , the canonical wimp annihilation cross section @xmath42 @xmath43s@xmath44 translates into @xmath45 one should keep in mind that this figure indicates the ballpark of the result and the velocity- as well as @xmath46dependent contributions affect @xmath47 , while the definitive answer is given by our numerical analysis .
+ * direct dm detection .
* tree level z@xmath0 exchange leads to spin dependent dm scattering on nuclei , which is constrained by a number of experiments .
for an ( approximately ) universal z@xmath0 coupling to quarks , @xmath48 where @xmath49 is the nucleon mass and @xmath50 are the quark contributions to the proton spin : @xmath51 , @xmath52 , @xmath53 .
taking @xmath54 cm@xmath4 as the benchmark bound on @xmath56 for @xmath57 gev @xcite , one finds @xmath58 this bound is satisfied automatically for thermally produced dark matter ( see eq .
( [ annih ] ) ) .
spin independent dm scattering is generated at one loop with the corresponding amplitude being suppressed both by a loop factor and the quark masses required by a helicity flip .
the resulting bound is weak @xcite .
we note that similar conclusions apply to the dm
nucleon interaction mediated by a pseudoscalar as recently studied in @xcite .
+ + * lep bounds . *
dark matter with @xmath59 can not be produced on shell at lep . however , there are still significant constraints on z@xmath0 due to the effective operators @xmath60 for various fermions @xmath61 and @xmath62 .
these operators are constrained by the precise measurements of the cross sections and angular distributions of the final state fermions . in the axial case ,
the resulting bound is @xcite @xmath63 comparing this to eq .
( [ annih ] ) , one finds that z@xmath0 couples much stronger to dm than it does to sm fermions , @xmath64 + + * perturbativity .
* as is clear from the above discussion , the dm z@xmath0 coupling can become quite strong .
then , our approximation is controllable only if @xmath65 we do not impose further constraints on the position of the landau pole of the coupling as we expect our model to be uv completed already in the multi tev range
. + + * dilepton and monojet lhc bounds .
* at the lhc , both dark matter and z@xmath0 can be produced on shell , which leads to strong bounds from cms and atlas .
the most important constraint is due to searches for dileptons with a large invariant mass .
we will use the cms z@xmath0 analysis of 3.6 fb@xmath44/8 tev and 5 fb@xmath44/7 tev @xcite as our benchmark constraint .
the result is summarized in fig.2 ( upper right panel ) of that paper .
for a sequential sm z@xmath0 ( ssm ) , that is having the same couplings as the standard model z - boson , the exclusion limit is around 2.5 tev . to adapt the results to our case
, one must take into account the difference in the z@xmath0 couplings as well as the reduced branching ratio for z@xmath0 decay into visible fermions , @xmath66 for our estimates it suffices to approximate @xmath67 by its ( universal ) axial component , @xmath68 .
the branching ratio for z@xmath0 decay into sm fermions is @xmath69 where @xmath70 accounts for the kinematic suppression in an axial
vector decay ( see e.g. @xcite ) .
these factors result in the dependence of the number of expected @xmath71 events on @xmath8 and @xmath72 .
the constraints on z@xmath0 relax significantly as @xmath8 decreases and @xmath73 as light as 500 gev becomes allowed given it decays predominantly invisibly . to estimate the resulting lhc bound on @xmath73 , we analytically approximate the @xmath71 production cross section in fig.2 of @xcite and calculate how much it should be reduced to comply with its experimental bound .
we find that the result can be cast in the form @xmath74 $ ] , for @xmath75 in tev and @xmath76 being an @xmath73range dependent constant : @xmath77 tev for @xmath78 tev .
for instance , a 500 gev z@xmath0 becomes allowed if the @xmath71 cross section reduces by about 3 orders of magnitude , whereas a 2.5 tev z@xmath0 is allowed with no suppression required .
z@xmath0 models are also constrained by monojet events with large missing energy , which is due to z@xmath0 decay into dark matter .
the atlas analysis of 10.5 fb@xmath44/8 tev data @xcite imposes the bound on the axial vector interaction ( d8operator of @xcite ) , @xmath79 for @xmath80 gev ( and a weaker bound for heavier dm ) .
inclusion of on - shell effects does not make the constraint significantly stronger @xcite . as a result , similarly to the direct dm detection constraint , it is satisfied when eq .
( [ annih ] ) is imposed . + + * combined constraints . *
the above estimates serve to single out the most important constraints , whose compatibility is to be analyzed .
we see that , once the correct dm relic abundance is imposed , the main factors restricting available parameter space are the lhc dilepton bound and perturbativity . indeed , the lhc constraint can always be satisfied by decreasing @xmath8 , which according to eq .
( [ annih ] ) increases @xmath81 until it hits the perturbative bound ( [ pert ] ) .
we find that all of the constraints are in fact compatible .
for instance , at @xmath82 gev , the allowed range of @xmath83 spans about two orders of magnitude , from @xmath13 to @xmath14 .
to go further , let us remind the reader that our estimate of the dm relic abundance constraint ( [ annih ] ) is rather simplistic .
it does not take into account resonant effects nor those due to thermal averaging .
the correct treatment is provided by the numerical package micromegas @xcite . using this tool , we find the same qualitative conclusion : all of the constraints are compatible . in our numerical study
, we impose the condition is very small and the z@xmath0 production cross section is suppressed altogether . ]
@xmath84 which allows for the z@xmath0 decay into dark matter and amounts to `` invisibility '' of the former .
two representative results are shown in fig . 1 .
in the left panel , we set @xmath85 gev , @xmath86 and scan parameter space @xmath87 satisfying the planck constraint @xcite .
as @xmath73 increases from @xmath88 , so does @xmath81 . in this region ,
the effect of resonant annihilation is important .
the resonance is quite broad , on the order of tens of gev , due to the thermal smearing .
away from the resonance , @xmath81 quickly turns non perturbative .
the lhc constraint excludes part of the parameter space close to the threshold , where the invisible z@xmath0 decay is inefficient .
further constraints from lep , direct dm detection and monojets are satisfied automatically in this panel .
the result is that the @xmath73 range 520 - 560 gev is allowed , while @xmath81 varies by two orders of magnitude , from 0.05 to 5 .
for heavier dm , the lhc bound becomes less severe and the coupling @xmath8 is allowed to be larger . for example , in the right panel of fig . 1
, the cms constraint is satisfied everywhere due to the suppressed z@xmath0 production with @xmath89 . the @xmath73 range satisfying planck extends over hundreds of gev .
the pattern observed in this figure is quite general : the allowed parameter space is not far from the resonance region , @xmath90 , with the latter being relatively broad , @xmath91 .
the lep , direct dm detection and monojet constraints are satisfied automatically in the region of interest , while the dilepton lhc bound cuts out part of the parameter space .
the dark matter candidate , axdm , belongs to the general wimp category as it has a tev scale mass and couplings in the range @xmath92
. a detailed scan of axdm parameter space is reserved for a subsequent publication .
in this paper , we have explored a very simple scenario in which dark matter couples to the sm fermions via an axial z@xmath0 ( `` axial dark matter '' ) .
the model is characterized by 2 couplings as well as the z@xmath0 and dm masses .
if the z@xmath0 couples much stronger to the dark sector compared to the visible sector , which may be due to a mixing of two u(1 ) s , all the phenomenological constraints can be satisfied .
in particular , the lhc constraints are loose due to invisible z@xmath0 decay and allow for @xmath73 as low as 500 gev , while dm scattering on nuclei is spin dependent and thus weakly constrained .
the correct dm relic density is obtained in regions not far from the resonance , @xmath90 .
all in all , we find that axdm is consistent with the thermal wimp paradigm .
* acknowledgements . *
the authors would like to thank g. arcadi , b. zaldivar , m. tytgat and k. tarachenko for very useful discussions .
this work was supported by the french anr tapdms * anr-09-jcjc-0146 * and the spanish micinn s consolider - ingenio 2010 programme under grant multi - dark * csd2009 - 00064*. y.m .
acknowledges partial support from the european union fp7 itn invisibles ( marie curie actions , pitn- ga-2011- 289442 ) and from the erc advanced grant higgs@lhc .
k. cheung and t. -c .
yuan , jhep * 0703 * , 120 ( 2007 ) ; y. mambrini , jcap * 1107 * ( 2011 ) 009 ; v. barger , d. marfatia and a. peterson , phys . rev .
d * 87 * , 015026 ( 2013 ) ; e. dudas , l. heurtier , y. mambrini and b. zaldivar , jhep * 1311 * , 083 ( 2013 ) .
s. chatrchyan _ et al . _
[ cms collaboration ] , phys .
b * 720 * , 63 ( 2013 ) .
g. aad _ et al .
_ [ atlas collaboration ] , jhep * 1211 * , 138 ( 2012 ) .
d. s. akerib _ et al .
_ [ lux collaboration ] , arxiv:1310.8214 [ astro-ph.co ] .
b. holdom , phys .
b * 166 * , 196 ( 1986 ) .
e. aprile _ et al .
_ [ xenon100 collaboration ] , phys .
lett . * 111 * , no . 2 , 021301 ( 2013 ) ; s. archambault _ et al .
_ [ picasso collaboration ] , phys .
b * 711 * , 153 ( 2012 ) ; e. behnke _ et al .
_ [ coupp collaboration ] , phys . rev .
d * 86 * , 052001 ( 2012 ) .
[ atlas collaboration ] , `` search for new phenomena in monojet plus missing transverse momentum final states using 10fb-1 of pp collisions at sqrts=8 tev with the atlas detector at the lhc , '' atlas - conf-2012 - 147 . | we consider the possibility that fermionic dark matter ( dm ) interacts with the standard model fermions through an axial z@xmath0 boson .
as long as z@xmath0 decays predominantly into dark matter , the relevant lhc bounds are rather loose .
direct dark matter detection does not significantly constrain this scenario either , since dark matter scattering on nuclei is spin dependent . as a result , for a range of the z@xmath0 mass and couplings ,
the dm annihilation cross section is large enough to be consistent with thermal history of the universe . in this framework ,
the thermal wimp paradigm , which currently finds itself under pressure , is perfectly viable . lpt - orsay-14 - 14 = 20pt * oleg lebedev@xmath1 and yann mambrini@xmath2 * + @xmath3__department of physics and helsinki institute of physics , gustaf hllstrmin katu 2 , fin-00014 university of helsinki , finland _ _
+ @xmath4__laboratoire de physique th ' eorique universit ' e paris - sud , f-91405 orsay , france _ _ |
in the last two decades , the phenomenon of chaos synchronization has been extensively studied in coupled nonlinear dynamical systems from both theoretical and application perspectives due to its significance in diverse natural and man - made systems @xcite . in particular ,
various types of synchronization , namely complete synchronization ( cs ) , phase synchronization ( ps ) , intermittent lag / anticipatory synchronizations and generalized synchronization ( gs ) have been identified in coupled systems .
all these types of synchronization have been investigated mainly in identical systems and in systems with parameter mismatch .
very occasionally it has been studied in distinctly nonidentical ( structurally different ) systems .
but in reality , structurally different systems are predominant in nature and very often the phenomenon of synchronization ( gs ) is responsible for their evolutionary mechanism and proper functioning of such distinctly nonidentical systems .
as typical examples , we may cite the cooperative functions of brain , heart , liver , lungs , limbs , etc .
, in living systems and coherent coordination of different parts of machines , between cardiovascular and respiratory systems @xcite , different populations of species @xcite , in epidemics @xcite , in visual and motor systems @xcite , in climatology @xcite , in paced maternal breathing on fetal @xcite etc
. it has also been shown that gs is more likely to occur in spatially extended systems and complex networks ( even in networks with identical nodes , due to the large heterogeneity in their nodal dynamics ) @xcite .
in addition , gs has been experimentally observed in laser systems @xcite , liquid crystal spatial light modulators @xcite , microwave electronic systems @xcite and has applications in secure communication devices @xcite . therefore understanding the evolutionary mechanisms of many natural systems
necessitates the understanding of the intricacies involved in the underlying generalized synchronization ( gs ) phenomenon .
the phenomenon of gs has been well studied and understood in unidirectionally coupled systems @xcite , but still it remains largely unexplored in mutually coupled systems .
only a limited number of studies are available on gs in mutually coupled systems even with parameter mismatches @xcite and rarely in structurally different dynamical systems with different fractal dimensions @xcite .
recent investigations have revealed that gs emerges even in symmetrically ( mutually ) coupled network motifs in networks of identical systems , and that it also plays a vital role in achieving coherent behavior of the entire network @xcite . as almost all natural networks
are heterogeneous in nature , the notion of gs has been shown to play a vital role in their evolutionary mechanisms @xcite . thus to unravel the role of gs in such large networks , it is crucial to understand the emergence of gs in heterogeneous network motifs composed of distinctly different nonidentical systems .
it is to be noted that the notion of ps has been widely investigated in mutually coupled essentially different ( low - dimensional ) chaotic systems @xcite , while the notion of gs in such systems has been largely ignored .
it is an accepted fact that ps is weaker than gs ( because ps does not add restrictions on the amplitude , and only the locking of the phases is crucial ) .
parlitz et al .
have shown that in general gs always leads to ps , if one can suitably define a phase variable , and that gs is stronger ( which has been studied in unidirectionally coupled chaotic systems ) .
that is , ps may occur in cases where the coupled systems show no gs @xcite .
further , the transition from ps to gs as a function of the coupling strength has been demonstrated in coupled time - delay systems with parameter mismatch which confirms that ps is weaker than gs @xcite . on the other hand , zhang and hu @xcite have demonstrated that gs is not necessarily stronger than ps , and in some cases ps comes after gs with increasing coupling strength depending upon the degree of parameter mismatch .
they have concluded that ps ( gs ) emerges first for low ( high ) degree of parameter mismatch and that they both occur simultaneously for a critical range of mismatch in low - dimensional systems @xcite . in general , the notion of gs and its relation with ps in mutually coupled systems , particularly in distinctly nonidentical systems with different fractal dimensions including time - delay systems , need much deeper understanding . in line with the above discussion , we have reported briefly the existence of gs in symmetrically coupled networks of distinctly nonidentical time - delay systems using the auxiliary system approach in a letter @xcite . in this paper
, we will provide an extended version of the letter with important additional results and explanations .
in particular , in this paper we will demonstrate the occurrence of a transition from partial to global gs in mutually coupled networks of structurally different time - delay systems ( for @xmath0 and @xmath1 ) with different fractal ( kaplan - yorke ) dimensions and in systems with different orders using the auxiliary system approach and the mutual false nearest neighbor ( mfnn ) method .
we use the mackey - glass ( mg ) @xcite , a piecewise linear ( pwl ) @xcite , a threshold piecewise linear ( tpwl ) @xcite and the ikeda time - delay @xcite systems to construct heterogeneous network motifs . the main reason to consider time - delay systems in this study
is that even with a single time - delay system , one has the flexibility of choosing systems with different fractal dimensions just by adjusting their intrinsic delay alone , which is a quite attracting feature of time - delay systems from modelling point of view @xcite .
further , time - delay systems are ubiquitous in several real situations , including problems in ecology , epidemics , physiology , physics , economics , engineering and control systems , @xcite which inevitably require delay for a complete description of the system ( note that intrinsic delay is different from connection delays which arise between different units due to finite signal propagation time ) . in our present work ,
we report that there exists a common gs manifold even in networks of distinctly different time - delay systems .
in other words , there exists a functional relationship even for systems with different fractal dimensions , which maps them to a common gs manifold .
further , we also wish to emphasize that our results are not confined to just scalar one - dimensional time - delay systems alone but we confirm that there exists a similar type of synchronization transitions even in the case of time - delay systems of different orders .
particularly , we demonstrate that synchronization phenomenon occurs in a system of a ikeda time - delay system ( first order time - delay system ) mutually coupled with a hopfield neural network ( a second order time - delay system ) , and in a system of a mg time - delay system ( first order time - delay system ) mutually coupled with a plankton model ( a third order system with multiple delays ) to establish the generic nature of our results .
stability of gs manifold in unidirectionally coupled systems is usually determined by examining the conditional lyapunov exponents of the synchronization manifold @xcite or the lyapunov exponents of the coupled system itself @xcite .
here , we will estimate the maximal transverse lyapunov exponent ( mtle ) to determine the asymptotic stability of the cs manifold of each of the systems with their corresponding auxiliary systems starting from different initial conditions , which in turn asserts the stability of gs between the original distinctly nonidentical time - delay systems .
further , we will also estimate the cross correlation ( cc ) and the correlation of probability of recurrence ( cpr ) to establish the relation between gs and ps ( here gs and ps always occur simultaneously in structurally different time - delay systems ) .
cc essentially gives a much better statistical average of the synchronization error , which is being widely studied to characterize cs .
further , cpr is a recurrence quantification tool @xcite , which effectively characterizes the existence of ps especially in highly non - phase - coherent hyperchaotic attractors usually exhibited by time - delay systems @xcite .
it is also to be noted that the auxiliary system approach has some practical limitations .
this method fails for systems whose dynamical equations are not known .
particularly , cs between response and auxiliary systems arises only when their initial conditions are set to be in the same basin of attraction . due to the above limitations of the auxiliary system approach
, we have also calculated the mutual false nearest neighbor ( mfnn ) which doubly confirms our results .
in addition , an analytical stability condition using the krasovskii - lyapunov theory is also deduced in suitable cases .
the remaining paper is organized as follows : in sec .
[ sec2 ] , we will describe briefly the notion of structurally different time - delay systems with different fractal dimensions with examples . in secs .
[ sec3 ] and [ sec4 ] , we will demonstrate the existence of a transition from partial to global gs in @xmath2 mutually coupled time - delay systems using the auxiliary system approach and the mfnn method , respectively .
an analytical stability condition is also deduced using the krasovskii - lyapunov theory in sec .
the transition from partial to global gs is demonstrated in @xmath3 mutually coupled time - delay systems with an array and a ring coupling configuration in sec .
[ sec5 ] . in sec .
[ sec6 ] , we will consider @xmath4 mutually coupled time - delay systems with array , ring , global and star configurations and discuss the occurrence of partial and global gs . the above synchronization transition in time - delay systems with different orders
is demonstrated in sec .
[ sec7 ] and finally we summarize our results in sec .
in this section , we consider structurally different first order scalar time - delay systems with different fractal dimensions . here
, structurally different time - delay systems refer to systems exhibiting chaotic / hyperchaotic attractors with different phase space geometry characterized by different degree of complexity . despite the similarity in the structure of the evolution equations ,
the nature of chaotic attractors , the number of positive les and their magnitudes characterizing the rate of divergence and the degree of complexity as measured by the kaplan - yorke dimension ( @xmath5 ) of the underlying dynamics are different even for the same value of time - delay because of the difference in the nonlinear functional form . as an illustration ,
first let us consider a symmetrically coupled arbitrary network of distinctly nonidentical scalar time - delay systems .
then the dynamics of the @xmath6th node in the network is represented as @xmath7 where @xmath8 , and @xmath9 is the number of nodes in the network , @xmath10 s and @xmath11 s are system s parameters , @xmath12 s the time - delays , the nonlinear function of the @xmath6th node is defined as @xmath13 , @xmath14 is the coupling strength and @xmath15 a laplacian matrix which determines the topology of the arbitrary network . for the mg time - delay system ,
we choose the nonlinear function defined as @xcite @xmath16 and for the pwl system with the nonlinear function given as @xcite @xmath17 for the tpwl system we choose the form of the nonlinear function as given by @xcite @xmath18 with @xmath19 and for the ikeda time - delay system the nonlinear function is given by @xcite @xmath20 hyperchaotic attractors of ( a ) mackey - glass,(b ) piecewise linear ( pwl ) , ( c ) piecewise linear with threshold nonlinearity ( tpwl ) and ( d ) ikeda time - delay systems for the choice of parameters given in table [ table1 ] . ]
the parameter values for the above time - delay systems are chosen throughout the paper as follows : + 1 ) for the mg systems : we choose @xmath21 , @xmath22 , and @xmath23 ; + 2 ) for the pwl systems : we choose @xmath24 , @xmath25 , @xmath26 , @xmath27 and @xmath28 ; + 3 ) the parameter values for the tpwl are fixed as @xmath29 , @xmath30 , @xmath31 , @xmath32 , @xmath33 and @xmath34 and + 4 ) for the ikeda time - delay system we choose @xmath35 , @xmath36 , @xmath37 .
the hyperchaotic attractors of the uncoupled mg , pwl , tpwl and ikeda time - delay systems are depicted in figs .
[ fig1](a)-(d ) , respectively .
the first few largest les of all the above four ( uncoupled ) time - delay systems are shown as a function of the delay time @xmath38 in figs .
[ fig2](a)-(d ) .
it is also clear from this figure that the number of positive les , and hence the complexity and dimension of the state space , generally increase with the delay time .
further , the degree of complexity , measured by their number of positive les of the dynamics ( attractors ) exhibited by all the four systems are distinctly different even for the same value of time - delay .
in fact , we have taken different values of delay for each of the systems , as pointed out by the arrows in fig .
[ fig2 ] , to demonstrate the existence of suitable smooth transformation that maps the strongly distinct individual systems to a common gs manifold .
[ table1 ] we wish to emphasize especially the structural difference , measured by their degree of complexity , between the hyperchaotic attractors of different scalar first order time - delay systems ( fig .
[ fig1 ] ) which we have studied in this paper and are detailed in table [ table1 ] : 1 ) the mg system has two positive les with @xmath39 for @xmath40 , 2 ) the pwl time - delay system has three positive les with @xmath41 for @xmath42 , 3 ) the tpwl time - delay system has four positive les with @xmath43 for @xmath44 and 4 ) the ikeda system has five positive les with @xmath45 for @xmath46 .
the above facts clearly indicate that the real state space dimension explored by the flow of a time - delay system , which is essentially infinite - dimensional in nature , and the associated degree of complexity are characterized by the form of nonlinearity and the value of delay time irrespective of the similarity of the underlying evolution equations of the scalar first order time - delay systems .
first few largest lyapunov exponents of ( a ) mackey - glass,(b ) piecewise linear ( pwl ) , ( c ) piecewise linear with threshold nonlinearity ( tpwl ) and ( d ) ikeda time - delay systems , as a function of the time - delay .
arrows point the value of the delay time we have considered in our analysis . ]
it has already been shown that the functional relationship between two different systems in gs is generally difficult to identify analytically .
however gs in such systems can be characterized numerically using various approaches , namely the mutual false nearest neighbor method @xcite , the statistical modeling approach @xcite , the phase tube approach @xcite , the auxiliary system approach @xcite , etc . among all these methods
the auxiliary system approach is extensively used to detect the presence of gs in unidirectionally coupled systems ( both in numerical and experimental studies due to its simple and powerful implementation ) .
abarbanel et al .
@xcite first introduced this approach to characterize and confirm gs in dynamical systems ( when the system equations are known ) .
the mathematical formulation of this concept was put forward by kocarev and parlitz @xcite for a drive - response configuration in low - dimensional systems .
they formulated it as the asymptotic convergence of the response and its auxiliary systems which are identically coupled to the drive system , starting from two different initial conditions from the same basin of attraction .
the asymptotic convergence indeed ensures the existence of an attracting synchronization manifold ( cs manifold between the response and auxiliary systems and gs manifold between the drive and response systems ) @xcite . in other words ,
gs between the drive @xmath47 and the response @xmath48 systems occur only when the response system is asymptotically stable , that is @xmath49 in the basin of the synchronization manifold one requires @xmath50 .
now , we will provide an extension of this approach to a network for mutually coupled systems . for simplicity
, we consider two mutually coupled nonidentical distinctly different time - delay systems represented by @xmath51 [ eqn2a ] where @xmath52 , @xmath53 ( @xmath54 ) and @xmath55 .
@xmath56 correspond to the driving signals .
system ( [ eqn2a ] ) is in gs if there exists a transformation @xmath57 such that the trajectories of the systems ( [ eqn2a]a ) and ( [ eqn2a]b ) are mapped onto a subspace ( synchronization manifold ) of the whole state space @xmath58 .
that is @xmath59 we also note here that since we are dealing with gs of nonidentical systems with different fractal dimensions , the transformation function @xmath57 refers to a generalized transformation ( not identity transformation ) and also there may exist a set of transformations @xmath57 that maps a given @xmath60 and @xmath61 to different subspaces of @xmath62 .
this indicates that the synchronization manifold @xmath63 is such that @xmath64 $ ] , which lies within a subset of the basin of attraction @xmath65 , approaches @xmath66 , so that @xmath67 is an attracting manifold . here
@xmath68 and @xmath69 are the basins of attraction of systems ( [ eqn2a]a ) and ( [ eqn2a]b ) , respectively .
the synchronization manifold @xmath67 could also be @xmath70 as special cases , but without ambiguity @xmath71 is the most general one for mutually coupled systems .
hence , gs exists between the systems ( [ eqn2a]a ) and ( [ eqn2a]b ) only when both coupled systems are asymptotically stable .
that is , @xmath72\subset b,~i=1,2 $ ] , one requires @xmath73 further , it is worth to emphasize that in ref .
@xcite , it is reported that a subharmonic entrainment takes place when there exists a relation between the interacting systems , which usually takes place for periodic synchronization with @xmath74 periods , @xmath75 .
but in this paper , the synchronization dynamics in all the cases we have considered exhibits chaotic / hyperchaotic oscillations and hence the transformation function @xmath76 refers to the existence of a function ( not a relation ) in our case . therefore the trajectories of eq .
( [ eqn2a ] ) starting from the basin of attraction @xmath77 asymptotically reach the synchronization manifold @xmath67 defined by the transformation function @xmath78 , which can be smooth if the systems ( [ eqn2a ] ) uniformly converge to gs manifold ( otherwise nonsmooth ) . the uniform convergence ( smooth transformation )
is confirmed by negative values of their local lyapunov exponents of the synchronization manifold @xmath67 @xcite .
schematic diagrams of the auxiliary system approach for networks of mutually coupled systems .
( a , b , d ) linear arrays with @xmath79 , ( c , e ) ring configurations with @xmath80 , and ( f , g ) global and star coupling configurations with @xmath4 . ]
now , we will demonstrate the existence of gs in symmetrically coupled arbitrary networks of distinctly nonidentical time - delay systems with different fractal dimensions using the auxiliary system approach .
we consider a symmetrically coupled arbitrary network as given in eq .
( [ eqn1 ] ) . to determine the asymptotic stability of each of the nodes in this network
, one can define a network ( auxiliary ) identical to eq .
( [ eqn1 ] ) ( starting from different initial conditions in the same basin of attraction ) , whose node dynamics is represented as @xmath81 the parameter values are the same as in eq.([eqn1 ] ) discussed in sec .
[ sec2 ] . in the following sections
, we will investigate the existence of transition from partial gs to global gs in networks of @xmath82 and @xmath1 systems with a linear array , ring , global and star coupling configurations . to start with , in order to demonstrate the existence of gs between two ( @xmath2 ) mutually coupled nonidentical distinctly different time - delay systems , we consider the network given in fig .
[ fig1a](a ) . here 1 and 2 are mutually coupled distinctly nonidentical time - delay systems and 1@xmath83 and 2@xmath83 are the associated auxiliary systems .
[ fig1a ] also depicts schematically different configurations for @xmath84 , which are discussed later on .
the state equations for @xmath2 coupled systems can be represented as @xmath85 now , let us couple the auxiliary system 1@xmath83 to 2 and 2@xmath83 to 1 unidirectionally as given in fig .
[ fig1a](a ) , such that the systems 1 and 1@xmath83 are driven by the same signal from system 2 , and systems 2 and 2@xmath83 are similarly driven by system 1 .
the corresponding dynamical equation for the auxiliary systems can be given as @xmath86 we choose the mg time - delay systems ( system 1 and 1@xmath83 ) with the nonlinear function @xmath87 given in eq .
( [ eqn1a ] ) and the pwl systems ( system 2 and 2@xmath83 ) with the nonlinear function @xmath88 given in eq .
( [ eqn1b ] ) .
the parameters of both systems are fixed as given in sec .
[ sec2 ] ( table [ table1 ] ) and for those parameter values both systems exhibit hyperchaotic attractors [ figs .
[ fig1](a ) and [ fig1](b ) ] with two [ fig . [ fig2](a ) ] and three [ fig .
[ fig2](b ) ] positive les , respectively . generally , in mutual coupling configuration the systems affect each other and attain a common synchronization manifold simultaneously above a threshold value of the coupling strength .
but interestingly in distinctly different coupled time - delay systems with different fractal dimensions , one of the systems first reaches the gs manifold for a lower value of @xmath14 , while the other system remains in a desynchronized state , which we call as a partial gs state . for a further increase of @xmath14
both systems organize themselves and enter into a common gs manifold , thereby achieving a global gs . in other words ,
when system 1 and 1@xmath83 are identically synchronized , system 1 is synchronized to a subspace ( synchronization manifold ) of the whole state space @xmath89 of both systems in a generalized sense , which we call as a partial gs .
similarly , when the systems 2 and 2@xmath83 are synchronized identically , then system 2 is synchronized to the common synchronization manifold .
this corroborates that both systems 1 and 2 share a common gs manifold in the phase space @xmath90 .
thus , when both auxiliary systems are completely synchronized with their original systems for an appropriate coupling strength , then there exists a function that maps systems 1 and 2 to the common ( global ) gs manifold .
( a ) mtles and ( b ) cc , cpr of the main and auxiliary systems for two mutually coupled mg - pwl systems as a function of @xmath91 for @xmath2 ( [ 2coup ] ) . ] in order to characterize the transition from partial to global gs and to evaluate the stability of the cs of each of the main and auxiliary systems , we have calculated the mtles of the main and auxiliary systems which in turn ensure the stability of gs manifold between the original systems .
we have also estimated the correlation coefficient ( cc ) of each of the main and the associated auxiliary systems , given by @xmath92 where the @xmath93 brackets indicate temporal average . if the two systems are in cs state , the correlation coefficient @xmath94 , otherwise @xmath95 .
further , the existence of ps ( between the main and auxiliary systems ) can be characterized by the value of the index cpr .
if the phases of the coupled systems are mutually locked , then the probability of recurrence is maximal at a time and cpr @xmath96 .
in contrast , one can expect a drift in the probability of recurrence resulting in low values of cpr characterizing the degree of locking between the coupled systems @xcite .
the coupled equations ( [ eqn1 ] ) and ( [ eqn2 ] ) are integrated using the runge - kutta fourth order method .
the mtles are the largest lyapunov exponents of the evolution equation of @xmath97 .
the lyapunov exponents are calculated using j. d. farmer s approach @xcite . in fig .
[ fig5 ] , we have plotted the various characterizing quantities based on our numerical analysis .
the red ( light gray ) continuous line in fig .
[ fig5](a ) shows the mtle ( @xmath98 ) of the mg systems @xmath99 and the blue ( dark gray ) dotted line depicts the mtle ( @xmath100 ) of the pwl systems @xmath101 as a function of the coupling strength .
figure [ fig5](b ) shows the cc and cpr of the main and auxiliary mg time - delay systems as red ( light gray ) filled and open circles , respectively , and the cc and cpr of the pwl systems are represented by the blue ( dark gray ) filled and open triangles .
initially , for @xmath102 , both @xmath103 and @xmath104 are nearly zero , indicating the desynchronized state when both @xmath98 and @xmath105 which confirm that cs ( gs ) is unstable .
if we increase the coupling strength , @xmath104 and @xmath106 start to increase towards unity and at @xmath107 , @xmath108 ( @xmath109 ) , where @xmath110 , which confirm the simultaneous existence of gs and ps in the pwl system , while the mg system continues to remain in a desynchronized state ( @xmath111 and @xmath112 ) ( partial gs ) .
further , if we increase the coupling strength to a threshold value @xmath113 , a global gs occurs where both @xmath103 and @xmath104 become unity and @xmath98 and @xmath100 become negative .
the transition of the mtle of the auxiliary and its original systems from positive to negative values as a function of the coupling strength strongly confirms the existence of an attracting manifold . to be more clear , for the value of the coupling strengths in range of global gs ,
a negative value of the mtle assures the convergence of the perturbed trajectories in the synchronization manifold ( cs between the main and auxiliary systems and gs manifold between the main systems ) .
the convergence corroborates the attracting nature of the synchronization manifold .
further , normally one may expect that the systems with lower dynamical complexity will converge to gs manifold first , followed by the system with higher dynamical complexity @xcite .
but in our studies we encounter a contrary behavior , where the pwl system with three positive les reaches the gs manifold first ( at @xmath114 ) and then the mg system ( with two positive les ) converges to the gs manifold at @xmath115 confirming the existence of a transition from partial to global gs in distinctly nonidentical time - delay systems .
( a , b ) the magnitude of difference in the trajectories between the systems ( @xmath116 ) for @xmath117 , and ( c , d ) for @xmath118 for @xmath2 mutually coupled mg and pwl systems . ]
( a - c ) the phase portraits of the systems ( @xmath119 , @xmath120 ) , ( @xmath121 , @xmath122 ) and ( @xmath119 , @xmath121 ) for @xmath117 , and ( d - f ) for @xmath118 for @xmath2 . ] we have also numerically computed the synchronization error ( @xmath123 ) and phase projection plots , which are depicted in figs .
[ fig4 ] and [ fig3 ] , respectively . in the absence of
the coupling all the systems evolve with their own dynamics . if we slowly increase the coupling strength the main and the auxiliary pwl systems become completely synchronized for @xmath124 .
the synchronization error @xmath125 and the linear relation between the systems @xmath101 in fig .
[ fig4](b ) and in fig .
[ fig3](b ) ( plotted for @xmath126 ) , respectively , confirm that the systems @xmath121 and @xmath127 are in a cs state , whereas the mg systems @xmath99 remain desynchronized as confirmed by the phase projection [ fig .
[ fig3](a ) ] for the same value of the coupling strength .
we also note here that for this value of coupling strength the systems @xmath119 and @xmath121 show certain degree of correlation as depicted in fig .
[ fig3](c ) .
if we increase @xmath14 further , the systems ( @xmath128 ) and ( @xmath129 ) reach the cs manifold ( for @xmath130 ) and one may expect that both @xmath119 and @xmath121 attain the common gs manifold , which we call as global gs . both synchronization errors @xmath131 and @xmath132 become zero as shown in figs .
[ fig4](c ) and [ fig4](d ) for @xmath133 .
this fact confirms the existence of a global gs state .
further , figs .
[ fig3](d ) and [ fig3](e ) show a linear relation between the systems ( @xmath134 ) and ( @xmath135 ) , respectively , which additionally confirms the existence of global gs .
the degree of correlation in the phase space between the systems @xmath119 and @xmath121 for the global gs state is depicted in fig .
[ fig3](f ) for the same value of coupling strength .
phase diagram in the ( @xmath136 ) plane for two mutually coupled mg - pwl systems showing partial ( blue / dark gray ) , global ( light gray ) gs and desynchronization ( white ) regimes . ] to obtain a global picture on the occurrence of transition from partial to global gs between the mg and pwl systems , we have plotted the values of @xmath137 as a 2-parameter diagram in the ( @xmath136 ) plane .
we have fixed the parameter values of the mg systems ( as given in sec .
[ sec2 ] ) and vary one of the parameter ( @xmath138 ) of the pwl system as a function of the coupling strength ( @xmath139 ) as depicted in fig .
the white region indicates the desynchronized state and the blue ( dark gray ) region corresponds to the partial gs region , where only one of the mutually coupled systems has reached the common gs manifold as indicated by the unit value of the @xmath137 .
the global gs is represented by light gray where both coupled systems are in gs manifold as confirmed by the unit value of @xmath137 of both systems .
we have also analytically investigated the existence of the transition from partial to global gs using the krasovskii - lyapunov functional theory .
in this connection , we consider the difference between the state variables ( synchronization error ) of the main and the associated auxiliary systems ( @xmath140 ) . for small values of @xmath141 , the evolution equation for the cs manifold of the mg systems ( @xmath142 ) can be written as @xmath143 \delta_{1\tau_{1 } } \label{eqn6}\ ] ] where @xmath144 and @xmath145 .
the cs manifold is locally attracting if the origin of eq .
( [ eqn6 ] ) is stable . from the krasovskii - lyapunov theory ,
a continuous positive definite lyapunov functional can be defined as @xmath146 where @xmath147 is an arbitrary positive parameter .
the lyapunov function @xmath148 approaches zero as @xmath149 .
the derivative of the above equation ( [ eqn6a ] ) can be written as @xmath150 \delta^{2}_{1}+\beta_{1 } [ f^{\prime}_{1}(x_{\tau_{1}})]\delta_{1 } \delta_{1\tau_{1}}+\mu\delta^{2}_{1}-\mu\delta_{1\tau_{1}}^{2 } , \label{eqn6b}\ ] ] which should be negative for stability of the cs manifold , that is @xmath151 .
this requirement results in a condition for stability as @xmath152 from eq .
( [ eqn6c ] ) , one can see that the stability condition depends on the nonlinear function @xmath153 of the individual systems .
hence from the form of the nonlinear function of the mg system [ eq .
( [ eqn1a ] ) ] , we have @xmath154 consequently , the stability condition becomes , @xmath155 it is not possible to find the exact value of the nonlinear function @xmath156 as it depends on the value of the variable @xmath157 .
however , it is possible to find the value of @xmath158 by identifying the maximal value of @xmath159 using the lyapunov - razumikin function ( which is a special class of the krasovskii - lyapunov theory ) . from the hyperchaotic attractor of the mg system , for the chosen parameter values
, one can identify the maximum value of @xmath160 .
so the stability condition becomes @xmath161 ( a sufficient condition for the global gs ) . from fig .
[ fig5 ] , the threshold value of the coupling strength to attain gs in the mg systems is @xmath162 for which the stability condition is satisfied .
now , one can find the evolution equation for the cs manifold corresponding to the pwl time - delay system ( @xmath163 ) as @xmath164 \delta_{2\tau_{2}}$ ] and with a similar procedure as above , the stability condition for the pwl systems becomes @xmath165 . now from the form of the piecewise linear function @xmath88 in eq .
( [ eqn1b ] ) , we have @xmath166 consequently , the stability condition becomes @xmath167 , corresponding to the inner regime of the nonlinear function where most of the system dynamics is confined for the asymptotic cs state @xmath168 . from fig .
[ fig5 ] , one can find that cs between the pwl systems occurs for a coupling strength @xmath169 , satisfying the sufficient stability condition for partial gs , @xmath170 .
next , we use the mutual false nearest neighbor method to confirm the existence of global gs in distinctly different time - delay systems .
the main idea of this technique consists of the fact of preserving the local neighborliness between the states of the interacting systems @xcite .
let us consider the trajectories of two systems which are connected by the relation @xmath171 ( condition for gs ) .
the mfnn method depends on the observation that in gs state two neighboring points in the phase space of the drive system @xmath172 correspond to two neighboring points in the phase space of the response system @xmath173 . for mutually coupled systems
, the inverse statement is also valid .
that is , all close states in the phase space of the system @xmath173 must correspond to close states of the system @xmath172 .
mfnn parameter ( @xmath174 ) as a function of coupling strength ( @xmath14 ) showing global gs state for @xmath2 mutually coupled mg - pwl time - delay systems ( corresponding to fig .
[ fig5 ] ) . ]
let us consider a set of embedded vector points ( obtained by attractor reconstruction using time - delay embedding methods @xcite ) in the spaces of the drive ( @xmath175 ) and response ( @xmath176 ) systems coming from finite segments of the trajectories sampled at uniform intervals of time .
now we can choose an arbitrary point @xmath177 in the phase space of the drive system .
let the nearest phase space neighbor of this point in the reconstructed attractor be @xmath178 . in gs state , one can also expect that the corresponding points of the response system @xmath179 will have @xmath180 as its close neighbor . from the gs relation ,
the distance between the two nearest neighbors in the phase space of the response system can be written as @xmath181 , where @xmath182 is the jacobian matrix of the transformation @xmath183 evaluated at @xmath177 .
similarly , we consider the point @xmath179 and locate its nearest neighbor from the time series as @xmath180 . again using the gs relation ,
the distance between the points of the response variables can be written as @xmath184 .
this suggests that the ratio for the mfnn parameter can be written as , @xmath185 where @xmath186 is the sampling time . in gs state
the mfnn parameter ( @xmath174 ) will be of the order of unity .
this method has been widely used to identify gs in mutually coupled systems .
the mfnn parameter ( @xmath174 ) for two mutually coupled mg and pwl time - delay systems is depicted in fig .
[ fig7 ] ( corresponding to fig . [ fig5 ] ) .
as can be seen from this figure the value of @xmath174 becomes close to unity above @xmath187 , which is indeed the critical coupling strength @xmath188 in fig .
[ fig5 ] , strongly confirming the existence of global gs .
now , we extend the above analysis to @xmath3 mutually coupled distinctly nonidentical systems in an array and a ring configuration [ figs . [ fig1a](b ) and
[ fig1a](c ) ] .
in addition to the above two time - delay systems , now we have considered the third system as tpwl with the nonlinear function @xmath189 as given in eq .
( [ eqn1c ] ) .
the system parameters are fixed as in sec .
[ sec2 ] . for this chosen set of parameter values
the system exhibits a hyperchaotic attractor ( fig . [
fig1](c ) ) with four positive les ( @xmath43 ) [ see fig .
[ fig2](c ) ] . ( a )
mtles and ( b ) cc , cpr of the main and auxiliary systems for @xmath3 distinctly nonidentical time - delay systems with an linear array configuration as a function of @xmath14 . ]
mfnn parameter ( @xmath174 ) for the linear array of @xmath3 mutually coupled time - delay systems as a function of the coupling strength shows the global gs state ( corresponding to fig .
[ fig8 ] ) .
see text for details . ]
first , we illustrate the transition from partial to global gs in a linear array of @xmath3 mutually coupled systems using the auxiliary system approach .
the schematic diagram for this configuration is given in fig .
[ fig1a](b ) . here
the systems 1 , 2 and 3 are mutually coupled with their neighboring systems as depicted in fig.[fig1a](b ) and the associated auxiliary systems are driven by the neighboring main systems .
for example , @xmath190 is driven by @xmath121 , while @xmath191 by @xmath119 and @xmath192 , and @xmath193 by @xmath121 .
mtles , cc and cpr of all the three main and auxiliary systems are shown in fig .
[ fig8](a ) and [ fig8](b ) as a function of the coupling strength . in the absence of the coupling ( @xmath102 ) , positive values of mtles ( @xmath194 ) and the low values of @xmath137 and @xmath195
indicate that the main and the corresponding auxiliary systems evolve independently . increasing the coupling strength from null value results in decreasing @xmath196 and increasing @xmath137 and @xmath195 .
if we increase the coupling strength to @xmath197 , the pwl systems with three positive les becomes synchronized first as evidenced by @xmath110 , @xmath108 and @xmath109 , indicating the onset of partial gs .
the mtle of the mg systems with two positive les becomes negative ( @xmath198 ) along with @xmath199 and @xmath200 at @xmath201 which confirms that the mg system reaches the common gs manifold , while the tpwl systems are not yet synchronized .
finally for @xmath202 , the tpwl system with four positive les attains the cs state with @xmath203 and @xmath204 and @xmath205 reaching unit values exactly at the same threshold value of the coupling strength ( @xmath206 ) indicating the existence of global gs .
this also confirms the simultaneous existence of gs and ps in an array of three mutually coupled systems .
( a ) mtles and ( b ) cc , cpr of the main and auxiliary systems for @xmath3 distinctly nonidentical time - delay systems with ring configuration as a function of @xmath14 . ] for @xmath84 , it is not possible to obtain accurate analytical results for all the coupled systems with respect to the numerical simulation results due to the complexity in the coupling architecture . we have also calculated the mfnn parameter ( @xmath174 ) for the linear array of three mutually coupled time - delay systems as a function of the coupling strength , which is depicted in fig .
[ fig9 ] ( corresponding to fig . [ fig8 ] ) .
the mfnn parameter for the mg and pwl systems ( systems @xmath207 and @xmath208 ) becomes close to unity at @xmath209 ( filled circle ) .
the mfnn parameter @xmath174 of the mg and tpwl systems ( systems @xmath207 and @xmath210 ) approximately reaches the unit value at @xmath211 ( filled triangle ) agreeing perfectly with the threshold values indicated by mtles , cc and cpr , all confirming the existence of global gs in the array .
next , we consider a system of three mutually coupled systems with a ring configuration along with their corresponding auxiliary systems ( here each auxiliary system is driven by the other two systems , for example , @xmath120 is driven by @xmath121 and @xmath192 , @xmath122 by @xmath119 and @xmath192 and @xmath212 by @xmath119 and @xmath121 , as depicted in the schematic diagram fig . [ fig1a](c ) ) .
cc , cpr and mtles for the mutually coupled ring of the above three distinctly nonidentical systems are shown in fig .
[ fig10](a ) and [ fig10](b ) . again
all the three systems ( pwl , mg and tpwl ) reach their cs ( gs ) manifold for the threshold values @xmath213 and @xmath214 , respectively , as indicated by the transition of their @xmath196 below zero [ fig .
[ fig10](a ) ] .
in addition all @xmath137 and @xmath195 reach unit values [ fig .
[ fig10](b ) ] exactly at the same threshold values of the coupling strength indicating the simultaneous existence of gs and ps in a ring of three mutually coupled time - delay systems .
it has been already known that structurally different time - delay systems will produce chaotic / hyperchaotic attractors with dissimilar phase space geometry characterized by multiple time - scales ( see fig .
[ fig1 ] for example ) . in order to preserve the local neighboring points of each of the attractor
we have reconstructed the attractors of different time - delay systems with almost identical time - scales .
due to the increased complexity of the topological properties of chaotic / hyperchaotic attractors of the network of ring , global and star configurations , it is not possible to obtain the reconstructed attractor with similar time - scales in structurally different time - delay systems and hence mfnn can not be estimated in these cases .
in this section , we demonstrate the existence of transition from partial to global gs in four ( only in @xmath4 for clear visibility of figures depicting synchronization transitions ) mutually coupled distinctly nonidentical time - delay systems in the following configurations : a ring , an array , a global and a star . in addition to the above three time - delay systems discussed in the previous sections , as a fourth system , we consider the ikeda time - delay system with the nonlinear function as in eq .
( [ eqn1e ] ) @xcite .
this system exhibits a hyperchaotic attractor [ fig .
[ fig1](d ) ] with five positive les [ fig .
[ fig2](d ) ] with a ky dimension @xmath215 for the chosen parameter values . ( a )
mtles and ( b ) cc , cpr of the main and auxiliary systems for @xmath4 distinctly nonidentical time - delay systems with linear array configuration as a function of @xmath14 . ]
mfnn parameter ( @xmath174 ) for the linear array of mutually coupled systems with @xmath4 as a function of the coupling strength showing the global gs state ( corresponding to fig .
[ fig11 ] ) . ]
first we demonstrate the existence of transition from partial to global gs in four mutually coupled time - delay systems ( mg , pwl , tpwl and ikeda ) in a linear array configuration .
the schematic diagram for this configuration is depicted in fig .
[ fig1a](d ) . in figs .
[ fig11](a ) and [ fig11](b ) , we have again presented the characterizing quantities @xmath196 , @xmath137 and @xmath216 of all the four systems along with their associated auxiliary systems as a function of coupling strength . for @xmath102 , @xmath194 , while @xmath137 and @xmath195 show low values corresponding to a desynchronized state . at @xmath217 and @xmath218 , @xmath100 of the pwl systems and @xmath98 of the mg systems
become negative , respectively , which confirm that these systems reach the cs ( gs ) state with their corresponding auxiliary ( main ) systems , whereas the other two systems are not yet synchronized ( @xmath219 ) . increasing the coupling strength
, we find that at @xmath220 the ikeda systems with five positive les attain the cs ( gs ) manifold when @xmath221 and finally the tpwl system becomes synchronized at @xmath222 , confirming the transition from partial to global gs in four mutually coupled time - delay systems in an array configuration .
the @xmath137 and @xmath195 of the corresponding systems show a clear transition to unit value for their corresponding threshold values of @xmath223 confirming the simultaneous existence of gs and ps [ fig .
[ fig11](b ) ] .
we have again calculated the mfnn parameter ( @xmath174 ) for @xmath4 mutually coupled linear array of time - delay systems by taking the mg time - delay system as a reference system as depicted in fig .
[ fig12 ] .
the unit value of the mfnn parameter @xmath174 of the respective @xmath223 values of pwl , tpwl and ikeda time - delay systems ( with respect to the mg time - delay system ) confirms the existence of the transition from partial to global gs .
( a ) mtles and ( b ) cc , cpr of the main and auxiliary systems for @xmath4 distinctly nonidentical time - delay systems with ring configuration as a function of @xmath14 . ]
next , we illustrate the existence of a transition from partial to global gs in a ring of @xmath4 mutually coupled distinctly nonidentical time - delay systems [ fig .
[ fig1a](e ) ] .
mtles , cc and cpr of all the four main and corresponding auxiliary systems are depicted in figs .
[ fig13](a ) and [ fig13](b ) . in the absence of the coupling ,
all the systems evolve independently as evidenced by the fact that @xmath224 , and low values of @xmath137 and @xmath195 .
the transitions of mtles ( @xmath225 ) from positive to negative values of the mg , pwl , tpwl and ikeda systems indicate that they reach their cs ( gs ) manifolds for the threshold values of the coupling strengths @xmath226 and @xmath227 , respectively .
gs and ps also occur simultaneously as indicated by the @xmath137 and @xmath195 [ in fig .
[ fig13](b ) ] of the systems in the ring configuration . ( a )
mtles and ( b ) cc , cpr of the main and auxiliary systems of four distinctly nonidentical time - delay systems with global coupling as a function of @xmath14 .
( c ) and ( d ) are the corresponding figures for star configuration . ] in this section , we study the transition from partial to global gs in global ( all - to - all ) and star coupling configurations with @xmath4 time - delay systems with different fractal dimensions .
the schematic diagrams of the global and star coupling configurations are depicted in figs .
[ fig1a](f ) and [ fig1a](g ) , respectively .
mtles , cc and cpr of all the four systems are depicted in figs .
[ fig14](a - b ) and [ fig14](c - d ) for global and star coupling configurations , respectively .
in the case of global coupling configuration , @xmath100 of the pwl systems and @xmath98 of the mg systems become negative at @xmath228 and @xmath229 , respectively , while the other two systems evolve independently .
when the coupling strength is increased to @xmath230 , the ikeda system gets synchronized ( @xmath221 ) and finally @xmath231 of the tpwl system becomes negative at @xmath232 , thereby confirming the onset of a global gs state in globally coupled network of distinctly nonidentical time - delay systems .
further , the values of @xmath137 and @xmath195 reach unit value at the same values of the coupling strength as above indicating the simultaneous occurrence of gs and ps [ fig .
[ fig14](b ) ] . in the star configuration , @xmath100 transits first from positive to negative values at @xmath217 showing a cs between pwl and its auxiliary system [ fig .
[ fig14 ] ] .
this is followed by the value of @xmath98 of mg systems becoming negative at @xmath233 .
next for the ikeda systems @xmath234 becomes negative at @xmath235 and finally the @xmath231 of tpwl systems becomes negative at @xmath236 , confirming the existence of global gs .
further , for the same values of coupling strength , @xmath137 and @xmath195 reach unit values confirming the simultaneous occurrence of gs and ps in four distinctly nonidentical coupled time - delay systems with different fractal dimensions .
in addition , we have also confirmed that there exist a similar type of synchronization transitions in other permutations on the order of the systems between mg , pwl , tpwl and ikeda systems in all coupling configurations .
now , we will demonstrate the genericity of the transition from partial to global gs by revealing it in coupled time - delay systems of different orders so as to prove that the reported phenomenon is not restricted to time - delay equations with structural similarity alone .
in particular , we will show the existence of a transition to global gs via partial gs in such systems using the measures cc and cpr .
mutual coupling is introduced in the @xmath237 variable of higher order systems .
we will demonstrate the above results in the following coupled systems . 1 . in a system consisting of a mutually coupled ikeda time - delay system ( which is a scalar first order time - delay system ) and a hopfield neural network @xcite ( which is a second order time - delay system ) , and 2 . in a system of
mutually coupled mg time - delay system ( which is a scalar first order time - delay system ) and a plankton model @xcite ( which correspond to a third order system with multiple delays ) , a class of delayed chaotic neural networks @xcite can be represented as a set of coupled ddes as given by the equation @xmath238+bf\left[x(t-\tau)\right ] , \label{eqn9}\ ] ] where @xmath239^t \in r^n$ ] is the state vector , the activation function @xmath240= \left(f_1\left [ x_1(t)\right],f_2\left [ x_2(t)\right ] , \cdots , f_n\left [ x_n(t)\right]\right)^t$ ] denotes the manner in which the neurons respond to each other .
@xmath241 is a positive diagonal matrix , @xmath242 is the feedback matrix , @xmath243 represents the delayed feedback matrix with a constant delay @xmath38 . the general class of delayed neural networks represented by the above eq .
( [ eqn9 ] ) unifies several well known neural networks such as the hopfield neural networks and cellular neural networks with delay .
the specific set of delayed neural networks ( eq .
( [ eqn9 ] ) ) which corresponds to the hopfield neural networks is for the choice of the activation function @xmath244=\tanh\left [ x(t)\right],\ ] ] and for the value of the matrices @xmath245 next , we will illustrate the existence of global gs via partial gs in a system of mutually coupled ikeda time - delay systems and the hopfield neural network .
the cc and cpr between the main and auxiliary systems are depicted in fig .
[ fig15 ] .
low values of cc and cpr in the absence of coupling indicates the asynchronous behavior of both the systems . upon increasing the coupling strength from zero , the second order system , that is the hopfield neural network , reaches the common synchronization manifold first ( partial gs ) at the threshold value of @xmath246 as indicated by the unit value of the cross correlation coefficient @xmath104 , while the ikeda system remains in its transition state .
the simultaneous existence of phase synchronization ( ps ) together with gs is also confirmed by the unit value of @xmath106 at the same @xmath247 .
further increase in the coupling strength results in the synchronization ( both gs and ps ) of the ikeda system to the common synchronization manifold for @xmath248 as evidenced from the unit value of @xmath103 and @xmath249 ( fig .
[ fig15 ] ) confirming the existence of global gs via partial gs in coupled systems of different orders .
cc and cpr of the main and auxiliary systems of a coupled ikeda time - delay system and hopfield neural network . ] cc and cpr of the main and auxiliary systems of a coupled mackey - glass time - delay system and the plankton model ( [ plankton ] ) . ] next , we illustrate the transition from partial to global gs in mutually coupled mg time - delay system , and a third order plankton model @xcite with multiple delays .
the normalized system of equations of a zoo - plankton model is represented as @xmath250-xy - l_1xz , \\ \dot{y}=&\,xy - b_2y - l_2yz , \\
\dot{z}=&\,-b_1z+l_1x(t-\tau_1)z(t-\tau_3)+l_2y(t-\tau_2)z(t-\tau_3)-n(x+y)z , \label{plankton}\end{aligned}\ ] ] where @xmath251 and @xmath252 are constants .
delays @xmath253 and @xmath254 are in general different may also be different , but for simplicity we have considered identical delays , @xmath255 , as studied in ref @xcite . here
@xmath256 and @xmath257 are the normalized quantities of the density of the susceptible phytoplankton , infected phytoplankton and zooplankton ( predator species ) , respectively .
low values of the cc and cpr for @xmath102 between the main and auxiliary systems as shown in fig .
[ fig16 ] confirm that the systems evolve independently in the absence of coupling between them .
as the coupling strength is increased the plankton model synchronizes first to the common synchronization manifold at @xmath258 as denoted by @xmath259 indicating partial gs ( fig . [ fig16 ] ) .
ps has also occurred simultaneously at the same threshold value of @xmath188 as indicated by the unit value of @xmath249 .
further increase in @xmath14 leads to the existence of global gs by synchronizing the mg time - delay system to the common synchronization manifold as both @xmath104 and @xmath106 attain unity at @xmath260 .
thus , we prove that the transition from partial to global gs phenomenon is not restricted to first order structurally similar systems alone but it is also valid for systems with different orders .
in conclusion , we have pointed out the existence of a synchronization transition from partial to global gs in distinctly nonidentical time - delay systems with different fractal dimensions in symmetrically coupled ( regular ) networks with a linear array , and also in ring , global and star configurations using the auxiliary system approach and the mutual false nearest neighbor method .
we have shown that there exists a smooth transformation function even for networks of structurally different time - delay systems with different fractal dimensions , which maps them to a common gs manifold . we have also found that gs and ps occur simultaneously in structurally different time - delay systems .
we have calculated mtles to evaluate the asymptotic stability of the cs manifold of each of the main and the corresponding auxiliary systems .
this in turn , ensures the stability of the gs manifold between the main systems .
in addition , we have also estimated the cc and the cpr to characterize the relation between gs and ps .
further , to prove the genericity of our results , we have also demonstrated the synchronization transition in systems with different orders such as coupled mg and the hopfield neural network model and a system of coupled ikeda and plankton models . the analytical stability condition for partial and global gs
is deduced using the krasovskii - lyapuov theory .
we wish to note that now we are working on the experimental realization of the existence of partial and global gs in distinctly nonidentical time - delay systems using nonlinear time - delayed electronic circuits .
d. v. senthilkumar acknowledges the supoport from serb - dst fast track scheme for young scientists .
m. lakshmanan(m .
l. ) has been supported by the dst , government of india sponsored irhpa research project .
m. l. has also been supported by a dst ramanna fellowship project and a dae raja ramanna fellowship .
9 abarbanel , h. d. i. , rulkov , n. f. & sushchik , m. m. [ 1996 ] generalized synchronization of chaos : the auxiliary system approach " _ phys .
e _ * 53 * , 4528 .
abarbanel , h. d. i. , brown , r. , sidorowich , j. j. & tsimring , l. s. [ 1993 ] the analysis of observed chaotic data in physical systems " _ rev .
_ * 65 * , 1331 .
abraham , e. r. [ 1998 ] the generation of plankton patchiness by turbulent stirring " _ nature _ * 391 * , 577 .
acharyya , s. & amritkar , r. e. [ 2013 ] generalized synchronization of coupled chaotic systems " _ eur .
j. special topics _ * 222 * , 939 .
amritkar , r. e. & rajgarajan , g. [ 2006 ] spatially synchronous extinction of species under external forcing " _ phys .
lett . _ * 96 * , 258102 .
appeltant , l. , soriano , m. c. , van der .
sande , g. , danckaert , j. , masser , s. , dambre , j. , schrauwen , s. , mirasso , c. r. & fischer , i. [ 2011 ] information processing using a single dynamical node as complex system " _ nature communications _ * 2 * , 468 .
blasius , b. , huppert , a. & stone , l. [ 1999 ] complex dynamics and phase synchronization in spatially extended ecological systems " _ nature ( london ) _ * 392 * , 239 .
boccaletti , s. , valladares , d. l. , kurths , j. , maza , d. & mancini , h. [ 2000 ] synchronization of chaotic structurally nonequivalent systems " _ phys .
e _ * 61 * , 3712 .
brown , r. [ 1998 ] approximating the mapping between systems exhibiting generalized synchronization " _ phys .
lett . _ * 81 * , 4835 .
chen , j. , lu , j .- a .
, wu , x. & zheng , w. x. [ 2009 ] generalized synchronization of complex dynamical networks via impulsive control " _ chaos _ * 19 * , 043119 .
dmitriev , b. s. , hramov , a , e. , krasovskii , a. a. , starodubov , a. v. , trubetskov , d. i. & zharkov , y. d. [ 2009 ] first experimental observation of generalized synchronization phenomena in microwave oscillators " _ phys .
lett . _ * 107 * , 074101 .
earn , d. t. d. , rohani , p. & grenfell , b. t. [ 1998 ] persistence , chaos and synchrony in ecology and epidemiology " _ proc .
london , ser .
b _ * 265 * , 1471 . farmer , s. f. [ 1998 ] rhythmicity , synchronization and binding in human and primate motor systems " _ j. physiol . _
* 509 * , 3 .
gakkhar , s. & singh , a. [ 2010 ] a delay model for viral infection in toxin producing phytoplankton and zooplankton system " _ commun .
nonlinear sci .
* 15 * , 3607 .
grenfell , b. t. , bjornstad , o. n. & kappey , j. [ 2001 ] travelling waves and spatial hierarchies in measles epidemics " _ nature ( london ) _ * 414 * , 716 .
guan , s. , wang , x. , gong , x. , li , k. & lai , c .- h . [ 2009 ] the development of generalized synchronization on complex networks " _ chaos _ * 19 * , 013130 .
guan , s. , gong , x. , li , k. , liu , z. & lai , c .- h . [ 2010 ] characterizing generalized synchronization in complex networks " _ new j. phys . _
* 12 * , 073045 .
hopfield , j. j. [ 1982 ] neural networks and physical systems with emergent collective computational abilities " _ proc .
_ * 79 * , 2554 .
hu , a. , xu , z. & guo , l. [ 2010 ] the existence of generalized synchronization of chaotic systems in complex networks " _ chaos _ * 20 * , 013112
. hung , y. c. , huang , y. t. , ho , m. c. & hu , c. k. [ 2008 ] paths to globally generalized synchronization in scale - free networks " _ phys .
e _ * 77 * , 016202 .
ikeda , k. , daido , h. & akimoto , o. [ 1980 ] optical turbulence : chaotic behavior of transmitted light from a ring cavity " _ phys .
lett . _ * 45 * , 709 .
kocarev , l. & parlitz , u. [ 1996 ] generalized synchronization , predictability , and equivalence of unidirectionally coupled dynamical systems " _ phys .
lett . _ * 76 * , 1816 .
koronvskii , a , a. , moskalenko , i , o. & hramov , a. e. [ 2011 ] nearest neighbors , phase tubes , and generalized synchronization " _ phys .
e _ * 84 * , 037201 .
lakshmanan , m. & senthilkumar , d. v. [ 2010 ] _ dynamics of nonlinear time - delay systems _
( springer , berlin ) .
mackey , m. c. & glass , l. [ 1977 ] oscillation and chaos in physiological control systems " _ science _ * 197 * , 287 .
maraun , d. & kurths , j. [ 2005 ] epochs of phase coherence between el nio / southern oscillation and indian monsoon " _ j. geophys .
lett . _ * 32 * , l15709 .
marwan , n. , romano , m. c. , thiel , m. & kurths , j , [ 2005 ] recurrence plots for the analysis of complex systems , " _ phys .
_ * 438 * , 237 - 329 .
meng , j. & wang , x. [ 2007 ] robust anti - synchronization of a class of delayed chaotic neural networks " _ chaos _ * 17 * , 023113 .
moskalenko , o. i. , koronovskii , a. a. , hramov , a. e. & boccaletti , s. [ 2012 ] generalized synchronization in mutually coupled oscillators and complex networks " _ phys . rev .
e _ * 86 * , 036216 .
moskalenko , o. i. , koronovskii , a. a. & hramov , a. e. [ 2010 ] generalized synchronization of chaos for secure communication : remarkable stability to noise " _ phys .
a _ * 374 * , 2925 .
murali , k. & lakshmanan , m. [ 1998 ] secure communication using a compound signal from generalized synchronizable chaotic systems " _ phys . lett .
a. _ * 241 * , 303 .
packard , n. h. , crutchfield , j. p. , farmer , j. d. & shaw , r. s. [ 1980 ] geometry from a time series " _ phys .
. lett . _ * 45 * , 712 .
parlitz , u. , junge , l. , lauterborn , w. & kocarev , l. [ 1996 ] experimental observation of phase synchronization " _ phys .
e _ * 54 * , 2115 .
parlitz , u. , junge , l. & kocarev , l. [ 1997 ] subharmonic entrainment of unstable periodic orbits and generalized synchronization " _ phys .
lett . _ * 79 * , 3158 .
pecora , l. m. , carroll , t. l. , jhonson , g. a. & mar , d. j. [ 1996 ] fundamentals of synchronization in chaotic systems , concepts , and applications . " _ chaos _ * 7 * , 520 .
pikovsky , a. s. , rosenblum , m. g. & kurths , j. [ 2001 ] _ synchronization - a unified approach to nonlinear science _
( cambridge university press , cambridge , england ) .
pyragas , k. [ 1996 ] weak and strong synchronization of chaos " _ phys .
e _ * 54 * , 4508(r ) .
rogers , e. a. , kalra , r. , schroll , r. d. , uchida , a. , lathrop , d. p. & rajarshi roy [ 2004
] generalized synchronization of spatiotemporal chaos in a liquid crystal spatial light modulator " _ phys .
* 93 * , 084101 .
rulkov , n. f. , sushchik , m. m. , tsimring , l. s. & abarbanel , h. d. i. [ 1995 ] generalized synchronization of chaos in directionally coupled chaotic systems " _ phys .
e _ * 51 * , 980 .
schfer , c. , rosenblum , m. g. , kurths , j. & abel , h. h. [ 1998 ] heartbeat synchronized with ventilation " _ nature ( london ) _ * 392 * , 239 .
schumacher , j. , haslinger , r. & pipa , g. [ 2012 ] statistical modeling approach for detecting generalized synchronization " _ phys .
* 85 * , 056215 .
sebe , j. y. , van berderode , j. f. , berger , a. j. & abel , h. h. [ 2006 ] inhibitory synaptic transmission governs inspiratory motoneuron synchronization " _ j. neurophysiol _ * 96 * , 391 .
senthilkumar , d. v. , lakshmanan , m. & kurths , j. [ 2006 ] phase synchronization in time - delay systems " _ phys .
* 74 * , 035205(r ) .
senthilkumar , d. v. , lakshmanan , m. & kurths , j. [ 2007 ] transition from phase to generalized synchronization in time - delay systems " _ chaos _ * 18 * , 023118 .
senthilkumar , d. v. , suresh , r. , lakshmanan , m. & kurths , j. [ 2013 ] global generalized synchronization in networks of different time - delay systems " _ europhys .
lett . _ * 103 * , 50010 .
shahverdiev , e. m. & shore , k. a. [ 2009 ] generalized synchronization in laser devices with electro - optical feedback " _ iet optoelectron . _ * 3 * , 274 .
shang , y. , chen , m. & kurths , j. [ 2009 ] generalized synchronization of complex networks " _ phys . rev .
e _ * 80 * , 027201 .
sorino , m. c. , guy van der sande , fischer , i. & mirasso , c. r. [ 2012 ] synchronization in simple network motifs with negligible correlation and mutual information measures " _ phys
lett . _ * 108 * , 134101 .
stein , k. , timmermann , a. & schneifer , n. [ 2011 ] phase synchronization of the el - nio - southern oscillation with the annual cycle " _ phys .
lett . _ * 107 * , 128501 .
suresh , r. , srinivasan , k. , senthilkumar , d. v. , raja mohamed , i. , murali , k. , lakshmanan , m. & kurths , j. [ 2013 ] zero - lag synchronization in coupled time - delayed piecewise linear electronic circuits " _ eur .
. j. special topics _ * 222 * , 729 .
uchida , a. , mcallister , r. , meucci , r. & rajarshi roy , [ 2003 ] generalized synchronization of chaos in identical systems with hidden degrees of freedom " _ phys .
* 91 * , 174101 .
van leeuwen , p. , geue , d. , thiel , m. , cysarz , d. , large , s. , romano , m. c. , wessel , n. , kurths , j. & grnemeyer , d. h. [ 2009 ] influence of paced maternal breathing on fetal maternal heart rate coordination " _ porc .
sci . usa _ * 106 * , 13661 .
zheng , z. , wang , x. & cross , m. c. [ 2002 ] transitions from partial to complete generalized synchronizations in bidirectionally coupled chaotic oscillators " _ phys .
e _ * 65 * , 056211 .
zheng , z. & hu , g. [ 2000 ] generalized synchronization versus phase synchronization " _ phys . rev .
* 62 * , 7882 . | we point out the existence of transition from partial to global generalized synchronization ( gs ) in symmetrically coupled regular networks ( array , ring , global and star ) of distinctly different time - delay systems of different orders using the auxiliary system approach and the mutual false nearest neighbor method . it is established that there exist a common gs manifold even in an ensemble of structurally nonidentical time - delay systems with different fractal dimensions and we find that gs occurs simultaneously with phase synchronization ( ps ) in these networks .
we calculate the maximal transverse lyapunov exponent to evaluate the asymptotic stability of the complete synchronization manifold of each of the main and the corresponding auxiliary systems , which , in turn , ensures the stability of the gs manifold between the main systems .
further we also estimate the correlation coefficient and the correlation of probability of recurrence to establish the relation between gs and ps .
we also deduce an analytical stability condition for partial and global gs using the krasovskii - lyapunov theory . |
high velocity clouds are detected primarily in ; no resolved optical counterparts such as stars have been detected in hvcs , and the only distance upper limits are for two hvcs seen in absorption against background halo stars .
the lack of distance constraints makes the nature of and models for hvcs extremely uncertain ; see the review of wakker & van woerden ( 1997 ) .
currently favored models include recycling of disk gas through a fountain ( _ e.g. _ bregman 1980 ) ; stripping from galactic satellites ; and infall of possibly primordial gas ( _ e.g. _ the local group model of blitz _
et al . _
these models place hvcs at from @xmath2 kpc to @xmath3 mpc respectively , a range of 100 in distance and @xmath4 in gas mass .
faint , diffuse optical recombination emission lines are observed from some hvcs .
h@xmath0 emission must be caused by ionization on the cloud , either from photo - ionization by lyman continuum radiation , or another process such as collisional ionization .
measurements of h@xmath0 flux can constrain hvc distances , given models for the ionization processes .
observing diffuse emission lines from hvcs is difficult because the emission is faint , night sky emission lines are strong and fluctuating , and hvcs are degrees across , larger than the field of view of most optical telescopes .
fabry - perot observations of diffuse emission provide large collecting solid angle with moderate to high spectral resolution , needed to separate hvc emission from sky lines .
fps have previously been used to detect h@xmath0 emission from four hvcs and the magellanic stream ( weiner & williams 1996 , tufte _ et al . _ 1998 , bland - hawthorn _ et al . _
chopping by several degrees between object and sky fields is necessary to obtain decent sky subtraction .
figure 1 shows example spectra from the wide field camera f - p at the las campanas 2.5-m , with a field of view of 25@xmath5 and etalon fwhm 1.2 ( @xmath6 ) .
the top panel is a extracted spectrum of a 900 sec exposure of hvc 343 + 32140 , one of the hvcs brightest in h@xmath0 , before subtraction of a sky - field spectrum .
the second panel is after a two - step sky subtraction : sky lines in the object - field and sky - field spectra are fit and subtracted , then the sky - field continuum is subtracted from the object - field continuum .
the h@xmath0 flux is strong , 1060 milli - rayleighs ( mr ) , as is [ n ii ] 6583 , with [ n ii]/h@xmath0 = 1.1 .
( 1 rayleigh = @xmath7 photons @xmath8 s@xmath9 into @xmath10 . )
the lower two panels show two hvcs very faint in h@xmath0 , hvc 16543280 ( 41 mr ) and hvc 230 + 61 + 165 ( 48 mr ) .
the h@xmath0 detections agree well with the velocities .
no [ n ii ] is detected .
note the tremendous difference in strength of hvc h@xmath0 and night - sky lines .
figure 2 compiles our results from the lco fp and the rutgers fp at the ctio 1.5-m , and hvc h@xmath0 detections from tufte _
et al._(1998 ) and bland - hawthorn _ et al . _ ( 1998 ) .
there is a wide range of hvc h@xmath0 intensity , but clouds in the same complex tend to have similar intensities , which suggests that the variations between complexes are due to hvc properties ( _ e.g. _ distances ) rather than extrinsic variations ( _ e.g. _ spatial variations in the ionizing field escaping from the galaxy ) . on the other hand ,
the magellanic stream points vary widely : some points have strong emission while others have weak or no h@xmath0 despite a high column density .
strong emission in the ms is often located near cloud edges .
a fundamental result of our h@xmath0 survey is that we have not found any cloud near or below our photon - limited noise ( generally @xmath11 mr at 2@xmath12 ) .
the faintest h@xmath0 detections , at 4150 mr , are well above our noise limit .
we always either detect h@xmath0 ( 15 of 20 hvcs ) , or are clobbered by residuals from sky - subtraction when hvcs are not well separated from sky lines , raising the detection limit .
if h@xmath0 emission from hvcs is due to photo - ionization by flux from the galaxy , we can infer distances to the hvcs .
we use the hvcs a and m with known distance brackets ( 410 kpc and @xmath13 kpc ; van woerden _
et al . _ 1999 ,
et al . _
1993 ) and h@xmath0 fluxes ( tufte _ et al . _
1998 ) to normalize a model for the ionizing flux escaping from the galaxy . figure 3 shows contours of the ionizing photon flux @xmath14 in the model ; it has a total ionizing luminosity of @xmath15 photons s@xmath9 , distributed in an exponential disk , and models the galactic absorbing layer with a one - sided face - on optical depth to ionizing photons of @xmath16 , yielding an overall , angle - averaged escape fraction @xmath17 .
( see also bland - hawthorn & maloney 1999 and bland - hawthorn , these proceedings . )
the inferred distances of hvcs are indicated ; the error bars are for systematic variations by a factor 1.5 up or down in h@xmath18 ratio ( statistical errors on the fluxes are much smaller ) .
the brightest clouds are within a few kpc of the galactic plane but fainter clouds are well away from the plane .
these clouds at @xmath19 kpc are inconsistent with a galactic fountain origin , especially given their high velocities .
however , the observed h@xmath0 in the magellanic stream can not be explained by this model ; ms ii and ms iv are much too bright compared to hvcs a and m. in fact , at a distance of @xmath20 kpc , the ms h@xmath0 emission can not be explained by a reasonable model of photoionization from the galaxy .
the h@xmath0 photon flux is @xmath21 , so that h@xmath0 of 300600 mr at ms ii requires @xmath22 and @xmath23 .
the required escape fraction is unrealistically large ( even if @xmath24 were increased by @xmath25 ) since most ionizing radiation must be absorbed in the galaxy to power regions .
( in agreement with the authors , we find that bland - hawthorn & maloney 1999 overestimated the ionizing flux incident on the ms . )
furthermore , there are strong spatial variations in the ms h@xmath0 , shown by the large dispersion of points in figure 2 .
photoionization from the galaxy should produce roughly the same h@xmath0 flux on any surface optically thick to ionizing radiation , since the ionizing photons travel nearly along our line of sight to the ms .
factors of 23 variation between different points are tolerable either from variations in the incident field or geometry of the absorbing surface , but factors of 30 are not . a second source of ionization in the galactic halo is needed to explain the ms h@xmath0 .
simply put , why are some points in the magellanic stream brighter than complexes a and m , despite being much further away ?
the only likely source is collisional ionization , probably from interaction with hot halo gas through ram pressure and turbulent mixing / thermal conduction ( weiner & williams 1996 ) .
the ms radial velocities are large and the space velocities with respect to halo gas are probably above 200 km s@xmath9 ; energy input from ram pressure goes as @xmath26 , so it may explain the brightness of the ms with respect to hvcs a and m , but it is barely adequate for reasonable halo gas densities ( weiner & williams 1996 ) ; perhaps the magellanic stream clouds are colliding with each other .
collisional ionization is required , but the brightest ms points remain a puzzle .
recent fuse detections of o vi absorption in the magellanic stream and in a few hvcs provide further evidence for collisional ionization in hvcs ( sembach _ et al . _ 2000 ) .
o vi can be produced collisionally in hot gas at @xmath27 k , which could arise from the interaction of an cloud with hot halo gas , but o vi can not realistically be produced in hvcs by photoionization from o stars .
given the existence of a second source of ionization in the galactic halo , simple photoionization models may not yield reliable distances .
however , robust predictions can still be made for a relation of h@xmath0 flux to distance .
hvcs which are bright in h@xmath0 ( as hvcs go ) are likely only a few kpc from the galactic plane .
the brightest hvcs plotted in figure 2 ( l , gcp , cr ) have inferred distances which give them low `` deviation velocities '' ( wakker & van woerden 1991 ) , so they are not too far from galactic differential rotation , despite having high @xmath28 ; these hvcs could well be recycled disk gas .
high [ n ii]/h@xmath0 ratios suggest high - metallicity gas photoionized by a hardened radiation field , as in the extraplanar diffuse ionized gas of ngc 891 ( rand 1998 ) .
conversely , if an hvc is very far from the galaxy , the only source of ionization is the metagalactic uv background
. then the hvc should be very faint in h@xmath0 , approaching the upper limits on h@xmath0 emission from isolated extragalactic h i clouds : @xmath29 mr at @xmath30 from the giovanelli - haynes cloud ( vogel _ et al . _
1995 ) and @xmath31 mr at @xmath30 from the leo ring ( donahue _ et al .
_ 1995 ) .
we do not find any hvcs that are very faint , even though our photon - limited noise is generally 1020 mr ( 2@xmath12 ) : we either detect h@xmath0 , or are limited by sky subtraction residuals .
the faintest hvcs we find are 16543280 , 4752129 , 22734 + 114 , and 230 + 61 + 165 , with h@xmath0 emission at 4148 mr . in the photoionization model of figure 3
their distances are @xmath32 , and 26 kpc from the sun .
we can push the distances to the limit by considering solely collisional ionization : the hvc h@xmath0 fluxes are some 5 - 15 times smaller than the fluxes typical of the magellanic stream .
if we attribute the h@xmath0 to interaction with hot halo gas , and assume that the hot gas density falls off as @xmath33 to @xmath34 , the hvcs can be at most 2 to 4 times farther than the ms , or 100 - 200 kpc . at distances of somewhere between 20 and at most 200 kpc , the h@xmath0-faint hvcs are not local group objects as defined by the blitz _
et al . _ ( 1999 ) model .
however , their low [ n ii]/h@xmath0 ratios suggest low metallicity , and they are not easily produced by a galactic fountain : for example , hvc 16543280 , with @xmath35 km s@xmath9 , has faint h@xmath0 , is at least 10 kpc away ( 20 kpc galactocentric radius ) , and has a large mass and kinetic energy ; perhaps it is infalling tidal debris .
the best explanation for these h@xmath0-faint hvcs is still likely to be gas , of uncertain origin , accreting onto the galaxy .
bland - hawthorn , j. & maloney , p.r .
1999 , , 510 , l33 bland - hawthorn , j. , veilleux , s. , cecil , g.n . ,
putman , m.e . ,
gibson , b.k . & maloney , p.r .
1998 , , 299 , 611 blitz , l. , spergel , d.n . ,
teuben , p.j . , hartmann , d. & burton , w. b. 1999 , , 514 , 818 bregman , j.n .
1980 , , 236 , 577 danly , l. , albert , c.e . and kuntz , k.d .
1993 , , 416 , l29 donahue , m. , aldering , g. , & stocke , j.t .
1995 , , 450 , l45 rand , r.j .
1998 , , 501 , 137 sembach , k.r . , _ et al .
_ 2000 , preprint astro - ph/0005012 tufte , s.l .
, reynolds , r.j . & haffner , l.m .
1998 , , 504 , 773 van woerden , h. , schwarz , u.j . ,
peletier , r.f . ,
wakker , b.p . & kalberla , p.m.w . 1999 ,
nature , 400 , 138 vogel , s.n . ,
weymann , r. , rauch , m. & hamilton , t. 1995 , , 441 , 162 wakker , b.p . & van woerden , h. 1991 , , 250 , 509 wakker , b.p . &
van woerden , h. 1997 , , 35 , 217 weiner , b.j .
, & williams , t.b .
1996 , , 111 , 1156 | we report results from a survey of high velocity clouds and the magellanic stream for faint , diffuse optical recombination emission lines .
we detect h@xmath0 emission with surface brightness from 41 to 1680 mr from hvcs , and from @xmath1 to 1360 mr in the ms . a simple model for the photoionizing radiation emergent from the galaxy , normalized to the hvcs a and m with known distances , predicts distances from a few to 40 kpc , placing the faintest hvcs in the galactic halo , too far away for a galactic fountain .
this model can not explain the bright and spatially varying h@xmath0 in the magellanic stream , which requires another source of ionization .
however , we do not find any hvcs super - faint in h@xmath0 ; even with another ionization source , we conclude that the detected hvcs are not more than 24 times the distance of the ms ( 100 - 200 kpc ) . |
squeezed states as well as number states ( fock states ) are so - called nonclassical states .
they allow the measurement , the communication and the processing of information in a way not possible with coherent states that are governed by vacuum fluctuations .
squeezed states of light and correlated photon pairs have been used in order to realize interferometric measurements with sensitivities beyond the photon counting noise @xcite , to demonstrate the einstein - podolski - rosen paradox @xcite , as a resource for quantum teleportation @xcite and for the generation of schrdinger kitten states for quantum information networks @xcite . in view of applications in long distance quantum communication , purification and distillation of
entangled photon number states and entangled two - mode squeezed states were experimentally demonstrated @xcite .
+ fock states are characterized by photon counting detectors , whereas squeezed states are traditionally characterized in the phase space of position and momentum - like operators @xcite .
appropriate operators are the non - commuting amplitude and phase quadratures of the field , @xmath0 and @xmath1 , respectively .
their variances obey a heisenberg uncertainty relation , @xmath2 , with the vacuum noise variance normalized to 1/4 .
a squeezed state is realized if the quantum noise of one quadrature is _ squeezed _ below the vacuum noise level , at the expense of increased noise of the other quadrature . for an overview
we refer to ref .
the relation between the quadrature operators and the photon number operator @xmath3 is revealed by the hamilton operator of the quantized harmonic oscillator : @xmath4 where @xmath5 is the reduced planck constant and @xmath6 the light s angular frequency .
one can think of the two expressions in equation ( [ hamiltonian ] ) corresponding to the particle picture and the wave picture of quantum physics respectively . these can be used to calculate the mean photon number of squeezed states .
note , that this number is always greater than zero .
consider now a squeezed state without displacement in phase - space .
such a state is called a squeezed _ vacuum _ state , and one can write @xmath7 . if such a state is pure , i.e. has minimum uncertainty , only _ even _ photon numbers can be observed @xcite .
furthermore , if the squeezing effect is strong a so - called odd - even _ photon number oscillation _ is realized , that includes not just the existence of photon pairs but also the existence of higher even - numbers of quantum - correlated photons .
+ this article presents the generation and characterization of squeezed vacuum states with strong odd - even photon number oscillations revealing a significant contribution of correlated photon numbers up to 10 and higher .
this result is made possible because our squeezed vacuum states show unprecedented quality , i.e. are strongly squeezed , are of high purity , and are in a well - defined spatial gaussian fundamental mode .
these properties are of great importance for applications in high - precision measurements as well as in quantum information . to the best of our knowledge
, our squeezed light source also shows the broadest squeezing bandwidth ever observed and thus constitutes a bright source of quantum - correlated photons .
at present the most successful squeezed light sources are based on parametric processes in nonlinear optical materials . in the regime of pulsed laser radiation
the optical kerr effect has recently enabled the observation of nearly 7db squeezing @xcite . in the regime of continuous - wave laser radiation ,
squeezing of vacuum fluctuations by up to 10db was observed utilizing optical parametric oscillation ( opo ) below threshold @xcite .
+ the squeezed vacuum states investigated here were produced via type i opo below threshold in a nonlinear standing wave cavity , similar to the experimental set - up in @xcite . in order to observe high levels of squeezing the entire experiment was optimized for low phase fluctuations and low optical loss .
a schematic diagram of the experimental setup is shown in fig.[experiment](a ) .
the main light source was a 2 watt continuous wave laser at 1064 nm .
it was used to generate a second harmonic light field to pump the parametric squeezed light source and to provide the local oscillator ( lo ) beam for balanced homodyne detection .
both laser fields were sent through travelling wave filter cavities that were kept on resonance using pound - drever - hall modulation - demodulation control schemes .
the filter cavities therefore acted as optical low - pass filters that reduced potential phase noise and provided clean gaussian fundamental modes .
the filtered second - harmonic pump beam had a power of up to 600mw and was subsequently injected into the opo cavity .
+ in order to minimize the internal optical loss of the squeezed light source we designed the opo as a monolithic ( bi - convex ) cavity , similar to the resonator used in @xcite .
it was made from a 7% doped mg : o linbo@xmath8 crystal measuring 6.5@[email protected]@xmath10 in dimensions with end faces of 8 mm radii of curvature ( see inlay in fig . [ experiment ] ) . while the back of the crystal had a high reflection coating for both the wavelengths involved , we chose a rather low power reflectivity of 88% at 1064 nm ( and @xmath111% at 532 nm ) for the coupling front face in order to increase the ratio of out - coupling rate and cavity round trip loss .
the crystal was controlled to keep its phase matching temperature close to 60c using peltier elements , and the main laser frequency was tuned to match the resonance condition of the opo cavity .
an assembly drawing of the crystal mounting is shown in fig . [ experiment](b ) . +
no control fields at or near the fundamental wavelength of 1064 nm were injected into the squeezed light source during data taking . thus all photons around 1064 nm leaving the opo cavity originated from the parametric down - conversion process .
the out - coupled squeezed states , i.e. the photons at 1064 nm , were separated from the 532 nm pump upon reflection at a dichroic beam splitter ( dbs ) and subsequently analyzed by means of balanced homodyne detection . to this end
the bright and spatially filtered lo beam was precisely adjusted to the spatial mode of the squeezed beam .
we achieved a fringe visibility of 99.8% at the homodyne detector s 50/50 beam splitter , which proves that the squeezed states were indeed in a well - defined gaussian fundamental mode .
+ a homodyne detector is the only known means to observe and characterize strongly squeezed states .
todays single photon detectors do not reach the same high quantum efficiencies and have an insufficient photon number resolution .
it has been shown that homodyne detection data of quantum states can indeed provide their _ full _ quantum information @xcite .
the phase angle @xmath12 between the local oscillator ( lo ) and the signal beam defines an arbitrary quadrature of the signal s quantum state @xmath13 @xcite . in our experiment
@xmath12 was controlled with a piezo actuated mirror in the local oscillator path which allowed us to precisely set the phase angle .
for example , to measure the squeezed quadrature @xmath14 or the anti - squeezed quadrature @xmath15 .
the collected data was readily converted into variances by a spectrum analyzer that was also used to select a certain fourier frequency @xmath16 and resolution bandwidth @xmath17 .
the vacuum noise reference was measured with the same settings but with the squeezed field input blocked . .
, scaledwidth=49.0% ] the traces in fig .
[ squeezing](a)-(c ) represent three different squeezed vacuum states from our opo source in terms of pairs of squeezed ( bottom trace ) and anti - squeezed ( top trace ) vacuum normalized quadrature variances @xmath18 and @xmath19 , respectively .
= 5mhz , with a resolution bandwidth of @xmath20100khz and video bandwidth of 100hz , and were normalized to the vacuum noise .
the data still include electronic detector dark noise , hence all traces represent direct observations without any data post - processing .
* ( a ) * , squeezed vacuum state that was generated with a second harmonic pump power of @xmath21=30mw .
a nonclassical noise reduction of 2.9db below vacuum noise was observed , with the respective anti - squeezing of 2.9db above shot noise . *
( b ) * , squeezed vacuum state generated with an increased pump power of @xmath21=150mw .
squeezing of 6.2db was measured with 6.7db of anti - squeezing . * ( c ) * , driving the opo at maximum pump power @xmath21=600mw yielded squeezing and anti - squeezing of 11.5db and 16.0db respectively .
the subtraction of the detector dark noise would further increase the value of maximum squeezing to 11.6db .
, scaledwidth=50.0% ] for a pump power of 30mw ( fig .
[ squeezing](a ) ) the squeezed variance ( @xmath22 ) was 2.9(@xmath230.1)db below and the anti - squeezed variance ( @xmath24 ) 2.9(@xmath230.1)db above the vacuum noise level . within the measurement error bars of @xmath230.1db
the state shows minimum uncertainty and therefore perfect purity . upon driving the opo with 150mw pump power ( fig .
[ squeezing](b ) ) we observed normalized variances of -6.2(@xmath230.1)db and 6.7(@xmath230.1)db , respectively .
the strongest and unprecedented squeezing level was measured with a pump power of approximately 600mw ( fig .
[ squeezing](c ) ) . with this
setting 16.0(@xmath230.1)db of anti - squeezing and a noise reduction of 11.5(@xmath230.1)db below the vacuum noise level was achieved . in practice , squeezed vacuum states generally can not have perfect purity due to both residual optical loss and residual phase noise .
the former results in an admixture of the vacuum state and the latter results in a mixture of phase space rotated states @xcite . here
, we make the attempt to describe all the three squeezed states measured by a fixed amount of optical loss assuming zero phase noise .
this approach is reasonable for the following reasons , i ) previous investigations have shown that our set - up is stable and phase noise is low @xcite , ii ) in contrast to squeezing in the audio - band @xcite , at mhz fourier frequencies no displaced or thermal states are present that could mix into the state , and iii ) no evidence was found for a non - gaussian statistic in our measured data .
all 3 pairs of variances observed can then be consistently calculated using @xmath25 , where @xmath26 represent the pure state variances before any decoherence has occurred .
here , @xmath27 represents the overall contribution of the vacuum state , in power . the detection efficiency @xmath28 accounts for all losses outside the squeezed light source and @xmath29
is the opo cavity escape efficiency accounting for all losses inside the squeezed light source .
we find that all of the measured variances ( shown in fig . [ squeezing ] )
can indeed be modelled by a vacuum noise contribution as low as @xmath30% , independent of the pump power used .
this result supports our assumption that phase noise is not significant in our set - up .
the loss value of 4.8% is consistent with an independent loss analysis of all the optical components used in our set - up , however , an independent measurement of the quantum efficiency of the photo diodes of the homodyne detector typically has large uncertainties , and we estimate this value to ( @xmath31)% . such a low optical loss should enable the observation of more than 13db of squeezing , however , pump powers above 600mw stably controlled and within a clean fundamental mode were not available in our experiment .
figure [ wignerfct ] presents the wigner function @xcite of the squeezed vacuum state with variances of -11.5db and 16db with respect to the vacuum noise variance .
the wigner function represents one of several possible phase - space probability distributions for quantum states . since it is gaussian for vacuum - squeezed states , it is particularly easy to obtain simply by measuring its second moments .
the entire statistical properties of the state can than be summarized in a finite dimensional space , namely , by the covariance matrix .
hence , for squeezed vacuum states mixed with the ordinary vacuum state only the variances of the squeezed and anti - squeezed quadrature are required to calculate the wigner function which is then given by @xmath32 where @xmath33 are again the variances of the quadratures @xmath34 and @xmath35 , respectively , with the vacuum noise variance of 1/4 . the vertical axis in fig.[wignerfct ] represents the quasi - probability density of the quadrature phase space .
the wigner function is a phase - space _ quasi_-probability distribution , since the quadratures can not simultaneously have precisely defined values .
what _ can _ be precisely measured is one projection of the wigner function onto the axis of the phase space at a time .
the projections are the probability distributions of the particular quadrature , and here , have a gaussian shape with squeezed or anti - squeezed variances with respect to the vacuum noise variance ( dotted curves ) .
contour lines of the wigner function have an elliptical shape and are thus squeezed in comparison with the circular contour lines of the vacuum state ( not shown in fig . [ wignerfct ] ) .
all photons that are present in a squeezed vacuum state produced by opo stem from parametric down conversion of second harmonic pump photons .
this is also true for a mixture of a squeezed vacuum state with the ordinary vacuum state . for strongly squeezed states
not just the detection of photon pairs but also of higher - order photon pairs is expected .
however , the overall loss of photons must be sufficiently low to enable such an observation . since the wigner function represents the full quantum information of a state it can be used to calculate the state s density matrix @xmath36 in the photon number ( fock ) basis , which is another complete description of the state .
the photon number distribution p(n ) can be obtained from the diagonal elements of the density matrix .
figure [ dm ] shows a plot of the first matrix elements ( absolute values ) up to photon numbers @xmath37 of the observed squeezed state with noise variances of -11.5db and 16db relative to the vacuum noise . in order to calculate the density matrix @xcite photon numbers of up to @xmath38
were considered , which ensured convergence of all the matrix elements .
plot of the absolute values of the density matrix elements with photon numbers up to 10 for the -11.5db squeezed and 16db anti - squeezed state .
the red diagonal elements give the probabilities of finding 0,1,2 n photons .
due to the state s high purity , the probabilities for finding an odd number of photons is significantly suppressed . , scaledwidth=50.0% ] similarly , the density matrices for the additional two states have also been generated .
their diagonal elements , i.e. the photon number distributions of the states , are depicted in fig .
[ photonnumber ] .
the formula used and the numerical values of the density matrix elements are given in the supplementary information .
for the 11.5db squeezed state ( red ) a strong odd / even photon number oscillation is visible .
even the probability of detecting @xmath39 photons is significantly higher than the probability of detecting @xmath40 photons .
this oscillation further continues for greater photon numbers but is not shown here . for our squeezed states with less squeezing
the oscillation flattens out at lower photon numbers , but the odd / even oscillation is even more pronounced at lower photon numbers .
for the 2.9db squeezed state the probability of finding @xmath41 photon is as small as 0.5% .
note that for _ displaced _ squeezed states additional photons are added to the state and the odd photon numbers can not all have zero probability , even if these state were pure .
however , _
weakly _ displaced squeezed states show photon number oscillations with larger periods .
such states were originally theoretically analyzed in @xcite .
+ the photon numbers calculated here are detected per bandwidth in hertz and per detection time interval in seconds . obviously , the higher the squeezing bandwidth the higher the number of photons per second . for applications in high speed quantum communication and information processing high photon counting rates or , equivalently , fields with squeezing over a broad fourier spectrum
are essential .
the half width half maximum ( hwhm ) _ squeezing bandwidth _ can be defined as the spectral width over which the squeezed variance increases from its lowest variance at small fourier frequencies @xmath22 to @xmath42 . here
the variances are again normalized to unity vacuum noise .
although continuous variable quantum states are often quoted to provide the possibility of high bandwidths , only modest values of the order of a few tens of mhz have been realized , e.g. in @xcite .
figure [ broadbandsqueezing ] shows the spectral analysis of the strongly squeezed field from our source with a hwhm squeezing bandwidth of 170mhz .
in contrast to fig .
[ squeezing ] here the electronic dark noise of our homodyne detector was subtracted . at @xmath43100mhz
the dark noise reached the measured quantum noise . at frequencies below 5mhz
the measurement was limited by the resolution bandwidth chosen . since our set - up
did not use control beams which might introduce technical noise at low frequencies , a white squeezing spectrum can be expected for lower frequencies if parasitic interferences with frequency shifted scattered photons are sufficiently suppressed as demonstrated in @xcite .
+ the spectrum of the squeezed and anti - squeezed quadrature variances ( @xmath44 and @xmath45 , respectively ) for the below - threshold opo can be calculated by @xcite @xmath46 where @xmath47 is the power of the second harmonic pump and @xmath48 is the pump power needed to reach the opo threshold .
the detection efficiency @xmath28 is calculated as the product of the propagation efficiency , the square of the homodyne visibility , and the quantum efficiency of the homodyne detector photo diodes .
the opo escape efficiency @xmath49 is determined by the transmittance @xmath50 of the out coupling mirror and the intra - cavity round trip loss @xmath51 .
the dimensionless parameter @xmath52 is the ratio of the fourier frequency @xmath16 and the decay rate @xmath53 of the cavity with round trip length @xmath54 .
the theoretical curves in fig .
[ broadbandsqueezing ] were modeled using equation ( [ sqzspec ] ) , with the best agreement obtained upon assumption of the following values : a pump beam at 53.5% of the threshold power , the output coupler transmittance of 12% , the round - trip loss within the crystal of approximately 0.1% , and a total detection efficiency @xmath55=95.2% in agreement with our analysis presented in section [ subsection1 ] .
the squeezing bandwidth of our source is given by the fourier frequency where the squeezed variance has increased to -2.7db .
according to the theoretical curve in fig . [ broadbandsqueezing ]
this bandwidth is 170mhz .
this result can be used to calculate the rate of down - converted photons from a full longitudinal mode of our source . to this end , we split the half free spectral range of our source of 5.5ghz into 55000 frequency bins of 100khz each . for every bin @xmath56 ,
the photon number distribution @xmath57 was calculated . by summing these distributions
, we find that the spectrally weighted representative photon number distribution of the central opo cavity mode around half the pump light frequency @xmath58 .
this distribution can now be used to calculate the mean number of photons per second and bandwidth averaged over the band of 5.5ghz .
multiplying with 5.5ghz provides the rate of down - converted photons from this mode of as high as @xmath59s@xmath60 , corresponding to a power of 52pw . + another relevant value is given by the mean number of correlated photons in the case of a single photon detection . for a parametric down - conversion source operated far below its threshold and with zero photon loss
this number is two because the detection of a single photon proves the presence of , in total , two photons . in order to calculate this number for our source
we set @xmath61 and calculate the mean value of the renormalized remaining distribution . for the 11.5db squeezed state presented in fig.[broadbandsqueezing ]
we obtain @xmath62 = 5.93 . implying that whenever a single photon detector clicks ( possibly without any photon number resolution ) this event proves the presence of , on average , almost 6 quantum correlated photons , including the detected photon(s ) .
in order to observe strong squeezing levels , we operated an optical parametric oscillator comparatively close to its threshold . in this regime not just single photon pairs , but a significant number of multiple or _ higher order _ photon pairs were produced .
this contrasts typical parametric down - conversion experiments in the photon counting regime where the probability of observing more than two quantum correlated photons is negligible .
the density matrix in the photon number basis of our -11.5db squeezed state not only reveals strong photon number oscillations but also quantum correlations between photons in the following way : whenever a photon was photo - electrically detected by the homodyne detector , there was an almost unity probability that an additional , odd number of photons were physically realized and were detected by the same detector . in photon counting experiments our squeezed state might therefore be used to overcome the problem of insufficient photon counter efficiencies .
the unprecedented quality of our squeezed states will also enable a great step forward in many applications currently under discussion .
in high power laser interferometers for gravitational wave detection @xcite the application of an -11.5db squeezed state reduces the shot noise equivalent to a laser power increase by a factor 14 .
the interference of two such states on a 50/50 beamsplitter would produce strong einstein - podolsky - rosen entanglement @xcite and its high bandwidth may result in the ultra - fast generation of quantum keys @xcite .
our squeezed light source may also find applications in future hybrid quantum information networks .
its well - defined fundamental gaussian spatial mode yielded a high fringe contrast with a spatially filtered local oscillator of 99.8% and enables efficient coupling to other systems , for example to atoms coupled to optical cavities @xcite , in order to realize quantum interfaces of light and matter .
we would like to thank b. hage , j. diguglielmo , a. thring , c. messenger and j. fiurek for many helpful discussions .
this work has been supported by the deutsche forschungsgemeinschaft ( dfg ) through sonderforschungsbereich 407 and the centre for quantum engineering and space - time research quest .
in order to calculate the density matrix elements @xmath63 we followed the prescription given in @xcite , whereby the entries can be obtained using the following equation : @xmath64 + 2\alpha}u^{n-2\alpha}}{a!(n-2a)!\left(a + \frac{m - n}{2}\right)!},\ ] ] where @xmath65 , @xmath66 , @xmath67 and @xmath68 .
note , that @xmath69 and @xmath70 correspond to the linear quadrature variances of the quantum state as introduced in equation ( [ sqzspec ] ) , but with the vacuum variance set to 1/2 .
this expression for the density matrix in the fock state basis can be derived from the covariance matrix by first noting that the husimi q - function in the coherent state basis is the generating function for the density matrix in the fock state basis .
the covariance matrix needs to be transformed from the quadrature basis into the coherent basis to obtain the above formula .
+ we calculated @xmath63 for the three states in fig.[squeezing]a - c . for the first state ( fig.[squeezing]a )
we chose vacuum normalized variances of -2.84db and 2.94db being within our experimental error bars .
these numbers correspond to an originally pure state with 3.05db squeezing and 3.05db anti - squeezing that was degraded by 4.8% of optical loss . for the remaining two states we chose the variances according to the values in fig.[squeezing]b ( -6.2db and 6.7db ) and fig.[squeezing]c ( -11.5db and 16.0db ) .
the calculation was done with photon numbers up to 170 to ensure the convergence of all entries , but the matrices presented below were truncated at n = m=10 . for clarity the diagonal elements representing the photon number distribution ( as shown in fig.[photonnumber ] ) are marked in red .
12 m. xiao , l. a. wu , and h. j. kimble , phys . rev
* 59 , * 278 ( 1987 ) .
p. grangier , r. e. slusher , b. yurke , and a. laporta , phys .
* 59 , * 2153 ( 1987 ) .
k. mckenzie , d. a. shaddock , d. e. mcclelland , b. c. , and p. k. lam , phys .
* 88 , * 231102 ( 2002 ) .
p. walther _
_ , nature * 429 , * 158 ( 2004 )
. h. vahlbruch _
et al . _ ,
* 95 , * 211102 ( 2005 ) .
et al . _ , nature phys . * 4 , * 472 - 476 ( 2008 ) .
a. einstein , b. podolsky , and n. rosen , phys . rev . * 47 , * 777 - 780 ( 1935 ) .
a. aspect , p. grangier , and g. roger , phys .
* 47 , * 460 - 463 ( 1981 ) .
z. y. ou , s. f. pereira , h. j. kimble , and k. c. peng , phys .
* 68 , * 3663 - 3666 ( 1992 ) .
k. wagner _
et al . _ ,
science * 321 , * 541 ( 2008 ) .
d. bouwmeester _
et al . _ ,
nature * 390 , * 575 - 579 ( 1997 ) .
a. furusawa _
_ , science * 282 , * 706 - 709 ( 1998 ) .
w. p. bowen _
a * 67 , * 032302 ( 2003 ) .
a. ourjoumtsev , r. tualle - brouri , j. laurat , and p. grangier , science * 312 , * 83 ( 2006 ) .
j. s. neergaard - nielsen , b. melholt nielsen , c. hettich , k. mlmer , and e. s. polzik , phys .
lett . * 97 , * 083604 ( 2006 ) .
_ , nature phys . * 3 , * 481 - 486 ( 2007 ) .
r. dong _ et al .
_ , nature phys .
* 4 , * 919 - 923 ( 2008 ) .
et al . _ , nature phys . * 4 , * 915 - 918 ( 2008 ) .
d. f. walls , nature * 306 , * 141 ( 1983 ) .
g. breitenbach , s. schiller , and j. mlynek , nature * 387 , * 471 - 475 ( 1997 ) . c. c. gerry , and p. l. knight , _ introductory quantum optics _ ( cambridge univ . press ,
cambridge , 2004 ) .
h. p. yuen , phys .
a * 13 , * 2226 - 2243 ( 1976 ) . v.
v. dodonev , o. v. manko , and v. i. manko , phys .
a * 49 , * 2993 - 2001 ( 1994 ) .
r. dong _ et al .
* 33 , * 116 ( 2008 ) .
y. takeno , m. yukawa , h. yonezawa , and a. furusawa , opt .
express * 15 , * 4321 - 4327 ( 2007 ) .
h. vahlbruch _
et al . _ , phys .
* 100 , * 33602 ( 2008 ) .
a. franzen , b. hage , j. diguglielmo , j. fiurek , and r. , phys .
* 97 , * 150505 ( 2006 ) .
h. vahlbruch _
et al . _ ,
* 97 , * 011101 ( 2006 ) .
h. vahlbruch , s. chelkowski , k. danzmann , and r. schnabel , new j. phys .
* 9 , * 371 ( 2007 ) .
e. p. wigner , phys .
rev . * * 4**0 , 749 - 759 ( 1932 ) . k. vogel and h. risken , phys .
a * * 4**0 , 2847 - 2849 ( 1989 ) .
j. bertrand and p. bertrand , found .
phys * * 1**7 , 397 - 405 ( 1987 ) . s. schiller , g. breitenbach , s. f. pereira , t. mller , and j. mlynek , phys .
* 77 , * 2933 - 2936 ( 1996 ) .
marek _ et al .
a * 76 , * 053820 ( 2007 ) .
w. schleich and j. a. wheeler , nature * 326 , * 574 - 577 ( 1987 ) .
g. breitenbach , t. mller , s. f. pereira , s. schiller , and j. mlynek , j. opt .
b * 12 , * 2304 ( 1995 ) .
s. chelkowski , h. vahlbruch , k. danzmann , and r. schnabel , phys .
rev . a * 75 , * 043814 ( 2007 ) .
j. diguglielmo , b. hage , a. franzen , j. fiurek , and r. , phys .
rev . a * 76 , * 012323 ( 2007 ) .
cerf , m. lvy , and g. van assche , phys .
a * 63 , * 052311 ( 2001 ) .
m. d. lukin , s. f. yelin , and m. fleischhauer , phys .
* 84 , * 4232 - 4235 ( 2000 ) .
j. simon , h. tanji , j. k. thompson , and v. vuleti , phys .
* 98 , * 183601 ( 2007 ) .
p. f. herskind , a. dantan , j. p. marler , m. albert , and m. drewsen , nature phys .
* 5 , * 494 - 498 ( 2009 ) . | squeezed states of light constitute an important nonclassical resource in the field of high - precision measurements , e.g. gravitational wave detection , as well as in the field of quantum information , e.g. for teleportation , quantum cryptography , and distribution of entanglement in quantum computation networks .
strong squeezing in combination with high purity , high bandwidth and high spatial mode quality is desirable in order to achieve significantly improved performances contrasting any classical protocols . here
we report on the observation of the strongest squeezing to date of 11.5db , together with unprecedented high state purity corresponding to a vacuum contribution of less than 5% , and a squeezing bandwidth of about 170mhz .
the analysis of our squeezed states reveals a significant production of higher - order pairs of quantum - correlated photons , and the existence of strong photon number oscillations . |
several groups have worked for the past two decades on the generation of reliable models of low - mass stars , but it was nt until the late 1990s that they arrived to realistic models of these objects .
the models of the group led by baraffe & chabrier are at present the most widely used ones , since they can reproduce very well many of the observational properties of low - mass stars .
for example , the mass - magnitude and the mass - luminosity relations of these stars are very nicely reproduced by the baraffe et al .
( 1998 ) models . those models , however , still have some problems reproducing the effective temperature scale and the mass - radius relation of these stars . in the case of the @xmath3 scale ,
baraffe et al . (
1998 ) find that at temperatures below @xmath2 3700k the models predict bluer v
i colors than the ones observed .
a possible reason provided by the authors for this mismatch is a missing source of opacity in the optical that causes the stars to be fainter in v than what the models predict . for the mass radius relation
, the models underestimate the radii of the stars by at least 10 % .
this conclusion is based on the observational results from eclipsing binaries with errorbars of 3 % or less ( see figure 1 ) .
the problem may be that the `` standard models '' do not include the effect of magnetic fields .
mullan & macdonald ( 2001 ) find that low - mass star models have larger radii and smaller @xmath3 when magnetic fields are taken into account .
magnetic fields are generally enhanced by stellar rotation , and in close binaries ( where we are measuring systematically larger radii ) the stars are spun up by orbital synchronization .
with the current observational techniques , double - lined detached eclipsing binaries are the only objects where we can measure simultaneously the mass and the radius of stars with error bars of less than 23 % . the technique is a well established one : the radial velocity ( rv ) curves of the binaries provide the masses as a function of the orbital inclination of the system . from their light curves ( lcs ) one
can then measure the orbital inclination of the system and the radius of each star . also , by measuring the lcs at different wavelengths one can estimate the effective temperature of the stars .
we have searched to date five photometry databases ( see companion paper in this proceedings by shaw & lpez - morales ) .
the result of that search are 41 new detached eclipsing binaries with masses below 1@xmath0 . after identifying the binaries from the lcs in those databases ,
we need to conduct follow - up observational campaigns to measure the optical and infrared light curves of the systems and their radial velocity curves .
this is an extensive project that requires of large amounts of telescope time .
currently we have been awarded time in the facilities listed in table 1 .
our final goal is to obtain full , high quality lcs and rv curves to be able to determine the masses and the radii of the stars in those binaries with errors smaller than 3% .
we have completed to date the optical ( vri ) light curves and radial velocity curves of three binaries : gu boo ( lpez - morales & ribas 2005 ) , rxj0239.1 ( torres et al .
, in prep ) , and nsvs01031772 ( hereafter nsvs0103 ; lpez - morales et al . , submitted ) .
near - ir light curves are also available for rxj0239.1 .
table 2 summarizes the masses , radii , and temperatures derived for the components of each binary .
the two stars in gu boo are almost identical to the stars in yy gem .
the stars in the other two binaries , with masses between 0.5 and 0.55 @xmath0 and 0.7 and 0.73 @xmath0 respectively , fill - in two current gaps in the mass - radius relation .
figure 1 shows the mass - radius relation of stars below 1@xmath0 .
the lines represent the predictions of different models , using 0.35 gyr isochrones and a metallicity z = 0.02 .
the open circles correspond to the previously known binaries cm dra ( lacy 1977 ; metcalfe et al .
1996 ) , cu cnc ( delfosse et al . 1999 ; ribas 2003 ) , tres - her0 - 07621 ( creevey et al .
2005 ) , and yy gem ( leung & schneider 1978 ; torres & ribas 2002 ) .
the filled squares show the location in this diagram of the components of gu boo , rxj0239.1 , and nsvs0103 . except for tres - her0 - 07621 ,
all the other stars show a clear trend towards larger radii than what the models predict .
all the stars in binaries are at least 10% larger than what any of the models predict .
figure 2 shows the mass
log(@xmath3 ) relation for gu boo , rxj0239.1 , and nsvs0103 ( open circles ) , yy gem ( filled circle ) , and cu cnc ( open triangles ) .
the top figure corresponds to a metallicity of z=0.01 , the bottom figure is for a metallicity of z=0.02 .
the age of both sets of isochrones is 0.35 gyrs .
the bottom figure ( z=0.02 ) agrees with the trend observed by baraffe et al .
( 1998 ) , where they find that below 37003800k the effective temperatures predicted by the models are larger than the ones observed in low - mass stars .
we present in this paper the first results of an extensive observing campaign primarily aimed at providing an accurate empirical m - r relation for low - mass stars .
our targets are low - mass eclipsing binaries , from where precise stellar masses and radii can be derived .
these systems also provide an estimation of the @xmath3 of the stars .
our current sample contains 41 new binaries with masses between 0.35 and 1.0@xmath0 . here
we present the parameters of the first three of those binaries , gu boo , rxj0239.1 , and nsvs0103 , which provide six new valuable data points .
the addition of those new data points to the mass radius relation diagram ( see figure 1 ) strengthens the trend already suggested by the other binaries ( cm dra , cu cnc , and yy gem ) .
that is , the models underestimate the radii of low - mass stars by at least 10 % .
this is at least the case for the components of binaries .
the few available measurements from single stars present a much larger scatter and larger error bars , preventing the identification of a clear trend , if any in fact exists .
a mismatch is also apparent when the @xmath3 of the binaries are compared to the predictions by the models ( see figure 2 ) . in this case the temperatures are lower than the prediction by the models below @xmath2 37003800k , as already noticed by baraffe et al .
( 1998 ) .
observations begin to reveal what appear to be two different populations of low - mass stars : non - active stars , whose parameters would be succesfully reproduced by the `` standard models '' ( i.e. those not including magnetic fields ) , and active stars , where the magnetic fields play an important role .
+ + + * acknowledgments*. we are grateful to the following institutions for providing support for this work : carnegie institution of washington , nasa astrobiology institute , southeastern association for research in astronomy , university of georgia at athens , instituto de astrofsica de canarias , national science fundation , and the university of north carolina at chapel hill .
we also thank our collaborators jerome a. orosz ( sdsu , usa ) , ignasi ribas ( ieec , spain ) , maria jess arvalo and carlos lzaro ( iac , spain ) , and guillermo torres ( harvard cfa , usa ) for their time and efforts towards this project
. m. l - m . acknowledges research and travel support from the carnegie institution of washington through a carnegie fellowship .
lll technique&&facility + photometry:&optical:&0.9 m @xmath12 telescope ( @xmath13 , usa ) + & & 1.0 m @xmath14 consortium telescope ( @xmath15 , chile ) + & & 2.5 m dupont telescope ( @xmath16 , chile ) + & near ir:&1.5 m carlos snchez telescope ( @xmath17 , spain ) + & & 2.5 m dupont telescope ( @xmath16 , chile ) + spectroscopy:&&2.5 m dupont telescope ( @xmath16 , chile ) + & & 4.0 m mayall telescope ( @xmath13 , usa ) + & & 6.5 m magellan telescopes ( @xmath16 , chile ) + lcccccc binary&m1 ( msun)&m2 ( msun)&r1 ( rsun)&r2 ( rsun)&teff1 ( k)&teff2 ( k ) + @xmath18&0.610 @xmath19 0.007&0.599 @xmath19 0.006&0.623 @xmath19 0.016&0.626 @xmath19 0.020&3920 @xmath19 130&3810 @xmath19 130 + @xmath20&0.730 @xmath19 0.009&0.693 @xmath19 0.006&0.741 @xmath19 0.004&0.703 @xmath19 0.002&4645 @xmath19 20&4275 @xmath19 15 + @xmath21&0.530 @xmath19 0.014&0.514 @xmath19 0.013&0.559 @xmath19 0.014&0.518 @xmath19 0.013&3750 @xmath19 150&3600 @xmath19 150 + | full tests to stellar models below 1@xmath0 have been hindered until now by the scarce number of precise measurements of the stars most fundamental parameters : their masses and radii . with the current observational techniques , the required precision to distinguish between different models ( errors @xmath1 2 - 3 % )
can only be achieved using detached eclipsing binaries where 1 ) both stars are similar in mass , i.e. q = m1/m2 @xmath2 1.0 , and 2 ) each star is a main sequence object below 1@xmath0 . until 2003 only three
such binaries had been found and analyzed in detail .
two new systems were published in 2005 ( creevey et al . ; lpez - morales & ribas ) , almost doubling the previous number of data points .
here we present preliminary results for 3 new low - mass detached eclipsing binaries .
these are the first studied systems from our sample of 41 new binaries ( shaw & lpez - morales , this proceedings ) .
we also provide an updated comparison between the mass radius and the mass@xmath3 relations predicted by the models and the observational data from detached eclipsing binaries .
we define low - mass stars as main sequence stars with masses between 1@xmath0 and the hydrogen burning limit ( 0.070.08@xmath0 ) .
low - mass stars are small cool objects , with radii between 1.0 and 0.1@xmath4 and effective surface temperatures between 6000 and 2500 k. they are also faint as their luminosities between 1 and @xmath5@xmath6 reveal .
they are the most abundant stars in the galaxy , where least 7 of every 10 stars are low - mass main sequence stars .
these objects play a role on studies of baryonic dark matter ( it is thought that low - mass stars , brown dwarfs , and stellar remnants are the main contributors to the baryonic dark matter in the universe ) , on dynamical studies of galaxies and star clusters , and on the detailed characterization of the stars in the solar neighborhood , where intensive searches for earth - like planets around low - mass stars are currently underway .
low - mass stars are also a subject of interest in other fields in physics , given the complicated physical processes that are taking place in these stars .
@xmath7 molecules , tio , @xmath7o , co , and cn become stable in the atmosphere of low - mass stars at temperatures below 50004000 k. below 2800 k , even more complex molecular compounds , such as @xmath8 , @xmath9 , and @xmath10 , also become stable . therefore any model trying to reproduce how the radiation is transmitted from the interior of the stars through their atmospheres
needs to take into account the effect of all these molecules and compounds .
the interior physics of low - mass stars is also challenging .
the interior of these stars is essentially a plasma of ionized h and he under partially or fully degenerate conditions , so one can not model them using ideal equations of state ( eoss ) . instead , we need more elaborate eoss , that include coulomb interactions between electrons and ions , ionization pressure effects , the effect of the electric fields generated by the ions , and so on . |
in spectral analysis , one of the most powerful tools is the method of the conjugate operator , also called mourre s commutator method after the seminal work of mourre in the early eighties .
this approach has reached a very high degree of precision and abstraction in @xcite ; see also @xcite for further developments . in order to study the nature of the spectrum of a selfadjoint operator @xmath0 , the main idea of the standard method
is to find an auxiliary selfadjoint operator @xmath1 such that the commutator @xmath2 $ ] is strictly positive when localized in some interval of the spectrum of @xmath0 . more precisely , one looks for intervals @xmath3 of @xmath4 such that @xmath5e(j)\geq a e(j)\ ] ] for some strictly positive constant @xmath6 that depends on @xmath3 , where @xmath7 denotes the spectral projection of @xmath0 on the interval @xmath3 . an additional compact contribution to ( [ mou ] ) is allowed , greatly enlarging the range of applications .
when strict positivity is not available , one can instead look for an @xmath1 such that the commutator is positive and injective , _
i.e. _ @xmath8>0.\ ] ] this requirement is close to the one of the kato - putnam theorem , _ cf . _
xiii.28 ) .
a new commutator method based on such an inequality was proposed in @xcite . by analogy to the method of the conjugate operator , it has been called the method of the weakly conjugate operator ( mwco ) . under some technical assumptions ,
both approaches lead to a limiting absorption principle , that is , a control of the resolvent of @xmath0 near the real axis . in the case of the usual method of the conjugate operator , this result is obtained locally in @xmath3 , and away from thresholds .
the mwco establishes the existence of the boundary value of resolvent also at thresholds , but originally applies only to situations where the operator @xmath0 has a purely absolutely continuous spectrum .
this drawback limits drastically the range of applications .
however , in some recent works @xcite the mwco has also been applied successfully to examples with point spectrum .
this review intend to present and illustrate some of these extensions through applications to the spectral theory of operators acting on groups and graphs .
compared to the huge number of applications based on an inequality of the form , the number of papers that contain applications of the mwco is very small .
let us cite for example the works @xcite that deal with the original form of the theory , and the papers @xcite that contain very close arguments .
the derivation of the limiting absorption principle of @xcite has been abstracted in @xcite .
the framework of @xcite is still the one of the mwco , but since its result applies a certain class of two - body schrdinger operators which have bound states below zero , it can also be considered as the first extension of the mwco dealing with operators that are not purely absolutely continuous .
the main idea of @xcite is that @xmath0 itself can add some positivity to .
the new requirement is the existence of a constant @xmath9 such that @xmath10>0.\ ] ] this inequality , together with some technical assumptions , lead to a limiting absorption principle which is either uniform on @xmath4 if @xmath11 or uniform on @xmath12 if @xmath13 .
the extensions of the mwco developed in @xcite are of a different nature . in these papers , the operators @xmath0 under consideration admit a natural conjugate operator @xmath1 that fulfills the inequality @xmath2\geq0 $ ] , namely , the commutator is positive but injectivity may fail . in that situation ,
the authors considered a decomposition @xmath14 , and the restrictions of @xmath0 and @xmath2 $ ] to these subspaces . in favorable circumstances
the injectivity can be restored in one of the subspace , and a comprehensible description of the vectors of the second subspace can be given .
this decomposition leads again to statements that are close to the ones of the mwco , but which apply to operators with arbitrary spectrum .
this extension is described in section [ ext ] .
we would also like to mention the references @xcite .
they pereform what can be considered as a unitary version of the mwco and extend the kato - putnam analysis of unitary operators to the case of unbounded conjugate operators . the main applications concern time - depending propagators .
the content of this review paper is the following . in section [ mofc ]
we recall the original method of the weakly conjugate operator , and then present an abstract version of the approach used in @xcite . in section [ secad ] we present applications of this approach to the study of adjacency operators on graphs .
a similar analysis for operators of convolution on locally compact groups is performed in section [ seclcg ] .
let us finally fix some notations . given a selfadjoint operator @xmath0 in a hilbert space @xmath15
, we write @xmath16 , @xmath17 , @xmath18 , @xmath19 and @xmath20 respectively for the continuous , absolutely continuous , singularly continuous , singular and pure point subspaces of @xmath15 with respect to @xmath0 .
the corresponding parts of the spectrum of @xmath0 are denoted by @xmath21 , @xmath22 , @xmath23 , @xmath24 and @xmath25
. * acknowledgements .
* m. m. was partially supported by the chilean science foundation _
fondecyt _ under the grant 1085162 .
s. r. and r. t. d a. thank the swiss national science foundation for financial support .
in this section we recall the basic characteristics of the method of the weakly conjugate operator , as originally introduced and applied to partial differential operators in @xcite .
we then present the abstract form of the extension developed in @xcite .
the method works for unbounded operators , but for our purposes it is enough to assume @xmath0 bounded .
we start by introducing some notations .
the symbol @xmath15 stands for a hilbert space with scalar product @xmath26 and norm @xmath27 .
given two hilbert spaces @xmath28 and @xmath29 , we denote by @xmath30 the set of bounded operators from @xmath28 to @xmath29 , and put @xmath31 .
we assume that @xmath15 is endowed with a strongly continuous unitary group @xmath32 .
its selfadjoint generator is denoted by @xmath1 and has domain @xmath33 . in most of the applications
@xmath1 is unbounded .
[ defc1 ] a bounded selfadjoint operator @xmath0 in @xmath15 belongs to @xmath34 if one of the following equivalent condition is satisfied : 1 . the map @xmath35 is strongly differentiable , 2 .
the sesquilinear form @xmath36 is continuous when @xmath33 is endowed with the topology of @xmath15 .
we denote by @xmath37 the strong derivative in ( i ) , or equivalently the bounded selfadjoint operator associated with the extension of the form in ( ii ) .
the operator @xmath37 provides a rigorous meaning to the commutator @xmath2 $ ] .
we write @xmath38 if @xmath37 is positive and injective , namely if @xmath39 for all @xmath40 .
the operator @xmath1 is _ weakly conjugate to _ the bounded selfadjoint operator @xmath0 if @xmath41 and @xmath42>0 $ ] . for @xmath38 let us consider the completion @xmath43 of @xmath15 with respect to the norm @xmath44 .
the adjoint space @xmath45 of @xmath43 can be identified with the completion of @xmath46 with respect to the norm @xmath47 .
one has then the continuous dense embeddings @xmath48 , and @xmath37 extends to an isometric operator from @xmath43 to @xmath45 . due to these embeddings
it makes sense to assume that @xmath49 restricts to a @xmath50-group in @xmath45 , or equivalently that it extends to a @xmath50-group in @xmath43 . under this assumption
( tacitly assumed in the sequel ) we keep the same notation for these @xmath50-groups .
the domain of the generator of the @xmath50-group in @xmath43 ( resp .
@xmath45 ) endowed with the graph norm is denoted by @xmath51 ( resp .
@xmath52 ) . in analogy with definition [ defc1 ]
the requirement @xmath53 means that the map @xmath54 is strongly differentiable , or equivalently that the sesquilinear form @xmath55 is continuous when @xmath51 is endowed with the topology of @xmath43 . here
, @xmath26 denotes the duality between @xmath43 and @xmath45 .
finally let @xmath56 be the banach space @xmath57 defined by real interpolation , see for example ( * ? ? ?
* proposition 2.7.3 ) .
one has then the natural continuous embeddings @xmath58 and the following results ( * ? ? ?
* theorem 2.1 ) : [ thmabstract ] assume that @xmath1 is weakly conjugate to @xmath0 and that @xmath42 $ ] belongs to @xmath59
. then there exists a constant @xmath60 such that @xmath61 for all @xmath62 , @xmath63 and @xmath64 .
in particular the spectrum of @xmath0 is purely absolutely continuous .
the global limiting absorption principle plays an important role for partial differential operators , _ cf .
_ @xcite .
however , for the examples included sections 3 and 4 , the space @xmath56 involved has a rather obscure meaning and can not be greatly simplified , so we shall only state the spectral results .
the extension proposed in @xcite relies on the following observation .
assume that @xmath0 is a bounded selfadjoint operator in a hilbert space @xmath15 .
assume also that there exists a selfadjoint operator @xmath1 such that @xmath65 , and that @xmath66=k^2\ ] ] for some selfadjoint operator @xmath67 , which is bounded by hypothesis .
it follows that @xmath2\geq0 $ ] but injectivity may not be satisfied .
so let us introduce the decomposition of the hilbert space @xmath68 with @xmath69 .
the operator @xmath37 is reduced by this decomposition and its restriction @xmath70 to @xmath71 satisfies @xmath72 .
formally , the positivity and injectivity of @xmath70 are rather promising , but are obviously not sufficient for a direct application of the mwco to @xmath70 .
however , let us already observe that some informations on @xmath73 can be inferred from .
indeed , it follows from the virial theorem ( * ? ? ?
7.2.10 ) that any eigenvector @xmath74 of @xmath0 satisfies @xmath75f\>=0 $ ] .
so @xmath76 , _ i.e. _ @xmath77 , and
one has proved : if @xmath78 and @xmath2=k^2 $ ] , then @xmath79 .
let us now come back to the analysis of @xmath70 .
the space @xmath43 can still be defined in analogy with what has been presented in the previous section : @xmath43 is the completion of @xmath71 with respect to the norm @xmath44 .
then , the adjoint space @xmath45 of @xmath43 can be identified with the completion of @xmath80 with respect to the norm @xmath47 .
but in order to go further on , some compatibility has be imposed between the decomposition of the hilbert space and the operators @xmath0 and @xmath1 .
let us assume that both operators @xmath0 and @xmath1 are reduced by the decomposition @xmath81 of @xmath15 , and let @xmath82 and @xmath83 denote their respective restriction to @xmath71 .
it clearly follows from the above lemma that @xmath82 has no point spectrum .
we are now in a suitable position to rephrase theorem [ thmabstract ] in our present framework .
we freely use the notations introduced above .
[ mo ] assume that @xmath78 and that @xmath84=k^2 $ ] .
assume furthermore that both operators @xmath0 and @xmath1 are reduced by the decomposition @xmath85 of @xmath15 and that @xmath70 belongs to @xmath86 .
then a limiting absorption principle holds for @xmath82 uniformly on @xmath4 , and in particular the spectrum of @xmath82 is purely absolutely continuous .
a straightforward consequence of this statement is that @xmath87 we will see in the applications below that theorem [ mo ] applies to various situations , and that it really is a useful extension of the original method of the weakly conjugate operator .
we present in this section the results of @xcite on the spectral analysis for adjacency operators on graphs .
we follow the notations and conventions of this paper regarding graph theory .
a _ graph _ is a couple @xmath88 formed of a non - void countable set @xmath89 and a symmetric relation @xmath90 on @xmath89 such that @xmath91 implies @xmath92 .
the points @xmath93 are called vertices and couples @xmath94 such that @xmath91 are called edges .
so , for simplicity , multiple edges and loops are forbidden in our definition of a graph .
occasionally @xmath88 is said to be a simple graph . for any @xmath93
we denote by @xmath95 the set of neighbours of @xmath96 .
we write @xmath97 for the degree or valence of the vertex @xmath96 and @xmath98 for the degree of the graph .
we also suppose that @xmath88 is uniformly locally finite , _
i.e. _ that @xmath99 .
a path from @xmath96 to @xmath100 is a sequence @xmath101 of elements of @xmath89 , usually denoted by @xmath102 , such that @xmath103 , @xmath104 and @xmath105 for each @xmath106 . throughout this section
we restrict ourselves to graphs @xmath88 which are simple , infinite countable and uniformly locally finite .
given such a graph we consider the adjacency operator @xmath0 acting in the hilbert space @xmath107 as @xmath108 due to ( * ? ? ?
* theorem 3.1 ) , @xmath0 is a bounded selfadjoint operator with @xmath109 and spectrum @xmath110 $ ] .
results on the nature of the spectrum of adjacency operators on graphs are quite sparse . some absolutely continuous examples are given in @xcite , including the lattice @xmath111 and homogeneous trees .
for cases in which singular components are present we refer to @xcite , @xcite , @xcite and @xcite .
we now introduce the key concept of @xcite .
sums over the empty set are zero by convention .
[ adaptat ] a function @xmath112 is semi - adapted to the graph @xmath88 if 1
. there exists @xmath113 such that @xmath114 for all @xmath115 with @xmath91 , 2 . for any @xmath115 one
has @xmath116=0.\ ] ] if in addition for any @xmath115 one has @xmath117\[\phi(z)-\phi(y)\ ] \[2\phi(z)-\phi(x)-\phi(y)\]=0,\ ] ] then @xmath118 is adapted to the graph @xmath88 . for a function @xmath118
semi - adapted to the graph @xmath88 we consider in @xmath15 the operator @xmath67 given by @xmath119f(y),\quad f\in\h,~x\in x.\ ] ] the operator @xmath67 is selfadjoint and bounded due to the condition ( i ) of definition [ adaptat ] .
it commutes with @xmath0 , as a consequence of condition ( [ semi ] ) .
we also decompose the hilbert space @xmath15 into the direct sum @xmath120 , where @xmath71 is the closure of the range @xmath121 of @xmath67 , thus the orthogonal complement of the closed subspace @xmath122 it is shown in ( * ? ? ?
4 ) that @xmath0 and @xmath67 are reduced by this decomposition , and that their restrictions @xmath82 and @xmath123 to the hilbert space @xmath71 are bounded selfadjoint operators .
a rather straightforward application of the general theory presented in [ ext ] gives [ unu ] assume that @xmath118 is a function semi - adapted to the graph @xmath88 .
then @xmath82 has no point spectrum .
[ doi ] let @xmath118 be a function adapted to the graph @xmath88 .
then the operator @xmath82 has a purely absolutely continuous spectrum .
the role of the weakly conjugate operator is played by @xmath124 and the assumptions imposed on @xmath118 make the general theory work . for a certain class of admissible graphs ,
the result of theorem [ doi ] on the restriction @xmath82 can be turned into a statement on the original adjacency operator @xmath0 .
the notion of admissibility requires ( among other things ) the graph to be directed .
thus the family of neighbors @xmath125 is divided into two disjoint sets @xmath126 ( fathers ) and @xmath127 ( sons ) , @xmath128 .
we write @xmath129 if @xmath130 and @xmath131 if @xmath132 . on drawings
, we set an arrow from @xmath100 to @xmath96 ( @xmath133 ) if @xmath131 , and say that the edge from @xmath100 to @xmath96 is positively oriented .
we assume that the directed graph subjacent to @xmath89 , from now on denoted by @xmath134 , is _
admissible _ with respect to these decompositions , _
i.e. _ ( i ) it admits a position function and ( ii ) it is uniform .
_ a position function _ is a function @xmath135 such that @xmath136 whenever @xmath129 .
it is easy to see that it exists if and only if all paths between two points have the same index ( which is the difference between the number of positively and negatively oriented edges ) .
the directed graph @xmath134 is called _ uniform _ if for any @xmath115 the number @xmath137 $ ] of common fathers of @xmath96 and @xmath100 equals the number @xmath138 $ ] of common sons of @xmath96 and @xmath100 .
thus the admissibility of a directed graph is an explicit property that can be checked directly , without making any choice . the graph @xmath88 is admissible if there exists an admissible directed graph subjacent to it .
[ first ] the adjacency operator of an admissible graph @xmath88 is purely absolutely continuous , except at the origin , where it may have an eigenvalue with eigenspace @xmath139 many examples of periodic graphs , both admissible and non - admissible , are presented in ( * ? ? ?
6 ) . in particular , it is explained that periodicity does not lead automatically to absolute continuity , especially ( but not only ) if the number of orbits is infinite .
@xmath140-products of graphs , as well as the graph associated with the one - dimensional xy hamiltonian , are also treated in @xcite .
we recall in figures [ graph2 ] , [ graph0 ] , and [ graph7 ] some two - dimensional @xmath141-periodic examples taken from ( * ? ? ?
more involved , @xmath142-periodic situations are also available .
in this section we consider locally compact groups @xmath89 , abelian or not , and convolution operators @xmath143 , acting on @xmath144 , defined by suitable measures @xmath145 belonging to @xmath146 , the banach @xmath147-algebra of complex bounded radon measures on @xmath89 .
using the method of the weakly conjugate operator , we determine subspaces @xmath148 and @xmath149 of @xmath144 , explicitly defined in terms of @xmath145 and the family @xmath150 of continuous group morphisms @xmath151 , such that @xmath152 and @xmath153 .
this result , obtained in @xcite , supplements other works on the spectral theory of operators on groups and graphs @xcite .
let @xmath89 be a locally compact group ( lcg ) with identity @xmath154 , center @xmath155 and modular function @xmath156 .
let us fix a left haar measure @xmath157 on @xmath89 , using the notation @xmath158 . on discrete groups
the counting measure ( assigning mass @xmath159 to every point ) is considered .
the notation _ a.e .
_ stands for almost everywhere " and refers to the haar measure @xmath157 .
we consider in the sequel the convolution operator @xmath143 , @xmath160 , acting in the hilbert space @xmath161 , _
@xmath162 where @xmath163 and @xmath164 the operator @xmath143 is bounded with norm @xmath165 , and it admits an adjoint operator @xmath166 equal to @xmath167 , the convolution operator by @xmath168 defined by @xmath169 . if the measure @xmath145 is absolutely continuous w.r.t .
the left haar measure @xmath157 , so that @xmath170 with @xmath171 , then @xmath172 is also absolutely continuous w.r.t . @xmath157 and @xmath173 , where @xmath174 for _ a.e . _ @xmath175 .
in such a case we simply write @xmath176 for @xmath177 .
we shall always assume that @xmath143 is selfadjoint , that is @xmath178 .
given @xmath160 , let @xmath179 be such that the linear functional @xmath180 is bounded .
then there exists a unique measure in @xmath181 associated to @xmath182 , due to the riesz - markov representation theorem .
we write @xmath183 for this measure , and we simply say that @xmath184 is such that @xmath185 .
we call _ real character _ any continuous group morphism @xmath151 . 1 .
a real character @xmath118 is _ semi - adapted to @xmath187 _ if @xmath188 , and @xmath189 .
the set of real characters that are semi - adapted to @xmath145 is denoted by @xmath190 .
a real character @xmath118 is _ adapted to @xmath145 _ if @xmath118 is semi - adapted to @xmath191 , and @xmath192 .
the set of real characters that are adapted to @xmath145 is denoted by @xmath193 .
the cases @xmath197 or @xmath198 are interesting ; in the first case @xmath143 has no eigenvalues , and in the second case @xmath143 is purely absolutely continuous . a more precise result is obtained in a particular situation .
[ precis ] let @xmath89 be a lcg and let @xmath186 .
assume that there exists a real character @xmath118 adapted to @xmath145 such that @xmath199 is equal to a nonzero constant on @xmath200 .
then @xmath143 has a purely absolutely continuous spectrum , with the possible exception of an eigenvalue located at the origin , with eigenspace @xmath201 .
corollary [ precis ] specially applies to adjacency operators on certain classes of cayley graphs , which are hecke - type operators in the regular representation , thus convolution operators on discrete groups . to see how the method of the weakly conjugate operator comes into play ,
let us sketch the proof of the second inclusion in theorem [ princ ] . in quantum mechanics in @xmath202 , the position operators
@xmath203 and the momentum operators @xmath204 satisfy the relation @xmath205 $ ] with @xmath206 , and the usual conjugate operator is the generator of dilations @xmath207 .
so if we regard @xmath208 as a position operator on @xmath89 , it is reasonable to think of @xmath209 $ ] as the corresponding momentum operator and to use @xmath210 as a tentative conjugate operator . in fact , simple calculations using the hypotheses of definition [ admis].(b ) show that @xmath211=k^2.\ ] ] therefore @xmath212 and the commutator @xmath213 $ ] is a positive operator .
however , in order to apply the theory of section [ mofc ] , we need strict positivity .
so , we consider the subspace @xmath214^\perp$ ] of @xmath15 , where @xmath213\equiv k^2 $ ] is strictly positive .
the orthogonal decomposition @xmath14 with @xmath69 , reduces @xmath143 , @xmath67 , and @xmath1 , and their restrictions @xmath82 , @xmath123 , and @xmath83 to @xmath71 are selfadjoint .
it turns out that all the other conditions necessary to apply theorem [ mo ] can also be verified in the hilbert space @xmath71 .
so we get the inclusion @xmath215 since @xmath208 is arbitrary , this implies the second inclusion of theorem [ princ ] .
the construction of weakly conjugate operators for @xmath143 relies on real characters .
so a small vector space @xmath216 is an obstacle to applying the method . a real character @xmath118 maps compact subgroups of @xmath89 to the unique compact subgroup @xmath217 of @xmath218 .
consequently , abundancy of compact elements ( elements @xmath93 generating compact subgroups ) prevents us from constructing weakly conjugate operators .
the extreme case is when @xmath89 is itself compact , so that @xmath219 .
actually in such a case all convolution operators @xmath143 have pure point spectrum .
we review now briefly some of the groups for which we succeeded in @xcite to apply the mwco .
if the group @xmath89 is unimodular , one can exploit commutativity in a non - commutative setting by using central measures ( _ i.e. _ elements of the center @xmath220 $ ] of the convolution banach @xmath147-algebra @xmath146 ) .
for instance , in the case of central groups @xcite , we have the following result ( @xmath221 stands for the closed subgroup generated by the set of compact elements of @xmath89 ) : [ centreaza ] let @xmath89 be a central group and @xmath222 a central measure such that @xmath223 is compact and not included in @xmath221
. let @xmath224 with @xmath225 and set @xmath226
. then @xmath227 .
let @xmath228 , where @xmath229 is the symmetric group of degree @xmath230 .
the group @xmath229 has a presentation @xmath231 , and its conjugacy classes are @xmath232 , @xmath233 and @xmath234 . set @xmath235 and choose @xmath236 two finite symmetric subsets of @xmath141 , each of them containing at least two elements .
then @xmath237 if @xmath238 .
let @xmath239 , where @xmath240 is the group ( with haar measure @xmath241 ) of @xmath242 unitary matrices of determinant @xmath243 .
for each @xmath244 $ ] let @xmath245 be the conjugacy class of the matrix @xmath246 in @xmath240 .
a direct calculation ( using for instance euler angles ) shows that @xmath247 for each @xmath248 $ ] with nonzero lebesgue measure .
set @xmath249 , @xmath250 , @xmath251 , @xmath252 , and @xmath253 .
then @xmath237 if @xmath238 . let @xmath89 be a central group , let @xmath254 , and set @xmath255 for some @xmath224 with @xmath225
. then @xmath145 satisfies the hypotheses of proposition [ centreaza ] , and we can choose @xmath256 such that @xmath257 ( note in particular that @xmath258 iff @xmath259 and that @xmath260 ) . thus @xmath261 . but @xmath163 belongs to @xmath262 iff @xmath263 for _ a.e .
_ @xmath93 .
this periodicity w.r.t .
the non - compact element @xmath264 implies that the @xmath265-function @xmath74 should vanish _
_ and thus that @xmath266 . if @xmath89 is abelian all the commutation relations in definition [ admis ] are satisfied
. moreover one can use the fourier transform @xmath267 to map unitarily @xmath143 on the operator @xmath268 of multiplication with @xmath269 on the dual group @xmath270 of @xmath89 .
so one gets from from theorem [ princ ] a general lemma on muliplication operators .
we recall some definitions before stating it .
the function @xmath271 is _ differentiable at @xmath272 along the one - parameter subgroup @xmath273 _ if the function @xmath274 is differentiable at @xmath275 . in such a case we write @xmath276 for @xmath277 .
higher order derivatives , when existing , are denoted by @xmath278 , @xmath279 .
we say that the one - parameter subgroup @xmath280 is in @xmath281 if @xmath282 is twice differentiable w.r.t . @xmath184 and @xmath283 . if , in addition , @xmath282 is three times differentiable w.r.t . @xmath184 and @xmath284 too , we say that @xmath184 belongs to @xmath285 .
we end up this section by considering a class of semidirect products .
let @xmath291 be two discrete groups with @xmath292 abelian ( for which we use additive notations ) , and let @xmath293 be a group morphism .
let @xmath294 be the @xmath295-semidirect produt of @xmath296 by @xmath292 .
the multiplication in @xmath89 is defined by @xmath297 so that @xmath298 in this situation it is shown in @xcite that many convolution operators @xmath176 , with @xmath299 of finite support , have a non - trivial absolutely continuous component .
for instance , we have the following for a type of wreath products .
take @xmath292 a discrete abelian group and put @xmath300 , where @xmath301 is an arbitrary discrete group and @xmath3 is a finite set on which @xmath292 acts by @xmath302
. then @xmath303 defines an action of @xmath292 on @xmath304 , thus we can construct the semidirect product @xmath305 . if @xmath306 and @xmath307 are finite subsets with @xmath308\ne\varnothing$ ] , then @xmath309 satisfies all the conditions of ( * ? ? ?
thus @xmath310 if @xmath311 . a. boutet de monvel and m. mntoiu .
the method of the weakly conjugate operator . in _
inverse and algebraic quantum scattering theory ( lake balaton , 1996 ) _ , volume 488 of _ lecture notes in phys .
_ , pages 204226 , springer , berlin , 1997 .
j. breuer .
singular continuous and dense point spectrum for sparse tree with finite dimensions . in _ probability and mathematical physics _
d. dawson , v. jaksic and b. vainberg ) , crm proc . and
lecture notes 42 , pages 6583 , 2007 . | in this review we present some recent extensions of the method of the weakly conjugate operator .
we illustrate these developments through examples of operators on graphs and groups . |
even though the set - sharing domain is , in a sense , remarkably precise , more precision is attainable by combining it with other domains .
in particular , freeness and linearity information has received much attention by the literature on sharing analysis ( recall that a variable is said to be free if it is not bound to a non - variable term ; it is linear if it is not bound to a term containing multiple occurrences of another variable ) .
as argued informally by sndergaard @xcite , the mutual interaction between linearity and aliasing information can improve the accuracy of a sharing analysis .
this observation has been formally applied in @xcite to the specification of the abstract @xmath0 operator for the domain @xmath1 . in his phd thesis
@xcite , langen proposed a similar integration with linearity , but for the set - sharing domain .
he has also shown how the aliasing information allows to compute freeness with a good degree of accuracy ( however , freeness information was not exploited to improve aliasing ) .
king @xcite has also shown how a more refined tracking of linearity allows for further precision improvements .
the synergy attainable from a bi - directional interaction between aliasing and freeness information was initially pointed out by muthukumar and hermenegildo @xcite . since then , several authors considered the integration of set - sharing with freeness , sometimes also including additional explicit structural information @xcite .
building on the results obtained in @xcite , @xcite and @xcite , but independently from @xcite , hans and winkler @xcite proposed a combined integration of freeness and linearity information with set - sharing .
similar combinations have been proposed in @xcite . from a more pragmatic point of view ,
codish et al .
@xcite integrate the information captured by the domains of @xcite and @xcite by performing the analysis with both domains at the same time , exchanging information between the two components at each step .
most of the above proposals differ in the carrier of the underlying abstract domain .
even when considering the simplest domain combinations where explicit structural information is ignored , there is no general consensus on the specification of the abstract unification procedure . from a theoretical point of view , once the abstract domain has been related to the concrete one by means of a galois connection , it is always possible to specify the best correct approximation of each operator of the concrete semantics .
however , empirical observations suggest that sub - optimal operators are likely to result in better complexity / precision trade - offs @xcite . as a consequence ,
it is almost impossible to identify `` the right combination '' of variable aliasing with freeness and linearity information , at least when practical issues , such as the complexity of the abstract unification procedure , are taken into account . given this state of affairs , we will now consider a domain combination whose carrier is essentially the same as specified by langen @xcite and hans and winkler @xcite .
( the same domain combination was also considered by bruynooghe et al .
@xcite , but with the addition of compoundness and explicit structural information . )
the novelty of our proposal lies in the specification of an improved abstract unification procedure , better exploiting the interaction between sharing and linearity . as a matter of fact
, we provide an example showing that all previous approaches to the combination of set - sharing with freeness and linearity are not uniformly more precise than the analysis based on the @xmath1 domain @xcite , whereas such a property is enjoyed by our proposal . by extending the results of @xcite to this combination ,
we provide a new abstraction function that can be applied to any logic language computing on domains of syntactic structures , with or without the occurs - check ; by using this abstraction function , we also prove the correctness of the new abstract unification procedure .
moreover , we show that the same notion of redundant information as identified in @xcite also applies to this abstract domain combination . as a consequence , it is possible to implement an algorithm for abstract unification running in polynomial time and
still obtain the same precision on all the considered observables : groundness , independence , freeness and linearity .
this paper is based on ( * ? ? ?
* chapter 6 ) , the phd thesis of the second author . in section [ sec : prelims ] , we define some notation and recall the basic concepts used later in the paper . in section [ sec : sfl - domain ] , we present the domain @xmath2 that integrates set - sharing , freeness and linearity . in section [ sec : sfl - asub - comparison ] , we show that @xmath2 is uniformly more precise than the domain @xmath1 , whereas all the previous proposals for a domain integrating set - sharing and linearity fail to satisfy such a property . in section [ sec : sfl - redundant ] , we show that the domain @xmath2 can be simplified by removing some redundant information . in section [ sec : exp - eval ] , we provide an experimental evaluation using the analyzer @xcite . in section [ sec : related ] , we discuss some related work .
section [ sec : conclusion ] concludes with some final remarks .
the proofs of the results stated here are not included but all of them are available in an extended version of this paper @xcite .
for a set @xmath3 , @xmath4 is the powerset of @xmath3 .
the cardinality of @xmath3 is denoted by @xmath5 and the empty set is denoted by @xmath6 .
the notation @xmath7 stands for the set of all the _ finite _ subsets of @xmath3 , while the notation @xmath8 stands for @xmath9 .
the set of all finite sequences of elements of @xmath3 is denoted by @xmath10 , the empty sequence by @xmath11 , and the concatenation of @xmath12 is denoted by @xmath13 .
let @xmath14 denote a possibly infinite set of function symbols , ranked over the set of natural numbers .
let @xmath15 denote a denumerable set of variables , disjoint from @xmath14 .
then @xmath16 denotes the free algebra of all ( possibly infinite ) terms in the signature @xmath14 having variables in @xmath15 .
thus a term can be seen as an ordered labeled tree , possibly having some infinite paths and possibly containing variables : every inner node is labeled with a function symbol in @xmath14 with a rank matching the number of the node s immediate descendants , whereas every leaf is labeled by either a variable in @xmath15 or a function symbol in @xmath14 having rank @xmath17 ( a constant ) .
it is assumed that @xmath14 contains at least two distinct function symbols , with one of them having rank @xmath17 .
if @xmath18 then @xmath19 and @xmath20 denote the set and the multiset of variables occurring in @xmath21 , respectively .
we will also write @xmath22 to denote the set of variables occurring in an arbitrary syntactic object @xmath23 .
suppose @xmath24 : @xmath25 and @xmath21 are _ independent _ if @xmath26 ; we say that variable @xmath27 _ occurs linearly in @xmath21 _ , more briefly written using the predication @xmath28 , if @xmath27 occurs exactly once in @xmath20 ; @xmath21 is said to be _ ground _ if @xmath29 ; @xmath21 is _ free _ if @xmath30 ; @xmath21 is _ linear _ if , for all @xmath31 , we have @xmath28 ; finally , @xmath21 is a _ finite term _ ( or _ herbrand term _ ) if it contains a finite number of occurrences of function symbols .
the sets of all ground , linear and finite terms are denoted by @xmath32 , @xmath33 and @xmath34 , respectively .
a _ substitution _ is a total function @xmath35 that is the identity almost everywhere ; in other words , the _ domain _ of @xmath36 , @xmath37 is finite . given a substitution @xmath35 , we overload the symbol ` @xmath36 ' so as to denote also the function @xmath38 defined as follows , for each term @xmath39 : @xmath40
if @xmath39 , we write @xmath41 to denote @xmath42 .
note that , for each substitution @xmath36 and each finite term @xmath39 , if @xmath43 , then @xmath30 . if @xmath44 and @xmath45 , then @xmath46 is called a _
binding_. the set of all bindings is denoted by @xmath47 .
substitutions are denoted by the set of their bindings , thus a substitution @xmath36 is identified with the ( finite ) set @xmath48 we denote by @xmath49 the set of variables occurring in the bindings of @xmath36 .
we also define @xmath50 .
a substitution is said to be _ circular _ if , for @xmath51 , it has the form @xmath52 where @xmath53 , , @xmath54 are distinct variables .
a substitution is in _ rational solved form _ if it has no circular subset .
the set of all substitutions in rational solved form is denoted by @xmath55 .
a substitution @xmath36 is _ idempotent _ if , for all @xmath18 , we have @xmath56 .
equivalently , @xmath36 is idempotent if and only if @xmath57 .
the set of all idempotent substitutions is denoted by @xmath58 and @xmath59 .
the composition of substitutions is defined in the usual way .
thus @xmath60 is the substitution such that , for all terms @xmath39 , @xmath61 and has the formulation @xmath62 as usual , @xmath63 denotes the identity function ( i.e. , the empty substitution ) and , when @xmath64 , @xmath65 denotes the substitution @xmath66 . for each @xmath67 and @xmath68
, the sequence of finite terms @xmath69 converges to a ( possibly infinite ) term , denoted @xmath70 @xcite .
therefore , the function @xmath71 such that @xmath72 is well defined .
note that , in general , this function is not a substitution : while having a finite domain , its `` bindings '' @xmath73 can map a domain variable @xmath74 into a term @xmath75 .
however , as the name of the function suggests , the term @xmath76 is granted to be _ rational _ , meaning that it can only have a finite number of distinct subterms and hence , be finitely represented .
consider the substitutions @xmath77 note that there are substitutions , such as @xmath78 , that are not idempotent and nonetheless define finite trees only ; namely , @xmath79 .
similarly , there are other substitutions , such as @xmath80 , whose bindings are not explicitly cyclic and nonetheless define rational trees that are infinite ; namely , @xmath81 .
finally note that the ` @xmath82 ' function is not defined on @xmath83 .
an _ equation _ is of the form @xmath84 where @xmath85 .
@xmath86 denotes the set of all equations .
a substitution @xmath36 may be regarded as a finite set of equations , that is , as the set @xmath87 .
we say that a set of equations @xmath88 is in _ rational solved form _ if @xmath89 . in the rest of the paper , we will often write a substitution @xmath67 to denote a set of equations in rational solved form ( and vice versa ) . as is common in research work involving equality
, we overload the symbol ` @xmath90 ' and use it to denote both equality and to represent syntactic identity .
the context makes it clear what is intended .
let @xmath91 .
we assume that any equality theory @xmath92 over @xmath16 includes the _ congruence axioms _ denoted by the following schemata : @xmath93 in logic programming and most implementations of prolog it is usual to assume an equality theory based on syntactic identity .
this consists of the congruence axioms together with the _ identity axioms _ denoted by the following schemata , where @xmath94 and @xmath95 are distinct function symbols or @xmath96 : @xmath97 the axioms characterized by schemata ( [ eq - ax : injective - functions ] ) and ( [ eq - ax : diff - funct ] ) ensure the equality theory depends only on the syntax .
the equality theory for a non - syntactic domain replaces these axioms by ones that depend instead on the semantics of the domain and , in particular , on the interpretation given to functor symbols .
the equality theory of clark @xcite , denoted @xmath98 , on which pure logic programming is based , usually called the _ herbrand _ equality theory , is given by the congruence axioms , the identity axioms , and the axiom schema @xmath99 axioms characterized by the schema ( [ eq - ax : occ - check ] ) are called the _ occurs - check axioms _ and are an essential part of the standard unification procedure in sld - resolution .
an alternative approach used in some implementations of logic programming systems , such as prolog ii , sicstus and oz , does not require the occurs - check axioms .
this approach is based on the theory of rational trees @xcite , denoted @xmath100 .
it assumes the congruence axioms and the identity axioms together with a _
uniqueness axiom _ for each substitution in rational solved form . informally speaking these state that , after assigning a ground rational tree to each variable which is not in the domain , the substitution uniquely defines a ground rational tree for each of its domain variables .
note that being in rational solved form is a very weak property .
indeed , unification algorithms returning a set of equations in rational solved form are allowed to be much more `` lazy '' than one would expect .
we refer the interested reader to @xcite for details on the subject . in the sequel we use the expression `` equality theory '' to denote any consistent , decidable theory @xmath92 satisfying the congruence axioms .
we also use the expression `` syntactic equality theory '' to denote any equality theory @xmath92 also satisfying the identity axioms .
we say that a substitution @xmath67 is _ satisfiable _ in an equality theory @xmath92 if , when interpreting @xmath36 as an equation system in rational solved form , @xmath101 let @xmath102 be a set of equations in an equality theory @xmath92 .
a substitution @xmath67 is called a _ solution for @xmath88 in @xmath92 _ if @xmath36 is satisfiable in @xmath92 and @xmath103 ; we say that @xmath88 is satisfiable if it has a solution .
if @xmath104 , then @xmath36 is said to be a _ relevant _
solution for @xmath88 .
in addition , @xmath36 is a _
most general solution for @xmath88 in @xmath92 _ if @xmath105 . in this paper ,
a most general solution is always a relevant solution of @xmath88 .
when the theory @xmath92 is clear from the context , the set of all the relevant most general solutions for @xmath88 in @xmath92 is denoted by @xmath106 .
[ ex : equations ] let @xmath107 and @xmath108 then , for any syntactic equality theory @xmath92 , we have @xmath109 . since @xmath67 , then @xmath36 and hence @xmath88 is satisfiable in @xmath100 . intuitively ,
whatever rational tree @xmath110 is assigned to the parameter variable @xmath111 , there exist rational trees @xmath112 , @xmath113 and @xmath114 that , when assigned to the domain variables @xmath74 , @xmath27 and @xmath115 , will turn @xmath36 into a set of trivial identities ; namely , let @xmath112 and @xmath113 be both equal to the infinite rational tree @xmath116 , which is usually denoted by @xmath117 , and let @xmath114 be the rational tree @xmath118 .
thus @xmath36 is a relevant most general solution for @xmath88 in @xmath100 .
in contrast , @xmath119 is just a relevant solution for @xmath88 in @xmath100 .
also observe that , for any equality theory @xmath92 , @xmath120 so that @xmath36 does not satisfy the occurs - check axioms
. therefore , neither @xmath36 nor @xmath88 are satisfiable in the herbrand equality theory @xmath98 .
intuitively , there is no finite tree @xmath112 such that @xmath121 .
we have the following useful result regarding ` @xmath82 ' and satisfiable substitutions that are equivalent with respect to any given syntactic equality theory .
[ prop : rt - preserves - vars - gterms - lterms ] let @xmath122 be satisfiable in the syntactic equality theory @xmath92 and suppose that @xmath123 . then @xmath124 given two complete lattices @xmath125 and @xmath126 , a _ galois connection _ is a pair of monotonic functions @xmath127 and @xmath128 such that @xmath129 the functions @xmath130 and @xmath131 are said to be the abstraction and concretization functions , respectively .
galois insertion _ is a galois connection where the concretization function @xmath131 is injective .
an _ upper closure operator _
( uco ) @xmath132 on the complete lattice @xmath125 is a monotonic , idempotent and extensive for each @xmath133 . ]
self - map .
the set of all uco s on @xmath134 , denoted by @xmath135 , is itself a complete lattice . for any @xmath136 , the set @xmath137 , i.e.
, the image under @xmath138 of the lattice carrier , is a complete lattice under the same partial order @xmath139 defined on @xmath134 .
given a galois connection , the function @xmath140 is an element of @xmath135 .
the presentation of abstract interpretation in terms of galois connections can be rephrased by using uco s . in particular
, the partial order @xmath141 defined on @xmath135 formalizes the intuition of an abstract domain being more precise than another one ; moreover , given two elements @xmath142 , their reduced product @xcite , denoted @xmath143 , is their @xmath144 on @xmath135 .
the set - sharing domain of jacobs and langen @xcite , encodes both aliasing and groundness information .
let @xmath145 be a fixed and finite set of variables of interest .
an element of the set - sharing domain ( a _ sharing set _ ) is a set of subsets of @xmath146 ( the _ sharing groups _ ) .
note that the empty set is not a sharing group .
[ def : sh ] let @xmath147 be the set of _ sharing groups_. the set - sharing lattice is defined as @xmath148 , ordered by subset inclusion .
the following operators on @xmath149 are needed for the specification of the abstract semantics .
[ def : aux - funcs - sh ] for each @xmath150 and each @xmath151 , we define the following functions : the _ star - union _
function @xmath152 , is defined as @xmath153 in @xcite it was shown that the domain @xmath149 contains many elements that are redundant for the computation of the actual _ observable _ properties of the analysis , definite groundness and definite independence .
the following formalization of these observables is a rewording of the definitions provided in @xcite .
[ def : sh - observables ] the _ groundness _ and _ independence _ observables ( on @xmath149 ) @xmath154 are defined , for each @xmath155 , by @xmath156 note that , as usual in sharing analysis domains , definite groundness and definite independence are both represented by encoding possible non - groundness and possible pair - sharing information .
the abstract domain @xmath157 @xcite is the simplest abstraction of the domain @xmath149 that still preserves the same precision on groundness and independence .
[ def : rhopsd ] the operator @xmath158 is defined , for each @xmath155 , by @xmath159 the _ pair - sharing dependency _ lattice is @xmath160 . in the following example we provide an intuitive interpretation of the approximation induced by the three upper closure operators of definitions [ def : sh - observables ] and [ def : rhopsd ] .
[ ex : con - ps ] let @xmath161 and consider , omitting inner braces .
for instance , we will write @xmath162 to denote @xmath163 . ] @xmath164 .
then @xmath165 when observing @xmath166 , the only information available is that variable @xmath111 does not occur in a sharing group ; intuitively , this means that @xmath111 is definitely ground .
all the other information encoded in @xmath167 is lost ; for instance , in @xmath167 variables @xmath168 and @xmath115 never occur in the _ same _ sharing group ( i.e. , they are definitely independent ) , while this happens in @xmath166 .
when observing @xmath169 , it should be noted that two distinct variables occur in the same sharing group if and only if they were also occurring together in a sharing group of @xmath167 , so that the definite independence information is preserved ( e.g. , @xmath168 and @xmath115 keep their independence ) . on the other hand ,
all the variables in @xmath146 occur as singletons in @xmath169 whether or not they are known to be ground ; for instance , @xmath170 occurs in @xmath169 although @xmath111 does not occur in any sharing group in @xmath167 . by noting that @xmath171 , it follows that @xmath172 preserves both the definite groundness and the definite independence information of @xmath167
; moreover , as the inclusion is strict , @xmath172 encodes other information , such as variable covering ( the interested reader is referred to @xcite for a more formal discussion ) .
one of the key concepts used in @xcite for the proofs of the correctness results stated in this paper is that of variable - idempotence .
for the interested reader , we provide here a brief introduction to variable - idempotent substitutions , although these are not referred to elsewhere in the paper .
the definition of idempotence requires that repeated applications of a substitution do not change the syntactic structure of a term and idempotent substitutions are normally the preferred form of a solution to a set of equations .
however , in the domain of rational trees , a set of solvable equations does not necessarily have an idempotent solution ( for instance , in example [ ex : equations ] , the set of equations @xmath88 has no idempotent solution ) . on the other hand ,
several abstractions of terms , such as the ones commonly used for sharing analysis , are only interested in the set of variables occurring in a term and not in the concrete structure that contains them .
thus , for applications such as sharing analysis , a useful way to relax the definition of idempotence is to ignore the structure of terms and just require that the repeated application of a substitution leaves the set of variables in a term invariant .
[ def : vsubst ] a substitution @xmath67 is _ _ variable - idempotent _ _ if and only if for all @xmath39 we have @xmath173 the set of variable - idempotent substitutions is denoted @xmath174 . as any idempotent substitution is also variable - idempotent , we have @xmath175 .
consider the following substitutions which are all in @xmath55 .
the abstract domain @xmath2 is made up of three components , providing different kinds of sharing information regarding the set of variables of interest @xmath146 : the first component is the set - sharing domain @xmath149 of jacobs and langen @xcite ; the other two components provide freeness and linearity information , each represented by simply recording those variables of interest that are known to enjoy the corresponding property .
let @xmath177 and @xmath178 be partially ordered by reverse subset inclusion .
the abstract domain @xmath2 is defined as @xmath179 and is ordered by @xmath180 , the component - wise extension of the orderings defined on the sub - domains . with this ordering
, @xmath2 is a complete lattice whose least upper bound operation is denoted by @xmath181 .
the bottom element @xmath182 will be denoted by @xmath183 .
when the concrete domain is based on the theory of finite trees , idempotent substitutions provide a finitely computable _ strong normal form _ for domain elements , meaning that different substitutions describe different sets of finite trees . in contrast , when working on a concrete domain based on the theory of rational trees , substitutions in rational solved form , while being finitely computable , no longer satisfy this property : there can be an infinite set of substitutions in rational solved form all describing the same set of rational trees ( i.e. , the same element in the `` intended '' semantics ) .
for instance , the substitutions @xmath184 for @xmath185 , @xmath186 , , all map the variable @xmath74 into the same infinite rational tree @xmath117 . ideally , a strong normal form for the set of rational trees described by a substitution @xmath67 can be obtained by computing the limit @xmath187 .
the problem is that @xmath187 can map domain variables to infinite rational terms and may not be in @xmath55 .
this poses a non - trivial problem when trying to define `` good '' abstraction functions , since it would be really desirable for this function to map any two equivalent concrete elements to the same abstract element .
as shown in @xcite , the classical abstraction function for set - sharing analysis @xcite , which was defined only for substitutions that are idempotent , does not enjoy this property when applied , as it is , to arbitrary substitutions in rational solved form . in @xcite , this problem is solved by replacing the sharing group operator ` @xmath188 ' of @xcite by an occurrence operator , ` @xmath189 ' , defined by means of a fixpoint computation . however , to simplify the presentation , here we define ` @xmath189 ' directly by exploiting the fact that the number of iterations needed to reach the fixpoint is bounded by the number of bindings in the substitution . [
def : occ ] for each @xmath67 and @xmath190 , the _ occurrence operator _
@xmath191 is defined as @xmath192 the operator ` @xmath193 ' is introduced for notational convenience only .
[ ex : occ ] let @xmath194 then @xmath195 so that @xmath196 , for @xmath197 , @xmath186 , @xmath198 and @xmath199 . as a consequence , supposing that @xmath200 , we obtain @xmath201 . in a similar way , it is possible to define suitable operators for groundness , freeness and linearity . as
all ground trees are linear , a knowledge of the definite groundness information can be useful for proving properties concerning the linearity abstraction .
groundness is already encoded in the abstraction for set - sharing provided in definition [ def : occ ] ; nonetheless , for both a simplified notation and a clearer intuitive reading , we now explicitly define the set of variables that are associated to ground trees by a substitution in @xmath55 .
[ def : gvars ] the _ groundness operator _
@xmath202 is defined , for each @xmath67 , by @xmath203 [ ex : gvars ] consider @xmath67 where @xmath204 then @xmath205 .
observe that @xmath206 although @xmath207 .
also , @xmath208 although @xmath209 for all @xmath210 . as for possible sharing
, the definite freeness information can be extracted from a substitution in rational solved form by observing the result of a bounded number of applications of the substitution .
[ def : fvars ] the _ freeness operator _
@xmath211 is defined , for each @xmath67 , by @xmath212 as @xmath67 has no circular subset , @xmath213 implies @xmath214 .
[ ex : fvars ] let @xmath215 and consider @xmath67 where @xmath216 then @xmath217 .
thus @xmath218 although @xmath207 .
also , @xmath219 although @xmath220 . as in previous cases ,
the definite linearity information can be extracted by observing the result of a bounded number of applications of the considered substitution .
[ def : lvars ] the _ linearity operator _
@xmath221 is defined , for each @xmath67 , by @xmath222 in the next example we consider the extraction of linearity from two substitutions .
the substitution @xmath36 shows that , in contrast with the case of set - sharing and freeness , for linearity we may need to compute up to @xmath223 applications , where @xmath224 ; the substitution @xmath225 shows that , when observing the term @xmath226 , multiple occurrences of domain variables have to be disregarded .
[ ex : lvars ] let @xmath227 and consider @xmath67 where @xmath228
then @xmath229 .
observe that @xmath230 .
this is because @xmath231 , @xmath232 so that @xmath233 and @xmath234 so that @xmath235 does not hold .
note also that @xmath236 holds for @xmath237 , @xmath238 , @xmath239 .
consider now @xmath240 where @xmath241 then @xmath242 .
note that we have @xmath243 although , for all @xmath64 , @xmath244 occurs more than once in the term @xmath245 .
the occurrence , groundness , freeness and linearity operators are invariant with respect to substitutions that are equivalent in the given syntactic equality theory .
[ prop : iff - rsubst - occ - f - g - l ] let @xmath246 be satisfiable in the syntactic equality theory @xmath92 and suppose that @xmath247 .
then @xmath248 moreover , these operators precisely capture the intended properties over the domain of rational trees .
[ prop : occ - f - g - l - rt - rsubst ] if @xmath67 and @xmath249 then @xmath250 it follows from ( [ case : rsubst - occ ] ) and ( [ case : rsubst - free ] ) that any free variable necessarily shares ( at least , with itself ) .
also , as @xmath251 , it follows from ( [ case : rsubst - ground ] ) , ( [ case : rsubst - free ] ) and ( [ case : rsubst - linear ] ) that any variable that is either ground or free is also necessarily linear .
thus we have the following corollary .
[ cor : f - g - l - rsubst ] if @xmath67 , then @xmath252 we are now in position to define the abstraction function mapping rational trees to elements of the domain @xmath2 .
[ def : abstrsfl ] for each substitution @xmath67 , the function @xmath253 is defined by @xmath254 with definition [ def : abstrsfl ] and proposition [ prop : iff - rsubst - occ - f - g - l ] , one of our objectives is fulfilled : substitutions in @xmath55 that are equivalent have the same abstraction .
[ cor : iff - rsubst - same - abstrsfl ]
let @xmath246 be satisfiable in the syntactic equality theory @xmath92 and suppose @xmath247
. then @xmath255 .
observe that the galois connection defined by the functions @xmath256 and @xmath257 is not a galois insertion since different abstract elements are mapped by @xmath257 to the same set of concrete computation states . to see this it is sufficient to observe that , by corollary [ cor : f - g - l - rsubst ] , any abstract element @xmath258 such that @xmath259 , as is the case for the bottom element @xmath183 , satisfies @xmath260 ; thus , all such @xmath261 s will represent the semantics of those program fragments that have no successful computations .
similarly , by letting @xmath262 , it can be seen that , for any @xmath263 such that @xmath264 we have , again by corollary [ cor : f - g - l - rsubst ] , @xmath265 . of course
, by taking the abstract domain as the subset of @xmath2 that is the co - domain of @xmath256 , we would have a galois insertion .
however , apart from the simple cases shown above , it is somehow difficult to _ explicitly _ characterize such a set . for instance , as observed in @xcite , if @xmath266 we have @xmath267 .
it is worth stressing that these `` spurious '' elements do not compromise the correctness of the analysis and , although they can affect the precision of the analysis , they rarely occur in practice @xcite .
the specification of the abstract unification operator on the domain @xmath2 is rather complex , since it is based on a very detailed case analysis . to achieve some modularity , that will be also useful when proving its correctness , in the next definition we introduce several auxiliary abstract operators .
[ def : aux - funcs - sfl ] let @xmath85 be finite terms such that @xmath268 . for each @xmath269
we define the following predicates : @xmath25 and @xmath21 are _ independent in @xmath270 _ if and only if @xmath271 holds for @xmath272 , where @xmath273 the function @xmath274 yields the set of variables of interest that may share with the given term . for each @xmath39 ,
@xmath275 the function @xmath276 strengthens the sharing set @xmath167 by forcing the coupling of @xmath74 with @xmath21 . for each @xmath155 and each @xmath277 ,
@xmath278 as a first correctness result , we have that the auxiliary operators correctly approximate the corresponding concrete properties . [ thm : soundness - of - sfl - preds ] let @xmath279 , @xmath280 and @xmath281 .
let also @xmath85 be two finite terms such that @xmath268 .
then @xmath282 let @xmath161 and consider the abstract element @xmath283 , where @xmath284 then , by applying definition [ def : aux - funcs - sfl ] , we obtain the following .
* @xmath285 does not hold whereas @xmath286 holds .
* @xmath287 holds but @xmath288 does not hold . * both @xmath289 and @xmath290 hold whereas @xmath291 does not hold ; note that , in the second case , the two arguments of the predicate do share @xmath27 , but this does not affect the independence of the corresponding terms , because @xmath27 is definitely ground in the abstract element @xmath270 . * let @xmath292 ; then @xmath293 does not hold because @xmath294 ; @xmath295 does not hold because @xmath74 occurs more than once in @xmath21 ; @xmath296 holds , even though @xmath27 occurs twice in @xmath21 , because @xmath27 is definitely ground in @xmath270 ; @xmath297 does not hold because both @xmath74 and @xmath115 occur in term @xmath21 and , as observed in the point above , @xmath291 does not hold . * for the reasons given in the point above , @xmath298 does not hold ; in contrast , @xmath299 holds . * @xmath300 and @xmath301 ; thus , both @xmath111 and @xmath74 may share one or more variables with @xmath115 ; since we observed that @xmath111 and @xmath74 are definitely independent in @xmath270 , this means that the set of variables that @xmath111 shares with @xmath115 is disjoint from the set of variables that @xmath74 shares with @xmath115 .
* let @xmath302 ; then @xmath303 an intuitive explanation of the usefulness of this operator is deferred until after the introduction of the abstract @xmath0 operator ( see also example [ ex : amgusfl - cyclicreduce ] ) .
we now introduce the abstract @xmath0 operator , specifying how a single binding affects each component of the domain @xmath2 in the context of a syntactic equality theory @xmath92 .
[ def : amgusfl ] the function @xmath304 captures the effects of a binding on an element of @xmath2 . let @xmath269 and @xmath277 , where @xmath305 .
let also @xmath306 where @xmath307 @xmath308 letting @xmath309 and @xmath310 , we also define @xmath311
l ' & { { \mathrel{\buildrel \mathrm{def } \over { = } } } } \bigl ( { \mathord{\mathit{vi}}}{\setminus}{\mathop{\mathrm{vars}}\nolimits}({\mathord{\mathit{sh } } } ' ) \bigr ) { \cup}f ' { \cup}l '' , \\ \intertext{where } l '' & { { \mathrel{\buildrel \mathrm{def } \over { = } } } } \begin{cases } l { \setminus}(s_x { \cap}s_t ) , & \text{if $ { \mathop{\mathrm{lin}}\nolimits}_{{\mathit{d}}}(x ) \land { \mathop{\mathrm{lin}}\nolimits}_{{\mathit{d}}}(t)$ ; } \\
l { \setminus}s_x , & \text{if $ { \mathop{\mathrm{lin}}\nolimits}_{{\mathit{d}}}(x)$ ; } \\
l { \setminus}s_t , & \text{if $ { \mathop{\mathrm{lin}}\nolimits}_{{\mathit{d}}}(t)$ ; } \\
l { \setminus}(s_x { \cup}s_t ) , & \text{otherwise . } \\
\end{cases } \intertext{then } { { \mathop{\mathrm{amgu}}\nolimits}_{\scriptscriptstyle s}}\bigl({\mathit{d } } , x \mapsto t\bigr ) & { { \mathrel{\buildrel \mathrm{def } \over { = } } } } \begin{cases } { \bot_{\scriptscriptstyle s } } , & \text{if \ ( { \mathit{d}}= { \bot_{\scriptscriptstyle s}}\lor \bigl(t = { \ensuremath{\mathcal{ft}}}\land x \in { \mathop{\mathrm{vars}}\nolimits}(t)\bigr ) \ ) ; } \\
\langle { \mathord{\mathit{sh } } } ' , f ' , l ' \rangle & \text{otherwise . }
\end{cases}\end{aligned}\ ] ] the next result states that the abstract @xmath0 operator is a correct approximation of the concrete one .
[ thm : soundness - of - amgusfl ] let @xmath279 and @xmath277 , where @xmath305 . then , for all @xmath280 and @xmath312 in the syntactic equality theory @xmath92 , we have @xmath313 we now highlight the similarities and differences of the operator @xmath314 with respect to the corresponding ones defined in the `` classical '' proposals for the integration of set - sharing with freeness and linearity , such as @xcite . note that , when comparing our domain with the proposal in @xcite , we deliberately ignore all those enhancements that depend on properties that can not be represented in @xmath2 ( i.e. , compoundness and explicit structural information ) . * in the computation of the set - sharing component , the main difference can be observed in the second , third and fourth cases of the definition of @xmath315 : here we omit one of the star - unions even when the terms @xmath74 and @xmath21 possibly share .
in contrast , in @xcite the corresponding star - union is avoided only when @xmath316 holds .
note that when @xmath316 holds in the second case of @xmath315 , then we have @xmath317 ; thus , the whole computation for this case reduces to @xmath318 , as was the case in the previous proposals .
* another improvement on the set - sharing component can be observed in the definition of @xmath319 : the @xmath320 operator allows the set - sharing description to be further enhanced when dealing with _ explicitly cyclic bindings _ , i.e. , when @xmath321 .
this is the rewording of a similar enhancement proposed in @xcite for the domain @xmath322 in the context of groundness analysis .
its net effect is to recover some groundness and sharing dependencies that would have been unnecessarily lost when using the standard operators .
when @xmath323 , we have @xmath324 . * the computation of the freeness component @xmath325 is the same as specified in @xcite , and is more precise than the one defined in @xcite . *
the computation of the linearity component @xmath263 is the same as specified in @xcite , and is more precise than those defined in @xcite . in the following examples we show that the improvements in the abstract computation of the sharing component allow , in particular cases , to derive better information than that obtainable by using the classical abstract unification operators .
[ ex : amgusfl - both - lin - share ] let @xmath326 and @xmath67 such that @xmath327 by definition [ def : abstrsfl ] , we have @xmath328 , where @xmath329 consider the binding @xmath330 . in the concrete domain ,
we compute ( a substitution equivalent to ) @xmath331 , where @xmath332 note that @xmath333 , where @xmath334 , so that the pairs of variables @xmath335 and @xmath336 keep their independence .
when evaluating the sharing component of @xmath337 , using the notation of definition [ def : amgusfl ] , we have @xmath338 since both @xmath339 and @xmath340 hold , we apply the second case of the definition of @xmath315 so that @xmath341 finally , as the binding is not cyclic , we obtain @xmath342 .
thus @xmath314 captures the fact that pairs @xmath343 and @xmath344 keep their independence .
in contrast , since @xmath345 does not hold , all of the classical definitions of abstract unification would have required the star - closure of both @xmath346 and @xmath347 , resulting in an abstract element including , among the others , the sharing group @xmath348 . since @xmath349
, this independence information would have been unnecessarily lost .
similar examples can be devised for the third and fourth cases of the definition of @xmath315 , where only one side of the binding is known to be linear .
the next example shows the precision improvements arising from the use of the @xmath320 operator .
[ ex : amgusfl - cyclicreduce ] let @xmath350 and @xmath351 . by definition [ def :
abstrsfl ] , we have @xmath328 , where @xmath352 let @xmath353 and consider the cyclic binding @xmath277 .
in the concrete domain , we compute ( a substitution equivalent to ) @xmath354 , where @xmath355 note that if we further instantiate @xmath225 by grounding @xmath27 , then variables @xmath74 , @xmath53 and @xmath356 would become ground too .
formally we have @xmath333 , where @xmath357 .
thus , as observed above , @xmath27 covers @xmath74 , @xmath53 and @xmath356 .
when abstractly evaluating the binding , we compute @xmath358 since both @xmath339 and @xmath298 hold , we apply the second case of the definition of @xmath315 , so that @xmath359 thus , as @xmath321 , we obtain @xmath360 note that , in the element @xmath361 ( which is the abstract element that would have been computed when not exploiting the @xmath320 operator ) variable @xmath27 covers none of variables @xmath74 , @xmath53 and @xmath356 . thus , by applying the @xmath320 operator , this covering information is restored . the full abstract unification operator @xmath362 , capturing the effect of a sequence of bindings on an abstract element , can now be specified by a straightforward inductive definition using the operator @xmath314 .
[ def : aunifysfl ] the operator @xmath363 is defined , for each @xmath279 and each sequence of bindings @xmath364 , by @xmath365 note that the second argument of @xmath362 is a _ sequence _ of bindings ( i.e. , it is not a substitution , which is a _ set _ of bindings ) , because @xmath314 is neither commutative nor idempotent , so that the multiplicity and the actual order of application of the bindings can influence the overall result of the abstract computation .
the correctness of the @xmath362 operator is simply inherited from the correctness of the underlying @xmath314 operator . in particular , any reordering of the bindings in the sequence @xmath366 still results in a correct implementation of @xmath362 .
the ` merge - over - all - path ' operator on the domain @xmath2 is provided by @xmath181 and is correct by definition .
finally , we define the abstract existential quantification operator for the domain @xmath2 , whose correctness does not pose any problem .
[ def : aprojsfl ] the function @xmath367 provides the _ abstract existential quantification _ of an element with respect to a subset of the variables of interest . for each @xmath368 and @xmath151 , @xmath369 the intuition behind the definition of the abstract operator
@xmath370 is the following . as explained in section [ sec : prelims ] ,
any substitution @xmath67 can be interpreted , under the given equality theory @xmath92 , as a first - order logical formula ; thus , for each set of variables @xmath371 , it is possible to consider the ( concrete ) existential quantification @xmath372 .
the goal of the abstract operator @xmath370 is to provide a correct approximation of such a quantification starting from any correct approximation for @xmath36 .
let @xmath373 and @xmath374 , so that , by definition [ def : abstrsfl ] , @xmath375 let @xmath376 and consider the concrete element corresponding to the logical formula @xmath372 . note that @xmath377 , where @xmath378 . by applying definition [ def : aprojsfl ]
, we obtain @xmath379 it is worth stressing that such an operator does not affect the set @xmath146 of the variables of interest .
in particular , the abstract element @xmath380 still has to provide correct information about variables @xmath27 and @xmath115 .
intuitively , since all the occurrences of @xmath27 and @xmath115 in @xmath372 are bound by the existential quantifier , the two variables of interest are un - aliased , free and linear .
note that an abstract _ projection _ operator , i.e. , an operator that actually modifies the set of variables of interest , is easily specified by composing the operator @xmath370 with an operator that simply removes , from all the components of @xmath2 _ and _ from the set of variables of interest @xmath146 , those variables that have to be projected out .
as we have already observed , example [ ex : amgusfl - both - lin - share ] shows that the abstract domain @xmath2 , when equipped with the abstract mgu operator introduced in section [ subsec : sfl - operators ] , can yield results that are strictly more precise than all the classical combinations of set - sharing with freeness and linearity information . in this section
we show that the same example has another interesting , unexpected consequence , since it can be used to formally prove that all the classical combinations of set - sharing with freeness and linearity , including those presented in @xcite , are not _ uniformly _ more precise than the abstract domain @xmath1 @xcite , which is based on pair - sharing . to formalize the above observation
, we now introduce the @xmath1 domain and the corresponding abstract semantics operators as specified in @xcite .
the elements of the abstract domain @xmath1 have two components : the first one is a set of variables that are known to be definitely ground ; the second one encodes both possible pair - sharing and possible non - linearity into a single relation defined on the set of variables .
intuitively , when @xmath381 and @xmath382 occurs in the second component , then @xmath74 and @xmath27 may share a variable ; when @xmath383 occurs in the second component , then @xmath74 may be non - linear .
the second component always encodes a symmetric relation ; thus , for notational convenience and without any loss of generality @xcite , we will represent each pair @xmath384 in such a relation as the sharing group @xmath385 , which will have cardinality 1 or 2 depending on whether @xmath386 or not , respectively .
[ def : asub ] the abstract domain @xmath387 is defined as @xmath388 , where @xmath389 for @xmath390 , let @xmath391 . then @xmath392 the partial order @xmath393 is extended on @xmath394 by letting @xmath395 be the bottom element .
let @xmath396 and @xmath397 .
then @xmath398 is a shorthand for the condition @xmath399 , whereas @xmath400 is a shorthand for @xmath401 .
it is well - known that the domain @xmath394 can be obtained by a further abstraction of any domain such as @xmath2 that is based on set - sharing and enhanced with linearity information .
the following definition formalizes this abstraction .
[ def : abstrasub ] let @xmath269 .
then @xmath402 where @xmath403 the definition of abstract unification in @xcite is based on a few auxiliary operators .
the first of these introduces the concept of abstract multiplicity for a term under a given abstract substitution , therefore modeling the notion of definite groundness and definite linearity .
[ def : chi ] let @xmath397 and let @xmath39 be a term such that @xmath404 .
we say that @xmath31 _ occurs linearly ( in @xmath21 ) in @xmath405 _ if and only if @xmath406 holds for @xmath407 , where @xmath408 it is worth noting that , modulo a few insignificant differences in notation , the multiplicity operator @xmath409 defined above corresponds to the abstract multiplicity operator @xmath410 , which was introduced in ( * ? ? ?
* definition 3.4 ) and provided with an executable specification in ( * ? ? ?
* definition 4.3 ) .
similarly , the next definition corresponds to ( * ? ? ?
* definition 4.3 ) .
[ def : soln ] for each @xmath411 and @xmath277 , where @xmath412 and @xmath413 are such that @xmath414 , the function @xmath415 is defined as follows @xmath416 the next definition corresponds to ( * ? ? ?
* definition 4.5 ) .
[ def : abstract - composition ] let @xmath417 , where @xmath418 and @xmath419 .
then @xmath420 , where @xmath421 we are now ready to define the abstract @xmath0 operator for the domain @xmath394 .
this operator can be viewed as a specialization of ( * ? ? ?
* definition 4.6 ) for the case when we have to abstract a single binding .
[ def : amguasub ] let @xmath422 and @xmath277 , where @xmath305 .
then @xmath423 by repeating the abstract computation of example [ ex : amgusfl - both - lin - share ] on the domain @xmath1 , we provide a formal proof that all the classical approaches based on set - sharing are not uniformly more precise than the pair - sharing domain @xmath1 . [
ex : amguasub - both - lin - share ] consider the substitutions @xmath122 and the abstract element @xmath279 as introduced in example [ ex : amgusfl - both - lin - share ] . by definition [ def : abstrasub ] ,
we obtain @xmath424 , where @xmath425 when abstractly evaluating the binding @xmath426 according to definition [ def : amguasub ] , we compute the following : @xmath427 where @xmath428 note that @xmath429 and @xmath430 , so that these pairs of variables keep their independence . in contrast , as observed in example [ ex : amgusfl - both - lin - share ] , the operators in @xcite will fail to preserve the independence of these pairs .
we now show that the abstract domain @xmath2 , when equipped with the operators introduced in section [ subsec : sfl - operators ] , is uniformly more precise than the domain @xmath1 .
in particular , the following theorem states that the abstract operator @xmath314 of definition [ def : amgusfl ] is uniformly more precise than the abstract operator @xmath431 .
[ thm : amgusfl - uniformly - more - precise - than - amguasub ] let @xmath279 and @xmath432 be such that @xmath433 .
let also @xmath277 , where @xmath305 .
then @xmath434 similar results can be stated for the other abstract operators , such as the abstract existential quantification @xmath370 and the merge - over - all - path operator @xmath181 .
it is worth stressing that , when sequences of bindings come into play , the specification provided in ( * ? ? ?
* definition 4.7 ) requires that the _ grounding _ bindings ( i.e. , those bindings such that @xmath435 ) are evaluated before the non - grounding ones .
clearly , if we want to lift the result of theorem [ thm : amgusfl - uniformly - more - precise - than - amguasub ] so that it also applies to the operator @xmath362 , the same evaluation strategy has to be adopted when computing on the domain @xmath2 ; this improvement is well - known @xcite and already exploited in most implementations of sharing analysis @xcite .
as done in @xcite for the plain set - sharing domain @xmath149 , even when considering the richer domain @xmath2 it is natural to question whether it contains redundancies with respect to the computation of the observable properties .
it is worth stressing that the results presented in @xcite and @xcite can not be simply inherited by the new domain .
the concept of `` redundancy '' depends on both the starting domain and the given observables : in the @xmath2 domain both of these have changed .
first of all , as can be seen by looking at the definition of @xmath314 , freeness and linearity positively interact in the computation of sharing information : _ a priori _ it is an open issue whether or not the `` redundant '' sharing groups can play a role in such an interaction . secondly ,
since freeness and linearity information can be themselves usefully exploited in a number of applications of static analysis ( e.g. , in the optimized implementation of concrete unification or in occurs - check reduction ) , these properties have to be included in the observables .
we will now show that the domain @xmath2 can be simplified by applying the same notion of redundancy as identified in @xcite .
namely , in the definition of @xmath2 it is possible to replace the set - sharing component @xmath149 by @xmath157 without affecting the precision on groundness , independence , freeness and linearity . in order to prove such a claim , we now formalize the new observable properties .
[ def : sfl - observables ] the ( overloaded ) _ groundness _ and _ independence _
observables @xmath437 are defined , for each @xmath438 , by @xmath439 the overloading of @xmath440 working on the domain @xmath2 is the straightforward extension of the corresponding operator on @xmath149 : in particular , the freeness and linearity components are left untouched .
[ def : rhopsd - for - sfl ] for each @xmath438 , the operator @xmath441 is defined by @xmath442 this operator induces the lattice @xmath443 .
as proved in @xcite , we have that @xmath444 ; by the above definitions , it is also clear that @xmath445 ; thus , @xmath440 is more precise than the reduced product @xmath446 .
informally , this means that the domain @xmath436 is able to _ represent _ all of our observable properties without precision losses .
the next theorem shows that @xmath440 is a congruence with respect to the @xmath362 , @xmath181 and @xmath370 operators .
this means that the domain @xmath436 is able to _ propagate _ the information on the observables as precisely as @xmath2 , therefore providing a completeness result .
[ thm : psd - precise - for - sfl ] let @xmath447 be such that @xmath448 .
then , for each sequence of bindings @xmath364 , for each @xmath449 and @xmath450 , @xmath451 finally , by providing the minimality result , we show that the domain @xmath436 is indeed the generalized quotient @xcite of @xmath2 with respect to the reduced product @xmath452 [ thm : psd - aunifysfl : minimality ] for each @xmath453 , let @xmath454 be such that @xmath455 .
then there exist a sequence of bindings @xmath364 and an observable property @xmath456 such that @xmath457 as far as the implementation is concerned , the results proved in @xcite for the domain @xmath157 can also be applied to @xmath436 .
in particular , in the definition of @xmath314 every occurrence of the star - union operator can be safely replaced by the self - bin - union operator . as a consequence ,
it is possible to provide an implementation where the time complexity of the @xmath314 operator is bounded by a polynomial in the number of sharing groups of the set - sharing component .
the following result provides another optimization that can be applied when both terms @xmath74 and @xmath21 are definitely linear , but none of them is definitely free ( i.e. , when we compute @xmath315 by the second case stated in definition [ def : amgusfl ] ) .
[ thm : psd - precise - for - sfl : inner - bin - unions - useless ] let @xmath155 and @xmath458 , where @xmath305 .
let @xmath459 , @xmath460 , @xmath461 , @xmath462 , @xmath463 , where @xmath464 , and @xmath465 then it holds @xmath466 therefore , even when terms @xmath74 and @xmath21 possibly share ( i.e. , when @xmath467 ) , by using @xmath436 we can avoid the expensive computation of at least one of the two inner binary unions in the expression for @xmath468 .
example [ ex : amgusfl - both - lin - share ] shows that an analysis based on the new abstract unification operator can be strictly more precise than one based on the classical proposal .
however , that example is artificial and leaves open the question as to whether or not such a phenomenon actually happens during the analysis of real programs and , if so , how often .
this was the motivation for the experimental evaluation we describe in this section .
we consider the abstract domain @xmath469 @xcite , where the non - redundant version @xmath436 of the domain @xmath2 is further combined , as described in ( * ? ? ?
* section 4 ) , with the definite groundness information computed by @xmath322 and compare the results using the ( classical ) abstract unification operator of ( * ? ? ?
* definition 4 ) with the ( new ) operator @xmath314 given in definition [ def : amgusfl ] . taking this as a starting point
, we experimentally evaluate eight variants of the analysis arising from all possible combinations of the following options : 1 . the analysis can be goal independent or goal dependent ; 2 .
the set - sharing component may or may not have widening enabled @xcite ; 3 . the abstract domain may or may not be upgraded with structural information using the @xmath470 operator ( see @xcite and ( * ? ? ?
* section 5 ) ) .
the experiments have been conducted using the analyzer @xcite on a gnu / linux pc system .
is a data - flow analyzer for ( constraint ) logic programs performing bottom - up analysis and deriving information on both call - patterns and success - patterns by means of program transformations and optimized fixpoint computation techniques .
an abstract description is computed for the call- and success - patterns for each predicate defined in the program .
the benchmark suite , which is composed of 372 logic programs of various sizes and complexity , can be considered representative .
the precision results for the goal independent comparisons are summarized in table [ tab : oldmodes - vs - newmodes - precision ] . for each benchmark
, precision is measured by counting the number of independent pairs as well as the numbers of definitely ground , free and linear variables detected . for each variant of the analysis ,
these numbers are then compared by computing the relative precision improvements and expressing them using percentages .
the benchmark suite is then partitioned into several precision equivalence classes and the cardinalities of these classes are shown in table [ tab : oldmodes - vs - newmodes - precision ] .
for example , when considering a goal independent analysis without structural information and without widenings , the value 5 found at the intersection of the row labeled ` @xmath471 ' with the column labeled ` i ' should be read : `` for five benchmarks there has been a ( positive ) increase in the number of independent pairs of variables which is less than or equal to two percent . ''
note that we only report on independence and linearity ( in the columns labeled ` i ' and ` l ' , respectively ) , because no differences have been observed for groundness and freeness .
the precision class labeled ` unknown ' identifies those benchmarks for which the analyses timed - out ( the time - out threshold was fixed at 600 seconds ) .
hence , for goal independent analyses , a precision improvement affects from 1.6% to 3% of the benchmarks , depending on the considered variant . when considering the goal dependent analyses , we obtain a single , small improvement , so that no comparison tables are included here : the improvement , affecting linearity information ,
can be observed when the abstract domain includes structural information . with respect to differences in the efficiency
, the introduction of the new abstract unification operator has no significant effect on the computation time : small differences ( usually improvements ) are observed on as many as 6% of the benchmarks for the goal independent analysis without structural information and without widenings ; other combinations register even less differences .
we note that it is not surprising that the precision and efficiency improvements occur very rarely since the abstract unification operators behave the same except under very specific conditions : the two terms being unified must not only be definitely linear , but also possibly non - free and share a variable . & & & & + prec .
class & & & & & & & & + @xmath472 & & 2 & & 2 & & 2 & & 2 + @xmath473 & & & & & & & & 1 + @xmath474 & 5 & 5 & 9 & 6 & 6 & 6 & 12 & 8 + same precision & 357 & 355 & 337 & 338 & 366 & 364 & 360 & 361 + unknown & 10 & 10 & 26 & 26 & & & & +
sharing information has been shown to be important for finite - tree analysis @xcite .
this aims at identifying those program variables that , at a particular program point , can not be bound to an infinite rational tree ( in other words , they are necessarily bound to acyclic terms ) .
this novel analysis is irrelevant for those logic languages computing over a domain of finite trees , while having several applications for those ( constraint ) logic languages that are explicitly designed to compute over a domain including rational trees , such as prolog ii and its successors @xcite , sicstus prolog @xcite , and oz @xcite .
the analysis specified in @xcite is based on a parametric abstract domain @xmath475 , where the @xmath476 component ( the herbrand component ) is a set of variables that are known to be bound to finite terms , while the parametric component @xmath477 can be any domain capturing aliasing , groundness , freeness and linearity information that is useful to compute finite - tree information .
an obvious choice for such a parameter is the domain combination @xmath2 . it is worth noting that , in @xcite , the correctness of the finite - tree analysis is proved by _ assuming _ the correctness of the underlying analysis on the parameter @xmath477 .
thus , thanks to the results shown in this paper , the proof for the domain @xmath478 can now be considered complete .
codish et al .
@xcite describe an algebraic approach to the sharing analysis of logic programs that is based on _ set logic programs_. a set logic program is a logic program in which the terms are sets of variables and standard unification is replaced by a suitable unification for sets , called
_ aci1-unification _
( unification in the presence of an associative , commutative , and idempotent equality theory with a unit element ) .
the authors show that the domain of _ set - substitutions _ , with a few modifications , can be used as an abstract domain for sharing analysis .
they also provide an isomorphism between this domain and the set - sharing domain @xmath149 of jacobs and langen .
the approach using set logic programs is also generalized to include linearity information , by suitably annotating the set - substitutions , and the authors formally state the optimality of the corresponding abstract unification operator @xmath479 ( lemma a.10 in the appendix of @xcite ) .
however , this operator is very similar to the classical combinations of set - sharing with linearity @xcite : in particular , the precision improvements arising from this enhancement are only exploited when the two terms being unified are definitely independent . as we have seen in this paper , such a choice results in a sub - optimal abstract unification operator , so that the optimality result can not hold . by looking at the proof of lemma a.10 in @xcite
, it can be seen that the case when the two terms possibly share a variable is dealt with by referring to an example : this one is supposed to show that all the possible sharing groups can be generated .
however , even our improved operator correctly characterizes the given example , so that the proof is wrong .
it should be stressed that the @xmath314 operator presented in this paper , though remarkably precise , is not meant to subsume all of the proposals for an improved sharing analysis that appeared in the recent literature ( for a thorough experimental evaluation of many of these proposals , the reader is referred to @xcite ) . in particular , it is not difficult to show that our operator is not the optimal approximation of concrete unification . in a very recent paper @xcite ,
j. howe and a. king consider the domain @xmath2 and propose three optimizations to improve both the precision and the efficiency of the ( classical ) abstract unification operator .
the first optimization is based on the same observation we have made in this paper , namely that the independence check between the two terms being unified is not necessary for ensuring the correctness of the analysis .
however , the proposed enhancement does not fully exploit this observation , so that the resulting operator is strictly less precise than our @xmath314 operator ( even when the operator @xmath320 does not come into play ) . in fact , the first optimization of @xcite is not uniformly more precise than the classical proposals .
the following example illustrates this point .
[ ex : amguhw - less - precise - than - classical - amgu ] let @xmath480 , @xmath330 and @xmath481 , where @xmath482 .
since @xmath74 and @xmath27 are linear and independent , @xmath314 as well as all the classical abstract unification operators will compute @xmath483 , where @xmath484 in contrast , a computation based on , results in the less precise abstract element @xmath485 , where @xmath486 the second optimization shown in @xcite is based on the enhanced combination of set - sharing and freeness information , which was originally proposed in @xcite . in particular , the authors propose a slightly different precision enhancement , less powerful as far as precision is concerned , which however seems to be amenable for an efficient implementation .
the third optimization in @xcite exploits the combination of the domain @xmath2 with the groundness domain @xmath322 .
in this paper we have introduced the abstract domain @xmath2 , combining the set - sharing domain @xmath149 with freeness and linearity information .
while the carrier of @xmath2 can be considered standard , we have provided the specification of a new abstract unification operator , showing examples where this operator achieves more precision than the classical proposals .
the main contributions of this paper are the following : * we have defined a precise abstraction function , mapping arbitrary substitutions in rational solved form into their _ most precise _ approximation on @xmath2 ; * using this abstraction function , we have provided the mandatory proof of _ correctness _ for the new abstract unification operator , _ for both finite - tree and rational - tree languages _ ; * we have formally shown that the domain @xmath2 is _ uniformly _ more precise than the domain @xmath1 ; we have also provided an example showing that all the classical approaches to the combinations of set - sharing with freeness and linearity fail to satisfy this property ; * we have shown that , in the definition of @xmath2 , we can replace the set - sharing domain @xmath149 by its non - redundant version @xmath157 . as a consequence ,
it is possible to implement an algorithm for abstract unification running in _ polynomial time _ and still obtain the same precision on all the considered observables , that is groundness , independence , freeness and linearity .
we recognize the hard work required to review technical papers such as this one and would like to express our real gratitude to the journal referees for their critical reading and constructive suggestions for preparing this improved version . ,
gori , r. , hill , p. m. , and zaffanella , e. 2001 .
finite - tree analysis for constraint logic - based languages . in _
static analysis : 8th international symposium , sas 2001 _ , p. cousot , ed .
lecture notes in computer science , vol .
springer - verlag , berlin , paris , france , 165184 . , hill , p. m. , and zaffanella , e. 1997 .
set - sharing is redundant for pair - sharing . in _
static analysis : proceedings of the 4th international symposium _ , p. van hentenryck , ed .
lecture notes in computer science , vol . 1302 .
springer - verlag , berlin , paris , france , 5367 . , hill , p. m. , and zaffanella , e. 2000 .
efficient structural information analysis for real clp languages . in _ proceedings of the 7th international conference on logic for programming and automated reasoning ( lpar 2000 ) _ ,
m. parigot and a. voronkov , eds .
lecture notes in artificial intelligence , vol . 1955 .
springer - verlag , berlin , runion island , france , 189206 . , zaffanella , e. , gori , r. , and hill , p. m. 2001 .
boolean functions for finite - tree dependencies . in _ proceedings of the 8th international conference on logic for programming , artificial intelligence and reasoning ( lpar 2001 ) _ , r. nieuwenhuis and a. voronkov , eds .
lecture notes in artificial intelligence , vol .
springer - verlag , berlin , havana , cuba , 579594 . , zaffanella , e. , and hill , p. m. 2000 . enhanced sharing analysis techniques : a comprehensive evaluation . in _ proceedings of the 2nd international acm sigplan conference on principles and practice of declarative programming _ ,
m. gabbrielli and f. pfenning , eds .
association for computing machinery , montreal , canada , 103114 .
freeness , sharing , linearity and correctness all at once . in _ static analysis , proceedings of the third international workshop _ , p. cousot , m. falaschi , g. fil , and a. rauzy , eds .
lecture notes in computer science , vol .
springer - verlag , berlin , padova , italy , 153164 .
an extended version is available as technical report cw 179 , department of computer science , k.u .
leuven , september 1993 . ,
codish , m. , and mulkers , a. 1994a .
abstract unification for a composite domain deriving sharing and freeness properties of program variables . in _ verification and analysis of logic languages , proceedings of the w2 post - conference workshop , international conference on logic programming _
, f. s. de boer and m. gabbrielli , eds .
santa margherita ligure , italy , 213230 . , codish , m. , and mulkers , a. 1994b . a composite domain for freeness , sharing , and compoundness analysis of logic programs .
technical report cw 196 , department of computer science , k.u .
leuven , belgium .
july . , codish , m. , and mulkers , a. 1995 . abstracting unification : a key step in the design of logic program analyses . in
_ computer science today : recent trends and developments _ , j. van leeuwen , ed .
lecture notes in computer science , vol .
springer - verlag , berlin , 406425 . , dams , d. , fil , g. , and bruynooghe , m. 1993 .
freeness analysis for logic programs and correctness ?
in _ logic programming : proceedings of the tenth international conference on logic programming _ , d. s. warren , ed .
mit press series in logic programming . the mit press , budapest , hungary , 116131 .
an extended version is available as technical report cw 161 , department of computer science , k.u .
leuven , december 1992 . , mulkers , a. , bruynooghe , m. , garca de la banda , m. , and hermenegildo , m. 1993 . improving abstract interpretations by combining domains . in _ proceedings of the acm sigplan symposium on partial evaluation and semantics - based program manipulation_. acm press , copenhagen , denmark , 194205 . also available as technical report cw 162 ,
department of computer science , k.u .
leuven , december 1992 .
abstract interpretation : a unified lattice model for static analysis of programs by construction or approximation of fixpoints . in _ proceedings of the fourth annual acm symposium on principles of programming languages_. acm press , new york , 238252 .
, ranzato , f. , and scozzari , f. 1998 .
complete abstract interpretations made constructive .
in _ proceedings of 23rd international symposium on mathematical foundations of computer science ( mfcs98 ) _ , j. gruska and j. zlatuska , eds .
lecture notes in computer science , vol . 1450 .
springer - verlag , berlin , 366377 . ,
bagnara , r. , and zaffanella , e. 1998 .
the correctness of set - sharing . in _
static analysis : proceedings of the 5th international symposium _ , g. levi , ed .
lecture notes in computer science , vol . 1503 .
springer - verlag , berlin , pisa , italy , 99114 . , bagnara , r. , and zaffanella , e. 2003 . on the analysis of set - sharing , freeness and linearity for finite and rational tree languages .
tech . rep .
2003.08 , school of computing , university of leeds .
available at http://www.comp.leeds.ac.uk/research/pubs/reports.shtml .
accurate and efficient approximation of variable aliasing in logic programs .
in _ logic programming : proceedings of the north american conference _ ,
e. l. lusk and r. a. overbeek , eds . mit press series in logic programming .
the mit press , cleveland , ohio , usa , 154165 . \1994 . a synergistic analysis for sharing and groundness which traces linearity . in _ proceedings of the fifth european symposium on programming _ , d. sannella , ed .
lecture notes in computer science , vol . 788 .
springer - verlag , berlin , edinburgh , uk , 363378 .
\1994 . depth-@xmath488 sharing and freeness . in _ logic programming : proceedings of the eleventh international conference on logic programming _ , p. van hentenryck , ed .
mit press series in logic programming . the mit press , santa margherita ligure , italy , 553568 . \1988 .
complete axiomatizations of the algebras of finite , rational and infinite trees . in _ proceedings , third annual symposium on logic in computer science_. ieee computer society press , edinburgh , scotland , 348357 .
an application of abstract interpretation of logic programs : occur check reduction . in _ proceedings of the 1986 european symposium on programming _ , b. robinet and r. wilhelm , eds .
lecture notes in computer science , vol .
springer - verlag , berlin , 327338 .
correctness , precision and efficiency in the sharing analysis of real logic languages .
ph.d . thesis ,
school of computing , university of leeds , leeds , u.k .
available at http://www.cs.unipr.it/~zaffanella/. , bagnara , r. , and hill , p. m. 1999 .
widening sharing . in _ principles and practice of declarative programming _
, g. nadathur , ed .
lecture notes in computer science , vol . 1702 .
springer - verlag , berlin , paris , france , 414431 . , hill , p. m. , and bagnara , r. 1999 .
decomposing non - redundant sharing by complementation . in _
static analysis : proceedings of the 6th international symposium _ ,
a. cortesi and g. fil , eds .
lecture notes in computer science , vol . 1694 .
springer - verlag , berlin , venice , italy , 6984 . | it is well - known that freeness and linearity information positively interact with aliasing information , allowing both the precision and the efficiency of the sharing analysis of logic programs to be improved . in this paper
we present a novel combination of set - sharing with freeness and linearity information , which is characterized by an improved abstract unification operator .
we provide a new abstraction function and prove the correctness of the analysis for both the finite tree and the rational tree cases .
moreover , we show that the same notion of redundant information as identified in @xcite also applies to this abstract domain combination : this allows for the implementation of an abstract unification operator running in polynomial time and achieving the same precision on all the considered observable properties . abstract interpretation ; logic programming ; abstract unification ; rational trees ; set - sharing ; freeness ; linearity . |
the properties of a giant resonance in nuclei are commonly determined from the distorted wave born approximation ( dwba ) analysis of its excitation cross - section by inelastic scattering of a certain projectile .
the transition potential required in actual implementation of dwba calculation is usually obtained by convoluting the projectile - nucleus interaction with the transition density associated with the giant resonance .
the relevant transition density can be obtained from a microscopic theory of the giant resonance , such as the hartree - fock ( hf ) based random phase approximation ( rpa ) .
however , the use of a macroscopic transition density @xmath3 greatly simplifies the application of the giant multipole resonance theory to the analysis of the experimental data .
the simple form of the transition density @xmath4 obtained by the scaling approximation , is a well - known example of the macroscopic transition density @xmath5 commonly used in the case of the isoscalar giant monopole resonance ( isgmr ) @xcite .
the transition density of eq .
( [ tr7 ] ) nicely agrees with the isgmr transition density obtained in microscopic hf - rpa calculations .
it has a one - node structure , satisfying the condition of particle number conservation @xmath6 unfortunately the scaling consideration can not be extended to the overtone of the isgmr , where @xmath3 has a two - node structure . to derive the macroscopic transition density @xmath3 in this more general case
, one can use the well - known method @xcite of determining @xmath7 from the local sum rule which is exhausted by one collective state with the appropriate choice of the transition operator @xmath8 .
however , in the quantum random phase approximation , the highly excited collective modes are strongly fragmented over a wide range of energy and a special averaging procedure must be employed to determine the macroscopic transition density corresponding to an average collective excitation . in this respect , the semiclassical fermi - liquid approach ( fla ) @xcite is more appropriate .
both the main isgmr and its overtone are well - defined within the fla as single resonance states .
this fact enables us to derive the transition operator @xmath9simply by maximizing the fraction of the energy - weighted sum rule ( fewsr ) exhausted by the single overtone . in the present work
we suggest a procedure to derive the macroscopic transition density for the isgmr overtone using both the hf based rpa and the fermi - liquid approaches .
we remark that some preliminary results of this investigation were presented in ref .
@xcite ( see also ref .
* * the transition density @xmath10 for a certain eigenstate @xmath11 of a nucleus with @xmath12 nucleons is given by @xmath13 where @xmath14 is the particle density operator and @xmath15 represents the ground state of the nucleus .
the transition density reflects the internal structure of the nucleus and does not depend on the external field .
however , a problem arises if one intends to derive the transition density @xmath16 for a group of the thin - structure resonances in the giant multipole resonance ( gmr ) region .
an appropriate averaging procedure is necessary in this case and @xmath16 can be evaluated if the nucleus is placed in an external field @xmath17 where the transition operator @xmath18 is so chosen that it provides a preferable excitation of the above mentioned thin - structure resonances .
let us introduce the local strength function @xmath19 and the energy smeared local strength function @xmath20 defined near the gmr energy @xmath21 by @xmath22 the corresponding strength functions are given by @xmath23 and @xmath24 let us assume , for the moment , that the operator @xmath25excites only a single state @xmath26 , within the energy interval @xmath27 .
the corresponding transition density @xmath28 is then given by the following _ exact _ expression @xmath29 we will extend expression ( [ tr1 ] ) to the case of a group of the thin - structure resonances in the gmr region which are excited by the operator @xmath25 and define the smeared transition density @xmath30 as @xmath31 note that eq .
( [ tr2 ] ) is associated with the strength in the region of @xmath27 and is consistent with the smeared strength function @xmath32 for a single resonance state .
that is ( see also eq .
( [ se0 ] ) ) , @xmath33 we also point out that with the lorentz s function @xmath34 the energy smeared @xmath35 is given by @xmath36 and the smeared transition density @xmath30 is obtained from @xmath37 the consistency condition , eq .
( [ se1 ] ) , then reads @xmath38 in the quantum rpa , the local strength function @xmath39 is related to the rpa green s function @xmath40 by @xcite @xmath41 \,d{\bf r}\,^{\prime } \ . \label{se4}\ ] ] for the isoscalar monopole and dipole excitations , the transition operator @xmath42 is taken in the form of @xmath43 and @xmath44 with an appropriate choice of the radial functions @xmath45 and @xmath46 , see below . in the following , the quantum transition density for the main isgmr and its overtone is evaluated using the eq .
( [ tr2 ] ) with @xmath47 taken separately for the isgmr and the overtone regions .
let us consider the local energy - weighted sum @xmath48 given by ( see eq .
( [ sre ] ) ) @xmath49 the continuity equation provides the following sum rule @xcite @xmath50 let us assume that only one state @xmath26 exhausts the sum rule eq .
( [ m1r2 ] ) .
then for the corresponding ( macroscopic ) transition density , @xmath51 , we have from eqs .
( [ m1r1 ] ) and ( [ m1r2 ] ) the following expression @xmath52 where the normalization coefficient @xmath53 can be found using the energy - weighted sum rule ( ewsr ) @xmath54 taking into account that @xmath55 we obtain @xmath56 thus , the macroscopic transition density @xmath57 of eq .
( [ tr5 ] ) coincides with the quantum transition density for a certain state @xmath26 if the single state @xmath26exhausts all the ewsr , eq . ( [ m1 ] ) , associated with the transition operator @xmath42 . in the case of the transition operator @xmath58 from eqs . ( [ fl0 ] ) and ( [ f1 ] ) ,
the ewsr is given by @xcite @xmath59 .
\label{m1l}\ ] ] assuming that the energy - weighted transition strength ( ewts ) @xmath60 fully exhausts the ewsr associated with @xmath61 , we obtain the expression for the macroscopic transition density for the state at @xmath62 from eq .
( [ tr5 ] ) @xcite @xmath63 \rho _ { { \rm eq}}(r)y_{lm}({\bf \hat{% r } } ) .
\label{trl}\ ] ] for @xmath64 we have from eq .
( [ trl ] ) the commonly used isgmr result of eq .
( [ tr7 ] ) .
the main ( lowest ) isgdr is the spurious state with the eigenenergy @xmath65 .
the next isgdr is the overtone . assuming that the ewts for the @xmath66 overtone equals the ewsr associated with the operator @xmath67 we obtain from eq .
( [ tr5 ] ) the macroscopic transition density as @xmath68 for the isoscalar dipole mode , the translation invariance condition is used for the derivation of @xmath69 .
this condition implies that the center of mass of the system can not be affected by internal excitation .
we thus have , @xmath70 from eqs .
( [ tr8 ] ) and ( [ cm1 ] ) one obtains , see also refs .
@xcite , @xmath71 where @xmath72 is the mean square radius . the free ( mixing ) parameter appearing in the transition operator @xmath73 ( similar to @xmath74 of eq .
( [ f1 ] ) for the @xmath75 case ) can be determined by an appropriate condition leading to a general method for the evaluation of the transition density for the overtone mode . in this work
we present this method for the case of the monopole mode , @xmath76 let us introduce the transition operator @xmath9 as @xmath77 the corresponding macroscopic transition density @xmath5 is obtained from eqs .
( [ tr5 ] ) and ( [ f4 ] ) as @xmath78 \rho _ { { \rm eq}}(r)y_{00}(% { \bf \hat{r}})\qquad { \rm for\qquad } l=0\quad ( { \rm overtone } ) .
\label{tr9}\ ] ] the determination of the parameter @xmath79 in eq .
( [ tr9 ] ) requires an additional consideration since for the @xmath64 case we have no fundamental condition such as eq .
( [ cm1 ] ) for the @xmath75 mode .
we note , however , that if we assume that the isgmr has the transition density of eq .
( [ tr7 ] ) and require that @xmath80 i.e. , the isgmr is not excited by the scattering operator of eq .
( [ f4 ] ) we have @xmath81 similar result is obtained by imposing the condition that the scattering operator @xmath82 does not excite the overtone of the isgmr , assuming the transition density of eq .
( [ tr9 ] ) . following the general requirement for the proper use of eq .
( [ tr5 ] ) in the derivation of the macroscopic transition density @xmath3 , we can determine the parameter @xmath79 from the condition that the transition operator @xmath9 provides for the single overtone the maximum fraction of the energy - weighted sum rule @xmath83 of eq .
( [ m1 ] ) .
the transition density @xmath16 and the strength function @xmath84 can also be evaluated within the semiclassical fermi - liquid approach . for a given multipolarity @xmath85 and overtone @xmath86 the fla transition density
is given by @xcite @xmath87 @xmath88 \rho _ { 0}y_{l0}({\bf \hat{r } } ) , \label{tr4}\ ] ] where @xmath89 is the bulk density , @xmath90 is the equilibrium nuclear radius and the parameter @xmath91 is determined by the translation invariance condition ( eq . ( [ cm1 ] ) ) in the case of the isoscalar dipole compression mode and is given by @xmath92 the wave numbers @xmath93 are derived from the boundary conditions of the fla model : the normal component of the tensor pressure @xmath94 , created by a sound wave , on the free surface of the nucleus should be equal to zero @xmath95 note that for the case of compression sound modes , the contribution from the surface tension pressure is negligible and it was omitted in eq .
( [ b1 ] ) . the boundary condition ( eq .
( [ b1 ] ) ) leads to the following secular equation ( see ref .
@xcite ) @xmath96_{r = r_{0}}=0,\quad \,d_{\mu } = { % \frac{4\,\mu } { m\,\rho _ { 0}c_{0}^{2}},}\qquad \label{sec2}\ ] ] where @xmath97 is the zero sound velocity and the coefficient @xmath98 determines the contribution from the dynamical fermi surface distortion associated with the collective motion in a fermi liquid . in the case of a quadrupole distortion of the fermi surface
, one has @xcite @xmath99 here , @xmath100 is the equilibrium pressure of a fermi gas , @xmath101 is the fermi energy , and @xmath102 and@xmath103 are the eigenfrequency and the relaxation time for sound excitations in the fermi liquid , respectively .
the relaxation time @xmath104 is assumed to be frequency dependent because of the memory effect in the collision integral @xcite . following refs .
@xcite we take @xmath105 where @xmath106 is the constant related to the differential cross section for the scattering of two nucleons in the nuclear interior . in the case of isoscalar sound mode , we will adopt @xmath107 mev @xcite .
the eigenfrequency @xmath102 is obtained from the dispersion relation @xmath108 where @xmath109 is given by @xmath110 here @xmath111 is the nuclear incompressibility coefficient and @xmath112 is the friction coefficient @xmath113 the smeared fla strength function @xmath114 can be obtained in a way similar to @xmath115 , obtained within the quantum approach of eq .
( [ se3 ] ) .
that is , @xmath116 the smearing function @xmath117 in eq .
( [ sfla ] ) is given by ( @xmath118 ) @xmath119 here , @xmath120 is the damping parameter due to the viscosity of the fermi liquid and @xmath121 , where the eigenfrequency @xmath122 is obtained as a solution to both the dispersion equation ( [ disp ] ) and the secular equation ( [ sec2 ] ) .
we point out that the amplitude @xmath123 in eq .
( [ tr4])@xmath124for the fla transition density @xmath125 is derived as the amplitude of the quantum oscillations @xmath126 where @xmath127 is determined by eq .
( [ sec2 ] ) and @xmath128 is the corresponding mass coefficient with respect to the density oscillations .
the collective mass coefficient @xmath128 can be found from the collective kinetic energy @xmath129 for the particle density oscillations .
the collective kinetic energy is derived as @xmath130 for the compression modes @xmath64 , the mass coefficient @xmath131 is given by @xcite @xmath132 _ { x = x_{0}},\qquad { \rm % for\qquad } l=0 , \label{b10}\ ] ] where @xmath133 .
we have carried out calculations for the isgmr in the frameworks of hf based rpa and the semiclassical fermi liquid approach as briefly outlined in the preceding sections .
we have evaluated the smeared fla strength function @xmath114 of eq .
( [ sfla ] ) and the hf - rpa smeared strength function @xmath115 of eq .
( [ se3 ] ) for the isgmr in several nuclei , for @xmath134 from eq .
( [ f4 ] ) . in the subsequent discussions ,
the hf - rpa results presented are obtained using the skyrme force skm@xmath135 @xcite . for the rpa calculations to be highly accurate we discretized the continuum in a large box of size 90 fm and
use a smearing parameter @xmath136 mev , in evaluating the rpa green s function ( see eq .
( [ se4 ] ) ) , and allow particle - hole excitations up to 500 mev ( see ref .
@xcite for the details ) . in case of the fla calculations
we have adopted the values of @xmath137 @xmath138 , @xmath139 and @xmath140 fm . in table
[ fla - rpa ] we compare the fla and rpa results for the centroid energies @xmath141 and @xmath142 corresponding to the main and overtone mode of the isgmr , respectively .
we see from this table that the fla and rpa results are in qualitative agreement . the small differences ( @xmath143 )
can be understood by the fact that the centroid energy mainly depends on the size of the system .
the values of the mean square radii and the higher moments of the ground state density distribution are smaller in the fla than the ones obtained from hf calculations . note also that the ratio @xmath144 in both models considered is greater than two ( @xmath145 ) .
we point out that @xmath146 is due to the fermi - surface distortion effect as noted earlier in ref .
@xcite .
we have performed a comparison of the macroscopic transition density @xmath147 with the ones obtained within the hf - rpa , @xmath148 , and the fla , @xmath149 , approaches for the main resonance @xmath150 and its overtone .
the fla transition density is given by eq .
( [ tr4 ] ) with the wave number @xmath151 obtained from the secular equation ( [ sec2 ] ) .
the contribution of the isgmr overtone to the ewsr for the case of the transition operator @xmath9 is given by @xmath152 where @xmath153 .
the eigenfrequency @xmath154 is obtained from the dispersion equation ( [ disp ] ) and the wave number @xmath155 is the second ( overtone ) solution to the secular equation ( [ sec2 ] ) .
the ewsr for the transition operator @xmath9 reads @xmath156 now , we will determine the parameter @xmath79 from the condition that the value of the ratio of the partial sum @xmath157 to the total sum@xmath158 is the maximum . in fig . 1
, we have plotted the ratio @xmath159 as a function of the parameter @xmath79 for the nucleus @xmath160pb obtained in the fla ( dashed line ) and the hf - rpa ( solid line ) models . as seen from fig .
1 , the maximum value of the ratio of @xmath157 to @xmath158 for the fla model is achieved for @xmath161 @xmath162 where the overtone exhausts about @xmath163 of the ewsr . in the case of hf - rpa , the maximum ratio is achieved for @xmath164 @xmath162 where the overtone , in the energy range of 25 - 50 mev , exhausts about @xmath165 of the ewsr .
it is interesting to note that if we use the fla and hf ground - state densities to calculate @xmath1 from eq .
( [ xi02 ] ) , we get @xmath166 and 79.0 @xmath167 , respectively .
these values are close to the corresponding ones ( 68.3 and 78.6 @xmath167 ) obtained by the condition of maximizing the ratio @xmath168 .
this means that the mixing parameter @xmath79 in the transition operator @xmath169 of eq .
( [ f4 ] ) can also be derived from the condition that the main isgmr gives a minimal contribution to the energy - weighted sum rule @xmath158 .
the fla as well as hf - rpa calculations show that the difference in the values of @xmath79 obtained in both conditions does not exceed @xmath170 .
we have also calculated the dependence of the parameter @xmath79 on the nuclear mass number @xmath12 .
following the same procedure we find for @xmath171zr , @xmath172sn and @xmath173sm nuclei the value of mixing parameter @xmath174 38.6 ( 48.5 ) , 45.9 ( 56.7 ) and 53.1 ( 66.1 ) @xmath167 from the fla ( rpa ) calculations , respectively .
these values can be well approximated by @xmath175 and 2.36@xmath176 @xmath167 for the fla and rpa approaches , respectively . in fig .
2 , we plot the fla and rpa results for the fraction energy - weighted transition strength as a function of the excitation energy obtained for the transition operator @xmath134 for the @xmath160pb nucleus .
we use @xmath177 68.3 and 78.6 @xmath167 in the fla and rpa calculations , respectively .
one can clearly see that the rpa calculation yields a wide resonance of width of @xmath178 mev around the excitation energy @xmath179 mev which corresponds to the isgmr overtone . whereas , in the case of fla , the transition operator @xmath180 ( eq . ( [ f4 ] ) ) with an appropriate value of @xmath1 gives rise to a well defined resonance for the overtone mode .
we also notice that the rpa results have the reminiscence of the isgmr main tone but it is practically eliminated in the fla calculations . in figs .
3@xmath91 and 3@xmath181 , we compare the radial macroscopic transition density @xmath182 of eq .
( [ tr9 ] ) , obtained using the hf ground - state density , and the corresponding fla and hf - rpa ones for the overtone of the isgmr .
the radial transition density @xmath183 for a certain multipolarity @xmath85 is given by @xmath184 the macroscopic transition densities for the overtone in figs .
3a and 3b are not the same . because , for an appropriate comparison , in fig .
3a , we have plotted @xmath185 obtained using @xmath186 and it is normalized to @xmath165 of the ewsr . on the other hand , @xmath185 plotted in fig .
3b corresponds to @xmath187 and is normalized to @xmath163 of the ewsr .
further , the rpa transition density is calculated by averaging over the energy range of 25 - 50 mev .
notice that the shift of the nodes of @xmath188 to the left with respect to the ones of @xmath189 is caused by the fact that in contrast to eq .
( [ tr9 ] ) used for the macroscopic transition density , in the fla we use sharp nuclear surface , see eq .
( [ tr4 ] ) .
we also looked into the energy dependence of the rpa transition density for the operator @xmath73 over the range of energy employed for the averaging .
we see that over the entire range considered , the transition density has two nodal structure and the distance between the nodes decreases with the increase in energy . for example , averaging over @xmath190 mev range , we find that at the excitation energies of 30 , 40 and 50 mev , the distances between the two nodes are 3.1 , 2.7 , and 2.4 fm , respectively , which reflects the fact that the microscopic transition density is state dependent .
we have used the microscopic transition densities for the operator @xmath73 to evaluate the cross - section for the isgmr overtone mode excited via inelastic scattering of @xmath191-particles with energies 240 and 400 mev .
we have used the folding model ( fm)-dwba to calculate the excitation cross - section ( see ref .
@xcite for details ) .
we find that for the @xmath191-particles with 400 mev energy , the calculated cross - section is about 7 - 10 times higher than the one obtained for @xmath191-particles with 240 mev .
note that for a monopole resonance the cross - section is maximum at @xmath192 .
the values of the cross - section at @xmath192 for the peak energy of the isgmr overtone are 0.5 and 3.5 mb/(sr mev ) for the case of 240 and 400 mev , respectively .
we point out that the maximum cross - section for the case of 240 mev @xmath191-particles is below the current experimental sensitivity of about 2 mb/(sr mev ) @xcite
. it may be possible to identify the isgmr overtone mode with 400 mev @xmath191-particles .
the transition density @xmath193 for the compression modes is distributed over the nuclear interior and has a node close to the nuclear surface for both the main isgmr and its overtone .
the transition density @xmath194 of the overtone has an additional node in the nuclear interior .
this feature of @xmath193 can be tested by evaluating the strength distribution @xmath195 of the electromagnetic operator @xmath196 .
the strength function @xmath197 for a certain eigenstate @xmath198 is given by @xmath199 where @xmath200 the strength function @xmath201 is related to the excitation function of electron - nucleus scattering in the born approximation .
we use eqs .
( [ slk1 ] ) and ( [ slk2 ] ) to calculate the energy - weighted sums @xmath202 and @xmath203 for the main isgmr and its overtone , respectively . in fig .
4 , we display the @xmath204 dependence of the fraction energy - weighted sums @xmath205 and @xmath206 for the @xmath160pb nucleus obtained from the fla ( dashed line ) and hf - rpa ( solid line ) approaches . it can be seen from fig .
4 that @xmath205 and @xmath207 depend strongly on @xmath204 .
a shift of the maximum of the ratio @xmath206 for the overtone to the higher value of wave number @xmath204 is due to the more complicated nodal structure of the transition density associated with the overtone as compared with the main resonance .
this shift can be exploited to separate the modes in electron - nucleus scattering . in fig .
5 we plot the surface and the volume contributions to the integral in eq .
( [ slk2 ] ) for the transition density associated with overtone mode ( see eq .
( [ tr4 ] ) ) . for smaller @xmath204
, there is a cancellation between the surface and the volume contributions leading to a peak structure for the overtone response as shown in fig .
starting from the local strength function @xmath39 and using the smearing procedure , we have extended the quantum expression for the transition density @xmath10 to the case of a group of the thin - structure resonances which are localized in the gmr region and are excited due to the specifically chosen transition operator @xmath208 .
our approach was applied to the study of the transition density of the isgmr overtone . in this case ,
an appropriate form of the transition operator @xmath58 is given by @xmath209 , see eq .
( [ f4 ] ) .
the mixing parameter @xmath79 was determined from the condition that the transition operator @xmath210 provides for the single overtone the maximum fraction of the energy - weighted sum rule @xmath83 of eq .
( [ m1 ] ) .
the mixing parameter @xmath79 depends on the nuclear mass number @xmath12 .
this dependence is well approximated by @xmath211 .
we have applied our smearing procedure ( using @xmath212 associated with the maximum fewsr of the overtone ) to the evaluation of the smeared out transition density @xmath30 of eq .
( [ tr2 ] ) within the hf - rpa .
we have shown that the smearing procedure for the isgmr overtone region provides a simple two nodal structure of @xmath30 ( see the solid line in fig .
3@xmath91 ) , as expected for the @xmath64 overtone .
moreover , the transition density @xmath213 , obtained by the averaging over many quantum states , resembles its macroscopic counterpart .
this fact is well illustrated in fig .
3@xmath91 by comparing the quantum smeared transition density @xmath30 with the macroscopic one @xmath51 of eq .
( [ tr9 ] ) .
an independent derivation of the smeared out transition density @xmath214 can be also obtained using the semiclassical approaches.in sec .
iv , we have applied a simple semiclassical fermi - liquid approximation to the evaluation of the smeared out ( in quantum mechanical sense ) transition density @xmath215 .
we have used the same form of the transition operator @xmath9 as in the case of quantum hf - rpa calculation to provide an additional check of the derivation of the mixing parameter @xmath79 from the energy - weighted sum @xmath83 .
we found a good agreement between the values of parameter @xmath79 obtained in both the quantum and the semiclassical approaches .
it is important to emphasize that equation ( [ xi02 ] ) together with eq .
( [ tr5 ] ) provides a simple expression for the macroscopic transition density that can be employed in the folding model - dwba analysis of excitation cross - section of the isgmr overtone .
the nodal structure of the semiclassical transition density @xmath216 is similar to that of both the quantum , @xmath217 , and the macroscopic , @xmath218 , cases , see figs . 3@xmath91 and 3@xmath181 .
a discrepancy occurs in the surface region , where the particular behavior of @xmath219 is due to the assumption of the sharp surface of the nucleus in the fla model .
this discrepancy is not so significant in the integral quantities like the strength functions .
this is illustrated in fig .
4 for the case of the nuclear response to the electromagnetic - like external field @xmath220 .
the ratios @xmath221 and @xmath222 for the isgmr and its overtone , respectively , show an distinct feature in the @xmath204-dependence .
namely , for a certain value of the wave number @xmath204 , the strength function for the overtone reaches a maximum whereas the contribution of the main resonance to the strength function is strongly suppressed . this fact can be exploited to separate the isgmr and the overtone modes in electron - nucleus scattering by varying the electron s momentum transfer @xmath204 .
this work was supported in part by the us department of energy under grant # doe - fg03 - 93er40773 .
one of us ( v.m.k . )
thank the cyclotron institute at texas a&m university for the kind hospitality .
h. s. khler , nucl . phys . *
a378 * , 159 ( 1982 ) .
j. bartel , p. quentin , m. brack , c. guet and h. -b .
hakansson , nucl . phys . *
a386 * , 79 ( 1982 ) .
d. h. youngblood , h. l. clark and y. w. lui , phys .
. lett . * 82 * , 691 ( 1999 ) .
a. kolomietz , o. pochivalov and s. shlomo , phys .
c * 61 * , 034312 ( 2000 ) .
d. h. youngblood ( private communication ) .
1 . the ratio @xmath223 of the partial contribution of the overtone to the ewsr as a function of the parameter @xmath79 in the transition operator @xmath9 ( eq . ( [ f4 ] ) )
obtained within the fla ( dashed line ) and the hf - rpa ( solid line ) approaches , for the monopole mode @xmath224 in the nucleus @xmath225 fig .
3@xmath91 . the hf - rpa transition density , @xmath227 multiplied by @xmath228 for the overtone of the isgmr in the nucleus @xmath229 ( solid line ) and the corresponding macroscopic transition density @xmath3 taken at @xmath230 @xmath231 ( dotted line ) .
3@xmath181 . the fla transition density , @xmath232 , multiplied by @xmath228 for the overtone of the isgmr in the nucleus @xmath233 ( dashed line ) and the corresponding macroscopic transition density @xmath3 taken at @xmath161 @xmath234 ( dotted line ) . fig .
the ratio @xmath235 for the main ( @xmath236 ) isgmr and its overtone ( @xmath237 ) , respectively , as a function of the wave number @xmath238 obtained for electromagnetic operator @xmath239 for the @xmath240pb nucleus .
the dashed and the solid lines represent the fla and hf - rpa results , respectively . fig .
partial contributions of the volume ( vol ) and surface ( surf ) terms of the fla transition density of the isgmr overtone ( see eq .
( [ tr4 ] ) ) to the integral in eq .
( [ slk2 ] ) ( dashed lines ) .
the solid line shows the sum of both the volume and the surface terms .
.[fla - rpa ] comparison of the fla and rpa results for the centroid energies ( in mev ) for the main ( @xmath141 ) and the overtone ( @xmath142 ) modes of isgmr . in the case of rpa calculations
the values of @xmath141 are obtained by integrating the strength function for the operator @xmath241 over the energy range of 0 - 60 mev and the values of @xmath142 are obtained using the operator @xmath242 ) and the energy ranges of 35 - 60 , 28 - 60 , 27 - 55 and 25 - 50 mev for @xmath243 , @xmath244 , @xmath245 and @xmath246 nuclei , respectively .
the experimental data for the main tone is taken from the ref .
[ cols="^,^,^,^,^,^,^,^ " , ] | we calculate the transition density for the overtone of the isoscalar giant monopole resonance ( isgmr ) from the response to an appropriate external field @xmath0 obtained using the semiclassical fluid dynamic approximation and the hartree - fock ( hf ) based random phase approximation ( rpa ) .
we determine the mixing parameter @xmath1 by maximizing the ratio of the energy - weighted sum for the overtone mode to the total energy - weighted sum rule and derive a simple expression for the macroscopic transition density associated with the overtone mode .
this macroscopic transition density agrees well with that obtained from the hf - rpa calculations .
we also point out that the isgmr and its overtone can be clearly identified by considering the response to the electromagnetic external field @xmath2 . |
in the human genome project ( from the year 1990 to 2003 ) sequences of chemical base pairs that make up human dna were intensively analyzed and determined , as it carries important genetic information .
dna is a polymer made up of a large number of deoxyribonucleotides , each of which composed of a nitrogenous base , a sugar and one or more phosphate groups @xcite .
similar ribonucleotides polymerize to form rna , which is also an important substance to produce a template for protein synthesis .
rna sometimes carries genetic information but rarely shows enzymatic functions @xcite .
one big issue of the post genome project is the proteomics since proteins play crucial roles in virtually all biological processes such as enzymatic catalysis , coordinated motion , mechanical support , etc .
a protein is a long polymer of amino acids , and folds into regular structures to show its biological function .
sugars and amino acids in these biological polymers contain carbon atoms , each of which is connected to four different groups , and consequently is able to take two kinds of stereostructures .
the two stereoisomers are mirror images or enantiomers of each other , and are called the d- and l - isomers .
two isomers should have the same physical properties except optical responses to polarized light .
therefore , a simple symmetry argument leads to the conclusion that there are equal amount of d- and l - amino acids or sugars in life .
but the standard biochemical textbook @xcite tells us that `` only l - amino acids are constituents of proteins '' ( @xcite , p.17 ) , and `` nearly all naturally occurring sugars belong to the d - series '' ( @xcite , p.346 ) .
there is no explanation why the chiral symmetry is broken in life and how the homochirality has been brought about on earth .
it is pasteur who first recognized the chiral symmetry breaking in life in the middle of the 19th century . by crystallizing optically inactive sodium anmonium racemates ,
he separated two enantiomers of sodium ammonium tartrates with opposite optical activities by means of their asymmetric crystalline shapes @xcite .
since the activity is observed even in the solution , it is concluded that the optical activity is due to the molecular asymmetry or chirality , not due to the crystalline symmetry . because two enantiomers with different chiralities are identical in every chemical and physical properties except the optical activity , `` artificial products have no molecular asymmetry '' and in 1860 pasteur stated that `` the molecular asymmetry of natural organic products '' establishes `` the only well - marked line of demarcation that can at present be drawn between the chemistry of dead matter and the chemistry of living matter '' @xcite . also by using the fact that asymmetric chemical agents react differently with two types of enantiomers , he separated enantiomers by fermentation
. therefore , once one has an asymmetric substance , further separation of two enantiomers or even a production of a single type of enantiomer may follow @xcite . but how has the first asymmetric organic compound been chosen in a prebiotic world ? this problem of the origin of homochirality in life has attracted attensions of many scientists in relation to the origin of life itself since the discovery of pasteur @xcite .
japp expected a `` directive force '' when `` life first arose '' @xcite , and various `` directive forces '' are proposed later such as different intensities of circularly polarized light in a primordial era , adsorption on optically active crystals , or the parity breaking in the weak interaction .
however , the expected degree of chiral asymmetry or the value of the enantiomeric excess ( ee ) turns out to be very small @xcite , and one needs a mechanism to amplify ee enormously to a level of homochirality .
another scenario for the origin of homochirality was suggested by pearson @xcite such that the chance breaks the chiral symmetry .
though the mean number of right- and left - handed enantiomers are the same , there is nonzero probability of deviation from the equal populations of both enantiomers .
the probability to establish homochirality in a macroscopic system is , of course , very small @xcite , but `` chance produces a slight majority of one type of '' enantiomers and `` asymmetric compounds when they have once arisen '' act as `` breeders , with a power of selecting of their own kind of asymmetry form '' @xcite . in this scenario ,
produced enantiomer acts as a chiral catalyst for the production of its own kind and hence this process should be autocatalytic .
nearly a century later , a long sought autocatalytic system with spontaneous amplification of chiral asymmetry is found by soai and his coworkers @xcite : in a closed reactor with achiral substrates addition of a small amount of chiral products with a slight enantiomeric imbalance yields the final products with an overwhelmingly amplified ee @xcite .
the autocatalytic system also shows significant ee amplification under a variety of organic and inorganic chiral initiators with a small enantiomeric imbalance @xcite .
when many reaction runs are performed without chiral additives , about half of the runs end up with the majority of one enantiomer , and the other half end up with the opposite enantiomer .
the probability distribution of the ee is bimodal with double peaks , showing amplification of ee and further indicating the occurrence of chiral symmetry breaking @xcite . as for the theoretical research on the chiral symmetry breaking , frank is the first to show that a linear autocatalysis with an antagonistic nonlinear chemical reaction can lead to homochirality @xcite .
his formulation with rate equations corresponds to the mean - field analysis of the phase transition in nonequilibrium situation @xcite , and other variants have been proposed @xcite .
all these analyses are carried out for open systems where the concentration of an achiral substrate is kept constant .
asymptotically the system approaches a unique steady state @xcite or even an oscilatory state @xcite , independent of the initial condition .
when the final state is chirally asymmetric , one may call this a chiral symmetry breaking in the sense of statistical mechanics , and it is indicated in the right upper corner in table 1 . on the other hand ,
the soai reaction is performed in a closed system and an achiral substrate is converted to chiral products irreversibly : the substrate concentration decreases in time , and eventually the reaction comes to a halt . after the reaction the ee value increases but its final value is found to depend on the initial state . even though the ee is amplified , the history dependent behavior is different from what we expect from phase transitions in statistical mechanics , so that we write down `` ee amplification '' in the left lower corner in table 1 .
we have analyzed theoretically the chiral symmetry breaking in a closed system in general , and found that with an irreversible nonlinear autocatalysis the system may show ee amplification as in the soai reaction .
furthermore , with an additional recycling back reaction , a unique final state with a finite ee value is found possible , which we call a chiral symmetry breaking as is shown in the right lower corner in table 1 @xcite .
there are also many theoretical works on the soai reaction in a closed system @xcite . here
we give a brief survey of various theoretical models of the ee amplification and the chiral symmetry breaking in a closed system , though it is by no means exhaustive . .
the ee amplification and the chiral symmetry breaking in an open and a closed system .
[ cols="^,^,^,^ " , ]
two stereostructural isomers are called d- and l - enantiomers for sugars and amino acids , but , for general organic compounds , @xmath0 and @xmath1 representation is in common .
we adopt @xmath2 representation hereafter .
we consider a production of chiral enantiomers @xmath0 and @xmath1 from an achiral substrate @xmath3 in a closed system .
actually , in the soai reaction , chiral molecules are produced by the reaction of two achiral reactants @xmath3 and @xmath4 as @xmath5 or @xmath6 .
but in a closed system a substrate of smaller amount controls the reaction , since it is first consumed up and holds the reaction to proceed further .
therefore , in order to grasp main features of chirality selection , it is sufficient to assume that an achiral substrate @xmath3 of a smaller amount converts to @xmath0 or @xmath1 .
the process may further involve formation of intermediate complexes , oligomers etc . , but for the purpose to discuss about relations and difference between the amplification and the symmetry breaking in chirality , we restrict our consideration in this section to the simplest case where only monomers are involved .
the spontaneous production of chiral molecules @xmath0 or @xmath1 from an achiral substrate @xmath3 is described by reactions @xmath7 we assume the same reaction rate @xmath8 for the nonautocatalytic spontaneous production of @xmath0 and @xmath1 enantiomers , since both of them are identical in every chemical and physical aspects : no advantage factor @xcite is assumed .
since the spontaneous production ( [ eq01 ] ) yields only a racemic mixture of two enantiomers , one has to assume some autocatalytic processes .
the simplest is a linear autocatalysis with a reaction coefficient @xmath9 as @xmath10 there may be a linear cross catalytic reaction or erroneous linear catalysis with a coefficient @xmath11 as @xmath12 as will be discussed later , the linear autocatalysis alone is insufficient not only to break the chiral symmetry but also to amplify ee . beyond these linear autocatalyses , nonlinear effects such as quadratic autocatalysis
have been considered as @xcite ; @xmath13 these higher order autocatalytic processes may actually be brought about by the dimer catalysts @xcite , but we confine ourselves to the simplest description ( [ eq04 ] ) and ( [ eq05 ] ) in terms of only monomers @xmath0 and @xmath1 in this section .
consideration on dimers is postponed in the following sections .
the reaction ( [ eq04 ] ) alone can give rise to the amplification of chirality as will be discussed .
one notes that all these processes ( [ eq01 ] ) to ( [ eq05 ] ) unidirectionally produce chiral enantiomers and thus the reaction comes to a halt when the whole achiral substrate is consumed ; the total process is irreversible .
concentrations @xmath14 and @xmath15 of two enantiomers @xmath0 and @xmath1 vary according to the reaction processes described in the previous subsection . in an open system ,
the achiral substrate is steadily supplied in such a way that its concentration @xmath16 is kept constant . on the contrary , in a closed system , achiral substrates are transformed to chiral products and only the total concentration @xmath17 of reactive chemical species is kept constant . by denoting the total concentration of monomeric chiral products with a subscript 1 as @xmath18 the concentration of the substrate @xmath16 is determined as @xmath19 which should be non - negative .
then , the development of reactions in a closed system can be depicted as a flow in the triangular region in the @xmath14-@xmath15 phase space @xmath20 the reaction schemes ( [ eq01 ] ) to ( [ eq05 ] ) induce rate equations for @xmath14 and @xmath15 as ; @xmath21a \nonumber \\
\frac{d s}{dt } = [ f(s)+k_1'r+k_2'rs]a \label{eq09}\end{aligned}\ ] ] with an effective rate coefficient @xmath22 since the right - hand side of eq.([eq09 ] ) for velocities @xmath23 and @xmath24 are always non - negative and in proportion to the concentration of the achiral substrate @xmath16 , both @xmath14 and @xmath15 never decrease and stop to grow after @xmath16 vanishes ( @xmath25 ) .
the asymptotic behavior of a two - dimensional autonomous dynamical system is , in general , known to have fixed points or lines where @xmath26 or to have limit cycles where @xmath27 .
since the system is irreversible as the concentrations of chiral products always increase at the cost of the substrate @xmath3 as @xmath28a \le 0 \label{eq11}\end{aligned}\ ] ] and no epimerization process @xmath29 is assumed , limit cycle is impossible : production stops when all the substrate molecules are transformed to chiral ones ; @xmath30 . since @xmath26 for @xmath31 , the diagonal line @xmath32 is a fixed line for this irreversible system .
any points on the fixed line are marginally stable because they have no guaranteed stability along the fixed line .
our interest is the dynamical behavior of ee or the chiral symmetry breaking and , in particular , as to whether ee increases asymptotically . in order to quantify the monomeric ee or the degree of chiral symmetry breaking
, we define a monomeric order parameter @xmath33 as @xmath34 which is zero for a symmetric or racemic state ( @xmath35 ) , and is nonzero for a symmetry - broken or chiral state .
the evolution equation of @xmath33 is immediately derived from eq.([eq09 ] ) as @xmath36 with coefficients @xmath37 however , as @xmath38 , both the coefficients @xmath3 and @xmath4 vanish asymptotically , and one can not determine the final value of ee from the evolution equation ( [ eq13 ] ) . on a fixed line @xmath32 the ee value varies from @xmath39 at a point @xmath40 to @xmath41 at ( c,0 ) , and the final values of @xmath33 can be anything between @xmath42 and @xmath43 . in order to obtain the final value of ee , one has to solve the evolution ( [ eq09 ] ) and especially to determine the trajectory of the dynamical flow in @xmath44 phase space . as the first step to analyze flow trajectories in a phase space ,
we consider the simplest case when reactions such as spontaneous production ( [ eq01 ] ) , linearly autocatalytic ( [ eq02 ] ) or quadratically autocatalytic ( [ eq04 ] ) reactions are active , respectively
. then the rate equations are simplified as @xmath45 with the rate coefficient @xmath46 defined in eq.([eq10 ] ) .
flow trajectories are obtained by integrating a trajectory equation @xmath47 in the following , an initial condition is set as @xmath48 ( or @xmath49 ) , and we discuss how the final ee value @xmath50 depends on the initial value @xmath51 for a few typical cases . ) , ( b ) linearly autocatalytic ( @xmath52 ) , and ( c ) quadratically autocatalytic ( @xmath53 ) reactions . ]
when chiral molecules are produced only spontaneously and the system has no autocatalytic processes ( @xmath54 , @xmath55 ) , the trajectory in the @xmath44 phase space is obtained as @xmath56 the flow diagram is a straight line with a unit slope , terminating at the fixed line @xmath32 , as shown in fig.[fig1](a ) .
half of the substrate @xmath57 initially present convert to @xmath0 and the remaining half to @xmath1 , and the ee decreases to the value @xmath58 .
the smaller the ratio @xmath59 of the initial amount of chiral species to the total amount of active reactants is , the more the ee decreases , as shown in fig.[fig2](a ) .
+ versus the initial ee @xmath51 for ( a ) spontaneous production ( @xmath60 ) , ( b ) linearly autocatalytic ( @xmath52 ) , and ( c ) quadratically autocatalytic ( @xmath53 ) reactions .
each curve corresponds to a fixed ratio of the initial amount of chiral species to the total amount of active reactants @xmath59 .
as the ratio decreases , the initial slope of the curve decreases in ( a ) , does not change in ( b ) , and increases in ( c ) . ] when only the linearly autocatalytic production of chiral molecules is active ( @xmath52 ) , the trajectory is obtained as @xmath61 and the flow diagram is a straight line radiating from the origin and terminating at the fixed line @xmath32 , as shown in fig.[fig1](b ) .
the ee value remains constant as @xmath62 and is independent of the initial ratio @xmath59 , as shown in fig.[fig2](b ) .
if the erroneous linear catalysis ( [ eq03 ] ) is at work in addition to the linear autocatalysis ( [ eq02 ] ) , rate equations are modified as @xmath63 these equations can be combined to yield @xmath64 with a fidelity factor @xmath46 defined by @xmath65 which is unity without the error @xmath66 and vanishes when @xmath67 .
the ee value @xmath33 is easily integrated as @xmath68 this result shows that the final value is less than the initial @xmath51 whenever there is a finite error process @xmath69 .
in particular , if @xmath70 , the final ee behaves the same with that of the spontaneous production . with only a quadratic autocatalysis ( @xmath71 ) , the trajectory is obtained as @xmath72 and the flow diagram is hyperbolae radiating out from the origin and terminating at the fixed line @xmath32 , as shown in fig.[fig1](c ) . below the diagonal @xmath35 line , the flow bents downward and the ratio @xmath73 decreases as time passes , and consequently the ee value @xmath33 increases .
in other words , by using the relation @xmath74 and @xmath75 , the ee value is easily found from eq.([eq24 ] ) to satisfy a relation @xmath76 with a constant @xmath3 determined by an initial condition . for a given value of @xmath77 , the ee value @xmath33 is determined graphically as a cross point of two curves representing both sides of eq.([eq25 ] ) , as shown in fig.[fig3 ] .
then , it is evident that @xmath33 increases as @xmath77 increases .
the asymptotic value corresponding to @xmath78 is obtained as @xmath79 \label{eq26}\end{aligned}\ ] ] the relation between the initial value @xmath51 of the ee and its final value @xmath50 is depicted in fig.[fig2](c ) for fixed values of the initial amount @xmath80 of chiral species relative to the total amount @xmath81 . the smaller the initial ratio @xmath59 , the more prominent the ee amplification .
the ee amplification curves possess a remarkable property such that they depend on the initial ratio @xmath59 but are independent of the rate coefficient @xmath82 .
the latter determines the time evolution of @xmath33 but not its final value .
if there are other processes involved , for instance the spontaneous production @xmath83 , then amplification curves should depend on the ratio @xmath84 .
is given at a cross point with two curves , @xmath85 and @xmath86 . ]
if there is a racemizing cross catalysis ( [ eq05 ] ) with @xmath87 , the trajectory equation is rewritten as @xmath88 with a crossing effect factor defined by @xmath89 which is unity without a cross catalysis term ( @xmath90 ) and vanishes when @xmath91 .
the solution of eq.([eq27 ] ) is given by @xmath92 from this result one can see that the asymptotic value @xmath50 decreases as @xmath93 decreases .
in particular , when @xmath94 , @xmath95 is independent of @xmath77 and does not vary ; @xmath96 .
namely , chirality amplification by @xmath82 is completely cancelled out by the racemization effect caused by @xmath87 .
in order to understand molecular mechanisms how the quadratic autocatalysis is brought about , the concept of dimer catalyst introduced by kagan and coworkers may be relevant @xcite . assume that monomers @xmath0 and @xmath1 react to form homodimers @xmath97 and @xmath98 with a rate @xmath99 and a heterodimer @xmath100 with a rate @xmath101 @xmath102 corresponding decomposition processes are described as @xmath103 homodimers thus formed are assumed to catalyze of their own enantiomeric type as @xmath104 whereas heterodimers have no preference in enantioselectivity as @xmath105 now , the state of the system is described not only by the concentrations of monomers @xmath14 and @xmath15 , but also by those of dimers denoted as @xmath106 $ ] , @xmath107 $ ] and @xmath108 $ ] respectively .
the rate equations are @xmath109 -2 \nu_{\mathrm{hom } } r^2 + 2 \nu_{\mathrm{hom } } ' [ r_2 ] \nonumber \\ & \quad -\nu_{\mathrm{het } } rs + \nu_{\mathrm{het } } ' [ rs ] + k_{\mathrm{het}}a[rs ] , \nonumber \\ \frac{d [ r_2]}{dt } & = \nu_{\mathrm{hom } } r^2- \nu_{\mathrm{hom } } ' [ r_2 ] , \nonumber \\ \frac{d [ rs]}{dt } & = \nu_{\mathrm{het } } rs- \nu_{\mathrm{het } } ' [ rs ] , \label{eq34}\end{aligned}\ ] ] and the corresponding ones for the @xmath1 enantiomer . the ee or the chiral order parameter is now defined by @xmath110-s-2[s_2]}{q } . \label{eq35}\end{aligned}\ ] ] where @xmath111 is the total concentration of chiral molecules given as @xmath112 + 2[rs]+2[s_2 ] }
. \label{eq36}\end{aligned}\ ] ] if the dimerization and decomposition proceed very fast , the dimer concentrations take quasi - equilibrium values which satisfy @xmath113=[\dot { rs } ] = [ \dot s_2 ] = 0 $ ] and are determined by the instantaneous monomer concentrations @xmath14 and @xmath15 as @xmath114=\frac{\nu_{\mathrm{hom}}}{\nu_{\mathrm{hom } } ' } r^2 , \quad [ rs]=\frac{\nu_{\mathrm{het}}}{\nu_{\mathrm{het } } ' } rs , \quad [ s_2]=\frac{\nu_{\mathrm{hom}}}{\nu_{\mathrm{hom } } ' } s^2 . \label{eq37}\end{aligned}\ ] ] then , the equations for monomers in eq.([eq34 ] ) reduce to eq.([eq09 ] ) with the coefficients @xmath82 and @xmath87 given as @xmath115 and with @xmath116 .
as the reaction proceeds , the whole achiral substrate is ultimately transformed to chiral products so that @xmath117 .
this represents a quadratic curve of fixed points in @xmath44 phase space .
when there is solely a homodimer catalytic effect @xmath118 without spontaneous production nor heterodimer racemization @xmath119 , the initial state with @xmath120 follows the hyperbolic flow trajectory ( [ eq24 ] ) and ends up on the curve of fixed points , @xmath121 , as shown in fig.[fig4](a ) .
the ratio @xmath122 increases along the flow trajectory if it is initially larger than unity .
thus the @xmath44 asymmetry increases , and the final value of the ee @xmath123 is larger than the initial value @xmath124 . figure [ fig4](b ) shows @xmath125 relation for cases of positive @xmath95 .
the enhancement is more evident when the initial ratio @xmath126 is smaller .
these essential features are the same with the monomeric system . and @xmath127 .
flow terminates on a curve of fixed points @xmath121 .
( b ) the final ee value @xmath128 as a function of its initial value @xmath129 for various amount of initial chiral species @xmath130 .
the less @xmath130 is , the more amplified @xmath128 is . ]
another model for the chiral asymmetry amplification is a variant of the frank model in a closed system @xcite . in its simplest form ,
the model consists of reaction schemes of a spontaneous production ( [ eq01 ] ) , a linear autocatalysis ( [ eq02 ] ) and a heterodimer formation ( [ eq31 ] ) .
then the rate equations for monomers are the same with those of the original frank model @xcite , but they are supplemented with the heterodimer formation @xmath131}{dt } & = \nu_{\mathrm{het } } rs \label{eq39}\end{aligned}\ ] ] .
red lines represent fixed lines
. initial amount of achiral substrate is @xmath132 , and the remaining @xmath133 is shared by @xmath0 and @xmath1 monomers .
there are no heterodimers initially .
rate coefficients are @xmath134 , @xmath135 , @xmath136 .
( b ) final ee @xmath128 versus its initial value @xmath129 for a given initial ratio @xmath126 . as @xmath126 decreases , the ee amplification increases . ] since every achiral substrate is consumed up eventually @xmath30 and all the reactions stop asymptotically , eq.([eq39 ] ) tells us that the product @xmath137 should vanish .
if there is more @xmath0 than @xmath1 initially , @xmath1 monomer disappears ultimately , for instance .
but @xmath1 molecules do not disappear nor decomposed back into achiral substrate .
they are only incorporated into the heterodimer @xmath100 .
the system is not determined solely by monomer concentrations @xmath14 and @xmath15 , ( or @xmath33 and @xmath77 ) but it depends on heterodimer concentration @xmath108 $ ] as well .
the flow takes place in a three - dimensional phase space of @xmath138 $ ] , as shown in fig.[fig5](a ) .
the asymptotics discussed above forces both the concentrations of the substrate @xmath139 $ ] and the product @xmath137 vanish ultimately .
these two conditions define a line of fixed points in the three - dimensional @xmath140 $ ] phase space .
if the initial state has a prejudice to @xmath0 enantiomer such as @xmath141 , then the system ends up on a fixed line @xmath142=c$ ] on a @xmath143 plane , as shown in fig.[fig5](a ) .
otherwise with @xmath144 , the system flows to another fixed line @xmath145=c$ ] on a @xmath146 plane .
the plane @xmath35 is a dividing surface that determines the final preference of chirality .
one can readily find that ee is amplified by the reaction involving antagonistic heterodimer , as shown in fig.[fig5](b ) .
for a given amount of initial total concentration of chiral species @xmath147 , the ee value @xmath95 increases by the completion of a reaction .
no heterodimer is assumed to exist initially , @xmath108_0=0 $ ] .
the amplification is more significant for smaller amount of @xmath126 and @xmath129 .
we note that the rate constant of heterodimerization @xmath101 has to be large in order to produce strong ee amplification .
minority enantiomer should be quickly incorporated into heterodimers before the achiral substrate is consumed up .
otherwise , both enantiomers consume up the achiral resource @xmath3 and slowly forms heterodimers afterwards but without significant ee amplification . for a large @xmath101
, minority enentiomers are incorporated into heterodimers in an early stage . therefore , flow trajectories first point upwards in fig.[fig5](a ) , indicating the rapid increase of heterodimer concentration @xmath108 $ ] .
the trajectory is in fact inclined such that the minority enantiomer decreases its monomeric concentration .
when it almost disappears , then the trajectory makes a quick turn in the vicinity of @xmath146 plane or @xmath143 plane , indicating the concentration growth of the surviving majority enantiomer .
versus the velocity @xmath148 for three typical cases : ( a ) @xmath149 , ( b ) @xmath150 , and ( c ) @xmath151 . ]
so far , we have discussed several models for the ee amplification . there , the irreversibility of the dynamical system is essential to make the final state dependent on the initial condition . in this section ,
we consider the effect induced by introducing reversibility in the system by including various types of recycling processes . for simplicity ,
we consider in terms of a monomeric system .
in addition to the catalytic production processes ( [ eq01 ] ) to ( [ eq05 ] ) , we include the recycling process such that enantiomeric products @xmath0 or @xmath1 undergo a back reaction to the substrate @xmath3 @xcite with a reaction rate @xmath152 as @xmath153 we call this reaction a linear recycling hereafter .
one can also assume a nonlinear type of recycling process @xcite such that different chiral species by their encounter react back to achiral molecules as @xmath154 this back reaction can be a combined effect of the following two processes @xmath155 the rate equations now read as @xmath21a - \mu rs - \lambda r \nonumber \\
\frac{d s}{dt } = [ f(s)+k_1'r+k_2'rs]a - \mu rs - \lambda s .
\label{eq42}\end{aligned}\ ] ] the term with @xmath156 has a same form with the contribution of frank s mutual inhibition @xcite . the corresponding time - evolution of @xmath33 is readily derived in the form of eq.([eq13 ] ) with coefficients @xmath3 and @xmath4 determined by symmetric quantities @xmath77 and @xmath157 as @xmath158q_1 .
\label{eq43}\end{aligned}\ ] ] the evolution eq.([eq13 ] ) of the order parameter has a similar form with the time - dependent landau equation @xcite which is fundamental in nonequilibrium phase transitions .
the asymptotic value of the order parameter @xmath159 is determined as the zero of the velocity @xmath160 .
the main difference from the standard model of phase transitions lies in the time - dependence in the coefficients @xmath161 and @xmath162 induced by that of the achiral concentration @xmath163 and the total chiral concentration @xmath164 . because the concentrations @xmath16 and @xmath77 are non - negative , @xmath161 can not exceed @xmath162 ; @xmath165 . in previous sections
we have focused our studies on the cases with @xmath166 . in these cases , the asymptotic value of @xmath16 vanishes , @xmath30 so that @xmath167 , and the asymptotic value of the order parameter @xmath33 can not be determined from eq.([eq13 ] ) .
another example with @xmath168 happens for a special case with @xmath169 and @xmath170 . in this case , we can calculate trajectories and find a fixed line shifted from @xmath25 to the one with a finite value of @xmath171 @xcite : flow trajectories are along lines passing through the origin , similar to those shown in fig.[fig1](b ) , but terminating at a shifted fixed line . since @xmath25 is no more a fixed line , trajectories not only point out from the origin , but also come out from the line @xmath25 .
in all the other cases , with either a linear or a nonlinear recycling process or with both , the coefficients @xmath3 and @xmath4 are no longer zero at the same time , and a definite value of the order parameter @xmath33 is obtained asymptotically .
the reason for the above is as follows .
if the nonlinear recycling exists as @xmath172 , @xmath4 becomes nonzero since @xmath173 .
if the linear recycling exists as @xmath174 , not all the achiral substrate transform to chiral products but a finite amount remains asymptotically as @xmath175 .
therefore , nonzero values of @xmath8 , @xmath11 or @xmath176 give contributions to the coefficients @xmath3 or @xmath4 . for nonzero @xmath3 and/or @xmath4 , the velocity @xmath177 has three typical behaviors as a function of the order parameter @xmath33 , as shown in fig.[fig6 ] . in the region with a positive ( negative ) velocity @xmath160 , @xmath33 increases ( decreases ) and moves to the right ( left ) , as indicated by arrows on the @xmath33 axis in fig.[fig6 ] .
when @xmath3 is negative , as in fig.[fig6](a ) or ( b ) , @xmath33 eventually approaches the racemic value @xmath178 within its range of definition @xmath179 : the racemic state is a stable fixed point for @xmath180
. equation ( [ eq43 ] ) tells us that this happens when the spontaneous production @xmath8 or error in a linear autocatalytic process @xmath11 are sufficiently strong , or quadratic catalysis @xmath82 is weaker than the cross catalysis @xmath87 .
the coefficient of the linear autocatalysis @xmath9 as well as the linear recycling process @xmath152 are absent in the coefficients @xmath3 and @xmath4 , and can not affect directly the chirality of the system .
these linear processes with @xmath9 and @xmath152 affect the chirality only implicitly through @xmath16 and @xmath77 .
when @xmath3 is positive , as in the case of fig.[fig6](c ) , the coefficient of the cubic term @xmath4 is also positive , and the velocity @xmath181 vanishes at three values of @xmath33 in the range of @xmath182 .
this is possible if a strong quadratic autocatalysis @xmath183 exists together with a linear recycling @xmath174 , or if a linear autocatalysis @xmath9 and a nonlinear recycling @xmath184 coexist . by following the direction indicated by the arrows for positive @xmath33
, the order parameter ends up at a definite value @xmath185 and the chiral symmetry breaking takes place . if the system starts with a negative @xmath33 , it stops at @xmath186 .
the final state of the system depends on the initial sign of @xmath51 , but independent of its magnitude @xmath187 .
we note that the asymptotic values of the coefficients @xmath188 and @xmath189 are independent of the initial condition and so is @xmath190 .
this situation is what we mean the `` chiral symmetry breaking '' in the sense of statistical mechanics .
we first discuss the chiral symmetry breaking when there is a nonlinear recycling , @xmath184 @xcite .
the rate eq.([eq42 ] ) tells us that @xmath191 is no more a fixed line , but we have many fixed points .
the simplest is the case with only a linear autocatalysis @xmath192 but nothing else ( @xmath193 ) . in this case
we have four fixed points in the @xmath44 phase space : two unstable ones , o and u , and two stable ones at @xmath194 or @xmath195 , as is shown in fig.[fig7](a ) . at the stable fixed points the final ee take values @xmath196 .
these values agree with those expected from eq.([eq44 ] ) with @xmath197 as determined from eq.([eq43 ] ) .
the nonlinear recycling ( 42 ) returns the same amount of enantiomers back to the achiral substrate , but the relative decrement in concentration is larger for the minority enantiomer than for the majority one .
the damage is further amplified by a large reduction of linear autocatalytic effect for the minority enantiomer , and it disappears at last . ) similar to the frank model , and ( b ) a quadratic autocatalysis and a linear recycling ( @xmath198 ) . ] as for the second example , we consider the case with a quadratic autocatalysis @xmath199 and a linear recycling @xmath174 . because the linear recycling takes place whenever there is nonzero concentrations of enantiomers
, a few achiral substrate always remain , @xmath200 .
thus the diagonal line @xmath201 is no more a fixed line .
instead there appear fixed points in general . in the present case with a finite @xmath174 ,
totally seven fixed points appear ; three stable ( o , s@xmath202 ) and four unstable ones ( u@xmath203 ) , as shown in fig.[fig7 ] . as reaction proceeds ,
the system approaches a state associated with one of the stable fixed points .
the origin o is not interesting , since all the chiral products are recycled back to the achiral substrate .
also its influence extends only in a small region around o , bounded by u@xmath204 , if the life time @xmath205 of chiral products is long enough .
close to this fixed point , influence of the spontaneous production process with @xmath8 should be taken into account , anyhow . when the initial system is sufficiently away from o , the system approaches s@xmath206 or s@xmath207 ; both correspond to homochiral states with @xmath196 . when the initial configuration is close to the racemic state or the diagonal line @xmath35 , the system approaches a racemic fixed point u@xmath208 at first .
but , while the recycling process returns the chiral enantiomers back to achiral substrate , the majority enantiomer increases its population at the cost of the minority one along the flow curve diverging from u@xmath208 to s@xmath206 or to s@xmath207 , and eventually the whole system becomes homochiral .
one may wonder whether recycling processes such as a linear or a nonlinear back reaction exist in relevant autocatalytic systems .
so far , we are not aware of their existence .
it is , however , possible that the back reaction rates @xmath152 or @xmath156 are nonzero but exceedingly small to be detected in laboratory experiments . concerning with the problem of homochirality in life , very small
@xmath152 or @xmath156 are not unimaginable , considering the geological time scale for its establishment on earth .
on the other hand , a possibility to provide a system with a recycling process is proposed theoretically @xcite : one simply let the reaction with ee amplification run in an open flow reactor . in an open system there is reactant flows between the system and the environment @xmath209 .
let the solution with the achiral substrate @xmath3 be supplied by an inflow with an influx @xmath210 as @xmath211 and the products @xmath0 and @xmath1 as well as the substrate @xmath3 be taken out by an outflow with a flow out rate @xmath152 as @xmath212 in terms of rate equations the processes are described as @xmath213a+f-\lambda a .
\label{eq47}\end{aligned}\ ] ] since the total concentration @xmath214 follows the time evolution @xmath215 , it approaches the steady state value @xmath216 with a relaxation time @xmath205 .
this is a consequence of unbiased outflow ( [ eq47 ] ) of all reactants with the same rate @xmath152 .
consequently , even though we are dealing with an open system under a flow , the analysis is similar to the closed system by replacing the total concentration @xmath81 with the steady state value @xmath216 . instead of recycling
, therefore , constant supply of the substrate allows the system to reach a certain fixed point with a definite value of the order parameter @xmath33 , independent of the initial condition .
we have presented various simple scenarios that we are aware of in relevance to the ee amplification of the soai reaction ; a quadratic autocatalytic model in a monomer or a homodimer system , or a linear autocatalytic model in an antagonistic heterodimer system .
all these models can realize ee amplification such that the final value of the ee @xmath123 depends on but is larger than the initial value @xmath124 , as schematically shown in fig.[fig8 ] .
the curve in the figure represents @xmath217 relation for a given initial ratio @xmath126 , namely the ratio of the amount @xmath130 of the total chiral initiators @xmath0 and @xmath1 relative to that of the total reactants @xmath81 . amplification is more enhanced if the ratio of the chiral initiator @xmath126 is smaller .
this plot also shows the possibility to increase the final ee by repeating the reaction . to achieve high ee values , soai et al .
performed experiments of consecutive autocatalytic reactions , where after each run of the reactions , resulting solutions are quenched by adding acid and purified and reused in the next reactions .
how the high ee value is attained in this type of experimental procedures can be explained illustratively in fig.8 . by choosing a specified initial condition @xmath218
, the autocatalytic reaction ends up giving a final ee as @xmath219 determined by the intersection between the vertical line starting from @xmath218 point and the round curve .
the consecutive reaction starts with the purified products with the ee @xmath220 as the initial ee @xmath221 .
this feature is illustrated by extending a horizontal line from @xmath222 on the round curve to the @xmath223 straight slope .
this second reaction yields the ee value @xmath224 , which is determined by the intersection between the vertical line starting from @xmath225 point and the round curve . to repeat these procedures , the ee approaches rapidly the perfect @xmath226 with the speed depending on the curvature of the curve , which is determined by @xmath126
this is a mechanism of ee amplification and consecutive autocatalytic reactions , in contrast with the chiral symmetry breaking in the sense of statistical mechanics , where the final ee is unique and is reached by a single procedure . in all the cases considered
, autocatalytic processes must be present , whether it be linear or nonlinear . to understand the actual mechanism of autocatalysis for the soai reaction , the identification of the process at a molecular level is necessary , but that is out of scope of the present review .
we would like to point out another interesting feature , namely the spatial aspect of the reaction @xcite . in usual experiments , the reactor is stirred to keep the system homogeneous .
if there is no stirring , inhomogeneous patterns may develop between domains of two enantiomers .
analysis of these patterns may give some insight into the mechanism of reaction dynamics .
let us imagine , for instance , a time evolution of patterns of chiral domains for diffusionless systems with ee amplification and with a chiral symmetry breaking .
in an initial stage chiral domains are nucleated randomly , and the domain pattern may look similar for both systems .
after domains come in touch with each other , they behave differently : for a system with ee amplification , domain pattern freezes because there remains no achiral substrate for further reactions . on the other hand , for a system with the chiral symmetry breaking ,
neighboring domains compete with each other at domain boundaries to extend their own territories , and the domain pattern evolves to selection slowly . when diffusion or hydrodynamic flow effects are taken into account , the system with ee amplification simply homogenizes and the ee value shall be averaged out . for the system with the chiral symmetry breaking , these transport processes speed up the chirality selection and enhance the final value of ee @xcite .
so far we have discussed the ee amplification by starting from an initial state with a finite enantiomeric imbalance , and the effect of spontaneous reaction is mostly neglected , @xmath134 . however , reaction experiments without adding chiral substances do produce chiral species @xcite , implying that the coefficient @xmath8 of the spontaneous reaction should be nonzero . in these experiments , it is found that values of the ee order parameter @xmath95 of many runs are distributed widely between @xmath43 and @xmath42 , and the probability distribution function has double peaks at positive and negative values of @xmath95 .
there is a theoretical study on the asymptotic shape of probability distribution for nonautocatalytic and linearly autocatalytic systems with a specific initial condition of no chiral enantiomers @xcite .
even though no ee amplification is expected in these cases , the probability distribution with a linear autocatalysis has symmetric double peaks at @xmath227 when @xmath8 is far smaller than @xmath9 ; @xmath228 with @xmath229 being the total number of all reactive chemical species , @xmath3 , @xmath0 and @xmath1
. this can be explained by the `` single - mother '' scenario for the realization of homochirality , as follows . from a completely achiral state ,
one of the chiral molecule , say @xmath0 , is produced spontaneously and randomly after an average time @xmath230 .
then , the second @xmath0 is produced by the autocatalytic process whereas for the production of the first @xmath1 molecule the spontaneous production is necessary .
the waiting time for the second @xmath0 is shorter than that for the first @xmath1 by a factor about @xmath231 .
if this factor is far smaller than @xmath232 , there is only a negligible chance for the production of @xmath1 enantiomer until all the achiral substrates turn into @xmath0 .
all the chiral products are the descendants of the first mother @xmath0 produced by the autocatalysis , and no second mother is born by the spontaneous production : this is a single - mother scenario .
it strongly reflects the stochastic character of the chemical reaction , and is not contained in the average description using rate equations .
further studies are necessary as for the stochastic aspects of the soai reaction @xcite . in relation to the origin of homochirality in life ,
an interesting question is whether the ee amplification is sufficient to achieve homochirality . from the experimental results of the soai reaction without chiral initiator ,
the magnitude of the final ee @xmath123 is distributed from a small value to a value close to unity .
it is rather difficult to imagine that a value close to unity has been accidentally selected on various places on earth in the prebiotic era .
it seems rather natural that chiral symmetry breaking took place and a unique value of @xmath233 has been finally selected . as has been discussed in the previous section , the flow in the open system can alter the system with ee amplification to the one with the chiral symmetry breaking .
so we may say that the system with ee amplification in an open flow is one possibility .
recently , there appears several theoretical proposals such as polymerization models @xcite or a polymerization - epimerization model @xcite to realize the chirality selection in polymer synthesis .
although we are still far away from satisfactory understanding of the origin of homochirality in life , the discovery of the autocatalytic reaction by soai gives us great impetus and a sense of reality to this problem . | theoretical models and rate equations relevant to the soai reaction are reviewed .
it is found that in a production of chiral molecules from an achiral substrate autocatalytic processes can induce either enantiomeric excess ( ee ) amplification or chiral symmetry breaking .
former terminology means that the final ee value is larger than the initial value but depends on this , whereas the latter means the selection of a unique value of the final ee , independent of the initial value .
the ee amplification takes place in an irreversible reaction such that all the substrate molecules are converted to chiral products and the reaction comes to a halt .
the chiral symmetry breaking is possible when recycling processes are incorporated .
reactions become reversible and the system relaxes slowly to a unique final state .
the difference between the two behaviors is apparent in the flow diagram in the phase space of chiral molecule concentrations .
the ee amplification takes place when the flow terminates on a line of fixed points ( or a fixed line ) , whereas symmetry breaking corresponds to the dissolution of the fixed line accompanied by the appearance of fixed points .
relevance of the soai reaction to the homochirality in life is also discussed . |
the @xmath0 factories - the belle detector taking data at the kekb collider at kek @xcite and the babar detector @xcite at the pep ii at slac - have in more than a decade of data taking outreached the initial expectations on the physics results .
they proved the validity of the cabibbo - kobayashi - maskawa model of the quark mixing and @xmath1 violation ( @xmath2 ) .
perhaps even more importantly , they pointed out few hints of discrepancies between the standard model ( sm ) predictions and the results of the measurements . facing the finalization of the data taking operations
the question thus arises about the future experiments in the field of heavy flavour physics , to experimentally verify the current hints of possible new particles and processes often addressed as the new physics ( np ) .
part of the answer are the planned super @xmath0 factories in japan and italy , that could perform a highly sensitive searches for np , complementary to the long expected ones at the large hadron collider .
the so called precision frontier represented by the two machines requires the achieved luminosities of the @xmath0 factories to be raised by @xmath3 . in the present paper
we summarize the plan and the status of the belle detector upgrade ( belle ii ) at the upgraded kekb ( super kekb ) @xmath4 collider . in the following section we first briefly discuss the necessary upgrade of the kekb accelerator . in sections [ sect03 - 1 ] to
[ sect03 - 3 ] we summarize the upgrade of the vital parts of the belle detector - the vertexing , the particle identification system and the electromagnetic calorimeter , respectively .
the upgrade is illustrated with examples of planned measurements that will greatly benefit from the improved collider and detector performance .
finally we draw short conclusions in sect .
[ sect04 ] .
the kekb accelerator is an asymmetric @xmath4 collider operating at and near the center of mass energy of 10.58 gev , corresponding to the mass of the @xmath5 resonance .
the asymmetry of the beams results in a lorentz boost factor of @xmath6 which enables the time dependent measurements in the system of @xmath0 mesons .
the history of the kekb luminosity is presented in fig . [ fig01 ] .
the highest luminosity ever reached in the accelerator ( @xmath7 @xmath8s@xmath9 ) is a result of the crab cavities installed in 2007 @xcite . the continuous injection scheme and a very stable operation made possible to collect data corresponding to the integrated luminosity of more than 1 ab@xmath9 .
the luminosity of the collider is governed by several factors .
the crucial ones for the upgrade of the kekb are denote the high energy electron and the low energy positron beam , her and ler , respectively . ] the beam currents ( @xmath10 ) , the vertical beta function at the interaction point ( @xmath11 ) and the beam - beam parameter @xmath12 . to start from the latter , the beam - beam parameter , @xmath13 , will remain almost unchanged at super kekb , @xmath14 .
the beta function , however , will be extremely reduced : @xmath15 5.9 mm/5.9 mm @xmath16 0.27 mm/0.41 mm . .
] the emittance will be reduced accordingly to match the current @xmath12 .
both beam currents will be also increased by roughly a factor of two . in terms of the @xmath4 bunches
the foreseen upgrade corresponds to the reduction of the current size in direction perpendicular to the beam direction from @xmath17 m , @xmath18 m to @xmath19 m , @xmath20 nm . to achieve the desired goal the main tasks during the upgrade will be the installation of longer bending radius in the ler , more arc cells in the her , re - design of the interaction region with the new final focusing quadrupoles closer to the interaction point , new beam pipe and a new damping ring ( see fig . [ fig02 ] ) .
the outstanding problems are a rather small dynamic aperture , larger touschek background and consequently a shorter lifetime of the beams , directly affecting the luminosity . to cope with these
, the upgrade includes an increased crossing angle of the two beams ( from 22 mrad to 83 mrad ) and a slightly smaller asymmetry of the beams ( from 3.6 gev/8 gev to 4 gev/7 gev ) .
the luminosity of the super kekb will reach @xmath21 @xmath8s@xmath9 . assuming the startup of the machine in 2014 , and a rather conservative increase of the starting luminosity to the design value , already in two years of data - taking the available data sample will correspond to 5 ab@xmath9 .
integrated luminosity of 50 ab@xmath9 is expected in 2020 . to illustrate the precision that could be achieved with such a large sample of @xmath0 meson decays we use the measurement of the lepton forward - backward asymmetry @xmath22 in @xmath23 decays .
this observable ( or even more so , the zero crossing - point of the @xmath24 , with @xmath25 ) is not very sensitive to the theoretical uncertainties arising from the unknown form factors @xcite . in fig .
[ fig03 ] the current belle measurement @xcite is compared to the expected sensitivity at belle ii with @xmath26 ab@xmath9 .
it can be seen that such a measurement will make possible a distinction among various models , for example the sm and the supergravity models with the reversed sign of the @xmath27 wilson coefficient . with 600 fb@xmath9 @xcite and 5 ab@xmath9 data ( expected accuracies shown at the sm prediction ) .
shaded regions correspond to the charmonium veto @xmath28 intervals.,scaledwidth=70.0% ]
a rough overview of the belle detector upgrade is sketched in fig .
[ fig04 ] . in the environment of the beams with luminosity of @xmath29 @xmath8s@xmath9 the detector will have to cope with an increased background ( 10 - 20 times compared to the present ) , which will be the cause of an increased occupancy and radiation damage .
the first level trigger rate is expected to increase from the current 0.5 khz to around 20 khz . for several detector components we nevertheless foresee an improved performance and a better overall hermiticity of the detector after the upgrade .
the task of vertexing will rely on two layers of depfet pixel detectors ( pxd ) and four layers of double sided silicon detectors ( ssvd ) .
the main tracking device , the central drift chamber ( cdc ) , will have a smaller cell size and an improved read - out system . the particle identification will be performed mainly by the time - of - propagation counter ( top ) in the barrel and the rich detector with aerogel radiator ( arich ) in the forward part . for the electromagnetic calorimeter ( ecl )
the electronics enabling a wave form sampling will be introduced , and some of the current csi crystals doped with tl are going to be replaced by the pure csi . the detector of muons and @xmath30 s ( klm ) will be upgraded with scintillator strips in the endcaps . a schematic view of the future semiconductor detector of belle ii is shown in fig . [ fig05](left ) and is composed of two layers of pixel detectors @xcite followed by four layers of double sided silicon strip detectors .
the improvement compared to the current detector is twofold : a better spatial resolution of the vertex determination ( for around 25% in the case of @xmath31 vertex ) , and an improved reconstruction efficiency of @xmath32 decays with pion signals in the detector , due to the increased radii of the layers .
the latter is important for the time dependent measurements of various decay modes with @xmath33 s in the final state ( increase of efficiency by around 30% ) .
since the dependence on the radius of the layers is opposite for the two mentioned improvements , a careful optimization of the design was performed . in @xmath34.,title="fig:",scaledwidth=50.0% ] in @xmath34.,title="fig:",scaledwidth=40.0% ] to illustrate the expected performance , a search for possible right - handed currents in @xmath34 is used as a benchmark mode . in
these decays only the @xmath33 direction is used together with the interaction point constraint to determine the @xmath0 meson decay vertex @xcite . while the indirect @xmath2 is heavily suppressed in the sm due to the helicity structure of the hamiltonian , it can be largely increased is some np models @xcite .
figure [ fig05](right ) shows a comparison of the current values of the direct and indirect @xmath2 parameters in this mode @xcite with the approximate expected precision including the statistical and systematic uncertainties .
the expected accuracies on @xmath35 with 5 ab@xmath9 and 50 ab@xmath9 are 0.09 and 0.03 , respectively .
while in the left - right symmetric models @xmath35 can be as high as 0.5 , the sensitivity with 50 ab@xmath9 of data is smaller than the sm predictions .
particle identification at belle ii will rely on the top @xcite counter in the barrel part , and arich detector in the forward @xcite .
the top detector will consist of quartz bars with mirrors on one side and microchannel plate photomultipliers on the other .
for high momentum ( 3 gev/@xmath36 ) kaons we expect around 10% better identification efficiency ( 90%-95% ) in the barrel at a similar misidentification probability as for the current detector ( 5% ) .
particle identification is of course crucial in several measurements , for example the measurements related to the so called direct @xmath2 puzzle , which arises from the observed difference between the direct @xmath2 asymmetry in @xmath37 and @xmath38 decays @xcite . while in the explicit calculations of the asymmetries @xmath39
several model uncertainties are present , a model independent sum rule was proposed @xcite to test the consistency of the sm .
it relates the asymmetries and the branching fractions of several decay modes : @xmath40[\tau_{b^0}/\tau_{b^+ } ] = a_{k^0\pi^0}[2{\cal{b}}_{k^0\pi^0}/{\cal{b}}_{k^+\pi^-}]+ a_{k^+\pi^0}[{2\cal{b}}_{k^+\pi^0}/{\cal{b}}_{k^+\pi^-}][\tau_{b^0}/\tau_{b^+}]$ ] .
figure [ fig06](left ) shows the current status of the measurements using the world average values of the measured observables @xcite , where @xmath41 is expressed as a function of the @xmath42 using the sum rule .
the predictions of the sum rule are in agreement with the direct measurements .
the main systematic uncertainty on the @xmath41 , which is the least precisely known asymmetry in the @xmath43 system , can be reduced wit a larger data sample , and hence the expected accuracy with 50 ab@xmath9 is 0.03 . assuming the current central values a discrepancy between the measurements and the sum rule prediction at @xmath44 ab@xmath9 would be significant ( fig .
[ fig06](right ) ) . and
@xmath42 @xcite compared to the sum rule @xcite prediction .
the width of the band is determined by the experimental uncertainties on other quantities entering the sum rule .
right : the same comparison assuming uncertainties expected with 50 ab@xmath9 of data.,scaledwidth=100.0% ] in the electromagnetic calorimeter ( ecl ) upgrade the replacement of the current electronics is foreseen .
the new one will enable amplitude - time measurements for the signals in the ecl and will thus help to suppress the background from clusters caused mainly by the off - time beam background ( we expect the reduction of this background by a factor of 7 ) . beside this , a partial replacement of the tl doped csi crystals with the radiation hard pure csi is being under the consideration . due to the increased rate of backgrounds at belle ii
the expected photon detection efficiency of the ecl is around 5%-10% lower , while keeping the background at the current level
. the importance of the ecl performance can be best illustrated by the measurement of the @xmath45 @xcite .
the method consists of full or partial reconstruction of the tagging @xmath0 meson , identification of hadrons or charged leptons from the @xmath46 decay , and examination of the distribution of the remaining measured energy in the event . for the signal , where the undetected particles are neutrinos , this distribution of energy measured in the ecl peaks at zero .
the leptonic @xmath0 meson decays are interesting since , for example , in the type ii two higgs doublet models , the sm branching fraction receives a contribution from the charged higgs boson exchange , expressed as a multiplicative factor : @xmath47 ^ 2 $ ] .
the expected sensitivity on @xmath48 with 50 ab@xmath9 of data is below 0.1 . with an increased statistical power of the data , and assuming the existing ecl performance ,
one obtains the five standard deviations discovery region for the charged higgs boson as shown in fig .
[ fig07 ] @xcite .
it can be seen that a large area of the @xmath49 plane ( compared to the current exclusion regions ) can be covered by this search , especially at larger values of the mass and @xmath50 . and @xmath51 values precision to @xmath523% each . ]
such a measurement is to some extent complementary to the measurements of the @xmath53 transition branching fraction which constrain the mass of the charged higgs boson almost independently of the @xmath50 value .
plane , from the measurement of @xmath45 with 50 ab@xmath9 @xcite .
other shaded regions show the current 95% c.l .
exclusion region.,scaledwidth=40.0% ]
in summary , we presented a short overview of the kekb accelerator and belle detector upgrade . while technologically most challenging , the preparation of the super @xmath0 factory at kek is well on the way .
the key features of the upgrade are illustrated by several measurements that will be possible at belle ii and will represent a highly sensitive search for np effects , complementary to searches at the lhc .
a comprehensive program of physics measurements can be found in @xcite .
the value of the super @xmath0 factory lies not only in a highly sensitive search of np in individual processes , but to a large extent in the possibility of performing measurements of various observables which through their correlations can help identifying the nature of np . as an example , fig .
[ fig08 ] @xcite shows correlations between the indirect @xmath2 in @xmath54 ( see sect .
[ sect03 - 1 ] ) and in the @xmath55 with the underlying penguin quark process @xmath56 ( a naive sm prediction is @xmath57 ) . in the minimal super gravity model ( msugra ) and supersymmetric grand unification theory with right - handed neutrinos ( susy su(5 ) ) the correlations between the two observables exhibit a different pattern , while the mass spectra of the particles predicted in the two models are similar .
if hints of new particles consistent with these predictions arise from the lhc , at belle ii one can distinguish the two models with already @xmath58 ab@xmath9 of data .
parameter in @xmath54 and @xmath55 decays @xcite for two models : the minimal super gravity model ( msugra ) and supersymmetric grand unification theory with right - handed neutrinos ( susy su(5 ) ) .
the points with error bars denote the expected sensitivity at belle ii with 5 ab@xmath9 of data .
, title="fig:",scaledwidth=45.0% ] parameter in @xmath54 and @xmath55 decays @xcite for two models : the minimal super gravity model ( msugra ) and supersymmetric grand unification theory with right - handed neutrinos ( susy su(5 ) ) .
the points with error bars denote the expected sensitivity at belle ii with 5 ab@xmath9 of data .
, title="fig:",scaledwidth=45.0% ] 0 a. abashian _
et al . _ ( belle coll . ) , nucl . instr
a479 * , 117 ( 2002 ) .
s. kurokawa _
et al . _ ,
. instr . meth .
* a499 * , 1 ( 2003 ) , and other papers in that volume . b. aubert _
et al . _
( babar coll . ) , nucl .
a479 * , 1 ( 2002 ) .
y. funakoshi , icfa beam dyn . newslett . * 40 * , 27 ( 2006 ) .
see for example g. burdman , phys .
rev . * d52 * , 6400 ( 1995 ) ; a. hovhannisyan , w .- s .
hou , n. mahajan , phys . rev . *
d77 * , 014016 ( 2008 ) . j
et al . _ ( belle coll . ) , phys . rev
* 103 * , 171801 ( 2009 ) .
see for example j. kemmer , g. lutz , nucl .
a253 * , 356 ( 1987 ) ; i. peric , nucl .
a579 * , 685 ( 2007 ) . y. ushiroda _
et al . _ ( belle coll . ) , phys . rev . *
d74 * , 111104 ( 2006 ) .
see for example d. atwood _ et al .
_ , phys . rev . * d71 * , 076003 ( 2005 ) ; b. grinstein _
et al . _ ,
phys . rev . * d71 * , 011504 ( 2005 ) .
e. barberio _
_ ( heavy flavor averaging group ) , arxiv:0808.1297 , and updates at http://www.slac.stanford.edu / xorg / hfag/. k. inami , nucl .
* a595 * , 96 ( 2008 ) . s. hashimoto _
et al . _ , kek - report 2004 - 4 . s .-
et al . _ ( belle coll . ) , nature * 452 * , 332 ( 2008 ) .
m. gronau , phys . lett . *
b627 * , 82 ( 2005 ) ; d. atwood , a. soni , phys . rev . *
d58 * , 036005 ( 1998 ) .
i. adachi _
et al . _ ( belle coll . ) , arxiv:0809.3834 .
a.g . akeroyd _
_ arxiv:1002.5012 . | we present the status of the kekb accelerator and the belle detector upgrade , along with several examples of physics measurements to be performed with belle ii at super kekb . |
we consider stochastic particle systems which are probabilistic models describing transport of a conserved quantity on discrete geometries or lattices .
many well known examples are introduced in @xcite , including zero - range processes and exclusion processes , which are both special cases of the more general family of misanthrope processes introduced in @xcite .
we focus on spatially homogeneous models with stationary product measures and without exclusion restriction , which can exhibit a condensation transition that has recently been studied intensively .
a condensation transition occurs when the particle density exceeds a critical value and the system phase separates into a fluid phase and a condensate .
the fluid phase is distributed according to the maximal invariant measure at the critical density , and the excess mass concentrates on a single lattice site , called the condensate .
most results on condensation so far focus on zero - range or more general misanthrope processes in thermodynamic limits , where the lattice size and the number of particles diverge simultaneously .
initial results are contained in @xcite , and for summaries of recent progress in the probability and theoretical physics literature see e.g. @xcite .
condensation has also been shown to occur for processes on finite lattices in the limit of infinite density , where the tails of the single site marginals of the stationary product measures behave like a power law @xcite . in general ,
condensation results from a heavy tail of the maximal invariant measure @xcite , and so far most studies focus on power law and stretched exponential tails @xcite . as a first result , we generalize the work in @xcite and provide a characterization of condensation on finite lattices in terms of the class of sub - exponential tails that has been well studied in the probabilistic literature @xcite . this characterization holds for a particular definition of condensation given in section [ sub sec : ips ] , which was also used in @xcite .
our main result is that all spatially homogeneous processes with stationary product measures that exhibit condensation on finite lattices with a finite critical density are necessarily non - monotone .
monotone ( attractive ) particle systems preserve the partial order on the state space in time , which enables the use of powerful coupling techniques to derive rigorous results on large scale dynamic properties such as hydrodynamic limits ( see @xcite and references therein ) .
these techniques have also been used to study the dynamics of condensation in attractive zero - range processes with spatially inhomogeneous rates @xcite , and more recently @xcite .
as we discuss in appendix [ sec : statmech ] , non - monotonicity in homogeneous systems with finite critical density can be related , on a heuristic level , to convexity properties of the canonical entropy . for condensing systems with zero - range dynamics
, it has been shown that this is related to the presence of metastable states , resulting in the non - monotone behaviour of the canonical stationary current / diffusivity @xcite .
this corresponds to a first order correction of a hydrodynamic limit leading to an ill - posed equation with negative diffusivity in the case of reversible dynamics .
heuristically , this is of course consistent with the concentration of mass in a small , vanishing volume fraction , but poses great technical difficulties to any rigorous proof of hydrodynamic limits for such particle systems .
first results in this direction only hold for sub - critical systems under restrictive conditions @xcite , and due to a lack of monotonicity there are no results for non - reversible dynamics .
condensing monotone particle systems would , therefore , provide interesting examples of homogeneous systems for which coupling techniques could be used to derive stronger results on hydrodynamic limits .
however , our result implies that this is not possible for condensing models with stationary product measures and a finite critical density on finite lattices . in the thermodynamic limit condensation
has been defined through the equivalence of ensembles , which can be established in generality for a class of long - tailed distributions with a finite critical density @xcite .
this class has also been studied before @xcite and includes the class of sub - exponential distributions , for which our results apply also in the thermodynamic limit . a detailed discussion of their connections and the resulting differences between condensation on finite lattices and in the thermodynamic limit is given in sections [ sec : subexp ] and [ tlimit ] .
we remark that for systems where the dynamics is directly defined on infinite lattices there are no rigorous results or characterizations of condensation to our knowledge , and we do not discuss this case here . for systems with infinite critical density condensation can still occur on finite lattices , and since non - monotonicity typically occurs above the critical density , such processes can also be monotone .
when the tail of the stationary measure is a power law and decays faster than @xmath0 with the occupation number @xmath1 , we prove that the process is still non - monotone . in section [ examples ]
we present preliminary results for tails that decay slower than @xmath0 , which strongly suggests that a monotone and condensing particle system exists ( see @xcite for further discussion ) .
the paper is organised as follows . in section [ sec : notation ] we introduce the background used to study condensation and monotonicity in particle systems , and state our main results . in section [ sec :
proof ] we prove our main theorem by induction over the size of the lattice , showing that the family of canonical stationary measures is necessarily not monotonically ordered in the number of particles . in section [ sec : char cond ] we discuss the differences between condensation on fixed lattices and in the thermodynamic limit , and prove equivalence of condensation on finite lattices with the tail of the maximal invariant product measure being sub - exponential . in section [ examples ]
we review examples of homogeneous processes that have been shown to exhibit condensation , and present some explicit computations for misanthrope processes and processes with power law tails .
we consider stochastic particle systems on fixed finite lattices @xmath2 , which are continuous - time markov chains on the countable state space @xmath3 . for a given configuration @xmath4
the local occupation , @xmath5 for @xmath6 , is a priori unbounded .
the jump rates from configuration @xmath7 to @xmath8 are denoted @xmath9 , and the dynamics of the process is defined by the generator @xmath10 for all continuous functions @xmath11 .
we assume the process conserves the total number of particles @xmath12 and conditioned on @xmath13 , the process is assumed to be irreducible , so that @xmath14 is the only conserved quantity .
the process is therefore ergodic on the finite state space @xmath15 for all fixed @xmath16 . on @xmath17
the process has a unique stationary distribution @xmath18 , and the family @xmath19 is called the canonical ensemble .
we focus on systems for which the stationary distributions are spatially homogeneous , the marginal distributions @xmath20 $ ] are identical for all @xmath21 .
this typically results from translation invariant dynamics on translation invariant lattices with periodic boundary conditions , but the actual details of the dynamics are not needed for our results . for these systems we define condensation in terms of the maximum occupation number @xmath22 [ condensation ] a stochastic particle system with canonical measures @xmath18 on @xmath17 with @xmath23 exhibits * condensation * if @xmath24 = 1 \ .
\label{eqn : cond1}\ ] ] this condition implies in particular the existence of all limits involved .
the interpretation of ( [ eqn : cond1 ] ) is that due to the sub - exponential tails of the measure , in the limit @xmath25 all but finitely many particles concentrate in a single lattice site and that the distribution of particles outside the maximum is non - degenerate .
as we will see in proposition [ prop1 ] below , the latter is in fact given by the maximal invariant measure when the system exhibits product stationary measures .
there are of course other possible definitions of condensation which are less restrictive or more appropriate in other situations . for inhomogeneous systems or thermodynamic limits ( with @xmath26 )
the condensed phase can be localized in particular sites or have a more complicated spatial structure ( see e.g. @xcite and for recent summaries @xcite ) . for our case of spatially homogeneous systems on finite lattices , @xmath27 with fixed @xmath28 ,
the above is the most convenient definition and has been used in previously studied examples @xcite .
a more detailed discussion is provided in section [ sec : char cond ] .
+ we use the natural partial order on the state space @xmath29 given by @xmath30 if and only if @xmath31 for all @xmath6 .
a function @xmath32 is said to be increasing if @xmath33 implies @xmath34 .
two measures @xmath35 on @xmath29 are stochastically ordered with @xmath36 , if for all increasing functions @xmath32 we have @xmath37 , where @xmath38 denotes the expectation of @xmath39 with respect to @xmath40 .
a stochastic particle system on @xmath29 with generator @xmath41 and semi - group @xmath42 is called monotone ( attractive ) if it preserves stochastic order in time , i.e. @xmath43 coupling techniques for monotone processes are an important tool to derive rigorous results on the large scale dynamics of such systems such as hydrodynamic limits .
there are sufficient conditions on the jump rates ( [ gene ] ) to ensure monotonicity for a large class of processes ( see e.g. @xcite for more details ) , however for our results we only need a simple consequence for the stationary measures of the process .
[ order ] if the stochastic particle system as defined in section [ sub sec : ips ] is monotone , then the canonical distributions @xmath18 are ordered in @xmath44 , i.e. @xmath45 the proof is completely standard but short , so we include it for completeness .
fix a monotone process as defined in section [ sub sec : ips ] .
consider two initial distributions @xmath40 and @xmath46 , concentrating on @xmath17 and @xmath47 respectively , given by @xmath48 = { \mathds{1}}(\eta_{1 } = n ) \quad \textrm{and}\quad \mu'[\xi ] = { \mathds{1}}(\xi_1 = n+1)\ , \ ] ] for @xmath49 and @xmath50 .
clearly @xmath51 , and so by monotonicity of the process this implies @xmath52 for all @xmath53 .
furthermore , by ergodicity we have @xmath54 all rigorous results on condensing particle systems so far have been achieved for processes which exhibit stationary product measures , for which the measures @xmath18 take a simple factorized form .
these can then be expressed in terms of un - normalized single - site weights @xmath55 , @xmath56 .
due to conservation of @xmath14 such processes exhibit a whole family of stationary homogeneous product measures @xmath57=\prod_{x\in\lambda }
\nu_\phi [ \eta_x ] \quad\mbox{with marginals}\quad \nu_\phi [ \eta_x ] = \frac{w(\eta_x ) } { z(\phi ) } \,\phi^{\eta_x}\ .
\label{eq : pm}\ ] ]
the measures are defined whenever the normalization @xmath58 is finite .
this is the case for all fugacity parameters @xmath59 where @xmath60 or @xmath61 $ ] , and @xmath62{w(n)}\big)^{-1}\ ] ] is the radius of convergence of ( [ zphi ] ) .
the family @xmath63 is also called the grand - canonical ensemble and @xmath64 the ( grand - canonical ) partition function . since the process is irreducible on @xmath17 for all @xmath65 we have @xmath66 for all @xmath67 .
the canonical distribution can be written as @xmath68=\nu^{l}_\phi [ \eta |s_l = n]\quad\mbox{for all } \phi\in d_\phi \ , \ ] ] which is independent of the choice of @xmath69 .
equivalently @xmath70 = \frac{1}{z_{l , n } } \prod_{x\in\lambda } w(\eta_{x})\quad\mbox{where}\quad z_{l , n } = \sum_{\eta\in \omega_{l , n } } \prod_{x\in\lambda } w(\eta_{x } ) \label{eqn : prod measure}\end{aligned}\ ] ] is the ( canonical ) partition function .
note that throughout the paper we characterize all measures by their mass functions since we work only on a countable state space @xmath29 and the measures @xmath18 concentrate on finite state spaces @xmath17 , which is the framework we rely on in this paper .
+ our results hold for systems with general stationary weights , @xmath66 for each @xmath71 , subject to the regularity assumption that @xmath72 \label{regu}\ ] ] exists , which is then necessarily equal to @xmath73 .
if @xmath74 then weights that satisfy are sometimes called long - tailed @xcite , which is discussed in more detail in section [ tlimit ] . for processes with such stationary product measures
there is a simple equivalent characterization of condensation which we prove in section [ proof equiv ] . [ prop1 ]
consider a stochastic particle system as defined in section [ sub sec : ips ] with stationary product measures as defined in section [ sec : mon ] satisfying regularity assumption .
then the process exhibits condensation according to definition [ condensation ] if and only if @xmath75 , @xmath76 $ ] , and @xmath77}{\nu_{\phi_c } [ \eta_1 = n ] } = \lim_{n\to\infty } \frac{z_{2,n}}{w(n)z(\phi_c ) } \in ( 0,\infty ) \quad\mbox{exists}\ . \label{paff}\ ] ] in this case , the distribution of particles outside of the maximum converges weakly ( equivalently in total variation ) to the critical product measure @xmath78 , _
i.e. _ for fixed @xmath79 we have @xmath80 \to \prod_{i=1}^{l-1}\nu_{\phi_{c}}[\eta_{i } = n_i ] \quad \textrm{as}\quad n \to \infty \ .\end{aligned}\ ] ] note that for @xmath81 we may rescale the exponential part of the weights to get @xmath82 and we can further multiply with a constant , so that in the following we can assume without loss of generality that @xmath83 the condition ( [ paff ] ) can also be written as @xmath84 where @xmath85 is the convolution product .
this is a standard characterization to define the class of sub - exponential distributions ( see e.g. @xcite ) .
sub - exponentiality implies that a large sum of two random variables is typically realized by one of the variables taking a large value ( see section [ sec : subexp ] for more details ) , which is of course reminiscent for the concept of condensation .
implications and simpler necessary conditions on @xmath86 which imply ( [ paff2 ] ) have been studied in detail , and we provide a short discussion in section [ sec : char cond ] . as a consequence of proposition [ prop1 ] ,
condensation is only a property of the tail of @xmath87 , therefore if a process with stationary product measures condenses in the sense of definition [ condensation ] for some @xmath23 , it condenses for all @xmath88 .
proposition [ prop1 ] provides a generalization of previous results on condensation on finite lattices @xcite and is used in the proof of our main result , which is the following . [ mainres ] consider a spatially homogeneous stochastic particle system as defined in section [ sub sec : ips ] which exhibits condensation in the sense of definition [ condensation ] , has stationary product measures that satisfy , and has finite critical density @xmath89 then the process is necessarily not monotone .
+ the same is true if ( [ fmean ] ) is replaced by the assumption that we use the notation @xmath90 if @xmath91 as @xmath92 . ]
@xmath93 with @xmath94 $ ] .
the class of distributions which fulfil ( called sub - expontial ) , and therefore exhibit condensation on finite lattices , is large ( see e.g. ( * ? ? ?
* table 3.7 ) ) , and includes in particular * power law tails @xmath95 where @xmath96 , * log - normal distribution @xmath97 where @xmath98 and @xmath99 , which always has finite mean , * stretched exponential tails @xmath100 for @xmath101 , @xmath102 , * almost exponential tails @xmath103 for @xmath104 . for the last two examples all polynomial moments are finite .
this covers all previously studied models on condensation in zero - range processes @xcite . as we will discuss in section [ tlimit ] all these examples also exhibit condensation in the thermodynamic limit , whenever they have finite first moment .
it can also be shown that the limit in ( [ paff2 ] ) is necessarily equal to @xmath105 and that in fact @xmath106 for any fixed @xmath23 ( see @xcite and proposition [ 2tol ] ) .
since we consider a fixed lattice @xmath27 , @xmath107 is not a necessary condition for condensation as opposed to systems in the thermodynamic limit .
even if the distribution of particles outside the maximum has infinite mean , condensation in the sense of definition [ condensation ] can occur . however , if @xmath108 ( e.g. for power law tails with @xmath109 ) , the distribution of particles outside the maximum can not be normalized , condition fails , and there is no condensation in the sense of our definition .
0.49 0.5 0.495 [ back : almostexp ] we will prove non - monotonicity in the next section by showing that expectations for a particular monotone decreasing observable @xmath110 under @xmath18 are not decreasing in @xmath44 .
the chosen function is related to ( but not equal to ) the number of particles outside the maximum ( condensate ) , which has been shown previously to exhibit non - monotone behaviour for a class of condensing zero - range processes in the thermodynamic limit @xcite .
when the number of particles @xmath111 just exceeds the critical value , typical configurations still appear homogeneous with a maximum occupation number of we use the notation @xmath112 if @xmath113 as @xmath114 . ] @xmath115 . only when the number of particles is increased further the system switches to a condensed state with a maximum that contains a non - zero fraction of all particles .
we present numerical evidence of this non - monotone switching behaviour for the background density @xmath116 in figure [ backnumerics ] .
this is a finite size effect which disappears in the limit @xmath117 , and for specific models it has been shown to be related to the existence of super - critical homogeneous metastable states @xcite . for large @xmath28 , the switching to condensed states occurs abruptly over a relatively small range of values for @xmath44 . since the @xmath18 are conditional product measures the correlations in the system are very weak , which causes metastable hysteresis effects and non - monotonicity of the canonical ensemble around the critical point .
metastable hysteresis has been established in @xcite for zero - range processes .
our result implies that this behaviour is generic for all condensing systems with finite critical density .
we also give a heuristic discussion of the connection to convexity properties of the entropy of the system in appendix [ sec : statmech ] .
there are several examples of homogeneous , condensing , monotone particle systems with finite critical density which have been studied on a heuristic level and which we summarize in section [ examples ] .
their stationary measures are not of product form and no explicit formulas are known , so these systems are therefore hard to analyse rigorously . for systems with non - product stationary measures , upward fluctuations in the density which are homogeneously distributed may be suppressed strongly enough , so that the metastable states do not exist .
such models may then also be monotone , and examples are given in section [ homnonpro ] .
we excluded the case @xmath118 in the presentation in section [ sec : resu ] for notational convenience , but it is easy to see that our results also hold in this case . with the convention @xmath119
we have @xmath120 and @xmath121 , and then existence of the limit
@xmath122 is equivalent to @xmath123=2\frac{w(n)w(0)}{z_{2,n } } \to \frac{2w(0)}{2z(0)}=1\ , \textrm { as } n \to \infty , \ ] ] i.e. condensation of all @xmath44 particles on a single site .
this can easily be extended to all @xmath23 with proposition [ 2tol ] .
considering only events with all @xmath44 particles on one site , or @xmath124 particles on one site and @xmath125 particle elsewhere , we have convergence from above @xmath126 this implies the non - monotonicity of @xmath18 as discussed in section [ sec : proof ] .
we assume that the process exhibits condensation in the sense of definition [ condensation ] and has stationary product measures , so the canonical measures @xmath127 are of the form .
furthermore , we assume the weights satisfy the regularity assumption , and without loss of generality @xmath128 , see .
we show that the family of canonical measures is not stochastically ordered in @xmath44 , which implies non - monotonicity of the process by lemma [ order ] . to achieve this
, we use the test function @xmath129 which indicates the event where all particles concentrate in the maximum at site @xmath28 . [
statement : monotonicity ] the function @xmath110 defined in ( [ eqn : dec observable ] ) is monotonically decreasing , which implies that @xmath130 whenever the family of canonical measures @xmath18 is stochastically ordered in @xmath44 .
fix configurations @xmath131 such that @xmath30 . if @xmath132 then @xmath133 has at least one particle outside of site @xmath28 , therefore so does @xmath134 which implies @xmath135 . if @xmath136 then necessarily @xmath137 since @xmath138 .
therefore @xmath39 is a decreasing function . using and the convention ( [ wlog ] )
, we find that the canonical expectation of the function is given by @xmath139 so if the canonical measures are monotone in @xmath44 , monotonicity of @xmath39 implies ( [ eq : zmono ] ) .
0.49 0.5 by proposition [ prop1 ] we know that for condensing systems the ratio @xmath140 converges .
this implies the sequence @xmath141 in lemma [ statement : monotonicity ] converges , as summarised in the following proposition using our notation ( see ( * ? ? ?
* theorem 1 and lemma 5 ) ) .
[ 2tol ] consider conditional product measures ( [ eqn : prod measure ] ) with weights @xmath66 for all @xmath71 , which satisfy * @xmath142 as @xmath114 , the regularity assumption ( [ wlog ] ) , + * @xmath143 , + * @xmath144 as @xmath145 . then @xmath146 and furthermore
, @xmath147 note that the limit in states that the probability of observing a large total number of particles under the critical product measure is asymptotically equivalent to the probability of observing a large number of particles on any one of the @xmath28 sites , precisely @xmath148}{l\,\nu_{\phi_c}[\h_1 = n ] } = 1\ , .
\end{aligned}\ ] ] this is further equivalent to the canonical probability of the maximum containing the total mass converges to the critical probability that @xmath149 sites are empty , i.e. @xmath150 \to \nu_{\phi_c}^{l-1 } [ \h \equiv 0 ] $ ] . to complete the proof
we show that a subsequence of @xmath141 converges from above , which contradicts the assumption of monotonicity by lemma [ statement : monotonicity ] .
we present a numerical illustration for the monotonicity properties of the function @xmath151 in figure [ testnumerics ] , which is normalized such that @xmath152 as @xmath25 .
the proof of the following lemma represents the most significant part of the proof of theorem [ mainres ] and is given in section [ sec : finite - mean ] for the case of finite mean and in section [ sec : power - law ] for the power law case .
[ res2 ] under the conditions of theorem [ mainres ] , and assuming without loss of generality @xmath153 , for each @xmath88 there exists a @xmath154 and a sequence @xmath155 with @xmath156 as @xmath157 such that @xmath158 therefore , we know that there exists some @xmath159 such that @xmath160 , which by lemma [ statement : monotonicity ] implies that the canonical measures are not stochastically ordered in @xmath44 , and thus the process can not be monotone by lemma [ order ] .
this completes the proof of theorem [ mainres ] . in order to prove that @xmath161 converges from above for some non - decreasing sequence @xmath162
we first specify a sequence on which we can bound the ratio @xmath163 below .
[ statement : ratio upper bound ] for weights @xmath164 with finite and non - zero first moment , _ i.e. _ @xmath165 , there exists a sequence @xmath166 with @xmath167 as @xmath168 such that for all @xmath169 @xmath170 for each @xmath171 , define @xmath172 as follows @xmath173 by definition @xmath162 is a non - decreasing sequence .
assume for contradiction that @xmath162 is bounded above , then for all @xmath174 we would have @xmath175 for some @xmath176 , and therefore @xmath177 contradicting the assumption of finite first moment .
for @xmath178 we have @xmath179 [ statement : 2 site limit above ] for weights @xmath164 with finite first moment , there exists a subsequence @xmath180 of the sequence defined in claim [ statement : ratio upper bound ] such that @xmath181 for all @xmath182 and @xmath183 sufficiently large . by neglecting at most a single term in the sum defining @xmath184
, the ratio @xmath185 can be bounded below as follows , @xmath186 we define @xmath187 to be the largest index where the ratio @xmath188 is minimized
. in particular @xmath189 by definition @xmath190 , and so @xmath191 .
there exists a subsequence @xmath192 such that @xmath193 , with a slight abuse of notation we denote the subsequences @xmath194 and @xmath195 simply by @xmath196 and @xmath197 .
suppose @xmath198 , by claim [ statement : ratio upper bound ] we have @xmath199 which together with contradicts proposition [ 2tol ] , therefore @xmath200 and @xmath201 for all @xmath183 .
applying claim [ statement : ratio upper bound ] we then have @xmath202 subtracting @xmath203 we get @xmath204 neglecting the final term in we have @xmath205 as @xmath206 , where @xmath207 is the critical density defined in . together with this completes the proof of claim [ statement : 2 site limit above ] . to complete the proof of lemma [ res2 ] we proceed by induction on the system size , @xmath28 .
we make the following inductive hypothesis ; 1 . [ h1 ] there exists a sequence @xmath208 such that @xmath209 for all @xmath210 and @xmath211 sufficiently large .
the case @xmath212 is given by claim [ statement : 2 site limit above ] .
analogously to the proof of claim [ statement : 2 site limit above ] we define @xmath213 by the same argument as in the proof of claim [ statement : 2 site limit above ] there exists a subsequence @xmath192 such that @xmath214 , again we denote the respective subsequences by @xmath197 and @xmath196 . for @xmath183
sufficiently large , we have @xmath215 where the final inequality follows from the inductive hypothesis ( h ) and claim [ statement : ratio upper bound ] .
subtracting @xmath216 we get @xmath217 now , following the proof of claim [ statement : 2 site limit above ] , multiply ( [ arghh ] ) by @xmath218 and neglect the second term on the second line .
then the first term vanishes , since @xmath219 in terms of the normalized grand - canonical measure @xmath220 $ ] , so we have @xmath221 where @xmath207 is the critical density as defined in ( [ fmean ] ) .
this implies that the first term in the second line of ( [ arghh ] ) , after multiplication with @xmath196 , converges to a strictly positive constant .
finally , the third line in ( [ arghh ] ) converges to zero after multiplying by @xmath196 since we have @xmath222 , which implies @xmath223 as @xmath206 , by . using the definition of @xmath197 in ( [ kmdef ] )
, this implies that there exists a constant @xmath102 such that for all @xmath183 large enough @xmath224 so ( h ) holds for @xmath225 , completing the induction .
this concludes the proof of lemma [ res2 ] for the case where the critical measure has finite mean .
we consider stationary weights of the form @xmath226 with @xmath227 , @xmath228 , and @xmath229 .
we prove non - monotonicity of @xmath141 for @xmath230 and @xmath231 for all @xmath71 via an exact computation .
the case @xmath232 can be done completely analogously but involves different expressions with logarithms in the resulting limits , and is presented in appendix [ bis2 ] .
the proof remains valid for general converging @xmath233 with only minor differences , which we explain in a remark at the end of this section .
convergence of @xmath234 from above or below for the exact power law depends on the parameter @xmath229 , as summarized in the next result .
[ lemma : infinite mean ] for stationary weights of the form @xmath235 and @xmath227 with @xmath229 @xmath236 where @xmath237 for @xmath238 we have @xmath239 which has the same sign as @xmath240 .
this result implies that whenever @xmath235 for @xmath241 and @xmath227 with @xmath230 lemma [ res2 ] holds with @xmath242 .
this completes the proof of lemma [ res2 ] in the case @xmath243 . to prove this result we make use of the full taylor series of @xmath244 at @xmath245 and integral approximations to compute the asymptotic behaviour of summations
. to simplify notation we assume that @xmath44 is even . for odd @xmath44
there is no term with multiplicity one and there exists an obvious modification .
first note that @xmath86 fulfils the regularity assumption and @xmath246 as @xmath247 for all @xmath88 @xcite , so by proposition [ prop1 ] a process with stationary measures @xmath18 will exhibit condensation . for @xmath212
we subtract @xmath203 from @xmath140 to get @xmath248 substituting the taylor expansion of @xmath244 we find @xmath249 in the last line the @xmath250 term was combined with the second term , and we adopt the usual convention that empty products are equal to one .
both summations in @xmath1 are over continuous and monotone functions @xmath251 , therefore we can use the usual integral approximation for decreasing ( increasing ) functions @xmath252 for all @xmath253 and @xmath254 . multiplying by @xmath255
we find the limit as @xmath145 of ( [ powerlaw difference ] ) to be @xmath256 it is shown in appendix [ appendixsign ] that this is positive ( and finite ) in the region @xmath230 and negative ( and finite ) in the region @xmath257 , completing the proof of lemma [ lemma : infinite mean ] for @xmath258 .
the result holds for general system size , @xmath88 , and is proved by induction .
the inductive hypothesis states @xmath259 similar to the case @xmath212 we write @xmath260 we first establish the limit of the function @xmath261 in equation . the inductive hypothesis can be written as @xmath262 which implies @xmath261 can be written as @xmath263 - l z(1)^{l } \right ) \ .\end{aligned}\ ] ] rearranging terms and noting that @xmath264 we then have @xmath265 after taylor expanding @xmath266 appearing in the first line above , it is easy to see that the limit of the first line is given by @xmath267 as @xmath247 . using the @xmath212 result to calculate the limit of the second line we find @xmath268 to identify the limit of @xmath269 in , we again make use of the taylor expansion of @xmath244 similarly to the two site case and we write @xmath270 changing the order of summations , separating the @xmath271 term and using @xmath222 we have @xmath272 for all @xmath273 and @xmath229 we have @xmath274 as @xmath145 , which implies that for any fixed @xmath275 we have @xmath276 . therefore , the following limits are equal @xmath277 using the inductive hypothesis we have @xmath278 is given by @xmath279 now applying the @xmath212
result it is possible to show that @xmath280 where the limit of the first line of was @xmath281 by the additional factor @xmath282 appearing in the summations . combining and we have @xmath283 concluding the induction so the result holds for all @xmath88 . from the recursion it is obvious that @xmath284 will have the same sign as @xmath240 , completing the proof of lemma [ lemma : infinite mean ] .
a slightly modified version of lemma [ lemma : infinite mean ] also holds if the stationary weights are of the form @xmath285 where @xmath286 .
the limit in only depends on the tail behaviour of the weights and is now given by @xmath287 .
briefly , this can be seen as follows , becomes @xmath288 taylor expanding @xmath244 and rearranging to find terms of the form @xmath289 and using the same argument to calculate the limit of @xmath269 we have @xmath290 for all @xmath273 and any @xmath275 , and the result follows .
similar modifications are required in the inductive step and the new limit in is given by @xmath291 for all @xmath88 .
this does not change the sign of the limit in and therefore lemma [ res2 ] still holds .
condensation arises in spatially homogeneous systems with stationary product measures due to the sub - exponential tail of the stationary weights @xmath87 , which has been studied extensively in previous work . in this section
we review relevant results on heavy - tailed distributions and discuss the links between condensation on finite fixed lattices and in the thermodynamic limit before we give the proof of proposition [ prop1 ] in section [ proof equiv ] .
sub - exponential distributions are a special class of heavy - tailed distributions , the following characterization was introduced in @xcite with applications to branching random walks , and has been studied systematically in later work ( see e.g. @xcite ) , for a review see for example @xcite or @xcite . a non - negative random variable @xmath292 with distribution function @xmath293 $ ]
is called heavy - tailed if @xmath294 , @xmath295 for all @xmath296 , and @xmath297 it is called sub - exponential if @xmath294 , @xmath295 for all @xmath296 , and @xmath298 here @xmath299 $ ] denotes the convolution product , the distribution function of the sum of two independent copies @xmath300 and @xmath301 . it has been shown @xcite that is equivalent to either of the following conditions , @xmath302}{\mathbb{p}\big[\max\{x_{i } : i \in \{1 , \ldots , l \ } \}>x\big ] } = 1 \textrm{\quad for all \quad } l { \;\geqslant\;}2 \ . & \end{aligned}\ ] ] the second characterization shows that a large sum of independent sub - exponential random variables @xmath303 is typically realized by one of them taking a large value , which is of course reminiscent of the condensation phenomenon .
it was further shown in @xcite that sub - exponential distributions also have the following properties , @xmath304 most results in the literature are formulated in terms of distribution functions and tails and apply to discrete as well as continuous random variables .
@xcite provides a valuable connection to discrete random variables in terms of their mass functions @xmath86 , @xmath305 .
+ assume the following properties for a sequence @xmath306 , * @xmath307 as @xmath114 , * @xmath308(normalizability ) , * @xmath309 exists .
then ( * ? ? ?
* theorem 1 ) asserts that @xmath310 and @xmath311 is the mass function of a discrete , sub - exponential distribution .
the implication @xmath312 is given in ( * ? ? ?
* lemma 5 ) .
sufficient ( but not necessary ) conditions for assumption ( c ) to hold are given in ( * ? ? ?
* remark 1 ) .
+ provided @xmath313 , then ( c ) holds if either of the following conditions are met : * @xmath314 for some constant @xmath315 , or * @xmath316 where @xmath317 is a smooth function on @xmath318 with @xmath319 and @xmath320 as @xmath321 , and @xmath322 .
+ case ( i ) includes distributions with power law tails , @xmath95 with @xmath96 .
the stretched exponential with @xmath323 , @xmath324 , and the almost exponential with @xmath325 , @xmath326 , are covered by case ( ii ) .
the class of sub - exponential distributions includes many more known examples than the list given in section [ sec : discussion ] ( see e.g. ( * ? ? ? * table 3.7 ) ) .
analogous to the characterisation of sub - exponential distributions , given by , for discrete distributions the existence of the limit @xmath327 is equivalent to the existence of the following condition @xmath328}{\mathbb{p}\big[\max\{x_{1},x_{2}\ } = n \big ] } \to 1 \textrm { as } n \to \infty \ .\ ] ] this holds , since we have the following equality of ratios @xmath329}{\mathbb{p}\big[\max\{x_{1},x_{2}\ } = n\big ] } = \frac{z_{2,n}}{2w(n)\sum_{n=0}^{n}{w(n ) } } = \frac{(w*w)(n)}{2w(n)\sum_{n=0}^{n}{w(n ) } } \ .\ ] ] specific properties of power law tails @xmath86 are used in @xcite to show condensation for finite systems in the sense of definition [ condensation ] . in proposition
[ prop1 ] , proved in section [ proof equiv ] , we extend this result to stationary product measures with general sub - exponential tails . in this context
, condensation is basically characterized by the property which assures emergence of a large maximum when the sum of independent variables is conditioned on a large sum . as summarized in the introduction , condensation in stochastic particle systems has mostly been studied in the thermodynamic limit with particle density @xmath330 , where @xmath331 such that @xmath332 . in that case conditions on the sum of @xmath28 independent random variables
become large deviation events , which have been studied in detail in @xcite .
+ in the thermodynamic limit , a definition of condensation is more delicate and the approach presented in @xcite follows the classical paradigm for phase transitions in statistical mechanics via the equivalence of ensembles ( see e.g. @xcite for more details ) .
a system with stationary product measures exhibits condensation if the critical density is finite , _
i.e. _ @xmath333 and the canonical measures @xmath18 are equivalent to the critical product measure @xmath334 in the limit @xmath331 such that @xmath332 for all super - critical densities @xmath335 .
the interpretation is again that the bulk of the system ( any finite set of sites ) is distributed as the critical product measure in the limit .
it has been shown in @xcite ( see also @xcite for a more complete presentation ) that the regularity condition and @xmath107 imply the equivalence of ensembles , which has therefore been used as a definition of condensation in ( * ? ? ?
* definition 2.1 ) .
therefore , any process that condenses for fixed @xmath336 with @xmath337 in the sense of definition [ condensation ] also condense in the thermodynamic limit .
this includes all previously studied examples @xcite , however there exists distributions that satisfy with @xmath337 but do not satisfy the conditions of proposition [ prop1 ] and do not condense for fixed @xmath336 .
this is illustrated by an example given below .
it is also discussed in ( * ? ? ?
* section 3.2 ) that assumption is not necessary to show equivalence , but weaker conditions are of a special , less general nature and are not discussed here . note also that equivalence of ensembles does not imply that the condensate concentrates on a single lattice site , the latter has been shown so far only for stretched exponential and power - law tails with @xmath107 in @xcite .
both definitions involve only the sequence @xmath18 of canonical measures and not on the dynamics of the underlying process .
since the canonical measures are fully characterised by the weights @xmath86 condensation can be viewed as a feature of the tails of the weights @xmath86 .
the condensation phenomena can also be studied for continuous random variables on the local state space @xmath338 , see for example @xcite .
the following continuous example , taken from @xcite is shown to satisfy but is not sub - exponential .
we show that the distribution has a finite mean and therefore exhibits condensation in the thermodynamic limit but not on a finite lattice in the sense of definition [ condensation ] . for a real - valued random variable @xmath292 with distribution function @xmath339
$ ] , assume @xmath340 .
let @xmath341 to be an increasing sequence with @xmath342 and @xmath343 be a continuous and piecewise linear function such that @xmath344 and @xmath345 for @xmath346 .
the sequence @xmath341 is defined iteratively as follows @xmath347 and @xmath348 for @xmath349 .
the mean is finite since @xmath350 where the final step uses the relation @xmath351 to bound the series from above . for all distributions satisfying which are not sub - exponential @xmath141 does not have a limit in @xmath352 as @xmath145 and with proposition [ prop1 ]
there is no condensation on finite lattices according to definition [ condensation ] . for a discretized version of the example given above with weights @xmath353 we have @xmath354 for @xmath355 as @xmath114 @xcite .
for this example , following the proof of proposition [ prop1 ] this implies that @xmath356\to 0 $ ] for @xmath355 as @xmath145 and all @xmath357 .
therefore , the @xmath212 bulk occupation number @xmath358 diverges in distribution as @xmath145 by receiving a diverging excess mass from the condensate due to the light tail of @xmath86 .
it can be shown that these weights satisfy and therefore exhibit condensation in the thermodynamic limit where the excess mass can be distributed on a diverging number of sites . for a process that exhibits condensation in the thermodynamic limit with a sub - exponential critical product measure , proposition [ prop1 ]
implies that condensation occurs also on finite lattices with @xmath359 .
theorem [ mainres ] then implies that this process is necessarily non - monotone for all fixed system sizes @xmath28 .
however , monotonicity for condensing processes with long - tailed but not sub - exponential stationary measures remains open .
let us first assume that the process exhibits condensation according to definition [ condensation ] and has canonical distributions of the form ( [ eqn : prod measure ] ) where the weights fulfil ( [ regu ] ) , i.e. @xmath360 $ ] as @xmath92 . in this part of the proof we establish that ; 1 .
@xmath74 , 2 .
@xmath361 has a limit as @xmath145 , 3 .
@xmath362 , which also implies @xmath363 as @xmath145 , and 4 . convergence of @xmath363 for some @xmath88 implies convergence for @xmath212 and therefore holds .
step ( 1 ) , show @xmath75 .
assume first that @xmath364 as @xmath114 .
for all @xmath365 and @xmath366 we have @xmath367 & = \frac{l}{z_{l , n}}\sum_{n=0}^{k}z_{l-1,n}w(n - n ) \\ &
{ \;\leqslant\;}l\frac{k+1}{z_{l , n}}\max_{0{\;\leqslant\;}n { \;\leqslant\;}k}\left(z_{l-1,n}\right)\max_{0{\;\leqslant\;}n { \;\leqslant\;}k } \left(w(n - n)\right ) \ .\end{aligned}\ ] ] let @xmath368
. the partition function @xmath369 is trivially bounded below by the event that site 1 takes @xmath370 particles and the second site takes the remaining @xmath371 particles , @xmath372 therefore @xmath373 { \;\leqslant\;}\frac{l}{w(0)^{l-2}}\frac{k+1}{w(n^{\star}+1 ) } \frac{w(n - n^{\star})}{w(n - n^{\star}-1)}\max_{0{\;\leqslant\;}n { \;\leqslant\;}k}\left(z_{l-1,n}\right)\to 0 \end{aligned}\ ] ] as @xmath145 , which implies condensation can not occur in the sense of definition [ condensation ] contradicting the initial assumption , therefore @xmath74 .
step ( 2 ) , prove @xmath141 converges as @xmath145 . by definition [ condensation ]
the limit @xmath374\ , \ ] ] exists and @xmath375 for @xmath376 sufficiently large .
for @xmath377 we have @xmath378 = l\frac{w(n)}{z_{l , n}}\sum_{n=0}^{k}z_{l-1,n}\frac{w(n - n)}{w(n ) } \ .\ ] ] since @xmath379 , @xmath376 is fixed , and @xmath375 , implies the convergence of @xmath141 as @xmath145 .
step ( 3 ) , prove @xmath362 .
by we have @xmath380 as @xmath381 , taking the limit as @xmath145 of this implies @xmath382 since we also have @xmath383 , this implies @xmath143 . using @xmath380 ,
then also implies @xmath384 as @xmath145 .
step ( 4 ) .
we have seen above that condensation implies @xmath74 , @xmath385 , and @xmath384 as @xmath145 , then ( * ? ? ?
* theorem 2.10 ) implies @xmath386 completing this part of the proof .
now , let us consider a stochastic particle system with canonical distributions of the form ( [ eqn : prod measure ] ) which fulfil ( [ wlog ] ) and ( [ paff2 ] ) with @xmath82 and @xmath387 .
we keep the notation for @xmath82 general in the following to clarify the argument .
it is immediate from proposition [ 2tol ] , and remembering that we set @xmath388 , that @xmath389=lw(n)/z_{l , n } \to z(\phi_c ) ^{-(l-1 ) } > 0\ .\ ] ] then we have for all fixed @xmath376 and @xmath366 @xmath390&=l\sum_{n=0}^k \frac{w(n - n)z_{l-1,n}}{z_{l , n } } = \sum_{n=0}^k z_{l-1,n}\frac{w(n - n)}{w(n)}\frac{lw(n)}{z_{l , n}}\nonumber\\ & \to \sum_{n=0}^k \frac{z_{l-1,n } \phi_c^n } { z(\phi_c ) ^{l-1 } } = \nu_{\phi_c } ( \eta_1 + \ldots + \eta_{l-1 } { \;\leqslant\;}k)\end{aligned}\ ] ] as @xmath25 . since @xmath391 is a non - degenerate probability distribution , this implies that @xmath392 as @xmath393 , which is ( [ eqn : cond1 ] ) . to compute the distribution outside the maximum we get for fixed @xmath394 and large enough @xmath44 @xmath395 = \frac{w(n_1 ) \cdots w(n_{l-1})w(n - n_{1 } { -}\ldots { -}n_{l-1})}{\pi_{l , n } [ m_l { = } \eta_l]\ z_{l , n } } } \nonumber\\ & & = \frac{1}{l\pi_{l , n } [ m_l = \eta_l]}w(n_{1 } ) \cdots w(n_{l-1})\frac{w(n - n_{1 } -\ldots -n_{l-1})}{w(n)}\frac{lw(n)}{z_{l , n}}\nonumber\\ & & \to w(n_1 ) \cdots w(n_{l-1})\phi_c^{n_{1 } + \ldots + n_{l-1 } } /z(\phi_c ) ^{l-1}\ , \end{aligned}\ ] ] as @xmath25 .
here we have used that spatial homogeneity of the measure and asymptotic uniqueness of the maximum according to imply @xmath396\to 1/l$ ] .
this completes the proof of proposition [ prop1 ] .
in this section we review several stochastic particle systems that exhibit condensation . by theorem [ mainres ] , if these processes are homogeneous and monotone with a finite critical density they do not have stationary product measures .
to prove monotonicity for the examples mentioned below it is sufficient to construct a basic coupling of the stochastic process which preserves the partial order and particles jump together with maximal rate .
for a definition of a coupling see @xcite and for the statement of strassen s theorem linking stochastic monotonicity and the coupling technique see @xcite .
the steps to construct a basic coupling are outlined in @xcite .
+ condensation in homogeneous particle systems has mostly been studied in the framework of misanthrope processes @xcite . at most
one particle is allowed to jump at a time and the rate that this occurs depends on the number of particles in the exit and entry sites .
the misanthrope process is a stochastic particle system on the state space @xmath397 defined by the generator @xmath398 here @xmath399 denotes the configuration after a single particle has jumped from site @xmath400 to site @xmath401 , which occurs with rate @xmath402 .
the purely spatial part of the jump rates , @xmath403 , are transition probabilities of a random walk on @xmath27 .
such models are usually studied in a translation invariant setting with periodic boundary conditions , typical choices are symmetric , totally asymmetric or fully connected jump rates with @xmath404 , @xmath405 , or @xmath406 , respectively .
misanthrope processes include many well - known examples of interacting particle systems , such as zero - range processes @xcite , the inclusion process @xcite , and the explosive condensation model @xcite .
it is known @xcite that misanthrope processes with translation invariant dynamics @xmath407 exhibit stationary product measures if and only if the rates fulfil @xmath408 and , in addition , either @xmath409 the corresponding stationary weights satisfy @xmath410 misanthrope processes are monotone ( attractive ) @xcite if and only if the jump rates satisfy @xmath411 in theorem [ mainres ] we have proved that processes that exhibit stationary product measures and condensation with finite mean or power law tails , @xmath95 , with @xmath412 $ ] are necessarily not monotone . for power law tails with @xmath413 $ ] convergence of @xmath141 is from below and our method does not disprove monotonicity of the measures @xmath18 or monotonicity of the underlying process . using the specific form of the stationary measures ,
it is clear that possible examples of monotone processes with stationary product measures of this form can not be of misanthrope type .
[ lemma : mismonotone ] a misanthrope process defined by the generator , that has stationary product measures and exhibits condensation is not monotone . gives necessary conditions for the monotonicity of the misanthrope process and implies with that @xmath414 is non - decreasing .
this implies that the ratio converges to @xmath415 $ ] , which is the regularity assumption . assuming the process condenses in the sense of definition [ condensation ] , then proposition [ prop1 ] implies @xmath74 .
now we have @xmath416 for all @xmath71 .
therefore , @xmath417 which implies @xmath418 we conclude that the critical partition function diverges and the critical measure @xmath391 does not exist , which is a necessary condition for condensation .
therefore condensation does not occur in misanthrope processes with stationary product measures . in @xcite
generalised misanthrope processes have been introduced where more than one particle is allowed to jump simultaneously .
they are defined via transitions @xmath419 for @xmath420 at rate @xmath421 and conditions on the jump rates for monotonicity are characterized .
this class provides candidates for possible monotone , condensing processes with product measures as we discuss in the next subsection . the generalised zero - range process ( gzrp ) @xcite is a stochastic particle system on the state space @xmath422 defined by the generator @xmath423 here @xmath424 is the configuration after @xmath425 particles have jumped from @xmath400 to @xmath426 .
the jump rates @xmath427 satisfy @xmath428 if @xmath429 , and we use the convention that empty summations are zero .
we consider translation invariant @xmath430 on a finite lattice @xmath431 and note that the process preserves particle number @xmath432 .
it is known @xcite that these processes exhibit stationary product measures if and only if the jump rates have the explicit form @xmath433 where @xmath434 are arbitrary non - negative functions with @xmath435 strictly positive .
the stationary weights are then given by @xmath436 .
monotonicity of the gzrp can be characterized in terms of @xmath437 the gzrp is monotone if and only if @xmath438 we note these conditions arise from a special case of the results in ( * ? ? ?
* theorem 2.11 ) on generalised misanthrope models , since @xmath427 depends only on the occupation of the exit site and not the entry site . in this class , which is also discussed in detail in @xcite , condensing processes which are monotone , homogeneous , and have stationary product measures with a power tail @xmath95 with @xmath413 $ ] are conjectured to exist . as an example , consider the gzrp with rates given by @xmath439 since @xmath427 is of the form the process exhibits stationary product measures with weights of the form @xmath440 for all @xmath96 and @xmath88 the ratio @xmath441 converges to @xmath442 as @xmath145 @xcite so by proposition [ prop1 ] the process exhibits condensation . to prove the process is monotone we must show the rates satisfy the conditions given in equation .
we first prove @xmath443 for all @xmath444 and @xmath445 .
since @xmath446 for all @xmath445 we can drop the @xmath447 term from the definition of @xmath448 .
we have @xmath449 \ .\end{aligned}\ ] ] since @xmath244 is increasing in @xmath400 for @xmath450 and @xmath451 we have @xmath452 we also need to show @xmath453 for all @xmath454 and @xmath241 . taking discrete derivatives in @xmath425 @xmath455 so @xmath456 is an increasing function in @xmath425 .
therefore , @xmath457 for all @xmath458 , and it suffices to show @xmath459 we present numerical evidence in figure [ gzrpnumerics ] which corroborates our claim that the process with rates ( [ gzrpex ] ) is indeed monotone for @xmath460 $ ] and is not for @xmath461 . the chipping model is a stochastic particle system on the state space @xmath462 , introduced in @xcite .
the dynamics are defined by the generator @xmath463 here @xmath464 denotes the configuration after all the particles at site @xmath400 have jumped to site @xmath401 , which occurs at rate @xmath125 , and single particles jump at rate @xmath465 .
the spatial part @xmath430 is again spatially homogeneous as described in section [ sec : mis ] .
it is easy to see that a basic coupling will preserve the partial order on the state space @xmath29 as defined in section [ sec : mon ] .
therefore , by strassen s theorem @xcite , the chipping model is a monotone process and lemma [ order ] implies that conditional stationary measures of the process are ordered in @xmath44 .
the condensation transition in the chipping model was established on a heuristic level in @xcite .
we have defined the critical density @xmath466 only for systems with product stationary measures ( see ) . in general , the critical density on a fixed system of size @xmath23 , with unique invariant measures @xmath467 , can be defined as the background density of bulk sites @xmath468 notice if @xmath467 are conditional product measures ( see ) then @xmath469 is consistent with and in particular independent of @xmath28 , which follows from proposition [ prop1 ] ( more explicitly ) . for the chipping model in the case
@xmath212 , the process reduces to a @xmath125-dimensional process on @xmath470 and the measure @xmath471 and @xmath472 can be computed explicitly to find @xmath473 in @xcite the critical density in the thermodynamic limit is defined as @xmath474 inspired by the fact that in case of condensation the second moment is either dominated by the condensate and scales like the system size @xmath28 , or it diverges since the maximal invariant measure does not have finite second moment .
it is shown by heuristic computations in a mean - field limit that @xmath475 this suggests that the critical density can depend on the system size @xmath28 for distributions with non - product stationary measures .
the @xmath476 scaling of the critical density can be intuitively understood in the two site chipping model with @xmath44 particles .
this process can be interpreted as a symmetric random walk on the state space @xmath477 with jumps @xmath478 at rate @xmath87 and random jumps to either boundary ( resetting , @xmath479 or @xmath44 ) at rate @xmath125 . after a reset the particle diffuses at rate @xmath87 and reaches a typical distance @xmath476 from the boundary until the next reset .
so this model is a monotone and spatially homogeneous process that heuristically exhibits a condensation transition with finite ( size dependent ) critical density , but it does not exhibit stationary product measures .
condensation is also observed in models where chipping is absent ( @xmath480 ) and the dynamics result in a single block of particles jumping on the lattice @xmath481 corresponding to the critical density @xmath121 .
condensation and non - monotonicity are also related to convexity properties of the entropy , which we briefly describe in the following in a non self - contained and non - rigorous discussion that is aimed at readers with a background in statistical mechanics . in the thermodynamic limit
the canonical entropy is defined as @xmath482 for the processes we consider , equivalence of canonical and grand - canonical ensembles has been established in @xcite for condensing or non - condensing systems , so @xmath483 is given by the ( logarithmic ) legendre transform of the pressure @xmath484 this takes a particularly simple form since the grand - canonical measures are factorisable , and is a strictly convex function for @xmath485 .
general results then imply that @xmath486 also has to be strictly convex below the critical density @xmath207 ( see e.g. @xcite ) , which holds for non - condensing systems and condensing systems with @xmath487
. for condensing systems with finite critical density @xmath483 is linear for @xmath488 , consistent with phase separation phenomena , where in this case the condensed phase formally exhibits density @xmath489 ( see e.g. @xcite for a general discussion ) .
it is not possible to derive general results for finite @xmath28 and @xmath44 , but if we assume that the ratio of weights @xmath490 is monotone increasing in @xmath1 , we can show that a monotone order of @xmath18 implies that @xmath491 is necessarily convex .
note that with ( [ regu ] ) our assumption implies that @xmath86 has exponential tails with @xmath81 or decays super - exponentially with @xmath492 , and in both cases the system does not exhibit condensation .
we can define @xmath493 so that @xmath494 is a monotone increasing test function on @xmath495 .
it is easy to see that for its canonical expectation we have for all @xmath23 and @xmath496 @xmath497 therefore , monotonicity of the canonical measures implies that the ratio of partition functions ( [ eq : exe ] ) is increasing and the discrete derivative of @xmath498 in @xmath44 is decreasing .
we expect that in the limit @xmath499 the monotonicity assumption on @xmath490 is not necessary , and @xmath500 is convex in @xmath44 for all non - condensing systems , consistent with strict convexity of @xmath483 . for condensing systems the weights @xmath87 decay sub - exponentially , and
if @xmath490 is monotone then it has to be decreasing in @xmath1 .
therefore the choice @xmath493 implies @xmath501 is not a monotone test function , and the above general arguments can not be used to relate non - convexity of @xmath500 to the absence of a monotone order in @xmath18 . for particular condensing systems , however , it has been shown that @xmath500 is typically convex for small @xmath502 and concave for larger @xmath111 @xcite .
these results focus on power law and stretched exponential tails for @xmath86 , and have been derived for zero - range processes where the ratio @xmath503 is equal to the canonical current .
non - monotone behaviour around the critical density therefore has implications for finite - size corrections and derivations of hydrodynamic limits as mentioned in the introduction .
consider stationary weights of the form @xmath504 with @xmath227 , we prove the non - monotonicity of @xmath141 in a similar fashion to the proof of lemma [ lemma : infinite mean ] summarised in the following lemma .
[ lemmabis2 ] for stationary weights of the form @xmath504 with @xmath227 we have @xmath505 for @xmath238 we have @xmath506 which is positive for all @xmath88 since @xmath507 .
first consider the case @xmath258 .
as in the proof of lemma [ lemma : infinite mean ] we will utilise the full taylor expansion of @xmath508 , integral bounds on monotone series , and assume @xmath44 is even , for @xmath44 odd there exists obvious modifications to the proof .
we have @xmath509 where the terms @xmath510 in the above summations cancel . substituting the taylor expansion of @xmath508 we find @xmath511 now we are in a position to apply the integral bounds , first noting that @xmath512 is decreasing for @xmath513 , constant and equal to @xmath125 for @xmath514 , and increasing for @xmath515 . multiplying both sides of and applying the integral bounds it is easy to show @xmath516
now consider the case @xmath517 and make the following inductive hypothesis @xmath518 as in the proof of lemma [ lemma : infinite mean ] write @xmath519 we fist establish the limit of @xmath520 in .
the inductive hypothesis can be rewritten as @xmath521 similar to the proof of lemma [ lemma : infinite mean ] @xmath520 can be written in the form @xmath522 since @xmath523 is decreasing for @xmath524 we can find upper and lower bounds of the first term , by pulling out the logarithm , of the form @xmath525 applying the same steps exactly as they appear in the proof of lemma [ lemma : infinite mean ] to the upper and lower bounds we have @xmath526 to identify the limit of @xmath527 in we again follow the steps given in the proof of lemma [ lemma : infinite mean ] , which implies @xmath528 combining this with we have @xmath529 from the recursion it is obvious that @xmath530 will have the same sign as @xmath531 , completing the proof of lemma [ lemmabis2 ] .
in this section , we compute the sign of @xmath240 for @xmath229 , where @xmath532 recall the definition of the pochhammer symbol @xmath533 and the hypergeometric function @xmath534 we now show @xmath535 which in particular implies @xmath536 by evaluating the hypergeometric formula . factorising the term @xmath537 from @xmath240 and
rearranging terms inside the summation we have @xmath538 now use the following identities to simplify the terms inside the summation @xmath539 which gives the required result . to complete the proof we use the following two relations for hypergeometric functions , euler s transform @xcite @xmath540 and gauss s second summation theorem @xcite @xmath541 therefore , @xmath542 to calculate the sign of @xmath240 we first note that the gamma function @xmath543 is positive for all @xmath296 and negative in the region @xmath544 , which implies @xmath545
this work was supported by the engineering and physical sciences research council ( epsrc ) , grant no .
p.c . acknowledges fellowship funding from the university of warwick , institute of advanced study .
the authors are grateful to ellen saada and thierry gobron for useful discussions and comments on the manuscript . | we study stochastic particle systems that conserve the particle density and exhibit a condensation transition due to particle interactions .
we restrict our analysis to spatially homogeneous systems on finite lattices with stationary product measures , which includes previously studied zero - range or misanthrope processes .
all known examples of such condensing processes are non - monotone , i.e. the dynamics do not preserve a partial ordering of the state space and the canonical measures ( with a fixed number of particles ) are not monotonically ordered . for our main result
we prove that condensing homogeneous particle systems with finite critical density are necessarily non - monotone . on finite lattices condensation
can occur even when the critical density is infinite , in this case we give an example of a condensing process that numerical evidence suggests is monotone , and give a partial proof of its monotonicity . |
we will consider the random matrix ensemble with an external source , @xmath8 defined on @xmath1 hermitian matrices @xmath9 .
the number @xmath7 is a large parameter in the ensemble .
the gaussian ensemble , @xmath10 , has been solved in the papers of pastur @xcite and brzin - hikami @xcite@xcite , by using spectral methods and a contour integration formula for the determinantal kernel . in the present work
we will develop a completely different approach to the solution of the gaussian ensemble with external source .
our approach is based on the riemann - hilbert problem and it is applicable , in principle , to a general @xmath11 .
we will assume that the external source @xmath2 is a fixed diagonal matrix with @xmath12 eigenvalues @xmath13 and @xmath14 eigenvalues @xmath15 , @xmath16 as shown by p. zinn - justin @xcite , for any @xmath17 , the @xmath18-point correlation function of eigenvalues of @xmath9 has the determinantal form , @xmath19 in our previous work @xcite we show that the kernel @xmath20 can be expressed in terms of a solution to the following matrix riemann - hilbert ( rh ) problem : find @xmath21 such that * @xmath22 is analytic on @xmath23 , * for @xmath24 , we have @xmath25 where @xmath26 and @xmath27 ( @xmath28 ) denotes the limit of @xmath29 as @xmath30 from the upper ( lower ) half - plane , * as @xmath31 , we have @xmath32 where @xmath33 denotes the @xmath34 identity matrix .
namely , @xmath35_{21}+e^{-nay}[y(y)^{-1}y(x)]_{31 } } { 2\pi i(x - y)}. \\ & = \frac{e^{-\frac{1}{2 }
n ( v(x ) + v(y))}}{2\pi i(x - y ) } \begin{pmatrix } 0 & e^{nay } & e^{-nay}\end{pmatrix } y(y)^{-1 } y(x ) \begin{pmatrix } 1 \\ 0 \\ 0 \end{pmatrix}. \end{aligned}\ ] ] the rh problem has a unique solution and the solution is expressed in terms of multiple orthogonal polynomials , see @xcite and section 2.1 below . for
now , let us mention that the @xmath36 entry @xmath37 satisfies @xmath38\end{aligned}\ ] ] where @xmath39 denotes expectation with respect to the measure ( [ m1 ] ) .
so it is the average characteristic polynomial for the random matrix ensemble .
it is the aim of this paper to analyze the rh problem as @xmath5 , by using the method of steepest descent / stationary phase of deift and zhou @xcite .
we focus here on the gaussian case @xmath40 .
our first result concerns the limiting mean eigenvalue density .
[ maintheo1 ] let @xmath41 , @xmath42 ( so @xmath7 is even ) and let @xmath4
. then the limiting mean density of eigenvalues @xmath43 exists , and it is supported by two intervals , @xmath44 $ ] and @xmath45 $ ] .
the density @xmath46 is expressed as @xmath47 where @xmath48 solve the cubic equation , @xmath49 ( pastur s equation ) .
the density @xmath50 is real analytic on @xmath51 and it vanishes like a square root at the edge points of its support , i.e. , there exist constants @xmath52 such that @xmath53 we obtain @xmath50 from an analysis of the equation @xmath54 the critical points of the mapping ( [ rem1 ] ) satisfy @xmath55 for @xmath4 , the four critical points are real , and they correspond to four real branch points @xmath56 , @xmath57 with @xmath58 .
we denote the three inverses of ( [ rem1 ] ) by @xmath59 , @xmath60 , where @xmath61 is chosen such that @xmath62 as @xmath63 .
then @xmath61 has an analytic continuation to @xmath64 \cup [ z_2 , z_1])$ ] and @xmath65 for @xmath66 .
then the density @xmath50 is @xmath67 see section 3 .
the assumption @xmath4 is essential for four real branch points and a limiting mean eigenvalue density which is supported on two disjoint intervals . for @xmath68 ,
two branch points are purely imaginary , and the limiting mean eigenvalue density is supported on one interval .
the main theorem on the local eigenvalue correlations continues to hold , but its proof requires a different analysis of the rh problem .
this will be done in part ii . in part
iii we will discuss the case @xmath69 .
the density @xmath50 can also be characterized by a minimization problem for logarithmic potentials .
consider the following energy functional defined on pairs @xmath70 of measures : @xmath71 the problem is to minimize @xmath72 among all pairs @xmath70 of measures on @xmath73 with @xmath74 .
there is a unique minimizer , and for @xmath4 , it can be shown that @xmath75 is supported on @xmath45 $ ] , @xmath76 is supported on @xmath44 $ ] and @xmath50 is the density of @xmath77 .
this minimal energy problem is similar to the minimal energy problem for angelesco systems in the theory of multiple orthogonal polynomials , see @xcite .
it is possible to base the asymptotic analysis of the rh problem on the minimization problem , as done by deift et al , see @xcite , for the unitarily invariant random matrix model .
however , we will not pursue that here .
our main results concern the universality of local eigenvalue correlations in the large @xmath7 limit .
this was established for unitarily invariant random matrix models @xmath78 by bleher and its @xcite for a quartic polynomial @xmath11 , and by deift et al @xcite for general real analytic @xmath11 .
the universality may be expressed by the following limit @xmath79 which is valid for @xmath80 in the bulk of the spectrum , i.e. , for @xmath80 such that the limiting mean eigenvalue density @xmath81 is positive .
the proof of ( [ univ1 ] ) established dyson s universality conjecture @xcite for unitary ensembles . in our case , we use a rescaled version of the kernel @xmath82 @xmath83 for some function @xmath84 .
the rescaling ( [ m16 ] ) is allowed because it does not affect the correlation functions @xmath85 ( [ m3 ] ) , which are expressed as determinants of the kernel . note that the kernel @xmath82 of ( [ m7 ] ) is non - symmetric and there is no obvious a priori scaling for it .
the function @xmath84 in ( [ m16 ] ) has the following form on @xmath86 @xmath87 with @xmath88 , where @xmath61 is as in the first remark after theorem [ maintheo1 ] .
[ maintheo2 ] let @xmath41 , @xmath42 , and let @xmath4 .
let @xmath89 and @xmath50 be as in theorem [ maintheo1 ] and let @xmath90 be as in ( [ m16 ] ) .
then for every @xmath91 and @xmath92 , we have @xmath93 our final result concerns the eigenvalue correlations near the edge points @xmath94 . for unitarily invariant random matrix ensembles ( [ univ0 ] ) the local correlations
near edge points are expressed in the limit @xmath5 in terms of the airy kernel @xmath95 provided that the limiting mean eigenvalue density vanishes like a square root , which is generically the case @xcite . in our non - unitarily invariant random matrix model , the limiting mean eigenvalue density vanishes like a square root , ( [ rho2 ] ) , and indeed we recover the kernel ( [ univ3 ] ) in the limit @xmath96 .
[ maintheo3 ] we use the same notation as in theorem [ maintheo2 ] .
let @xmath97 and @xmath98 be the constants from ( [ rho2 ] ) . then for every @xmath99 we have @xmath100 and @xmath101 similar limits hold near the edge points @xmath102 and @xmath103 .
as said before , our results follow from an asymptotic analysis of the rh problem ( [ m4])([m6 ] ) , which involves @xmath34 matrices . in the past ,
asymptotics for rh problems has mostly been restricted to @xmath104-matrix valued rh problems , see e.g.@xcite and references cited therein .
the first asymptotic analysis of a @xmath34 matrix rh problem appeared in @xcite in an approximation problem for the exponential function . in the present work we use some of the ideas of @xcite .
as in @xcite a main tool in the analysis is an appropriate three sheeted riemann surface . to motivate the choice of the riemann surface we describe in section [ sec2 ] the recurrence relations and differential equations that are satisfied by a matrix @xmath105 , which is an easy modification of @xmath22 , see ( [ m13 ] ) below .
the riemann surface is studied in section [ sec3 ] and we obtain from it the functions @xmath106 and @xmath107 , @xmath60 , that are necessary for the transformations of the rh problem .
the first transformation @xmath108 normalizes the rh problem at infinity and at the same time introduces oscillating diagonal entries in the jump matrices on the cuts @xmath44 $ ] and @xmath45 $ ] , see section [ sec4 ] .
the second transformation @xmath109 involves opening of lenses around the cuts , which results in a rh problem for @xmath110 with rapidly decaying off - diagonal entries in the jump matrices on the upper and lower boundaries of the lenses , see section [ sec5 ] .
the next step is the construction of a parametrix , an approximate solution to the rh problem . in section [ sec6 ]
we ignore all jumps in the rh problem for @xmath110 , except those on the cuts @xmath44 $ ] , @xmath45 $ ] .
this leads to a model rh problem , which we solve by lifting it to the riemann surface via the functions @xmath111 .
this leads to the parametrix away from the edge points @xmath112 .
a separate construction is needed near the edge points .
this is done in section [ sec7 ] where we build the local parametrices out of airy functions .
the final transformation @xmath113 is done in section [ sec8 ] and it leads to a rh problem for @xmath114 whose jump matrices are uniformly close to the identity matrix .
then we can use estimates on solution of rh problems , see @xcite , to conclude that @xmath114 is close to the identiy matrix , with error estimates . having that we can give the proofs of the theorems in section [ sec9 ] .
our approach proves simultaneously large @xmath7 asymptotics of the @xmath36 entry of @xmath22 , which by ( [ y11 ] ) is equal to the average characteristic polynomial .
this polynomial is called a multiple hermite polynomial for the case of @xmath115 , see @xcite and section [ sec2 ] below . since its asymptotics
may be of independent interest , we consider it briefly in section [ sec10 ] below .
more information on multiple orthogonal polynomials and their asymptotics can be found in @xcite , see also the surveys @xcite and the references cited therein .
in order to motivate the introduction of the riemann surface associated with ( [ rem1 ] ) we discuss here the recurrence relations and differential equations that are satisfied by the solution of the riemann - hilbert problem ( [ m4])([m6 ] ) in case @xmath115 . it also reveals the integrable structure .
we note however , that the results of this subsection are not essential for the rest of the paper . for the recurrence relations we need to separate the indices @xmath12 and @xmath14 in the asymptotic behavior ( [ m6 ] ) from the exponent @xmath7 in the weight functions @xmath116 , @xmath117 of ( [ m5 ] ) .
in this section we put @xmath118 where @xmath119 is fixed , and we let @xmath120 be the solution of the riemann - hilbert problem ( [ m4 ] ) , ( [ m6 ] ) with @xmath115 and @xmath116 , @xmath117 given by ( [ m5a ] ) .
let @xmath121 be a monic polynomial of degree @xmath122 such that for @xmath123 , @xmath124 the polynomial @xmath125 is unique and it is called a multiple hermite polynomial , see @xcite .
denote for @xmath123 , @xmath126 the solution to the rh problem is @xmath127 with the constants @xmath128 and where @xmath129 denotes the cauchy transform of @xmath130 , @xmath131 the recurrence relations and differential equations are nicer formulated in terms of the function @xmath132 the function @xmath133 solves the following rh problem : * @xmath105 is analytic on @xmath23 , * for @xmath24 , we have @xmath134 * as @xmath31 , we have @xmath135 [ lax ] we have the recurrence relations , @xmath136 and the differential equation , @xmath137 the proof of proposition [ lax ] is given in the appendix [ ai ] below .
we look for a wkb solution of the differential equation ( [ lax2 ] ) of the form @xmath138 where @xmath139 is a diagonal matrix . by substituting this form into ( [ lax2 ] )
we obtain the equation , @xmath140 where @xmath2 is the matrix of coefficients in ( [ lax2 ] ) . by dropping the last term
we reduce it to the eigenvalue problem , @xmath141 the characteristic polynomial is @xmath142 & = \left|\begin{matrix } \xi - z & t_1 & t_2 \\ { } -1 & \xi - a & 0 \\ { } -1 & 0 & \xi+a \end{matrix}\right|\\ & = \xi^3-z\xi^2+(t_1+t_2-a^2)\xi+(t_1-t_2+za)a , \end{aligned}\ ] ] where @xmath143 and @xmath144 .
the spectral curve @xmath145 defines a riemann surface , which in the case of interest in this paper ( where @xmath146 and @xmath147 ) reduces to @xmath148 this defines the riemann surface that will play a central role in the rest of the paper .
the riemann surface is given by the equation ( [ bp0 ] ) or , if we solve for @xmath149 , @xmath150 there are three inverse functions to ( [ bp1 ] ) , which we choose such that as @xmath63 , @xmath151 we need to find the sheet structure of the riemann surface ( [ bp0 ] ) .
the critical points of @xmath152 satisfy the equation @xmath153 which is biquadratic .
the roots are @xmath154 the value @xmath155 is critical , in the sense that for @xmath156 all the roots are real , while for @xmath157 , two are real and two are purely imaginary . in this paper
we will consider the case @xmath156 .
as noted before , we will consider the cases @xmath158 and @xmath155 in parts ii and iii
. set @xmath159 then the critical points are @xmath160 .
the branch points on the @xmath149-plane are @xmath56 and @xmath57 , where @xmath161 we can show that @xmath61 , @xmath162 , and @xmath163 have analytic extensions to @xmath164 \cup [ z_2,z_1])$ ] , @xmath165 $ ] and @xmath166 $ ] , respectively . also on the cut
@xmath45 $ ] , @xmath167 and @xmath168 is real . on the cut
@xmath44 $ ] , @xmath169 and @xmath170 is real .
figure 1 depicts the three sheets of the riemann surface ( [ bp0 ] ) .
we define @xmath171 \cup [ z_2 , z_1].\ ] ] [ rho - density ] we have @xmath172 for @xmath173 and @xmath174 moreover , there are @xmath52 such that @xmath175 * proof * : the fact that @xmath172 for @xmath66 was already noted in ( [ bp8 ] ) and ( [ bp9 ] ) .
we have for @xmath176 $ ] , @xmath177 thus @xmath178 where @xmath179 is a contour encircling the interval @xmath45 $ ] once in the positive direction . letting the contour go to infinity and using the asymptotic behavior ( [ wkb5 ] ) of @xmath180 as @xmath63
, we find the value of the second integral in ( [ rhoint ] ) .
the value of the first integral follows in the same way .
for ( [ rhoconst ] ) we note that near the branch point @xmath181 , we have for a constant @xmath182 , @xmath183 as @xmath184 .
similarly , near @xmath185 we have for a constant @xmath186 , @xmath187 as @xmath188 ( with main branches of the square root ) . by symmetry
, we have similar expressions near @xmath102 and @xmath103 and ( [ rhoconst ] ) follows .
next , we need the integrals of the @xmath189-functions , @xmath190 which we take so that @xmath191 and @xmath192 are analytic on @xmath193 $ ] and @xmath194 is analytic on @xmath195 $ ] . from ( [ wkb5 ] )
it follows that , as @xmath196 , @xmath197 where @xmath198 , @xmath199 , @xmath200 are some constants , which we choose as follows .
we choose @xmath198 and @xmath199 such that @xmath201 and then @xmath200 such that @xmath202 then we have the following jump relations : @xmath203 , \\ { \lambda}_{1+}(x ) - { \lambda}_{1-}(x )
= -2\pi i , & \qquad x \in ( -\infty , -z_1 ] , \\ { \lambda}_{2+}(x ) - { \lambda}_{2-}(x ) = \pi i , & \qquad x\in ( -\infty , z_2 ] , \\ { \lambda}_{1+}(x ) = { \lambda}_{2-}(x ) , \quad { \lambda}_{1-}(x ) = { \lambda}_{2+}(x ) , & \qquad x \in [ z_2,z_1 ] , \\ { \lambda}_{1+}(x ) = { \lambda}_{3-}(x ) , \quad { \lambda}_{1-}(x ) -
\pi i = { \lambda}_{3+}(x ) , & \qquad x \in [ -z_1,-z_2 ] . \\
{ \lambda}_{3+}(x ) - { \lambda}_{3-}(x ) = \pi i , & \qquad x \in ( -\infty , -z_1 ]
. \end{aligned}\ ] ] note that due to the first two equations of ( [ wkb7 ] ) we have that @xmath204 is analytic on the complex plane with cuts on @xmath44 $ ] and @xmath45 $ ] ( recall that @xmath7 is even ) .
furthermore , we also see that @xmath205 ( resp . , @xmath206 )
is analytic on the complex plane with a cut on @xmath45 $ ] ( resp . , @xmath44 $ ] ) , see figure [ fig1 ] .
for later use , we state the following two propositions .
[ laineq1 ] on @xmath207 $ ] we have @xmath208 , and on @xmath209 $ ] , we have @xmath210 . it is easy to see that @xmath211 for @xmath212 . since @xmath213 and @xmath214 for @xmath60 , it is then clear that @xmath215 for @xmath212 .
we also have that @xmath216 for @xmath217 , from which it follows that @xmath218 .
similarly we find that @xmath210 on @xmath209 $ ] .
[ laineq2 ] 1 .
the open interval @xmath219 has a neighborhood @xmath220 in the complex plane such that @xmath221 for every @xmath222 .
the open interval @xmath223 has a neighborhood @xmath224 in the complex plane such that @xmath225 for every @xmath226 .
the function @xmath227 is purely imaginary on @xmath219 .
its derivative is @xmath228 , and this has negative imaginary part .
the cauchy riemann equations then imply that the real part of @xmath229 increases as we move from the interval @xmath219 into the upper half - plane .
thus @xmath230 for @xmath149 near @xmath219 in the upper half - plane .
similarly , @xmath230 for @xmath149 near @xmath219 in the lower half - plane . by proposition [ laineq1 ]
we have @xmath231 on @xmath45 $ ] . by continuity
, the inequality continues to hold in a complex neighborhood of @xmath45 $ ] .
this proves part ( a ) .
the proof of part ( b ) is similar .
using the functions @xmath232 and the constants @xmath233 , @xmath60 , from the previous section , we define @xmath234 then by ( [ m4 ] ) and ( [ deft ] ) , we have @xmath235 , @xmath236 , where @xmath237 the jump relations ( [ wkb7 ] ) allow us to simplify the jump matrix @xmath238 on the different parts of the real axis . on @xmath45
$ ] , ( [ jt ] ) reduces to @xmath239 and on @xmath44 $ ] to @xmath240 on @xmath241 \cup [ -z_2,z_2]\cup [ z_1,\infty)$ ] , ( [ jt ] ) reduces to @xmath242 \cup [ -z_2,z_2 ] \cup [ z_1,\infty ) .
\end{aligned}\ ] ] the asymptotics of @xmath243 are , because of ( [ m6 ] ) , ( [ wkb6 ] ) , and ( [ deft ] ) , @xmath244 thus @xmath243 solves the following rh problem : * @xmath243 is analytic on @xmath23 , * @xmath245 * as @xmath31 , @xmath246 using ( [ deft ] ) in ( [ m7 ] ) we see that the kernel @xmath82 can be expressed in terms of @xmath243 as follows @xmath247
the second transformation of the rh problem is opening of lenses . consider a lens with vertices @xmath248 , see figure [ fig2 ] .
the lens is contained in the neighborhood @xmath220 of @xmath219 , see proposition [ laineq2 ] .
we have the factorization , @xmath249 set @xmath250 then ( [ ft14 ] ) and ( [ st2 ] ) imply that @xmath251.\ ] ] similarly , consider a lens with vertices @xmath252 that is contained in @xmath224 ( see proposition [ laineq2 ] ) and set @xmath253 then ( [ ft14 ] ) and ( [ st4 ] ) imply that @xmath254.\ ] ] set @xmath255 then we have jumps on the boundary of the lenses , @xmath256 where the contours are oriented from left to right ( that is , from @xmath102 to @xmath103 , or from @xmath185 to @xmath181 ) , and where @xmath257 ( @xmath258 ) denotes the limiting value of @xmath110 from the left ( right ) if we traverse the contour according to its orientation .
the jump matrix @xmath259 in ( [ st7 ] ) has the form @xmath260$-lens},\\
j_s(z)&= \begin{pmatrix } 1 & 0 & 0 \\ e^{n({\lambda}_1(z)-{\lambda}_2(z ) ) } & 1 & -e^{n({\lambda}_3(z)-{\lambda}_2(z ) ) } \\ 0 & 0 & 1 \end{pmatrix}\;\text{\rm on the lower boundary of the $ [ z_2,z_1]$-lens},\\ j_s(z)&= \begin{pmatrix } 1 & 0 & 0 \\ 0 & 1 & 0 \\ e^{n({\lambda}_1(z)-{\lambda}_3(z ) ) } & e^{n({\lambda}_2(z)-{\lambda}_3(z ) ) } & 1 \end{pmatrix}\;\text{\rm on the upper boundary of the $ [ -z_1,-z_2]$-lens},\\ j_s(z)&= \begin{pmatrix } 1 & 0 & 0 \\ 0 & 1 & 0 \\ e^{n({\lambda}_1(z)-{\lambda}_3(z ) ) } & -e^{n({\lambda}_2(z)-{\lambda}_3(z ) ) } & 1 \end{pmatrix}\;\text{\rm on the lower boundary of the $ [ -z_1,-z_2]$-lens}. \end{aligned}\ ] ] on @xmath261\cup[-z_2,z_2]\cup[z_1,\infty)$ ] , @xmath110 has the same jump as @xmath243 , so that @xmath262\cup[-z_2,z_2]\cup[z_1,\infty).\ ] ] thus , @xmath110 solves the following rh problem : * @xmath110 is analytic on @xmath263 , where @xmath264 is the boundary of the lenses , * @xmath265 * as @xmath31 , @xmath266 the kernel @xmath82 is expressed in terms of @xmath110 as follows , see ( [ ft16 ] ) and the definitions ( [ st2 ] ) and ( [ st4 ] ) . for @xmath267 and @xmath268 in @xmath219
we have @xmath269 while for @xmath267 and @xmath268 in @xmath223 we have @xmath270 since @xmath271 and @xmath272 are complex conjugates on @xmath219 , we can rewrite ( [ st12 ] ) for @xmath273 as @xmath274 where @xmath275 as in ( [ m20 ] ) .
similarly , we have for @xmath276 , @xmath277
as @xmath96 , the jump matrix @xmath278 is exponentially close to the identity matrix at every @xmath149 outside of @xmath44 \cup [ z_2,z_1]$ ] .
this follows from ( [ st8 ] ) and proposition [ laineq2 ] for @xmath149 on the boundary of the lenses , and from ( [ st9 ] ) , ( [ ft4 ] ) and proposition [ laineq1 ] for @xmath149 on the real intervals @xmath279 , @xmath280 and @xmath281 . in this section
we solve the following model rh problem , where we ignore the exponentially small jumps : find @xmath282\cup[z_2,z_1])\to { { \mathbb c}}^{3\times
3}$ ] such that * @xmath9 is analytic on @xmath283\cup[z_2,z_1])$ ] , * @xmath284 * as @xmath31 , @xmath285 this problem is similar to the rh problem considered in ( * ? ? ?
* section 6.1 ) .
we also follow a similar method to solve it .
we lift the model rh problem to the riemann surface of ( [ bp0 ] ) with the sheet structure as in figure [ fig1 ] .
consider to that end the range of the functions @xmath111 on the complex plane , @xmath286 \cup [ z_2,z_1])),\\ { \omega}_2&=\xi_2({{\mathbb c}}\setminus [ z_2,z_1 ] ) , \\ { \omega}_3&=\xi_3({{\mathbb c}}\setminus [ -z_1,-z_2 ] ) .
\end{aligned}\ ] ] then @xmath287 , @xmath288 , @xmath289 give a partition of the complex plane into three regions , see figure [ fig3 ] .
the regions @xmath288 , @xmath289 are bounded , @xmath290 , @xmath291 , with the symmetry conditions , @xmath292 denote by @xmath293 the boundary of @xmath294 , @xmath295 , see figure [ fig3 ] . then we have @xmath296)&=\xi_{2-}([z_2,z_1 ] ) = { \gamma}_{2}^+\equiv { \gamma}_2\cap\{{{\operatorname{im\,}}}z\ge 0\ } , \\ \xi_{1-}([z_2,z_1])&=\xi_{2+}([z_2,z_1 ] ) = { \gamma}_{2}^-\equiv { \gamma}_2\cap\{{{\operatorname{im\,}}}z\le 0\ } , \\ \xi_{1+}([-z_1,-z_2])&=\xi_{3-}([-z_1,-z_2 ] ) = { \gamma}_{3}^+\equiv { \gamma}_3\cap\{{{\operatorname{im\,}}}z\ge 0\}\\ \xi_{1-}([-z_1,-z_2])&=\xi_{3+}([-z_1,-z_2 ] ) = { \gamma}_{3}^-\equiv { \gamma}_3\cap\{{{\operatorname{im\,}}}z\le 0\}. \end{aligned}\ ] ] we are looking for a solution @xmath9 in the following form : @xmath297 where @xmath298 , @xmath299 , @xmath300 are three analytic functions on @xmath301 .
to satisfy jump condition ( [ mod2 ] ) we need the following relations for @xmath302 : @xmath303 since @xmath304 , @xmath305 , @xmath306 , then to satisfy ( [ mod3 ] ) we demand @xmath307 equations ( [ mod9])([mod10 ] ) have the following solution : @xmath308 with cuts at @xmath309 , @xmath310 .
the constants @xmath311 are determined by the equations @xmath312 . by ( [ bp4 ] ) ,
@xmath313 hence @xmath314 by taking into account the cuts of @xmath299 we obtain that @xmath315 hence @xmath316 similarly , @xmath317 hence @xmath318 is the same as @xmath319 , @xmath320 thus , the solution to the model rh problem is given as @xmath321 with cuts on @xmath45 $ ] and @xmath44 $ ] .
the model solution @xmath322 will be used to construct a _
parametrix _ for the rh problem for @xmath110 outside of a small neighborhood of the edge points .
namely , we will fix some @xmath323 and consider the disks of radius @xmath324 around the edge points . at the edge points @xmath322
is not analytic and in a neighborhood of the edge points the parametrix is constructed differently .
we consider small disks @xmath325 with radius @xmath326 and centered at the edge points , and look for a local parametrix @xmath327 defined on the union of the four disks such that * @xmath327 is analytic on @xmath328 , * @xmath329 * as @xmath96 , @xmath330 we consider here the edge point @xmath181 in detail .
we note that by ( [ rhoconst ] ) and ( [ wkb5a ] ) we have as @xmath184 , @xmath331 so that @xmath332 as @xmath184 .
then it follows that @xmath333^{2/3}\ ] ] is analytic at @xmath181 , real - valued on the real axis near @xmath181 and @xmath334 .
so @xmath335 is a conformal map from @xmath336 to a convex neighborhood of the origin , if @xmath324 is sufficiently small ( which we assume to be the case ) .
we take @xmath337 near @xmath181 such that @xmath338 then @xmath337 and @xmath73 divide the disk @xmath339 into four regions numbered i , ii , iii , and iv , such that @xmath340 , @xmath341 , @xmath342 , and @xmath343 for @xmath149 in regions i , ii , iii , and iv , respectively .
recall that the jumps @xmath259 near @xmath181 are given by ( [ st3 ] ) , ( [ st8 ] ) , and ( [ ft4 ] ) : @xmath344 . \end{aligned}\ ] ] we write @xmath345 then the jumps for @xmath346 are @xmath347 where @xmath348 . \end{aligned}\ ] ] we still have the matching condition @xmath349 since @xmath350 on @xmath351 , which follows from proposition [ laineq1 ] . the rh problem for @xmath346 is essentially a @xmath104 problem , since the jumps ( [ lp5 ] ) are non - trivial only in the upper @xmath104 block .
a solution can be constructed in a standard way out of airy functions .
the airy function @xmath352 solves the equation @xmath353 and for any @xmath354 , in the sector @xmath355 , it has the asymptotics as @xmath63 , @xmath356 the functions @xmath357 , @xmath358 , where @xmath359 , also solve the equation @xmath360 , and we have the linear relation , @xmath361 write @xmath362 and we use these functions to define @xmath363 then @xmath364 where @xmath365 is an analytic prefactor that takes care of the matching condition ( [ match2 ] ) .
explicitly , @xmath365 is given by @xmath366 a similar construction works for a parametrix @xmath327 around the other edge points .
in the third and final transformation we put @xmath367 then @xmath114 is analytic on @xmath368 , where @xmath369 consists of the four circles @xmath370 , @xmath371 , the parts of @xmath337 outside the four disks , and the real intervals @xmath372 , @xmath373 , @xmath374 , see figure [ fig5 ] .
there are jump relations @xmath375 where @xmath376 from ( [ lp1b ] ) it follows that @xmath377 uniformly on the circles , and from ( [ st8 ] ) , ( [ st9 ] ) , ( [ ft4 ] ) and propositions [ laineq1 ] and [ laineq2 ] it follows that @xmath378 for some @xmath379 as @xmath5 , uniformly on the remaining parts of @xmath380 .
so we can conclude @xmath381 as @xmath196 , we have @xmath382 from ( [ tt2 ] ) , ( [ tt4 ] ) , ( [ tt5 ] ) and the fact that we can deform the contours in any desired direction , it follows that @xmath383 uniformly for @xmath384 , see @xcite . by cauchy s theorem
, we then also have @xmath385 and thus @xmath386 which is the form we will use below .
consider @xmath387 .
we may assume that the circles around the edge points are such that @xmath267 is outside of the four disks .
then ( [ tt1 ] ) shows that @xmath388 and it follows easily from ( [ tt7 ] ) and the fact that @xmath389 is real analytic in a neighorhood of @xmath267 that @xmath390 uniformly in @xmath7 .
thus by ( [ st14 ] ) we have that @xmath391 \\ & = e^{n(h(y)-h(x ) ) } \left [ \frac{\sin(n { { \operatorname{im\,}}}({\lambda}_{1+}(x ) - { \lambda}_{1+}(y)))}{\pi ( x - y ) } + o(1 ) \right ]
\\ \end{aligned}\ ] ] and the @xmath392 holds uniformly in @xmath7 . letting @xmath393 and noting that by ( [ wkb5a ] ) and ( [ defrho ] ) @xmath394 we obtain by lhopital s rule , @xmath395 which proves theorem [ maintheo1 ] if @xmath396 . the proof for @xmath397 is similar , and also follows because of symmetry . for @xmath398
, we have that @xmath399 decreases exponentially fast . for example
, for @xmath212 , we have that @xmath400 this follows from ( [ ft16 ] ) and the observation that that @xmath401 as @xmath393 if @xmath212 .
it is clear that ( [ pf3b ] ) implies @xmath402 we also have ( [ pf3c ] ) if @xmath267 is one of the edge points .
in fact , for an edge point @xmath267 it can be shown as in the proof of theorem [ maintheo3 ] that @xmath403 this completes the proof of theorem [ maintheo1 ] .
we give the proof for @xmath404 , the proof for @xmath405 being similar .
we let @xmath406 then we have ( [ pf2 ] ) , and so by the definition ( [ m16 ] ) of @xmath90 , @xmath407 because of ( [ pf2b ] ) we have by the mean value theorem , @xmath408 for some @xmath409 between @xmath267 and @xmath268 .
using ( [ pf4 ] ) we get @xmath410 and @xmath411 inserting ( [ pf7 ] ) into ( [ pf5 ] ) , we obtain @xmath412 which proves theorem [ maintheo2 ] .
we only give the proof of ( [ m18 ] ) , since the proof of ( [ m19 ] ) is similar .
we take @xmath97 as in ( [ rhoconst ] ) and we recall that @xmath413 .
take @xmath414 and let @xmath415 assume @xmath416 so that we can use formula ( [ st14 ] ) for @xmath20 .
then we have that @xmath267 belongs to @xmath339 , for @xmath7 large enough , so that by ( [ tt1 ] ) , ( [ tildep ] ) and ( [ ep7 ] ) @xmath417 and similarly for @xmath418
. then we get from ( [ st14 ] ) and ( [ m16 ] ) @xmath419 since @xmath420 , we have as @xmath5 , @xmath421 which implies that @xmath422 .
we use the second formula of ( [ ep5 ] ) to evaluate @xmath423 ( since @xmath424 ) , and it follows that @xmath425 similarly @xmath426 the factor @xmath427 comes from the inverse of @xmath428 , since @xmath429 by wronskian relations .
next , we recall that @xmath430 , so that by ( [ pf9 ] ) @xmath431 the explicit form ( [ ep8 ] ) for @xmath365 readily gives @xmath432 combining ( [ pf15 ] ) and ( [ pf16 ] ) , we have @xmath433 inserting ( [ pf13 ] ) , ( [ pf14 ] ) , and ( [ pf17 ] ) into ( [ pf11 ] ) , we obtain @xmath434 since @xmath435 , we have now completed the proof of ( [ m18 ] ) in case @xmath416 . for the remaining cases where
@xmath436 and/or @xmath437 , we have to realize that we have not specified the rescaled kernel @xmath438 for @xmath267 and/or @xmath268 outside of @xmath44 \cup [ z_2,z_1]$ ] , since in ( [ m20 ] ) @xmath84 is only defined there .
we define @xmath439 we will assume in the rest of the proof that @xmath440 and @xmath441 .
the case where @xmath442 and @xmath443 have opposite signs follows in a similar way : then we have to combine the calculations given below with the ones given above .
so let @xmath444 and let @xmath267 and @xmath268 be as in ( [ pf9 ] ) .
for the kernel @xmath82 we start from the expression ( [ ft16 ] ) in terms of @xmath243 . since @xmath440 , we have @xmath445 , and so we have by ( [ st6 ] ) , ( [ tt1 ] ) , ( [ tildep ] ) and ( [ ep7 ] ) , @xmath446 then @xmath447 as before , we have @xmath422 as @xmath5 .
now we use the first formula of ( [ ep5 ] ) to evaluate @xmath423 so that @xmath448 we have as in ( [ pf20 ] ) @xmath449 so that @xmath450 we have @xmath451 and as before we have ( [ pf17 ] ) . inserting ( [ pf21 ] ) and
( [ pf24 ] ) into ( [ pf16 ] ) and using the limits ( [ pf17 ] ) , ( [ pf22 ] ) and ( [ pf25 ] ) , we arrive at ( [ m18 ] ) in the case @xmath444 .
this completes the proof of theorem [ maintheo3 ] .
as noted in section [ sec2 ] , see also @xcite , we have that the @xmath36 entry of the solution @xmath22 of the rh problem ( [ m4])-([m5 ] ) is a monic polynomial @xmath452 of degree @xmath7 satisfying @xmath453 for the case @xmath454 , @xmath455 , this polynomial is called a multiple hermite polynomial @xcite .
the asymptotic analysis of the rh problem done in sections [ sec4][sec9 ] , also yields the strong asymptotics of the multiple hermite polynomials ( as @xmath5 with @xmath7 even and @xmath456 ) in every part of the complex plane .
we describe these asymptotics here .
recall that @xmath452 is the average characteristic polynomial of the random matrix ensemble ( [ m1 ] ) , see ( [ y11 ] ) .
we will partition the complex plane into 3 regions : 1 . outside of the lenses and of the disks @xmath325 , @xmath371 .
2 . inside of the lenses but outside of the disks .
3 . inside of the disks . we will derive the large @xmath7 asymptotics of the multiple hermite polynomials in these 3 regions . in this region
, we have by ( [ st6 ] ) and ( [ tt1 ] ) , @xmath457 hence by ( [ deft ] ) @xmath458 by restricting this matrix equation to the element @xmath36 we obtain that @xmath459 where @xmath460 and as @xmath31 , @xmath461 in the sum over @xmath462 in ( [ l3 ] ) the term @xmath463 dominates and we obtain because of ( [ mod17 ] ) that @xmath464 where for the square root we use the principal branch ( the one that is positive for @xmath465 ) , with two cuts , @xmath44 $ ] and @xmath45 $ ] . in this region
, we get from ( [ st2 ] ) , ( [ st4 ] ) and ( [ tt1 ] ) , @xmath466 where @xmath467 is the matrix on the right in ( [ st2 ] ) and ( [ st4 ] ) .
hence by ( [ deft ] ) @xmath468 consider @xmath149 the upper lens region on @xmath45 $ ] .
then @xmath469 hence @xmath470 and the first column of the matrix @xmath471 is @xmath472 see ( [ mod7 ] ) . by restricting equation ( [ l8 ] ) to the @xmath36 entry , and using ( [ mod17 ] ) and ( [ tt6 ] )
, we obtain that in the upper lens region on @xmath45 $ ] @xmath473 e^{-n{\lambda}_1(z)+nl_1}\\ { } & \qquad + \left[\frac{\xi_2 ^ 2(z)-a^2}{\sqrt{(\xi_2 ^ 2(z)-p^2)(\xi_2 ^ 2(z)-q^2 ) } } + o\left(\frac{1}{n}\right)\right ] e^{-n{\lambda}_2(z)+nl_1 } , \end{aligned}\ ] ] where @xmath474 in the same way we obtain that in the lower lens region on @xmath45 $ ] , @xmath475 e^{-n{\lambda}_1(z)+nl_1}\\ { } & \qquad -\left[\frac{\xi_2 ^ 2(z)-a^2}{\sqrt{(\xi_2 ^ 2(z)-p^2)(\xi_2 ^ 2(z)-q^2 ) } } + o\left(\frac{1}{n}\right)\right ] e^{-n{\lambda}_2(z)+nl_1}. \end{aligned}\ ] ] for @xmath476 real , @xmath477 $ ] , both ( [ l13 ] ) and ( [ l13a ] ) can be rewritten in the form @xmath478 + o\left(\frac{1}{n}\right)\right\}e^{-n\,{{\operatorname{re\,}}}{\lambda}_{1+}(x)+nl_1},\ ] ] where @xmath479 and @xmath480 by using equation ( [ defrho ] ) , we can also rewrite ( [ l15 ] ) in terms of the eigenvalue density function @xmath46 , @xmath481 + o\left(\frac{1}{n}\right)\right\}e^{-n\,{{\operatorname{re\,}}}{\lambda}_{1+}(x)+nl_1}.\ ] ] equation ( [ l15a ] ) clearly displays the oscillating behavior of @xmath452 on the interval @xmath482 $ ] .
it also shows that the zeros of @xmath483 are asymptotically distributed like @xmath484 , the limiting probability distribution of eigenvalues .
similar formulae can be derived on the interval @xmath485 $ ] .
consider the disk @xmath339 . in the regions i and iv , we have by ( [ tildep ] ) , ( [ tt1 ] ) and ( [ tt6 ] ) @xmath486 hence by ( [ deft ] ) , ( [ ep7 ] ) , and ( [ ep8 ] ) @xmath487 where @xmath488 by restricting equation ( [ l20 ] ) to the @xmath36 entry , and using the first expression of ( [ ep5 ] ) ( in region i ) or the fourth expression of ( [ ep5 ] ) ( in region iv ) to evaluate @xmath489 , and ( [ mod17 ] ) to evaluate @xmath322
, we obtain that @xmath490 { } & \qquad \qquad \left .
+ n^{-1/6}c(z){{\operatorname{ai}}}'(n^{2/3}{\beta}(z))(1+o(n^{-1}))\ , \right]\ , e^{-n{\alpha}(z)+nl_1 } , \end{aligned}\ ] ] where @xmath491 and @xmath492 the same asymptotics , ( [ l22 ] ) , holds in regions ii and iii as well .
thus , ( [ l22 ] ) holds in the full disk @xmath339 .
it may be verified that the functions @xmath493 and @xmath494 are analytic in @xmath339 .
this approach allows one to derive a formula similar to ( [ l22 ] ) in all the other disks @xmath325 as well .
from orthogonality equation ( [ m8 ] ) , we obtain that as @xmath31 , @xmath495 where for @xmath123 , @xmath496 is defined in ( [ m9 ] ) and @xmath497 this implies that @xmath498 where @xmath499 and @xmath500 .
set @xmath501 then by ( [ m15 ] ) , @xmath502 ( i.e. , no jump on the real line ) and as @xmath31 , @xmath503 where @xmath504 since @xmath505 is analytic on the complex plane , equation ( [ rec6 ] ) implies , by the liouville theorem , that @xmath506 where @xmath507 thus , we obtain the matrix recurrence equation , @xmath508 by restricting it to the element @xmath36 we obtain that @xmath509 and by restricting it to the element @xmath510 we obtain that @xmath511 similar to ( [ rec10 ] ) , we have another recurrence equation , @xmath512 where @xmath513 and @xmath514 by restricting ( [ rec13 ] ) to the elements @xmath36 and @xmath515 , we obtain the equations , @xmath516 and @xmath517
set @xmath518 it follows from ( [ m15 ] ) , that @xmath519 has no jump on the real axis , so that it is analytic on the complex plane . by differentiating ( [ rec3 ] )
we obtain that as @xmath31 , @xmath520 since @xmath519 is analytic , we obtain that @xmath521_{\rm pol}\\ { } & + \begin{pmatrix } 0 & 0 & 0 \\ 0 & -a & 0 \\ 0 & 0 & a \end{pmatrix } , \end{aligned}\ ] ] where @xmath522_{\rm pol}$ ] means the polynomial part of @xmath523 at infinity . from ( [ de1 ] )
we get the differential equation , @xmath524 and ( [ de4 ] ) reduces to @xmath525
equations ( [ rec10 ] ) , ( [ rec13 ] ) , ( [ de5 ] ) form a lax pair for multiple hermite polynomials .
their compatibility conditions are @xmath526 this gives the equations , @xmath527 since @xmath528 , we obtain that @xmath529 this proves the first equation in ( [ lax1 ] ) and equation ( [ lax2 ] ) .
similarly we obtain that @xmath530 and this proves the second equation in ( [ lax1 ] ) .
proposition [ lax ] is proved .
aptekarev , multiple orthogonal polynomials , j. comput . appl .
99 ( 1998 ) , 423447 .
aptekarev , a. branquinho , w. van assche , multiple orthogonal polynomials for classical weights , trans .
( 2003 ) , 38873914 .
a.i . aptekarev and h. stahl , asymptotics of hermite - pad polynomials , progress in approximation theory , ( a.a .
gonchar and e.b .
saff , eds ) , springer - verlag , new york , 1992 , pp .
p. bleher and a. its , semiclassical asymptotics of orthogonal polynomials , riemann - hilbert problem , and the universality in the matrix model , ann .
( 1999 ) , 185266 .
p. bleher and a. its , double scaling limit in the random matrix model .
the riemann - hilbert approach , commun .
pure appl .
56 ( 2003 ) , 433516 .
p.m. bleher and a.b.j .
kuijlaars , random matrices with external source and multiple orthogonal polynomials , int .
research notices 2004 , no 3 ( 2004 ) , 109129 .
e. brzin and s. hikami , spectral form factor in a random matrix theory , phys .
e 55 ( 1997 ) , 40674083 e. brzin and s. hikami , correlations of nearby levels induced by a random potential , nuclear physics b 479 ( 1996 ) , 697706 .
e. brzin and s. hikami , extension of level spacing universality , phys .
e 56 ( 1997 ) , 264269 .
e. brzin and s. hikami , universal singularity at the closure of a gap in a random matrix theory , phys .
e 57 ( 1998 ) 41404149 .
e. brzin and s. hikami , level spacing of random matrices in an external source , phys .
e 58 ( 1998 ) , 71767185 .
p. deift , orthogonal polynomials and random matrices : a riemann - hilbert approach , courant lecture notes in mathematics vol . 3 , amer . math . soc .
, providence r.i
p. deift , t. kriecherbauer , k.t - r mclaughlin , s. venakides , and x. zhou , uniform asymptotics of polynomials orthogonal with respect to varying exponential weights and applications to universality questions in random matrix theory , commun .
pure appl .
52 ( 1999 ) , 13351425 .
p. deift , t. kriecherbauer , k.t - r mclaughlin , s. venakides , and x. zhou , strong asymptotics of orthogonal polynomials with respect to exponential weights , commun .
pure appl .
math 52 ( 1999 ) , 14911552 .
p. deift and x. zhou , a steepest descent method for oscillatory riemann - hilbert problems .
asymptotics for the mkdv equation , ann . of math
. 137 ( 1993 ) , 295368 .
dyson , correlations between eigenvalues of a random matrix , commun .
phys 19 ( 1970 ) , 235250 .
a.a . gonchar and e.a .
rakhmanov , on the convergence of simultaneous pad approximants for systems of functions of markov type ( russian ) , trudy mat .
inst . steklov 157 ( 1981 ) , 3148 ; english transl . in proc .
steklov math .
inst . 3 ( 1983 ) , 3150 .
kuijlaars , riemann - hilbert analysis for orthogonal polynomials , in : orthogonal polynomials and special functions ( e. koelink en w. van assche eds ) , lecture notes in mathematics vol .
1817 , springer - verlag , 2003 , pp . 167210 .
a.b.j . kuijlaars and k. t - r mclaughlin , generic behavior of the density of states in random matrix theory and equilibrium problems in the presence of real analytic external fields , commun .
pure appl . math .
53 ( 2000 ) , 736785 a.b.j .
kuijlaars , w. van assche , and f. wielonsky , quadratic hermite - pad ' approximation to the exponential function : a riemann - hilbert approach , preprint math.ca/0302357 .
mehta , random matrices , 2nd edition , academic press , boston , 1991 .
nikishin and v.n .
sorokin , rational approximations and orthogonality , transl math .
92 , amer .
providence r.i . , 1991 .
j. nuttall , asymptotics of diagonal hermite - pad polynomials , j. approx .
theory 42 ( 1984 ) , 299386 .
pastur , the spectrum of random matrices ( russian ) , teoret .
( 1972 ) , 102112 .
w. van assche and e. coussement , some classical multiple orthogonal polynomials , j. comput . appl .
127 ( 2001 ) , 317347 .
w. van assche , j. geronimo , and a.b.j .
kuijlaars , riemann - hilbert problems for multiple orthogonal polynomials , special functions 2000 : current perspectives and future directions ( j. bustoz et al . , eds . ) , kluwer , dordrecht , 2001 , pp . 2359 .
p. zinn - justin , random hermitian matrices in an external field , nuclear physics b 497 ( 1997 ) , 725732 .
p. zinn - justin , universality of correlation functions of hermitian random matrices in an external field , comm .
( 1998 ) , 631650 . | we consider the random matrix ensemble with an external source @xmath0 defined on @xmath1 hermitian matrices , where @xmath2 is a diagonal matrix with only two eigenvalues @xmath3 of equal multiplicity .
for the case @xmath4 , we establish the universal behavior of local eigenvalue correlations in the limit @xmath5 , which is known from unitarily invariant random matrix models .
thus , local eigenvalue correlations are expressed in terms of the sine kernel in the bulk and in terms of the airy kernel at the edge of the spectrum .
we use a characterization of the associated multiple hermite polynomials by a @xmath6-matrix riemann - hilbert problem , and the deift / zhou steepest descent method to analyze the riemann - hilbert problem in the large @xmath7 limit . |
the cmb radiation in a wide frequency range has a planck spectrum with @xmath2 ( fixsen et al .
the last dramatic event that could have influenced a part of the spectrum was the recombination of the primeval hydrogen , since after it the residual ionization was very low and the radiation fields practically do nt interact with the non - relativistic matter any more . the properties of the post
recombination universe can be studied by observing the microwave background radiation .
the fluctuations in the cmb are supposed to be the precursors of the largest structures observed today . to understand the nature of the microwave background anisotropies it is necessary to have a correct picture of the recombination process itself .
the problem was first studied shortly after the discovery of the cmb by peebles ( 1968 ) and at the same time by zeldovich , kurt and sunyaev ( 1968 ) . since than several authors have adressed the problem , but no one could improve on the basic approximations used in these works .
however many details were worked out .
matsuda , sato & takeda ( 1971 ) examined the effects of the collision processes and found that they are negligible with respect to the radiation processes .
jones & wyse ( 1985 ) improved on the calculations to allow for the presence of non baryonic matter .
krolik ( 1989 , 1990 ) showed that two previously equally neglected scattering effects in the ly@xmath3 line almost completely cancel each other .
sasaki & takahara ( 1993 ) concluded that taking induced recombination into account changes the time history of the recombination at very low ionization grades , and lowers the residual ionization .
they however used an approximation for the recombination cross sections and did not discuss the divergence of the sum of the recombination rates .
dellantonio & rybicki ( 1993 ) followed the progress of recombination numerically , using a multilevel hydrogen atom model ; all angular momentum states were treated individually up to @xmath4 .
their recombination curve was similiar to that of jones & wyse ( 1985 ) .
in an other article rybicki & dellantonio ( 1994 ) calculated the time - dependence of the ly@xmath5 line profile in a homogenious expanding universe .
they found that the usual quasi - static approximation for the line shape is reasonable and does not cause substantial error in the solutions .
they also determined the recombination history for several cosmologies .
hu , scott , sugiyama & white ( 1995 ) discussed the effect of physical assumptions for cmb anisotropies and on the recombination process .
recombination , as it has been pointed out in the papers quoted above , sets in when all energy levels of hydrogen , except for the ground state are still in equilibrium with radiation .
when one hydrogen atom enters the ground state , a photon with energy greater than or equal to the energy of a @xmath6 transition is emitted .
photons , capable of exciting electrons in the ground state , are lost via two main routes : 1 . by cosmological
redshift of @xmath6 photons , 2 . by two
photon transition @xmath7 .
of these two competing processes the first is purely cosmological , the second is atomic .
their relative importance depends on the cosmology . by theoretical examination of the recombination process one
computes the time dependence of the ionization rate , the residual ionization , the position and width of the last scattering layer and the distortion of the cmb spectrum .
though the amount of the residual ionization is important for the further evolution of the universe ( peebles 1993 , lepp & shull 1984 ) , it can not be measured directly .
a possible observable consequence of the hydrogen recombination in the early universe is the distortion of the microwave background radiation spectrum .
it was first calculated for the flat cosmological model by peebles ( 1968 ) . in this work
we concentrate on the determination of this distortion for different cosmological models .
recently there has been a considerable observational and theoretical activity to determine the spectrum of the cosmic background radiation and carrying out galaxy counts in the far infrared / submillimeter range ( puget et al .
1996 , schlegel et al .
1997 , guideroni et al .
1997 ) . because the hydrogen recombination changes the cbm spectrum in the same spectral range a detailed exact recalculation of the frequency distribution of the recombination photons is important . in this paper
we calculate the recombination cross sections exactly by using a continuous , physical cut off for the highly excited states of hydrogen , take into account the induced recombination and explain , why the time dependence of the recombination process is so hardly effected by different techniques and by different effective ionization curves .
the outline of the paper is as follows . in 2 we discuss the recombination process and derive the recombination equation .
we give a new parametrization of the effective recombination coefficients , taking into account the induced recombination as well . in the third section
we solve the recombination equation .
the spectrum of the recombination photons and its dependence on the cosmological parameters is given in the fourth , and in the fifth closing section we discuss our results .
the number of recombinations in unit time can be calculated from the recombination equation as a function of the density of free electrons @xmath8 , temperature @xmath9 , and cosmological , and atomic constants . in this section
we derive this equation following peebles ( 1968 ) . at the beginning of the hydrogen recombination
the helium is already completely recombined .
the mass fraction of the helium is 25 % of the total baryonic mass . as was remarked by novikov and zeldovich ( 1967 ) the direct recombinations to the ground state
are inhibited , while the new born energetic photons ionize again almost immediately when there are already some hydrogen atoms . in the following we neglect completely the direct recombinations to the ground state .
the states with principal quantum number @xmath10 play a key role in the recombination processes .
first we calculate the rate of recombinations to @xmath11 for given free electron number density @xmath12 , temperature t and @xmath13 . here
@xmath14 is the density of atoms in the state with the principal quantum number @xmath15 and angular momentum quantum number @xmath16 .
second , we determine the net number of @xmath17 transitions in unit time by given @xmath18 and @xmath19 .
the two rates must be equal , so we can eliminate @xmath13 from the calculations .
the binding energy of the @xmath10 state is b@xmath20 . during the recombination process
there are a large number of photons with energy less than @xmath21 , therefore the excited states of the hydrogen atoms are in thermodynamical equilibrium above the second level , i.e. @xmath22 in the case of gaseous nebulae the mean free path of low energetic photons is larger than the dimensions of the ionized region , the system is far from being in equilibrium above the @xmath23 level in contrast to the recombining universe .
the partition sum @xmath24 is divergent .
as analysed by hummer and mihalas ( 1988 ) a number of effects limit the range of summation . in the considered
temperature and density range the action of free protons turns out to be the most important , the action of neutral atoms are much smaller .
the free protons destroy the state @xmath25 with a probability @xmath26 , where @xmath27\,$ ] , with @xmath28 ( n@xmath29 in @xmath30 ) .
this means that the recombined electrons become unbound with @xmath31 probability , before they would begin to move towards the state , corresponding to thermal equilibrium .
the highly excited ( @xmath15 larger than @xmath32 ) states are practically completely destroyed .
the occupation numbers of the excited states in thermal equilibrium are : @xmath33 owing to this fact , the number of electrons in bound states is finite , and thermal equilibrium between the @xmath34 bound states and the continuum is possible .
this approximation is good down to z @xmath35 300 .
the @xmath10 levels freeze out between the redshifts 300 and 250 .
the ground state of the hydrogen atom is about 10.2 ev deeper than the @xmath36 level .
consequently , it s occupation is greater than the occupation of the excited states together : @xmath37 .
this means , that the density of free electrons plus the density of hydrogen atoms in ground state can be taken as equal to the total proton number : @xmath38 .
the total baryon number ( determined by @xmath39 ) is the sum of p and the number of baryons in the helium nuclei . the number of recombinations to the level @xmath40 and the number of ionizations from @xmath40 in unit time are @xmath41 where @xmath42 and @xmath43 are the recombination and ionization coefficients , @xmath44 is the density of free protons .
the usual definition of these quantities are @xmath45 here @xmath46 and @xmath47 are the velocities .
the averages must be taken over the maxwell
distribution for electrons and the planck distribution for photons .
the ionization coefficients are : @xmath48 the relation between the recombination and ionization cross sections is : @xmath49 , where @xmath50 and @xmath51 are the photon and electron momenta . by using , instead of the electron velocity , the energy of the recombination photon in the maxwell distribution and expressing the recombination cross section through the ionization cross section , one arrives at @xmath52 where @xmath53 is the electron mass .
the integrals in @xmath54 and @xmath55 are not the same .
but , due to the principle of detailed balance ( milne 1924 ) the recombinations and ionizations in thermodynamic equilibrium must exactly cancel each other .
if we let contributions from the induced recombinations modify the recombination coefficients ( e.g. mihalas 1984 , sasaki & takahara 1993 ) instead of ( [ rcoefo ] ) we get @xmath56 the term @xmath57 under the integral in the denominator comes from the induced recombinations . without it the principle of detailed balance would not be fulfilled .
summing up over all @xmath58 levels and making use of equation ( [ nnl ] ) the total rate of recombinations is : @xmath59 the total rate of ionizations from the excited states of hydrogen is : @xmath60 the net number of recombinations to excited levels in unit time and unit volume is the difference of these two expressions . the number density of the electrons changes , not only because of the recombination process , but also in consequence to the universal expansion .
the simplest way to take this effect into account is to write on equation for the ionization rate , i.e. the ratio of free electrons to the number of free protons plus hydrogen atoms : @xmath61.\ ] ] each @xmath63 transition liberates a @xmath64 photon , which can excite the ground state , and there are always enough low energy photons to ionize from an excited state .
the transition to the ground state is final , when the number of these @xmath65 photons is also diminished .
photons can escape from the @xmath64 line either by redshift or by two
photon 2s @xmath66 1s transitions . in a stationary state
the numbers of emitted and absorbed @xmath67 photons are equal .
consequently , the number of photons per mode in the @xmath64 resonance line is : @xmath68 because of the general expansion of the universe a frequency @xmath69 is shifted in unit time by @xmath70 where h is the hubble
the number of recombination photons removed from the line by redshift in unit time is : @xmath71 on the other hand , the rate of two photon decay in the 2s states has been calculated by spitzer & greenstein ( 1951 ) .
the number of net decays in unit time is @xmath72 with @xmath73 . in thermal equilibrium
the numbers of two photon decays and the two
photon @xmath74 excitations are equal . at this point
we have four unknown quantities : @xmath75 and four equations ( [ rec2s ] ) , ( [ gamm ] ) , ( [ rsr ] ) and ( [ rph2 ] ) ; so we are left with @xmath76 \nonumber \\ & & \cdot { 8 \pi \nu^3_{12 } c^{-3 } h + a_{2s,1s } ( p - n_e ) \over 8 \pi \nu^3_{12 } c^{-3 } h + ( a_{2s,1s } + \beta ) ( p - n_e ) } .\end{aligned}\ ] ] the time dependence of @xmath77 are calculated from the friedmann equation with the density parameter @xmath78 , the baryon density parameter @xmath79 and the hubble constant @xmath80 ( @xmath81 , the cosmological constant ) . to solve the equation ( [ receq ] )
we need the recombination coefficient @xmath82 , wich is parametrized as a function of the temperature in several ways .
peebles ( 1968 ) used @xmath83 on the basis of boardman s ( 1964 ) data for the seven lowest states .
zeldovich , kurt and sunyaev ( 1968 ) used @xmath84 .
peebles ( 1993 ) using osterbrock s ( 1989 ) data got the parametrization : @xmath85 . recently hummer ( 1994 ) tabulated very accurately the recombination rates .
his results confirm the parametrization given by pquignot et al .
( 1991 ) in the form of @xmath86 sasaki & takahara ( 1993 ) had used the asymptotic form for the recombination cross sections which , also valid for large principal quantum numbers @xmath15 , was less than 20% accurate for the important low lying levels .
when calculating the recombination coefficients without taking into account the induced recombination , the @xmath87 falls out of the denominator of the integral in ( [ rcoef ] ) . in this case
the @xmath88 series converges as @xmath89 , so one can neglect the contribution of the terms with large @xmath15 .
if induced recombination is taken into account , the denominator under the integral in ( [ rcoef ] ) goes to zero for large @xmath15 , and the series of the @xmath90 diverges as @xmath91 . here we calculate the total recombination coefficient from ( [ recr ] ) , wherein the factor @xmath92 , the probability that the states with principal quantum number @xmath15 are not destroyed , makes the sum finite . the probability that a state is destroyed depends on the baryon density of the universe .
consequently , the effective recombination coefficient also density dependent .
our procedure is then the following .
we calculate the first several hundred @xmath42 s using the exact results given by storey & hummer ( 1991 ) . for large @xmath15 ( @xmath93 ) , when the recombination coefficients have already reach their asymptotic values ( the difference between the exact and asymptotic value @xmath94 )
, we use the @xmath95 s given e.g.in rybicki & lightman ( 1979 ) .
the density dependence comes in through the @xmath96 probabilities .
the effective recombination coefficient is parametrized as a function of the temperature , @xmath97 , and the electron number density , @xmath98 , in the form @xmath99 for a given temperature @xmath12 and @xmath100 depend on the cosmological parameters .
however this dependence is mild , not larger than 10 % for the usual range of parameters .
we compute the @xmath101 function together with @xmath12 by solving ( [ receq ] ) .
the solution of this equation for n@xmath29 and @xmath102 will be given in the next section .
for the sake of clarity we present our recombination rate already here ( fig .
the form of the @xmath103 function is not far from the usual @xmath104 function , but its amplitude is about four times larger than it would be without taking the induced recombination into account .
furthermore we compare our @xmath105 with ( [ alpeq ] ) , and with sasaki & takahara ( 1993 ) , whereby our function is somewhat steeper than that of these authors .
crosses : present work ( for @xmath106 h = 1 ) ; continuous line : @xmath107 fit ; diamonds : sasaki & takahara 1993 ; points : eq .
( [ alpeq ] ) ; dashed line : peebles 1993 .
we solved the equation ( [ receq ] ) for recombination numerically using equations ( [ alpar ] ) and ( [ ionr ] ) .
the results with and without taking into account the induced recombination are compared .
llll @xmath108 & 1 & 1 & 0.5 + @xmath39 & 1 & 0.06 & 0.1 + @xmath109 & 1 & 1 & 1 + @xmath110&@xmath111 & @xmath112 & @xmath113 + @xmath114 & @xmath115 & @xmath116 & @xmath117 + @xmath118 & 2.5567 & 2.5662 & 2.5784 + + + + lll @xmath108 & 0.1 & 0.3 + @xmath39 & 0.1 & 0.01 + @xmath109 & 0.75 & 0.5 + @xmath110&@xmath119 & @xmath120 + @xmath114&@xmath121 & @xmath122 + @xmath118 & 2.6297 & 2.6486 + at @xmath123 200 @xmath124 formation becomes possible .
since free electrons and protons serve as catalysts for the formation of molecular hydrogen ( peebles 1993 ) , the value of the ionized fraction at this epoch is a very important . at about @xmath123
500 @xmath125 becomes much smaller than a@xmath126 .
it is independent of the cosmology , because both @xmath125 and a@xmath127 are atomic quantities . in this case the fraction in equation ( [ receq ] ) is one .
the second term in the square bracket is also small , so instead of the equation ( [ receq ] ) for @xmath128 we can write @xmath129 when the induced recombinations are not taken into account the density dependent term does not appear and the residual ionization scales with @xmath130 .
the logarithmic term contains another combination of the cosmological parameters , therefore the scaling law is mildly violated .
the residual ionizations at @xmath131 are shown in table 1 . when the induced recombination is taken into account the residual ionization is reduced by a factor of about 2.6 .
small deviations from this factor are due to the violation of the scaling law .
the total number of ions is proportional to @xmath132 , so it is almost independent of the baryon density .
we refined the previous treatment ( sasaki & takahara 1993 ) of the induced recombination by using the exact recombination cross sections and applying a physical cutoff in the summation on the hydrogen states .
this causes a 10 - 20 % change in the residual ionization . the recombination history for different sets of comological parameters is shown in fig . 2 .
the quotient of the fractional ionization at different cosmologies grows with time and approximately follows the x @xmath133 scaling law .
we compare our result with dellantonio and rybicki ( 1993 ) , who did not take induced recombinations into account . for corresponding parameters our recombination curve runs significantly under their curve , because the last phase of the recombination processes is accelerated by the induced recombinations .
a)= @xmath134 = 1,= @xmath39 = 1,= h = 1 + b ) @xmath134 = 0.1 , @xmath39 = 0.1 , h = 1 + c ) @xmath134 = 1 , @xmath39 = 0.01 , h = 1 + d)@xmath134 = 0.3 , @xmath39 = 0.01 , h = 0.5 [ fig3 ] in fig . 3 we show the ratio of fractional ionization calculated with ( @xmath135 ) and without ( @xmath136 taking induced recombination into account . in the first part of the process the induced free bound transitions and the numerical value of the effective recombination coefficient have little effect on the course of the process , because the depopulation of the @xmath36 states is much slower than the recombination to the excited states , the thermal equilibrium corresponding to the occupation of @xmath36 is always maintained . down to @xmath137
the @xmath138 ratio is one . between @xmath137 and @xmath139 the change owing to induced recombinations
depends on the cosmology .
later , when @xmath140 , especially when the baryon density @xmath79 is small , the depopulation of @xmath36 states is faster than the recombination to the excited states and this last process determines the net recombination rate . at this stage the induced recombinations play an important role . as can be seen in fig . 3 , the @xmath141 ratio depends hardly on the cosmological parameters . from the @xmath142 function
one can estimate the probability that a cmb photon has not been scattered since a given redshift @xmath143 . instead of this probability , usually its differential is calculated .
this quantity determines the probability density that the radiation was last scattered at @xmath143 and is expressed as @xmath144 . here
@xmath145 is the thomson scattering optical depth : @xmath146 , and @xmath147 is the thomson cross section . the parameters , determining the position and width of this layer are given in table 2 .
these results are in reasonable agreement with those of white , scott & silk ( 1994 ) .
this occurs because the last thomson scattering happens early on , when the induced recombinations do nt play any role .
llllll @xmath108 & 1 & 1 & 1 & 0.1 & 0.1 + @xmath39 & 1 & 0.1 & 0.01 & 0.1 & 0.1 + @xmath109 & 1 & 1 & 1 & 1 & 0.5 + @xmath148 & @xmath149 & @xmath150&@xmath151&@xmath152 & @xmath153 + @xmath154&@xmath155 & @xmath156&@xmath157&@xmath158 & @xmath159 +
the evolution of the photon spectrum during recombination can be calculated from the continuity equation in frequency space ( peebles 1968 ) : @xmath160 @xmath161 is the number density of photons with frequency @xmath162 , and @xmath163 is the net rate of production of photons per unit volume and unit frequency interval .
the planck
spectrum fulfills this equation without the source term , so ( [ photeq ] ) is also true for @xmath164 . here
we introduce new dependent variables with the definition @xmath165 and define the independent variables @xmath166 as @xmath167 instead of the equation ( [ photeq ] ) we have @xmath168 in the absence of sources the spectrum moves to the left ( with `` velocity '' 1 ) in the @xmath169 plane .
when the right hand side of ( [ lapleq ] ) differs from zero , the solution can be written in integral form as : @xmath170 with @xmath171 . from the solution of equation ( [ receq ] )
we know the time dependence of the free electron density @xmath172 . using the equations
( [ rph2 ] ) and ( [ rsr ] ) the net two photon transition rate and the @xmath173 redshift rate can be calculated .
the spectrum of the photons emerging from the @xmath174 transition can be taken from spitzer & greenstein ( 1951 ) .
the appropriately normalized two photon spectrum in terms of the variable @xmath175 is : @xmath176 the @xmath177 . for our purposes
the frequency distribution of the @xmath178 photons can be approximated by a delta function on the @xmath179 axis , so @xmath180 with @xmath181 , the ratio of @xmath64 redshift rate to the two photon decay rate can be calculated from equations ( [ rsr ] ) and ( [ rph2 ] ) .
this ratio depends on the cosmology . for the @xmath182 transition neglecting the @xmath183 in the denominator of equation ( [ rsr ] ) and using the numerical values , one arrives at @xmath184 the redshift of the new
born photons is given by equation ( [ lapleq ] ) in terms of the variables @xmath175 and @xmath185 .
the spectrum at different redshifts is shown in figure 4 ( @xmath186 , @xmath187 , @xmath188 ) and figure 5 ( @xmath189 , @xmath190 , @xmath188 ) . when the baryon density is low , the 2s @xmath191 1s transition plays a minor role .
these results disagree with dellantonio & rybicki ( 1993 ) , who state that the energy distribution of the photons emitted by the @xmath7 transition rather strongly peaked at @xmath192 and the @xmath193 transitions give no more than 1% difference in the free electron densities and the line strength .
our results ( fig . 4 and fig .
5 ) show that the spectrum of @xmath193 photons is broad ( see also spitzer & greenstein 1951 ) .
though for certain cosmologies the contribution of these photons is small , but as it is demonstrated in figure 6 , for other set of parameters this contribution is quite considerable . in the models with high @xmath108 and low @xmath39 the two photon transitions are negligible , but for @xmath194 they are more important , and for the flat , high @xmath39 model they are dominant . in spite of
the above mentioned disagreement our recombination history curve ( fig .
2 ) agrees well with that of dellantonio & rybicki ( 1993 ) , because by computing this function they took into account also the 2s @xmath66 1s transitions .
figure 7 shows a part of the cmb radiation spectrum for different cosmological parameters , where the distortion due to recombinations is the most important .
the hydrogen recombination begins when in the background radiation there are less energetic photons than hydrogen atoms .
consequently , the spectrum of the photons issuing from the recombination has a maximum near the planck - curve .
the other maximum in this spectrum corresponds to the two photon transitions , is longwards from the first maximum at @xmath195 ev , and is under the planck curve .
the @xmath196 photons influences the distortion in two ways .
first , the short wavelength part of their spectrum is above the planck - curve and give a direct contribution to the distortion ( figure 6 ) .
second , the number of recombinations are given by the number of hydrogen atoms .
if a considerable part of recombinations happens by two photon decays , there are less redshifted ly@xmath5 photons , the amplitude and shape of their spectrum changes . for given @xmath197 and h there are more photons above the planck
curve when both @xmath198 and r are large .
when @xmath199 and h are given , the spectrum with the larger @xmath79 lies above all the others .
for a fixed value of @xmath108 and @xmath79 the spectrum with larger @xmath109 is larger . in figure 7
we compare our spectrum with that of dellantonio & rybicki ( 1993 ) . in the 120 @xmath200 m @xmath201 170 @xmath200 m range
there are about a factor of 23 more photons as seen by comparing our curve with the diamonds .
that may be caused by the different technique in computing the recombination process .
it seems us somewhat arbitrary to handel the 10 lowest level in a different way as the others .
burgess ( 1958 ) used this method to compute the recombination spectrum in nebulae . in the case of nebulae this method gives good results , because there is no thermal equilibrium above the n = 2 states , the electrons recombine on the low lying states and cascade down .
background photons stimulate recombination processes and that give an important contribution to the effective recombination coefficient .
the resulting recombination coefficient is about a factor of four larger than calculated without the induced recombinations ( figure 1 ) and depends , besides the temperature , on the free electron ( proton ) density as well .
in contrast to the recombining universe in a gaseous nebulae with the same temperature and free electron density the recombination coefficient is given by ( [ alpeq ] ) , because there are no photons that should stimulate recombination on excited states .
the photons , except the ly@xmath5 photons , emerging from recombination in the nebulae leave the nebula without interaction .
for any given principal quantum number the @xmath202 recombination coefficients depend considerably on the angular momentum quantum number @xmath16 .
there is a peak around @xmath203 and @xmath202 vanishes when @xmath16 is high .
transitions between states with high @xmath16 s are slow , which could influence the relative occupation of s and p states . since this last effect is probably small ( see hu et al .
1995 ) , we have completely neglected it here .
in the main part of the recombination process the free bound transitions are more rapid than the depopulation of the @xmath36 state and the thermodynamic equilibrium is maintained by reionization .
consequently , the course of the recombination process does not depend on the details of the @xmath204 function .
because of that , authors , using quite different effective recombination coefficients , come up with similiar time dependence for recombination .
when , however , the free electron density is already low , the depopulation of the @xmath36 states is more rapid than the recombinations of the @xmath205 states and the speed of the recombination process is determined by the free bound transitions of the excited states and the induced recombinations are important . as a consequence ,
our ionization curve ( figure 2 ) is for small ionization grades steeper than that of dellantonio & rybicki ( 1993 ) , who did not take this effect into account .
the last scattering of the cmb photons happens , with high probability , in the first phase of the recombination , when the number of free bound transitions to the excited states is large .
the induced recombinations and the numerical value of the effective recombination coefficient do not influence the position of the last scattering layer .
our results agree with those of white , scott & silk ( 1994 ) . on the other hand
the residual ionization is a quantity determined by the details of the recombinations of the excited states at low n@xmath29 . at @xmath206 ,
whether induced recombination is taken into account or not , the fractional ionization is proportional to @xmath207 . however , because of the induced recombinations , the residual ionization is reduced by a factor of about 2.6 .
the exact value of this factor depends on the cosmological parameters .
a consequence of the low residual ionization and the higher number of photons in the tail of the cmb spectrum is that the final abundance of molecular hydrogen will be much smaller than that estimated by lepp & shull ( 1983 ) .
the hydrogen recombination gives contribution to the cmb spectrum in the @xmath208 190 @xmath200 m range .
the amplitude and form of this distortion depends on the number of photons issuing from the recombination , and their distribution between the two decay way , the @xmath7 two photon mode and the redshifted ly@xmath5 mode .
the induced recombinations only in the last phase influence the recombination process , when the ionization grade already very low .
this gives a very small contribution to the distortion of the spectrum .
the contribution of helium recombination to the photon spectrum was discussed by lyubarsky & sunyaev ( 1983 ) and by fahr & loch ( 1991 ) . according to their results the recombination of he sets in at a redshift of about four times larger than that for hydrogen recombination .
but the photon energy is also four times larger and the photons emerging from helium recombination could also distort the cmb spectrum in the same frequency range as the h recombination .
however , as it was pointed out by peebles ( 1995 ) , at those redshifts where he ii recombines there is already a trace of recombined hydrogen , which can absorb some of the photons created during helium recombination .
thus direct recombination into the ground state is possible for helium and the saha formula is a good approximation to describe the helium recombination history .
the optical depth due to ionization between z@xmath209 and z@xmath210 is @xmath211 the redshift of the helium recombination is z@xmath209 , the cross section for hydrogen ionization is @xmath212 , and the number density of hydrogen atoms , as computed from the saha formula , is n@xmath213 .
the distance to the radiation source at the moment of emission , r(z@xmath209 ) , is calculated from the friedmann equation . the integral reaches 1 at about z@xmath214 .
consequently , in contrast to the results mentioned above , helium recombination does not leave any trace on the cmb spectrum . as is discussed in hu et al .
( 1995 ) , the difference between the temperature of the photons and the kinetic temperature of the electrons has an effect below z @xmath215 200 .
we follow the recombination process only down to z = 220 , so this effect is neglected .
the possible @xmath216 and the @xmath217 two photon transitions could also have some importance .
the number of bound electrons in the @xmath218 states is of the same order of magnitude that for n@xmath219 .
the two photon decay probabilities for these states are smaller than for @xmath13 .
moreover some of these photons will have more energy than b@xmath210
b@xmath209 = 10.2 ev .
however , the contribution of these transitions is not entirely negligible , they can change the balance between the two routes , redshift and two photon decay , for the elimination of energetic photons and change the spectrum of the recombination photons .
this correction will be discussed in detail elsewhere . if there were some energy input prior to the hydrogen recombination and a `` y - distortion '' of the spectrum ( zeldovich & sunyaev 1969 ) it would be easier to observe the consequences of h recombination ( lyubarsky & sunyaev 1983 ) .
however , the observations permit a very low upper limit for the y - distortion ( mather et al . 1994 ,
fixsen et al . 1996 ) .
the distortion seems to occur at wavelengths where other sources give considerable contributions . by doing the calculations we had hoped that the zodiacal forground emission , the dust and molecular emission from the interstellar medium would be reliably modeled , subtracted and the calculated spectral distortion would be measurable .
recently that substraction has been done ( schlegel , finkbeiner & davis 1997 ) and it turned out that in the considered wavelength range the measured upper limit of the background is at least two order of magnitude higher as the spectrum calculated here . as it was pointed out by guideroni et al .
( 1997 ) , it could originate from early starlight scattered by dust .
boardman , w. j. , 1964 , apjs 9 , 185 burgess , a. , 1958 , mnrs , 18 , 477 dellantonio , j. p. rybicki , g. b. , 1993 , in observational cosmology , asf conference series vol .
51 fixsen , d. j. , cheng , e. s. , gales , j. m. , mather , j. c. , shafer , r. a. & wright , e. l. , 1996 , apj 473 , 576 guideroni , b. , bouchet , f. r. , puget , j .- l . ,
lagache , g. & hivon , e. , 1997 , nature 390 , 257 hu , w. , scott , d. , sugiyama , n. & white , m. , 1995 , phys . rev .
d 52 , 5498 hummer , d. g. , 1994 , mnrs 268 , 109 hummer , d. g. & mihalas , d. , 1988 , apj 331 , 794 fahr , h. j. & loch , r. , 1991 , a&a 246 , 1 krolik , j. h. , 1989 , apj 338 , 594 krolik , j. h. , 1990 , apj 353 , 21 jones , b. j. t. & wyse , r. f. g. , 1985 , a&a 149 , 144 lepp , s. & shull , j. m. , 1984 , apj 280 , 465 lyubarski , y. e. & sunyaev , r. a. , 1983 , ap&ss 123 , 171 matsuda , t . ,
sato , h. & takeda , h. , 1971 , progr .
46 , 216 mihalas , d. 1984 , foundation of radiation hydrodynamics , oxford university press , n.y .
milne e. a. , 1924 , phil . mag .
47 , 209 novikov , i. d. & zeldovich , ya .
b. , 1967 , the structure and evolution of the universe , the university of chicago press , chicago and london pquignot , d. , petitjean , p. & boisson , c.,1991 , a&a 251 , 680 puget , j .- l .
, abergel , a. , bernard , j .-
, bulanger f. , burton , w. b. , dsert , f .,- x .
& hartmann , d. , 1996 , a&a 308 , l5 rybicki , g. b. & dellantonio , i. , p. , 1994
, apj 427 , 603 rybicki , g. b. & lightman , a. p. , 1979
, radiative processes in astrophysics , john wiley new york sasaki , s. & takahara , f. , 1993 , pasj 45 , 655 schlegel , d.j . ,
finkbeiner , d.p .
& davis , m. , 1997 , baas 191 , 87.04 ( astro - ph/9710327 ) spitzer , l. a. & greenstein , j. l. , 1951 , apj 114 , 407 storey , p. j. & hummer , d. g. , 1991 , computer physics communications 66 , 129 white , m. , scott , d. & silk , j. , 1994 , annu .
astron . astrophys . | we solve the recombination equation by taking into account the induced recombinations and a physical cut off in the hydrogen spectrum .
the effective recombination coefficient is parametrized as a function of temperature and free electron density and is about a factor of four larger than without the induced recombination .
this accelerates the last stage of the recombination processes and diminishes the residual ionization by a factor of about 2.6 .
the number and energy distribution of photons issuing from the hydrogen recombination are calculated .
the distortion of the cosmic microwave background ( cmb ) spectrum depends strongly on the cosmological parameters @xmath0 and differs essentially from the planck - spectrum for wavelengths @xmath1 . |
the rapid proliferation of smart mobile devices has triggered an unprecedented growth of the global mobile data traffic .
hetnets have been proposed as an effective way to meet the dramatic traffic growth by deploying short range small - bss together with traditional macro - bss , to provide better time or frequency reuse@xcite . however
, this approach imposes a significant challenge of providing expensive high - speed backhaul links for connecting all the small - bss to the core network@xcite .
caching at small - bss is a promising approach to alleviate the backhaul capacity requirement in hetnets@xcite .
many existing works have focused on optimal cache placement at small - bss , which is of critical importance in cache - enabled hetnets .
for example , in @xcite and @xcite , the authors consider the optimal content placement at small - bss to minimize the expected downloading time for files in a single macro - cell with multiple small - cells .
file requests which can not be satisfied locally at a small - bs are served by the macro - bs .
the optimization problems in @xcite and @xcite are np - hard , and low - complexity solutions are proposed . in @xcite ,
the authors propose a caching design based on file splitting and mds encoding in a single macro - cell with multiple small - cells .
file requests which can not be satisfied locally at a small - bs are served by the macro - bs , and backhaul rate analysis and optimization are considered .
note that the focuses of @xcite are on performance optimization of caching design . in @xcite ,
the authors consider caching the most popular files at each small - bs in large - scale cache - enabled small - cell networks or hetnets , with backhaul constraints .
the service rates of uncached files are limited by the backhaul capacity . in @xcite ,
the authors propose a partion - based combined caching design in a large - scale cluster - centric small - cell network , without considering backhaul constraints . in @xcite ,
the authors consider two caching designs , i.e. , caching the most popular files and random caching of a uniform distribution , at small - bss in a large - scale cache - enabled hetnet , without backhaul constraints .
file requests which can not be satisfied at a small - bs are served by macro - bss . in @xcite ,
the authors consider random caching of a uniform distribution in a large - scale cache - enabled small - cell network , without backhaul constraints , assuming that content requests follow a uniform distribution .
note that the focuses of @xcite are on performance analysis of caching designs .
on the other hand , enabling multicast service at bss in hetnets is an efficient way to deliver popular contents to multiple requesters simultaneously , by effectively utilizing the broadcast nature of the wireless medium@xcite . in @xcite and @xcite ,
the authors consider a single macro - cell with multiple small - cells with backhaul costs .
specifically , in @xcite , the optimization of caching and multicasting , which is np - hard , is considered , and a simplified solution with approximation guarantee is proposed . in @xcite ,
the optimization of dynamic multicast scheduling for a given content placement , which is a dynamic programming problem , is considered , and a low - complexity optimal numerical solution is obtained .
the network models considered in @xcite do not capture the stochastic natures of channel fading and geographic locations of bss and users .
the network models considered in @xcite are more realistic and can reflect the stochastic natures of signal and interference .
however , the simple identical caching design considered in @xcite does not provide spatial file diversity ; the combined caching design in @xcite does not reflect the popularity differences of files in each of the three categories ; and the random caching design of a uniform distribution in @xcite can not make use of popularity information .
hence , the caching designs in @xcite may not lead to good network performance . on the other hand , @xcite consider analysis and optimization of caching in large - scale cache - enabled single - tier networks .
specifically , @xcite considers random caching at bss , and analyze and optimize the hit probability .
reference @xcite considers random caching with contents being stored at each bs in an i.i.d .
manner , and analyzes the minimum offloading loss . in @xcite ,
the authors study the expected costs of obtaining a complete content under random uncoded caching and coded caching strategies , which are designed only for different pieces of a single content .
in @xcite , the authors consider analysis and optimization of joint caching and multicasting .
however , the proposed caching and multicasting designs in @xcite may not be applicable to hetnets with backhaul constraints . in summary , to facilitate designs of practical cache - enabled hetnets for massive content dissemination , further studies are required to understand the following key questions .
@xmath0 how do physical layer and content - related parameters fundamentally affect performance of cache - enabled hetnets ?
@xmath0 how can caching and multicasting jointly and optimally assist massive content dissemination in cache - enabled hetnets ? in this paper , we consider the analysis and optimization of joint caching and multicasting to improve the efficiency of massive content dissemination in a large - scale cache - enabled hetnet with backhaul constraints .
our main contributions are summarized below .
@xmath0 first , we propose a hybrid caching design with certain design parameters , consisting of identical caching in the macro - tier and random caching in the pico - tier , which can provide spatial file diversity .
we propose a corresponding multicasting design for efficient content dissemination by exploiting broadcast nature of the wireless medium .
@xmath0 then , by carefully handling different types of interferers and adopting appropriate approximations , we derive tractable expressions for the successful transmission probability in the general region and the asymptotic region , utilizing tools from stochastic geometry .
these expressions reveal the impacts of physical layer and content - related parameters on the successful transmission probability .
@xmath0 next , we consider the successful transmission probability maximization by optimizing the design parameters , which is a very challenging mixed discrete - continuous optimization problem .
we propose a two - step optimization framework to obtain a near optimal solution with superior performance and manageable complexity .
specifically , we first characterize the structural properties of the asymptotically optimal solutions . then , based on these properties , we obtain the near optimal solution , which achieves better performance in the general region than any asymptotically optimal solution , under a mild condition .
@xmath0 finally , by numerical simulations , we show that the near optimal solution achieves a significant gain in successful transmission probability over some baseline schemes .
we consider a two - tier hetnet where a macro - cell tier is overlaid with a pico - cell tier , as shown in fig .
[ fig : system ] . the locations of the macro - bss and the pico - bss
are spatially distributed as two independent homogeneous poisson point processes ( ppps ) @xmath1 and @xmath2 with densities @xmath3 and @xmath4 , respectively , where @xmath5 .
the locations of the users are also distributed as an independent homogeneous ppp @xmath6 with density @xmath7 .
we refer to the macro - cell tier and the pico - cell tier as the @xmath8st tier and the @xmath9nd tier , respectively .
consider the downlink scenario .
each bs in the @xmath10th tier has one transmit antenna with transmission power @xmath11 ( @xmath12 ) , where @xmath13 .
each user has one receive antenna .
all bss are operating on the same frequency band of total bandwidth @xmath14 ( hz ) .
consider a discrete - time system with time being slotted and study one slot of the network .
we consider both large - scale fading and small - scale fading . due to
large - scale fading , a transmitted signal from the @xmath10th tier with distance @xmath15 is attenuated by a factor @xmath16 , where @xmath17 is the path loss exponent of the @xmath10th tier . for small - scale fading , we assume rayleigh fading channels@xcite .
let @xmath18 denote the set of @xmath19 files ( e.g. , data objects or chucks of data objects ) in the hetnet .
for ease of illustration , assume that all files have the same size .
each file is of certain popularity , which is assumed to be identical among all users .
each user randomly requests one file , which is file @xmath20 with probability @xmath21 , where @xmath22 .
thus , the file popularity distribution is given by @xmath23 , which is assumed to be known apriori .
in addition , without loss of generality ( w.l.o.g . ) , assume @xmath24 .
the hetnet consists of cache - enabled macro - bss and pico - bss .
each bs in the @xmath10th tier is equipped with a cache of size @xmath25 to store different files .
assume @xmath26 .
each macro - bs is connected to the core network via a wireline backhaul link of transmission capacity @xmath27 ( files / slot ) , i.e. , each macro - bs can retrieve at most @xmath28 different files from the core network in each slot .
note that @xmath29 , @xmath30 and @xmath28 reflect the storage and backhaul resources in the cache - enabled hetnet .
we are interested in the case where the storage and backhaul resources are limited , and may not be able to satisfy all file requests . in this section ,
we propose a joint caching and multicasting design with certain design parameters , which can provide high spatial file diversity and ensure efficient content dissemination . to provide high spatial file diversity , we propose a _ hybrid caching design _ consisting of identical caching in the 1st tier and random caching in the 2nd tier , as illustrated in fig .
[ fig : system ] .
let @xmath31 denote the set of @xmath32 files cached in the @xmath10th tier . specifically ,
our hybrid caching design satisfies the following requirements : ( i ) _ non - overlapping caching across tiers _ : each file is stored in at most one tier ; ( ii ) _ identical caching in the 1st tier _ : each macro - bs stores the same set @xmath33 of @xmath29 ( different ) files ; and ( iii ) _ random caching in the 2nd tier _ : each pico - bs randomly stores @xmath30 different files out of all files in @xmath34 , forming a subset of @xmath34 .
thus , we have the following constraint : @xmath35 to further illustrate the random caching in the 2nd tier , we first introduce some notations .
we say every @xmath30 different files in @xmath36 form a combination .
thus , there are @xmath37 different combinations in total .
let @xmath38 denote the set of @xmath39 combinations .
combination @xmath40 can be characterized by an @xmath41-dimensional vector @xmath42 , where @xmath43 indicates that file @xmath44 is included in combination @xmath45 and @xmath46 otherwise .
note that there are @xmath30 1 s in each @xmath47 .
denote @xmath48 as the set of @xmath30 files contained in combination @xmath45 . each pico - bs stores one combination at random , which is combination @xmath40 with probability @xmath49 satisfying : ( for the 2nd tier ) .
then , based on the insights obtained , we shall focus on reducing complexity while maintaining superior performance . ]
@xmath50 denote @xmath51 . to facilitate the analysis in later sections , based on @xmath52
, we also define the probability that file @xmath53 is stored at a pico - bs , i.e. , @xmath54 where @xmath55 denotes the set of @xmath56 combinations containing file @xmath53 .
denote @xmath57 .
note that @xmath52 and @xmath58 depend on @xmath34 .
thus , in this paper , we use @xmath59 and @xmath60 when emphasizing this relation .
therefore , the hybrid caching design in the cache - enabled hetnet is specified by the design parameters @xmath61 . to efficiently utilize backhaul links and ensure high spatial file diversity
, we only retrieve files not stored in the cache - enabled hetnet via backhaul links .
let @xmath62 denote the set of @xmath63 files which can be retrieved by each macro - bs from the core network .
thus , we have the following constraint : @xmath64 therefore , the file distribution in the cache - enabled hetnet is fully specified by the hybrid caching design @xmath61 . in this part
, we propose a multicasting design associated with the hybrid caching design @xmath61 .
first , we introduce the user association under the proposed hybrid caching design . in the cache - enabled hetnet ,
a user accesses to a tier based on its desired file .
specifically , each user requesting file @xmath65 is associated with the nearest macro - bs and is referred to as a macro - user . while , each user requesting file @xmath44 is associated with the nearest pico - bs storing a combination @xmath66 ( containing file @xmath67 ) and is referred to as a pico - user .
the associated bs of each user is called its serving bs , and offers the maximum long - term average receive power for its desired file@xcite .
note that under the proposed hybrid caching design @xmath61 , the serving bs of a macro - user is its nearest macro - bs , while the serving bs of a pico - user ( affected by @xmath52 ) may not be its geographically nearest bs .
we refer to this association mechanism as the _ content - centric association _ in the cached - enabled hetnet , which is different from the traditional _ connection - based association _
@xcite in hetnets .
now , we introduce file scheduling in the cache - enabled hetnet
. each bs will serve all the cached files requested by its associated users .
each macro - bs only serves at most @xmath28 uncached files requested by its associated users , due to the backhaul constraint for retrieving uncached files .
in particular , if the users of a macro - bs request smaller than or equal to @xmath28 different uncached files , the macro - bs serves all of them ; if the users of a macro - bs request greater than @xmath28 different uncached files , the macro - bs will randomly select @xmath28 different requested uncached files to serve , out of all the requested uncached files according to the uniform distribution .
we consider multicasting in the cache - enabled hetnet for efficient content dissemination .
suppose a bs schedules to serve requests for @xmath68 different files .
then , it transmits each of the @xmath68 files at rate @xmath69 ( bit / second ) and over @xmath70 of total bandwidth w using fdma .
all the users which request one of the @xmath68 files from this bs try to decode the file from the single multicast transmission of the file at the bs .
note that , by avoiding transmitting the same file multiple times to multiple users , this content - centric transmission ( multicast ) can improve the efficiency of the utilization of the wireless medium and reduce the load of the wireless links , compared to the traditional connection - based transmission ( unicast ) . from the above illustration
, we can see that the proposed multicasting design is also affected by the proposed hybrid caching design @xmath61 .
therefore , the design parameters @xmath61 affect the performance of the proposed joint caching and multicasting design .
in this paper , w.l.o.g . , we study the performance of the typical user denoted as @xmath71 , which is located at the origin .
we assume all bss are active .
suppose @xmath71 requests file @xmath67 .
let @xmath72 denote the index of the tier to which @xmath71 belongs , and let @xmath73 denote the other tier .
let @xmath74 denote the index of the serving bs of @xmath71 .
we denote @xmath75 and @xmath76 as the distance and the small - scale channel between bs @xmath77 and @xmath78 , respectively .
we assume the complex additive white gaussian noise of power @xmath79 at @xmath71 .
when @xmath78 requests file @xmath67 and file @xmath67 is transmitted by bs @xmath80 , the signal - to - interference plus noise ratio ( sinr ) of @xmath78 is given by @xmath81 when @xmath71 requests file @xmath82 ( @xmath83 ) , let @xmath84 ( @xmath85 ) and @xmath86 ( @xmath87 ) denote the numbers of different cached and uncached files requested by the users associated with bs @xmath88 , respectively .
when @xmath71 requests file @xmath44 , let @xmath89 denote the number of different cached files requested by the users associated with bs @xmath90 .
note that @xmath91 are discrete random variables , the probability mass functions ( p.m.f.s ) of which depend on @xmath92 , @xmath93 and the design parameters @xmath94 .
in addition , if @xmath95 , bs @xmath80 will transmit file @xmath67 for sure ; if @xmath96 , for given @xmath97 , bs @xmath80 will transmit file @xmath67 with probability @xmath98 . given that file @xmath67 is transmitted , it can be decoded correctly at @xmath71 if the channel capacity between bs @xmath80 and @xmath71 is greater than or equal to @xmath99 .
requesters are mostly concerned with whether their desired files can be successfully received .
therefore , in this paper , we consider the successful transmission probability of a file requested by @xmath71 as the network performance metric . by total probability theorem ,
the successful transmission probability under the proposed scheme is given by : @xmath100 where @xmath101 is given by , and @xmath102 and @xmath103 are given by and , respectively .
note that in and , each term multiplied by @xmath104 represents the successful transmission probability of file @xmath67 .
later , we shall see that under the proposed caching and multicasting design for content - oriented services in the cache - enabled hetnet , the successful transmission probability is sufficiently different from the traditional rate coverage probability studied for connection - oriented services @xcite .
in particular , the successful transmission probability considered in this paper not only depends on the physical layer parameters , such as the macro and pico bs densities @xmath105 and @xmath106 , user density @xmath93 , path loss exponents @xmath107 and @xmath108 , bandwidth @xmath14 , backhaul capacity @xmath28 and transmit signal - to - noise ratios ( snrs ) @xmath109 and @xmath110 , but also relies on the content - related parameters , such as the popularity distribution @xmath92 , the cache sizes @xmath29 and @xmath30 , and the design parameters @xmath111 . while , the traditional rate coverage probability only depends on the physical layer parameters . in addition
, the successful transmission probability depends on the physical layer parameters in a different way from the traditional rate coverage probability .
for example , the content - centric association leads to different distributions of the locations of serving and interfering bss ; the multicasting transmission results in different file load distributions at each bs @xcite ; and the cache - enabled architecture makes content availability related to bs densities .
in this section , we study the successful transmission probability @xmath112 under the proposed caching and multicasting design for given design parameters @xmath113 .
first , we analyze the successful transmission probability in the general region .
then , we analyze the asymptotic transmission probability in the high snr and user density region . in this part
, we would like to analyze the successful transmission probability in the general region , using tools from stochastic geometry .
in general , file loads @xmath114 , @xmath115 , @xmath116 , @xmath117 , @xmath118 and sinr @xmath119 are correlated in a complex manner , as bss with larger association regions have higher file load and lower sinr ( due to larger user to bs distance ) @xcite . for the tractability of the analysis , as in @xcite and @xcite , the dependence is ignored . therefore , to obtain the successful transmission probability in , we analyze the distributions of @xmath114 , @xmath115 , @xmath116 , @xmath117 , @xmath118 and the distribution of @xmath119 , separately .
first , we calculate the p.m.f.s of @xmath114 and @xmath115 for @xmath120 as well as the p.m.f.s of @xmath116 and @xmath117 for @xmath83 . in calculating these p.m.f.s
, we need the probability density function ( p.d.f . ) of the size of the voronoi cell of @xmath121 w.r.t .
file @xmath122 .
note that this p.d.f .
is equivalent to the p.d.f . of the size of the voronoi cell
to which a randomly chosen user belongs . based on a tractable approximated form of this p.d.f . in @xcite , which is widely used in existing literature@xcite , we obtain the p.m.f.s of @xmath114 , @xmath115 , @xmath116 and @xmath117 .
[ p.m.f.s of @xmath114 @xmath115 , @xmath116 and @xmath117 ] the p.m.f.s of @xmath114 and @xmath115 for @xmath120 and the p.m.f.s of @xmath116 and @xmath117 for @xmath96 are given by @xmath124=g(\mathcal f_{1,-n}^c , k^c-1),\quad k^c=1,\cdots , k_1^c,\label{eqn : k-1-c}\\ & \pr \left[\overline{k}_{1,n,0}^b = k^b\right]=g(\mathcal f_{1}^b , k^b),\quad k^b=0,\cdots , f_1^b,\label{eqn : k-1-b - bar}\\ & \pr \left[\overline{k}_{1,n,0}^c = k^c\right]= g(\mathcal f_{1}^c , k^c),\quad k^c=0,\cdots , k_1^c,\label{eqn : k-1-c - bar}\\ & \pr \left[k_{1,n,0}^b = k^b\right]=g(\mathcal f_{1,-n}^b , k^b-1),\quad k^b=1,\cdots , f_1^b,\label{eqn : k-1-b } \ ] ] where @xmath125 , @xmath126 @xmath127 and @xmath128 .
[ lem : pmf - k - m ] please refer to appendix a. next , we obtain the p.m.f .
of @xmath118 for @xmath44 . in calculating the p.m.f . of @xmath118 , we need the p.d.f . of the size of the voronoi cell of @xmath121 w.r.t .
file @xmath129 when @xmath121 contains combination @xmath66 . however , this p.d.f .
is very complex and is still unknown . for the tractability of the analysis , as in @xcite , we approximate this p.d.f . based on a tractable approximated form of the p.d.f . of the size of the voronoi cell to which a randomly chosen user belongs@xcite , which is widely used in existing literature@xcite
. under this approximation
, we obtain the p.m.f . of @xmath118 .
[ p.m.f . of @xmath118 ]
the p.m.f . of @xmath118 for @xmath53
is given by @xmath131\nonumber\\ & = \sum_{i\in \mathcal i_n}\frac{p_i}{t_n}\sum_{\mathcal x\in \left\{\mathcal s \subseteq
\mathcal n_{i ,-
n } : |\mathcal s|=k^c-1\right\ } } \prod\limits_{m\in \mathcal x}\left(1-\left(1+\frac{a_m\lambda_u}{3.5t_m\lambda_2}\right)^{-4.5}\right)\prod\limits_{m\in { \mathcal n_{i ,-
n}\setminus \mathcal x}}\left(1+\frac{a_m\lambda_u}{3.5t_m\lambda_2}\right)^{-4.5},\nonumber\\ & \hspace{12cm}k^c=1,\cdots , k_2^c,\label{eqn : k - pmf}\end{aligned}\ ] ] where @xmath132 .
[ lem : pmf - k ] please refer to appendix b. the distributions of the locations of desired transmitters and interferers are more involved than those in the traditional connection - based hetnets .
thus , it is more challenging to analyze the p.d.f . of @xmath119 .
when @xmath71 is a macro - user , as in the traditional connection - based hetnets , there are two types of interferers , namely , i ) all the other macro - bss besides its serving macro - bs , and ii ) all the pico - bss .
when @xmath71 is a pico - user , different from the traditional connection - based hetnets , there are three types of interferers , namely , i ) all the other pico - bss storing the combinations containing the desired file of @xmath71 besides its serving pico - bs , ii ) all the pico - bss without the desired file of @xmath71 , and iii ) all the macro - bss . by carefully handling these distributions
, we can derive the p.d.f .
of @xmath119 , for @xmath133 and @xmath134 , respectively .
then , based on lemma [ lem : pmf - k - m ] and lemma [ lem : pmf - k ] as well as the p.d.f . of @xmath119 , we can derive the successful transmission probability @xmath112 .
[ performance ] the successful transmission probability @xmath112 of @xmath78 is given by @xmath135 \pr[\overline{k}_{1,n,0}^b = k^b]f_{1,{k^c+\min\{k_1^b , k^b\}}}\nonumber\\ & + \sum_{n\in \mathcal f_1^b}a_{n } \sum_{k^c=0}^{k_1^c}\sum_{k^b=1}^{f_1^b } \pr [ \overline{k}_{1,n,0}^c = k^c]\pr [ k_{1,n,0}^b = k^b ] \frac{\min\{k_1^b , k^b\}}{k^b } f_{1,{k^c+\min\{k_1^b , k^b\}}}\nonumber\\ & + \sum_{n\in \mathcal f_2^c}a_{n } \sum_{k^c=1}^{k_2^c } \pr[k_{2,n,0}=k^c]f_{2,k^c}(t_n),\end{aligned}\ ] ] where the p.m.f.s of @xmath114 @xmath115 , @xmath116 , @xmath117 and @xmath118 are given by lemma [ lem : pmf - k - m ] and lemma [ lem : pmf - k ] , @xmath136 and @xmath137 are given by and , and @xmath138 is given by .
here , @xmath139 and @xmath140 denote the complementary incomplete beta function and the beta function , respectively .
[ thm : generalkmulti ] please refer to appendix c. from theorem [ thm : generalkmulti ] , we can see that in the general region , the physical layer parameters @xmath107 , @xmath108 , @xmath14 , @xmath105 , @xmath106 , @xmath93 , @xmath109 , @xmath110 , and the design parameters @xmath141 jointly affect the successful transmission probability @xmath112 .
the impacts of the physical layer parameters and the design parameters on @xmath112 are coupled in a complex manner . in this part , to obtain design insights , we focus on analyzing the asymptotic successful transmission probability in the high snr and user density region . note that in the remaining of the paper , when considering the high snr region , we assume @xmath142 and @xmath143 for some @xmath144 and @xmath145 , and let @xmath146 . on the other hand , in the high user density region where @xmath147 , discrete random variables @xmath148 , @xmath149 and @xmath150 in distribution .
define @xmath151 , @xmath152 , and @xmath153 .
note that when @xmath154 , @xmath155 and @xmath156 become functions of @xmath58 instead of @xmath52 . from theorem [ thm : generalkmulti ] , we have the following lemma . when @xmath146 and @xmath157 , we have @xmath158 , where @xmath159 and @xmath160 . here ,
@xmath161 and @xmath162 are given by and , and @xmath138 is given by .
please refer to appendix d. [ lem : asym - perf ] note that @xmath163 represents the successful transmission probability for file @xmath164 ( given that this file is transmitted ) , and @xmath165 represents the successful transmission probability for file @xmath44 , in the asymptotic region . for given @xmath166 ,
we interpret lemma [ lem : asym - perf ] below .
when @xmath167 , the successful transmission probability of file @xmath168 is the same as that of file @xmath169 . in other words , when backhaul capacity is sufficient , storing a file at a macro - bs or retrieving the file via the backhaul link makes no difference in successful transmission probability .
when @xmath170 , the successful transmission probability of file @xmath168 is greater than that of file @xmath169 .
in other words , when backhaul capacity is limited , storing a file at a macro - bs is better than retrieving the file via the backhaul link .
note that @xmath171 is an increasing function ( please refer to appendix e for the proof ) .
thus , for any @xmath172 satisfying @xmath173 , the successful transmission probability of file @xmath174 is greater than that of file @xmath175 .
that is , a file of higher probability being cached at a pico - bs has higher successful transmission probability . later , in section
[ sec : new - opt ] , we shall see that the structure of @xmath176 facilitates the optimization of @xmath177 .
next , we further study the symmetric case where @xmath178 in the high snr and user density region . from lemma [ lem : asym - perf ]
, we have the following lemma . when @xmath179 , @xmath146 and @xmath157 , we have @xmath180 , where @xmath181 and @xmath182 . here ,
@xmath138 is given by , and @xmath183 , @xmath184 and @xmath185 are given by @xmath186 [ lem : asym - perf - v2 ] please refer to appendix d. from lemma [ lem : asym - perf - v2 ] , we can see that in the high snr and user density region , when @xmath179 , the impact of the physical layer parameters @xmath187 , @xmath188 and @xmath14 , captured by @xmath189 , @xmath190 and @xmath191 , and the impact of the design parameters @xmath141 on the successful transmission probability @xmath192 can be easily separated .
later , in section [ sec : new - opt ] , we shall see that this separation greatly facilitates the optimization of @xmath177 .
[ fig : verification - kmulti ] plots the successful transmission probability versus the transmit snr @xmath193 and the user density @xmath93 . fig .
[ fig : verification - kmulti ] verifies theorem [ thm : generalkmulti ] and lemma [ lem : asym - perf ] ( lemma [ lem : asym - perf - v2 ] ) , and demonstrates the accuracy of the approximations adopted .
[ fig : verification - kmulti ] also indicates that @xmath176 provides a simple and good approximation for @xmath194 in the high transmit snr region ( e.g. , @xmath195 db ) and the high user density region ( e.g. , @xmath196 ) .
in this section , we formulate the optimal caching and multicasting design problem to maximize the successful transmission probability @xmath177 , which is a mixed discrete - continuous optimization problem . to facilitate the solution of this challenging optimization problem in the next section
, we also formulate the asymptotically optimal caching and multicasting design problem to maximize the asymptotic successful transmission probability @xmath176 in the high snr and user density region .
the caching and multicasting design fundamentally affects the successful transmission probability via the design parameters @xmath141 .
we would like to maximize @xmath197 by carefully optimizing @xmath141 .
[ performance optimization][prob : opt ] @xmath198 where @xmath197 is given by .
note that problem [ prob : opt ] is a mixed discrete - continuous optimization problem with two main challenges .
one is the choice of the sets of files @xmath33 and @xmath34 ( discrete variables ) stored in the two tiers , and the other is the choice of the caching distribution @xmath59 ( continuous variables ) of random caching for the 2nd tier .
we thus propose an equivalent alternative formulation of problem [ prob : opt ] which naturally subdivides problem [ prob : opt ] according to these two aspects .
[ equivalent optimization][prob : opt - eq ] @xmath199
@xmath200 for given @xmath34 , the optimization problem in is in general a non - convex optimization problem with a large number of optimization variables ( i.e. , @xmath201 optimization variables ) , and it is difficult to obtain the global optimal solution and calculate @xmath202 . even given @xmath202 , the optimization problem in is a discrete optimization problem over a very large constraint set , and is np - complete in general . therefore , problem [ prob : opt - eq ] is still very challenging . to facilitate the solution of the challenging mixed discrete - continuous optimization problem
, we also formulate the optimization of the asymptotic successful transmission probability @xmath176 given in lemma [ lem : asym - perf ] , i.e. , which has a much simpler form than @xmath197 given in theorem [ thm : generalkmulti ] .
equivalently , we can consider the asymptotic version of problem [ prob : opt - eq ] in the high snr and user density region .
[ asymptotic optimization][prob : opt - asymp - eq ] @xmath203 the optimal solution to the optimization in is written as @xmath204 and @xmath205 is given by @xmath206 where the optimal solution to the optimization in is written as @xmath207 .
the optimal solution to problem [ prob : opt - asymp - eq ] is given by @xmath208 , which is the asymptotic optimal solution to problem [ prob : opt - eq ] ( problem [ prob : opt ] ) .
based on lemma 2 in @xcite , we know that the optimization in is equivalent to the following optimization for given @xmath34 @xmath209 where the optimal solution is written as @xmath210 .
in addition , any @xmath207 in convex polyhedron @xmath211 is an optimal solution to the optimization in , where is given by : @xmath212 the vertices of the convex polyhedron @xmath213 can be obtained based on the simplex method , and any @xmath214 can be constructed from all the vertices using convex combination .
thus , when optimizing the asymptotic performance for given @xmath34 , we can focus on the optimization in instead of the optimization in .
in this section , we propose a two - step optimization framework to obtain a near optimal solution with manageable complexity and superior performance in the general region . we first characterize the structural properties of the asymptotically optimal solutions . then , based on these properties , we obtain a near optimal solution in the general region . in this part
, we study the continuous part and the discrete part of the asymptotic optimization in problem [ prob : opt - asymp - eq ] , respectively , to obtain design insights into the solution in the general region .
as the structure of @xmath215 is very complex , it is difficult to obtain the closed - form optimal solution @xmath210 to the optimization in . by exploring the structural properties of @xmath215
, we know that files of higher popularity get more storage resources . [ structural property of optimization in ] given any @xmath216 satisfying @xmath217 and @xmath218 , if @xmath219 , then @xmath220.[lem : mono - general - asym ] please refer to appendix e. now , we focus on obtaining a numerical solution to the optimization in . for given @xmath216 satisfying @xmath217
, the optimization in is a continuous optimization of a differentiable function @xmath215 over a convex set . in general , it is difficult to show the convexity of @xmath171 in . a stationary point to the optimization in can be obtained using standard gradient projection methods@xcite . here , we consider the diminishing stepsize@xcite satisfying @xmath221 and propose algorithm [ alg : local ] . in step 2 of algorithm [ alg : local ] , @xmath222 , where @xmath223 is given by .
step 3 is the projection of @xmath224 onto the set of the variables satisfying the constraints in and .
it is shown in @xcite that @xmath225 in algorithm [ alg : local ] converges to a stationary point of the optimization in as @xmath226 .
on the other hand , as illustrated in the discussion of lemma [ lem : asym - perf ] , @xmath171 is actually a cumulative density function ( c.d.f . ) , and is concave in most of the cases we are interested in . if @xmath171 in is concave w.r.t .
@xmath227 , the differentiable function @xmath215 is concave w.r.t .
@xmath58 , and hence , the optimization in is a convex problem .
then , @xmath225 in algorithm [ alg : local ] converges to the optimal solution @xmath210 to the optimization in as @xmath226 . in other words , under a mild condition ( i.e. , @xmath171 is convex ) , we can obtain the optimal solution @xmath210 to the optimization in using algorithm [ alg : local ] .
[ alg : local ] next , we consider the symmetric case , i.e. , @xmath228 , in the high snr and user density region . in this case , we can easily verify that @xmath229 ( given in lemma [ lem : asym - perf - v2 ] ) is convex and slater s condition is satisfied , implying that strong duality holds . using kkt conditions , we can obtain the closed - form solution to the optimization in in this case .
[ asymptotically optimal solution when @xmath230 for given @xmath34 , when @xmath179 , @xmath146 and @xmath157 , the optimal solution to the optimization in is given by @xmath231^+,1\right\ } , \
n\in \mathcal f_2^c,\label{eqn : opt - k - infty}\end{aligned}\ ] ] where @xmath232^+\triangleq \max\{x,0\}$ ] and @xmath233 satisfies @xmath234^+,1\right\}=k_2^c.\label{eqn : opt - k - infty - v}\end{aligned}\ ] ] here , @xmath184 and @xmath185 are given by and , respectively .
[ lem : solu - opt - infty ] please refer to appendix f. as the water - level in the traditional water - filling power control , the root @xmath233 to the equation in can be easily solved .
thus , by lemma [ lem : solu - opt - infty ] , we can efficiently compute @xmath210 when @xmath235 .
lemma [ lem : solu - opt - infty ] can be interpreted as follows . as illustrated in fig .
[ fig : optstructure ] , @xmath210 given by lemma [ lem : solu - opt - infty ] has a reverse water - filling structure .
the file popularity distribution @xmath236 and the physical layer parameters ( captured in @xmath237 and @xmath238 ) jointly affect @xmath233 .
given @xmath233 , the physical layer parameters ( captured in @xmath237 and @xmath238 ) affect the caching probabilities of all the files in the same way , while the popularity of file @xmath53 ( i.e. , @xmath104 ) only affects the caching probability of file @xmath67 ( i.e. , @xmath239 ) .
from lemma [ lem : solu - opt - infty ] , we know that for any @xmath172 such that @xmath219 , we have @xmath240 , as @xmath241 . in other words , files in @xmath34 of higher popularity get more storage resources in the 2nd tier .
in addition , there may exist @xmath242 such that @xmath243 for all @xmath53 and @xmath244 , and @xmath245 for all @xmath53 and @xmath246 .
in other words , some files in @xmath34 of lower popularity may not be stored in the 2nd tier . for a popularity distribution with a heavy tail ,
more different files in @xmath34 can be stored in the 2nd tier .
given @xmath247 , the optimization in is a discrete optimization . it can be shown that the number of possible choices for @xmath248 satisfying is given by @xmath249 .
thus , a brute - force solution to the discrete optimization in is not acceptable .
now , we explore the structural properties of the discrete optimization in to facilitate the design of low - complexity asymptotically optimal solutions .
[ structural properties of optimization in ] there exists an optimal solution @xmath250 to the optimization in satisfying the following conditions : ( i ) @xmath251 , where @xmath252 ; and ( ii ) there exists @xmath253 , such that @xmath254 and @xmath255 , where @xmath256.[thm : opt - prop ] please refer to appendix g. theorem [ thm : opt - prop ] can be interpreted as follows . property ( ii )
indicates that there is an optimal solution @xmath250 to the optimization in satisfying that the files in @xmath257 , @xmath258 and @xmath259 are consecutive , and the files in @xmath257 are more popular than those in @xmath258 .
this can be easily understood from fig .
[ fig : cachestru ] .
it can be shown that the number of possible choices for @xmath248 satisfying the properties in theorem [ thm : opt - prop ] is given by @xmath260 , which is much smaller than the number of possible choices just satisfying , i.e. , @xmath261 . by restricting to the choices for @xmath262 satisfying the properties in theorem [ thm : opt - prop ]
, we can greatly reduce the complexity for solving the optimization in without losing any optimality . in some special cases ,
we can obtain extra properties other than those in theorem [ thm : opt - prop ] .
\(i ) if @xmath263 , then @xmath264 in theorem [ thm : opt - prop ] satisfies @xmath265 ; ( ii ) if @xmath266 , then @xmath264 in theorem [ thm : opt - prop ] satisfies @xmath267 .
[ lem : opt - prop ] please refer to appendix h. lemma [ lem : opt - prop ] can be interpreted as follows .
property ( i ) implies that the most popular files are served by the 1st tier ( cf .
case 1 in fig .
[ fig : cachestru ] ) , if @xmath268 .
this condition holds when @xmath269 and @xmath270 are above some thresholds , respectively . in this case ,
macro - bss intend to multicast the most popular files , as they can offer relatively higher receive power , and hence higher successful transmission probability for the most popular files .
note that when the condition in ( i ) holds , by theorem [ thm : opt - prop ] and lemma [ lem : opt - prop ] , we can directly determine @xmath271 .
property ( ii ) implies that the most popular file , i.e. , file 1 , is not served by the 1st tier ( cf .
cases 2 - 4 in fig .
[ fig : cachestru ] ) , if @xmath266 .
this condition holds when @xmath269 and @xmath270 are below some thresholds , respectively . in this case ,
pico - bss intend to multicast the most popular file , as they can offer relatively higher receive power , and hence higher successful transmission probability for the most popular file .
when the condition in ( ii ) holds , we can use lemma [ lem : opt - prop ] together with theorem [ thm : opt - prop ] to reduce the set of possible choices for @xmath262 , to further reduce the complexity for solving the optimization in without losing any optimality .
first , we consider the near optimal solution for the continuous part ( for given @xmath34 ) . as illustrated in section [ sec : new - opt]-a , based on @xmath210 obtained using algorithm [ alg : local ] or lemma [ lem : solu - opt - infty ] ( when @xmath235 ) , we can determine @xmath272 .
as illustrated in section [ subsec : asym - opt - prob ] , any @xmath273 is an optimal solution to the optimization in . in other words , for given @xmath34
, we have a set of asymptotically optimal solutions in the high snr and user density region . substituting @xmath274 satisfying into @xmath275 in theorem [ thm : generalkmulti ] ,
we have @xmath276 given in .
for given @xmath34 ( and @xmath210 ) , we would like to obtain the best asymptotically optimal solution which maximizes the successful transmission probability @xmath276 in the general region among all the asymptotically optimal solutions in @xmath277 .
[ optimization of @xmath274 under given @xmath34 ( and @xmath278 ) ] @xmath279 the optimal solution is denoted as @xmath280.[prob : asm - improvement ] problem [ prob : asm - improvement ] is a linear programming problem . to reduce the complexity for solving problem [ prob : asm - improvement ] ,
we first derive some caching probabilities which are zero based on the relationship between @xmath52 and @xmath58 . in particular , for all @xmath281 and @xmath282 , we have @xmath283 ; for all @xmath284 and @xmath285 , we have @xmath283 .
thus , we have @xmath286 where @xmath287
. then , we can compute the remaining caching probabilities for the combinations in @xmath288 using the simplex method ( refer to step 6 of algorithm [ alg : sympk ] for details ) . therefore , using the above approach , for given @xmath34 , we can obtain the best asymptotically optimal solution @xmath280 in @xmath272 to the optimization in .
next , we consider the near optimal solution for the discrete part . specifically , after obtaining @xmath289 using the above approach for the continuous part , we consider the optimization of @xmath290 over the set of @xmath291 satisfying theorem [ thm : opt - prop ] ( and lemma [ lem : opt - prop ] ) .
let @xmath292 and @xmath293 denote the optimal solution and the optimal value . finally , combining the above discrete part and continuous part ,
we can obtain the near optimal solution @xmath294 to problem [ prob : opt ] ( problem [ prob : opt - eq ] ) , as summarized in algorithm [ alg : sympk ] .
we can show that in the general region , under a mild condition ( i.e. , @xmath171 is convex ) , the near optimal solution @xmath294 obtained by algorithm [ alg : sympk ] achieves the successful transmission probability @xmath295 greater than or equal to that of any optimal solution to problem [ prob : opt - asymp - eq ] , i.e. , any asymptotically optimal solution to problem [ prob : opt ] ( problem [ prob : opt - eq ] ) .
we have @xmath296 , for all @xmath297 , where @xmath298 is an optimal solution to problem [ prob : opt - asymp - eq ] .
[ alg : sympk ]
in this section , we compare the proposed near optimal design given by algorithm [ alg : sympk ] with three schemes .
baseline 1 ( most popular ) refers to the design in which each macro - bs selects the most @xmath299 popular files to store and fetch , and each pico - bs selects the most @xmath30 popular files to store @xcite .
baseline 2 ( i.i.d .
file popularity ) refers to the design in which each macro - bs selects @xmath299 files to store and fetch , and each pico - bs selects @xmath30 files to store , in an i.i.d .
manner with file @xmath67 being selected with probability @xmath104 @xcite .
note that under this scheme , each ( macro or pico ) bs may cache multiple copies of one file , leading to storage waste , and each macro - bs may fetch multiple copies of one file , leading to backhaul waste .
baseline 3 ( uniform comb . dist . ) refers to the design in which each macro - bs randomly selects a combination of @xmath299 different files to store and fetch , and each pico - bs randomly selects a combination of @xmath30 different files to store , according to the uniform distribution @xcite . under the three baseline schemes ,
each user requesting file @xmath67 is associated with the bs which stores file @xmath67 and offers the maximum long - term average receive power at this user .
in addition , the three baseline schemes also adopt the same multicasting scheme as in our design . in the simulation
, we assume the popularity follows zipf distribution , i.e. , @xmath300 , where @xmath301 is the zipf exponent .
[ fig : simulation - large ] illustrates the successful transmission probability versus different parameters . from fig .
[ fig : simulation - large ] , we can observe that the proposed design outperforms all the three baseline schemes .
in addition , the proposed design , baseline 2 and baseline 3 have much better performance than baseline 1 , as they provide file diversity to improve the network performance , when the storage and backhaul resources are limited and the cache - enabled hetnet with backhaul constraints may not be able to satisfy all file requests . specifically , fig .
[ fig : simulation - large ] ( a ) illustrates the successful transmission probability versus the cache sizes @xmath29 and @xmath30 .
we can see that the performance of all the schemes increases with @xmath29 and @xmath30 .
this is because as @xmath29 and @xmath30 increase , each bs can store more files , and the probability that a randomly requested file is cached at a nearby bs increases .
[ fig : simulation - large ] ( b ) illustrates the successful transmission probability versus the zipf exponent @xmath301 .
we can observe that the performance of the proposed design , baseline 1 and baseline 2 increases with the zipf exponent @xmath301 faster than baseline 3 .
this is because when @xmath301 increases , the tail of popularity distribution becomes small , and hence , the average network file load decreases .
the performance increase of baseline 3 with @xmath301 only comes from the decrease of the average network file load . while , under the proposed design , baseline 1 and baseline 2 , the probability that a randomly requested file is cached at a nearby bs increases with @xmath301 .
thus , the performance increases of the proposed design , baseline 1 and baseline 2 with @xmath301 are due to the decrease of the average network file load and the increase of the chance of a requested file being cached at a nearby bs . fig .
[ fig : simulation - large ] ( c ) illustrates the successful transmission probability versus the user density @xmath93 .
we can see that the performance of all the schemes decreases with @xmath93 .
this is because the probability of a cached file being requested by at least one user increases , as @xmath93 increases .
in this paper , we considered the analysis and optimization of caching and multicasting in a large - scale cache - enabled hetnet with backhaul constraints .
we proposed a hybrid caching design and a corresponding multicasting design to provide high spatial file diversity and ensure efficient content dissemination . utilizing tools from stochastic geometry
, we analyzed the successful transmission probability in the general region and the asymptotic region .
then , we formulated a mixed discrete - continuous optimization problem to maximize the successful transmission probability by optimizing the design parameters . by exploring the structural properties
, we obtained a near optimal solution with superior performance and manageable complexity , based on a two - step optimization framework .
the analysis and optimization results offered valuable design insights for practical cache - enabled hetnets .
when typical user @xmath71 requests file @xmath133 , let random variable @xmath302 denote whether file @xmath303 is requested by the users associated with serving macro - bs @xmath121 . specifically , when @xmath71 requests file @xmath304 , we have @xmath305 and @xmath306 . when @xmath71 requests file @xmath307 , we have @xmath308 and @xmath309 . thus , we have to prove , , and , it remains to calculate @xmath310 $ ] .
the p.m.f . of @xmath311 depends on the p.d.f . of the size of the voronoi cell of macro - bs @xmath121 ,
i.e. , the p.d.f . of the size of the voronoi cell
to which a randomly chosen user belongs @xcite .
thus , we can calculate the p.m.f .
of @xmath311 using lemma 3 of @xcite as follows therefore , we complete the proof .
when typical user @xmath71 requests file @xmath53 , let random variable @xmath312 denote whether file @xmath313 is requested by the users associated with serving pico - bs @xmath121 when pico - bs @xmath121 contains combination @xmath66 . when @xmath71 requests file @xmath314 and serving pico - bs @xmath121 contains combination @xmath66 , we have @xmath315 .
thus , we have the probability that pico - bs @xmath121 contains combination @xmath66 is @xmath316 .
thus , by the law of total probability , we have thus , to prove , it remains to calculate @xmath317 $ ] .
the p.m.f . of @xmath318 depends on the p.d.f . of the size of the voronoi cell of pico - bs @xmath121 w.r.t .
file @xmath319 when pico - bs @xmath121 contains combination @xmath66 , which is unknown .
we approximate this p.d.f .
based on the known result of the p.d.f . of the size of the voronoi cell
to which a randomly chosen user belongs @xcite . under this approximation
, we can calculate the p.m.f .
of @xmath318 using lemma 3 of @xcite as follows
based on ( [ eqn : succ - prob - def ] ) , to prove theorem [ thm : generalkmulti ] , we calculate @xmath320 and @xmath321 , respectively . when @xmath71 is a macro - user , as in the traditional connection - based hetnets , there are two types of interferers , namely , i ) all the other macro - bss besides its serving macro - bs , and ii ) all the pico - bss .
thus , we rewrite the sinr expression in as follows : where @xmath322 and @xmath323 .
next , we calculate the conditional successful transmission probability of file @xmath324 requested by @xmath78 conditioned on @xmath325 when the file load is @xmath68 , i.e. , where @xmath326 is obtained based on ( [ eqn : sinr_v3 ] ) , ( b ) is obtained by noting that @xmath327 , and ( c ) is due to the independence of the rayleigh fading channels and the independence of the ppps . to calculate @xmath328 according to ,
we first calculate @xmath329 and @xmath330 , respectively .
the expression of @xmath329 is calculated as follows : where @xmath331 is obtained by utilizing the probability generating functional of ppp ( * ? ? ?
* page 235 ) , and @xmath332 is obtained by first replacing @xmath333 with @xmath334 , and then replacing @xmath335 with @xmath336 .
similarly , the expression of @xmath330 is calculated as follows : substituting ( [ eq : lt_k1_n_m ] ) and ( [ eq : lt_k1_m1 ] ) into ( [ eq : condi_cp_k_m ] ) , we obtain @xmath328 as follows : now , we calculate @xmath320 by first removing the condition of @xmath328 on @xmath325 .
note that we have the p.d.f .
of @xmath337 as @xmath338 .
thus , we have : therefore , by ( [ eqn : succ - prob - def-1 ] ) and by letting @xmath339 in ( [ eq : cp_k_n_m ] ) , we have when @xmath71 is a pico - user , different from the traditional connection - based hetnets , there are three types of interferers , namely , i ) all the other pico - bss storing the combinations containing the desired file of @xmath71 besides its serving pico - bs , ii ) all the pico - bss without the desired file of @xmath71 , and iii ) all the macro - bss .
thus , we rewrite the sinr expression in as follows : where @xmath341 is the point process generated by pico - bss containing file combination @xmath342 , @xmath343 is the point process generated by pico - bss containing file combination @xmath344 , @xmath345 , @xmath346 and @xmath347 . due to the random caching policy and independent thinning ( * ? ? ?
* page 230 ) , we obtain that @xmath341 is a homogeneous ppp with density @xmath348 and @xmath343 is a homogeneous ppp with density @xmath349 .
next , we calculate the conditional successful transmission probability of file @xmath44 requested by @xmath78 conditioned on @xmath350 when the file load is @xmath68 , denoted as @xmath351.\ ] ] similar to ( [ eq : condi_cp_k_m ] ) and based on ( [ eqn : sinr_v2 ] ) , we have : to calculate @xmath352 according to , we first calculate @xmath353 , @xmath354 and @xmath329 , respectively .
similar to ( [ eq : lt_k1_n_m ] ) and ( [ eq : lt_k1_m1 ] ) , we have : substituting ( [ eq : lt_k1_n ] ) , ( [ eq : lt_k1_p1 ] ) and ( [ eq : lt_k1_p2 ] ) into ( [ eq : condi_cp_k ] ) , we obtain @xmath352 as follows : now , we calculate @xmath355 by first removing the condition of @xmath352 on @xmath350 .
note that we have the p.d.f .
of @xmath356 as @xmath357 , as pico - bss storing file @xmath67 form a homogeneous ppp with density @xmath358 .
thus , we have : therefore , by ( [ eqn : succ - prob - def-2 ] ) and by letting @xmath359 in ( [ eq : cp_k_n ] ) , we have
when @xmath360 , @xmath361 and @xmath362 .
when @xmath157 , discrete random variables @xmath148 , @xmath149 and @xmath150 in distribution . thus , when @xmath142 , @xmath143 , @xmath146 , and @xmath157 , we can show @xmath363 and @xmath364 . thus , we can prove lemma [ lem : asym - perf ] .
when @xmath142 , @xmath143 , @xmath146 , @xmath157 , and @xmath365 we have : where @xmath189 , @xmath184 and @xmath185 are given by , and . noting that @xmath366 ( @xmath367 is a constant ) , we can solve integrals in and .
thus , by lemma [ lem : asym - perf ] , we can prove lemma [ lem : asym - perf - v2 ] .
to prove lemma [ lem : mono - general - asym ] , we first have the following lemma .
[ monotonicity of @xmath171 ] @xmath171 is an increasing function of @xmath227 .
[ lem : monotonicity - f-2-k ] by replacing @xmath369 with @xmath370 in ( [ eqn : f-2-k - infty ] ) , we have : when @xmath371 and @xmath372 , @xmath373 is a decreasing function of @xmath374 . because @xmath375 , @xmath376 and @xmath377 , and @xmath378 and @xmath379 are decreasing functions of @xmath227 .
the integrand is an increasing function of @xmath227 for all @xmath380 .
therefore , we can show that @xmath171 is an increasing function of @xmath227 .
now , we prove lemma [ lem : mono - general - asym ] .
let @xmath381 denote an optimal solution to probelm [ prob : opt - asymp - eq ] .
consider @xmath382 satisfying @xmath241 .
suppose @xmath383 . based on lemma [ lem : monotonicity - f-2-k ]
, we have @xmath384 .
now , we construct a feasible solution @xmath385 to problem [ prob : opt - asymp - eq ] by choosing @xmath386 , @xmath387 , @xmath388 , and @xmath389 for all @xmath390 .
thus , by lemma [ lem : asym - perf ] and the optimality of @xmath381 , we have : since @xmath241 , by ( [ eqn : contradiction - thm2-a ] ) , we have @xmath391 , which contradicts the assumption .
therefore , by contradiction , we can prove lemma [ lem : mono - general - asym ] .
for given @xmath34 , when @xmath228 , @xmath146 , and @xmath392 , the lagrangian of the optimization in is given by where @xmath393 and @xmath394 are the lagrange multipliers associated with , @xmath395 is the lagrange multiplier associated with , @xmath396 , and @xmath397 .
thus , we have since strong duality holds , primal optimal @xmath278 and dual optimal @xmath398 , @xmath399 , @xmath233 satisfy kkt conditions , i.e. , ( i ) primal constraints : , , ( ii ) dual constraints @xmath400 and @xmath394 for all @xmath44 , ( iii ) complementary slackness @xmath401 and @xmath402 for all @xmath44 , and ( iv ) @xmath403 for all @xmath44 . by ( ii ) , ( iii ) , and ( iv ) , when @xmath404 , we have @xmath400 , @xmath405 , and @xmath406 ; when @xmath407 , we have @xmath408 , @xmath405 , , and @xmath409 ; when @xmath410 , we have @xmath408 , @xmath394 , and @xmath411 . therefore , we have . combining , we can prove lemma [ lem : solu - opt - infty ] .
by constraints ( [ eqn : cache - constr ] ) and ( [ eqn : backhaul ] ) , we have @xmath413 , @xmath414 and @xmath415 . to prove property ( i ) of theorem [ thm : opt - prop ]
, it remains to prove @xmath416 .
suppose there exists an optimal solution @xmath381 to probelm [ prob : opt - asymp - eq ] satisfying @xmath417 , then we have : now , we construct a feasible solution @xmath385 to problem [ prob : opt - asymp - eq ] , where @xmath418 consists of the most @xmath28 popular files of @xmath258 , @xmath386 , @xmath419 , @xmath420 for all @xmath421 and @xmath422 for all @xmath423 .
by lemma [ lem : asym - perf ] , we have : thus , @xmath424 is not an optimal solution , which contradicts the assumption .
therefore , by contradiction , we can prove @xmath425 for any optimal solution @xmath424 to probelm [ prob : opt - asymp - eq ] . since @xmath426 , @xmath414 and @xmath427 , we have @xmath428 .
therefore , we can prove property ( i ) of theorem [ thm : opt - prop ] .
first , we prove that there exists an optimal solution @xmath430 to problem [ prob : opt - asymp - eq ] , such that files in @xmath431 are consecutive . by lemma [ lem : asym - perf ] and @xmath425 shown in the proof of property@xmath412 ,
we have : let @xmath432 denote the most ( least ) popular file in @xmath431 .
suppose for any optimal solution @xmath424 to probelm [ prob : opt - asymp - eq ] , files in @xmath431 are not consecutive , i.e. , there exists @xmath433 satisfying @xmath434 .
now , we can construct a feasible solution @xmath435 to probelm [ prob : opt - asymp - eq ] where files in @xmath436 are consecutive as follows .
@xmath0 if @xmath437 , choose @xmath438 , @xmath439 , @xmath440 and @xmath420 for all @xmath441 . by lemma [ lem : asym - perf ]
, we have : where the inequality is due to @xmath442 and @xmath437 . @xmath0
if @xmath443 , choose @xmath444 , @xmath445 , @xmath446 and @xmath420 for all @xmath441 .
by lemma [ lem : asym - perf ] , we have : where the inequality is due to @xmath447 and @xmath448 .
@xmath0 if @xmath449 , choose @xmath438 , @xmath439 , @xmath440 and @xmath420 for all @xmath441 . by lemma [ lem : asym - perf ]
, we have : by ( [ eqn : controdiction - thm2-b2-leq ] ) , ( [ eqn : controdiction - thm2-b2-geq ] ) and ( [ eqn : controdiction - thm2-b2-eq ] ) , we know that if @xmath437 or @xmath443 , @xmath424 is not an optimal solution , which contradicts the assumption ; and if @xmath449 , we can always construct an optimal solution @xmath450 , satisfying that files in @xmath436 are consecutive .
thus , we can prove that there exists an optimal solution @xmath430 to problem [ prob : opt - asymp - eq ] , such that files in @xmath431 are consecutive .
in addition , by ( [ eqn : optimal - value - appendix - b2 ] ) , we know that whether file @xmath451 belongs to @xmath257 or @xmath258 makes no difference in the optimal successful transmission probability
. therefore , we can prove the property ( ii ) of theorem [ thm : opt - prop ] .
we prove that if @xmath268 , the most popular file @xmath452 belongs to @xmath431 for any optimal solution @xmath453 to problem [ prob : opt - asymp - eq ] .
suppose that there exists an optimal solution @xmath453 to problem [ prob : opt - asymp - eq ] , such that the most popular file @xmath452 belongs to @xmath454 .
let @xmath455 denote a file in @xmath456 .
now , we can construct a feasible solution @xmath435 to probelm [ prob : opt - asymp - eq ] , where @xmath457 , @xmath458 , @xmath459 and @xmath420 for all @xmath460 . by lemma [ lem : asym - perf ]
, we have : since @xmath461 and @xmath462 , we have @xmath463 .
thus , @xmath381 is not an optimal solution to problem [ prob : opt - asymp - eq ] , which contradicts the assumption . by contradiction
, we prove that if @xmath268 , the most popular file @xmath452 belongs to @xmath431 for any optimal solution @xmath453 to problem [ prob : opt - asymp - eq ] , and hence @xmath264 in theorem [ thm : opt - prop ] ( ii ) satisfies @xmath265 .
we prove that if @xmath464 , the most popular file @xmath452 belongs to @xmath454 for any optimal solution @xmath453 to problem [ prob : opt - asymp - eq ] .
suppose that there exists an optimal solution @xmath453 to problem [ prob : opt - asymp - eq ] , such that file @xmath452 belongs to @xmath465 .
let @xmath455 denote the most popular file in @xmath454 . based on lemma [ lem : mono - general - asym ]
, we have @xmath466 for any @xmath467 , and hence @xmath468 .
now , we can construct a feasible solution @xmath435 to probelm [ prob : opt - asymp - eq ] , where @xmath469 , @xmath470 , @xmath471 and @xmath420 for all @xmath467 . by lemma [ lem : asym - perf ]
, we have : since @xmath472 and @xmath473 , we have @xmath463 .
thus , @xmath453 is not an optimal solution to problem [ prob : opt - asymp - eq ] , which contradicts the assumption .
therefore , we prove that if @xmath464 , the most popular file @xmath452 belongs to @xmath454 for any optimal solution @xmath453 to problem [ prob : opt - asymp - eq ] , and hence @xmath264 in theorem [ thm : opt - prop ] ( ii ) satisfies @xmath267 .
x. wang , m. chen , t. taleb , a. ksentini , and v. leung , `` cache in the air : exploiting content caching and delivery techniques for 5 g systems , '' _ communications magazine , ieee _ , vol .
52 , no . 2 ,
pp . 131139 , february 2014 .
k. shanmugam , n. golrezaei , a. dimakis , a. molisch , and g. caire , `` femtocaching : wireless content delivery through distributed caching helpers , '' _ information theory , ieee transactions on _ , vol .
59 , no .
12 , pp . 84028413 , dec 2013 .
j. li , y. chen , z. lin , w. chen , b. vucetic , and l. hanzo , `` distributed caching for data dissemination in the downlink of heterogeneous networks , '' _ ieee transactions on communications _
63 , no .
10 , pp . 35533568 , oct 2015 .
d. liu and c. yang , `` cache - enabled heterogeneous cellular networks : comparison and tradeoffs , '' in _ ieee int .
conf . on commun .
( icc ) _ , kuala lumpur , malaysia , june 2016 . [ online ] .
available : http://arxiv.org/abs/1602.08255 z. chen , j. lee , t. q. s. quek , and m. kountouris , `` cooperative caching and transmission design in cluster - centric small cell networks , '' _ corr _ , vol .
abs/1601.00321 , 2016 .
[ online ] .
available : http://arxiv.org/abs/1601.00321 s. t. ul hassan , m. bennis , p. h. j. nardelli , and m. latva - aho , `` caching in wireless small cell networks : a storage - bandwidth tradeoff , '' _ ieee communications letters _
pp , no .
99 , pp . 11 , 2016 .
s. tamoor - ul - hassan , m. bennis , p. h. j. nardelli , and m. latva - aho , `` modeling and analysis of content caching in wireless small cell networks , '' _ corr _ , vol .
abs/1507.00182 , 2015 .
[ online ] .
available : http://arxiv.org/abs/1507.00182 b. zhou , y. cui , and m. tao , `` stochastic content - centric multicast scheduling for cache - enabled heterogeneous cellular networks , '' _ submitted to wireless communications , ieee transactions on _ , vol .
abs/1509.06611 , 2015 .
[ online ] .
available : http://arxiv.org/abs/1509.06611 y. cui , d. jiang , and y. wu , `` analysis and optimization of caching and multicasting in large - scale cache - enabled wireless networks , '' _ corr _ , vol .
abs/1512.06176 , 2015 .
[ online ] .
available : http://arxiv.org/abs/1512.06176 s. singh , h. s. dhillon , and j. g. andrews , `` offloading in heterogeneous networks : modeling , analysis , and design insights , '' _ ieee trans .
wireless commun .
_ , vol . 12 , no . 5 , pp .
24842497 , march 2013 .
s. m. yu and s .- l .
kim , `` downlink capacity and base station density in cellular networks , '' in _ modeling optimization in mobile , ad hoc wireless networks ( wiopt ) , 2013 11th international symposium on _ , may 2013 , pp . | heterogeneous wireless networks ( hetnets ) provide a powerful approach to meet the dramatic mobile traffic growth , but also impose a significant challenge on backhaul .
caching and multicasting at macro and pico base stations ( bss ) are two promising methods to support massive content delivery and reduce backhaul load in hetnets . in this paper , we jointly consider caching and multicasting in a large - scale cache - enabled hetnet with backhaul constraints .
we propose a hybrid caching design consisting of identical caching in the macro - tier and random caching in the pico - tier , and a corresponding multicasting design . by carefully handling different types of interferers and adopting appropriate approximations
, we derive tractable expressions for the successful transmission probability in the general region as well as the high signal - to - noise ratio ( snr ) and user density region , utilizing tools from stochastic geometry .
then , we consider the successful transmission probability maximization by optimizing the design parameters , which is a very challenging mixed discrete - continuous optimization problem . by using optimization techniques and exploring the structural properties
, we obtain a near optimal solution with superior performance and manageable complexity .
this solution achieves better performance in the general region than any asymptotically optimal solution , under a mild condition .
the analysis and optimization results provide valuable design insights for practical cache - enabled hetnets . cache , multicast , backhaul , stochastic geometry , optimization , heterogenous wireless network |
charge balance functions were suggested as a means for addressing fundamental questions concerning hadronization in relativistic heavy ion collisions @xcite .
the most pressing issue concerns whether hadronization is delayed in such reactions beyond the characteristic time scale of 1 fm / c , i.e. , is a new phase of matter created ?
a delayed hadronization of a gluon - rich medium would mean that many charge - anticharge pairs would be created late in the reaction and then be more tightly correlated to one another in momentum space .
charge balance functions are designed to identify such charge / anticharge pairs on a statistical basis .
unfortunately , the ability to identify balancing partners is compromised by two effects .
first , surplus charge , originating from the non - zero baryon number and charge of the colliding nuclei , pollutes the balance function .
secondly , interactions of a balancing pair with the other charges effectively polarize the other particles and distort the shape of the balance function . in this paper , the magnitude of such distortions are exhibited within the context of simple blast - wave models , and means for eliminating or reducing these distortions are presented .
charge balance functions are based on conditional distributions , @xmath0 here , @xmath1 counts the incidences where a particle of charge @xmath2 is observed with momentum in a region defined by @xmath3 while a particle of charge @xmath4 is observed that satisfies the momentum constraint @xmath5 .
the second constraint could be any function of the momenta of the two particles , e.g. , the relative rapidity .
put into words , the balance function measures the chance of observing an extra particle of opposite charge given the observation of the first particle .
balance functions are related to charge fluctuations which can be used to investigate similar issues @xcite .
the advantage of balance functions is that they represent a more differential measure . for a neutral system ,
every charge has an opposite balancing charge and the balance function would integrate to unity . @xmath6 the normalization is reduced if not all particles carrying the charge are included , e.g. only @xmath7 and @xmath8 are evaluated , thus neglecting the chance that the electric charge is balanced by a kaon or a baryon , or that the detector has less than a perfect acceptance .
if @xmath5 refers to the relative rapidity , and @xmath3 includes all measured particles , @xmath9 provides the probability that a balancing charge was observed with relative rapidity @xmath10 .
since much of the charge observed in a relativistic heavy ion collision should be produced at hadronization , a delayed hadronization should result in a tighter space - time correlation between balancing charges .
due to the large collective flow fields in these reactions , a tighter correlation in space - time translates into a tighter correlation between the final momenta .
therefore , a delayed hadronization should be signaled by a narrower balance function when plotted as a function of relative momentum or relative rapidity .
one of the most enticing results from rhic is the preliminary measurement of balance functions by the star collaboration @xcite . in accordance with expectations for delayed hadronization
, the balance functions appear to narrow with increasing centrality of the collision .
however , given the nascent stage of these observations and of the phenomenology , it should be emphasized that numerous questions remain concerning the interpretation of such a measurement . to that end
, several issues were pursued in a previous paper , including the effects of hanbury - brown twiss correlations , detector acceptance and the relation to charge fluctuations @xcite . in the same spirit as that paper ,
more issues will be addressed in this study . in the next section , the benefits analyzing balance functions in other observables , e.g , the invariant relative momentum , will be addressed .
in addition to allowing one to analyze the contribution from specific resonances , it will be shown that such observables help clarify other issues such as the interplay of collective flow and cooling .
balance function analyses are based on the assumption that all charges have balancing partners .
this is not true in relativistic heavy ion collisions due to the presence of the initial protons and neutrons which bring about an imbalance of baryon number , electric charge and isospin . in section [ sec : surplus ] , the distorting influence of the surplus positive charge is investigated and a modified balance function observable is proposed that would eliminate such effects .
the subsequent section contains a detailed study of the effects of inter - pair correlations . by extending the model presented in @xcite to balance functions in @xmath11
, it appears that the hanbury - brown twiss ( hbt ) correlations cause a more noticeable distortion , especially in the most central collisions .
the source of these residual effects is analyzed in detail , and the degree to which these distortions can be accounted for is discussed .
the final section presents a summary of what further work must be done in analyzing and interpreting this class of observables .
in reference @xcite balance functions were evaluated as a function of relative rapidity . like two - particle correlation functions ,
the balance function is a six - dimensional quantity and new insights can be gained by performing different cuts or binnings . specifically , we focus on performing analyses in terms of the invariant relative momentum , i.e. the relative momentum as measured by an observer moving with the velocity of the two - particle center of mass .
we find that these variables yield clearer insight for interpreting the physics of the balancing charges , as well as providing a better illumination of the distorting effects which are the subject of this study .
the relative momentum of the two particles is defined , @xmath12 here , the total momentum of the pair is @xmath13 and the center - of - mass energy of the pair is @xmath14 . for two particles of the same mass ,
the last term can be neglected .
the invariant momentum is then @xmath15 for pion - correlation studies , it is conventional to define three projections of the relative momentum , @xmath16 , @xmath17 and @xmath18 @xcite .
these components measure the projections of @xmath19 along the beam axis , the outwards direction ( defined by the pair s transverse momentum ) and the sidewards direction ( perpendicular to the pair s transverse momentum and to the beam axis ) .
motivated by the semi - boost invariant nature of the collision geometry at rhic , @xmath16 is usually measured in a reference frame moving with the beam velocity of the pair .
although not typically invoked in correlations studies , one can also perform a second outwards boost to a frame where the total spatial momentum of the pair is zero . in this frame the three components , @xmath17 , @xmath20 and @xmath16 sum to @xmath11 .
@xmath21 in terms of laboratory momenta @xmath13 and @xmath19 , these components are : @xmath22 here , @xmath23^{1/2}$ ] is the transverse momentum of the pair .
these components differ from the common convention for hbt in that @xmath17 is defined as the relative momentum in the pair frame , whereas in hbt the usual convention is to ignore the second boost which means that the three components do not sum to @xmath11 .
in fact , the factor @xmath24 in the definition of @xmath17 is simply the lorentz gamma factor corresponding to the transverse boost to the two - particle rest frame .
analyzing balance function in terms of @xmath11 simplifies interpretation with thermal models by eliminating the sensitivity to collective flow .
blast wave models are based on thermal emission from sources which move to account for the collective flow of the exploding matter .
collective flow affects the spectra , but leaves the invariant momentum differences unchanged if the two particles originate from the same space - time point of the blast wave .
hence , plotting the balance function in invariant momentum variables would minimize the confusion associated with the collective flow as the width would only depend on the local thermal properties of the individual sources .
if a particle and its balancing particle were always emitted close to one another in coordinate space , the width of the balance function would principally be a function of the breakup temperature with no sensitivity to collective flow , assuming a uniform detector acceptance . to illustrate the complications of using rapidity differences rather than @xmath11 , one may consider a thermal source where the width in @xmath25 is determined by the temperature .
the separation of two tracks in rapidity is then , @xmath26 where @xmath27 is the transverse mass of the particles .
since collective flow affects the distribution of transverse masses , the balance function widths for localized thermal sources would depend on the collective flow in the data when plotted in relative rapidity .
although it is easy to account for collective flow in a theoretical model , the interpretation of experimental results is simplified by performing the analysis in @xmath11 .
furthermore , assuming thermal emission with highly localized charge conservation , the balance function would be isotropic with respect to the direction of the relative momentum , e.g. , the width in @xmath18 would equal the width in @xmath16 .
for early production of charge , one expects string dynamics or diffusion to lead to an anisotropic balance function as the balancing charges should separate significantly in coordinate space along the beam axis due to the extremely large velocity gradient along the beam axis at early times , @xmath28 .
thus , in addition to the width of the balance function in @xmath11 , the behavior of the anisotropy as a function of the collision s centrality provides a crucial test of the mechanism for charge creation and transport . to illustrate the sensitivity of a balance function in terms of these analyses , we consider a simple blast wave model where the collective transverse motion is assumed to rise linearly with the radius . of the numerous parameterizations of the blast wave model , it is assumed that the sources have transverse rapidities governed by a simple distribution , @xmath29 here , @xmath30 is the transverse rapidity , @xmath31 .
the distribution of longitudinal rapidities is assumed to be uniform . for our calculation
we assume that these sources emit isotropically in the source frame according to a temperature @xmath32 .
a balancing positive and negative pion are assumed to be emitted from sources with the same longitudinal and transverse rapidity .
figure [ fig : tempdependence ] illustrates the sensitivity to the temperature by presenting balance functions for three temperatures , 90 mev , 120 mev and 150 mev .
the balance function is clearly narrower for lower breakup temperatures .
the results of fig . [ fig : tempdependence ] are insensitive to the choice of @xmath33 .
however , it must be stressed that the sensitivity would return if the balance function is analyzed in a finite acceptance .
the balance function could also be binned in any of the three projections , @xmath16 , @xmath17 and @xmath18 , rather than in @xmath34 .
if the balancing pairs were to always originate from sources with the same collective velocity , the balance function would be identical in all three variables .
however , if the balancing particles were to diffuse relative to one another , the shape of the balance function might become decidedly non - isotropic .
for instance , if charge is created early in a rhic collision , the balancing charges might easily separate along the beam axis and ultimately be emitted from regions with different rapidities .
figure [ fig : sigmaeta ] presents the widths of balance functions assuming that the balancing particles independently dissipated and were each ultimately emitted with sources moving with a spread of rapidities characterized by @xmath35 .
a gaussian form for the diffusion was assumed for the distribution of source rapidities , @xmath36 .
@xmath37 this extends the distribution of @xmath16 while leaving the distribution of @xmath18 unaffected and the distribution of @xmath17 only slightly affected by boost effects . as can be seen in fig .
[ fig : sigmaeta ] , the disparity in the three widths should be easily observed for this example where the temperature was chosen to be 120 mev and the radial collective velocities were between zero and @xmath38 .
it should be difficult to discern the difference between thermal broadening and dissipation of balancing charges into regions with different collective flow .
however , other observables provide insight into breakup temperature , mainly the comparison of proton and pion spectra @xcite .
once one knows the breakup temperature , it is possible to fit parameters that describe the diffusive spread , e.g. @xmath35 . furthermore , a thermal fit to data where the diffusive terms are set to zero provides an upper bound for the breakup temperature . for the reasons above ,
much of the analysis of the following sections will be given in @xmath11 .
an additional advantage of using @xmath11 is that it allows one to identify the contributions from specific resonances which contribute peaks to the balance function when plotted in @xmath11 .
it is our hope that experimental analyses will also switch to these variables .
not all charges have balancing partners . in a au+au collision at rhic ,
the two gold nuclei provide 158 unbalanced protons and 236 unbalanced neutrons .
these pollute the balance function by providing unbalanced electric charge , baryon number and isospin . for detectors like star
, these effects are lessened by the fact that most of the surplus charge is at high rapidity and outside the experimental acceptance .
however , the effect should become more significant if the balance function is constructed for a set of charges , e.g. @xmath39 , for which there is a significant imbalance of one charge vs. the opposite charge . our goal in this section is to offer a revised procedure for producing balance functions from data that would subtract the pollution due to the surplus charge .
more precisely , we wish to define a balance function that would ignore any additional unbalanced charges that are not correlated with one another or with pair - wise created charges . in order to demonstrate the effects of the polluting surplus charge
, we introduce a notation where distributions , @xmath40 , count charges which are divided into three categories .
the subscripts `` + '' and `` - '' will refer to positive and negative charges which are created in tandem . the subscripts `` @xmath41 '' will denote the surplus positive charge . the balance function will be re - evaluated after inserting the replacements , @xmath42 into eq .
( [ eq : balancedef ] ) .
@xmath43 in deriving eq .
( [ eq : balancedef_wdelta ] ) an explicit symmetry between the positive and negative charges has been assumed , i.e. , @xmath44 , @xmath45 , and @xmath46 .
the first term in this expression is proportional to the unpolluted balance function , @xmath47 .
the second term can be simplified by assuming that the surplus charges are uncorrelated with the other charges and that they are also uncorrelated with themselves , aside from overall conservation of charge .
@xmath48 where @xmath49 represents the maximum integrated surplus charge .
the balance function can then be expressed as : @xmath50 if the charges used to construct the balance function obey strict charge conservation , a perfect detector would satisfy the normalization conditions , @xmath51 after inserting eq .
( [ eq : normconditions ] ) into eq .
( [ eq : balancepollution ] ) , one can see that the normalization of the balance function is unchanged by the surplus charge .
however , the shape is altered as the balance function is comprised of two components .
the first term in eq . ( [ eq : balancepollution ] ) describes the separation of balancing charges , while the second term is governed by the separation of two random balancing charges .
the relative weights of the two terms is determined by the fraction of the charge which owes itself to a surplus in the initial state .
thus , the effect of the surplus charge is to dampen the contribution from the balancing charges and to average in a second contribution . for the star detector at rhic ,
this second term is fairly small even for protons as the number of surplus protons is less than 10 per unit rapidity in central collisions @xcite . given that charge conservation constraints would suggest @xmath52 , the effect of the extra charge is to first dampen the balance function by approximately 15% , and secondly to add in a second component whose width is characteristic of the acceptance , and whose magnitude is only one or two percent of the contribution from balancing charges . in order to eliminate the contribution from surplus charge and
determine @xmath47 from experiment , one can consider an object similar as to what is used to create the balance function numerator using mixed events .
this object will be referred to as @xmath53 and will be constructed from mixed events , where @xmath54 signifies that the charges @xmath2 and @xmath4 which satisfy the momentum constraints @xmath3 and @xmath5 are chosen from separate events .
@xmath55 since the counts for different events are independent , @xmath56 , @xmath57 and @xmath58 .
thus , @xmath59 becomes @xmath60 one could define a similar object using pairs from the same event , @xmath61
again if one assumes that the only correlation between surplus charges involves a constraint on the overall number of surplus charges , @xmath62 one can then generate the desired balance function @xmath63 by combining @xmath64 and @xmath59 , @xmath65 where @xmath66 is @xmath67 if the surplus charges are positive and is @xmath68 if the surplus charges are negative .
it should be emphasized that this derivation assumed that the detector has equal acceptance for positive and negative charges .
the ability of @xmath47 to ignore the polluting surplus charge is based on the assumption that the surplus charges are uncorrelated with one another and are equally correlated with the `` + '' and `` - '' charges .
as discussed in the next section and in @xcite , such correlations can be important , especially at small relative momentum . to illustrate the importance of these corrections
, we generate a @xmath39 balance function from a simple model of a boost - invariant emission of particles governed by a temperature of 120 mev and a maximum transverse velocity of 0.7@xmath69 .
it is assumed that the number of protons per unit rapidity is 28 and that the number of antiprotons is 21 to be consistent with measurements from rhic @xcite .
the polluted balance function as described in eq .
( [ eq : balancepollution ] ) is displayed in fig .
[ fig : surplus ] along with the corrected balance function @xmath47 .
this calculation is generated by assuming that particles were emitted from sources with random rapidities , but that two balancing particles are emitted from sources with the same velocity .
the parameter @xmath49 used in eq .
( [ eq : balancepollution ] ) is assumed to be 158 .
two additional modifications have been added to eq .
( [ eq : balancepollution ] ) in order to more fairly illustrate the magnitude of the effect of the surplus charge .
first , the function @xmath47 was scaled down by 40% to account for the fact that the charge of an antiproton is often balanced by a neutron or by a @xmath70 .
secondly , the simulated momenta were put through an acceptance filter which crudely mocks the acceptance of the star detector at rhic .
particles were required to have a @xmath71 greater than 100 mev / c and a momentum of magnitude less than 700 mev / c .
the pseudo - rapidities were confined to a region of midrapidity , @xmath72 .
as illustrated in fig .
[ fig : surplus ] , the effects of the extra charge are mainly to dampen the balance function . the importance of correcting for the surplus charge would certainly be magnified if one were to analyze balance functions from sps or ags collisions where the fraction of extra protons is much higher .
these corrections are not - model dependent , and the corrected balance functions exactly reproduce @xmath47 .
however , it should be emphasized that this statements relies on the assumption that the surplus charge is uncorrelated with other surplus charges , and with the pair - wise created charges .
the balance function is implicitly predicated on the assumption that there are no residual correlations between a given charge and all other charges besides its balancing partner , i.e. , all other charges are statistically eliminated from the distribution by the like - sign subtraction .
not all correlations cause problems . for instance , flow correlations tend to be identical between particles of the opposite charge or the same charge and thus fall out of the balance function . on the other hand final - state interactions
involve all the other charges and depend sensitively on the relative signs of the charges .
this distortion can rise linearly with the multiplicity since the number of charges with which a given charge can correlate rises linearly with the multiplicity .
however , correlation functions tend to approach unity at higher multiplicity in accordance with expectations for increasing source size .
this makes the resulting multiplicity dependence of the distortion non - trivial .
a method for estimating the distortion to the balance function from residual interactions was provided in @xcite .
the same method is applied here . for every balancing pair , @xmath73 and @xmath74
, one must consider the correlation weight with other pairs whose momenta are @xmath75 and @xmath76 .
the weight , @xmath77 can be estimated : @xmath78 ideally , the balance function would isolate the @xmath79 pair and the interaction with the @xmath80 pair would cancel from the subtraction , @xmath81 . the correlations will lead to distortions if @xmath82 i.e. , distortions are caused only by those interactions which differ between same - sign and opposite - sign particles .
for instance , an isoscalar exchange of between pions would not bring on a distortion , but a coulomb interaction or identical - particle interference would provide a source for distortion .
we simulate these effects for @xmath83 balance functions with the same blast wave model described in section [ sec : qinv ] .
the source of the @xmath79 pairs was chosen to move with one randomly chosen velocity , while the @xmath80 source was chosen to move with a different velocity .
the distributions were calculated for the balance function numerators using the four particles , but rather than incrementing he distributions by unity , the distributions were incremented by the weight described in eq .
( [ eq : cweight ] ) . by considering the contributions due to the extra weight separately from the usual contributions arising between the balancing particles , and weighting them appropriately for the given value of @xmath84 ,
balance functions were calculated with and without the distortions .
in addition to the blast - wave parameters , the residual - interaction distortion is sensitive to the form for the two particle correlations and to the pion multiplicity .
the multiplicity of charged pions was chosen to be 300 for both positive and negative pions , roughly consistent with measurements at rhic @xcite .
the correlation functions were chosen to correspond to gaussian sources of radius , @xmath85 fm , again consistent with measurements at rhic @xcite .
a more sophisticated treatment would account for the @xmath71 dependence and the directional dependence of the radius .
the correlation calculation presented here accounted for both identical - particle interference and for the mutual coulomb interaction .
the results of the calculation are shown in fig.s [ fig : balancehbt_qinv ] and [ fig : balancehbt_dely ] .
a crude filter for the star detector at rhic was applied and it was assumed that 70% of the pions had their charge balanced by other pions , which affects the normalization of the balance function .
residual interactions can either strengthen or diminish the balance function depending on the relative momentum . at very small relative momentum
, the balance function at small relative momentum rises due to the coulomb enhancement of the @xmath83 correlation function and the coulomb repulsion of the @xmath86 and @xmath87 correlation functions . for values of @xmath11 larger than a few mev / c but less than @xmath88 mev / c , the identical - particle interference which enhances the @xmath89 and @xmath87 correlation functions , diminishes the balance function since same - sign pairs contribute negatively to the balance function . at larger relative momenta
, coulomb effects again dominate .
the effects are less dramatic when the balance function is viewed as a function of relative rapidity . the distortion of the balance function in fig .
[ fig : balancehbt_qinv ] is dominated by coulomb effects at large momentum .
the correlation weights are driven by the squared quantum wave function .
however , the correlation for large values of @xmath90 can be understood by considering the classical analog to the wave function . as shown in ref .
@xcite , the classical analog is : @xmath91 where @xmath92 is the fine structure constant , the product of the charges of the two species is @xmath93 , @xmath94 is the reduced mass and @xmath95 and @xmath96 are the initial and final relative momenta .
thus , the effects of coulomb interactions , in the classical limit , only diminish as a function of @xmath97 . by averaging @xmath98 over a gaussian source characterized by the gaussian source size @xmath99 , one can find the asymptotic form for the classical correlation function . @xmath100
the classical result for the @xmath83 correlation function is compared to the quantum result for a six - fm source in fig .
[ fig : corrtail ] .
the agreement is remarkable for @xmath101 mev / c when @xmath102 , especially for the opposite - sign case where there is no identical - particle interference .
even though the correlation function goes to zero proportional to @xmath103 , the phase space is increasing as @xmath104 .
thus , the coulomb interaction remains important to remarkably large momenta . for more central collisions , the value of @xmath99 rises , but the number of particles with which a given particle is correlated also rises .
if the multiplicity scales as @xmath105 , it is clear that the coulomb distortion will become acute for central collisions .
strong - interaction distortions have not been considered in these calculations . in terms of the @xmath106 phase shifts , @xmath107 , the contribution to the correlation function from strong interactions
can be approximated by the relation @xcite , @xmath108 the strength of the strong - interaction correlation falls much more quickly with @xmath99 than does coulomb - induced correlation .
if @xmath105 were to scale linearly with multiplicity , the effect of the strong interaction on the balance function would be roughly independent of multiplicity or centrality .
this difference in the behavior derives from the fact that a given pion interacts with only its neighbors through the strong interaction while it may interact with nearly all particles through the coulomb interaction .
if the breakup density is independent of centrality , the number of neighbors stays constant and the distortion to the balance function from strong interactions should not be strongly centrality dependent
. on the other hand , the coulomb distortion interaction should be much stronger for central collisions than for peripheral collisions .
the ingredients for calculating the distortion were the correlation weights and spectra along with the procedure for generating the pairs , @xmath109 and @xmath110 . in principal ,
the spectra and correlation weights can be taken or inferred from data without introducing a theoretical model .
however , the generation of the pairs can not be extracted directly from data due to the correlations between @xmath73 and @xmath74 and those between @xmath75 and @xmath76 .
since particles are produced pairwise , it is necessary to include these correlations because the inter - pair interaction must attract pairs rather than single particles .
i.e. , the net charge in the medium can not change , but it can be polarized . at face value
, this is an explicit model dependence .
however , the parameters that govern the correlation between @xmath2 and @xmath4 and between @xmath69 and @xmath111 are precisely those parameters used to model the undistorted balance function .
thus , no additional model parameters would be introduced to calculate the distortion .
thus , the distortion from residual interactions can not be subtracted from experimental results in a model - independent fashion , but it can be modeled theoretically without additional parameters . given the significant effects from inter - pair correlations , it is imperative that the balance function analyses correct for these distortions . fortunately , the corrections can be confidently modeled , and the robustness of the balance function is not compromised . however , this conclusion is predicated on an understanding of the two - particle correlations . since a correlation of a fraction of a percent can significantly alter the balance function , the issue of strong - interaction corrections to the balance functions should be revisited .
strong - interaction effects can be divided into two categories .
the first category would be @xmath112-channel interaction which have particle - antiparticle channels , e.g. @xmath113 .
but , this source should not be considered as a distortion since the two pions are indeed a balancing pair . for instance , if all pions resulted from @xmath114 decays , the balance function would peak at the invariant mass of the @xmath115 , and provide an important clue as to the creation mechanism for pions .
such resonant contributions can be calculated in a microscopic model or in a thermal calculation based on the canonical ensemble .
a second source of strong - interaction effects is the interaction with other bodies through non - resonant interactions .
since the strong interaction is short range , this interaction should involve only a few neighbors . for large sources ,
the coulomb interaction provides a larger effect on two - particle correlation function than does the strong interaction .
nonetheless , it would be worthwhile to better quantify the significance or insignificance of the strong interaction .
charge balance functions were developed with the hope of identifying balancing charges on a statistical basis .
two effects prevent the like - sign subtraction from accomplishing this goal to high precision . as shown in sec .
[ sec : surplus ] , the excess nucleons coming from the colliding nuclei provide a modest pollution to the balance function in measurements at mid - rapidity at rhic .
fortunately , these effects can be easily subtracted .
the second source of distortion derives from the inter - pair interaction of balancing charges with other particles in the medium .
these distortions become more important in high multiplicity events . as shown in sec .
[ sec : hbt ] , for high multiplicity events these distortions are most strongly affected by the coulomb interaction .
these effects are also more noticeable for balance functions calculated in @xmath11 than they are for balance functions calculated in relative rapidity .
although it is difficult to subtract these distortions in a model - independent fashion , it is straight - forward to include these effects in a theoretical treatment .
in addition to the typical parameters one would use to model balance functions , modeling the distortion requires only an additional understanding of two - particle correlations .
as these correlations can be extracted from measurement , the distortion from inter - pair interactions can be modeled quite confidently . in central collisions ,
this distortion can be a 20% effect , and if the correlation functions are understood to the 90% level , the residual systematic uncertainty is probably of the order of one or two percent .
bass , p. danielewicz and s. pratt , phys .
. lett . * 85 * , 2689 ( 2000 ) .
s. jeon and s. pratt , phys .
c65 , 044902 ( 2002 ) .
m. asakawa , u. heinz and b. muller , phys .
lett . * 85 * , 2072 ( 2000 ) .
m. asakawa , u.w .
heinz and b. muller , nucl .
a*698 * , 519 ( 2002 ) .
v. koch , acta phys .
b*33 * , 4219 ( 2002 ) . v. koch , m. bleicher and s. jeon , nucl .
a*698 * , 261 ( 2002 ) .
m. bleicher , s. jeon and v. koch , phys .
c * 62 * , 061902 ( 2000 ) .
s. jeon and v. koch , phys . rev . lett . * 85 * , 2076 ( 2000 ) .
m. abdel - aziz and s. gavin , proceedings of quark matter 2002 , to appear in nucl .
a. c. pruneau , s. gavin and s. voloshin , phys .
c*66 * , 044904 ( 2002 ) .
s. gavin and j.i .
kapusta , phys .
c*65 * , 054910 ( 2002 ) .
l. ray ( star collaboration ) , proceedings of quark matter 2002 , nantes , france , to appear in nucl .
a. s. pratt , phys .
rev . d*33 * , 1314 ( 1986 ) .
bertsch , nucl .
a*498 * , 173c ( 1989 ) .
t. csrg , j. zimnyi , j. bondorf , h. heiselberg , and s. pratt , phys . lett . *
b241 * , 301 ( 1990 ) .
k. adcox , et al .
lett . * 88 * , 242301 ( 2002 )
. j.h . lee ( brahms collaboration ) , proceedings of quark matter 2002 , nantes , france , to appear in nucl .
a. t. chujo ( phenix collaboration ) , proceedings of quark matter 2002 , nantes , france , to appear in nucl .
a. c. adler , et al . ,
87 * , 082301 ( 2001 ) .
kim , r.t .
de souza , c.k .
gelbke , w.c . gong and s.
pratt , phys .
c*45 * , 387 ( 1992 ) .
jennings , d.h .
boal and j.c .
shillcock , phys .
c*33 * , 1303 ( 1986 ) .
blast - wave predictions of @xmath83 balance functions are displayed for three temperatures , assuming the balancing pions are always emitted thermally from sources with identical source velocities .
when plotted in @xmath11 , the shape depends only on the breakup temperature .
calculations are shown for @xmath116 ( squares ) , @xmath117 ( triangles ) and @xmath118 mev ( circles ) .
, scaledwidth=70.0% ] the width of the balance function is shown for the three momentum components .
the calculations assumed a blast wave scenario with the collective velocities of the source points for balancing pions being separated longitudinally according to a gaussian distribution of width @xmath35 .
the calculations assumed a breakup temperature of 120 mev and a maximum transverse collective velocity of 0.7@xmath69 .
, scaledwidth=70.0% ] proton - antiproton balance functions are shown for a blast wave model with and without corrections for the surplus unbalanced protons
. the corrected balance function ( circles ) is constructed assuming a breakup temperature of 120 mev , a maximum transverse velocity of 0.7@xmath69 , and @xmath119 .
the calculation was scaled down by 40% to account for balancing of the proton s charges by other species . the distorted balance function ( squares )
is based on the proton excess as measured by the brahms collaboration .
both balance functions were filtered through the star acceptance .
, scaledwidth=70.0% ] as a function of @xmath11 , @xmath83 balance functions from a blast - wave model are shown with ( circles ) and without ( squares ) the distorting effects of inter - pair interactions .
the model assumed a breakup temperature of 120 mev , a maximum transverse velocity of 0.7@xmath69 , and @xmath119 .
the undistorted balance function was scaled by 70% to account for balancing by other species , and both balance functions were filtered by the star acceptance . the significant enhancement for momenta between 60
mev / c and 400 mev / c owes itself to the coulomb interaction between pions .
, scaledwidth=70.0% ] correlations for same - sign ( squares ) and opposite - sign ( circles ) pions are shown for a gaussian source of size @xmath121 fm in the upper panel . to illustrate the coulomb effects that force @xmath122 to approach unity as @xmath123 , @xmath124 is multiplied by @xmath125 and displayed in the lower panel .
the lines represent the constant expected for coulomb interactions described in eq .
( [ eq : corrtailgauss ] ) .
since phase space increases as @xmath125 , inter - pair correlations distort the balance functions for relative momenta of several hundred mev / c .
, scaledwidth=70.0% ] | charge balance functions provide insight into critical issues concerning hadronization and transport in heavy - ion collisions by statistically isolating charge / anti - charge pairs which are correlated by charge conservation .
however , distortions from residual interactions and unbalanced charges cloud the observable . within the context of simple models ,
the significance of these effects is studied by constructing balance functions in both relative rapidity and invariant relative momentum .
methods are presented for eliminating or accounting for these distortions . |
agent - based models represent an efficient way in exploring how individual ( microscopic ) behaviour may affect the global ( macroscopic ) behaviour in a competing population .
this theme of relating macroscopic to microscopic behaviour has been the focus of many studies in physical systems , e.g. , macroscopic magnetic properties of a material stem from the local microscopic interactions of magnetic moments between atoms making up of the material .
in recent years , physicists have constructed interesting models for non - traditional systems and established new branches in physics such as econophysics and sociophysics .
the minority game ( mg ) proposed by challet and zhang @xcite and the binary - agent - resource ( b - a - r ) model proposed by johnson and hui @xcite , for example , represent a typical physicists binary abstraction of the bar attendance problem proposed by arthur @xcite . in mg ,
agents repeatedly compete to be in a minority group .
the agents have similar capabilities , but are heterogeneous in that they use different strategies in making decisions .
decisions are made based on the cumulative performance of the strategies that an agent holds .
the performance is a record of the correctness of the predictions of a strategy on the winning action which , in turn , is related to the collective behaviour of the agents .
thus , the agents interact through their decision - making process , creation of the record of winning actions , and strategy selection process .
interesting quantities for investigations include the statistics of the fraction of agents making a particular choice @xmath1 every time step and the variance or standard deviation ( sd ) @xmath2 of this number @xcite .
these quantities are related in that knowing the distribution of @xmath3 , one may obtain @xmath2 .
the mg , suitably modified , can be used to model financial markets and reproduce stylized facts .
the variance , for example , is a quantity related to the volatility in markets @xcite .
recently , we proposed a theory of agent - based models based on the consideration of decision - making and strategy dynamics @xcite .
the importance of the strategy selection dynamics has been pointed out by dhulst and rodgers @xcite .
this approach @xcite , which we refer to as the strategy - ranking theory ( srt ) , emphasizes on how the strategies performance ranking pattern changes as the game proceeds and the number of agents using a strategy in a certain rank for making decisions .
it is recognized that the srt has the advantages of including tied strategies into consideration and avoiding the troublesome in considering each strategy s performance separately .
the theory , thus , represents a generalization of the crowd - anticrowd theory @xcite to cases with tied strategies and strategy ranking evolutions two factors that are particularly important in the so - called informationally efficient phase of the mg .
the theory has been applied successfully to explain non - trivial features in the mean success rate of the agents in ( i ) mg with a population of non - networked @xcite or networked agents @xcite , ( ii ) mg with some randomly participating agents @xcite , and ( iii ) b - a - r model with a tunable resource level @xcite . in this conference paper
, we aim to illustrate the basic ideas of srt .
in particular , we present results based on srt in evaluating the distribution of @xmath1 and @xmath2 , in the efficient phase of mg in non - networked and networked populations . validity of the results of our theory is tested against results obtained by numerical simulations .
while the srt was developed within the context of mg , many of the ideas are should also be appliable to a wide range of agent - based models .
the basic mg @xcite comprises of @xmath4 agents competing to be in a minority group at each time step .
the only information available to the agents is the history .
the history is a bit - string of length @xmath0 recording the minority ( i.e. , winning ) option for the most recent @xmath0 time steps .
there are a total of @xmath5 possible history bit - strings .
for example , @xmath6 has @xmath7 possible histories of the winning outcomes : @xmath8 , @xmath9 , @xmath10 and @xmath11 . at the beginning of the game
, each agent picks @xmath12 strategies , with repetition allowed .
they make their decisions based on their strategies .
a strategy is a look up table with @xmath5 entries giving the predictions for all possible history bit - strings .
since each entry can either be ` 0 ' or ` 1 ' , the full strategy pool contains @xmath13 strategies .
adaptation is built in by allowing the agents to accumulate a merit ( virtual ) point for each of her @xmath12 strategies as the game proceeds , with the initial merit points set to zero for all strategies .
strategies that predicted the winning ( losing ) action at a given time step , are assigned ( deducted ) one virtual point . at each turn
, the agent follows the prediction of her best - scoring strategy . in case of tied best - scoring strategies
, a random choice will be made to break the tie . in the present work
, we will focus on the regime where @xmath14 , i.e. , the efficient phase .
in mg literature , a parameter @xmath15 is defined with @xmath16 characterizing the efficient phase @xcite .
features in this regime is known to be dominated by the crowd effect @xcite .
a quantitative theory in this regime would have to include the consideration of frequently occurred tied strategies into account , as the dynamics in this regime is highly sensitive to the agents strategy selection . in what follows
, we introduce the basic physical picture of the strategy ranking theory and apply it to evaluate the distribution in the fraction of agents making a particular choice @xmath17 and the variance @xmath18 from an analytic expression for non - networked and networked populations .
to put our discussions into proper context , we will first present the numerical results of the quantities that we are focusing on .
let @xmath1 be the fraction of agents taking the action 1 " ( or 0 " ) at time step @xmath19 .
as the game proceeds , there will be a time series @xmath1 .
we may then analyze these values of @xmath1 by considering the distribution or probability density function @xmath17 , where @xmath20 is the probability of having a value within the interval @xmath3 to @xmath21 . in using the mg for market modelling
, @xmath1 can be taken to be the fraction of agents deciding to buy ( or sell ) an asset at time @xmath19 . in the context of the el farol bar attendance problem @xcite , @xmath1
may be taken to be the fraction of agents attending the bar .
note that every realization of the mg may have a different distribution of strategies among the agents and a different initial bit - string to start the game .
these details do not affect the main results reported here , especially when we consider cases deep into the efficient phase , i.e. , when @xmath22 . to illustrate the point ,
we have carried out detailed numerical simulations for the simplest case of @xmath23 and @xmath24 .
figure 1 shows the numerical results ( squares ) of @xmath17 for systems with two different sizes ( @xmath25 and @xmath26 ) , with the aim of emphasizing the size effect on @xmath17 . notice that the distribution consists of a few peaks ( five peaks for the case of @xmath23 and @xmath24 ) , indicating that as the game proceeds the number @xmath1 jumps among values characterized by these peak values . for larger population ,
the peaks are sharper . also shown in fig.1 are the results of the strategy ranking theory ( lines ) .
the theoretical results are in reasonably agreement with numerical results .
we defer the discussion on obtaining the theoretical results to the next section . besides the typical results shown in fig.1 , we have studied the variance @xmath18 in the following way .
we carried out numerical simulations in many realizations using different values of @xmath0 and @xmath4 , with @xmath4 up to @xmath27 and @xmath0 up to @xmath28 .
for each run , a value of @xmath18 is obtained . to facilitate comparison with theory
, we select those data that are deep in the efficient phase , i.e. , with @xmath29 and plotted them ( black dots ) in fig.2 to show the dependence of @xmath30 on @xmath0 .
the data points do not show significant scatter , and essentially fall on a line .
also included in the figure are two ( dashed ) lines corresponding to two approximations within the crowd - anticrowd theory @xcite .
these approximations assume that all the strategies can be ranked at every time step without tied virtual points .
one of them assumes that the popularity rankings , i.e. , ranking based on the number of agents using a strategy , of a strategy and its anti - correlated partner are uncorrelated and gives an expression for @xmath30 for cases with @xmath24 as @xcite @xmath31.\ ] ] another approximation is that the ranking of strategies are highly correlated .
for example , the anti - correlated partner of the momentarily most - popular strategy is the least - popular one , and so on .
this leads to another expression within the crowd - anticrowd theory @xcite : @xmath32.\ ] ] we note that for small values of @xmath0 , the numerical data fall within the two crowd - anticrowd approximations , with neither of the approximations capturing the @xmath0-dependence of @xmath30 .
as will be discussed later , the strategy ranking theory gives an _ analytic _ expression for @xmath30 that captures the @xmath0-dependence very well in the small @xmath0 regime where the criteria @xmath33 is satisfied to a fuller extent .
we proceed to discuss how we could obtain the analytic results shown in figs.1 and 2 , within the strategy ranking theory .
details of the theory can be found in @xcite .
here we briefly summarize the key ideas , with the aim to make the theory physically transparent .
we note that in mg and other agent - based models of competing populations , it is the interplay between decision - making , strategy selections , and collective response that leads to the non - trivial and often interesting global behaviour of a system . with this in mind , the strategy performance ranking pattern is of crucial importance . at any time step
, the strategies can be classified into @xmath34 ranks , according to the virtual points of the strategies .
the momentarily best - performing strategy ( or strategies ) belongs ( belong ) to rank-1 , and so on . at the beginning of the game
, all strategies are tied that thus they all belong to the same rank .
this is also the case when the strategies are all tied during the game .
thus , the lower bound of @xmath35 is zero .
it is also noted that there are two different kinds of behaviour in the ranking pattern _ after _ a time step : ( i ) the number of different ranks
_ increases _ and such a time step is called an even " time step , and ( ii ) the number of different ranks _ decreases _ and such a time step is called an odd " time step .
take , for example , a time step at which the strategies are all tied before decision .
regardless of the history based on which the agents decide _ and _ the wining outcome after the agents decided , the strategies split into two ranks , i.e. , @xmath35 increases from 0 to 1 after the time step .
half of the strategies belong to the better rank and half to the worse rank , as half of the strategies would have predicted the correct winning outcome for the history concerned . generally speaking ,
the underlying mechanism for this splitting is that _ there is no registered virtual point or stored information in the strategies for the history concerned_. we call this kind of time steps even " time steps because this is what would happen when the population encounters a history for decision that had occurred an even number of times since the beginning of the game , not counting the one that is currently in use for decisions .
the parameter @xmath35 has another physical meaning .
it is the number of history bit - strings that have occurred an odd number of times since the beginning of the game , regardless the current history in use for decisions .
since there are at most @xmath5 history bit - strings for a given @xmath0 , the upper bound of @xmath35 is @xmath5 .
thus we have @xmath36 .
therefore , every time step as the game proceeds can be classified as even " or odd " , together with a parameter @xmath35 . for @xmath37 when all the strategies are tied , the time step is necessarily an even time step . for @xmath38 where there are @xmath39 ranks
, the time step is necessarily an odd time step since all the histories have occurred an odd number of times , including the current history in use for decisions . noting that the total number of strategies is @xmath13 , there are in general several strategies in a certain ranking . in this way , the theory takes explicit account of cases of tied strategies . for even time steps ( regardless of the value of @xmath35 )
, there is no registered virtual points in the strategies for the current history .
therefore , _ even time steps are characterized by agents making random decisions _
@xcite . using a random walk argument
, the distribution @xmath40 is a normal distribution independent of @xmath35 , with a mean @xmath41 and a variance @xmath42 , i.e. , @xmath43 it turns out that the part of the distribution around @xmath44 shown in fig.1 originates from the even time steps . for odd time steps ,
there are registered virtual points or stored information in the strategies for the current history .
this is the origin of the crowd effect @xcite , which is fundamental to the understanding of collective response in the class of agent - based models based on mg . in this case , the momentarily better performing strategies have predicted the correct action in the last occurrence of the current history in use for decision .
there will then be more agents using these better - performing strategies for decisions .
however , the number is too large , hence forming a crowd , that the winning action in the last occurrence becomes the losing action in this turn .
this is the anti - persistent nature or double periodicity of mg @xcite . using the strategy ranking theory , we know that there are @xmath45 ranks among the strategies for time steps labelled @xmath35 .
the ratio of the fractions of strategies in different ranks is given by @xcite @xmath46 , which are simply the numbers in the pascal triangles .
given that the agents use their best - performing strategy for decision , we can readily count the number of agents using a strategy in a particular rank . as mentioned , the better - performing strategies are more likely to lead to wrong predictions at odd time steps .
this can be modelled by a winning probability at odd time steps of the form of @xmath47 for a strategy belonging to rank-@xmath48 , for a given value of @xmath35 @xcite . putting the information together
, we arrive at the probability density function @xmath49 for @xmath50 . the distribution @xmath49 is given by normal distributions centered at the mean values of @xmath51 with a variance @xmath52 applying eq .
( 4 ) to the results for @xmath23 in fig.1 , we immediately identify that the peaks in @xmath17 at @xmath53 and @xmath54 are originated from odd time steps corresponding to @xmath55 and the peaks at @xmath56 and @xmath57 are originated from odd time steps corresponding to @xmath58 .
these peaks are more noticeable in fig .
1(b ) when the population size is large . in eq .
( 5 ) , the binomial coefficients should formally be expressed in terms of gamma functions , so that when the lower index in the coefficient becomes negative , @xmath59 vanishes .
this is the case for @xmath60 , and the corresponding distribution will then be very sharp .
this is , for example , the case for the sharp peaks at @xmath53 and @xmath54 in fig .
( 1 ) . to obtain an expression for the overall @xmath17 , including both even and odd time steps and all possible values of @xmath35
, we need to take a weighted average over the occurrence of odd and even time steps @xcite .
the resulting expression is @xmath61,\ ] ] where the factor @xmath62 is the probability of having @xmath35 history bit - strings occurred an odd number of times .
the factor @xmath63 is the probability that given @xmath35 , the time step is odd .
applying eq .
( 6 ) to the case of @xmath23 , we obtain the results ( lines ) shown in fig . 1 .
we note that the expression in eq .
( 6 ) is also applicable to @xmath64 , as long as the efficient phase criteria is satisfied .
the calculation of the variance follows from the definition @xmath65 where @xmath66 is the mean value of @xmath3 and the average @xmath67 represents a time average . replacing the time average by invoking the probability density function @xmath17 , we have @xmath68 + ( 1-\frac{\kappa}{2^m } ) \sigma_{even}^2 \right\}\\ & \approx & \sum_{\kappa=0}^{2^m } \frac{c_{\kappa}^{2^m}}{2^{2^m } } ( \frac{\kappa}{2^m } ) ( \frac{c_{\kappa-1}^{2 \kappa - 1}}{2^{2\kappa}})^2 \nonumber \\ & = & \sum_{\kappa=0}^{2^m } \frac{c_{\kappa}^{2^m}}{2^{2^m } } ( \frac{\kappa}{2^m } ) \left(\frac{1}{2 } \prod_{q=1}^{\kappa } ( 1 - \frac{1}{2q } ) \right)^{2},\end{aligned}\ ] ] where the approximation is valid for @xmath69 .
( 9 ) is an _
expression for @xmath18 .
the last two expressions are equivalent and one may use whichever convenient in obtaining numerical values from eq . ( 9 ) .
several remarks are worth mentioning .
firstly , we note that the expression of @xmath18 is closely related to the analytic expression for the winning probability reported in @xcite , from which an alternative approach arriving at the same result is possible @xcite .
secondly , the results from eq . ( 9 ) are plotted ( open squares ) in fig.2 .
we note that the strategy ranking theory does capture the @xmath0-dependence of @xmath30 , with good agreement with numerical simulation results in the range where the criteria @xmath33 is better fulfilled @xcite .
the success of the theory stems from the inclusion of tied strategies , as each rank typically consists of a number of strategies . in the simplest case of @xmath23 , for example
, there are tied strategies in _ every _ time step of the game .
the better agreement with numerical results when compared with the crowd - anticrowd approximations is thus an indication of the importance of ( i ) the tied strategies and ( ii ) the time evolution of the ranking pattern from time step to time step . in mg , both
the number of tied strategies , i.e. , number of strategies belonging to the same rank , and the time evolution of strategy ranking pattern can be readily found .
thirdly , the result eq .
( 9 ) is interesting in that there have been much effort in trying to re - scale numerical results of @xmath18 as a function of the parameter @xmath15 so that results from systems of different values of @xmath4 and @xmath0 can be collapsed onto a single curve .
( 9 ) suggests that @xmath30 is a complicated function of @xmath0 , deep in the efficient phase .
in particular , as one increases the population size at fixed and small @xmath0 , one should approach the result given by eq.(9 ) assuming a uniform initial distribution of strategies to the agents .
it is , in fact , possible to include the effects of a finite population size @xmath4 into the strategy - ranking theory starting from eq .
( 8) by incorporating the so - called market impact effects @xcite .
systems in the real world are characterized by connected agents @xcite .
the connections are often used for collecting information from the neighbours . recently
, several interesting attempts @xcite have been made to incorporate information sharing mechanisms among the agents into mg and b - a - r models . as an illustration of the application of srt to networked mg , we focus on the model proposed by anghel _
_ @xcite . as in the mg , anghel _ et al .
_ s model @xcite features @xmath4 agents who repeatedly compete to be in a minority group .
communications between agents are introduced by assuming that the agents are connected by an undirected random network , i.e. , classical random graph , with a connectivity @xmath70 being the probability that a link between two randomly chosen agents exists .
the links are used as follows .
each agent compares the cumulated performance of his predictor , which is the suggested action from his own best - performing strategy at each time step , with that of his neighbours , and then follows the suggested action of the best performing predictor among his neighbours and himself .
the @xmath71 limit of the model reduces to the mg .
note that the identity of the best - performing strategy changes over time . for @xmath72 the predictor s
performance is generally _ different _ from the agent s performance .
it has been reported that the efficiency of the population as a whole , characterized by either @xmath18 @xcite or by the average winning probability per agent per turn @xcite , shows a _ non - monotonic _ dependence on the connectivity @xmath70 with the most efficient performance occurring at a small but finite value of @xmath70 .
in other words , a small fraction of links is beneficial but too many of them are bad .
we have explained the feature successfully within the framework of srt @xcite .
the most important point is that , from our understanding of the non - networked mg ( e.g. , see fig . 1 ) , the performance of an agent actually depends on how similar the @xmath24 strategies that he is holding , with the best performing ones holding two identical strategies .
the links then act in two ways depending on the connectivity . for low connectivity ,
the links bring the agents with two anti - correlated strategies to have the chance to use other strategies so that these agents will not always join the crowd at odd time steps and hence with their winning probability enhanced . for high connectivity , there are so many links that many agents are linked to the momentarily best - performing predictor or predictors .
as discussed in previous section , the higher ranking strategies have a smaller chance of predicting the correct minority outcome .
when the connectivity is high , there are many links so that agents have access to strategies that are more likely to lose .
this leads to a drop in the average winning probability of the agents @xcite .
figure 3 shows how the distribution @xmath17 changes with the connectivity @xmath70 at two small values of @xmath70 .
the range of small @xmath70 is particularly of interest since for a large population ( @xmath73 ) the non - monotonic feature occurs for @xmath74 .
the symbols ( open circles ) give the results from numerical simulations .
the peaks of the distribution @xmath17 shifts as @xmath70 is varied . applying srt and incorporating the effects of the presence of links
, we found that @xmath17 can again be represented by a weighted sum of distributions characterized by different kinds of time steps .
in particular , for @xmath75 , the parameters of the distributions in eq .
( 4 ) can be found @xcite to be @xmath76 , @xmath77 , and @xmath78 . the variances are given by eq .
( 5 ) as @xmath79 and @xmath80 .
similarly for @xmath81 , we have @xmath82 and @xmath83 , with the same variances .
the values of these parameters are obtained by considering the different winning probabilities of the strategies in different ranks and the change in the number of agents using a strategy of a certain rank due to the presence of the links .
the solid lines in fig .
3 show the distributions obtained by srt .
the theory captures the shifts in @xmath17 with the connectivity @xmath70 .
in the present work , we illustrated the basic ideas in constructing a strategy ranking theory for a class of multi - agent models incorporating the effects of tied strategies and strategy selections .
we showed how the theory can be applied to mg in the efficient phase to evaluate the distribution @xmath17 in the fraction of agents making a particular decision and the associated variance @xmath18 .
in particular , an analytic expression is given for @xmath18 in a non - networked population .
the theory is also applied to a version of networked mg in which there exists non - trivial dependence on the performance of the agents as a function of the connectivity . besides @xmath17 and @xmath18 , the theory can also be applied to evaluate other quantities such as the average winning probability of the agents . in closing , while srt is developed with models based on the mg in mind , the general approach , namely that of focusing on the ranking pattern of the strategies and how the pattern evolves in time , should be a key ingredient in the construction of theories for a large class of agent - based models . | the minority game ( mg ) is a basic multi - agent model representing a simplified and binary form of the bar attendance model of arthur .
the model has an informationally efficient phase in which the agents lack the capability of exploiting any information in the winning action time series .
we illustrate how a theory can be constructed based on the ranking patterns of the strategies and the number of agents using a particular rank of strategies as the game proceeds .
the theory is applied to calculate the distribution or probability density function in the number of agents making a particular decision . from the distribution ,
the standard deviation in the number of agents making a particular choice ( e.g. , the bar attendance ) can be calculated in the efficient phase as a function of the parameter @xmath0 specifying the agent s memory size . since situations with tied cumulative performance of the strategies often occur in the efficient phase and they are critical in the decision making dynamics , the theory is constructed to take into account the effects of tied strategies .
the analytic results are found to be in better agreement with numerical results , when compared with the simplest forms of the crowd - anticrowd theory in which cases of tied strategies are ignored .
the theory is also applied to a version of minority game with a networked population in which connected agents may share information .
* paper to be presented in the 10th annual workshop on economic heterogeneous interacting agents ( wehia 2005 ) , 13 - 15 june 2005 , university of essex , uk . * |
dendritic spines are small protrusions from many types of neurons located on the surface of a neuronal dendrite .
they receive most of the excitatory inputs and their physiological role is still unclear although most spines are thought to be key elements in neuronal information processing and plasticity @xcite .
spines are composed of a head ( @xmath0 @xmath1 m ) and a thin neck ( @xmath2 @xmath1 m ) attached to the surface of dendrite ( see fig .
1 ) . the heads of spines have an active membrane , and as a consequence , they can sustain the propagation of an action potential with a rate that depends on the spatial density of spines @xcite .
decreased spine density can result in cognitive disorders , such as autism , mental retardation and fragile x syndrome @xcite .
diffusion over branched smooth dendritic trees is basically determined by classical diffusion and the mean square displacement ( msd ) along the dendritic axis grows linearly with time .
however , inert particles diffusing along dendrites enter spines and remain there , trapped inside the spine head and then escape through a narrow neck to continue their diffusion along the dendritic axis .
recent experiments together with numerical simulations have shown that the transport of inert particles along spiny dendrites of purkinje and piramidal cells is anomalous with an anomalous exponent that depends on the density of spines @xcite .
based on these results , a fractional nernst - planck equation and fractional cable equation have been proposed for electrodiffusion of ions in spiny dendrites @xcite . whereas many studies have been focused to the coupling between spines and dendrites , they are either phenomenological cable theories @xcite or microscopic models for a single spine and parent dendrite @xcite . more recently a mesoscopic non - markovian model for spines - dendrite interaction and an extension including reactions in spines and variable residence time
have been developed @xcite .
these models predict anomalous diffusion along the dendrite in agreement with the experiments but are not able to relate how the anomalous exponent depends on the density of spines @xcite . since these experiments have been performed with inert particles ( i.e. , there are not reaction inside spines or dendrites ) we conclude that the observed anomalous diffusion is due exclusively to the geometric structure of the spiny dendrite .
recent studies on the transport of particles inside spiny dendrites indicate the strong relation between the geometrical structure and anomalous transport exponents @xcite
. therefore , elaboration such an analytical model that establishes this relation can be helpful for further understanding transport properties in spiny dendrites .
the real distribution of spines along the dendrite , their size and shapes are completely random @xcite , and inside spines the spine necks act as a transport barrier @xcite .
for these reasons we reasonably assume that the diffusion inside spine is anomalous .
so , we propose in this paper models based on a comb - like structure that mimic a spiny dendrite ; where the backbone is the dendrite and the fingers ( lateral branches ) are the spines .
the models predict anomalous transport inside spiny dendrites , in agreement with the experimental results of ref .
@xcite , and also explain the dependence between the mean square displacement and the density of spines observed in @xcite .
geometry of the comb structure consisting of a one - dimensional backbone and lateral branches ( fingers ) @xcite makes it possible to describe anomalous diffusion in spiny dendrites structure in the framework of the comb model . in this case
dynamics inside fingers corresponds to spines , while the backbone describes diffusion inside dendrites .
the comb model is an analogue of a 1d medium where fractional diffusion has been observed and explained in the framework of a so - called continuous time random walk @xcite .
usually , anomalous diffusion on the comb is described by the @xmath3 distribution function @xmath4 , and a special behavior is that the displacement in the @xmath5direction is possible only along the structure backbone ( @xmath5-axis at @xmath6 ) .
therefore , diffusion in the @xmath5-direction is highly inhomogeneous .
namely , the diffusion coefficient is @xmath7 , while the diffusion coefficient in the @xmath8direction ( along fingers ) is a constant @xmath9 . due to this geometrical construction ,
the flux of particles along the dendrite is @xmath10 and the flux along the finger describes the anomalous trapping process that occurs inside the spine @xmath11 where @xmath12 is the density of particles and @xmath13 is the riemann - liouville fractional derivative , where the fractional integration @xmath14 is defined by means of the laplace transform @xmath15= s^{-\gamma}\tilde{f}(s)\ , .\ ] ] so , inside the spine , the transport process is anomalous and @xmath16 , where @xmath17 . making use of the continuity equation for the total number of particles @xmath18 where @xmath19 one has the following evolution equation for transport along the spiny dendrite @xmath20 the riemann - liouville fractional derivative in eq .
( [ eb ] ) is not convenient for the laplace transform . to ensure feasibility of the laplace transform , which is a strong machinery for treating fractional equations , one reformulates the problem in a form suitable for the laplace transform application . to shed light on this situation ,
let us consider a comb in the @xmath21 @xcite .
this model is described by the distribution function @xmath22 with evolution equation given by the equation @xmath23it should be stressed that @xmath24 coordinate is a supplementary , virtue variable , introduced to described fractional motion in spines by means of the markovian process .
thus the true distribution is @xmath25 with corresponding evolution equation @xmath26 a relation between @xmath12 and @xmath27 can be expressed through their laplace transforms ( see derivation in the appendix ) @xmath28 where @xmath29 $ ] and @xmath30 $ ] .
therefore , performing the laplace transform of eq .
( [ 3d_comb2 ] ) yields @xmath31 and substituting relation ( [ 3d_comb3 ] ) , dividing by @xmath32 and then performing the laplace inversion , one obtains the comb model with the fractional time comb model @xmath33 where @xmath34 and the caputo derivative , while the latter fractional derivative is not indexed @xmath35 .
note , that it is also convenient to use eq .
( [ lcaputo ] ) as a definition of the caputo fractional derivative . ]
@xmath36 can be defined by the laplace transform for @xmath17 @xcite @xmath37=s^{\gamma}\tilde{f}(s)-s^{\gamma-1}f(t=0)\ , .\ ] ] the fractional transport takes place in both the dendrite @xmath5 direction and the spines @xmath8 coordinate . to make fractional diffusion in dendrite normal
, we add the fractional integration @xmath38 by means of the laplace transform ( [ res_1c ] ) , as well @xmath39= s^{\gamma-1}\tilde{f}(s)$ ] .
this yields eq .
( [ 3d_comb4 ] ) , after generalization @xmath40 , @xmath41 performing the fourier - laplace transform in ( [ 3d_comb5 ] ) we get @xmath42where the fourier - laplace image of the distribution function is defined by its arguments @xmath43=p(k_x , k_y , s)$ ] . if @xmath44 , inversion by fourier over @xmath8 gives @xmath45 taking @xmath6 the above equation provides@xmath46 which inserted into ( [ tfl ] ) yields @xmath47we can calculate the density of particles at a given point @xmath5 of the dendrite at time @xmath48 namely @xmath49 by integrating over @xmath8@xmath50then @xmath51 so that @xmath52 eq .
( [ moment2 ] ) predicts subdiffusion along the spiny dendrite which is in agreement with the experimental results reported in @xcite . it should be noted that this result is counterintuitive .
indeed , subdiffusion in spines , or fingers should lead to the slower subdiffusion in dendrites , or backbone with the transport exponent less than in usual comb , since these two processes are strongly correlated .
but this correlation is broken due to the fractional integration @xmath38 in eq .
( [ 3d_comb5 ] ) . on the other hand ,
if we invert ( [ pks ] ) by fourier - laplace we obtain the fractional diffusion equation for @xmath53 @xmath54which is equivalent to the generalized master equation @xmath55if the laplace transform of the memory kernel is given by @xmath56 , which corresponds to the waiting time pdf in the laplace space given by @xmath57 that is @xmath58 as @xmath59 .
the above waiting time pdf is the effective pdf corresponding to the whole comb and takes into account the particle trapping inside spines .
let us employ the notation for a dynamical exponent @xmath60 used in @xcite .
if @xmath61 then the msd grows as @xmath62 . on the other hand , it has been found in experiments that @xmath60 increases with the density of spines @xmath63 and the simulations prove that @xmath60 grows linearly with @xmath63 .
indeed , the experimental data admits almost any growing dependence of @xmath60 with @xmath63 due to the high variance of the data ( see fig 5.d in @xcite ) .
equation ( [ moment2 ] ) also establishes a phenomenological relation between the second moment and @xmath63 .
when the density spines is zero then @xmath64 , @xmath65 and normal diffusion takes place .
if the spine density @xmath63 increases , the anomalous exponent of the pdf ( [ pdf ] ) @xmath66 must decrease ( i.e. , the transport is more subdiffusive due to the increase of @xmath63 ) so that @xmath60 has to increase as well .
so , our model predicts qualitatively that @xmath60 increases with @xmath63 , in agreement with the experimental results in @xcite .
in this section we consider a fractal comb model @xcite to take into account the inhomogeneity of the spines distribution .
the comb model is a phenomenological explanation of an experimental situation , where we introduce a control parameter that establishes a relation between diffusion along dendrites and the density of spines . suggesting more sophisticated relation between the dynamical exponent and the spine density
, we can reasonably suppose that the fractal dimension , due to the box counting of the spine necks , is not integer : it is embedded in the @xmath67 space , thus the spine fractal dimension is @xmath68 . according the fractal geometry ( roughly speaking ) , the most convenient parameter is the fractal dimension of the spine volume ( mass ) @xmath69 .
therefore , following nigmatulin s idea on a construction of a `` memory kernel '' on a cantor set in the fourier space @xmath70 @xcite ( and further developing in refs .
@xcite ) , this leads to a convolution integral between the non - local density of spines and the probability distribution function @xmath12 that can be expressed by means of the inverse fourier transform @xcite @xmath71 $ ] .
therefore , the starting mathematical point of the phenomenological consideration is the fractal comb model @xmath72=0\ , .\ ] ] performing the same analysis in the fourier - laplace space , presented in previous section , then eq . ( [ pks ] ) reads @xmath73 where @xmath74 .
contrary to the previous analysis expression ( [ msd_a ] ) does not work any more , since superlinear motion is involved in the fractional kinetics .
this leads to divergence of the second moment due to the lvy flights .
the latter are described by the distribution @xmath75 , which is separated from the waiting time probability distribution @xmath76 . to overcome this deficiency
, we follow the analysis of the lvy walks suggested in @xcite .
therefore , we consider our exact result in eq .
( [ nupks ] ) as an approximation obtained from the joint distribution of the waiting times and the lvy walks .
therefore , a cutoff of the lvy flights is expected at @xmath77 .
this means that a particle moves at a constant velocity inside dendrites not all times , and this laminar motion is interrupted by localization inside spines distributed in space by the power law .
performing the inverse laplace transform , we obtain solution in the form of the mittag - leffler function @xcite @xmath78 where @xmath79 . for the asymptotic behavior
@xmath80 the argument of the mittag - leffler function can be small .
note that in the vicinity of the cutoff @xmath77 this corresponds to the large @xmath81 ( @xmath82 ) , thus we have @xcite @xmath83 therefore , the inverse fourier transform yields @xmath84 where @xmath85 is determined from the normalization condition ) and ( [ solpxt ] ) also follows from the plausible finite result of eq .
( [ solpxt ] ) , which is the normalized distribution @xmath86 , where @xmath77 . ] .
now the second moment corresponds to integration with the cutoff at @xmath87 that yields @xmath88 where @xmath89 is a generalized diffusion coefficient .
transition to absence of spines means first transition to normal diffusion in fingers with @xmath90 and then @xmath91 that yields @xmath92
the present analysis establishes that the fractional dynamics in spiny dendrites can be described by two parameters , related to the fractal geometry of spines @xmath93 and fractional kinetics inside the spines @xmath94 . summarizing
, the most general phenomenological description can be performed in the framework of the fractional fokker - planck equation ( ffpe ) @xmath95 where , for the present analysis @xmath96 and @xmath97 ; in general case @xmath98 and @xmath99 are arbitrary . for @xmath100
, we arrive at the first model , presented in sec .
ii , where we deal with a one temporal control parameter @xmath94 only . in this case ,
anomalous transport in dendrites , described by the dynamical exponent @xmath60 , is characterized by anomalous transport inside spines , described by the transport exponent @xmath94 .
the obtained relation @xmath101 also establishes a relation between the dynamical exponent and the density of spines and is in agreement with recent experiments @xcite . in the second model we suggested a more sophisticated relation between the dynamical exponent and the spine density . in this case @xmath102
depends on fractal dimension of spines , and this leads to an essential restriction for eq .
( [ ffpe ] ) .
the first one is a cutoff of the lvy flights at @xmath77 that leads to a consequence of laminar and localized motions @xcite and yields a finite second moment @xmath103 .
when @xmath104 and @xmath100 the ffpe ( [ ffpe ] ) corresponds to the continuous comb model , namely spine dendrites with the maximal density of spines . for @xmath104 and @xmath105 this model corresponds to smooth dendrites .
apparently , another physically sound transition to limiting case is possible for @xmath106 and @xmath64 that corresponds first to the transition to the continuous model , and then the transition to @xmath64 .
this physical control of the parameters ensures an absence of superdiffusion in eq .
( [ ffpe ] ) .
another important question is what happens in intermediate cases .
a challenging question here is what is the fractal dimension of the spine volume .
we conclude our consideration by presenting the physical reason of the possible power law distribution of the waiting time pdf @xmath76 in eq .
( [ pdf ] ) . at this point
we paraphrase some arguments from ref .
@xcite with the corresponding adaptation to the present analysis .
let us consider the escape from a spine cavity from a potential point of view , where geometrical parameters of the cavity can be related to a potential @xmath107 .
for example , let us consider spines with a head of volume @xmath108 and the cylindrical spine neck of the length @xmath109 and radius @xmath110 , and the diffusivity @xmath111 @xcite . in this case , the potential is @xmath112 , which `` keeps '' a particle inside the cavity , while @xmath113 plays a role of the kinetic energy , or the `` boltzmann temperature '' , where @xmath114 is a mean survival time a particle inside the spine .
therefore , escape probability from the spine cavity / well is described by the boltzmann distribution @xmath115 .
this value is proportional to the inverse waiting , or survival time @xmath116 as admitted above , potential @xmath107 is random and distributed by the exponential law @xmath117 , where @xmath118 is an averaged geometrical spine characteristic .
the probability to find the waiting time in the interval @xmath119 is equal to the probability to find the trapping potential in the interval @xmath120 , namely @xmath121 .
therefore , from eq .
( [ gimp2 ] ) one obtains @xmath122 here @xmath123 establishes a relation between geometry of the dendrite spines and subdiffusion observed in @xcite and support application of our comb model with anomalous diffusion inside spines , which is a convenient implement for analytical exploration of anomalous transport in spiny dendrites in the framework of the continuous - time - random - walk framework .
this research has been partially supported by the generalitat de catalunya with the grant sgr 2009 - 00164 ( vm ) , by ministerio de economa y competitividad with the grant fis2012 - 32334 ( vm ) , by the israel science foundation isf ( ai ) , and by the us - israel binational science foundation bsf ( ai ) .
( [ 3d_comb3 ] ) is a relationship between the distributions @xmath27 and @xmath12 in the laplace space .
both distributions are related through the expression @xmath124 if we transform the above equation by fourier - laplace we get @xmath125 then , eq . ( [ 3d_comb3 ] ) is nothing but a relation between @xmath126 and @xmath127 . to find @xmath128 we transform eq .
( [ 3d_comb1 ] ) by fourier - laplace and after collecting terms we find @xmath129 where the initial condition has been assumed @xmath130 for simplicity . setting @xmath131
one gets @xmath132 inverting eq .
( [ a2 ] ) by fourier over @xmath133 we obtain @xmath134 and setting @xmath135 @xmath136 combining ( [ a3 ] ) and ( [ a4 ] ) one has @xmath137 and inverting fourier over @xmath138 and @xmath139 one finally recovers eq .
( [ 3d_comb3 ] ) .
henry , t.a.m .
langlands , and s.l .
wearne , fractional cable models for spiny neuronal dendrites phys .
100 : 128103 , ( 2008 ) ; t.a.m .
langlands , b.i .
henry , and s.l .
wearne , fractional cable equation models for anomalous electrodiffusion in nerve cells : infinite domain solutions .
59 : 761 , ( 2009 ) ; t.a.m .
langlands , b.i .
henry , and s.l .
wearne , fractional cable equation models for anomalous electrodiffusion in nerve cells : finite domain solutions .
siam j. appl .
71 : 1168 , ( 2011 ) .
baer and j. rinzel , propagation of dendritic spikes mediated by excitable spines : a continuum theory .
j. neurophysiol .
65 : 874 , ( 1991 ) ; h .- y . wu and s.m .
baer , analysis of an excitable dendritic spine with an activity - dependent stem conductance . j. math .
36 : 569 , ( 1998 ) ; p.c .
bressloff and s. coombes , solitary waves in a model of dendritic cable with active spines .
siam j. appl .
61 : 432 , ( 2000 ) ; saltatory waves in the spike - diffuse - spike model of active dendritic spines .
91 : 028102 , ( 2003 ) .
sabatini , m. maravall , and k. svoboda , ca(2 + ) signaling in dendritic spines .
11 : 349 , ( 2001 ) ; b. l. bloodgood and b. l. sabatini , neuronal activity regulates diffusion across the neck of dendritic spines .
science 310 : 866 , ( 2005 ) . z. schuss , a. singer , and d. holcman , the narrow escape problem for diffusion in cellular microdomains .
104 : 16098 , ( 2007 ) ; l. dagdug , a.m. berezhkovskii , y.a .
makhnovskii , v.yu .
zitserman , transient diffusion in a tube with dead ends .
phys . 127 : 224712 , ( 2007 ) .
s. fedotov and v. mndez , non - markovian model for transport and reactions of particles in spiny dendrites .
lett . 101 : 218102 , ( 2008 ) ; s. fedotov , h. al - shamsi , a. ivanov and a. zubarev , anomalous transport and nonlinear reactions in spiny dendrites .
e 82 : 041103 , ( 2010 ) . | we suggest a modification of a comb model to describe anomalous transport in spiny dendrites .
geometry of the comb structure consisting of a one - dimensional backbone and lateral branches makes it possible to describe anomalous diffusion , where dynamics inside fingers corresponds to spines , while the backbone describes diffusion along dendrites .
the presented analysis establishes that the fractional dynamics in spiny dendrites is controlled by fractal geometry of the comb structure and fractional kinetics inside the spines .
our results show that the transport along spiny dendrites is subdiffusive and depends on the density of spines in agreement with recent experiments . |
compact remnants of massive stars in star clusters , which are neutron stars ( ns ) and black holes ( bh ) , form a dynamically interesting sub - population due to their tendency of segregating towards the cluster s center and augmenting their population density therein . in this respect ,
the bhs are special in that they undergo a `` runaway '' mass segregation .
these remnant bhs are typically several 10s of @xmath0 heavy , enough to form a spitzer - unstable sub - system , provided a significant number of them are retained in their parent cluster . due to this instability (
also called the mass - stratification instability , @xcite ) , the continually sinking bhs can not come to an energy equipartition with the local surrounding stars and finally end - up in a central , highly concentrated sub - cluster made purely of bhs , which is self - gravitating and dynamically nearly isolated from the rest of the stellar cluster @xcite .
such a dense environment of bhs is dynamically very active due to the formation of bh - bh binaries via 3-body encounters @xcite and their hardening by super - elastic encounters @xcite with their surrounding bhs .
studies of the dynamics of pure bh sub - clusters using monte - carlo and direct n - body integration methods indicate that the dynamical bh - bh merger events they generate are likely to contribute a significant gravitational wave ( gw ) detection rate to the future `` advanced ligo '' ( adligo ) and `` lisa '' gw observatories @xcite .
such studies show that a bh sub - cluster is typically self - depleted in a few gyr due to the super - elastic dynamical encounters and the resulting escape of the bhs @xcite .
the energy extracted from the tight bh - bh binaries heats - up and expands the cluster s core @xcite , which can be detectable by the future optical missions such as the `` gaia '' mission .
furthermore , the bhs can be important for dynamically formed bh x - ray sources due to their encounters with the surrounding stars @xcite .
x - ray observations have indicated the presence of bh x - ray binary candidates in gcs @xcite .
the presence of a bh sub - cluster within a star cluster , therefore , has the potential to give rise to a plethora of physical phenomena , all of which have significance to upcoming prime missions such as the gaia , adligo and the present and future x - ray missions .
is it possible to obtain any direct observational signature of the presence of a bh sub - cluster within a star cluster ?
we predict here , for the first time , that within a few kpc from the galactic center , rapid tidal stripping of star clusters by the strong tidal field can expose its bh sub - cluster .
this would happen when the timescale of the preferential removal of stars from the outer regions of the cluster is shorter or comparable to the encounter - driven self - depletion timescale of its central bh sub - cluster ( see above ) .
such a dissolved phase of the cluster would consist of a few stars orbiting around a cluster of bhs and would observationally appear as a highly super - virial star cluster with a large mass - to - light - ratio . as we discuss here
, a number of direct n - body computations of model star clusters indeed support the formation of such systems .
these objects comprise a predicted , new class of compact stellar populations which we name `` dark star clusters '' .
the importance of dark star clusters ( hereafter dsc ) is twofold : on one hand , if they are found to exist , then they guarantee that star clusters are potential sites for gw emission and formation of bh x - ray binaries and on the other hand , they naturally constrain the uncertain natal kicks of bhs @xcite , as dscs can form only if a significant number of bhs retain in the cluster following their progenitor supernovae .
this , in turn , restricts the theoretical models of core collapse supernovae @xcite .
we compute the evolution of model star clusters subjected to the galactic tidal field using the direct n - body integration method .
for our purposes , we use the state - of - the - art n - body integration code `` nbody6 '' @xcite , which , apart from utilizing a highly sophisticated numerical integration scheme @xcite , also follows the evolution of the individual stars until their remnant phases , using an analytic but well - tested stellar evolution recipe @xcite .
a unique feature of nbody6 is its use of highly accurate regularization methods in resolving close encounters @xcite .
furthermore , the code exploits the remarkable hardware - accelerated computing capacity of graphical processing units ( gpus ) in integrating the centers of masses .
nbody6 currently incorporates general - relativistic effects only through analytic prescriptions of gw energy - loss .
we follow the evolution of initial plummer clusters @xcite of single stars , having masses between @xmath1 and half - mass radii between @xmath2 .
all the clusters initially consist of zero - age - main - sequence ( zams ) stars with their masses @xmath3 chosen from the canonical initial mass function ( imf ; @xcite ) @xmath4 , where @xmath5 for @xmath6 and @xmath7 ( salpeter index ) for @xmath8 .
their metallicities are chosen to be solar , as suitable for our galaxy s disk .
we assume for now that all the supernova remnants ( _ i.e. _ , bhs and nss ) receive low natal kicks in general such that they remain bound to their parent clusters at their formation .
such models follow circular orbit around a point mass of @xmath9 representing the milky way bulge .
[ fig : qevol ] ( top panel ) shows an example of the evolution of the virial coefficient for one of our computed model clusters with initially @xmath10 stars and @xmath11 pc , located at @xmath12 kpc galactocentric distance .
the orange curve shows the time - evolution of the virial coefficient , @xmath13 , for all the members bound to the cluster including the bhs and the nss , which mostly remains constant at @xmath14 , as it should be for the quasi - static relaxation of a self - gravitating system through two - body encounters @xcite . the green curve in fig . [ fig : qevol ] ( top ) shows the time - evolution of the virial coefficient taking into account only the luminous objects , _
i.e. _ , the nuclear burning stars and the white dwarfs ( hereafter wd ) , which are those an observer sees .
the bhs formed have masses @xmath15 and the nss are typically of @xmath16 , which are significantly more massive than the majority of the remaining luminous members and hence are segregated to the cluster s center . as the lower mass luminous stars in the outer regions
are stripped by the external field , the gravitational potential of these central invisible remnants becomes increasingly important .
therefore , the kinetic energy of the luminous sub - system increasingly exceeds the corresponding self - equilibrium ( or quasi - static ) value as the constituents perceive a potential that becomes increasingly deeper than their self - potential .
this correspondingly raises their exclusive virial coefficient , @xmath17 , above 0.5 as in fig .
[ fig : qevol ] ( top panel , green curve ) .
the cluster thus evolves to a dsc state ( see sec .
[ dcpop ] ) .
note that while @xmath17 reaches a very high value , the system as a whole remains bound since @xmath18 throughout ( except at the final dissolved state , not shown in the figure ) .
notably , nss form a few factors more often than the bhs and contribute significantly in elevating @xmath17 .
however , the nss being lighter than the bhs , their sub - population occupies a more extended zone in the cluster s center .
this causes them to get stripped earlier than the bhs ( _ c.f . _
[ fig : qevol ] , bottom panel ) so that in the late evolutionary phase , typically when @xmath19 , it is mostly the bh population that contributes to the augmented @xmath17 , thereby determining the lifetime of the dsc phase ( see sec .
[ dcpop ] ) . nevertheless , at large enough distances from the galactic center , where slower tidal stripping causes the dsc phase to appear later than the self - depletion of the bh sub - cluster ( see sec . [ intro ] ; @xcite ) , the nss constitute the primary dark component of the corresponding dsc state ( see sec . [ dcpop ] ) .
this self - depletion process , although operative for both the ns and the bh sub - populations , is more efficient for the latter as it is significantly more concentrated .
[ fig : tfram ] ( top ) shows the expected increasing trend in the lifetime of the dscs with initial cluster mass @xmath20 ( @xmath21 , @xmath22 ) .
the dsc phase can be defined when the cluster appears unbound , _
i.e. _ , @xmath23 or when it appears significantly super - virial , which we take when @xmath24 , and denote the corresponding lifetimes by @xmath25 with @xmath26 and 0.75 respectively . fig .
[ fig : tfram ] ( bottom ) shows @xmath27 against @xmath28 ( @xmath29 , @xmath22 ) .
@xmath25 also increases with increasing @xmath28 since the dsc takes a longer time to get depleted in a weaker external field .
notably , for @xmath30 kpc , the dsc state becomes ns - dominated . beyond @xmath31 kpc , it takes more than the age of our galaxy ( @xmath32 gyr ) for the representative cluster to evolve to its dsc phase ( _ i.e. _ , @xmath24 ) .
it then follows that for the present - day galaxy the dscs are formed from clusters with @xmath33 within @xmath34 kpc , to be taken as representative numbers , and have lifetimes @xmath35 myr and @xmath36 myr as conservative estimates .
although these estimates are based on a point - mass tidal field , note that the dscs progenitor clusters form in the galactic disk and orbit on nearly circular paths ( in the equatorial plane ) and hence would experience the same external field with an axisymmetric disk - like distribution of the same mass .
the @xmath37 dependence can however be moderately modified as the clusters see a mass increasing moderately with @xmath28 . to estimate the galactic population of dscs
, we take the average star cluster formation rate ( hereafter cfr ) over the entire galactic disk ( @xmath38 kpc ) to be @xmath39 @xcite following a schechter initial cluster mass function @xcite over the mass range @xmath40 @xcite which is assumed to remain invariant over the last few hundred myr .
this implies @xmath41 clusters form per myr which have properties ( @xmath42 and @xmath43 kpc ) that must have them evolve to dscs if a sufficiently high fraction of bhs are retained . assuming a steady state conversion to the dsc phase with lifetime @xmath35 myr ( see above ) , the expected number of @xmath19 dscs in the galaxy within @xmath43 kpc is @xmath44 .
for dscs with @xmath45 , the corresponding number is @xmath46 .
hence , a significant number of dscs can be expected in the inner galactic zone
. there has been recent concern with the widely debated irs 13e ( hereafter irs13e ) ; an extremely compact stellar association of a few young , massive stars at a close projection to the galactic center that apparently survives the extreme tidal field by being bound by an invisible mass @xcite . while a @xmath47 intermediate mass black hole ( imbh ) was widely believed to be this invisible mass @xcite , this possibility has recently been ruled out with a significant confidence by @xcite through their newer proper motion measurements , leaving the nature of the invisible component of irs13e currently ambiguous .
this status of irs13e prompts us to consider whether its dark component can be an ensemble of stellar - mass bhs instead . to that end
, we perform preliminary n - body calculations to determine whether a dsc configuration resembling irs13e can be a possible fate of a star cluster very close to the galactic center .
these computed clusters follow circular orbits around the galactic smbh ( a central mass of @xmath48 which is a combination of the mass of the smbh and that of the nuclear star cluster within @xmath49 pc ) within @xmath50 few pc , where star formation has been shown to possibly lead to a top - heavy stellar imf @xcite .
since we are primarily interested in the final state , we begin the n - body calculations from an evolved phase of the cluster , for computational ease .
therefore , we initiate the computations with plummer clusters made of stars which are pre - evolved until @xmath51 myr age . this age is slightly earlier than when the most massive star ( of @xmath52 ) evolves to a bh .
all the stars are thus still on their main sequence .
observations indicate a very flat imf ( @xmath53 ; see @xcite ) for stars close to the galactic center and also a dearth of low - mass stars .
the latter is evident from a declining density of b - stars away from the central smbh @xcite and a significant lack of coronal x - ray emission @xcite from the sgra@xmath54 field .
it is currently unclear from the literature what would be the lower - mass limit of such an imf , which , in turn , determines the @xmath55 of the mass function ( mf ) and its lower limit @xmath56 at cluster - age @xmath57 . in this preliminary study
, we simply take the total number of stars in the cluster @xmath58 , @xmath55 and @xmath56 at age @xmath57 as free parameters .
the upper mf limit is chosen to be the canonical @xmath59 and the stellar metallicity is solar .
these limits are the zams values and they are appropriately reduced during the @xmath51 myr pre - evolution by the nbody6 s built - in stellar evolution prescription ( see above ) . like the computations in sec .
[ darkcl ] , we retain all the bhs ( of @xmath60 as obtained from within nbody6 ) in the cluster . at the beginning of the computations ,
the models are taken compact to a similar extent as irs13e @xcite , with half - mass radii between @xmath61 and they orbit within @xmath62 .
we find that to reach a irs13e - like state , _
i.e. _ , a state where typically @xmath32 luminous , young stars are very tightly bound to a cluster of @xmath63 bhs , it takes @xmath64 and a rather extreme lower cutoff of @xmath65 for mf index @xmath66 ( at age @xmath57 ) , this @xmath55 being close to the upper limit of the @xcite index . a steeper mf requires even higher @xmath56 . as a demonstration , it can be seen in fig .
[ fig : irs13 ] ( top panel ) that this cluster eventually evolves to a configuration consisting of @xmath67 bhs , comprising a @xmath47 dark component , orbited by @xmath68 young stars .
although the half - mass radius of the system expands at the beginning of the evolution , primarily driven by the mass loss through the massive stars winds and their supernovae , it finally collapses to @xmath69 pc , as shown in fig .
[ fig : irs13 ] ( middle panel ) , by the time the system arrives at the above configuration .
the luminous members in this state include 1 - 2 o - stars ( those still on the main sequence ) and a few helium and wolf - rayet stars , thereby being of similar variety as observed in irs13e .
the final state of the system , therefore , resembles irs13e in terms of compactness and stellar content @xcite .
[ fig : irs13 ] ( lower panel ) shows that the luminous sub - system becomes highly super - virial , _
i.e. _ , a dsc , as the irs13e - like state occurs .
the calculations presented in this section are preliminary which suggest an intermediate state of a cluster , close to the galactic center , that would evolve to an irs13e - like configuration .
it remains an open question whether any reasonable initial cluster evolves to such an intermediate phase , which depends on the low - mass limit of the galactic - central imf , its index and the initial mass and compactness of the cluster , at a given @xmath28 .
a scan over these parameters , beginning with much larger and more compact clusters , is necessary to determine such possible initial configuration(s ) , which is much more compute - intensive .
although the above @xmath56 appears too high for a lower cut - off , its progenitor cluster would have a smaller limit taking into account the rapid tidal stripping over the 3.5 myr pre - evolution .
tidal stripping also implies a steeper initial index , in better agreement with the observed one .
although we over - simplify by effectively ignoring the dynamical evolution of the cluster during the stellar pre - evolution , the dynamical history is not instrumental in determining the occurrence of a irs13e - type state which happens merely due to the competition between stellar evolution and tidal dissolution of the cluster .
it is enough that the cluster remains bound by the time most of the stars become bhs quenching the wind mass loss ; the system would then core - collapse to a compact configuration irrespective of its history .
this is in contrast with the `` classical '' dscs discussed in sec .
[ darkcl ] whose formation depends crucially on mass segregation . a potential drawback of this `` in - situ '' model is that it is likely to significantly over - produce the number of massive stars seen in the galactic center . given these drawbacks , the calculations in this section are only to suggest that irs13e might be a dsc but are by no means conclusive .
our calculations ( sec . [ darkcl ] ) signify that a gravitationally bound star cluster naturally evolves to an apparent super - virial state , while remaining bound as a whole , as a consequence of the interplay between the dynamics and the evolution of stars , provided a significant number of stellar remnants survive in the bound system after their formation via supernovae .
the existence of such dark star clusters is a first - time prediction and serves as an excellent cross - check of the retention of supernova remnants in star clusters , the effects of which is widely explored .
the very presence of dscs would kill two birds with one stone by having consequences on the widely debated theoretical stellar collapse models @xcite due to the implied direct constraints on natal kicks , and by securing star clusters as potential sources for the forthcoming `` advanced ligo '' @xcite gw detector @xcite .
given that the dscs must have a significant population in the inner region of our galaxy , it is tempting to conduct a survey of intermediate - aged ( see below ) stellar assemblies to identify them .
dark star clusters can be observationally distinguished from actually dissolving clusters ( which are also super virial and hence expanding and are generally young ) through their compact sizes in spite of their intermediate ages , as shown in fig .
[ fig : lagrev ] . that the apparent super - virial state of a dsc is not due to an imbh , would be indicated by the absence of a central cusp in its velocity - dispersion profile , which would be there otherwise @xcite
the present studies justify dscs as predicted , new type of compact stellar populations .
an immediate improvement over this study is to explore any effects of varying orbital eccentricity and different initial profile types ( _ e.g. _ , using king instead of plummer profiles ) .
also , the effect of a varying star ( and hence cluster ) formation rate over the galactic disc needs to be incorporated to determine the predicted galactic dsc population , as an improvement over our assumption of a uniform average cluster formation .
another development would be to consider a disk - like mass distribution instead of a central point - mass ( but see sec .
[ dcpop ] ) .
the preliminary computations in sec . [ irs13e ] suggest that the dark component of irs13e can be an ensemble of stellar mass bhs and irs13e may therefore perhaps be a dsc .
however , this conclusion should , for now , be taken as being suggestive rather than conclusive due to the drawbacks discussed ( sec .
[ irs13e ] ) .
we postpone a more detailed and self - consistent study on this issue in a future paper . ,
@xmath11 pc and @xmath12 kpc .
the lower ( orange ) curve represents the virial coefficient , @xmath13 , for the whole cluster , _
i.e. _ , including the luminous stars , nss and bhs bound to the cluster .
the upper ( green ) curve represents the virial coefficient , @xmath17 , of the luminous objects exclusively .
the monotonic growth of @xmath17 implies the approach towards an apparent highly super - virial state the `` dark star cluster '' phase . *
bottom : * the corresponding evolution of the numbers of the luminous stars , @xmath70 , the nss , @xmath71 and of the bhs , @xmath72.,title="fig:",width=434,height=264 ] , @xmath11 pc and @xmath12 kpc . the lower ( orange ) curve represents the virial coefficient , @xmath13 , for the whole cluster , _
i.e. _ , including the luminous stars , nss and bhs bound to the cluster . the upper ( green ) curve represents the virial coefficient , @xmath17 , of the luminous objects exclusively .
the monotonic growth of @xmath17 implies the approach towards an apparent highly super - virial state the `` dark star cluster '' phase .
* bottom : * the corresponding evolution of the numbers of the luminous stars , @xmath70 , the nss , @xmath71 and of the bhs , @xmath72.,title="fig:",width=434,height=264 ] , of the dscs with increasing initial plummer cluster mass @xmath20 ( @xmath73 kpc , @xmath11 pc ) . * bottom : * @xmath25 ( @xmath74 ) as a function of @xmath28 for a plummer cluster with @xmath29 and @xmath22.,title="fig:",width=434,height=302 ] , of the dscs with increasing initial plummer cluster mass @xmath20 ( @xmath73 kpc , @xmath11 pc ) . *
bottom : * @xmath25 ( @xmath74 ) as a function of @xmath28 for a plummer cluster with @xmath29 and @xmath22.,title="fig:",width=434,height=302 ] , and of the bhs , @xmath72 , for a model star cluster computation , leading to an irs13e solution . the direct n - body computation starts with a plummer cluster of pre - evolved stars having @xmath75 , @xmath76 pc and @xmath77 pc , where the stars are taken from a @xmath78 mf within @xmath79 zams mass ( see sec . [ irs13e ] ) .
the evolution of the lagrangian radii of the system is shown in the middle panel .
the final state of the cluster resembles that of irs13e .
the evolution of the viral coefficient @xmath13 , for the whole system , and @xmath17 , for the luminous sub - system is shown in the bottom panel.,title="fig:",width=415,height=245 ] , and of the bhs , @xmath72 , for a model star cluster computation , leading to an irs13e solution .
the direct n - body computation starts with a plummer cluster of pre - evolved stars having @xmath75 , @xmath76 pc and @xmath77 pc , where the stars are taken from a @xmath78 mf within @xmath79 zams mass ( see sec .
[ irs13e ] ) .
the evolution of the lagrangian radii of the system is shown in the middle panel .
the final state of the cluster resembles that of irs13e .
the evolution of the viral coefficient @xmath13 , for the whole system , and @xmath17 , for the luminous sub - system is shown in the bottom panel.,title="fig:",width=423,height=245 ] , and of the bhs , @xmath72 , for a model star cluster computation , leading to an irs13e solution .
the direct n - body computation starts with a plummer cluster of pre - evolved stars having @xmath75 , @xmath76 pc and @xmath77 pc , where the stars are taken from a @xmath78 mf within @xmath79 zams mass ( see sec .
[ irs13e ] ) .
the evolution of the lagrangian radii of the system is shown in the middle panel .
the final state of the cluster resembles that of irs13e .
the evolution of the viral coefficient @xmath13 , for the whole system , and @xmath17 , for the luminous sub - system is shown in the bottom panel.,title="fig:",width=415,height=245 ] . the half - mass radius ( black curve , top panel ) remains nearly constant throughout and the core also remains compact following an initial brief contraction phase.,title="fig:",width=434,height=283 ] .
the half - mass radius ( black curve , top panel ) remains nearly constant throughout and the core also remains compact following an initial brief contraction phase.,title="fig:",width=434,height=283 ] | among the most explored directions in the study of dense stellar systems is the investigation of the effects of the retention of supernova remnants , especially that of the massive stellar remnant black holes ( bh ) , in star clusters . by virtue of their eventual high central concentration , these stellar mass bhs potentially invoke a wide variety of physical phenomena , the most important ones being emission of gravitational waves ( gw ) , formation of x - ray binaries and modification of the dynamical evolution of the cluster .
here we propose , for the first time , that rapid removal of stars from the outer parts of a cluster by the strong tidal field in the inner region of our galaxy can unveil its bh sub - cluster , which appears like a star cluster that is gravitationally bound by an invisible mass .
we study the formation and properties of such systems through direct n - body computations and estimate that they can be present in significant numbers in the inner region of the milky way .
we call such objects `` dark star clusters '' ( dscs ) as they appear dimmer than normal star clusters of similar mass and they comprise a predicted , new class of entities
. the finding of dscs will robustly cross - check bh - retention ; they will not only constrain the uncertain natal kicks of bhs , thereby the widely - debated theoretical models of bh - formation , but will also pin - point star clusters as potential sites for gw emission for forthcoming ground - based detectors such as the `` advanced ligo '' .
finally , we also discuss the relevance of dscs for the nature of irs 13e . |
charged particle multiplicities measured in high - multiplicity ( hm ) @xmath0 collisions at cern large hadron collider ( lhc ) energies reach values that are of the same order as those measured in heavy - ion collisions at lower energies ( _ e.g. _ , well above those observed at rhic for cu + cu collisions at @xmath9 = 200 gev @xcite ) .
the bjorken energy density relation @xcite connects high multiplicity events with high energy density . within that approach at the lhc
, @xmath0 collisions could reach an energy density of 5 - 10 gev/@xmath10 , comparable to those in @xmath11 collisions at rhic @xcite .
it is , therefore , a valid question whether @xmath0 collisions also exhibit any behavior of the kind observed in heavy - ion collisions @xcite .
bjorken first suggested the idea of possible deconfinement in @xmath0 collisions @xcite .
it has also been suggested by van hove @xcite and recently by campanini @xcite that an anomalous behavior of average transverse momentum ( @xmath12 ) as a function of multiplicity could be a signal for the occurrence of a phase transition in hadronic matter , _
i.e. _ , formation of a _ mini quark - gluon plasma _ ( mqgp ) .
the hadronic interaction model epos ( partons off - shell remnants and splitting of parton ladders ) , has also been used to describe the production of mqgp features in high energy density @xmath0 collisions @xcite .
another indication of collective phenomena might be the observed long - range , near - side angular correlation ( ridge ) in hm @xmath0 collisions at center - of - mass energy @xmath2 = 7 tev for charged particle multiplicities well above the mean multiplicity .
cms @xcite and atlas @xcite constructed a two - particle correlation function and measured its value for different @xmath13 and @xmath14 angular separations .
when looking at particles in a specific range of @xmath15 and high multiplicity ( @xmath16 gev/_c _ and @xmath17 ) , a clear ridge - like structure emerges at @xmath14 @xmath18 0 and @xmath19 , that is not reproduced by existing monte carlo ( mc ) event generators @xcite .
the origin of this unexpected _ ridge - like _ structure found in the two - particle correlation analysis , albeit attracting much theoretical attention , is still under debate @xcite .
identified particle production has been studied in detail by the atlas @xcite , alice @xcite and cms @xcite collaborations in @xmath0 collisions at the lhc .
meson ( @xmath20 , @xmath21 , @xmath22 ) and baryon ( @xmath23 , @xmath24 , @xmath25 , @xmath26 ) yields , rapidity and multiplicity distributions have been measured with different event selections [ minimum bias , inelastic ( inel ) or non - single diffractive events ( nsd ) ] . for minimum bias event selection , different pythia parameter sets @xcite have difficulty reproducing ( multi)strange particle production , predicting too few strange particles and harder @xmath15 spectra , the differences with data increasing with the mass of the strange particle @xcite . up to now
, none of the mc event generators is able to describe the softer @xmath15 and the huge rise of particle production with energy .
this has led to a concerted effort to improve the available mc generators . in a string fragmentation phenomenology
, it has been proposed that the observed strong enhancement of strange particle production in nuclear collisions could be naturally explained via strong longitudinal color field effects ( slcf ) @xcite .
recently , an extension of color glass condensate ( cgc ) theory has proposed a more detailed dynamical `` glasma '' model @xcite of color ropes . in the string models , strong longitudinal fields ( flux tubes , effective strings ) decay into new ones by quark anti - quark ( @xmath27 ) or diquark anti - diquark ( qq-@xmath28 ) pair production and subsequently hadronize to produce the observed hadrons . due to confinement
, the color of these strings is restricted to a small area in transverse space . with increasing energy of the colliding particles , the number of strings grows and they start to overlap , forming clusters this can introduce a possible dependence of particle production on the energy density @xcite . we have studied @xcite the effect of strong longitudinal color fields ( slcf ) in @xmath0 collisions up to lhc energies in the framework of the hijing / bb v2.0 model , which combines ( collinear factorized ) pqcd multiple minijet production with soft longitudinal string excitation and hadronization .
the default vacuum string tension , @xmath29 = 1 gev / fm , is replaced by an effective energy dependent string tension , @xmath30 gev / fm that increases monotonically with center - of - mass energy .
the exponent @xmath31 is found to succeed at describing well the energy dependence of multiparticle observables for rhic , tevatron , as well as lhc data @xcite . in the hijing / bb v2.0 model
the rapid growth of @xmath32 at mid - rapidity with energy is due to the interplay of copious minijet production with increasing strong color field contributions .
however , the large ( strange)baryon - to - meson ratios recently measured at lhc energies , especially at @xmath2 = 7 tev , are not well described using this set of parameters . in this work
we will address this question and in addition we will discuss a possible dependence of the strength of strong color field on the event multiplicity .
we will show that the model predicts a very strong enhancement of ( multi)strange baryon - to - meson ratios in hm events .
if observed , this could be interpreted as a possible signature for formation of a deconfined but out of local thermal equilibrium mini quark - gluon plasma .
for a uniform chromo - electric flux tube with field ( _ e _ ) , the pair production rate @xcite per unit volume for a ( light)heavy quark ( @xmath33 ) is given by @xmath34 where @xmath35 ( diquark ) , @xmath36 ( strange ) , @xmath37 ( charm ) or @xmath38 ( bottom ) .
the _ current quark masses _ are @xmath39 = 0.45 gev @xcite , @xmath40 gev , @xmath41 gev , and @xmath42 gev @xcite .
the _ constituent quark masses _ of light non - strange quarks are @xmath43 = 0.23 gev , of the strange quark is @xmath44=0.35 gev @xcite , and of the diquark is @xmath45 gev @xcite . an enhanced rate for spontaneous pair production is naturally associated with `` _ strong chromo - electric fields _ '' , such that @xmath46 1 _ at least some of the time_. in a strong longitudinal color electric field , the heavier flavor suppression factor @xmath47 varies with string tension via the well known schwinger formula @xcite , @xmath48 for @xmath49 , @xmath36 , @xmath37 or @xmath38 and @xmath50 , @xmath51 . in the model calculations
, we assume the following effective masses : @xmath52 = 0.5 gev , @xmath53 = 0.28 gev , and @xmath54 = 1.30 gev . therefore , the above formula implies a suppression of heavier quark production according to @xmath55 : @xmath51 : @xmath56 : @xmath36 : @xmath37 @xmath18 1 : 1 : 0.02 : 0.3 : 10@xmath57 for the vacuum string tension @xmath29 = 1 gev / fm . for a color rope ( or cluster ) , on the other hand , if the _
average string tension _
value @xmath1 increases , the suppression factors @xmath47 increase ( i.e. , this implies a higher rate of @xmath58 pair production ) . using the hijing / bb v2.0 model , we have shown that it is important to consider that high energy @xmath0 collisions can have a substantial contribution from slcf effects @xcite . in the model phenomenology
, the degree of collectivity is described by the overlap of individual strings ( clusters ) , quantified by the infrared sensitive variable , the string tension ( @xmath1 ) . a reduction mechanism of strange quark suppression was introduced by assuming that the effective string tension increased with increasing reaction energy according to a power law : @xmath59 where @xmath5 = 1 gev / fm is the vacuum string tension value and @xmath60 = 1 gev@xmath61 is a scale factor .
in addition to describing well the energy dependence of charged particle density at mid - rapidity from sps to lhc energies , we have shown that this dynamical mechanism improves the description of the strange meson / hyperon data at tevatron and lhc energies @xcite .
however , using a set of parameters corresponding to those obtained from eq .
[ eq : eq3 ] , results in an over - prediction of the recently measured yields @xcite of @xmath62 particles ( by up to a factor of two ) and of @xmath26 particles ( by up to a factor of four ) .
a possible reason for this could be a too strong energy dependence used for the mean string tension values ( see eq .
[ eq : eq3 ] ) . therefore ,
here we consider a weaker energy dependence of the form , @xmath63 this leads to a value for the mean string tension of @xmath64 gev / fm at 0.9 tev and @xmath65 gev / fm at 7 tev .
the results obtained using this new parametrization will be discussed in section iii . because the threshold for strange quark production in a deconfined phase ( or in a mqgp ) is much smaller than in a hadron gas , a larger enhancement in strange particle production has been suggested as an indication of possible mqgp formation @xcite .
equation [ eq : eq2 ] , which describes strangeness suppression factors @xmath66 , shows that an increased value @xmath1 up to @xmath67 gev / fm leads to a saturation of suppression factors , @xmath66 .
for example , in @xmath0 collisions at 7 tev the suppression factor @xmath66 increases from 0.4 ( corresponding to @xmath68 gev / fm ) to approximately 0.85 ( corresponding to @xmath69 gev / fm ) , and has only a modest further increase up to 0.89 for @xmath70 gev / fm .
therefore , if we assume a different energy density in mb and hm @xmath0 collision events ( corresponding also to a different type of ropes formation ) could lead to different effective @xmath1 values .
the @xmath1 value obtained from eq .
[ eq : eq4 ] @xmath1= @xmath71 can be associated with mb events , since it is deduced from inclusive data . for hm events we will assume an extreme value of @xmath72 gev / fm , which corresponds to a saturation of strangeness suppression factors . to better characterize the sensitivity to the parameter @xmath1
, we also present the result corresponding to an intermediate value @xmath1= @xmath73 .
note that a value @xmath72 gev / fm is also supported by recent calculations at finite temperature ( @xmath74 ) of potentials associated with a @xmath27 pair separated by a distance @xmath75 @xcite . the finite temperature ( @xmath74 ) form of the @xmath27 potential has been calculated by means of lattice qcd @xcite . at finite temperature
, there are two potentials associated with a @xmath27 pair separated by a distance @xmath75 : the free energy @xmath76 and internal energy @xmath77 .
the free and internal energies actually correspond to slow and fast ( relative ) motion of the charges , respectively @xcite .
infrared sensitive variables such as string tension are very helpful to identify specific degrees of freedom of the plasma .
since the confinement of color in non - abelian theories is due to the magnetic degree of freedom , the magnetic component is expected to be present in the plasma as well @xcite . in the presence of the _ chromo - magnetic scenario _
it was shown that the effective string tension of the free energy @xmath1 = @xmath78 decreases with @xmath74 , to near zero at critical temperature ( @xmath79 ) @xcite .
in contrast , the effective string tension of the internal energy ( corresponding to a fast relative motion of the charges ) @xmath1 = @xmath80 remains nonzero below about @xmath81 with a peak value at @xmath79 about 5 times the vacuum tension @xmath82 ( @xmath80 = 5 @xmath82 = 5 gev / fm ) @xcite .
all details of the hijing / bb v2.0 model are extensively discussed in the literature @xcite . here
we focus our analysis at two energies of interest : @xmath83 tev and @xmath84 tev , where data for charged particles @xcite and identified particles @xcite have been reported for mb events . except for the new energy dependence of the string tension ( eq .
[ eq : eq4 ] ) , all the other parameters of the models are as in ref . @xcite .
note that this modification also leads to a relatively small increase ( approximately 10 - 12% ) for the predicted density of charged particles at mid - rapidity , in comparison with previous results reported in ref .
@xcite in the entire energy region of interest .
this increase is now also supported by the new experimental data with lhc energy @xmath85 tev @xcite .
[ h ! ] tev ( left panel ) and @xmath85 tev ( right panel ) for non - single - diffractive ( nsd ) @xmath0 collisions .
the solid and dashed histograms are the results with and without scf , respectively .
the data are from refs .
@xcite ( cms ) and from ref .
@xcite ( totem ) . only statistical error bars are shown .
[ fig : fig1 ] , title="fig:",height=264 ] 0.5 cm charged hadron multiplicity measurements are the first results of the lhc physics program .
the new data on charged particle pseudorapidity distributions @xcite , over a limited @xmath86 range for non single diffractive interactions ( nsd ) , are compared to model calculations in fig .
[ fig : fig1 ] .
the data show a sizeable increase of the central pseudorapidity density with c.m.s . energy . as the colliding energy increases
, the rate of multiple parton interactions ( mpi ) also increases , producing a rise in the central multiplicity .
the increase with energy in our phenomenology is due to the interplay of the increased mini - jet production in high colliding energy with slcf effects .
a scenario with slcf effects ( solid histograms ) reproduces well the measured multiplicity distributions . without slcf effects ( _ i.e. _ , @xmath87 gev / fm ) the model strongly overestimates the central charged particle density ( dashed histograms ) .
data over a larger rapidity range are needed to determine the shape of the falling density in the fragmentation region . for completeness , the new totem data @xcite at forward pseudorapidity are also included .
[ h ! ] ) at lhc energies .
the calculated spectra include the combined effects of slcf and jj loops .
the solid and dashed histograms have the same meaning as in fig .
[ fig : fig1 ] .
the data are from refs .
@xcite ( cms ) , @xcite ( alice preliminary ) , @xcite ( atlas ) .
statistical error bars on the data points are smaller than the markers .
[ fig : fig2 ] , title="fig:",height=264 ] 0.5 cm the measured transverse momentum distributions for nsd events at @xmath2 = 0.9 tev and @xmath2 = 7 tev are shown in fig .
[ fig : fig2 ] in the range @xmath88 gev / c , where both hard and soft processes are expected to contribute .
the data of cms @xcite and atlas @xcite are measured in larger pseudorapidity intervals @xmath89 and @xmath90 , respectively .
in contrast , alice measurements @xcite are in a very central region @xmath91 .
the calculations were performed using alice acceptance but , as can inferred from the data shown in fig .
[ fig : fig1 ] , the difference in pseudorapidity range has a negligible effect on the @xmath15 spectra .
the model calculations including slcf effects give a good description of the spectral shape at low @xmath15 ( @xmath92 gev / c ) for both energies . at high
@xmath15 ( @xmath93 gev / c ) the calculations lead to a somewhat harder spectrum than that observed . in our phenomenology
this could indicate that jet quenching , _
i.e. _ , suppression of high @xmath15 particles like that observed at rhic energies in nucleus - nucleus collisions , could also appear in @xmath0 collisions , particularly for events with large multiplicity @xcite .
the @xmath94 single particle inclusive @xmath15 spectra measurements are important for understanding collision dynamics , since the various particles show different systematic behavior , as observed at rhic energy @xcite . detailed theoretical predictions for single inclusive hadron production ( including hyperons ) are discussed in this section .
baryon - to - meson ratios are experimental observables that can be used at the lhc to investigate multi - parton interactions and help understanding of the underlying physics @xcite .
unexpectedly high ratios observed in @xmath11 collisions at rhic energies have been discussed in terms of recombination and coalescence mechanisms @xcite . in @xmath0 collisions , however , a coalescence / hadronization scenario is not favored due to low phase space density in the final state .
the hijing / bb model @xcite , with slcf effects and junction - anti - junction ( @xmath95 ) loops included , provides an alternative dynamical explanation of the heavy - ion data at rhic energies .
we have shown that the model also predicts an increasing yield of ( multi)strange particles , thereby better describing the experimental data in @xmath11 and @xmath0 collisions .
[ h ! ] ) for mesons ( @xmath96 ) and baryons ( @xmath23 ) at @xmath2 = 0.9 tev ( left panel ) and at @xmath2 = 7 tev ( right panel ) are compared to data .
the data ( open symbols , left panel ) are from ref .
@xcite ( alice ) .
the preliminary data ( open symbols , right panel ) are from ref .
@xcite ( alice ) .
preliminary data ( closed symbols ) are from cms collaboration and are plotted using a normalization factor 0.78 discussed in ref .
statistical error bars on the data points are smaller than the markers . for clarity
the data and theoretical calculations for the proton results are divided by a factor of ten .
[ fig : fig3 ] , title="fig:",height=264 ] 0.5 cm figure [ fig : fig3 ] compares the alice results to the predicted mid - rapidity spectra for positive pions ( solid histograms ) , kaons ( dashed histograms ) , and protons ( dotted histograms ) in minimum bias @xmath0 collisions .
there is agreement in the @xmath15 region of interest @xmath97 gev/_c _ at both energies .
the over - prediction for proton and kaon production below @xmath98 gev/_c _ is consistent with possible presence of radial flow , which seems to be larger at 7 tev than at 0.9 tev .
the radial flow could appear as a consequence of a hydrodynamic type evolution with flux tube initial conditions @xcite , not embedded in our model .
[ h ! ] ) for ( multi)strange hyperons ( @xmath24 , @xmath62 , @xmath26 ) at @xmath2 = 0.9 tev ( left panel ) and at @xmath2 = 7 tev ( right panel ) are compared to data .
the data are from the alice @xcite and cms collaborations @xcite .
error bars include only the statistical uncertainties .
[ fig : fig4 ] , title="fig:",height=264 ] 0.5 cm we extend our analysis to the production of ( multi)strange baryons . in fig .
[ fig : fig4 ] we show the hijing / bb v2.0 model predictions of @xmath15 spectra at mid - rapidity ( @xmath99 ) for @xmath24 ( solid histograms ) , @xmath25 ( dashed histograms ) and @xmath100 ( dotted histograms ) baryons at 0.9 tev and 7 tev . for ( multi)strange particles ,
the data indicate a stronger radial flow at 7 tev than at 0.9 tev . for the @xmath15 region of interest ,
@xmath101 gev/_c _ , the model results are in agreement with data at both energies .
figure [ fig : fig5 ] shows a comparison of model predictions with alice data @xcite of non - strange baryon over meson ratios ( @xmath102 ) at 0.9 tev and 7 tev .
the ratios have been calculated by dividing the spectra reported in refs .
@xcite . within our phenomenology the measured ratios
are reasonably described in a scenario with sclf effects ( solid histograms ) . the larger string tension parameterization results in a predicted increase of the ratio @xmath102 by a factor of @xmath18 5 at 7 tev .
note that the models pythia @xcite and epos @xcite can not reproduce the observed high baryon - to - meson ratios ( see fig .
6 and fig . 7 in ref .
@xcite ) .
[ h ! ] at @xmath2 = 0.9 tev ( left panel ) and at @xmath2 = 7 tev ( right panel ) .
the solid and dashed lines have the same meaning as in fig .
[ fig : fig1 ] . experimental results at 0.9 tev are from ref .
@xcite ( alice ) and at 7 tev from ref .
@xcite(alice preliminary ) .
the ratios have been calculated by us , dividing the spectra from fig .
[ fig : fig3 ] .
error bars include only the statistical uncertainties .
[ fig : fig5 ] , title="fig:",height=264 ] 0.5 cm the strange - particle ratios could also reveal manifestations of new collective phenomena . in the epos model such an increase
is obtained if the production of a _ mini - plasma _ is assumed in @xmath0 collisions @xcite , @xcite .
if confirmed by future measurements , these observables could open a perspective on new physics in @xmath94 interactions .
[ h ! ] collisions at @xmath2 = 7 tev for ratio @xmath103 ( left panel ) are compared to data .
the calculated ratios ( solid histogram ) include the combined effects of scf and jj loops .
the dashed histogram are the results without scf effects .
pythia simulation results from ref .
@xcite are also included ( left panel ) .
the data are from ref .
@xcite ( alice ) .
statistical error bars on the data points are smaller than the markers .
the model predictions for the ratio of multi - strange hyperons to mesons @xmath104 ( right panel ) .
[ fig : fig6 ] , title="fig:",height=264 ] 0.5 cm to investigate possible differences in the production mechanisms of multi - strange baryons that do or do not contain a non - strange quark we study the ratio of @xmath105-to-@xmath106 as a function of transverse momentum , @xmath15 . due to low statistics
we will consider the combined ratio ( @xmath107)-to-(@xmath108 ) at 7 tev : the predictions are shown in fig .
[ fig : fig6]a .
the model results without slcf effects ( dashed histogram ) underestimate the data by a factor of approximately 3 .
note that recent calculations with the pythia event generator @xcite strongly underestimate the production rates @xcite and fail to describe the above ratio @xcite .
in contrast , the hijing / bb v2.0 model in a scenario with slcf effects ( solid histograms ) does describes the data .
this ratio could also help to study the possible saturation of the s - quarks , which would be indicated by a flattening at high @xmath15 .
however , the currently available data do not allow for firm conclusions .
it has been argued that strangeness production could be suppressed in @xmath0 collisions by the limited volume of the colliding system , which requires localized strangeness conservations @xcite .
such a canonical suppression does not , however , explain the suppression of the @xmath109 meson production in @xmath0 collisions because @xmath109 has a net strangeness of zero @xcite .
the study of the @xmath104 ratio is therefore also of great interest to distinguish between possible dynamical production mechanisms .
the hijing / bb v2.0 model predictions are presented in fig .
[ fig : fig6]b .
the results predict a strong increase ( up to an order of magnitude ) for the scenario that includes slcf effects ( solid histogram ) . in our approach ,
the dynamical mechanism that leads to such high values of baryon - to - meson ratios is slcf appearing at the initial stage of the interaction .
the slcf mechanism strongly modifies the fragmentation processes ( strangeness suppression factors ) and thus results in a large increase of ( strange)baryons .
this interpretation is also supported by more sophisticated theoretical calculations , in a scenario in which a time - dependent pulse for the initial strength of the color field is considered @xcite .
an observed large enhancement of the baryon - to - meson ratios would be consistent with slcf playing an important role in multiparticle production in @xmath0 collisions at lhc energies and suggesting that high energy density fluctuations can reach very high densities , potentially comparable to those reached in central au + au collisions at rhic energies @xcite . to test the above assumptions , and in order to study possible new phenomena in @xmath0 collisions , we examine in this paper the dependence of particle production as a function of the total charged particle multiplicity ( @xmath110 ) and compare the results with those from minimum bias event selection .
the mb event selection is defined here as the existence of one charged particle in the rapidity interval @xmath111 . since ( multi)strange particle production in heavy - ion reactions is enhanced relative to that in mb @xmath0 collisions @xcite
, one might ask whether the production rates in hm @xmath0 collisions may already exhibit any feature like an enhancement due to slcf effects . the increase is quantified by calculating the @xmath15 dependence at mid - rapidity ( @xmath112 ) of the baryon / meson yield ratios , _
e.g. _ , ( multi)strange baryon ( @xmath24 , @xmath62 , and @xmath26 ) over mesons ( pions ) . [ h ! ] , ( a ) ] , [ @xmath113 , ( b ) ] [ @xmath114 , ( c ) ] [ @xmath115 , ( d ) ] . the results are plotted at mid - rapidity ( @xmath116 ) taking @xmath6 gev / fm in the mb event selection ( dotted histograms ) and hm events with @xmath117 ( dashed histograms ) . for the hm events
selection the predictions are also shown considering @xmath69 gev / fm ( solid histograms ) .
the experimental ratios in ( a ) have been calculated by dividing the spectra from ref .
[ fig : fig7 ] , title="fig:",height=510 ] 0.5 cm figure [ fig : fig7 ] shows the model predictions for various baryon - to - meson ratios as a function of transverse momentum . shown
are the results for non - strange baryon over non - strange meson production [ @xmath118 , part ( a ) ] , strange baryon over non - strange meson production [ @xmath113 , part ( b ) ] multi - strange baryon over non - strange meson production [ @xmath114 , part ( c ) ] , and multi - strange baryon over @xmath109 mesons [ @xmath115 , part ( d ) ] for mb and hm ( @xmath117 ) events .
the results are presented for two scenarios : mb and hm ( @xmath117 ) events assuming @xmath1= 2 gev / fm , and for hm events assuming @xmath69 gev / fm .
one first notes that , for a constant value of @xmath1 , there are negligible differences between the predicted ratios for mb and hm events . on the contrary , assuming a larger value of @xmath1 leads to large increases that depend on particle species .
the ratio of non - strange baryon over non - strange mesons @xmath118 ) shows a multiplicative enhancement of approximately @xmath119 relative to mb and hm results with @xmath1= 2 gev / fm ( fig [ fig : fig7 ] a ) .
however , the enhancement for ( multi)strange baryon over meson ratio increases with increasing mass of hyperons : up to a factor of four for @xmath113 ( fig .
[ fig : fig7 ] b ) , up to a factor of ten for @xmath120 ( not shown here ) , and up to a factor of twenty for @xmath114 ( fig . [
fig : fig7 ] c ) and @xmath115 ( fig .
[ fig : fig7 ] d ) .
recently , new ( preliminary ) data for the ratio of non - strange baryons over mesons , @xmath118 as function of charged particle multiplicity have been reported by the cms collaboration@xcite .
these data do not show a dependence of the measured ratio on the multiplicity of the event , consistent with a scenario assuming a constant value for @xmath1 .
however , the measurements are performed within a very limited @xmath15 range ( @xmath121 gev / c ) and the selection of events is neither nsd nor inel and make comparison with theory difficult .
data on ( multi)strange particle production extending over a larger @xmath15 range are more sensitive to scf and would allow drawing a more definitive conclusion . with high statistics measurements of identified particles , the lhc collaborations could test the model predictions and lend credence to the idea that new phenomena or possible out of equilibrium mqgp has been formed in hm @xmath0 collisions .
note that the alice collaboration recently reported measurements of the inclusive @xmath122 yield as a function of charged particle density at mid - rapidity ( @xmath123 ) in mb and hm events at 7 tev .
hm events were selected with different bins , up to four times mb multiplicity density . in these hm events an enhancement by a factor of about eight for @xmath122 yields at mid - rapidity ( @xmath124 ) was found relative to those in mb events @xcite .
preliminary alice data @xcite on @xmath24 and @xmath22 nuclear modification factors in central 0 - 5% pb+pb collisions at 2.76a tev appear also to agree qualitatively with our predictions of enhanced hyperon / meson yield ratios and their nuclear quenching pattern at transverse momentum @xmath125 gev/_c _ ( see fig . 8 from ref .
@xcite ) . in this paper
we propose that a similar enhancement of the baryon - to - meson ratios may be observed in rare hm @xmath0 collisions . just as in the pb+pb collisions , the hyperon / meson enhancement and the quenching pattern of @xmath24 , @xmath22 and @xmath126 in hm @xmath0 collision events relative to those in mb , corroborated with other bulk flow correlations , could provide further evidence of out of equilibrium mqgp production in rare hm @xmath0 reactions .
a mini quark - gluon plasma differs from the strong coupling plasma ( sqgp ) produced in central nucleus - nucleus ( @xmath11 ) collisions , mostly by its small initial transverse size @xmath127 fm compared to the significantly larger nuclear transverse size @xmath128 fm .
extensive hydrodynamic @xcite and transport calculations @xcite have shown that collective flow signatures , such as perfect fluid elliptic flow , require several fm/_c _ to develop .
similarly , jet quenching observables depend strongly on the jet path length and thus , in @xmath0 collisions quenching will be much weaker than in pb+pb central collisions .
in contrast , signatures associated with strangeness equilibration @xcite,@xcite are generated on much faster time scales .
therefore , hyperon - to - meson ratios and their @xmath15 dependence may serve as the best probe of possible out of equilibrium mqgp formation . in order to explore better the enhancement of the baryon / meson production ratio as a possible signature of mqgp in @xmath0 collisions , and especially its possible dependence on event multiplicity ,
the particle yields at mid - rapidity ( @xmath129 ) are studied in six bins of charged particle multiplicity : @xmath130 ; @xmath131 ; @xmath132 ; @xmath133 ; @xmath134 ; @xmath135 .
we select an experimental observable that would be sensitive to possible new phenomena and is well adapted to the low statistics expected from low yield particle production .
we consider the integrated values @xmath136 of the baryon - to - meson yield ratios shown in fig .
[ fig : fig7 ] .
the domain of the integration is taken over the range @xmath97 gev/_c _ , where the increase appears to be strongest .
the calculations of @xmath136 are performed for the above six bins of multiplicity .
we also define @xmath137 for mb events , _
i.e. _ , the integrated ratio without selection on multiplicity .
for @xmath137 the mean value of the string tension is taken as @xmath138 gev / fm .
for the multiplicity bins , the calculations are performed for both @xmath138 gev / fm ( @xmath139 ) and @xmath69 gev / fm ( @xmath140 ) . since we expect a gradual transition as function of multiplicity an intermediate value @xmath141 gev / fm ( @xmath142 ) is also considered .
( see text for explanation ) .
the results are plotted as a function of multiplicity in five bins : @xmath131 ; @xmath132 ; @xmath143 ; @xmath144 ; @xmath117 .
the values @xmath145 are given for @xmath113 ( left panel ) and for @xmath114 ( right panel ) .
the results for the relative increase @xmath146 ( triangles ) and @xmath147 ( stars ) are obtained using @xmath141 gev / fm and @xmath69 gev / fm , respectively .
the integrated values @xmath137 are obtained using @xmath148 gev / fm @xcite .
[ fig : fig8 ] , title="fig:",height=264 ] 0.5 cm in figure [ fig : fig8 ] , the theoretical predictions for the relative increase in the integrated values @xmath149 to the integrated values @xmath137 , _
i.e. _ , the ratio @xmath150 , are shown .
note that the statistical fluctuations are too large in the first bin ( @xmath151 ) so it is not included in the plot .
the calculations are presented for the ratios @xmath152= @xmath153 , @xmath154= @xmath155 , and @xmath156 . as examples , the predictions for ( multi)strange - baryon - to - meson ratios are given , _
i.e. _ , @xmath113 and @xmath114 . for the @xmath113 ratio ,
the model predicts an almost constant value of @xmath157 with no multiplicity dependence , while a slight dependence is predicted for the ratio @xmath158 of @xmath114 . for
the ratio @xmath113 an enhancement of a factor of 4 of @xmath158 over @xmath152 is predicted ( fig .
[ fig : fig8]a ) .
in contrast , for @xmath114 ( fig . [
fig : fig8]b ) higher values ( up to a factor of approximately 40 ) are predicted in the scenario with a possible transition to an out of equilibrium mqgp state , @xmath158 ( solid line ) , than in a scenario without , @xmath152 ( dotted line ) .
if the assumption of the dependence of @xmath1 on the total charged particle multiplicity ( @xmath110 ) is valid , we expect to see in the data a transition from @xmath152 ( dotted line ) at low multiplicity to @xmath154 ( dashed line ) , to @xmath158 values ( solid line ) for higher multiplicity events .
these results strongly suggest that the experimental data could contain a signature for a possible transition to a mqgp phase in hadronic collisions .
the shape of an observed transition as a function of multiplicity could contain information on the nature of the underlying physics , _
e.g. _ , if the transition is smooth or has a net threshold .
we showed that the baryon - to - meson ratios have an enhancement up to the highest lhc energy ( 14 tev ) @xcite .
based on eq .
[ eq : eq4 ] the string tension value has a predicted modest increase from @xmath159 gev / fm to @xmath160 gev / fm , when going from @xmath85 tev to @xmath161 tev . in addition , as shown in ref .
@xcite , a saturation sets in near @xmath162 tev .
therefore , we expect relatively small further increases in the strange particle ratios and in the predicted values for @xmath157 in hm @xmath0 collisions at higher lhc energy .
in this work within the phenomenology of the hijing / bb v2.0 model we discussed observables sensitive to possible new phenomena , such as strong longitudinal color fields in hm @xmath0 collision events . for mb bias events we show that a good description is obtained for charged and identified particle production , taking an energy dependence of mean string tension values @xmath4 .
the predictions for baryon / meson production ratios in @xmath0 collisions at @xmath85 tev are discussed .
we analyze the dependence of these ratios on the degree of collectivity in the reaction dynamics , characterized by an infrared sensitive variable , the string tension @xmath1 .
the formation of a collective phase in high multiplicity @xmath0 collisions would be made evident within our phenomenology by a relative increase in the baryon / meson ratios with increasing multiplicity , particularly those ratios involving strange and multi - strange particles .
the experimental data could show a multiplicity dependent transition as indicated by comparing results obtained with a lower value of @xmath6 gev / fm to those corresponding to a higher value of @xmath69 gev / fm originating from possible production of a transitory out of equilibrium mqgp phase .
this transition is most likely gradual as a function of multiplicity and center - of - mass energy @xmath2 .
corroborated with other observables sensitive to collective behavior , such as _ ridge structure _ , enhanced radial flow , high @xmath15 particle suppression ( _ jet quenching _ ) in @xmath0 collisions , observation of saturation of hyperon / meson ratios could provide possible signatures of out of equilibrium mqgp phase formed in high multiplicity @xmath0 collision events at ultra - high energies .
0.2 cm * acknowledgments : * we thank s. das gupta for useful discussions and continue support .
vtp acknowledges discussion with m. petrovici and access to computer facilities ifin - hh , bucharest , romania and at columbia university , new york , where parts of these calculations were performed .
this work was supported by the natural sciences and engineering research council of canada .
this work was also supported by the division of nuclear science , of the u. s. department of energy under contract no .
de - ac03 - 76sf00098 and de - fg02 - 93er-40764 .
n. armesto , n. borghini , s. jeon , u. a. wiedemann ( eds . ) , s. abreu , v. akkelin , j. alam and j. l. albacete _ et al .
_ , j. phys .
g * 35 * , 054001 ( 2008 ) .
k. werner , i. karpenko , t. pierog , m. bleicher and k. mikhailov , phys .
c * 83 * , 044915 ( 2011 ) .
l. cunqueiro , j. dias de deus and c. pajares , eur .
j. c * 65 * , 423 ( 2010 ) .
d. velicanu ( cms collaboration ) , j. phys .
g * 38 * , 124051 ( 2011 ) .
et al . _ ( atlas collaboration ) , atlas - conf-2011 - 055 , 19@xmath163 international workshop on deep inelastic scattering and related subjects , 11 - 15 april , 2011 , newport news , va , usa m. y. azarkin , i. m. dremin and a. v. leonidov , mod .
lett . a * 26 * , 963 ( 2011 ) , [ erratum - ibid .
a * 26 * , 1309 ( 2011 ) . ]
r. c. hwa and c. b. yang , phys .
c * 83 * , 024911 ( 2011 ) .
a. dumitru , k. dusling , f. gelis , j. jalilian - marian , t. lappi and r. venugopalan , phys .
b * 697 * , 21 ( 2011 ) .
t. a. trainor and d. t. kettler , phys .
c * 84 * , 024910 ( 2011 ) .
e. shuryak , arxiv:1009.4635 [ hep - ph ] .
m. chojnacki ( alice collaboration ) , j. phys .
g * 38 * , 124074 ( 2011 ) .
m. floris ( alice collaboration ) , j. phys .
g * 38 * , 124025 ( 2011 ) .
d. d. chinellato ( alice collaboration ) , j. phys .
g * 38 * , 124079 ( 2011 ) .
b. abelev _ et al . _
( alice collaboration ) , phys .
b * 712 * , 309 ( 2012 ) .
r. rougny , to be published in proceedings of hadron collider physics symposium 2011 ( hcp 2011),14 - 18 nov 2011 : paris , france , arxiv:1201.5563v2 [ hep - ex ] .
v. khachatryan _ et al . _
( cms collaboration ) , cms pas fsq-12 - 014 ( 2012 ) .
m. n. chernodub and v. i. zakharov , phys .
lett . * 98 * , 082002 ( 2007 ) .
v. khachatryan _ et al . _
( cms collaboration ) , jhep * 1002 * , 041 ( 2010 ) .
v. khachatryan _ et al . _
( cms collaboration ) , phys .
* 105 * , 022002 ( 2010 ) .
h. ricaud , a. kalweit and a. maire , j. phys .
g * 37 * , 094049 ( 2010 ) .
r. j. fries , b. muller , c. nonaka and s. a. bass , phys . rev .
c * 68 * , 044902 ( 2003 ) .
v. greco , c. m. ko and p. levai , phys .
c * 68 * , 034904 ( 2003 ) .
r. c. hwa and c. b. yang , phys .
c * 67 * , 034902 ( 2003 ) .
k. werner , t. hirano , i. karpenko , t. pierog , s. porteboeuf , m. bleicher and s. haussler , nucl .
. suppl . * 196 * , 36 ( 2009 ) .
k. werner , i. karpenko and t. pierog , j. phys .
conf . ser .
* 230 * , 012026 ( 2010 ) .
v. skokov and p. levai , pos * eps - hep2009 * , 456 ( 2009 ) ; p. levai and v. skokov , phys .
d * 82 * , 074014 ( 2010);p .
levai , d. berenyi , a. pasztor and v. v. skokov , j .
g * 38 * , 124155 ( 2011 ) .
i. kraus , j. cleymans , h. oeschler and k. redlich , j. phys .
g * 37 * , 094021 ( 2010 ) .
f. becattini , j. phys .
g * 36 * , 064019 ( 2009 ) .
n. xu ( star collaboration ) , j. phys .
g * 35 * , 044031 ( 2008 ) .
b. abelev _ et al . _
( alice collaboration ) , phys .
b * 712 * , 165 ( 2012 ) .
s. schuchman , work presented at 7@xmath163 int .
workshop on high @xmath15 physics at the lhc , 26 - 29 march , 2012 , frankfurt , germany ( private communication ) . | we use the framework of the hijing / bb v2.0 model to simulate high - multiplicity ( hm ) @xmath0 collision events at the large hadron collider ( lhc ) to study observables sensitive to possible collective phenomena , such as strong longitudinal color fields ( slcf ) modeled by an enhanced string tension ( @xmath1 ) .
we focus on the hyperon / meson yield ratios at center - of - mass ( c.m . )
energy @xmath2 = 7 tev , in the transverse momentum region , @xmath3 gev/_c_. for minimum bias events these ratios are well described assuming an energy dependence @xmath4 + ( @xmath5= 1 gev / fm ) , giving a value @xmath6 gev / fm at @xmath2 = 7 tev .
we compare minimum bias ( mb ) events to simulated hm events assuming that @xmath7 gev / fm could grow to an extreme value of @xmath8 gev / fm that saturates the strangeness suppression factor . with this assumption
the model predicts a very strong enhancement of ( multi)strange baryon / meson ratios in hm events .
if observed , such an enhancement could be also interpreted as a possible signature for formation in hm @xmath0 collision events of a deconfined but out of local thermal equilibrium _
mini quark - gluon plasma _
( mqgp ) . |
since the discovery of the fascinatingly robust transport properties of graphene @xcite , there has been an increasing interest in other two - dimensional systems with similar spectral properties , such as the surface of topological insulators @xcite . in all these systems the transport is dominated by a band structure , in which two bands touch each other at nodes . if the fermi energy is exactly at or close to these nodes , the point - like fermi surface and interband scattering lead to particular transport properties , such as a robust minimal conductivity .
based on these results , an extension of the nodal spectral structure to three - dimensional ( 3d ) systems is of interest @xcite . in 3d the fermi surface is a sphere with radius @xmath0 rather than the circular fermi surface in 2d , which is either occupied by electrons ( @xmath1 ) or by holes ( @xmath2 ) . for @xmath3
the conductivity vanishes in the absence of impurity scattering in contrast to the minimal conductivity of the 2d system . on the other hand
, sufficiently strong impurity scattering leads to a conductivity at the node @xmath3 .
thus , an important difference between 2d and 3d weyl fermions is that there exists a metal - insulator transition in the latter , which is driven by increasing disorder @xcite .
this transition is similar to the metal - insulator transition caused by decreasing random gap fluctuations in a system of 2d dirac fermions @xcite . on the other hand ,
it is quite different from an anderson transition from a metallic state at weak scattering to an insulating state at strong scattering , which is caused by anderson localization at strong scattering @xcite .
there is agreement between the various approaches , based on self - consistent , perturbative and numerical methods , on the existence of a transition from a 3d weyl semimetal at weak scattering to a diffusive metallic behavior at stronger random scattering @xcite .
this transition can be characterized by a vanishing density of states at the weyl node ( i.e. , the scattering rate or the imaginary part of the self - energy ) and a nonzero density of states in the diffusive phase .
however , the transport properties for the weak scattering regime are still under discussion .
in particular , a recent study indicates that there is a metal - metal transition rather than a insulator - metal transition for 3d weyl fermions with a critical point @xcite .
we will address this problem in the subsequent calculation , using a weak scattering approach ( wsa ) .
calculations of quantum transport consist usually of two steps : determining the scattering time ( or scattering rate ) within a self - consistent solution of the dyson equation , also known as the self - consistent born approximation ( scba ) , and determining the conductivity by a self - consistent solution of the bethe - salpeter equation ( bse ) .
this approach , in particular the solution of the bse , is rather complex due to the existence of many modes .
not all of them are relevant for the transport properties because some decay quickly . from this point of view
it is easier to project at the beginning only onto those modes which do not decay quickly but control the transport properties on large scales . for a system with spectral nodes in a disordered environment these modes
are a result of a spontaneously broken chiral symmetry @xcite .
we will employ this idea here to 3d weyl fermions in order to calculate the conductivity .
for this purpose it is important to identify the underlying symmetries of the two - particle green s function .
then spontaneous symmetry breaking is characterized by its non - vanishing order parameter which is the scattering rate .
there is a metallic phase with long range correlations ( i.e. diffusion ) , whereas in the insulating phase the symmetry remains unbroken .
the paper is organized as follows : in sec .
ii we define the model and discuss the symmetry properties of the two - particle green s function . then the dc conductivity is calculated within a weak scattering approach ( sect .
iii ) , using an expansion in powers of the scattering rate @xmath4 .
this provides us a formula for the dc conductivity , which is discussed in sect .
iv at the node ( sect . iva ) and away from the node ( sect .
our discussion includes a comparison with the results of the boltzmann approach and with results from an approach based on the scba and bse of refs . @xcite .
the three - dimensional weyl hamiltonian for electrons with momentum @xmath5 is expanded in terms of pauli matrices @xmath6 ( @xmath7 ) as h = v_f - u_0 with = ( _ 1,_2,_3 ) [ ham00 ] with fermi velocity @xmath8 .
@xmath9 is a disorder term , represented by a random potential with mean @xmath10 ( fermi energy ) and variance @xmath11 .
the average hamiltonian @xmath12 generates a spherical fermi surface with radius @xmath0 , and with electrons ( holes ) for @xmath1 ( @xmath2 ) .
physical quantities are expressed in such units that @xmath13 .
the electronic conductivity , obtained as the response to a weak external field with frequency @xmath14 @xcite ( ) = -^2_rr_k^2 a_r0 ( ) , [ conda1 ] is given by the correlation function of the green s functions @xmath15 a_rr()=_0 tr_2[(h + z)^-1_rr(h - z)^-1_rr ] with z=/2+i , [ corr1 ] where @xmath16 represents disorder average and @xmath17 is the trace with respect to the pauli matrix structure .
this expression , often called the two - particle green s function , has two different energies @xmath18 for the same hamiltonian @xmath19 to create two independent green s functions @xmath15 .
now we represent this two - particle green s function by two different hamiltonians and one energy : we define the pair of hamiltonians h_=p_1_1+p_2_2 ( p_3_3-u_0 ) , where @xmath20 . the matrix transposition @xmath21 relates @xmath22 and @xmath23 through the identity _ 1 h_^t_1 = -h _ , [ id1 ] since @xmath24 .
this allows us to write for the correlation function ( [ corr1 ] ) a_rr()=-_0 tr_2[(h_+ + z)^-1_rr_1(h_-^t+z)^-1_rr_1 ] .
[ corr1a ] instead of two different energies @xmath18 and the same hamiltonian @xmath19 , the two - particle green s function has now the same energy @xmath25 but different hamiltonians , namely @xmath22 and @xmath26 .
the relation ( [ id1 ] ) and the representation ( [ corr1a ] ) reveals an internal structure of the model which leads to the hamiltonian = .
[ 8by8a ] the green s functions @xmath27 and @xmath28 in eq .
( [ corr1a ] ) are just the first and the third diagonal element of the green s function @xmath29 .
this indicates that the transport properties of the original hamiltonian @xmath19 , which requires two different energies @xmath18 , are related to the transport properties of the extended hamiltonian @xmath30 at the same energy @xmath25 .
the extended hamiltonian @xmath30 , its symmetries and its relation to diffusive transport were studied previously @xcite . in particular , it was found , together with property ( [ id1 ] ) , that the matrix = [ symm_tr ] with scalar variables @xmath31 anticommutes with @xmath30 : @xmath32 .
this relation implies a non - abelian chiral symmetry @xcite : e^she^s = h [ symmetry ] which is a symmetry relation for the extended hamiltonian @xmath30 in eq .
( [ 8by8a ] ) .
the term proportional to @xmath25 in the green s function @xmath33 breaks this symmetry due to @xmath34 , and therefore , @xmath35 $ ] plays the role of an order parameter for spontaneous symmetry breaking : ( z)-g(-z)=-2zg(z)g(-z)=-2z(h^2-z^2)^-1 .
[ op ] since the diagonal elements of this expression are proportional to the density of states at the node when we take the limit @xmath36 , a non - vanishing density of states indicates spontaneous symmetry breaking .
the role of a non - vanishing density of states at the node as an order parameter for a diffusive metallic phase was also discussed in refs .
@xcite . following the recipe of ref .
@xcite the correlation function ( [ corr1a ] ) can be expressed as a diffusion propagator .
this is used in the next section , where we focus on the long - range behavior of @xmath37 to calculate the conductivity @xmath38 .
the scattering rate @xmath4 is defined by = .
[ scatt_rate ] this definition is motivated by the assumption of a complex self - energy for the average one - particle green s function @xmath39 , whose imaginary part is the scattering rate ( cf .
[ sect : self - consist ] ) .
the corresponding scattering time @xmath40 is @xmath41 .
@xmath4 can either be calculated , for instance , within the scba @xcite or it can be taken from experimental measurements .
as discussed in the previous section , a non - vanishing scattering rate indicates spontaneous symmetry breaking .
since the broken symmetry is continuous , there exists a massless mode .
the latter is reflected by the relation _ ra_rr()=tr_2_rr=- tr_2[(h+/2 ) ^-1_rr- ( h-/2 ) ^-1_rr ] [ relation1 ] which diverges for a vanishing symmetry breaking term @xmath14 due to long - range correlations .
for the correlation function ( [ corr1a ] ) a similar but more elaborate analysis yields a diffusion propagator @xcite , whose fourier components read _
[ diff_prop ] this agrees with ( [ relation1 ] ) for @xmath42 when we use the self - consistent approximation @xmath43 of app .
[ sect : self - consist ] .
this is not an accident but a consequence of the fact that the self - consistent approximation represents the saddle point of the corresponding functional integral @xcite .
the prefactor of the @xmath44 term is the diffusion coefficient d = _ k tr_2 ( ) . [ 3ddiff ] thus , the dc limit @xmath45 of the conductivity formula in ( [ conda1 ] ) and the correlation function ( [ diff_prop ] ) reproduce the einstein relation = d , [ einstein_r ] which gives with the right - hand side of eq .
( [ 3ddiff ] ) for 3d weyl fermions the integral ( , e_f)=2 ^ 2_0^ [ cond2 ] with momentum cut - off @xmath46 .
thus , the conductivity depends on the disorder strength @xmath11 only through the scattering rate @xmath4 .
a diffusion propagator can also be calculated from the bse , as demonstrated recently for 3d weyl fermions @xcite
. however , the derivation of the propagator ( [ diff_prop ] ) from the symmetry ( [ symmetry ] ) has the advantage that it is simpler and that we obtain the diffusion coefficient @xmath47 in ( [ 3ddiff ] ) directly as a quadratic form of green s functions .
in the following we present and discuss the results which are obtained from the conductivity @xmath48 in eq .
( [ cond2 ] ) .
this expression is subtle in the limit of a vanishing scattering rate @xmath4 , since the latter appears as @xmath49 in front of an integral that diverges for @xmath50 .
this makes the conductivity very sensitive to the order of the limits @xmath51 and @xmath50 in the case when the scattering rate vanishes at the node .
since the conductivity depends on @xmath4 and @xmath52 separately , we consider first properties exactly at the node @xmath3 , where results are simple , and then the more complex results when @xmath4 depends on @xmath52 .
for the second part we employ the scba to determine the function @xmath53 and calculate the corresponding conductivity . at the node ( @xmath3 ) the dc conductivity in eq .
( [ cond2 ] ) is reduced to the expression = 2 ^ 2_0^ = ( =/ ) , [ cond000 ] which becomes for @xmath54 ~ .
[ cond001 ] in contrast to the 2d case , where @xmath55 , the 3d case gives a linearly increasing behavior with respect to the scattering rate .
this result was derived directly ( i.e. , without using eq .
( [ cond2 ] ) ) by fradkin some time ago @xcite . with a disorder dependent scattering rate
he also obtained a transition for a critical disorder strength @xmath56 , where the conductivity vanishes for @xmath57 and increases linearly for @xmath58 .
the linear behavior indicates an unconventional transport because in the classical boltzmann approach for one - band metals the conductivity _ decreases _ with increasing scattering rate : @xmath59 ( @xmath60 : electron density , @xmath61 : electron mass ) @xcite .
this remains true when we include the band structure of the weyl fermions in the boltzmann approach : @xmath62 is nonzero at the node for any scattering rate and even diverges with vanishing disorder as @xcite _
b= , [ cond_b ] where @xmath11 is related to the density of impurities @xmath63 and the impurity potential @xmath64 by the equation @xmath65 @xcite .
the disagreement between the expressions in ( [ cond000 ] ) and ( [ cond_b ] ) can be explained by interband scattering , caused by particle - hole creation processes , which has been ignored in the boltzmann approach . on the other hand ,
the increasing behavior of ( [ cond000 ] ) for small @xmath4 turns into a decreasing behavior for larger @xmath4 , as one can see in fig .
[ fig : cond1 ] , indicating a crossover from quantum transport for weak scattering to conventional boltzmann transport for stronger scattering . in this context it is also interesting to study the finite - size effects of the conductivity by considering a cubic system of finite length @xmath66 .
the @xmath67function @xmath68 describes the finite - size scaling of the conductivity .
it can be calculated from eq .
( [ cond2 ] ) by replacing the lower integration boundary with @xmath69 , which assumes non - periodic boundary conditions .
then @xmath70 in eq . ( [ cond000 ] ) is a monotonically increasing function for increasing size @xmath66 with ( , l)~^*- for @xmath71 , where @xmath72 is the expression ( [ cond001 ] ) .
the corresponding @xmath67function reads ~2[1-(,l)/^ * ] , [ betaf ] which vanishes at the @xmath4-dependent fixed point @xmath72 .
this differs from the 2d case only by different fixed points @xmath72 , where in 2d it is a universal constant @xmath73 @xcite and in 3d it is the @xmath4-dependent expression ( [ cond001 ] ) .
it should be noticed that @xmath72 is not a critical point because it is an attractive fixed point .
but since for 3d weyl fermions @xmath72 depends on the scattering rate @xmath4 , we have a line of fixed points for @xmath74 .
thus , the endpoint @xmath75 for @xmath76 has the feature of a critical point because any change of @xmath4 drives us away from this endpoint , as illustrated in fig .
[ fig : betaf ] .
it indicates a transition from an insulator ( @xmath76 ) to a metal ( @xmath77 ) .
the transition is driven by increasing disorder , since the scattering rate is a monotonic function of the disorder strength @xmath11 .
@xmath4 is also the order parameter for spontaneous symmetry breaking ( [ op ] ) , which can be calculated from eq .
( [ scatt_rate ] ) within scba . from the solution of the self - consistent equation ( [ spe0 ] ) we get for @xmath78 and @xmath79 the linear behavior ~(-1)(-1 ) ( = g/2 ^ 2 ) with the step function @xmath80 .
for @xmath81 we have no spontaneous symmetry breaking .
thus , @xmath4 as well as the dc conductivity vanish strictly .
when we approach @xmath82 from above there is linear behavior for the scattering rate , which agrees with the numerical calculation of kobayashi et al .
@xcite . at the critical point itself
we obtain from the einstein relation ( [ einstein_r ] ) a finite diffusion coefficient @xmath83 the results of the dc conductivity from previous self - consistent studies , based on a combination of scba and a self - consistent solution of the bse @xcite , are summarized and compared with our results of the wsa in table [ table1 ] . for sufficiently large scattering rates
the boltzmann approach , the solution of the bse and the result of the wsa agree reasonably well , reflecting a rather conventional transport .
this indicates that quantum effects , such as particle - hole pair creation , are dominated by impurity scattering . on the other hand , for smaller values of the scattering rate
the conductivity exhibits a larger variety of results : the boltzmann conductivity has a simple @xmath84 behavior , which is also found with the solution of the bse in ref .
@xcite , with a different constant prefactor though .
in contrast , the approximative analytic solution of the bse in ref .
@xcite has a characteristic dip down to zero at a critical @xmath56 and increases for @xmath58 and for @xmath85 : = _ 1|1/g-1/g_c| , _
[ scba_cond ] when we compare this result with the wsa conductivity in eq .
( [ cond000 ] ) it should be noticed that the latter was obtained by sending @xmath51 first and then @xmath50 .
as mentioned at the beginning of this section the value of the conductivity depends on the way we take these two limits .
although nothing has been said in ref .
@xcite about the order of the two limits to get ( [ scba_cond ] ) , we will study in sect .
[ sect : away ] the case when @xmath52 and @xmath4 go to zero simultaneously in eq .
( [ cond2 ] )
. then we obtain a result similar to ( [ scba_cond ] ) .
.the scattering rate @xmath4 and the conductivity at the node calculated with three different methods .
the translation from ref .
@xcite is @xmath65 and the scba coefficient @xmath86 is given in eq .
( [ scba_cond ] ) . [ cols="^,^,^,^,^",options="header " , ] the conductivity as a function of the fermi energy is plotted at fixed scattering rates in fig .
[ fig : cond2 ] . as we increase the scattering rate the effect of the node is washed out and the conductivity becomes flatter .
this is similar to the behavior in fig .
[ fig : cond1 ] . in other words ,
impurity scattering supports transport near the node whereas it suppresses it further away .
thus , we can distinguish a regime close to the node , where the conductivity increases with the scattering rate , and a more conventional regime further away from the node , where the conductivity decreases with the scattering rate , as also described by the boltzmann approach .
so far we have considered the case that @xmath4 and @xmath52 are independent . however , in general the scattering rate depends on @xmath87 and @xmath52 .
for instance , the self - consistent calculation in app . [
sect : self - consist ] , based on the scba or saddle - point approximation , creates a scattering rate in eq .
( [ scatt3 ] ) that depends on the fermi energy : = re .
[ scattering1 ] the behavior of the conductivity in ( [ cond2 ] ) is affected by this result , since the limits @xmath50 and @xmath51 are not independent anymore .
( [ scattering1 ] ) has two typical regimes , namely @xmath88 near the critical point @xmath89 , which leads to @xmath90 , and @xmath91 for @xmath92 , which leads to a non - vanishing conductivity for @xmath51 .
this implies that the conductivity vanishes only for @xmath93 , whereas it nonzero above and below @xmath93 .
thus , there is no insulating phase but only an insulating point for @xmath89 , in agreement with the analytic result of ref .
the behavior of the conductivity as a function of @xmath94 is plotted for different values of the fermi energy in fig .
[ fig : cond3 ] .
it should be noticed , though , that the transport behavior is determined by the @xmath52 dependence of the scattering rate of the special form in eq .
( [ scattering1 ] ) . using
another form of the scattering rate as a function of the fermi energy can lead to a substantially different behavior of the conductivity near the node .
an example was observed in ref .
@xcite within a numerical solution of the sbca and the bse , where the scattering rate is exponentially small for @xmath95 . in this case
a vanishing conductivity was found at the node also for @xmath95 .
we have studied the dc conductivity of 3d weyl fermions in the presence of random scattering .
the relevant parameters in the conductivity ( [ cond2 ] ) are the scattering rate @xmath4 , which is an order parameter for spontaneous chiral symmetry breaking , and the fermi energy @xmath52 .
exactly at the node @xmath3 there is a metal - insulator transition with a diffusive metal for @xmath77 and an insulator for @xmath76 .
the conductivity is linearly increasing with @xmath4 up to a maximal value and decreases for stronger scattering rates , as illustrated in fig .
[ fig : cond1 ] .
this non - monotonic behavior is in contrast to the constant conductivity in the corresponding 2d system .
it reflects the fact that the increased phase space of the 3d weyl fermions suppresses the conductivity for weak scattering but also that stronger scattering implies a screening of the node such that the boltzmann approach eventually becomes applicable .
further away from the node the behavior depicted in fig .
[ fig : cond2 ] agrees qualitatively with that of the 2d system @xcite , which was also obtained in a quasiclassical approach @xcite .
the latter diverges as one approaches the node , which indicates that the full quantum approach is necessary near the node . the critical behavior at the node describes an unconventional phase transition for @xmath96 which is driven by quantum fluctuations : in contrast to a conventional transition the symmetry broken phase with @xmath77 is characterized by robust diffusion whereas the phase with unbroken symmetry ( @xmath76 ) has a subtle behavior in terms of the conductivity because it is very sensitive to the limit @xmath51 .
thus , it is possible that we either have an insulating phase with vanishing conductivity when the scattering rate vanishes slowly with @xmath52 or a metallic phase when the scattering rate vanishes sufficiently fast with @xmath52 . in the case of an scba calculation for @xmath4
there is only a quantum critical point in the transport properties and metallic behavior above and below this critical point .
it can not be ruled out , though , that a different calculation of @xmath4 leads to a different behavior .
thus , our discussion of the delicate limits @xmath97 clarifies some of the contradicting results in the literature about the presence of a metal - insulator transition for 3d weyl fermions @xcite .
moreover , the fact that the conductivity depends only on @xmath4 and @xmath52 allows us to determine @xmath4 independently with other approximations than the scba , and to insert the result into the conductivity ( [ cond2 ] ) .
a possible step in this direction is a correction to the scba @xcite or perturbative renormalization - group approach in @xmath98expansion .
the latter gives @xmath99 @xcite .
this result would lead to a vanishing conductivity below the critical point . in the regime @xmath100 disorder
may also affect physical properties of weyl fermions in another way .
the reason is the existence of non - uniform solutions of the scba with an exponentially small contributions to @xmath4 , similar to lifshitz tails in the density of states of disordered systems @xcite . in the case of 2d weyl fermions
this has been discussed in ref .
@xcite . the problem , however , is always that the self - consistent equation is nonlinear and has many non - uniform solutions . under certain plausible assumptions for the solutions @xmath4 , the existence of exponentially small contributions for
@xmath95 has been discussed for 3d weyl fermions in ref .
the corresponding states might be localized then with no contribution to the conductivity at @xmath101 .
however , for @xmath102 thermally activated electrons may hop between patches of localized states and provide a hopping conductivity @xcite .
whether or not resonant tunneling without spontaneous symmetry breaking can occur in this case is an open question .
0.5 cm acknowledgement : i am grateful to david schmeltzer for an extended discussion of weyl fermions .
the first step is to study spontaneous symmetry breaking of the symmetry ( [ symmetry ] ) by a non - zero scattering rate within scba , following a similar approach as given in ref .
the average one - particle green s function then reads ( h+z)^-1(h+ z+)^-1 , [ scba ] and the self - energy is given by = -g(h++i)^-1 .
the imaginary part of the self - energy @xmath103 is a scattering rate @xmath4 . then the self - energy reads in our case with the momentum cut - off @xmath46 = ( = g/2 ^ 2 , = e_f + ) . [ scba1 ] for small @xmath52 near the node we expand eq .
( [ scba1 ] ) in powers of @xmath104 up to second order to obtain ~-e_f+ + i .
[ scatt3 ] the real part of @xmath103 provides a shift of the fermi energy : ~-e_f + i = -e_f + e^i/4 , where the sign is chosen such that we have a positive scattering rate . at the node @xmath3 the self - consistent eq .
( [ scba1 ] ) reduces to @xmath105 with @xmath106 % , \ \ \ \beta=\beta % \sqrt{\eta^2+\delta^2/4 } \ .\ ] ] there are two solutions , namely @xmath76 and @xmath107 with = , =/ .
[ spe0 ] a nonzero @xmath4 reflects spontaneous symmetry breaking with respect to ( [ symmetry ] ) .
such a solution exists for ( [ spe0 ] ) only at sufficiently large @xmath87 .
moreover , @xmath4 vanishes continuously as we reduce @xmath87 .
a nonzero @xmath4 is proportional to the density of states at the fermi level .
however , even for @xmath76 there can be a nonzero local density of states due to localized energy levels , which are not counted in @xmath4 within the scba . for @xmath108
we obtain the linear behavior @xmath109 | we calculate the transport properties of three - dimensional weyl fermions in a disordered environment .
the resulting conductivity depends only on the fermi energy and the scattering rate .
first we study the conductivity at the spectral node for a fixed scattering rate and obtain a continuous transition from an insulator at weak disorder to a metal at stronger disorder . within the self - consistent born approximation the scattering rate depends on the fermi energy .
then it is crucial that the limits of the conductivity for a vanishing fermi energy and a vanishing scattering rate do not commute . as a result , there is also metallic behavior in the phase with vanishing scattering rate and only a quantum critical point remains as an insulating phase .
the latter turns out to be a critical fixed point in terms of a renormalization - group flow . |
the search for a scientific understanding of _ time _ has a wide scope in physics , ranging from classical mechanics to quantum field theory , from particle mechanics to cosmology , statistical physics and beyond .
the research on time inevitably touches foundational issues and one can even suspect it can not be fully understood since time is so essential to our perception of reality . nevertheless , we can hope to improve our knowledge about time as _ time goes on _ ... instead of to deal with the subtleties of the physical meaning of _ time _ ( for detailed discussion see @xcite and @xcite and references quoted therein ) , here i m devoted to a simpler task : to show that _ it is theoretically possible an elementary quantum system be both isolated and non - time reversal invariant_. this possibility contradicts a common sense among physicists , namely , that isolated systems are ( ever ) time reversal invariant and that irreversibility is just a statistical phenomenon ( coded in the second law of thermodynamics ) . to be more precise ,
consider a quantum system and denote its state ( density operator ) in time @xmath0 by @xmath1 here , i use _ schrdinger s picture_. it is generally accepted that if the system is _ closed _ , then its time evolution is given by von neumann s equation with some time - dependent hamiltonian @xmath2:@xmath3
\label{equation_von - neumann}\]]accordingly , the system is said to be _ isolated _ when it is closed and its hamiltonian is constant .
so , according with this view _
open quantum systems _ are exactly those whose time evolution is not governed by ( time - dependent ) von neumann equation ( [ equation_von - neumann ] ) . in books and papers those conceptions may be used implicitly ; for example , authors can relate the semigroup structure of time evolution to ( markovian ) open systems _
although the above definitions for the concepts of _ isolated _ , _ closed _ and _ open _ can be mathematically perfect , they lack direct correspondence with those physical meanings we give to them . indeed , _
when we say a physical system is closed we mean that it does not exchange matter with other systems _ ; also , _ we say a physical system is isolated when it does not interact with any other system_. in this sense , one realizes that the concept of isolated system must be related to the principle of inertia , meaning that _ after arbitrary preparation , an isolated system has its energy and momenta remaining constant_. assuming the operationalist point of view , i define the concept of isolated quantum systems in terms of the expectation values of its energy - momentum tensor operator : a quantum system is isolated when the expectation values of its energy - momentum tensor operator with respect to any time - dependent state is constant .
i argue that this definition is not trivially irrelevant , even though it is a truism that there is no isolated system ( within universe ) at least , they could not be observed even if they existed .
nevertheless , many systems can be regarded isolated in practice , at least for a short interval of time , and this is sufficient to the concept be ( eventually ) useful . beyond the almost triviality of give a definition for the concept of isolated quantum system , i m going to highlight a special class of them , namely , _
isolated quantum systems that are non - time reversal invariant _ .. ] quantum systems of this sort i call _ dispersive quantum systems _ :
a dispersive quantum system is a quantum system which is isolated and non - time reversal invariant .
perhaps , a distinguished example of dispersive quantum system is the whole universe .
however , it would be remarkable if we could discover a dispersive quantum system being also elementary , because its non - time reversal invariance would be fundamental in the sense of not being an emergent ( statistical ) property .
the _ dispersive qubit _ ( section [ sec:2 ] ) and its application to _ neutrino oscillation _
( section [ sec:3 ] ) are intended to instantiate that !
the structure of the paper is simple .
section [ sec:1 ] starts with general quantum mechanics focusing on time evolution and a definition of irreversibility , followed by a review of definitions and results about markovian quantum systems , dynamical semigroups and completely positiveness .
after , it is obtained the equation for the liouvillian s non - hamiltonian part that characterizes those completely positive markovian quantum systems that are isolated . in section [ sec:2 ] , it is presented an explicit model exhibiting the mentioned features . in section [ sec:3 ]
i apply previous developments to describe neutrino oscillation , with the introduction of a new parameter i call _ dispersive parameter_. in the final section [ sec:4 ] , some remarks are discussed .
i denote by @xmath4 the set of natural numbers including zero and set @xmath5 . for a hilbert space @xmath6
, i denote by @xmath7 the set of densely defined operators in @xmath6 , by @xmath8 the space of bounded operators in @xmath6 and by @xmath9 the space of bounded trace class operators in @xmath6 .
finally , i use _ natural units _
, so @xmath10 .
in quantum mechanics , _ physical systems _ are described in terms of _ observables _ and _ states _ with the use of a _ separable hilbert space _
@xmath6 : for a system @xmath11 without _ superselection sectors _ , observables are identified with _ ( densely defined ) self - adjoint operators _ in @xmath6 and states are identified with _ density operators _ , _ i.e. _ , positive trace class _ _ _ _ operators with trace one in @xmath6 .
i denote the set of observables by @xmath12 and the set of states by @xmath13 .
the fundamental postulate of quantum mechanics states that _ the expectation value of an observable _ @xmath14 _ when the system is in the state _
_ is given by _
_ is bounded what is automatic in the finite dimensional case , which is the one we are interested .
] @xmath16 [ [ time - evolution . ]
] time evolution .
+ + + + + + + + + + + + + + + consider the system was prepared at time @xmath17 and evolves without interference since then . in schrdinger
s picture , system s _ time evolution _ from an instant @xmath18 to a later instant @xmath19 must be described by a map from the space of states to itself,@xmath20i call this _ time evolution map _ and its physical interpretation is simple : _ given two instants _
@xmath21 _ _ , if _ _ @xmath22 _
_ is the system s state at instant _ _ @xmath23 _ _ , then _ _
@xmath24 _ _ is the system s state at the instant _ _ @xmath25 .
it is natural to assume the following property i call _ factorization _ : @xmath26 a physical system is _ _
time reversal invariant _
_ when there is a map @xmath27 satisfyingi ) idempotence:@xmath28ii ) time - reversing equation:@xmath29i call @xmath30 the _ time reversing map_. by physical reasons , one may require additional properties on @xmath30 , such as antilinearity .
its physical meaning is natural : @xmath30 defines a correspondence among the system s states which reverts the direction of time evolution : @xmath31this equation is the quantum analog of what can in principle be done to a classical ( non - magnetic ) system : if the velocities of all particles of a classical mechanical system are reversed ( what can be represented by a map in phase space ) , then this system would behave _ as if _ it was running backwards in time !
a system is _ non - time reversal invariant _ when it is not time reversal invariant .
time reversal invariance holds for some physical systems , but one does not have any reason to assume this property must hold for all closed physical systems .
( actually , i m going to describe a class of _ isolated non - time reversal invariant quantum systems . _ ) if the system is time reversal invariant , equations ( [ def_reversible_idempotence ] ) and ( [ def_reversible_time - reversing - equation ] ) imply @xmath32 is invertible , for all @xmath21 so , _ time - evolution map s invertibility is a necessary condition for the system to be invertible_:@xmath33 in this case , one can define the _ extended time evolution map _ in @xmath34:@xmath35the extended time evolution map satisfies _ extended factorization_:@xmath36proof : i have to analyze the six possible orderings for instants @xmath37 ; the case @xmath38 follows directly from factorization ( [ form_time - evolution_factorization ] ) ; here , i verify explicitly the case @xmath39 only : @xmath40 for a time reversal invariant system , there is no intrinsic distinction between past and future , since any pair of states related by time evolution map are equally related by the time evolution map s inverse . however , an observer assigns a time arrow to an non - time reversal invariant system through preparation time @xmath17 : before @xmath17 the system interacts with environment and after @xmath17 the system evolves without external interaction @xcite , @xcite . [
[ markovian - systems . ] ] markovian systems .
+ + + + + + + + + + + + + + + + + + a system is said to be _ markovian _ when its time evolution map depends of the time interval between instants only:@xmath41 in this case , the system s time evolution is given by the _
( one - parameter ) quantum dynamical semigroup _ in @xmath42 with domain @xmath43:@xmath44directly from factorization ( [ form_time - evolution_factorization ] ) , it follows the _ semigroup property _ : @xmath45proof : @xmath46 .
physically , time evolution is markovian when it does not depend on the past ( or future ) history of the quantum system and its environment there is no memory about the way it reaches its present state . as far as i know
, markovian property holds for closed systems and , as it was demonstrated by davies in @xcite , it holds also for open systems under special conditions .
if the system is markovian and time reversal invariant , its quantum dynamical semigroup can be extended to a one - parameter group in @xmath47 , with its group property being a consequence of the semigroup property ( [ def_quantum - dynamical - semigroup]):@xmath48 [ [ quantum - dynamical - semigroups - generator . ] ] quantum dynamical semigroup s generator .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + for technical reasons ( to appear in subsection [ subsec:1.2 ] ) , from now on assume the quantum dynamical semigroup can be extended to a _
_ semigroup of positive trace preserving superoperators in the space _ _ @xmath49__of bounded trace class operators in _ _
@xmath6,@xmath50semigroup property:@xmath51positivity:@xmath52trace preserving property:@xmath53also , one assume the quantum dynamical semigroup is @xmath54-continuous from above:@xmath55 the space @xmath49 of trace class operators in @xmath6 is a banach space w.r.t .
the trace - norm @xmath56 , @xmath57moreover , @xmath58 ( trace class operators are bounded ) and the space of finite - rank operators in @xmath6 is @xmath59-dense in @xmath49 . for details ,
see @xcite . denotes the space of _ bounded operators _ in @xmath6 , but here it denotes space of _ densely defined operators _ in @xmath6 . ]
the above conditions are sufficient to guarantee that the quantum dynamical semigroup has an infinitesimal generator : [ teorema_generator ] for the quantum dynamical semigroup ( def_quantum - dynamical - semigroup ) under conditions ( def_time - evolution - map_semigroup ) , ( [ def_time - evolution - map_positivity ] ) , ( [ def_time - evolution - map_trace ] ) and ( def_time - evolution - map_continuity ) , it holds : i ) there exists an operator @xmath60 with @xmath59-dense domain @xmath61 such that@xmath62ii ) the quantum dynamical semigroup @xmath63 is a limit of exponentials related to @xmath60:@xmath64where@xmath65iii ) @xmath60 is bounded if and only if @xmath66 . in this case , @xmath63 is the exponential of @xmath60:@xmath67further , in this situation it holds@xmath68 in quantum mechanics , _ the generator _ @xmath60 _ _ _ _ of the system s quantum dynamical semigroup is called the system s _ liouvillian superoperator_. for a proof , see @xcite or @xcite . the limit ( [ def_quantum - dynamical - semigroup - generator ] ) combined with the fact that the operator norm and trace - norm satisfy @xcite@xmath69implies the _ system s equation of motion _ , called _ markovian quantum master equation _
@xcite:@xmath70 from the mathematical point of view , theorem ( [ teorema_generator ] ) generalizes
_ stone s theorem _ , which establishes the existence and uniqueness of the infinitesimal generator for an one - parameter group of unitary operators in hilbert spaces for details see ru . from the physical point of view , theorem ( [ teorema_generator ] ) generalizes for completely positive markovian quantum systems the von neumann equation for time - dependent states ( [ equation_von - neumann ] ) .
the hille - yosida theorem characterizes those operators in @xmath6 which are generators of some dynamical semigroup ru , @xcite : a densely defined operator @xmath60 in a banach space @xmath71 is the generator of some trace - norm continuous one - parameter semigroup of bounded operator in @xmath71 if and only if there are constants @xmath72 and @xmath73 such that@xmath74where @xmath75 is the operator norm in @xmath76 .
it is natural to ask about the general form of the liouvillian of a markovian quantum system .
fortunately , this question has an answer for the special class of quantum dynamical semigroups which are _
completely positive_. this is the point of the following subsection .
_ completely positiveness _
( see appendix sec : a ) is a property of _ quantum operations _
, a concept of the theory of generalized measurements @xcite , @xcite .
this is a special property of a large class of quantum systems , including subsystems of systems whose time evolution is unitary and do satisfy some special conditions @xcite pe .
such systems have a wide range of applications , ranging from quantum information theory @xcite to astrophysics @xcite @xcite ( to give only two examples ) .
if the liouvillian of a completely positive quantum dynamical semigroup is bounded , then it has a standard form , according to the following theorems due to lindblad @xcite ( bounded liouvillian ) and also gorini - kossakowski - sudarshan @xcite ( finite dimensional hilbert space ) .
although the finite dimensional case can be viewed as a special case of the general theorem due to lindblad , it can be better scrutinized ; so , i will state them separately . as far as i know
, those theorems have not been extended to the case of unbounded liouvillians , a typical situation in physics .
[ teorema_liouvillian - bounded ] let @xmath60 be the liouvillian of the quantum dynamical semigroup @xmath63 of a completely positive markovian quantum system @xmath11 with hilbert space @xmath6.if @xmath60 is bounded ( equivalently , @xmath66 ) , then there exist a bounded self - adjoint operator @xmath77 in @xmath6 and a countable family of bounded operators @xmath78 in @xmath6 satisfying @xmath79such that@xmath80 + \frac{1}{2}\sum_{j}\left ( v_{j}\sigma v_{j}^{\ast } -\frac{1}{2}\left ( v_{j}^{\ast } v_{j}\sigma + \sigma v_{j}^{\ast }
v_{j}\right ) \right ) \ \ , \ \forall \sigma \in \mathcal{t}% \left ( \mathcal{h}\right ) \label{form_liouvillian - infinite}\ ] ] [ teorema_liouvillian - finite ] let @xmath60 be the liouvillian of the quantum dynamical semigroup @xmath63 of a completely positive markovian quantum system @xmath11 with hilbert space @xmath6 having finite dimension @xmath81 . in this case @xmath82 has dimension @xmath83.let @xmath84 be a complete set in @xmath85 with @xmath86 which is orthonormal w.r.t . trace , _
i.e._,@xmath87then , there exist a self - adjoint operator @xmath77 in @xmath6 and a unique positive complex matrix @xmath88 such that@xmath80 + % \sum_{i , j=1}^{n^{2}-1}a_{ij}\left ( f_{i}\sigma f_{j}^{\ast } -\frac{1}{2}% \left ( f_{j}^{\ast } f_{i}\sigma + \sigma f_{j}^{\ast } f_{i}\right ) \right ) \
\ , \ \forall \sigma \in \mathcal{t}\left ( \mathcal{h}\right ) \label{form_liouvillian - finite}\]]@xmath89further , the matrix @xmath90 is unique for each family @xmath91 and the operator @xmath77 is unique if it is required @xmath92 .
the conditions on @xmath93 in theorem teorema_liouvillian - finite imply that all operators different from @xmath94 are traceless:]@xmath95 expression ( [ form_liouvillian - finite ] ) of theorem teorema_liouvillian - finite can be transformed in the expression ( form_liouvillian - infinite ) of theorem [ teorema_liouvillian - bounded ] by a diagonalization of the positive - semidefinite matrix @xmath90 , as done in @xcite and @xcite .
the operator @xmath77 is called _ hamiltonian _ and the term @xmath96 $ ] is called the _ hamiltonian part _ of the system s liouvillian.the _ dissipator superoperator _ is defined by ( in agreement with @xcite):@xmath97\]]in the finite dimensional case,@xmath98people call @xmath90 the system s _ kossakowski matrix _ and note that it has dimension @xmath99 when the hilbert space has finite dimension @xmath100 . in this subsection
i analyze hamiltonian expectation value values related to time - dependent states for completely positive markovian quantum systems .
our considerations are restricted to the finite dimensional case , so @xmath101 and the system s liouvillian is given by lindblad s form ( [ form_liouvillian - finite ] ) .
consider a quantum system @xmath11 with hamiltonian @xmath77 and dissipator superoperator having the form ( def_dissipator - superoperator_finite - dimensional ) .
the system s dissipation operator is defined by the action of dissipator superoperator on the hamiltonian:@xmath102 [ teorema_hamiltonian - time - derivative ] let @xmath1 be a time - dependent state of the system @xmath11
. then the related time - dependent hamiltonian s expectation value value satisfies the equation@xmath103 the markovian quantum master equation ( equation_markovian - master - equation ) with liouvillian having the form ( form_liouvillian - finite ) reads@xmath104 + % \mathcal{d}\left ( \rho \left ( t\right ) \right)\]]since derivation commutes with trace , it holds@xmath105 \right\ } + tr\left\ { h\mathcal{d}\left ( \rho \left ( t\right ) \right ) \right\}\]]using that trace is invariant under cyclic permutation of factors , it follows@xmath106 \right\ } = tr\left\ { hh\rho \left ( t\right ) -h\rho \left ( t\right ) h\right\ } = tr\left\ { hh\rho \left ( t\right ) -hh\rho \left ( t\right ) \right\ } = tr\left\ { 0\right\ } = 0\]]therefore@xmath107using ( [ form_liouvillian - finite ] ) to expand the expression inside trace , it follows@xmath108 [ teorema_disclosedness ] the hamiltonian s expectation value value is constant for whatever be the system s time - dependent state if and only if the system s dissipation operator is identically zero . from proposition [ teorema_hamiltonian - time - derivative ]
, the hamiltonian s expectation value value is constant whatever be the system s time - dependent state if and only if @xmath109 ; this is equivalent to @xmath110 and this is equivalent to @xmath111 .
the above result motivates the following definition : a markovian dispersive quantum system is a markovian quantum system whose dissipation operator is identically zero . in the finite dimensional case , corollary ( [ teorema_disclosedness ] )
characterizes dispersive quantum systems within the class of completely positive markovian quantum systems by a linear equation for its dissipation matrix:@xmath112just to emphasize , this equation is a necessary and sufficient condition for the system s hamiltonian to belong to the kernel of its dissipator superoperator . in the next section ,
i present a simple example of markovian dispersive quantum system , showing that the class of such systems is non empty .
the theoretical discussion of previous section does not follow the usual reasoning used in the modeling of physical systems .
instead , it just set the mathematical framework for modeling of quantum systems .
in general , one associates to a quantum system its characteristic _ _ hamiltonian _ _ and , in each specific situation that system is studied , its equation of motion is constructed taking into account the interaction with other systems and all relevant contributions to the system s dynamics due to its environment .
this procedure is used when we want to model the system s behavior under specific conditions , or when we want to specify conditions for the system to behave according to some prescription . for completely positive markovian quantum systems
, one has to propose a liouvillian and , to verify if that can be the generator of the system s quantum dynamical semigroup , perform one of the two following procedures : i ) check the hille - yosida theorem s condition ( [ equation_hille - yosida]);ii ) solve the system s equation of motion ( equation_markovian - master - equation ) , built the dynamical semigroup and compute its generator . here
i define _ dispersive qubit _ as a two - level quantum system having a special dynamics .
the hilbert space of this system is @xmath113 with its usual structure of vector space and inner product . according with the general framework , _ observables _
are defined by _ self - adjoint operators _ and _ states _ are defined by _ density operators _ in @xmath113 .
the system s equation of motion is the markovian master equation ( [ equation_markovian - master - equation ] ) with liouvillian having form ( [ form_liouvillian - finite ] ) in terms of a _
hamiltonian _ @xmath77 and a _ dissipator superoperator _ @xmath114:@xmath104 + % \mathcal{d}\left ( \rho \right ) \label{model_equation - motion}\]]the hamiltonian has non - degenerated spectrum , with eigenvalues @xmath115 . using dirac s notation , the corresponding normalized eigenvectors of the hamiltonian are written as @xmath116 and @xmath117 ; using the ordered basis @xmath118 , the space of linear operators @xmath119 is identified with thespace of @xmath120 complex matrices @xmath121 ; in particular , the hamiltonian is given by the following diagonal matrix@xmath122i denote the identity matrix of @xmath121 and _ pauli matrices _
by@xmath123pauli matrices with identity matrix divided by @xmath124 form a basis for @xmath125 which is orthonormal with respect to trace:@xmath126therefore , one can use pauli matrices to describe the system s liouvillian and the dissipator superoperator , as prescribed by theorem teorema_liouvillian - finite and defined by ( def_dissipator - superoperator_finite - dimensional):@xmath127 + \mathcal{d}\left ( \rho \right)\]]where@xmath128[model_hamiltoniano - dissipator ] and @xmath90 is the system s kossakowski matrix .
defining@xmath129the system s _ dissipation operator _
( def_energy - dissipator - operator ) is given by:@xmath130 finally , to complete the definition of _ dispersive qubit system _ i require that the dissipation operator is null : @xmath131this condition is an equation for kossakowski matrix .
below , i deal with a special case . from now on
, i will deal with the following special case of dispersive qubit . for a fixed @xmath132 , called here _ dispersive parameter _ , i define the kossakowski matrix : @xmath133finally , the explicit expression for the system s liouvillian is@xmath134 in matricial terms:@xmath135 [ [ time - evolution.-1 ] ] time evolution .
+ + + + + + + + + + + + + + + one can solve the equation of motion ( [ model_equation - motion ] ) and analyze the time evolution of states to get further details about what happens to the system as time goes on .
writing the density matrix for a generic time - dependent state@xmath136the system s equation of motion ( [ model_equation - motion ] ) becomes @xmath137 the solution of this equation is@xmath138under conditions which guarantee self - adjointness , positivity and trace one for @xmath120 complex matrices , then the general form of the system s time - dependent states is:@xmath139where if and only if they hold for @xmath140.]@xmath141 the system s quantum dynamical semigroup follows from ( model_time - dependent - states ) : @xmath142to verify the consistence of the model , we mention that @xmath143 is actually a continuous semigroup ( in the sense of ( def_time - evolution - map_continuity ) ) and its generator is @xmath60:@xmath144 [ [ irreversibility . ] ] irreversibility .
+ + + + + + + + + + + + + + + + the condition of positiveness will be violated in ( model_time - dependent - states ) when @xmath145 and @xmath146 if one extrapolates this solution for times before@xmath147this fact suggests the model is non - time reversal invariant , since the quantum dynamical semigroup can not be naturally extended to a one parameter group ! actually , the dispersive qubit is non - time reversal invariant .
it is sufficient to show that there is no idempotent solution to the time - reversing equation ( [ def_reversible_time - reversing - equation ] ) .
so , assume that equation ( [ def_reversible_time - reversing - equation ] ) has some solution @xmath148 and write@xmath149for @xmath150 , denote @xmath151 as given by ( model_stationary - state ) ; then , from ( def_reversible_time - reversing - equation ) it follows@xmath152therefore @xmath153 and this means @xmath154 can not be idempotent .
this completes the proof .
the dispersive qubit has pure states which evolve to mixed states .
more precisely , an initial pure state remains pure if and only if it is one of the hamiltonian s eigenstates .
the density matrix for a pure state has the form@xmath155i note that @xmath156 and this condition is necessary for any density matrix representing a pure state.the state @xmath157 is one of the two eigenstates of hamiltonian if and only if @xmath158 or @xmath159 ; from ( [ model_time - dependent - states ] ) , those states are pure and from ( [ model_time - dependent - states ] ) remain constant ( and pure).however , any initially pure state @xmath160 with @xmath161 and @xmath162 evolves to impure states , since@xmath163 the system s stationary states are defined by the limit@xmath164they are explicitly @xmath165 the von neumann entropy @xcite of the state ( model_time - dependent - states ) is given by:@xmath166 & = & -tr\left\ { \rho \left ( t\right ) \log
\rho \left ( t\right ) \right\ } \\ & = & \log 2-a\log \left ( 1+\sqrt{1 - 4\left ( a\left ( 1-a\right ) -\left\vert b\right\vert ^{2}e^{-2\lambda t}\right ) } \right ) + \\ & & -\left ( 1-a\right ) \log \left ( 1-\sqrt{1 - 4\left ( a\left ( 1-a\right ) -\left\vert b\right\vert ^{2}e^{-2\lambda t}\right ) } \right)\end{aligned}\]]as
one can verify , @xmath167 $ ] increases with time when @xmath168 and@xmath169 = \log 2-a\log \left ( 1+\sqrt{1 - 4a\left ( 1-a\right ) } \right ) -\left ( 1-a\right ) \log \left ( 1-\sqrt{1 - 4a\left ( 1-a\right ) } \right)\]]for the special case @xmath170 , this limit reaches the entropy s maximum value @xcite:@xmath169 = \log 2\ ] ] let @xmath71 be an observable of the dispersive qubit , _
i.e. _ , a self - adjoint operator in @xmath113 .
it has two eigenvalues ( which can be equal ) @xmath171 and corresponding orthonormal eigenvectors @xmath172 and @xmath173:@xmath174from above conditions and with an eventual redefinition of @xmath172 and @xmath173 , it follows that there exists @xmath175 $ ] such that@xmath176 in the basis @xmath177 , we have@xmath178and@xmath179 i remark that the observable @xmath71 is not compatible with the hamiltonian @xmath77 when @xmath180 and @xmath181 , because they can not be simultaneously diagonalized . [
[ time - evolution - of - eigenstates - of - x . ] ] time evolution of eigenstates of @xmath71 .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + if the initial state of the system is that corresponding to the eigenstate @xmath182 of @xmath71@xmath183then , according with ( [ model_time - dependent - states_initial ] ) the time - dependent state is given by@xmath184if @xmath180 and @xmath181 , than the state @xmath185 is a mixture for all @xmath186 , since it can not be put in the form ( [ model_pure - state ] ) . in particular
, the system evolves from the pure state @xmath187 to the mixture@xmath188 [ [ expectation - values - of - x . ] ] expectation values of @xmath71 .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the expectation value of @xmath71 w.r.t .
a generic time - dependent state ( model_time - dependent - states ) is @xmath189{l } a\left [ x_{1}\cos ^{2}\theta + x_{2}\sin ^{2}\theta \right ] + be^{-\left ( \lambda + i\delta\right ) t}\left [ \left ( x_{1}-x_{2}\right ) \sin
\theta \cos \theta \right ] + \\ + \bar{b}e^{-\left ( \lambda -i\delta\right ) t}\left ( x_{1}-x_{2}\right ) \sin
\theta \cos \theta + \left ( 1-a\right ) \left [ x_{1}\sin ^{2}\theta + x_{2}\cos ^{2}\theta \right ] \end{array}% \right .
\label{model_observable - expectation - value}\]]in particular , the stationary expectation value of @xmath71 is@xmath190combining the above formulas , we get the expectation value of @xmath71 w.r.t . the time - dependent state which started as the @xmath71 s vector - state @xmath172:@xmath191and@xmath192 [ [ probabilities - of - transition - and - of - surviving . ] ] probabilities of transition and of surviving .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the transition probability from the @xmath71 s vector - state @xmath172 to the @xmath71 s vector - state @xmath193 after a time @xmath194 is given by the expectation value of the projection operator @xmath195:@xmath196 explicitly : @xmath197 \cos ^{2}\theta
\sin ^{2}\theta \end{aligned}\ ] ] the surviving probability of the @xmath71 s vector - state @xmath172 after a time @xmath194 is given by the expectation value of the projection operator @xmath198:@xmath199explicitly : @xmath200one can easily verify that@xmath201 for later use , i rewrite above formulas:@xmath202 \sin ^{2}\left ( 2\theta \right ) \label{model_observable_transition - probability}\]]@xmath203 \sin ^{2}\left ( 2\theta \right ) \label{model_observable_surviving - probability}\ ] ] i remark the role of the dispersive parameter @xmath204 of dispersive qubit : it changes the initially _ time - dependent superposition _ of the states of @xmath71 to a final ( assymptotic ) _ time - indepented mixture _ of them ! to ilustrate the time evolution of the special dispersive qubit , i plot in the fig.[fig : graph-0 ] and fig.[fig : graph-1 ] the graphs of the time - dependent transition and surviving probabilities corresponding to @xmath205 and @xmath206 for @xmath158 and @xmath207 . _ is there any application of the previous concepts and developments ? maybe ... _ the following proposition shows that the choice ( model_liouvillian - matrix ) for the kossakowski matrix can not be otherwise : the dispersive qubit s dissipator operator ( model_dissipator - superoperator ) is zero if and only if its dissipation matrix @xmath90 satisfies @xmath208 we have to find all solutions to the equation for kossakowski matrix s coefficients@xmath209subject to the conditions which guarantee self - adjointness@xmath210and one of the following two sets of conditions which guarantee positive - semi definiteness for @xmath211 matrices:@xmath212for @xmath213 , the matrix @xmath214 is positive since it obviously self - adjoint and has only non - negative eigenvalues ( namely , @xmath215 and @xmath216 ) satisfies conditions ( [ model_eq_sa ] ) and ( [ model_eq_positivity]-@xmath217 ) . ] ; by direct verification we see it satisfies equation ( model_eq_dispersion):@xmath218now , it is very tedious to write down the calculations to get all solutions for the above matrix equation ( [ model_eq_dispersion ] ) subject to ( model_eq_sa ) with one of the above two positivity conditions ( [ model_eq_positivity ] ) ; besides , such calculations are not directly relevant to the purposes of this paper ; so i omit those here .
neutrinos are neutral leptons which occur in one out of three flavors ( related to the others three charged leptons : electron , muon and tauon ) .
they have very tiny masses and interact extremely feebly , being sensible to weak interaction and gravity only .
those characteristics mean that neutrinos are quasi free particles or more precisely , they propagate almost unperturbed during the ( eventually large ) time lapse between their production and detection .
so , it is natural to pay attention in neutrinos in the search to find a system combining both isolatedness and non - time reversal invariance .
the phenomenology of neutrinos is not completely theoretically understood @xcite . actually , the standard model of particle physics and what must extend / replace it are at stake @xcite .
specifically , the phenomenon of _ neutrino oscillation _ ( defined as the dynamic change of flavor ) is a compelling evidence that those particles have masses different from zero , an explicit contradiction with the standard model @xcite , @xcite , @xcite,@xcite , @xcite . the theoretical mechanism explaining neutrino oscillation was first devised by pontecorvo @xcite and @xcite around the 1960s ;
the first experimental evidence of neutrino oscillation was obtained in the super - kamiokande experiment in 1998 @xcite , and since them several experiments have been realized around the world to measure the parameters associated with neutrinos @xcite , @xcite , @xcite , @xcite .
further , there are empirical evidences @xcite as well as theoretical reasons @xcite for the existence of ( at least ) one more neutrino flavor ( besides the three standard flavors ) , called _ sterile neutrino_. presumably , sterile neutrino is a hight - handed particle which mix itself with the other neutrino species and interacts only through gravity what makes it be a very ghostly particle . from the original proposal to understand unexpected data concerning neutrino oscillation ,
some speculate that sterile neutrino can also explain the large disparity of leptons masses as well as be the reason behind the matter - antimatter asymmetry and , further , that it can be the main missing ingredient of the universe ( as the predominant component of dark matter ) @xcite , @xcite .
the possibility of new physics beyond standard model is enlarged by unusual ideas to describe neutrino dynamics and mixing of flavors : non - standard interactions @xcite and non - unitary time evolution of flavor - states @xcite , @xcite .
as long as there still are deep open questions about neutrinos , i think it is opportune investigate the possibility that the dynamics of neutrinos can be `` dispersive '' , i.e. , intrinsically non - time reversal invariant more precisely : that the time evolution of neutrinos is non - time reversal invariant even when they propagate isolately ( in vacuum ) .
specifically , i describe below the neutrino oscillation between two flavors and @xmath219 is relevant ] throughout a phenomenological approach similar to the original ones @xcite @xcite , called _ theory of massive and mixed neutrino _ in the monograph @xcite my basic reference . ) . ] in this theory , neutrino flavor states are superpositions of eigenstates of the relativistic mass operator ( * ? ? ?
* chapter 4 ) ; in the quantum mechanical approximation , the theory reduces to the case described in the subsection _ [ section - model - observables]_. [ [ dispersive - theory - of - massive - and - mixed - neutrino . ] ] dispersive theory of massive and mixed neutrino .
+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + the oscillation between two neutrino flavors is described by a two - level quantum system , identical to the _ dispersive quibit _ defined in the previous section .
the flavor states , denoted by @xmath220 and @xmath221 , are superpositions of the hamiltonian s eigenvectors , denoted by @xmath222 and @xmath223 through relation ( [ model_mixing ] ) , where @xmath224 is called the _ mixing angle _ @xcite . here
, i assume the dynamics is given by lindblad s equation ( [ model_equation - motion ] ) with hamiltonian and dissipator superoperator ( [ model_hamiltoniano - dissipator ] ) and kossakowski matrix ( [ model_dissipation - matrix ] ) with a dispersive parameter @xmath204 ( [ model_liouvillian]):@xmath225where ( as in subsection _ [ section - model - observables]_)@xmath226 since neutrinos are produced with speed near the light velocity ( @xmath227 ) , their energies must be given by the relativistic formula which combines mass and momentum @xcite : @xmath228 i assume the following conditions @xcite
: \i ) the masses of neutrinos are small compared to the momenta they are produced:@xmath229 ii ) the momenta of neutrinos are approximately equal:@xmath230 with above assumptions , it holds the approximations@xmath231and@xmath232 the time lapse between the production and the detection of the neutrinos is approximately given in terms of the distance @xmath60 between the source and the detector by ( @xmath227)@xmath233 finally , substituting ( [ neutrino_energy - approx ] ) and ( [ neutrino_time ] ) in the formulas for the probability of transition ( model_observable_transition - probability ) and surviving ( model_observable_surviving - probability ) it follows with the insertion of constants @xmath234 and @xmath235:@xmath236 \sin ^{2}\left ( 2\theta \right ) \label{neutrino_probability - transition}\]]and@xmath237 \sin ^{2}\left ( 2\theta \right ) \label{neutrino_probability - surviving}\ ] ] in the case @xmath158 , these formulas are reduced to the standard ones @xcite .
assuming @xmath158 , experimental data from kamland for the oscillation between antineutrino - eletron @xmath238 to the antineutrino - muon @xmath239 gives the following values for physical constants @xcite:@xmath240finally , i leave to the experts the analysis of experimental data taking into account the dispersive parameter : _ is it possible that the data can eventually corroborate the hypothesis that neutrino dynamics have a dispersive parameter different from zero ? _
to summing up , the effect of non - unitary time evolution in quantum systems can be threefold : _ dissipation / gain _ ( which means variation of system s energy ) , _ impurification _ ( which means time evolution from pure states to mixed states , what is measured by the systems entropy ) and _ decoherence _ ( which happens only to composed systems and was not discussed here ) . in general ,
those phenomena are related to non - time reversal invariance and occur typically in open systems the environment being the system s partner to the exchanges of matter , energy , momenta and information ( entropy ) .
that is our expectation for completely positive markovian quantum systems with non - zero dissipator superoperator ( which means a deviation from unitary time evolution ) ; however , for the class of dispersive quantum systems , time evolution implies impurification even when the systems are isolated . surely , i do nt know if the concept of dispersive quantum system can help us deepen our understanding about time reversal invariance and the meaning of _
irreversibility_. nevertheless , the existence of an actual elementary dispersive quantum systems would be remarkable , since for them the non - time reversal invariance ( and the property of the system s entropy be non - decreasing with time ) can not have a statistical meaning .
if one does not forget the essential difference concerning statistical interpretation , it can be useful to think about dispersive quantum systems as the quantum analogs of classical isolated thermodynamical systems ( such as a low density gas in free expansion ) because both can be isolated and non - time reversal invariant .
i glimpse some developments to be done from what was presented here : i ) the characterization of dispersive quantum systems with infinite degrees of freedom;ii ) the study of decoherence in the context of dispersive quantum systems;iii ) improving the modeling of neutrino oscillation.iv ) the building of new models and the search to get , in the laboratory , instances of dispersive quantum systems .
finally , the issue deserves further theoretical as well as experimental researches , if not due to the possibility to describe actual physical systems ( like neutrinos ) , at least because we ignore answers to this simple question : _ if dispersive quantum systems can not exist in nature ( even approximately ) , are there physical principles forbidding them to exist ? _
here , i present the definition of a _ completely positive quantum dynamical semigroup _ following @xcite ( using a different notation ) . denote by @xmath241 the c*-algebra of @xmath242 complex matrices with identity @xmath243 , for any positive integer @xmath244 .
[ def_completely - positive - map ] let @xmath245 and @xmath246 be c*-algebras .
an operator @xmath247 is said to be completely positive if and only if for all positive integer @xmath244 the following map is positive:@xmath248 j. uffink , _ three concepts of irreversibility and three versions of the second law in time and history _ , in : f. stadler and m. stoelzner ( eds ) heussenstamm , _ proceedings of the 28 . international ludwig wittgenstein symposium _ , ontos verlag(2006 ) : 275 - 288 .
b. pontecorvo , _ mesonium and antimesonium _ , sov .
jetp 6 ( 2957 ) 429 ; _ inverse beta processes and nonconservation of lepton charge _ ,
_ ibid _ 7 ( 1958 ) 172 ; _ neutrino experiments and the question of leptonic - charge conservation _ , _ ibid _ , 26 ( 1968 ) 984 .
a. aguilar et al .
( lsnd collaboration ) , _ evidence for neutrino oscillations from the observation of _
@xmath238 _ e appearance in a _ @xmath255 _ _ beam__. phys .
d 64 ( 2001 ) 112007 ; _ event excess in the miniboone search for @xmath256 oscillations _ , phys . rev .
( 2010 ) 181801 . | a _ dispersive quantum system _ is a quantum system which is both isolated and non - time reversal invariant .
this article presents precise definitions for those concepts and also a characterization of dispersive quantum systems within the class of completely positive markovian quantum systems in finite dimension ( through a homogeneous linear equation for the non - hamiltonian part of the system s liouvillian ) . to set the framework ,
the basic features of quantum mechanics are reviewed focusing on time evolution and also on the theory of completely positive markovian quantum systems , including kossakowski - lindblad s standard form for liouvillians .
after those general considerations , i present a simple example of dispersive two - level quantum system and apply that to describe neutrino oscillation .
example.eps gsave newpath 20 20 moveto 20 220 lineto 220 220 lineto 220 20 lineto closepath 2 setlinewidth gsave .4 setgray fill grestore stroke grestore |
there exists a generic minimal tree - level diagram , with two external scalars and a heavy intermediate fermion , that can generate naturally suppressed standard model ( sm ) neutrino masses ; see figure [ fig : nu_tree_generic ] .
the internal fermion line in this diagram has a single mass insertion , which can be of the majorana type or the dirac type .
the minimal ( and best known ) models that produce this diagram are the type - i @xcite and type - iii @xcite seesaws , where the sm is augmented by an @xmath0 singlet / triplet fermion with zero hypercharge . in these cases ,
lepton number symmetry is broken by the ( majorana ) mass insertion . however , the underlying mechanism is more general , and alternative extensions of the sm can realize the basic diagram in a number of ways .
the set of these minimal tree - level diagrams can be partitioned according to the nature of the mass insertion ( equivalently , to the origin of lepton number violation ) .
any given representative diagram of the distinct subsets has a number of properties that can be determined without recourse to a specific model .
these properties can , in turn , be used to guide one in the search for viable realizations of these minimal seesaws . in this work we aim to catalogue the minimal models that produce small neutrino masses via one of these tree - level diagrams . to achieve this goal we reverse - engineer the models . in the process
we rediscover " some models that have already appeared in the literature , and discover a number of additional models which , to the best of our knowledge , have not previously appeared . as one moves beyond the minimal realizations of figure [ fig : nu_tree_generic ] ,
the models typically require additional fields to be added to the sm .
thus , it naively appears that the tree - level diagram can be realized in a large number of ways .
however , as we will see , if one restricts their attention to natural models , in which no tuning is needed to achieve small vacuum expectation values ( vevs ) , the list of candidate models is finite and quite short .
we provide a comprehensive version of this list , apparently exhausting the variant seesaws of this type .
interestingly a number of the corresponding models only realize viable seesaws when the new physics occurs near the tev scale , and are therefore largely amenable to discovery ( or exclusion ) at the lhc .
these models realize neutrino mass by low - energy effective operators with mass dimension @xmath1 . before proceeding we note that some models described in this paper employ scalars in non - fundamental representations of @xmath0 .
the demands of perturbative unitarity place general upper bounds on the quantum numbers of larger scalar multiplets @xcite ( also see @xcite ) .
however , all multiplets appearing here are consistent with these constraints .
bounds from flavor changing processes in models with large multiplets can also be found in ref .
@xcite .
the plan of this paper is as follows . in section
[ sec : mass_insert ] we consider diagrams with a lepton number violating majorana mass insertion .
section [ sec : l_vertex ] considers diagrams with a dirac mass insertion , for which lepton number symmetry is broken by a vertex .
loop effects are briefly discussed in section [ sec : loops ] , and we conclude in section [ sec : conc ] . in an appendix we provide details for some non - minimal cases with a majorana mass insertion .
the busy " reader is advised that our main results are contained in the two tables presented in the text .
readers interested primarily in the particle content of the models can refer to these tables ; the minimal natural models with a mass insertion of the majorana ( dirac ) type are the first three ( five ) entries in table [ l_mass_result ] ( table [ l_vertex_result ] ) .
in the generic tree - level diagram of figure [ fig : nu_tree_generic ] , the mass insertion can be of the majorana type or the dirac type . adopting the standard convention for a type - i seesaw , in which the beyond - sm fermion @xmath2 is assigned the same lepton number value as the sm leptons ,
these two cases correspond to lepton number violation by the mass insertion , or by a vertex , respectively . in this section
we consider models with a lepton number violating mass insertion , for which the generic tree - level diagram takes the form shown in figure [ fig : l_massinsert_nu_tree_generic ] .
inspection of the figure reveals the following generic features : * the internal fermion transforms as @xmath3 under the sm gauge symmetry ; that is , @xmath4 should form a real representation of @xmath5 .
* the multiplet @xmath6 should contain an electrically neutral component , constraining @xmath7 to be an odd number .
* the two external scalars can be distinct .
however , minimal models occur when they have the same quantum numbers : @xmath8 .
* the quantum numbers of the scalars must satisfy the following conditions : y_s =- y_l=1 r_sr_2 .
we focus on the minimal case with @xmath9 here , but comment on the more general case at the end of the section .
the first condition ensures that a lepton - number violating majorana mass appears in the lagrangian , while the last condition ensures that the requisite yukawa couplings appear : & & _ s s_1 l + _ r^c+ , where @xmath10 denotes a sm lepton doublet .
integrating out the heavy fermions , and inserting the scalar vevs , gives the generic form for the seesaw - suppressed sm neutrino masses in these models : m__s^2 .
this mass has the familiar seesaw form , and for good reason ; the simplest models that realize figure [ fig : l_massinsert_nu_tree_generic ] occur when @xmath11 is the sm scalar . in this case , one can have @xmath12 or @xmath13 , corresponding to the well known type - i @xcite and type - iii @xcite seesaw mechanisms , respectively .
these give rise to the famous @xmath14 weinberg operator @xcite in the low - energy effective theory ( see e.g. @xcite ) .
it should be clear that the basic mechanism for generating seesaw suppressed neutrino masses is more generic than the type - i and type - iii seesaws .
indeed , on the basis of the above - mentioned conditions , it seems that one can select any odd value of @xmath7 , and then choose an appropriate @xmath15 to obtain suppressed neutrino masses via figure [ fig : l_massinsert_nu_tree_generic ] .
however , there are additional considerations when one moves away from the simplest cases . in models with @xmath16
, one has @xmath17 and @xmath18 , so @xmath15 forms a non - trivial representation of @xmath0 .
thus , when symmetry breaking occurs , the vev of @xmath15 contributes to the @xmath19 and @xmath20 boson masses .
this modifies the tree - level value of the @xmath21-parameter away from the sm value of @xmath22 , leading to a bound on the vev of roughly @xmath23 gev @xcite .
direct searches , on the other hand , typically constrain the mass of scalars with non - trivial @xmath0 quantum numbers to satisfy @xmath24 gev .
therefore @xmath15 is expected to be a heavy scalar with a small vev : @xmath25 .
there are two ways in which one can add a heavy scalar to the sm and obtain a small tree - level vev . in the first approach
, one must tune the parameters in the scalar potential , to ensure that multiple dimensionful quantities that are @xmath26 gev conspire to deliver ( @xmath27 .
this approach has limited appeal . in the second approach ,
the quantum numbers of @xmath15 are such that the scalar potential contains a term that is linear in @xmath15 .
we denote this term as @xmath28 , where @xmath29 denotes generic dimensionful quantities ( field operators or constants ) .
then @xmath15 develops a nonzero vev of the form @xmath30 .
thus , provided @xmath31 , the vev @xmath32 can be naturally suppressed , ensuring that the constraints from the @xmath21-parameter are satisfied without fine tuning . and
the triplet @xmath33 obtains the vev @xmath34 .
we restrict our attention to models possessing a generalized version of this natural vev suppression . ]
the second approach is clearly more desirable .
we therefore restrict our attention to models in which the small vev of any new scalar is naturally suppressed without fine - tuning .
these models achieve a _
natural _ tree - level seesaw via heavy fermion exchange ( with a majorana mass insertion , in the present case ) . to find candidate models that meet this constraint
, one can consider the direct product of @xmath35 sm scalars , namely @xmath36 . here
@xmath37 denotes a generic sm scalar , including hermitian- or charge - conjugate fields , and the integer @xmath35 satisfies @xmath38 .
a viable model is found if the direct product contains a term @xmath39 for some @xmath38 , provided a scalar field with the quantum numbers @xmath40 contains an electrically neutral component .
this fixes the quantum numbers of the new scalar , @xmath41 , and determines a new model with a natural tree - level seesaw .
the demand of naturalness , in the sense defined above , turns out to be a rather stringent constraint .
a comprehensive study of the aforementioned direct products reveals a single candidate ; namely h^h(1,4,-1 ) , where @xmath42 is the charge conjugate field .
thus , if we add the scalar field @xmath43 to the sm , and also add a real fermion @xmath4 with quantum numbers @xmath44 such that @xmath45 , we obtain a natural tree - level seesaw .
there are two candidates for the new fermion field , namely @xmath13 and @xmath46 .
the former case is non - minimal as the standard type - iii seesaw is also present .
absent hierarchical yukawa couplings of a conspiratorial nature , the type - iii term is expected to dominate the neutrino mass matrix ( we will return to this matter below ) .
we therefore conclude that one can extend the sm by adding the fields s_1~(1,4,1 ) _
r~(1,5,0 ) , to arrive at a new seesaw model that is distinct from the standard type - i and type - iii seesaws .
actually , this model was recently proposed in ref .
@xcite , where it was shown to provide a viable model of naturally suppressed seesaw neutrino masses .
the lagrangian contains the terms & & _ s s_1 l + _ r^c + h^3 s_1 , where the last term ensures that @xmath15 develops a naturally suppressed vev : , the linear term in @xmath47 forces a naturally suppressed vev of this form . ]
s_1 .[s1_vev_quad ] putting all the pieces together , one arrives at the final expression for the seesaw suppressed neutrino masses : m _ & & _
s^2 ~ , [ eq : quintuplet_nu_mass ] where , for simplicity , we consider dimensionless couplings of @xmath48 in the last expression .
the first expression has the standard seesaw form , @xmath49 , bearing testament to the relation between this model and the more familiar seesaws . denoting @xmath50 as an approximate common scale for the new physics
, the second expression gives @xmath51 .
thus , in this model , the tree - level diagram in figure [ fig : l_massinsert_nu_tree_generic ] produces an effective low - energy operator is studied in ref .
@xcite . ] with mass - dimension @xmath52 , namely : @xmath53 .
the new fields can therefore be relatively light ; dimensionless couplings of order @xmath54 allow @xmath55 to be as low as @xmath56 gev @xcite .
the scale of new physics is thus expected to be significantly smaller than that found in the minimal realizations of figure [ fig : l_massinsert_nu_tree_generic ] , namely the type - i and type - iii seesaws .
the payoff for the extra complexity , it seems , is that interesting regions of parameter space for the resulting model can be explored at the lhc , and there is a realistic chance that the full parameter space of the model can be probed in the not - too - distant " future .
this exhausts the list of minimal models that produce natural seesaw neutrino masses via figure [ fig : l_massinsert_nu_tree_generic ] .
the natural models of this type are the type - i and type - iii seesaws , and the model with a quintuplet fermion introduced in ref .
these models are minimal , in the sense that they comprise minimal extensions of the sm , and that there is only one type of tree - level diagram generating neutrino masses .
there are additional models with a lepton number violating mass insertion that are less minimal .
these models generate diagrams with @xmath57 . in what follows
we briefly discuss an example to illustrate some of the differences .
additional general analysis for these models appears in appendix [ app : mass_non_minimal ] .
return to the case where the sm is extended to include the scalar @xmath43 .
instead of considering the minimal scenario with @xmath46 , let us now take @xmath13 .
then there are three distinct tree - level diagrams that contribute to light neutrino masses .
the first of these has two external sm scalars , the second has two external scalars @xmath15 , and the third has one external @xmath15 and one external sm scalar .
the light neutrino mass has multiple contributions , which can be written as m_&&_h^2 + 2_h_s + _
s^2 .[eq : type_3_variant ] the first piece here is the usual type - iii seesaw term . if all dimensionless couplings are of a similar order of magnitude , this type - iii term is expected to dominate the pure @xmath15 " term and the mixing term , due to the relation @xmath58 ; see the appendix . another way to say this ,
is that the three terms in eq .
correspond to low - energy operators with mass dimension @xmath14 , @xmath59 and @xmath52 , respectively .
given that all three operators arise from tree - level diagrams , one would normally expect the operator with the lowest mass dimension to dominate .
the model is therefore a more complicated version of the type - iii seesaw , with additional subdominant contributions to the neutrino mass matrix . strictly speaking
, one can imagine that for some reason the coupling @xmath60 is much smaller than the coupling @xmath61 , so the hierarchy in vevs is overcome . in this case the mixing term and the type - iii term can be subdominant to the pure @xmath15 piece , giving @xmath62 , with additional subdominant type - iii ( and mixed ) contributions .
this region of parameter space produces a seesaw that is distinct from the aforementioned minimal variants , and the model can be studied as a viable theory of neutrino masses .
of course , the need for hierarchical yukawa couplings runs counter to the spirit of the seesaw mechanism .
however , the model does have a certain benefit , as the phenomenology of the new scalar can be distinct , and there is a better chance of the new physics being experimentally accessible than in the pure type - iii seesaw @xcite .
.[l_mass_result ] natural seesaw models with a majorana mass insertion .
the first three entries are minimal while last two entries are not ; model @xmath63 is essentially model @xmath64 with an additional field . [
cols="^,^,^,^,^,^ " , ]
before concluding , we note that , in addition to the tree - level diagrams in the models we have described , there will also be loop processes that contribute to the light neutrino masses . generally speaking , one expects loop effects to be subdominant to the tree level results . however , one must be careful to account for the details of the given model . in the present class of models
, there always exists a coupling @xmath65 , where @xmath66 can denote the sm scalar .
consequently one expects that the two external scalars ( @xmath66 ) can be joined together to form a one - loop diagram with two external sm scalars ; see figure [ fig : loop ] .
this one - loop diagram generates an effective operator in the low - energy theory with mass dimension @xmath14 , i.e. the weinberg operator .
many of the tree - level diagrams presented here produce low - energy operators with mass dimension @xmath1 .
for these models there will be regions of parameter space in which the tree - level diagram is not the dominant effect .
for example , ref .
@xcite finds regions of parameter space in which the loop - diagram dominates the tree - diagram for model ( c ) in table [ l_mass_result ] .
one expects that similar effects are possible in the models presented here when @xmath67\ge 7 $ ] . in general
both tree - level and loop effects will be present , with the loop - effects dominating for larger values of the beyond - sm masses . if the loop - effect dominates the tree - level mass in a given model , one obtains radiative neutrino masses with sub - dominant tree - level corrections .
our interest is in tree - level seesaw masses and one can crudely estimate the parameter space for which the tree - level diagram is dominant as follows . in a model with @xmath67=d$ ] , the tree - level mass is roughly @xmath68 , while the loop - mass is @xmath69 .
demanding that the former exceeds the latter gives @xmath70 , so that @xmath71 tev for @xmath72 .
the region of parameter space for which the tree - level mass dominates is therefore eminently testable for models with @xmath52 . while tree - level dominance in models with @xmath73 is , at best ,
marginally permitted , and may already be excluded .
certainly the lhc should definitively determine whether the tree - level masses are viable for models ( d ) and ( e ) , and likely for models ( b ) and ( c ) also .
meaningful regions of parameter space will also be probed for model ( a ) .
let us emphasize , however , that this rough estimate should only be used as a guide .
there is a generic minimal tree - level diagram , with two external scalars and a heavy intermediate fermion , that can naturally achieve small neutrino masses via a seesaw .
the diagram has a mass insertion on the internal fermion line , and the set of such diagrams can be partitioned according to whether the mass insertion is of the majorana or dirac type .
we have shown that , upon demanding that requisite small vevs are naturally suppressed , this set is finite , and amenable to a systematic description .
we have undertaken this task , and found a number of original seesaw models in the process .
our results show that there are only three minimal models with a lepton - number symmetry breaking ( majorana ) mass insertion ; namely the type - i and type - iii seesaws , and the quintuplet seesaw of ref .
five additional minimal models also exist , in which the mass insertion is of the dirac type , and lepton number symmetry is broken by a vertex ; these are the @xmath59 model of ref .
@xcite , two models with @xmath52 ( one of which was presented in ref .
@xcite ) , and two further models with @xmath73 . our work appears to exhaust the list of these minimal non - tuned tree - level seesaws . in a partner paper
we shall present similar generalizations of the inverse seesaw mechanism @xcite .
a more detailed analysis of the new @xmath52 model , and a discussion of dark matter candidates in the class of models considered here @xcite , will appear elsewhere .
the author thanks s. s. c. law .
this work was supported by the australian research council .
models that generate neutrino mass by the seesaw diagram in figure [ fig : l_massinsert_nu_tree_generic ] with @xmath57 also generate diagrams with two external @xmath15 vevs and two external @xmath74 vevs .
these models are therefore non - minimal , in the sense that there are distinct tree - level diagrams contributing to the neutrino masses .
one can always remove one of the scalars from the model and yet retain a viable theory of seesaw neutrino masses . in these models
the intermediate fermion must be real , @xmath44 , with odd @xmath75 .
the quantum numbers of the scalars are fixed as @xmath76 .
thus , to ensure that @xmath77 , only models with @xmath78 can have two scalars with distinct quantum numbers . for these models
one always has @xmath79 and @xmath80 , so the term @xmath81 is always allowed .
provided one of the scalars obtains a naturally suppressed vev via its couplings with the sm scalar @xmath37 in @xmath82 , the other scalar @xmath83 can acquire a naturally suppressed vev via this quartic coupling . for the case of @xmath84 one has the model of ref .
@xcite , while for @xmath85 one has @xmath86 , and @xmath15 has suitable quantum numbers to allow @xmath87 . both @xmath88 and @xmath74 therefore
develop naturally suppressed vevs .
for @xmath89 no coupling linear in @xmath90 can appear in @xmath91 , so one can not be assured of generating naturally suppressed vevs for the beyond - sm scalars [ which form larger representations of @xmath0 ] .
the list of natural non - minimal models of this type is therefore exhausted by the last two entries in table [ l_mass_result ] .
p. minkowski , phys .
b * 67 * , 421 ( 1977 ) ; t. yanagida , in _ workshop on unified theories _
, kek report 79 - 18 p.95 ( 1979 ) ; m. gell - mann , p. ramond , r. slansky , in _ supergravity _ ( north holland , amsterdam , 1979 ) eds .
van nieuwenhuizen , d. freedman , p.315 . ; s. l. glashow , in _ 1979 cargese summer institute on quarks and leptons _
( plenum press , new york , 1980 ) eds .
m. levy , j .- l .
basdevant , d. speiser , j. weyers , r. gastmans and m. jacobs , p.687 ; r. barbieri , d. v. nanopoulos , g. morchio and f. strocchi , phys .
b * 90 * , 91 ( 1980 ) ; r. n. mohapatra and g. senjanovic , phys .
lett . * 44 * , 912 ( 1980 ) .
r. foot , h. lew , x. g. he and g. c. joshi , z. phys .
c * 44 * , 441 ( 1989 ) .
k. hally , h. e. logan and t. pilkington , phys .
d * 85 * , 095017 ( 2012 ) [ arxiv:1202.5073 [ hep - ph ] ] .
k. earl , k. hartling , h. e. logan and t. pilkington , arxiv:1303.1244 [ hep - ph ] .
w. konetschny and w. kummer , phys .
b * 70 * , 433 ( 1977 ) ; t. p. cheng and l. f. li , phys .
d * 22 * , 2860 ( 1980 ) ; m. magg and c. wetterich , phys .
b * 94 * , 61 ( 1980 ) ; j. schechter and j. w. f. valle , phys .
d * 22 * , 2227 ( 1980 ) ; g. lazarides , q. shafi and c. wetterich , nucl .
b * 181 * , 287 ( 1981 ) ; c. wetterich , nucl .
b * 187 * , 343 ( 1981 ) ; r. n. mohapatra and g. senjanovic , phys .
d * 23 * , 165 ( 1981 ) .
y. liao , phys .
b * 694 * , 346 ( 2011 ) [ arxiv:1009.1692 [ hep - ph ] ] .
k. kumericki , i. picek and b. radovcic , phys .
d * 86 * , 013006 ( 2012 ) [ arxiv:1204.6599 [ hep - ph ] ] ; i. picek and b. radovcic , phys .
b * 719 * , 404 ( 2013 ) [ arxiv:1210.6449 [ hep - ph ] ] . also see : y. liao , jhep * 1106 * , 098 ( 2011 ) [ arxiv:1011.3633 [ hep - ph ] ] . b. ren , k. tsumura and x. -g .
he , phys .
d * 84 * , 073004 ( 2011 ) [ arxiv:1107.5879 [ hep - ph ] ] .
k. s. babu , s. nandi and z. tavartkiladze , phys .
d * 80 * , 071702 ( 2009 ) [ arxiv:0905.2710 [ hep - ph ] ] .
e. del nobile , r. franceschini , d. pappadopulo and a. strumia , nucl .
b * 826 * , 217 ( 2010 ) ; [ arxiv:0908.1567 [ hep - ph ] ] . c. -k .
chua and s. s. c. law , phys . rev .
d * 83 * , 055010 ( 2011 ) [ arxiv:1011.4730 [ hep - ph ] ] ; a. delgado , c. garcia cely ,
t. han and z. wang , phys .
d * 84 * , 073007 ( 2011 ) [ arxiv:1105.5417 [ hep - ph ] ] ; s. s. c. law and k. l. mcdonald , phys .
b * 713 * , 490 ( 2012 ) [ arxiv:1204.2529 [ hep - ph ] ] .
j. hisano and k. tsumura , arxiv:1301.6455 [ hep - ph ] .
i. picek and b. radovcic , phys .
b * 687 * , 338 ( 2010 ) [ arxiv:0911.1374 [ hep - ph ] ] ; k. kumericki , i. picek and b. radovcic , phys . rev .
d * 84 * , 093002 ( 2011 ) [ arxiv:1106.1069 [ hep - ph ] ] .
fileviez perez , h. h. patel , m. j. ramsey - musolf and k. wang , phys .
d * 79 * , 055024 ( 2009 ) [ arxiv:0811.3957 [ hep - ph ] ] .
s. s. c. law and k. l. mcdonald , phys .
d * 87 * , 113003 ( 2013 ) [ arxiv:1303.4887 [ hep - ph ] ] .
s. s. c. law and k. l. mcdonald , arxiv:1305.6467 [ hep - ph ] . | there exists a generic minimal tree - level diagram , with two external scalars and a heavy intermediate fermion , that can generate naturally small neutrino masses via a seesaw .
this diagram has a mass insertion on the internal fermion line , and the set of such diagrams can be partitioned according to whether the mass insertion is of the majorana or dirac type .
we show that , once subjected to the demands of naturalness ( i.e. precluding small scalar vacuum expectation values that require fine - tuning ) , this set is finite , and contains a relatively small number of elements . some of the corresponding models have appeared in the literature .
we present the remaining original models , thus generalizing the type - i and type - iii seesaws , and apparently exhausting the list of their minimal non - tuned variants .
minimal tree - level seesaws with a heavy intermediate fermion + kristian l. mcdonald + arc centre of excellence for particle physics at the terascale , + school of physics , the university of sydney , nsw 2006 , australia + [email protected] |
the temporal flux variability from active galactic nuclei ( agn ) , detectable in nearly all wavelength regimes , contains information on the underlying emission processes and source geometry that is otherwise difficult to probe @xcite .
however , precise details of the physical mechanism generating the observed nuclear variability in agn remain unclear @xcite .
current and future large - scale photometric time - domain surveys have motivated many recent studies of the optical broadband variability properties of various agn subclasses using large numbers of well - sampled light curves .
this has been especially useful for agn identification and selection @xcite . beyond the optical , large - scale surveys of agn variability
have been pursued at many other wavelengths , including the radio @xcite , ultraviolet @xcite , and @xmath2-ray regimes @xcite .
fewer studies of agn variability have focused on the infrared ( ir ) , due in part to the expectation that non - variable dust and the host galaxy dominate the emission at these wavelengths ( e.g. see * ? ? ?
explanations of agn variability in the optical as due to localized temperature fluctuations in inhomogeneous accretion disks @xcite or global accretion rate changes @xcite predict little flux variability from disk emission in the ir .
reprocessing of variable optical emission by a dusty torus has been suggested by @xcite as a possible cause for the nir variability , although the time - lags between the optical and nir are difficult to detect .
for example @xcite have shown small time lags between nir and x - ray data for the agn 3c273 , with the nir variability leading the x - ray by @xmath41.5 days , implying this nir variability originates near the nucleus . a study of nir variability in the spitzer deep wide - field survey botes field by @xcite showed that although only 1.1% of objects appear to be variable in the near - ir ( nir ) , the vast majority of the variable objects were agn .
small samples of agn monitored in both the optical and nir have shown that while nuclear variability is prevalent , the amplitude of variations decreases towards longer wavelengths .
for example , @xcite have produced realistic models of the optical to nir variability for ngc 4151 , which show a small amplitude time lag and decreasing amplitude variability as a function of wavelength . however , this is not true for blazars , whose nir flux is dominated by non - thermal emission from a relativistic jet , rather than circum - nuclear dust which is not expected to be non - variable @xcite .
recent results from optical / nir monitoring of @xmath5 @xmath2-ray blazars have shown that flat - spectrum radio quasars are more variable in the nir than in the optical @xcite . @xcite
have used 7 year photometric light curves of 7 blazars , spanning the optical to nir bands ( @xmath6 ) , to find that variability amplitudes increase with wavelength for these objects . in this study
we search for nir variability of luminous agn using well sampled light curves from the two micron all sky survey ( 2mass ) calibration data , finding one highly variable blazar ( sdssj14584479 + 3720215 ) .
we explore the utility of nir variability - based selection of agn , including its use in identifying counterparts to @xmath5 @xmath2-ray sources .
we also study the nature of non - agn contaminants with similar nir colors and variability properties , laying groundwork for the agn variability science possible with current and future multi - epoch ir surveys .
the 2mass survey observed the full sky in the near - ir over the timespan of june 1997 to february 2001 using the @xmath7 , @xmath8 , and @xmath9 bands @xcite .
photometric zero - points for calibrating the survey were based on hourly observations of a set of 40 standard fields , each @xmath10 in size , and spaced evenly throughout the sky .
these fields were visited repeatedly over the course of the survey , resulting in 562 to 3692 epochs of observation per field .
this produced @xmath0 light curves for 113,030 individual objects , known as the 2mass calibration point source working database ( hereafter cal - pswdb , * ? ? ?
* ; * ? ? ?
these light curves are the most precise nir time domain survey to date and provide a novel dataset in which to study the infrared variability properties for a wealth of astrophysical phenomena .
this dataset has produced the best sampled nir light curve for a rr lyr star @xcite , a hunt for stellar flares at long wavelengths @xcite , a large sample of young stellar objects ( yso s , * ? ? ?
* ; * ? ? ?
* ) , and many well characterized binary star systems ( * ? ? ?
* ; * ? ? ?
* davenport 2014 in prep ) . here
we make use of these data to search for nir variability of agn with unprecedented precision . the wide - field infrared survey explorer ( wise , * ? ? ?
* ) has created the deepest mid - ir survey of the entire sky in four photometric bands , from 3.4 @xmath11 m to 22 @xmath11 m .
several recent studies have demonstrated the utility of wise colors for effectively separating agn from stars and unresolved galaxies .
we spatially matched the cal - pswdb objects to the wise all sky data release using a match radius of 15 .
this resulted in 52,148 objects with @xmath0 light curves and a detection in at least one wise filter .
the distribution of these point sources in wise color color space is shown in figure [ wcolor ] .
for comparison , we show the wise `` blazar strip '' , defined in @xcite as the gaussian kernel density estimate of the wise colors of blazars from @xcite .
to find previously known agn in our data , we spatially matched these 52k objects to the `` milliquas catalog , '' version 3.3 .
we recovered 27 known agn with light curves in cal - pswdb , 25 of which had wise colors shown in figure [ wcolor ] .
these agn had between 26 and 3482 epochs of cal - pswdb photometry , with an average of 924 epochs .
of these , 19 have been spectroscopically confirmed by the sloan digital sky survey ( sdss ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) , one of which has been identified as a blazar ( sdssj14584479 + 3720215 ) .
-band ( top ) and @xmath12 color ( bottom ) light curves for the known blazar , sdssj14584479 + 3720215 .
the median photometric error is shown for comparison ( vertical bar).,width=336 ] this set of known agn with well sampled cal - pswdb light curves creates a unique dataset in which to search for agn variability in the nir . to determine if these objects were variable in the 2mass bands , we computed the error - weighted root mean square variability for each light curve , @xmath13 , following the prescription in @xcite .
objects were classified as variable if they had @xmath14 mag . of the `` milliquas '' agn in our dataset , only the known blazar , sdssj14584479 + 3720215 , exhibited variability above this threshold .
the @xmath7-band and @xmath12 color light curves for this blazar , with 1972 epochs of cal - pswdb photometry spanning over three years , are presented in figure [ lc ] .
the remaining 26 known agn in our sample displayed no significant nir variability using this metric .
we found strong variability in sdssj14584479 + 3720215 in all three 2mass bands with our densely sampled light curves .
however , as seen in figure [ lc ] no significant variation in @xmath12 color was found for this object , indicating the underlying variability was `` grey '' over this wavelength regime . a larger sample of such well studied variable objects is needed to test if this is truly a generic property of blazars in the nir .
characterizing photometric variability timescales has proven to be a very efficient means of selecting a clean sample of agn from time domain surveys ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the so - called `` damped random walk '' ( drw ) model provides a robust fit to agn light curves as a function of two observable free parameters : the characteristic timescale @xmath15 , and the variability amplitude sf@xmath16 .
stars are easily removed when fit with this model , as their primary sources of variability are typically stochastic ( e.g. flares or flickering ) or strongly periodic ( such as pulsations or rotation ) in nature .
combining drw variability models with multi - wavelength photometry provides even greater power in separating blazars from `` normal '' agn @xcite . from our cal - pswdb
wise matched sample we selected all light curves having at least 50 good @xmath7-band epochs ( ph_qual = a , b or c ) , median @xmath7-band magnitudes brighter than 16.5 mag , and wise colors of @xmath17 to remove main sequence stars @xcite .
we also required sources to be detected as variable using the @xmath18 cut described above , and removed objects known to be periodic ( binaries and pulsating variables ) or long period variable stars ( * ? ? ? * davenport 2014 in prep ) .
this yielded 22 variable cal - pswdb targets to search for drw timescales , which included our known blazar object , but did not select any other previously identified agn . for completeness
we also analyzed the other 26 known `` milliquas '' sources for drw variability .
for each object we down - sampled the light curves to a single data point per night , using the median of the @xmath7-band photometry , which reduced the light curves to an average of 144 epochs .
we then ran the drw fitting code from @xcite and @xcite on the reduced @xmath7-band light curves for all 47 objects .
we used the probability thresholds outlined in @xcite to determine which light curves had significant drw variability .
the known blazar had a weakly constrained drw timescale of @xmath19 days , and a significant driving amplitude of @xmath20 mag year@xmath21 .
the relative likelihood for the drw versus an infinite or unconstrained timescale was log(l@xmath22/l@xmath23 ) = 1.4 , where values greater than 0.05 are considered significant .
similarly , the relative likelihood for the drw model versus pure noise was log(l@xmath22/l@xmath24 ) = 153 , which was much greater than the threshold for significance of 0 .
none of the other 26 previously identified agn showed any signs of significant drw variability .
however , 7 of the variability - selected objects did show significant drw variability .
we discuss the origin of these objects in the following section .
we additionally fit these light curves using the continuous auto - regressive moving average , or carma(p , q ) , models of @xcite , where @xmath25 is the auto - regressive order and @xmath26 the moving average order .
the drw is a special case of the carma models , equivalent to a carma(1,0 ) , or a first - order continuous auto - regressive process with no moving average . in the carma formalism , a drw model s power spectral density is described as a single lorentzian function with a characteristic break frequency . as mentioned above , the drw has been shown to empirically model the behavior of qso light curves well , but is otherwise inflexible . by looking at higher order carma models we search for more complex behavior in the light curve power spectral density , such as stochastic variability on multiple timescales or quasi
periodic oscillations . for sdssj14584479 + 3720215
we found a moderate preference for the carma(2,0 ) model compared to a drw when using the deviance information criterion for model selection .
the drw model provides a sufficient fit to the data , in that the sequence of residuals is independently and normally distributed .
a markov chain monte carlo analysis of the carma(1,0 ) model provided 1-@xmath3 confidence levels on the timescale of @xmath27 days , and on the standard deviation of the driving noise of between 0.14 and 0.20 magnitudes .
the recovered drw timescales for the seven variability - selected objects we found in the previous section had an average timescale of @xmath28 days , significantly shorter than the typical timescale seen in the optical for agn @xcite , as well as for our blazar target .
this was reproduced in the carma(1,0 ) models as well .
these seven objects all reside in field # 90009 , which was centered on the star forming region @xmath1 ophiuchus , making these likely ysos . of these ysos
, five had x - ray detections in the literature , one coming from a targeted x - ray monitoring campaign of the field @xcite , and four from the chandra x - ray source catalog @xcite .
the other two ysos in our sample did not have a published x - ray counterpart within 0.1@xmath29 .
nir light curve properties for ysos within this field in the cal - pswdb have been studied in great detail by @xcite and @xcite .
these seven ysos had mid - ir colors that placed them within the blazar strip , nir variability amplitudes comparable to our known blazar , and light curves best parameterized by a carma(1,0 ) model .
we thus highlight ysos as a potentially important source of contamination for classifying agn from future multi - wavelength time domain searches .
while using longer wavelength filters can improve the efficiency of selecting agn from normal stars ( e.g. * ? ? ?
* ) , ysos remain a potential contaminant for object classification ( see also color spaces in fig 7 of * ? ? ?
. however , these objects may be distinguished from bona fide agn based on their @xmath410x shorter characteristic timescales under a carma(1,0 ) model .
ysos are also mostly confined to the galactic disk and dense star forming regions , and as such can be avoided by surveys .
the _ fermi gamma - ray space telescope _ provides the deepest survey to date in the 100 mev to 100 gev regime , and has discovered 886 @xmath2-ray agn in the 2nd _ fermi _ agn catalog @xcite .
the vast majority of these agn are blazars .
however , @xmath430% of _ fermi _ sources in the _ fermi _ 2nd point source catalog ( 2fgl ; * ? ? ?
* ) are unidentified , largely due to the poor angular resolution of _ fermi _
s pair - conversion large area telescope ( lat ; * ? ?
all reliably - associated agn in the 2nd _ fermi _ agn catalog are confidently detected by _
fermi _ with test statistic @xmath30 , corresponding to approximately @xmath31 detections . for a _ fermi _ counts map of a region of interest ( roi ) centered on a new possible source , the ts detection significance is the defined to be the ratio of likelihoods ts = 2(log@xmath32(source ) - log@xmath32(nosource ) ) , where @xmath32(source ) is the likelihood of a model of all known sources in the roi and a new source at the central location , while @xmath32(nosource ) is the likelihood for the same model but no new source at the central location @xcite . the detection significance is approximately @xmath33 .
thousands of potential faint @xmath2-ray blazars lie below the @xmath30 ( @xmath31 ) detection threshold , but are difficult to localize due to extremely low photon counts .
these faint @xmath2-ray blazars have large positional uncertainties , with error ellipses with radius @xmath3420 for 3@xmath3 fermi detections using 4 years of photon data @xcite .
the strong @xmath0 variability exhibited by the blazar sdssj14584479 + 3720215 relative to the quasars in our sample suggests that this nir variability is likely to be indicative of strongly beamed jets , which are characteristic of blazars , and are known to produce strong @xmath2-ray emission .
this allows for nir - variability based identification of previously unidentified or undetected @xmath2-ray blazars .
sdssj14584479 + 3720215 was not included in the _ fermi _ 2fgl catalog , and thus had @xmath35 is 2 years of photon data .
however , motivated by the strong nir variability we observe , we investigated whether this blazar can be detected at high significance using more photon data .
we downloaded approximately 4.5 years of available pass 7 _ fermi _ photon data of the source event class , spanning mission elapsed time 239557417 to 383570220 seconds .
we selected photons in the 200 mev to 100 gev energy range , within a 10@xmath29 region of interest ( roi ) centered on the blazar . using the standard binned likelihood analysis scripts provided in the fermi science tools package , we performed photon event selection with a roi - based zenith angle cut of @xmath36100@xmath29 and a rock angle cut of @xmath3652@xmath29 .
we produced a spatial model of the likely @xmath2-ray sources within 15@xmath29 of the blazar based on the 2fgl catalog , using the 2fgl best fit values for the template spectral model of these sources .
we included the blazar sdssj14584479 + 3720215 as an additional point source , modeled as a simple power - law spectrum .
the spectral normalization for all sources within 15@xmath29 was allowed to vary , and all spectral parameters were set as free parameters within the roi .
we included the appropriate 2-year pass 7 galactic diffuse emission and extragalactic isotropic diffuse emission models in our source model fit to the observed counts map .
the blazar sdssj14584479 + 3720215 is detected with @xmath37 , approximately a 5@xmath3 detection , with a @xmath2-ray power - law spectral slope of @xmath38 , typical of _ fermi _ blazars @xcite .
this detection significance is consistent with the @xmath39 found in a preliminary version of the _ fermi _ 4-year point source catalog by @xcite , produced using 4 years of photon data .
we note minor discrepancies in our modeling of the brightest object in the field , noticeable in the lower - left corner of the difference image .
these are well - known issues in _ fermi _ source analysis caused by the finite resolution of the counts maps , and do not affect our detection .
we have conducted a search for nir variability from agn using the 2mass c al - pswdb light curves .
no significant variability was found for 26 of the 27 known agn in our sample .
one previously known agn , showed strong variations in the 2mass filters , and we have presented the 1972 epoch nir photometric light curve for this blazar ( sdssj14584479 + 3720215 ) .
this previously known blazar is a benchmark object , displaying almost 1 magnitude of variability in the @xmath0 bands , but with no significant variation in @xmath12 color .
this is the best sampled nir light curve of a blazar ever measured .
we note a handful of other agn candidates from the 2mass cal - pswdb were recently identified by @xcite , including the blazar sdssj14584479 + 3720215 . our detection of sdssj14584479 + 3720215 as a @xmath2-ray source , selected by infrared variability , has implications on the association and classification of multi - wavelength counterparts of _ fermi _ sources .
current selection methods such as mid - ir color @xcite and optical variability @xcite suffer from low efficiency .
this is often due to `` normal '' type 1 quasars , which have similar colors in the mid - ir and are also variable in the optical .
as this pilot study demonstrates , these issues may be surmounted by the inclusion of ir variability information , since normal agn are not significantly variable at this wavelength regime .
a search for other such light curves in the 2mass cal - pswdb reveals a large number of contaminating ysos .
this study establishes a baseline for the agn variability science possible with current and future multi - epoch ir missions such as the wise , the wide field infrared survey telescope ( wfirst ; * ? ? ?
* ) , and vista variables in the va lctea ( vvv ; * ? ? ?
* ) surveys .
we thank p. plavchan and j. parks for helpful discussions of yso variability and sharing an early version of their manuscript .
the authors acknowledge support from nasa adp grant nnx09ac77 g and nasa nnx14ak26 g .
jjr is supported by nasa fermi grant nnx14aq23 g .
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 .
this research has made use of the nasa / ipac extragalactic database ( ned ) which is operated by the jet propulsion laboratory , california institute of technology , under contract with the national aeronautics and space administration .
this publication makes use of data products from the wide - field infrared survey explorer , which is a joint project of the university of california , los angeles , and the jet propulsion laboratory / california institute of technology , funded by the national aeronautics and space administration .
lcccccc j@xmath40 & 1063 & 16.86 & 1.45 & 14.91 & 1.33 & 3.55 + j@xmath41 & 30 & 16.96 & 1.43 & 15.80 & 1.41 & 3.05 + j@xmath42 & 339 & 17.00 & 1.51 & 15.29 & 0.42 & 2.71 + j@xmath43 & 86 & 17.03 & 1.52 & 15.61 & 1.27 & 3.38 + j@xmath44 & 11 & 17.08 & 1.18 & 15.39 & 1.48 & 2.39 + j@xmath45 & 254 & 16.99 & 1.48 & & & + j@xmath46 & 2580 & 16.72 & 1.75 & 13.58 & 0.97 & 2.85 + j@xmath47 & 68 & 17.08 & 1.49 & 15.22 & 1.49 & 3.27 + j@xmath48 & 778 & 16.23 & 1.56 & 13.39 & 0.87 & 2.58 + j@xmath49 & 1873 & 16.68 & 1.41 & 14.89 & 0.51 & 2.91 + j@xmath50 & 9 & 17.01 & & 15.20 & 1.32 & 3.20 + j@xmath51 & 19 & 17.09 & 1.76 & 15.78 & 1.43 & 3.44 + j@xmath52 & 49 & 17.02 & 1.41 & 15.77 & 0.59 & 3.18 + j@xmath53 & 2737 & 16.29 & 1.10 & 13.38 & 1.24 & 2.85 + j@xmath54 & 18 & 17.10 & & 16.50 & 1.20 & 3.84 + j@xmath55 & 1212 & 16.91 & 1.56 & 14.63 & 0.20 & 2.64 + j@xmath56 & 916 & 16.40 & 1.13 & 13.80 & 1.59 & 2.97 + j@xmath57 & 1568 & 16.79 & 1.64 & 14.16 & 1.18 & 3.09 + j@xmath58 & 1500 & 16.50 & 1.48 & 13.88 & 0.74 & 2.92 + j@xmath59 & 58 & 17.06 & 1.38 & 15.32 & 0.98 & 3.56 + j@xmath60 & 1927 & 16.33 & 1.77 & 14.03 & 1.10 & 2.85 + j@xmath61 & 1191 & 15.51 & 1.73 & 12.46 & 1.07 & 2.52 + j@xmath62 & 21 & 16.99 & 1.49 & & &
+ j@xmath63 & 888 & 16.75 & 1.32 & 14.34 & 1.45 & 2.90 + j@xmath64 & 15 & 16.88 & & 15.22 & 1.20 & 2.36 + j@xmath65 & 7 & 16.99 & 1.46 & 15.06 & 1.39 & 2.80 + j@xmath66 & 358 & 16.95 & 1.52 & 14.62 & 0.86 & 2.41 [ agntable ] , d. j. , ballet , j. , burnett , t. , & fermi large area telescope collaboration .
2014 , in american astronomical society meeting abstracts , vol .
223 , american astronomical society meeting abstracts 223 , 256.01 | active galactic nuclei ( agns ) are well - known to exhibit flux variability across a wide range of wavelength regimes , but the precise origin of the variability at different wavelengths remains unclear . to investigate the relatively unexplored near - ir variability of the most luminous agns , we conduct a search for variability using well sampled @xmath0-band light curves from the 2mass survey calibration fields .
our sample includes 27 known quasars with an average of 924 epochs of observation over three years , as well as one spectroscopically confirmed blazar ( sdssj14584479 + 3720215 ) with 1972 epochs of data .
this is the best - sampled nir photometric blazar light curve to date , and it exhibits correlated , stochastic variability that we characterize with continuous auto - regressive moving average ( carma ) models .
none of the other 26 known quasars had detectable variability in the 2mass bands above the photometric uncertainty .
a blind search of the 2mass calibration field light curves for agn candidates based on fitting carma(1,0 ) models ( damped - random walk ) uncovered only 7 candidates .
all 7 were young stellar objects within the @xmath1 ophiuchus star forming region , five with previous x - ray detections
. a significant @xmath2-ray detection ( 5@xmath3 ) for the known blazar using 4.5 years of fermi photon data is also found .
we suggest that strong nir variability of blazars , such as seen for sdssj14584479 + 3720215 , can be used as an efficient method of identifying previously - unidentified @xmath2-ray blazars , with low contamination from other agn . |
the observed bias in biopolymers composed from homochiral l - amino acids and d - sugars towards a single handedness or chirality is a remarkable feature of biological chemistry .
nowadays , there is a firm consensus that the homochirality of biological compounds is a condition associated to life that probably emerged in the prebiotic phase of evolution through processes of spontaneous mirror symmetry breaking ( smsb ) @xcite .
this could have proceeded by incorporating steps of increasing complexity thus leading to chemical systems and enantioselective chemical networks @xcite .
theoretical proposals for the emergence of homochirality in abiotic chemical evolution , are based either on deterministic or on chance events @xcite .
however , the current state of knowledge strongly suggests that the emergence of chirality must be based on reactions leading to spontaneous mirror symmetry breaking .
smsb are transformations yielding chiral outcomes as non - thermodynamic final stable states , and in the absence of any chiral polarization or external chiral physical forces @xcite .
this is provided by enantioselective autocatalysis , but not by the simple linear asymmetric induction reactions @xcite on which past discussions on deterministic or chance phenomena were based for the justification of biological homochirality .
systems capable of smsb lead to a stochastic distribution of final chiral signs between successive experiments .
nowadays this deterministic versus chance debate is restricted to more specific scenarios @xcite .
the smsb abiotic scenario for the emergence of single homochirality in the biological world implies that single asymmetry emerges provided a small chiral fluctuation with respect to the idealized racemic state can be amplified @xcite to a state capable of biotic evolution .
relevant features common to such systems are that they take into account the small fluctuations about the racemic state and that they display _ non - linear _ kinetic effects . these stochastic scenarios are theoretically well understood on general grounds @xcite and equally important , are experimentally feasible in the laboratory @xcite . on the theoretical side , for describing systems capable of smsb , the frank model has been widely invoked to justify the emergence of biological homochirality @xcite .
the original model @xcite consists of an irreversible enantioselective autocatalysis eq.([frank1953 ] ) and an irreversible mutual inhibition reaction eq.([mutual ] ) between the product enantiomers , in an open - flow system : @xmath0 this model has been extensively studied , allowing for reversible transformations and in diverse open - system scenarios @xcite .
the significance of the mutual inhibition step eq .
( [ mutual ] ) is that it makes smsb possible for first - order enantioselective autocatalysis , such as that of eq .
( [ frank1953 ] ) .
although enantioselective autocatalyses of quadratic and higher orders may by themselves lead theoretically to smsb , they correspond to reactions of unrealistically high molecularity . for example , biological replicators of interest for enantioselective autocatalytic transformations , have their mechanisms composed by consecutive bi - molecular reactions .
this means that , by themselves , these replicators _ can not amplify _ the initial enantiomeric excess ( @xmath1 ) .
however , the coupling to a mutual inhibition reaction between the enantiomers can enable smsb for some reaction and system parameters .
therefore , the chemically significant scenarios for the study of the emergence of chirality are those consisting of simple linear enantioselective autocatalyses coupled to reaction networks that include a mutual inhibition step . originally proposed as an alternative to the frank model , the limited enantioselectivity ( les ) model is composed of entirely reversible steps : an enantioselective autocatalytic reaction eq.([lesscheme ] ) , a limited enantioselective autocatalysis eq.([limited ] ) , and direct production of enantiomers from an achiral precursor eq .
( [ direct ] ) @xcite : @xmath2 note that the _ inverse _ reaction in eq.([limited ] ) with rate @xmath3 provides the necessary chiral inhibition step , thus effectively replacing frank s mutual inhibition eq .
( [ mutual ] ) leading to the inert product @xmath4 .
the dynamic stability properties of racemic and chiral states in fully reversible versions of frank and in les , including hetero- and homo - dimerization , in both open and closed systems , are reported in detail in @xcite .
typically , rate equation theory ( in the mean field approximation ) is used to cast chemical reaction schemes in terms of coupled differential equations for the temporal evolution of the concentrations of the chemical species involved . in this deterministic kinetic dynamics , _ initial conditions _ must be taken to simulate the inherent statistical chiral fluctuations about the ideal racemic composition @xcite .
in contrast , real chemical reactions are inherently stochastic in nature : the reagents in solution ( or attached on surfaces ) must encounter each other before they can react , and the probability per unit time for the reaction to occur is related to the corresponding reaction rate constant , which depends on the temperature .
the molecular nature of chemical reagents and reactions gives rise to the concept of intrinsic _ reaction noise _ , and is typically multiplicative @xcite .
despite the fact that stochastic and deterministic kinetics must coincide in the macroscopic limit ( i.e. , for large numbers of molecules ) , stochastic methods can be used to address the question of whether such internal noise affects the final outcome of the underlying reaction , and in what way it might do so . the answer to
this question depends on the specific process studied .
thus , for example , reaction noise explains the anomalous scaling in reactions undergoing dynamic phase transitions from active to absorbing states @xcite .
the influence that reaction noise may have in schemes purporting to lead to smsb is readily investigated . in the case of the frank model with _
reversible _ autocatalysis ( defined by the pair of eqs.([lesscheme],[mutual ] ) ) , reaction noise induces complete chiral amplification starting from ideally racemic initial conditions in spatially extended domains , but this result requires the parameter @xmath5 , controlling the symmetry breaking transition , to satisfy the same condition as for the deterministic model , namely @xmath6 @xcite .
that is , the _ final outcome _ of the reaction scheme is governed solely by this simple condition , irrespective of whether internal noise is included in the temporal evolution or not . on the other hand ,
neither chiral bias nor external physical chiral polarizations need be invoked to achieve homochirality when reaction noise is included @xcite .
hence , the initial and subsequent chiral fluctuations intrinsic to the system @xcite are included automatically .
stochastic methods are necessary to describe kinetic dynamics in the case of small volumes and/or small numbers of reacting molecules @xcite , as is the case , for example , in compartmentalized cellular processes @xcite .
therefore , the differences in the evolution of the @xmath1 between deterministic and stochastic kinetics should provide better insights regarding asymmetric inductions and smsb processes in living systems . in this respect ,
internal noise has been considered recently in a closed - mass model which results from taking strictly irreversible enantioselective autocatalysis , eq.([frank1953 ] ) together with the direct production and decay of the enantiomers , eq.([direct ] ) @xcite .
stability analyses for the deterministic model show that the final stable state is necessarily racemic . yet
, the claim was made that reaction noise stabilizes the homochiral states , making these the most probable outcome of the system . according to this , in systems governed by stochastic kinetics ,
i.e. , before coinciding with the limit of deterministic kinetics , the linear decay of the enantiomers to the compound a ( eq . ( [ direct ] ) ) is to play the role of mutual inhibition for achieving smsb , i.e. , the homochiral states are supposed to result without requiring additional non - linearities " nor even chiral inhibition itself . in this paper
we analyze these claims in detail . as the need for chiral inhibition in smsb
has been questioned @xcite , we draw a careful distinction between linear racemization and non - linear mutual inhibition in sec [ sec : inhibs ] . the constraints dictated by detailed balance , often overlooked in the modeling of biological homochirality , are discussed in sec [ sec : micro ] . in sec [ sec : results ] we consider the influence , for both well - mixed and small spatially coupled systems , that reaction noise has on the stationary states of the chemical scheme when detailed balance is properly accounted for . in sec
[ sec : cycles ] we address the problem of coupling chemical reaction schemes to external energy sources for driving _ unidirectional _ cyclic reactions , a minimum requirement for biological systems .
conclusions are drawn in sec [ sec : disc ] .
details of the calculation of the probability distribution for the enantiomeric excess are relegated to an appendix .
as the scheme in @xcite ( i ) dispenses with chiral inhibition and ( ii ) overlooks the principle of detailed balance , we consider these two concepts below , bringing in some closely related reaction schemes for illustrative purposes .
the frank model @xcite involves enantioselective autocatalysis coupled to a reaction between the two enantiomers of product / catalyst yielding an achiral addition product .
the term mutual chiral inhibition " was coined for this reaction because it represents the decrease of chiral compounds in a racemic ratio .
this leads to an increase in the value of the enantiomeric excess @xmath1 . when this occurs faster than the reverse enantioselective autocatalysis , namely when @xmath7 , a cooperative effect drives the amplification of the @xmath1 in the enantioselective autocatalysis : eq.([frank1953 ] ) or eq.([lesscheme ] ) . as remarked above , in
les this mutual chiral inhibition is manifested via the inverse non - enantioselective autocatalysis , eq.([limited ] ) .
compared to frank , in les no inert product p nor achiral heterodimer is formed , but instead the recycling of one enantiomer back to the achiral precursor a plus the mirror image enantiomer .
this reverse reaction implies the disappearance of only one stoichiometric part of the racemic mixture ( either d or l ) , but in the dynamics of the system there is a non - linear dependence on the heterochiral interaction @xmath8\times[d]$ ] , just as in the case of the mutual inhibition stage of frank - like systems .
the importance for smsb of coupling enantioselective autocatalysis of first order ( eq.([frank1953 ] ) or eq.([lesscheme ] ) ) with such inhibition stages [ eq .
( [ mutual ] ) or eq.([limited ] ) ] is due to the fact that the autocatalysis _ by itself _ can not yield smsb . in lieu of these inhibitions
, there is no amplification of chirality , and in the best of cases , ( open flow systems , or systems with heterogeneous energy distributions , etc . ) , the production of chiral matter can only maintain the initial @xmath1 value .
the significance of the heterochiral inhibition stage , when coupled to an enantioselective autocatalysis of first order , is that the overall reaction network is then able to lead to smsb .
this chiral state is a non - thermodynamic one , but is the more stable state of the system .
the reaction of eq .
( [ direct ] ) , has been reported @xcite as being able to play the role of the needed inhibition stage for smsb in enantioselective autocatalysis of first order , by appealing to the presence of reaction noise .
the character of such an inhibition stage , however , is easily appreciated by re - expressing it as follows : @xmath9 the reactions of eqs.([direct],[direct1 ] ) are identical and the direct transformation between enantiomers in eq.([direct2 ] ) represents an equivalent overall transformation , as far as l and d are concerned .
what actually differs between all three is the temporal or spatial resolution or scale at which we can resolve them into individual steps or else as overall , collective reactions .
thus the first , eq.([direct ] ) , describes two individual reactions , but because of the degenerate character of enantiomerism , the existence of one reaction implies necessarily the existence of the enantiomeric one . the second eq .
( [ direct1 ] ) is identical to ( [ direct ] ) , but it can also include a single reaction going through a non - stable intermediate , and the third eq .
( [ direct2 ] ) is a single reaction with no intermediate species made explicit .
all these reactions are the basic transformations representing the chemical process known as racemization " .
the common effect of such reactions determines , in the case of closed systems with homogeneous energy distribution , that any initial @xmath1 value must decrease towards the unavoidable racemic mixture .
note that the three eqs .
( [ direct]-[direct2 ] ) lead to the decrease the total chiral matter , but the underlying dynamics does not involve a non - linear dependence on the racemic composition @xcite .
this non - linearity is expressed in deterministic kinetics by the dependence on the product of the concentrations of both enantiomers , and in stochastic kinetics by the existence of a non - elastic heterochiral collision between the enantiomers .
the rate constants in les are constrained by the principle of microreversibility @xcite : @xmath10 the les model has had a controversial reputation in the past because the constraints dictated by microreversibility or detailed balance , have not always been correctly considered nor properly taken into account @xcite . using eq.([lesconstraint ] ) one proves that les in a closed to mass flow system at uniform temperature can not lead to either temporary nor permanent chiral symmetry breaking : the racemic state is the only stable outcome @xcite . in order to overcome these microreversibility constraints
necessarily requires extending the reaction model via coupling to external energy sources and/or to external reagents .
the new reactions or energy fluxes thus introduced alter the overall set of ( original ) transformations and can allow for the ( partial ) lifting of the original microreversibility constraints .
thus for example , when the enantioselective and the limited enantioselective autocatalyses are individually _ localized _ within regions of low and hot temperatures , respectively , in a thermal gradient , mirror symmetry can be broken permanently @xcite . alternatively ,
when the reverse reaction of the non - enantioselective autocatalysis is driven by an external reagent , les in a uniform temperature can break mirror symmetry permanently @xcite .
both these modifications maintain les far from equilibrium and also lift the constraints eq.([lesconstraint ] ) on some of the reaction rates .
even so , this does not imply we can set any of the inverse rate constants to zero ( and we can not : doing so would violate eq .
( [ lesconstraint ] ) ) .
artifacts in mathematical modeling can and do arise when ( i ) reactions are approximated by irreversible transformations and especially when ( ii ) irreversible and reversible reactions are combined _ together _ in the same scheme @xcite , as was done in @xcite . note that the original frank scheme eqs([frank1953],[mutual ] ) involves only irreversible reactions ( understood as approximations ) , but here there is no constraint dictated by microreversibility : in this case both rates @xmath11 and @xmath12 are independent .
the rates for the modified frank model with reversible autocatalysis and mutual inhibition : eqs .
( [ lesscheme],[mutual ] ) , are also independent .
in contrast , the individual rate constants for _ reversible _ autocatalysis in concert with reversible non - catalytic production must obey @xmath13 we emphasize that , and as expressed by wegscheider s rule @xcite , the microreversibility constraint eq .
( [ constraint ] ) requires us to include _ both _ forward and inverse chemical reactions in the autocatalysis eq .
( [ lesscheme ] ) , since direct production eq .
( [ direct ] ) is taken to be reversible in @xcite .
that is , one reaction is reversible if and only if the other one is .
thus , we can not set @xmath14 in the presence of eq .
( [ direct ] ) . if however , we insist on combining irreversible autocatalysis @xmath14 with irreversible direct production @xmath15 @xcite , then the microreversibility constraint is satisfied consistently , and in the most trivial way , since @xmath16 and we are free to vary the forward reaction rates @xmath17 and @xmath18 independently .
but such an irreversible scheme @xcite corresponds to reactions under strict kinetic control .
on asymptotically long time scales , the inverse reactions become relevant and the closed mass system will necessarily racemize @xcite .
we consider the role of reaction noise in both well mixed and small spatially coupled systems when detailed balance is taken into account .
if we eliminate the reverse autocatalysis ( set @xmath14 ) from the following reactions @xmath19 we recover the scheme proposed in @xcite , which is itself a minor variation of the model in ref .
@xcite . when this reverse step is overlooked
, then the resultant reaction noise would appear to stabilize the homochiral states , provided a certain parameter @xmath20 .
here we include the obligatory reverse reaction as dictated by microreversibility , eq .
( [ constraint ] ) , and reconsider carefully the role of reaction noise on the stationary states of the system . in passing we
note that the models considered up to this point are variations of either the basic frank paradigm or of the les model , obtained by combining some elements of the former with some elements of the latter , now taking reversible steps , or instead taking irreversible steps , etc .
we thus approximate the scheme eqs .
( [ autocata ] , [ noncat ] ) by means of a stochastic differential equation for the time dependence of the enantiomeric excess @xmath21-[l]}{[d ] + [ l]}$ ] .
we consider a closed mass well - mixed system of volume @xmath22 and total number of molecules @xmath23 . taking the limit , @xmath24 , as in @xcite
, we arrive at the following equation for @xmath25 ( see appendix [ sec : stochastic ] ) : @xmath26 where @xmath27 is gaussian white noise with zero mean and unit variance . the normalized stationary distribution of eq . ( [ stochastic ] )
is given by @xmath28 we plot @xmath29 for various values of @xmath30 in fig [ fig1 ] .
the distribution @xmath29 is _ always _ peaked around the racemic state @xmath31 since the parameter @xmath32 .
as the total number of molecules @xmath23 increases , the distribution becomes ever more sharply peaked around @xmath31 .
in particular , the probability for homochiral states @xmath33 is strictly zero . stationary probability distribution for the chiral order parameter , eq .
( [ pstheta ] ) .
different values of @xmath34 correspond going from the broadest to the narrowest distribution .
@xmath35 for homochiral states .
, scaledwidth=45.0% ] the deterministic part of eq.([stochastic ] ) has one fixed point at the racemic state @xmath36 , in accord with stability analyses for the deterministic kinetic rate equations .
the amplitude of the noise is maximum for the racemic state , and vanishes at the homochiral states .
nevertheless , we can not arrange for the noise amplitude to be larger than that of the deterministic term , since @xmath37 ; see eq .
( [ pstheta ] ) .
this means that the racemic state is stable in the presence of reaction noise , and is surrounded by gaussian fluctuations that become negligible for increasing total number @xmath23 of molecules in the system , see fig [ fig1 ] and fig .
[ time - series ] .
temporal series for the enantiomeric excess @xmath1 and the chiral mass proportion obtained from gillespie simulations for different values of @xmath17 ( see legend ) . after a very brief transient ,
the curves fluctuate about the racemic state .
the parameters are : @xmath38 ( and hence @xmath39 ) , the number of molecules is 1000 ( initial condition is 10 l , 10 d , 980 a ) . the @xmath1 is defined as @xmath40-[d])/([l]+[d])$ ] and the chiral mass proportion as @xmath40+[d])/([l]+[d]+[a])$].,scaledwidth=30.0% ] as shown in fig .
[ time - series ] , stochastic simulations of the scheme eqs .
( [ autocata],[noncat ] ) using the gillespie algorithm @xcite reveal that the magnitude of the fluctuations about the racemic composition depend on the rate @xmath17 .
thus we observe that the reaction noise is somewhat more erratic for @xmath41 in comparison with the smoother fluctuations that result when @xmath42 .
note moreover the dependence of the total _
chiral mass proportion _ , defined as @xmath40+[d])/([l]+[d]+[a])$ ] : the fraction of total system mass which is chiral . increased non - catalytic production leads to a greater proportion of chiral matter .
the other rates were set to @xmath38 as in @xcite , and we include @xmath39 as dictated by microreversibility .
this implies that smaller @xmath17 thus leads to a greater recycling of the enantiomers back to achiral precursor via reverse autocatalysis , leading to smaller net chiral matter than when @xmath17 is large .
the racemizing tendency of the forward rate of non - catalytic production can also be appreciated in fig [ r1pp ] which shows the distribution in the enantiomeric excesses for different values of @xmath17 .
the greater the @xmath17 , the more sharply peaked is the distribution about the racemic outcome ; compare to fig [ fig1 ] .
distribution of the enantiomeric excess @xmath1 obtained from gillespie simulations for different values of @xmath17 ( see legend ) . after a brief initial transient , the distributions center about the racemic state .
the parameters are : @xmath43 , @xmath38 ( and hence @xmath39 ) , the number of molecules is 1000 ( initial condition is 10 l , 10 d , 980 a ) .
we obtain the @xmath44-[d])/([l]+[d])$ ] at @xmath45 .
we perform @xmath46 realizations .
binning is in intervals of @xmath47 in the enantiomeric excess .
, scaledwidth=35.0% ] the reaction scheme can be generalized and studied by spatially coupling a number of well - mixed systems . in the manner of @xcite , space can be discretized into a set of @xmath48 patches of volume @xmath22 , with the patches indexed by @xmath49 . here
we consider such spatial coupling " of @xmath50 well - mixed patches . within each such patch the reactions eqs .
( [ autocata],[noncat ] ) take place , with identical reaction rates for each patch , while all the molecules @xmath51 can diffuse from one patch to the other , with a common spatial coupling constant @xmath52 ( or , intra - patch diffusion " ) : @xmath53 stochastic simulations ( gillespie algorithm ) indicate that each patch racemizes independently .
the curves for the two patch system do not depend on the spatial coupling constant @xmath52 , see fig .
( [ two - patch ] ) . the distribution of the total enantiomeric excess obtained from gillespie simulations of a two - patch model for @xmath54 and different values of @xmath52 ( see legend ) .
asymptotically , the distributions center around the racemic state .
the parameters are : @xmath38 ( and hence @xmath55 ) , @xmath43 , the initial number of molecules per patch is 1000 ( 10 l , 10 d , 980 a ) and we obtain the total enantiomeric excess @xmath56-[d_1]+ [ l_2]-[d_2])/([l_1]+[d_1]+[l_2]+[d_2])$ ] at @xmath57 and perform @xmath58 realizations .
binning is in intervals of @xmath47 in enantiomeric excess .
spatial coupling applies to l , d , and a. for comparison , we show the distribution for a single uncoupled patch with @xmath54 ( 100000 realizations).,scaledwidth=35.0% ]
in order to maintain an irreversible self - replication " , the system in @xcite would have to be driven by an external source of energy to maintain the steady state of the system far from equilibrium .
this is a necessary , but not sufficient , condition for achieving smsb .
additional effort is required to include explicitly such a hypothetical source as an integral part of the overall model , and then to demonstrate its feasibility . indeed , perhaps the most challenging aspect of any proposal for modeling biological homochirality at the molecular level is not so much in the design of the intermediate reaction scheme itself , but rather in defining the very nature of the essential external energy source and its coupling to the reactions that comprise the intermediate chemical system .
the sought - after irreversible self - replication " corresponds to a unidirectional cyclic reaction . but
setting up such a cyclic reaction , by no means implies that the elementary autocatalytic reaction be irreversible " , i.e. , by simply putting @xmath14 .
the true cyclic behavior requires maintaining a steady unidirectional flow of matter in the system .
this unidirectional cycling behavior is a general property of steady states maintained by an energy flux . a brief review of the meaning of _ cycle _ , using onsager s original triangle reaction " as an example , @xcite is warranted .
consider a closed mass system held at constant temperature , containing species a , b and c reacting according to the scheme : @xmath59 at equilibrium there is no net flow around system since the forward and inverse reactions are in detailed balance ( @xmath60_{eq}$ ] denotes the equilibrium concentration ) : @xmath61_{eq } - k_2[b]_{eq } = k_3[b]_{eq } - k_4[c]_{eq } = k_5[c]_{eq } - k_6[a]_{eq } = 0.\ ] ] the principle of microreversibility implies the following wegscheider condition for the rate constants : @xmath62 an external energy source could drive a net flow of material around the system , brought about by a flow of energy from a high potential source to a low potential sink , passing through the closed mass intermediate system of eq.([onsager ] ) .
the steady state condition would then require that @xmath63 - k_2[b ] = k_3[b ] - k_4[c ] = k_5[c ] - k_6[a ] = { \cal f } > 0,\ ] ] where @xmath64 is the flow , the rate at which material is cycling around the system .
this flow of material around a closed reaction loop is what is meant by a cycle @xcite .
note that this cycle is irreversible : the net matter flow is unidirectional .
we emphasize two important points : ( i ) eq.([steadyflow ] ) involves the non - equilibrium concentrations of the species involved , and ( ii ) the unidirectional matter flow @xmath64 depends on _ all _ the forward and reverse reaction rates @xmath65 .
the cycle is _ not _ established by simply putting the reverse rates to zero ( this is prohibited by the wegscheider condition eq .
( [ microreverse ] ) ) , but rather from an energy flow that traverses the closed mass system . by way of example
, morowitz offers a kinetic model for unidirectional cycles in onsager s network using photochemical reactions @xcite . in sec [ sec : results ] , we demonstrated that the reaction scheme eqs .
( [ autocata],[noncat ] ) when obeying eq .
( [ constraint ] ) , does not lead to smsb , regardless of the inclusion of reaction noise .
we can legitimately circumvent this latter constraint by going to an out - of - equilibrium scenario . to assess whether irreversible cycling can lead to smsb
, we consider this scheme in a uniform temperature distribution driven by a constant concentration of external reactants , x and y. see eq .
( [ drivecata ] ) , an open system with x and y matter exchange with the surroundings , and depicted in fig [ cycle ] .
the resultant reaction network is cyclic one with permanent consumption and production of y and x ( or of x and y , depending on the flow direction in the cycle ) : @xmath66 the unidirectional cyclic network discussed in the text is composed of the _ micro - reversible _ reactions eqs.([drivecata ] , [ noncat2 ] ) in a uniform temperature distribution with y and x matter exchange with the environment .
a is an achiral compound and d and l the enantiomer pair of a chiral compound .
the cycle is driven by the external reagents.,scaledwidth=50.0% ] it is straightforward to carry out a dynamic stability analysis for this system .
the presence of external constant concentration reactants now lifts the constraint eq .
( [ constraint ] ) , and the far - from - equilibrium reaction model depends on the _ two _ independent parameters : @xmath67}{k_2[y]}.\ ] ] because the reactants are external to the intermediate system , we can control the matter flow , e.g. , in the forward sense , see fig .
[ cycle ] : @xmath68[y ] - k_{-2}[l][x ] & = & { \cal f}_l > 0,\\ k_2[a][y ] - k_{-2}[d][x ] & = & { \cal f}_d > 0.\\\end{aligned}\ ] ] the matter flow @xmath69 is partitioned @xmath70 , among the @xmath71 or @xmath72 autocatalytic branch of the reaction network , respectively .
nevertheless , even with the enantioselective autocatalyses driven _ unidirectionally _ in this way ( @xmath73 ) , a stability analysis proves that the racemic state is the only stable outcome , and for all @xmath74 . by marked contrast , whereas detailed balance implies the racemic outcome for the les model in a closed - mass system at uniform temperature , driving les by external reagents can lead to smsb @xcite . the crucial fundamental difference between eqs([drivecata],[noncat2 ] ) and les driven by external reagents @xcite is in the inverse limited enantioselective catalytic step eq.([limited ] ) , which implies a _ non - linear chiral inhibition _ between the two enantiomers @xcite .
the reaction models analyzed in this paper have served as useful vehicles for examining a number of basic issues relevant for framing proposals , coherent with fundamental chemical and physical principles , aimed at modeling biological homochirality at the molecular level .
we summarize our main findings here .
the following points are , to some extent , inextricably interrelated . * .
thermodynamics dictates that the enantioselective autocatalysis and direct production / decay of the enantiomers must have identical ratios of the forward and reverse reaction rate constants , regardless of whether the system is in equilibrium or far from it .
once detailed balance is accounted for correctly , we have proven , by employing standard methods ( stochastic differential equations , the fokker - planck equation ) and numerical simulations , that the resultant model including reaction noise can never break chiral symmetry . on the contrary ,
the final stable outcome is always the racemic state . and
, this holds whether the system is well - mixed or coupled spatially . * .
we have proven that the presence of reaction noise does not lead to any new final stable state not already in accord with the stability analysis of the deterministic model .
note : the regime where stochastic kinetics is expected to be important corresponds to the case of small volumes and _ small _ numbers @xmath23 of molecules . in deriving our stationary probability distribution
, we take the limit @xmath24 , as do the authors of ref . @xcite . in this
limit , reaction noise has only a minor quantitative , but not qualitative , effect . * .
we have argued that the linear decay reaction of eq.([direct ] ) can not act as a mutual inhibition stage for smsb when coupled to a first - order enantioselective autocatalysis .
this strongly suggests , that a chiral inhibition reaction , or a set of coupled reactions generating a chiral inhibition dynamics , must have a _
non - linear _ chiral dependence @xcite , as is the case in both the les and frank models . * .
we have shown explicitly how to establish a unidirectional net flow of matter in the reversible autocatalytic reaction eq.([drivecata ] ) , by coupling it to external reagents . in spite of this
, a stability analysis proves that the manifestly out - of - equilibrium model eqs.([drivecata],[noncat2 ] ) leads inexorably to the racemic state . and
this result is intimately related to the absence of chiral inhibition .
there is a widespread and active research effort devoted to understanding the origins of biological homochirality that crosses the traditional boundaries between physics , chemistry and biology .
the fundamental concepts treated here deserve careful consideration in scenarios for candidate reaction schemes proposed as models for the emergence of biological homochirality .
the research of jmr and dh is supported in part by the coordinated project ctq2013 - 47401-c2 - 1/2-p ( mineco ) .
ms , jmr and dh form part of the cost action cm1304 on emergence and evolution of complex chemical systems " .
we cast the fully reversible kinetic scheme defined by eqs .
( [ autocata],[noncat ] ) in terms of stochastic differential equations to quantify the role played by internal reaction noise .
the mapping of chemical reactions to master equations and then on to fokker - planck ( fp ) equations is an established technique @xcite , as is the correspondence of fp with stochastic differential equations . defining the state vector @xmath75
where @xmath76 denote the time - dependent concentrations of molecules a , d and l , respectively , we find that our scheme may be approximated by the stochastic differential equation ( defined in the ito sense ) @xcite : @xmath77 where @xmath78 @xmath79 an the @xmath80 @xmath81 are gaussian white noises with zero mean and correlation , @xmath82 @xmath22 is the system volume .
the rate of inverse autocatalysis is _ not _ an independent variable , but obeys the constraint : @xmath83 the number of chemical degrees of freedom @xmath84 can be effectively reduced from three to one @xcite .
this is so because firstly , the total number of molecules is conserved by our reaction scheme , thus so is the total concentration @xmath85 .
secondly , the total chiral matter @xmath86 is a _ fast degree of freedom _ relative to the enantiomeric excess @xmath25 @xcite .
simulations of the fully reversible scheme eqs .
( [ autocata],[noncat ] ) using the gillespie algorithm @xcite confirm that @xmath87 approaches a stable fixed point value surrounded by small gaussian fluctuations ( see , e.g. , fig [ time - series ] ) .
we therefore substitute @xmath88 into the equation for @xmath89 derived below .
we thus carry out the change of variables on eq.([ito ] ) : @xmath90 employing ito s formula @xcite : @xmath91_{ij } \partial_i \partial_j f(\vec{x } ) ] dt + \sum_{ij } \bm{g}(\vec{x})_{ij } \partial_i f(\vec{x } ) dw_j(t).\ ] ] from eq .
( [ ito - change ] ) it is straightforward to demonstrate that @xmath92 is identically zero , as it must be . from @xmath93
we solve for the fixed point @xmath94 : @xmath95}}{2k_a + ( 1 + \bar{\theta}^2)k_{-a}},\ ] ] since @xmath96 , we take the positive root . note the total chiral matter @xmath97 depends _ weakly _ on the most probable stationary value @xmath98 for the chiral order parameter .
the most probable value of @xmath99 is determined from the stochastic differential equation for @xmath89 .
we prove below that _ self - consistency _ requires taking @xmath100 in eq .
( [ chi - star ] ) .
we derive the stochastic equation obeyed by @xmath89 and substitute @xmath101 into this equation .
we express the result in terms of the total number of molecules @xmath102 and for @xmath24 .
the enantiomeric excess or chiral order parameter @xmath25 obeys the equation @xmath103 where @xmath27 is gaussian white noise with zero mean and unit variance . from the fokker - planck equation corresponding to eq ( [ theta - eq ] ) we readily solve for the steady state probability distribution @xmath29 for @xmath25 @xcite .
we find @xmath104 and the normalization constant @xmath105 where @xmath106 is the hypergeometric function @xcite . from @xmath107
we conclude ( see fig [ fig1 ] ) that the most probable value for the chiral order parameter is @xmath108 , corresponding to the racemic state , thus establishing the self - consistency of employing this value in eq .
( [ chi - star ] ) .
systems capable of smsb yield large net chiral excesses , mostly in the vicinity of homochirality , as common non - thermodynamic states @xcite .
for example , in closed systems with relatively high exergonic transformations , prolonged chiral excursions i.e. , kinetically controlled chiral outcomes , are possible .
even final stationary chiral states are possible in either open or in closed systems unable to come into thermodynamic equilibrium with their surroundings ; see , for example , refs .
. this amplification does not correspond to the term as used in chemical asymmetric synthesis , because in the absence of any external chiral force or polarization , the racemic mixture is a metastable , but not the most stable , state of the system .
frank , on spontaneous asymmetric synthesis , biochim .
et biophys .
acta * 11 * ( 1953 ) 459 .
r. plasson , d.k .
kondepudi , h. bersini , a. commeyras and k. asakura , emergence of homochirality in far - from - equilibrium systems : mechanisms and role in prebiotic chemistry , chirality * 19 * , 589 - 600 ( 2007 ) . c. blanco , j.m rib , j. crusats , z. el - hachemi , a. moyano and d. hochberg , mirror symmetry breaking with limited enantioselective autocatalysis and temperature gradients , phys .
phys . * 15 * , 1546 ( 2013 ) . c. blanco , j. crusats , z. el - hachemi , a. moyano , d. hochberg and j.m .
rib , spontaneous emergence of chirality in the limited enantioselective model : autocatalytic cycle driven by an external reagent , chemphyschem * 14 * , 2432 - 2440 ( 2013 ) .
such a non - linear coupling between competing species is present in any type of model for biological survival / depletion of competing species : see j.m .
rib , d. hochberg , competitive exclusion principle in ecology and absolute asymmetric synthesis in chemistry , chirality * 27 * , 722 - 727 ( 2015 ) . | applying the constraints dictated by the principle of detailed balance , we analyze a recent proposal for spontaneous mirror symmetry breaking ( smsb ) based on enantioselective autocatalysis coupled to a linear decay of the enantiomers and in the presence of reaction noise .
we find the racemic state is the final stable outcome for both deterministic as well as for stochastic dynamics , and for both well - mixed and small spatially - coupled systems .
the racemic outcome results even when the autocatalytic cycles are driven irreversibly by external reagents , in manifestly non - equilibrium conditions .
our findings suggest that first - order autocatalysis coupled to reactions involving _ non - linear _ heterochiral dynamics is a necessary pre - condition for any mechanism purporting to lead to molecular homochirality . |
clearly , the high energy @xmath0-ray emission is an important piece in the blazar puzzle because the @xmath0-ray observations of blazars provide a new probe of dense radiation field released through accretion onto a supermassive black hole in the central engine ( bregman 1990 ) .
the energetic gamma ray experiment telescope ( _ egret _ ) which works in the 0.110gev energy domain has now detected and identified 66 extragalactic sources in 3th catalog ( mukherjee et al 1999 ) .
all these objects are blazar - type agns whose relativistic jets are assumed to be close to the line of sight to the observer . it seems unambiguous that the intense gamma - ray emission is related with highly relativistic jet .
it has been generally accepted that the luminous gamma - ray emission is radiated from inverse compton , but the problem of seed photons remains open for debate .
the following arguments have been proposed : ( 1 ) synchrotron photons in jet ( inhomogeneous model of synchrotron self compton ) ( maraschi , ghisellini & celotti 1992 ) ; ( 2 ) optical and ultraviolet photons directly from the accretion disk ( dermer & schlikeiser 1993 ) ; ( 3 ) diffusive photons in broad line region ( blr ) ( sikora , begelman & rees 1994 , blandford & levinson 1995 ) ; ( 4 ) the reflected synchrotron photons by electron mirror in broad line region , namely , the reflected synchrotron inverse compton ( rsc ) ( ghisellini & madau 1996 ) . these mechanisms may operate in different kinds of objects , however there is not yet a consensus on how these mechanisms work . also it is not clear where the @xmath0-ray emission is taking place largely because of uncertainties of soft radiation field in the central engine . on the other hand , vhe observations ( kerrick et al 1995 ,
chadwick et al 1999 , roberts et al 1999 , aharonian et al 1999 ) are making attempts to explore the radiation mechanism because they may provide some restrictive constraints ( begelman , rees & sikora 1994 , mastichiadis & kirk 1997 , tavecchio , maraschi & ghisellini 1998 , coppi & aharonian 1999 , harwit , protheroe and biermann 1999 ) .
based on the simple version of ssc model , stecker , de jager & salamon ( 1996 ) predicted a large number of low redshift x - ray selected bl lacs as tev candidates , taking into account that the presence of intergalactic infrared radiation field including cosmic background leads to strong absorption of tev photons from cosmological emitters ( stecker & de jager 1998 ) .
it is suggested to form an extended pair halo in cosmological distance due to the _ external _ absorption ( aharonian , coppi , & voelk 1994 ) .
however , so far only three x - ray selected bl lacs have been found to be tev emitters by whipple telescope ( @xmath2gev ) , in addition , photons higher than 0.3tev in the x - ray - selected pks 2155 - 304 with redshift @xmath3 has been detected photons 0.3tev by durham mrk 6 telescope ( chadwick et al 1999 ) .
the recent measurements of intergalactic infrared field is quite different from the previous observations ( madau et al 1998 , steidel 1998 ) .
although this _ external _ absorption is definitely important , the critical redshift @xmath4 beyond which cosmological back ground radiation and intergalactic infrared fields will absorb vhe photons remains uncertain .
especially the recent vhe observations show that mrk 501 emits 25 tev photons ( aharonian et al 1999 ) .
evidently this suggests that the _ external _ absorption can not efficiently attenuate the vhe photons from reaching us across distances of 100 mpc .
it is highly desired to accurately probe the star formation rate in order to determine the critical redshift @xmath4 .
thus it seems significant to study the _ intrinsic _ mechanism for the deficiency of tev photons from @xmath5-ray loud agns disregarding the absorption by intergalactic infrared radiation field
. a larger lorentz factor of the jet implies higher density of the external photons in the blob , if the reflection of clouds in broad line region works , and therefore stronger absorption of high energy @xmath5-rays ( celotti , fabian & rees 1998 ) .
here we argue based on the hypothesis of ghisellini & madau ( 1996 ) that the energy density of reflected synchrotron photon is high enough for pair production via interaction of gamma - ray photons by inverse compton scattering with reflected synchrotron photons if the bulk velocity is high enough .
further we apply the present constraint to the representative individual objects , mrk 421 and 3c 279 .
ghisellini & madau ( 1996 ) have calculated the energy density of reflected synchrotron ( rsy ) emission , and compared with the other reflected components .
they draw a conclusion that the energy density of rsy component dominates over 10 times of that of reflected component of accretion disk radiation . in this section
we make an attempt to use the observables quantities to express the intrinsic constraints on very high energy emission .
the overall @xmath6 spectrum of blazars shows that there are two power peaks : the first is low energy one between ir / soft x - ray band , and the second is high energy one peaking in the mev / gev range ( von montigny et al 1995 , sambruna , maraschi & urry 1996 , comastri et al 1997 , kubo et al 1998 ) .
this characteristic can be explained by the simple context of one - zone homogeneous ssc or ec model .
the low energy peak denoted @xmath7 is caused by synchrotron radiation of relativistic electrons , and the second peak denoted @xmath8 , or @xmath9 results from the compton scattering off the synchrotron or reflected synchrotron photons by the same population of electrons , respectively .
we take the two peaks and their corresponding fluxes as four observable quantities . from the rsc model
the magnetic field @xmath10 can be approximately expressed by the observational quantities .
the observed frequency of synchrotron photon is @xmath11 ( @xmath12 ) , and the frequency of reflected synchrotron compton photons reads @xmath13 , and we can get the estimation of magnetic field @xmath10 @xmath14 while in pure ssc model the magnetic field is approximately as @xmath15 , where @xmath8 is the frequency of photons emitted by ssc . the doppler factor @xmath16 $ ] , where @xmath17 is the cosine of the orientated angle of jet relative to the observer .
equation ( 1 ) is similar to the model of sikora , begelman & rees ( 1994 ) ( also see sambruna , maraschi & urry 1996 ) . comparing with the above two formula
, we learn that rsc model needs stronger magnetic field than ssc model does whereas the energy of relativistic electrons is lower in rsc model than in ssc model . the reflected synchrotron compton ( rsc )
mainly depends on two parameters : the reflection albedo , namely , the thomson scattering optical depth ( @xmath18 ) , and the lorentz factor @xmath19 of the relativistic jet . in the case of power - law distribution of electrons ,
@xmath20 , ( @xmath21 ) , where @xmath22 is the number density of relativistic electrons , and @xmath0 is the lorentz factor of electron , the synchrotron emission coefficiency is approximately given by @xmath23 .
here @xmath24 , and @xmath25 is tabulated in pacholczyk ( 1970 ) within the frequency range @xmath26 , where @xmath27 .
the average energy density per frequency @xmath28 in a region with dimension @xmath29 in the jet comoving frame can be obtained @xmath30 where @xmath31 .
the number density of synchrotron photons can be obtained by @xmath32 and @xmath33 reads @xmath34 here @xmath35 is planck constant , and @xmath36 .
we have employed relationship @xmath37 to derive equation ( 3 ) .
the mean energy density is expressed by @xmath38 for @xmath39 . defining @xmath40 as @xmath41 we have @xmath42 where @xmath43 denotes @xmath44 .
since the opening angle of jet ( @xmath45 ) is much less than @xmath46 , it is then reasonable to assume that the blr reflection approximates to plane mirror with thickness @xmath47 and electron number density @xmath48 .
the distance distribution of reflected synchrotron photons has been discussed by ghisellini & madau ( 1996 ) . the angular distribution has not been issued .
since the thickness of mirror is zero , the energy density of reflected synchrotron emission sharply increases when blob is close to the mirror . in fact if we drop the assumption of zero - thickness of mirror , this characteristic will disappear .
we will deal with this sophisticate model in future . because the reflected synchrotron emission is isotropic in observer s frame , the blob receives the reflected photon beamed within a solid angle @xmath45 .
the subsequent section will pay attention to this effects . neglecting the angle - dependent distribution of reflected photon field , we approximate the doppler factor @xmath49 ( @xmath50 ) . for simplicity
, we assume the mirror ( reflecting clouds in broad line region ) has thomson scattering optical depth @xmath51 .
the received photon density @xmath52 in the jet comoving frame can be then approximately written as @xmath53 where we use @xmath54 ( there are two doppler shifts due to mirror effects ) .
the convenient form of cross section of photon - photon collision is ( coppi & blandford 1990 ) @xmath55 where @xmath56 is the usual @xmath56-function .
this approximation is only valid for case of isotropic radiation field .
the rsy component is seen by the blob within the solid angle @xmath57 .
although the cross section of photon - photon interaction holds , the interacting possibility among photons reduces by a factor of @xmath58 due to the beaming effects , which effectively reduces the opacity .
thus the pair production optical depth for photon with energy @xmath59 reads @xmath60 here @xmath61 is @xmath62 we will show the validity of the above approximation in the next subsection .
supposing that rsc operates efficiently in @xmath5-ray loud blazars , we can get @xmath63 from the observations @xmath64 from equation ( 11 ) we have @xmath65 and @xmath66 from equation ( 11 ) we know that the observed @xmath63 represents the reflection ratio and doppler factor of jet motion as long as the compton catastrophe does not occur .
if we set @xmath67 , we get @xmath68 for @xmath69 .
this value is the lowest one in the model of sikora , begelman & rees ( 1994 ) who suggest @xmath70 .
in fact we can roughly adopt @xmath18 as the covering factor which is usually taken to be 0.1 in fitting the broad emission line by photoionization model .
inserting @xmath10 and @xmath71 [ eqs(1 ) and ( 6 ) ] into @xmath72 ( eq .
10 ) , and letting @xmath73 and @xmath74 , we have the pair production optical depth for @xmath75 in the observer s frame @xmath76 where @xmath77 ( @xmath78 for @xmath79 ) .
there are five observational parameters : @xmath7 , @xmath80 , @xmath81 , @xmath82 and @xmath63 ; and the unknown doppler factor @xmath83 . for the typical value of parameters , @xmath84 , @xmath85hz , and @xmath86hz ,
@xmath87 day , and @xmath88 , we have @xmath89 where @xmath90 .
figure 1 shows the opacity due to pair production of photons with very high energy encountering with the reflected synchrotron photons .
the equation ( 15 ) tells us the constraints on vhe from jet : ( 1 ) smaller @xmath63 , i.e. stronger reflection , will leads to the absorption of tev photons .
this parameter represents the energy density reflected by the blr cloud including the bulk relativistic motion . from this estimation
we know that tev photon will be absorbed by the reflected synchrotron photons provided that @xmath91 .
( 2 ) @xmath92 is sensitive to @xmath7 and @xmath80 .
( 3 ) @xmath92 is proportional to @xmath93 , in contrast to the usual down - limit ( see mattox et al 1993 , and dondi & ghisellini 1995 ) , providing the upper limit doppler factor of bulk motion from @xmath94 , @xmath95 this is a new constraint , which is expressed by the observational quantities .
it lends us a simple and efficient way to select tev candidates from known blazars in term of their known characteristics .
the received photons reflected by blr in comoving frame is anisotropic , therefore , the pair opacity should be carefully treated .
we have made important approximations that the blr is thought to be a plane mirror and treated the photon - photon interaction in an approximate way .
now let us show the validity of this approximation .
we adopt the geometry shown in fig 1c of ghisellin & madau ( 1996 ) .
they show that the energy density of reflected synchrotron photon strongly depends on the location of emitting blob .
we should admit that the aximal symmetry holds in the reflected synchrotron emission .
we approximate the thomson scattering event by isotropic scattering with cross section @xmath96 and neglect recoil , which is a very good approximation when @xmath97 . the angular distribution of reflected synchrotron emission is given by @xmath98 ( @xmath99 is a constant ) , the function @xmath100 determines the angular distribution of reflected synchrotron photons ( ghisellini & madau 1996 ) @xmath101 where @xmath102^{-2}$ ] , and @xmath103 , where @xmath104 is the distance of blob to the center .
figure 2 shows the angular distribution in blob comoving frame .
it can be seen that the geometry effect of reflecting mirror isotropizes the radiation at some degrees , but the beaming effect still dominates .
it is still a good approximation that the radiation is beamed with a cone of solid angle @xmath45 .
thus the pair opacity can be written as ( gould & schrder 1967 ) @xmath105 where @xmath106 and the photon - photon cross section @xmath107 reads @xmath108,\ ] ] where @xmath109 is the speed of the electron and positron in the center of momentum frame @xmath110^{1/2}$ ] . performing the integral we have @xmath111 where @xmath112 reads @xmath113 with @xmath114 and @xmath115 is the integral of @xmath116 over the entire broad line region , which can be evaluated as @xmath117 + \sin \theta_0 \ln \left[\frac{\sin \phi_1}{\sin \phi_2}\right ] \right\},\end{aligned}\ ] ] with @xmath118 , @xmath119 , and @xmath120 .
the function @xmath112 is plotted in fig 3 . since the beamed radiation field reduces the effective cross section of photon - photon interaction by a factor @xmath121 , it would be convenient to check our approximation by the quantity @xmath122 .
we can easily find that it is close to [email protected] when @xmath124 , suggesting our approximation is accurate enough .
it should be pointed out that the present treatments of reflected synchrotron radiation can be conveniently extended to the inclusion of the radiation from the secondary electrons if we further study the pair cascade in the future .
the last two subsections are devoted to the _ internal _ absorption of tev photons , the developments of pair cascade due to the present mechanism will be treated in a preparing paper ( wang , zhou & cheng 2000 )
. however it would be useful to compare the dimensions and radiation of the pair cloud due to the _ internal _ absorption and the pair halo suggested by aharonian , coppi , & veolk ( 1994 ) , who argue the formation of pair halo due to the interaction of tev photons from agns with infrared photons of cosmological background radiation .
this _ external _
absorption produces pairs , which are quickly isotropized by an ambient random magnetic field , forming a extended halo of pairs with typical dimension of ( @xmath125mpc ) . without specific mechanism
we know that the time scale of halo formation is of about @xmath126 yr .
usually this absorption is regarded as the main mechanism of deficiency of tev emission from _
egret_-loud blazars ( stecker & de jager 1998 ) .
let us simply estimate the scale of pair halo before it is isotropized by the ambient magnetic field .
assuming the intergalactic magnetic field @xmath127 gauss , then the mean free path of pair electrons in halo reads @xmath128 the initial halo is of such a dimension , which is much larger than that of intrinsic absorption case .
aharonian , coppi , & volk ( 1994 ) have suggested some signatures of such an extended halo , especially for the light curves in high energy bands ( coppi & aharonian 1999 ) .
anyway this is much larger than that of the present internal pair cloud .
thus it is easier to distinguish the two cases .
we have set a new constraint on the very high energy emission in term of observable quantities . as the applications of the present model
, we would like to address some properties of very high energy from blazars .
the broadband continuum of blazars show attractive features which indicate the different processes powering the objects .
the ratio @xmath129 of @xmath0-ray luminosity to optical in flat spectrum radio quasars ( fsrqs ) is quite different from that in bl lacs ( dondi & ghisellini 1995 ) .
comastri et al ( 1997 ) confirmed this result in a more larger samples and found this mean ratio is roughly of unity in bl lacs and @xmath130 , namely @xmath131 in fsrqs .
ghisellini et al ( 1993 ) , using the classical limit of ssc model , show that there is a systematical difference in doppler factors @xmath83 between bl lacs and core - dominated quasars , @xmath132 for bl lacs and @xmath133 for core - dominated quasars .
these differences have been confirmed by gijosa & daly ( 1996 ) who assume that the particles and magnetic field are in equipartition .
this difference would lead to more prominent difference of reflected synchrotron photon energy density , suggesting a different mechanism in these objects .
the two systematically different features in @xmath63 and doppler factor @xmath83 strongly suggest that the different mechanism of @xmath0-ray radiation may operate in these objects . from eq.(15 ) it is believed that the deficiency of tev emission in radio - loud quasars may be _ intrinsic _ due to the present mechanism .
whipple observatory had ever searched for tev gamma - ray emission for 15 egret - agns with low redshift , but only mrk 421 has positive signal(kerrick et al 1995 ) . even at present stage only three x - ray
selected bl lacs have been reported as tev emitters , mrk 421 , mrk 501 and 1es2344 + 514 ( cataness et al 1997 ) , and pks 2155 - 304 is a potential tev emittor ( chadwick et al 1999 ) .
we can apply the present model to the two representative sources : mrk 421 and 3c 279 for specific illustration .
_ mrk 421 : _ this is an x - ray selected bl lac object , and has been detected gev @xmath0-ray emission by egret ( lin et al 1992 ) , and the first tev emission by whipple ( punch et al 1992 ) .
it has been extensively and frequently observed by telescopes from radio to tev bands(kerrick et al 1995 , macomb et al 1995 , takahashi et al 1996 , krennrich et al 1999 ) .
tev observations of mrk 421 by whipple show that the tev photon did not flare much more dramatically than the x - rays , suggesting that the enhanced high - energy electrons were scattering off a part of the synchrotron spectral energy distribution that remained constant ( takahashi et al 1996 ) . roughly speaking this object
satisfies the energy equipartition for the two power peaks(zdziarski & krolik 1993 , macomb et al 1995 ) , suggesting @xmath134 ( macomb et al 1995 ) , and pure ssc model agrees with the observations ( krennrich et al 1999 ) , suggesting @xmath135 .
this indicates the rsc process is not important .
the synchrotron component peaks in luminosity at uv to soft x - ray energies and continues into kev x - rays(maraschi , ghisellini & celotti 1994 ) .
the gamma - ray emission extends from 50 mev to an astounding tev .
data combined over several periods ( lin et al 1992 ) reveal a hard gev spectrum ( @xmath136 ) by egret and a steeper one at tev energies ( @xmath137 ) from the whipple observatory ( schubnell et al 1994 ) , implying a spectral break .
the multiwavelegenth spectrum shows that @xmath138hz , @xmath139hz , and @xmath140 ( macomb et al 1995 , kubo et al 1998 ) .
the shortest timescale of @xmath0-ray variability is about @xmath141 minutes ( gaidos et al 1996 ) .
if we take @xmath88 and @xmath142 , we have @xmath143 for tev photons .
this means @xmath144 , which is consistent with pure ssc model . from this
we can estimate the scattering medium @xmath145 [ @xmath146 when we take @xmath147 and @xmath148 , suggesting the mirror effects can be ruled out in this object .
this result agrees to the absence of any evident emission lines in mrk 421 .
interestingly , celotti , fabian & rees ( 1998 ) have suggested from rapid tev variability of mrk 421 that its accretion rate is lower than @xmath149 eddington rate .
they thus propose that advection - dominated accretion flow or ion pressure supported tori ( ichimaru 1977 , rees et al 1982 , narayan & yi 1994 ) may power the luminosity . _
3c 279 : _ this is a typical fsrq .
the first whipple observation shows negative signal in tev ( kerrick et al 1995 ) , however the multiwavelength simultaneous observations by _ egret _ , _ asca _ , _ rxte _ , _ rosat _ , _ iue _ in 1996 january - february show an intensive flare with very flat spectrum in _ egret _ band ( wehrle et al 1998 ) , showing @xmath150 at the high state of @xmath0-ray emission .
it has been argued that rsc may explain the 1996 gamma - ray flare ( wehrle et al 1998 ) .
the observed flare show that the synchrotron emission peaks at @xmath151hz , @xmath152hz , @xmath153 , and @xmath1548hr ( wehrle et al 1998 ) .
if we take @xmath88 , then we have @xmath155 .
therefore it is expected that no tev emission occurs in this object .
however it is interesting to note that this is due to the _ intrinsic _ mechanism .
we hope that there will be some effects due to the presence of pair production in the vhe flare ( wang , zhou & cheng 2000 ) . the q1633 + 382 ( mattox et al 1993 ) is quite similar to 3c 279 , but it shows much smaller @xmath156 .
the strong reflected synchrotron photons as seed photons may appear in this object , however its redshift ( @xmath157 ) is too large to detect vhe photons due to the absorptions of back ground photons .
here we suggest that the deficiency of vhe emission may be _
it is expected to make simultaneous observations at other bands to test its light curves in order to reach a decision .
the present paper focuses our attention on the effects of blr mirror on the attenuation of @xmath0-ray in blazars .
the mirror effect mainly depends on two parameters : lorentz factor of the bulk motion ( @xmath19 ) and the thomson scattering depth ( @xmath18 ) of broad line region .
based on the calculations , we would like to draw the conclusions : \1 ) the parameters @xmath63 and doppler factor @xmath83 in fsrqs are systematically greater than that in bl lacs .
this will cause the more stronger `` _ intrinsic _ '' absorption of vhe photons in fsrqs than that in bl lacs .
it is predicted that there is general absence of very high energy emission in fsrqs , owning to the attenuation of vhe photons by the blr reflection of synchrotron emission .
\2 ) the mirror model provides a new constraint on relativistic bulk motion .
intrinsic _ absorption of tev photons may operate in some objects , especially in fsrqs .
this constraint is cause by the motion of blob itself . although the origin of @xmath0-ray emission in blazars still remains open , vhe observations sets strong constraints on blazar s radiation mechanism .
these constraints are : ( 1 ) brightness temperature exceeding the kellermann - pauliny - toth ( begelman et al 1994 ) , ( 2 ) multiwavelength light curves based on the homogeneous model ( mastichiadis & kirk 1997 ) , ( 3 ) high energy variations in x - ray and @xmath0-ray including interaction with background ir radiation ( coppi & aharonian 1999 ) .
these constraints are mainly based on ssc model .
the deficiency of vhe photons from high redshift @xmath5-ray loud blazars may be explained by the interaction of the cosmic background radiation fields with the vhe photons .
however the possible alternative mechanism may be due to the _ intrinsic _ attenuation by the reflected synchrotron photons .
three bl lac objects have been found to show @xmath158 and @xmath159 emission , indicating the existence of broad line region in these objects ( vermeulen et al 1995 , corbett et al 1996 ) , even mrk 421 has been detected weak luminosity of a broad emission line ( morgani , ulrich & tadhunter 1992 ) .
the increasing evidence of the presence of broad emission lines in bl lacs lends the possibility that the reflected external photons might be the main source of seed photon in this kind of blazars . distinguishing the two different mechanisms
might be traced by the following - up observations in other wavebands because a pair cascade process may be developed , forming a pair halos in the _ external _ absorption ( aharonian , coppi & voelk 1994 ) .
such an extended halo due to _ external _ absorption may be of very long variable timescale at least @xmath160 yr ( corresponding to one mean free path ) [ see equation ( 24 ) ] . however if the _
intrinsic _ absorption works in the central engine , the time - dependent synchrotron self - compton model including pair cascade ( wang , zhou & cheng 2000 ) could predict the interesting spectrum and light curves , which may interpret the variations of pks 2155 - 304 ( urry et al 1997 ) .
the other radiative properties of such an extended pair halo are needed to be studied in order to distinguish the _ intrinsic _ absorption from the _ external _ one .
aharonian , f.a . ,
coppi , p.s . & voelk , h.j . , 1994 , , 423 , l5 aharonian , f.a .
, et al for the hegra collaboration 1999 , , 349 , 29 begelman , m. , rees , m.j . & sikora , m. , 1994 , , 429 , l57 blandford , r.d . & levinson , a. , 1995 , , 441 , 79 bregman , j. , 1990 , , 2 , 125 cataness , m. , et al , 1997 , , 487 , l143 chadwick , p.m. , et al 1999 , , 513 , 161 celotti , a. , fabian , a.c . & rees , m.j . , 1998 , , 293 , 239 comastri , a. , fossati , g. , ghisellini , g. & molendi , s. , 1997 , 480 , 534 coppi , p.s . & aharonian , f. , 1999 , apj , 521 , l33 coppi , p.s .
& blandford , r.d . , 1990 , , 245 , 453 corbett , e.a . , robinson , a. , axon , d.j . ,
hough , j.h . ,
jeffries , r.d . ,
thurston , m.r . & young , s. , 1996 , , 281 , 737 dermer , c.d .
& schlikeiser , r. , 1993 , , 416 , 458 dondi , l. & ghisellini , g. , 1995 , , 273 , 583 harwit , m. , protheroe , r.j . ,
& biermann , p.l .
, , 524 , 91 gaidos , j.a . , et al , 1996 , , 383 , 319 ghisellini , g. & madau , p. , 1996
, , 280 , 67 ghisellini , g. , padovani , p. , celotti , a. , maraschi , l. , 1993 , , 407 , 65 gould , r.j .
& schrder , g.p . , 1967 ,
, 155 , 1404 gijosa , a. & daly , r.a . , 1996 , , 461 , 600 ichimaru , s. , 1977 , apj , 214 , 840 kerrick , a.d . , et al , 1995 , , 452 , 588 krennrich , f. et al , 1999 , , 511 , 149 kubo , h. et al , 1998 , , 504 , 693 lin , y.c .
, et al , 1992 , , 401 , l61 macomb , d.j .
et al , 1995 , , 449 , l99 madau , p. , pozzetti , l. , & dickinson , m. , 1998 , , 498 , 106 maraschi , l. , ghisellini , g. & celotti , a. , 1992 , , 397 , l5 maraschi , l. , ghisellini , g. & celotti , a. , 1994 , in multiwavelength continuum emission of agn .
courvoisier & a. bleche ( dordrecht : kluwer ) , 221 mastichiadis , a. & kirk , j.g .
, 1997 , , 320 , 19 mattox , j.r .
, et al , 1993 , , 410 , 609 morgani , r. , ulrich , m .- h .
, tadhunter , c.n . , 1992 , , 254 , 546 mukherjee , r. , hartman , r.c . , lin , y.c .
& sreekumar , p. , 1999
, , 31 , 699 narayan , r. & yi , i. , 1994 , , 412 , l13 pacholczyk , a.g . , 1970 , radio astrophysics , freeman ( san francisco ) punch , m. et al , 1992 , , 358 , 477 rees , m.j . , begelman , m.c . ,
blandford , r.d . , & phinney , e.s . , 1982 , , 295 , 17 roberts , m.d . , et al , 1999 , , 343 , 691 sambruna , r.m . ,
maraschi , l. & urry , m.c . , 1996 , , 463 , 444 schubnell , m. , et al , 1994 , in aip conf .
304 , second compton symposium , ed .
fichtel , n. gehrels & j.p . norris ( new york : aip ) , 602 sikora , m. , begelman , m. & rees , m.j .
, 1994 , , 421 , 153 stecker , f.w .
& de jager , o.c . , 1998 , , 476 , 712 stecker , f.w .
, de jager , o.c . & salamon , m.h . , 1996 , , 473 , l75 steidel , c.c , et al , 1998 , in a. banday et al .
evolution of large scale structure , to be published .
takahashi , f. , et al , 1996 , , 470 , l89 tavecchio , f. , maraschi , l. & ghisellini , g. , 1998 , , 509 , 608 urry , c.m .
et al , 1997 , , 486 , 799 vermeulen , r.c .
, et al , 1995 , , 452 , l5 von montigny , c. et al , 1995 , , 440 , 525 wang , j .- m . , zhou , y .- y .
& cheng , k.s . , 2000 , in preparation wehrle , a.w . , et al , 1998 , , 497 , 178 zdziaski , a.a
& krolik , j.h . , 1993 , , 409 , l33 | photons with very high energy up to tev ( vhe ) emitted from active galactic nuclei ( agns ) provide some invaluable information of the origin of @xmath0-ray emission .
although 66 blazars have been detected by _
egret _ , only three low redshift x - ray selected bl lacs ( mrk 421 , mrk 501 , and 1es 2344 + 514 ) are conclusive tev emitters ( pks 2155 - 304 is a potential tev emitter ) since vhe photons may be absorbed by cosmological background infrared photons ( _ external _ absorption ) .
based on the `` mirror '' effect of clouds in broad line region , we argue that there is an _ intrinsic _ mechanism for the deficiency of tev emission in blazars . employing the observable quantities we derive the pair production optical depth @xmath1 due to the interaction of vhe photons with the reflected synchrotron photons by electron thomson scattering in broad line region .
this sets a more strong constraints on very high energy emission , and provides a sensitive upper limit of doppler factor of the relativistic bulk motion .
it has been suggested to distinguish the _ intrinsic _ absorption from the _ external _ by the observation on variation of multiwavelegenth continuum . |
in the mid 1980 s the jones polynomial of a link was introduced @xcite , @xcite .
it was defined as the normalized trace of an element of a braid group , corresponding to the link , in a certain representation .
although a topological invariant , the extrinsic nature of its computation made the relationship between topological configurations lying in the complement of the knot and the value of the jones polynomial obscure . in order to elucidate this connection , witten @xcite introduced topological quantum field theory .
tqft is a cut and paste technique which allows for localized computation of the jones polynomial .
following his discovery various approaches to topological quantum field theory were introduced ( @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite , @xcite and others ) . in the work of witten ,
it is obvious that the vector spaces of topological quantum field theory are quantizations of the space of connections on the manifold
. initial considerations of tqft focused on the formal rules of combination of vector spaces and vectors .
as more intrinsic developments of tqft appeared there was a shift towards the use of the kauffman bracket skein module .
the kauffman bracket skein module of a 3-manifold can be thought of as a quantization of the @xmath0characters of the manifold @xcite .
this approach to tqft recovers some of the initial feel of witten s work .
although the study of @xmath1-manifold invariants is substantial in itself , few applications to classical @xmath1-manifold topology have been found . in this paper
, we use quantum invariants to develop obstructions to embedding one 3-manifold in another . in order to extract quantum invariants from a 3-manifold
a good deal of extra information about its boundary is needed .
the big challenge is to sort out what is information about the manifold and what is just data about the parameterization chosen for the boundary .
a very similar problem arises in differential geometry .
the riemannian curvature tensor is a classifying invariant of the metric structure of the manifold , but it comes in the form of a tensor that depends intimately on the choice of local coordinates . to sidestep this , invariant functions of the coefficients of the tensor are used in order to obtain numbers that are independent of the choice of local coordinates .
similarly , we take the quantum invariant of a manifold and perform a construction to produce an invariant that is independent of the extra information about how the boundary is parametrized . in the next section
needed definitions are recalled and the tqft is constructed . the path we take is the elegant approach of roberts @xcite , along with a simplification of sikora @xcite .
this is a compact and eminently computable approach to topological quantum field theory .
the vector spaces are built out of framed links in the manifold , so that they have a topological feel much like singular homology and fundamental group . in the following section we prove a theorem that gives an obstruction to embedding one 3-manifold in another and illustrate it with an application of quantum obstructions .
if a 3-manifold contains a punctured lens space then it can not be embedded in the 3-sphere .
this observation lies at the heart of the technique of scharlemann cycles - a technique that has been extremely fruitful in resolving questions about dehn surgery on knots .
our proposition generalizes this criterion and opens the door to the use of quantum invariants in similar combinatorial arguments .
there is a construction of topological quantum field theory based on the witten - reshetikhin - turaev invariants due to blanchet , habegger , masbaum and vogel @xcite .
the vector space associated to a surface is a quotient of the vector space with basis given by all connected oriented 3-manifolds having that surface as boundary .
there is a pairing between the vector space of a surface and the vector space associated to the same surface with the opposite orientation . specifically , if @xmath2 then @xmath3 where @xmath4 is the witten - reshetikhin - turaev invariant of level @xmath5 @xcite , normalized so that the invariant of the 3-sphere is 1 . to obtain the vector space of a surface
take the quotient of the vector space above by the radical of this form .
the invariant of a connected 3-manifold with boundary is just its image in the quotient .
we take a different approach here , with equivalent results , that makes fundamental computations more direct .
the full apparatus of topological quantum field theory is not needed , as the invariant we derive is independent of framing .
hence , we set up a projective topological quantum field theory .
the approach we take is based on chapter 6 of @xcite , clarified by using a theorem from @xcite .
the construction uses the kauffman bracket skein module at a root of unity , modulo fusion .
let @xmath5 be an odd prime .
let @xmath6 be a primitive 8rth root of unity and @xmath7 .
we will work over @xmath8 $ ] , the cyclotomic numbers corresponding to @xmath6 .
the quantized integer @xmath9 $ ] denotes @xmath10 which is a unit in the ring @xmath11 $ ] for @xmath12 .
let @xmath13 be the positive square root of @xmath14 ^ 2 $ ] .
it can be shown that @xmath15 $ ] .
explicitly , @xmath16 let @xmath17 be an orientable @xmath1-manifold .
a framed link in @xmath17 is an embedding of a disjoint union of annuli into @xmath17 . in our diagrams
we draw only the core of an annulus lying parallel to the plane of the paper , i.e. with blackboard framing .
two framed links in @xmath17 are equivalent if there is an isotopy of @xmath17 taking one to the other .
let @xmath18 denote the set of equivalence classes of framed links in @xmath17 , including the empty link . in the vector space
@xmath8 \mathcal{l}$ ] , with basis @xmath18 , define @xmath19 to be the smallest subspace of @xmath8 \mathcal{l}$ ] containing all expressions of the form @xmath20 and @xmath21 , where the framed links in each expression are identical outside the balls pictured in the diagrams .
the kauffman bracket skein module @xmath22 is defined to be the quotient @xmath23 \mathcal{l } / s(m).\ ] ] we assume that the reader is familiar with the jones - wenzl idempotents and the kauffman triads as described on pages 136 - 138 of @xcite .
we refer the reader to pages 150 - 152 of @xcite for a computation of the number @xmath24 that is defined whenever @xmath25 is an admissible triple .
notice that @xmath24 is a unit in @xmath26 $ ] . by labeling an edge of a trivalent graph with the letter @xmath27 we mean that it carries the @xmath27-th jones - wenzl idempotent .
fusion ( see @xcite , page 156 ) is the relation : @xmath28}{\theta(a , b , c ) } \hspace{2pt } \raisebox{-22pt}{\includegraphics{fus2.eps}},\ ] ] where the sum is over all c so that the triads are admissible .
the result of enforcing fusion in the kauffman bracket skein module @xmath22 is denoted @xmath29 , and will be referred to as the reduced kauffman bracket skein module of level r. the element @xmath30 is the skein in the solid torus , @xmath31 \hspace{2pt } \raisebox{-15pt } { \includegraphics{fus1.eps}}.\ ] ] when one plugs @xmath30 into a component of a framed link in @xmath17 , then in @xmath29 one can alter the framed link by handle slides along that component and not change the element of @xmath29 .
the result of plugging in copies of @xmath30 into each of the components of the link @xmath32 is denoted by @xmath33 .
notice that the kauffman bracket of the result of plugging @xmath30 into a trivial framed knot in @xmath34 is @xmath13 .
let @xmath35 be the kauffman bracket of the skein in the three sphere obtained by plugging @xmath30 into the framed knot shown below .
we will also need the turaev - wenzl identity ( see @xcite ) .
suppose a diagram contains the figure below , where @xmath30 is plugged into the horizontal circle , and the vertical arc carries the @xmath36th jones - wenzl idempotent .
the diagram represents zero in the reduced kauffman bracket skein module unless @xmath37 .
( 56,56 ) ( 4,24)@xmath30 ( -22,45)@xmath36 we review a result of sikora that is used in the construction of the tqft .
suppose that @xmath17 and @xmath38 are compact oriented 3-manifolds and @xmath39 is an orientation preserving homeomorphism .
we say that framed links @xmath40 and @xmath41 are adapted to @xmath42 if 1 .
the result of surgery along @xmath32 is @xmath38 , the result of surgery along @xmath43 is @xmath17 .
2 . there are regular neighborhoods @xmath44 and @xmath45 , and an extension of @xmath42 , @xmath46 that is a homeomorphism taking the meridian of @xmath32 to the framing of @xmath43 , and the framing of @xmath32 to the meridian of @xmath43 . given a pair of links adapted to @xmath42 and the extension
, we can define a map from @xmath29 to @xmath47 , which we denote @xmath48 given a framed link @xmath49 in @xmath17 push it off of @xmath32 , then map it forward to @xmath38 using the extension of @xmath42 to get @xmath50 .
the skein in @xmath38 induced by the union of @xmath50 and @xmath51 is @xmath52 .
any two links adapted to @xmath42 differ by the kirby - roberts moves for surgery on a manifold with boundary @xcite . from this
we see that the map @xmath53 is well defined up to a power of @xmath35 .
[ @xcite ] the map @xmath54 is an isomorphism .
let @xmath55 denote the connected sum of @xmath17 and @xmath56 .
the kauffman bracket skein modules of @xmath55 and of the disjoint union of @xmath17 and @xmath56 are not isomorphic .
however , there is a natural isomorphism between @xmath57 and @xmath58 .
let @xmath38 and @xmath59 be the result of removing a ball from each of @xmath17 and @xmath56 .
there is a map @xmath60 coming from inclusion .
as the kauffman bracket skein module of the result of removing a ball from a manifold is isomorphic to the kauffman bracket skein module of the manifold , this induces , @xmath61 the map @xmath62 is an isomorphism .
this follows immediately from the following lemma .
if a @xmath1-manifold @xmath17 contains a family of embedded spheres then any skein in @xmath29 can be represented by a linear combination of links that miss these spheres .
let @xmath63 be a sphere embedded in @xmath17 and suppose that @xmath32 is a link in @xmath17 which intersects @xmath63 transversely in one point .
one can isotope @xmath32 so it is the same away from the sphere and has two added twists close to the sphere .
this implies that @xmath64 for some complex number @xmath65 , thus @xmath66 .
take a representative of any skein in @xmath29 .
one can assume that it is a linear combination of trivalent graphs admissibly colored with jones - wenzl idempotents @xcite .
use fusion repeatedly to make sure that the graphs do not have more than one edge intersecting each of the spheres embedded in @xmath17 . use the light bulb trick as above to see that the label on that edge has to be zero .
hence , we can be fluid about connected sums and disjoint unions so long as we work in the reduced skein module .
recall that a _ projective topological quantum field theory _ consists of a functor and a partition function , defined for each level @xmath5 .
the functor @xmath67 maps the category of _ marked _ surfaces and homeomorphisms to the category of vector spaces over @xmath8 $ ] , and linear maps defined up to scalar multiplication by a power of @xmath35 .
the partition function @xmath4 assigns to every _ marked _
3-manifold @xmath17 a vector @xmath68 in the vector space corresponding to its boundary , well defined up to scalar multiplication by a power of @xmath35 .
these have to satisfy the following axioms . 1 .
* dimension * : the empty surface is mapped to the vector space @xmath8 $ ] . @xmath69\ ] ] 2 .
* disjoint union * : disjoint union of surfaces is mapped to the tensor product of vector spaces .
@xmath70 3 .
* duality * : oppositely oriented marked surfaces are mapped to naturally dual vector spaces . @xmath71
this duality determines a pairing : @xmath72\ ] ] 4 .
* gluing * : gluing two @xmath1-manifolds along surfaces in their boundaries corresponds to contracting vectors .
+ let @xmath73 and @xmath74 be marked @xmath1-manifolds with @xmath75 and @xmath76 where the markings agree on @xmath77 . then @xmath78 can be written as @xmath79 , and @xmath80 can be written as @xmath81 . if @xmath82 then @xmath83 5 . * mapping cylinder * : the value of the partition function on the mapping cylinder of a homeomorphism corresponds to a linear map given by that homeomorphism .
+ let @xmath84 and let @xmath85 , where @xmath86 , so that @xmath87 . then + @xmath88 we will now construct a projective tqft .
fix a level @xmath5 .
_ vector spaces .
_ let @xmath77 be a closed oriented surface . a _ marking _ of @xmath77
is a choice of a handlebody @xmath89 with @xmath90 , up to homeomorphisms that are the identity on @xmath77 .
we assign to the marked surface @xmath91 the vector space @xmath92 .
_ linear maps .
_ let @xmath93 be an orientation preserving homeomorphism of the marked surfaces @xmath91 and @xmath94 .
given links @xmath95 and @xmath96 adapted to @xmath42 , we let @xmath97 .
our morphism is defined up to multiplication by a power of @xmath35 .
suppose now that a surface is not connected . in this case , a marking for the surface is a choice of a handlebody @xmath98 for each component .
the vector space associated with the surface is @xmath99 .
we can realize this tensor product by taking any combination of connected sums and disjoint unions of the @xmath98 , since @xmath100 is canonically isomorphic to @xmath99 .
_ duality .
_ let @xmath91 be a marked surface .
the marked surface with the orientations of @xmath77 and @xmath89 reversed is denoted by @xmath101 .
we can form the double @xmath102 , which is homeomorphic to @xmath103 , where @xmath104 is the genus of @xmath77 .
skeins @xmath105 and @xmath106 determine a skein in @xmath107 obtained by taking their disjoint union . in order to pair @xmath108 and @xmath109 ,
choose a family of spheres that cut @xmath103 down to punctured balls , and represent the union of @xmath108 and @xmath109 by a linear combination of links missing the spheres .
since the corresponding links live in punctured balls we can take their kauffman bracket to get a complex number times the empty link , finally multiply the coefficient by @xmath110 .
the final multiplication makes the answer coincide with a number computed in the 3-sphere . to be more precise ,
choose an isomorphism between @xmath111 and @xmath112 coming from a surgery diagram , and the answer will differ from the one we have given by multiplication by a power of @xmath35 .
we have defined a pairing , @xmath113.\ ] ] this is a duality pairing .
since @xmath114 , we can see the pairing as defined on @xmath115 . choose a trivalent spine for the handlebody @xmath89 .
a coloring @xmath116 of the spine is a choice of an integer between @xmath117 and @xmath118 for each edge .
when the numbers at each vertex are admissible , we can form a skein by running that number of arcs between the vertices and putting the appropriate kauffman triad at each vertex .
we denote that skein by @xmath119 .
the skeins @xmath119 where the coloring gives an admissible triad at each vertex form a basis for @xmath120 @xcite .
this can easily be verified by computing @xmath121 . if @xmath122 the answer is @xmath117 , and @xmath123 , where @xmath124 is a unit in @xmath11 $ ] .
the dual basis @xmath125 are the skeins with @xmath126 where @xmath127 is kronecker s delta .
it is easy to see that @xmath128 .
suppose now that @xmath129 .
there is a standard embedding of @xmath130 in @xmath131 , where @xmath104 is the sum of the genera of the components of @xmath132 . in the case
that @xmath130 is a handlebody , the double of @xmath130 is homeomorphic to @xmath131 , and this is the standard embedding of @xmath130 . if @xmath130 is a connected sum of handlebodies , take the connected sum of the doubles of each of the handlebodies .
choose the balls along which the sum is taken to lie inside the handlebodies @xmath98 .
it is clear that this embedding is unique up to homeomorphism and the complement of @xmath130 is a disjoint union of handlebodies .
choose a trivalent spine for each @xmath98 .
a basis for @xmath133 is given by all admissible colorings of the disjoint union of the spines , which we denote by @xmath134 or @xmath119 , where @xmath135 is the restriction of the coloring @xmath116 to the @xmath136th spine .
we can see our pairing quite nicely in the three - sphere . to this end
we say that @xmath130 is embedded in @xmath34 in the standard way if its complement is a disjoint union of handlebodies , so that each handlebody is unknotted and there is a family of disjoint balls in @xmath34 , each containing exactly one of the complementary handlebodies .
such an embedding is unique up to topological equivalence . to get from a standard embedding of @xmath130 in @xmath34 to a standard embedding in @xmath131
we can do surgery on a link that lies in the complementary handlebodies . specifically , choose an unknotted circle with zero framing , linking each 1-handle of @xmath130 lying in the complementary handlebodies so that the aggregate of such circles forms the unlink which we denote @xmath137 .
these are the same as the circles used in the constructions of @xcite , @xcite .
surgery along this link yields @xmath131 with @xmath130 embedded in a standard way .
in the three - sphere picture we can see the skeins @xmath134 lying in @xmath130 . to see a skein on the right hand side of the pairing , push it through the boundary of @xmath130 and take its disjoint union with @xmath138 .
notice that this is just the map of skein modules associated to the identity on the boundary of the complementary handlebodies in @xmath131 to the boundary of the complementary handlebodies in @xmath34 . to pair @xmath134 with @xmath139 ,
take the kauffman bracket of the union @xmath140 .
the result of the pairing is @xmath141 , where @xmath124 is a unit in @xmath11 $ ] depending on the coloring @xmath116 .
let @xmath17 be a compact , connected , oriented three - manifold .
a marking for @xmath17 is a choice of handlebodies , one for each connected component of @xmath142 .
we denote the connected sum of these handlebodies by @xmath130 .
if @xmath142 is empty then the marking is @xmath34 .
since there is only one choice we suppress it from the notation .
the _ preinvariant _ of @xmath143 is the pair consisting of @xmath29 and the empty skein .
let @xmath40 and @xmath144 be framed links adapted to the identity map on @xmath142 .
let @xmath145 be the associated isomorphism .
we define @xmath146 once again , from the kirby calculus for manifolds with boundary , this is well defined up to multiplication by a power of @xmath35 .
notice that the orientation on @xmath142 is always the orientation inherited from @xmath17 . using the main result from @xcite
we see that if @xmath17 is any manifold then @xmath147 can be written as @xmath148 where the @xmath149 are algebraic integers .
the dual basis allows us to rewrite this answer without any fractions : @xmath150 where the @xmath151 are just the @xmath149 times a unit depending on @xmath116 . if @xmath152 then @xmath153 and the basis consists of the empty skein . in this case
@xmath154 $ ] and , up to multiplication by a power of @xmath35 , it is equal to the witten - reshetikhin - turaev invariant @xmath155 @xcite , @xcite .
suppose that @xmath156 and @xmath157 are marked @xmath1-manifolds , and that there is a surface @xmath77 with @xmath158 and @xmath159 .
we will assume that the genus of @xmath77 is @xmath104 .
suppose further that the handlebodies in @xmath160 and @xmath161 corresponding to @xmath77 and @xmath162 are a handlebody @xmath89 for @xmath77 and @xmath163 for @xmath162 .
the invariants of the manifolds are @xmath164 and @xmath165 where the first factor in each tensor corresponds to @xmath77 or @xmath162 .
the invariant of the result of gluing @xmath156 and @xmath157 with respect to the marking @xmath130 , coming from deleting the handlebodies @xmath89 and @xmath163 and taking the disjoint union of the remaining handlebodies , is @xmath166 to see this , glue @xmath160 to @xmath161 and then choose the obvious surgery diagram in @xmath163 to take the union to a connect sum of handlebodies corresponding to the deletion of @xmath89 and @xmath163 . using ideas of roberts @xcite
, we can establish the equivalence of this projective tqft and the one derived from the constructions in @xcite .
there is a natural equivalence of the functors from the compact marked oriented surfaces to vector spaces , that takes every manifold @xmath17 to a surgery diagram for @xmath17 in the handlebody that is its marking .
it is easy to check that this equivalence preserves the invariants of the three - manifolds ( projectively ) .
in this section we introduce three ideals in @xmath11 $ ] derived from quantum invariants .
the first two are invariants of the manifold , and can be used as obstructions to embedding one manifold into another .
the third ideal is not a manifold invariant but it can be easily computed and is used to estimate one of the first two ideals . if an ideal is all of @xmath11 $ ] we say it is trivial .
it is clear that an ideal is trivial if it contains a unit @xcite .
any element of @xmath173 can be written as a sum , taken over all the manifolds @xmath56 with @xmath174 , @xmath175 since for every such @xmath56 , the element @xmath176 is in the ideal @xmath168 , thus also @xmath177 , and so is the linear combination of such elements .
it follows form theorem 2 of yamada @xcite , where he computes @xmath186 .
in particular for any lens space there exists @xmath5 for which @xmath186 is not a unit . if @xmath187 then for any odd prime @xmath5 the ideal is zero .
otherwise take @xmath5 to be an odd prime dividing @xmath188 , and the ideal will be either @xmath117 or nontrivial .
let @xmath143 be a marked @xmath1-manifold , denote by @xmath104 the sum of the genera of the handlebodies making up @xmath130 , and suppose that @xmath189 with respect to a basis for @xmath130 made up of admissibly colored trivalent spines .
define the ideal @xmath190 to be generated by all coefficients @xmath191 . @xmath192.\ ] ] the ideal @xmath190 is only an invariant of the marked three - manifold .
for instance , if we mark the solid torus @xmath193 with itself , then the surgery diagram is empty and the ideal is generated by @xmath13 . if we mark the solid torus so that its longitude corresponds to the meridian of the marking solid torus and its meridian corresponds to the longitude of the marking torus , then the surgery diagram consists of the core of the marking torus with @xmath30 plugged in , hence the ideal is 1 .
we are less interested in @xmath190 as an invariant than we are as a tool to estimate @xmath171 .
we restrict our attention now to the case when @xmath142 is connected .
the ideal @xmath173 can be estimated from the ideal @xmath194 for a particular marking @xmath130 . here
@xmath195 denotes the disk sum of @xmath17 and @xmath196 along a disk in @xmath142 .
the construction is an extension for analyzing the invariant of a @xmath1-manifold with boundary of roberts `` chain mail '' @xcite .
let @xmath198 be a surface of genus @xmath104 .
let @xmath199 be a disk and let @xmath200 , and denote the double of @xmath201 by @xmath202 .
the boundary of @xmath195 can be naturally identified with @xmath203 .
we will use the handlebody @xmath204 for the marking @xmath130 of @xmath195 , obviously @xmath205 .
we will show that for any @xmath56 with @xmath206 , the turaev - viro invariant of the manifold @xmath207 can be expressed as an integral linear combination of the coefficients @xmath191 where , @xmath208 notice that @xmath104 is one half of the genus of @xmath130 .
we achieve this by constructing an efficacious surgery description in the three - sphere .
first we find a surgery diagram in @xmath34 for the double of @xmath209 .
next , we locate a framed link in @xmath209 that is adapted to the marking of @xmath210 by @xmath203 .
finally we combine these diagrams to effect the computation .
the double of @xmath204 can be constructed in two steps .
first identify @xmath211 with @xmath212 .
second , fill the boundary so that the curves @xmath213 , where @xmath214 , bound meridian disks in the filling torus .
alternatively , one can do @xmath117-surgery along a fiber of @xmath215 . a surgery diagram for @xmath215 is drawn below for genus @xmath216 . in general , the number of clasped pairs woven through the middle circle is equal to @xmath104 .
let @xmath217 be a regular neighborhood of the boundary of the disk @xmath218 in @xmath77 .
notice that @xmath219 $ ] .
let @xmath220 be a smooth function , equal to @xmath221 on the disk @xmath222 , and @xmath117 outside of @xmath223 .
suppose that on @xmath217 the function @xmath42 has the circles @xmath224 as level sets , and that it decreases from @xmath221 to @xmath117 as you move from the boundary component in @xmath218 to the boundary component outside of @xmath218 , without any critical points .
extend @xmath42 to a smooth function on @xmath17 .
perturb @xmath42 so that it is morse and then move all index @xmath221 critical points below all of the index @xmath225 critical points , and cancel all local minima and maxima .
let @xmath226 be a level between the index @xmath221 and index @xmath225 critical points .
notice that @xmath227)$ ] is the result of adding @xmath221 handles to @xmath204 along @xmath211 .
we get @xmath17 from @xmath227)$ ] by adding @xmath225-handles along some curves @xmath228 in @xmath229 . using this picture
we can build a surgery diagram for @xmath230 .
let @xmath231 be the result of adding @xmath232 trivial @xmath221-handles from above to @xmath233 , so that it has the same genus as @xmath229 .
choose a homeomorphism from @xmath227)$ ] to the part of @xmath204 between @xmath212 and @xmath231 , so that @xmath234 is the identity .
let @xmath235 be the image of the curves along which you add the @xmath225-handles .
the surgery diagram will consist of @xmath236 curves @xmath235 , along with @xmath232 @xmath117-framed curves @xmath237 above @xmath231 linking each added trivial @xmath221-handle in @xmath204 .
notice that there is an orientation reversing map of order @xmath225 on the result of surgery along the curves @xmath237 , that is the identity on @xmath231 .
this establishes that surgery on this diagram yields @xmath195 .
from this diagram and the marking @xmath130 we can compute @xmath238 where the @xmath119 are standard basis elements coming from an admissibly colored trivalent spine for @xmath201 and @xmath104 is the genus of @xmath201 . to do this , apply the turaev - wenzl identity @xmath232 times to cancel out the curves @xmath237 . to see @xmath231 in the surgery diagram for the double of @xmath209 ,
take a disk whose boundary is where the middle curve of our initial surgery diagram was , and let it swallow the arcs of the clasped curves going through it upwards . finally add enough trivial handles above to obtain a surface homeomorphic to @xmath231 .
visualize the surgery diagram for @xmath195 by seeing the curves @xmath235 lying on @xmath231 and the curves @xmath237 linking the trivial handles .
since a regular neighborhood of the version of @xmath231 lying in @xmath34 , along with balls containing the trivial one handles and the linking curves , is homeomorphic to @xmath209 , we can expand the surgery curves @xmath235 and the trivial linking handles @xmath237 to get the same formal sum @xmath239 as above , where the @xmath119 now lie in @xmath34 .
we can perform an analogous construction based on a disk outside the middle curve , only swallowing the clasping curves downwards , and add trivial handles downwards . if @xmath56 is any @xmath1-manifold with @xmath240 we can find a surgery diagram for the disk sum of @xmath56 with @xmath241 from a heegaard surface , and construct a surface @xmath242 analogous to @xmath231 .
we can place this surgery diagram on the surface outside that we just constructed ( with linking curves below the trivial handles ) . in the picture
we just show the tubes that swallow the clasping curves below .
trivial handles below can be added as needed . in this way we can build a surgery diagram for
any @xmath243 expand the surgery curves in terms of a standard basis @xmath244 associated to @xmath204 , with @xmath245 lying in the copy of @xmath201 outside the middle curve .
similar to equation ( [ msum ] ) , @xmath246 the @xmath247 clasped circles are trivial and unlinked .
furthermore , the disks these circles bound cut the heegaard surfaces down to planar surfaces .
hence , the turaev - wenzl identity can be applied enough times to get rid of all the fractional part of the invariant .
therefore , @xmath248 is an integral linear combination of @xmath249 .
this implies that @xmath250 where @xmath130 is the marking we constructed here .
as an immediate consequence of propositions [ mark ] and [ obstruct ] we see that if the ideal @xmath251 is nontrivial then @xmath17 does not embed in the @xmath1-sphere .
this leads to the following example .
let @xmath17 be the result of gluing two solid tori along annuli in their boundaries .
choose the annuli so that their cores are @xmath252 curves .
the disk sum of @xmath17 with @xmath196 can be realized as surgery on a diagram in the cylinder over a punctured torus .
the ideal @xmath253 is the principal ideal generated by the integer @xmath225 in the ring @xmath254 $ ] .
as this ideal is nontrivial , @xmath17 can not be embedded in @xmath34 . | we use topological quantum field theory to derive an invariant of a three - manifold with boundary .
we then show how to use the structure of this invariant as an obstruction to embedding one three - manifold into another . |
vacuum polarization and pair production have been issues of continuous concern since the early works by sauter , heisenberg and euler , and weisskopf @xcite , and then by schwinger @xcite ( for a review and references , see ref .
the task of directly computing , without relying on the electromagnetic duality , the effective action in electric field backgrounds , however , has been a challenging problem due to the vacuum instability .
dunne and hall used the resolvent method to directly find the effective action in time - dependent electric fields @xcite . in the previous paper @xcite , employing the evolution operator method , we found the exact one - loop effective actions of scalar and spinor qed at zero temperature in a constant or a pulsed electric field of sauter - type , which satisfy the exact relation @xmath0 ( with @xmath1 for scalar and @xmath2 for spinor ) between the imaginary part of the effective lagrangian density @xmath3 and the mean number of created pairs @xmath4 at state @xmath5 . even finding the pair production rate by time - dependent or spatially localized electric fields is methodologically nontrivial , which has recently been intensively studied @xcite . to calculate the effective action and
thereby schwinger pair production at finite temperature is another challenging problem in qed . in a constant pure magnetic field
the qed effective action was studied at finite temperature @xcite and at finite temperature and density @xcite . however , the presence of an additional electric field raised schwinger pair production at debate depending on the formalism employed .
some calculations in thermal field theory reported that the effective action in both a constant magnetic and electric field had an imaginary part having dependence on temperature @xcite .
however , in the real - time formalism , no imaginary part was found in the qed effective action in the presence of both magnetic and electric field @xcite .
the recent calculation of effective action in the imaginary - time formalism shows an imaginary part only at two - loop but not at one - loop in both a constant electric and magnetic field @xcite . in nonequilibrium quantum field theory of scalar qed , a calculation in the real - time formalism
shows thermal enhancement of pair production @xcite . in this paper
, using the evolution operator method , we find the pair - production rate at finite temperature in time - dependent electric fields both in scalar and spinor qed .
the evolution operator , unitarily transforming the particle and antiparticle operators from the ingoing vacuum to the outgoing vacuum , carries all the information of quantum evolution .
in fact , the evolution operator is completely determined by the bogoliubov coefficients .
the advantage of the evolution operator is the readiness to calculate the probability for transitions among multiparticle states .
this allows us to compute the mean number of created pairs at finite temperature in scalar and spinor qed by counting the pairs from the vacuum and the induced production and the stimulated annihilation in a thermal ensemble of bosons and fermions .
we find that the mean number of created pairs is factorized into the mean number of created pairs at zero temperature and the initial thermal distribution for bosons and fermions .
the organization of this paper is as follows . in sec .
ii , rewriting the bogoliubov transformation as a unitary transformation by the evolution operator , we find the mean number of created pairs at zero temperature in scalar and spinor qed . in sec .
iii , we calculate the mean number of created pairs at finite temperature and apply it to the sauter - type electric field .
we first consider scalar qed for spinless charged bosons under an external electric field with the gauge field @xmath7 . the electric field is assumed to be acting on for a finite period of time so that the ingoing and the outgoing vacua are well - defined .
thus , at @xmath8 , before the external electric field being turned on , the scalar field is free and the hamiltonian takes the usual form @xmath9 where @xmath10 is the initial frequency at momentum @xmath11 . here
, the gauge is chosen @xmath12 , so that the ingoing vacuum @xmath13 is the minkowski vacuum @xmath14 , annihilated by @xmath15 and @xmath16 for each momentum @xmath11 .
similarly , the outgoing vacuum at @xmath17 is defined by @xmath18 and @xmath19 .
these operators are related through the bogoliubov transformations @xcite @xmath20 where @xmath21 .
to express the outgoing vacuum as multiparticle states of the ingoing vacuum , we rewrite the bogoliubov transformations ( [ sc - out - in ] ) as a unitary transformation @xcite @xmath22 here , the evolution operator @xmath23 is factorized into the overall phase factor and the two - mode squeeze operator as @xcite @xmath24 where @xmath25 from the charge neutrality of the vacuum , equal numbers of particles and antiparticles are produced at zero temperature and they carry the opposite momenta due to the momentum conservation .
the multiparticle state of @xmath5-pairs can be concisely denoted as @xmath26 .
the probability for @xmath5-pairs with momentum @xmath11 to be created from the vacuum is @xmath27 note that @xmath28 and @xmath29 for each @xmath11 .
thus , at zero temperature , the mean number of pairs created from the vacuum for each momentum per unit volume is @xmath30 next , in spinor qed , before the interaction of an external electric field , the spinor field is free without the gauge potential @xmath31 , and has the hamiltonian given by @xmath32 where @xmath33 and @xmath34 are particle and antiparticle operators in the ingoing vacuum . after the interaction of the electric field
, the ingoing vacuum evolves to the outgoing vacuum , whose particle and antiparticle operator are @xmath35 and @xmath36 .
the bogoliubov transformations between the ingoing and the outgoing particle and antiparticle operators , @xmath37 , are similarly given by @xmath38 where @xmath39 and @xmath40 with @xmath41 .
the bogoliubov transformation can be also written as a unitary transformation @xcite @xmath42 where @xmath43 here , the three parameters @xmath44 and @xmath45 are determined by the bogoliubov coefficients as @xmath46 note that the pair production on spinor qed is restricted to only one pair of particle and antiparticle for a given quantum number @xmath5 due to the pauli exclusion principle .
thus , the mean number of pairs created from the vacuum for each state @xmath5 at zero temperature is calculated as @xmath47
we now calculate the mean number of pairs at finite temperature from the probability for each transition , as in the case of zero temperature .
as there is no mode - mixing , we separately calculate the mean number of created pairs for each mode . for an initial thermal ensemble at @xmath49 , which might not be charge neutral , the mean number of produced pairs consists of the vacuum pair production , the induced pair production , and the stimulated pair annihilation as shown in fig . 1 : @xmath50 , \label{mean - t1}\end{aligned}\ ] ] where @xmath51 is the partition function for the initial ensemble and @xmath52 is the transition probability from @xmath53 to @xmath54 . using @xmath55 , which implies @xmath56 , the mean number of pairs created at finite temperature , eq .
( [ mean - t1 ] ) , can be written as @xmath57 \nonumber \\ ~ & = & \frac{1}{2z_{n } } { \rm tr}(u_{n,{\rm in } } n_{n,{\rm in}}u_{n,{\rm in}}^{\dagger } -n_{n,{\rm in } } ) e^{-\beta h_{n,{\rm in } } } , \label{mean - t2}\end{aligned}\ ] ] where @xmath58 for scalar particles and @xmath59 for spinor particles and @xmath60 for both . from the bogoliubov transformations ( [ sc - u - tr ] ) and ( [ sp - u - tr ] )
, we have @xmath61 .
thus , the mean number of pairs created at finite temperature can be written concisely as @xmath62 first , for scalar particles , from eq .
( [ sc - out - in ] ) , @xmath63 and @xmath64 , the mean number of scalar pairs created in momentum @xmath11 at finite temperature is given by @xmath65 second , for spinor particles , from eq .
( [ sp - out - in ] ) , @xmath66 and @xmath67 , we find the mean number of spinor pairs created in state @xmath5 at finite temperature @xmath68 , in @xmath69 state for a pulsed sauter - type electric field of duration t@xmath70 : ( a ) dependence on temperature @xmath71 with e@xmath72 , and ( b ) dependence on electric field strength @xmath73 with t@xmath72 for the upper and the lower lines , and with t@xmath74 for the middle lines , where dimensionless parameters @xmath75 t , @xmath76e , and @xmath77 t are used.,title="fig:",width=268 ] , in @xmath69 state for a pulsed sauter - type electric field of duration t@xmath70 : ( a ) dependence on temperature @xmath71 with e@xmath72 , and ( b ) dependence on electric field strength @xmath73 with t@xmath72 for the upper and the lower lines , and with t@xmath74 for the middle lines , where dimensionless parameters @xmath75 t , @xmath76e , and @xmath77 t are used.,title="fig:",width=268 ] + as an interesting model for discussions , we consider the sauter - type electric field @xmath78 with the gauge choice , @xmath79 , which allows the exact solution leading to the exact mean number of pairs created , @xmath80 , at zero temperature both in scalar and spinor qed @xcite .
also the approximation scheme of the wkb or worldline instanton method has been developed @xcite . because the sauter - type electric field acts effectively for a finite period of time @xmath81 , the ingoing thermal states are stable and well defined as required in our formalism .
2 shows the mean number of created pairs eqs.([sc - result],[sp - result ] ) at finite temperature with the electric field @xmath73 , the duration @xmath81 and temperature @xmath82 scaled in terms of the critical strength , the compton time and the electron mass , respectively .
the production of scalar pairs is thermally enhanced , while the production of fermion pairs is thermally suppressed as expected by the pauli blocking , which is consistent with the calculation by density matrix method @xcite .
the pure thermal effect on the mean number of created pairs , which is @xmath83 , is given by @xmath84 being the bose - einstein or fermi - dirac distribution .
for @xmath85 , the distribution @xmath86 approximately equals to the boltzmann factor @xmath87 .
thus , the mean number of created pairs at finite temperature , eqs.([sc - result],[sp - result ] ) , is reduced to the zero temperature result , eqs.([sc - mean],[sp - mean ] ) , for a temperature much lower than the rest mass
. our result eq.([sp - result ] ) could be compared with the calculation in imaginary - time formalism @xcite , where the thermal effect appears only at two - loop because thermal one - loop fluctuations are on - shell . on the other hand , eq .
( [ sp - result ] ) , an off - shell calculation , is equivalent to the thermal loop times the vacuum one - loop , in fact , part of two - loops
. one interesting comment to be pointed out is that the momentum integral over the distribution function in @xmath88 leads to the factor @xmath89 in ref .
@xcite , though the momentum integral of the distribution is intertwined with the vacuum pair - production rate . to show rigorously the connection between our result and ref .
@xcite requires calculating the effective action at finite temperature along the line of ref.@xcite , which will be addressed in the future .
in this paper , using the evolution operator method , we found that the mean number of created pairs in state @xmath5 at finite temperature is given by @xmath90 for scalar and spinor qed in external electric fields , the total mean number density of created pairs at finite temperature is given by @xmath91 where @xmath92 and @xmath93 are the mean number of created pairs at zero temperature for scalar and spinor qed , respectively .
k. would like to appreciate the hospitality of hanyang university , and h. k. l. and y. y. would like to appreciate the hospitality of kunsan national university .
the work of s. p.
k. was supported by the korea research foundation grant funded by the korean government ( moehrd ) ( krf-2007-c00167 ) and the work of h. k. l. was supported by the korea science and engineering foundation ( kosef ) grant funded by the korea government ( most ) ( no .
r01 - 2006 - 000 - 10651 - 0 ) .
g. v. dunne , `` heisenberg - euler effective lagrangians : basics and extensions , '' _ from fields to strings : circumnavigating theoretical physics _ , edited by m. shifman , a. vainshtein , and j. wheater , ( world scientific , singapore , 2005 ) , vol .
445 - 522 , hep - th/0406216 . | we use the evolution operator method to find the schwinger pair - production rate at finite temperature in scalar and spinor qed by counting the vacuum production , the induced production and the stimulated annihilation from the initial ensemble .
it is shown that the pair - production rate for each state is factorized into the mean number at zero temperature and the initial thermal distribution for bosons and fermions . |
let @xmath2 be the undirected finite path with vertex set @xmath3 and edge set @xmath4 . given two positive measures @xmath5 on @xmath6 with @xmath7 , the dirichlet form and variance associated with @xmath8 and @xmath9
are defined by @xmath10[g(i)-g(i+1)]\nu(i , i+1)\ ] ] and @xmath11 where @xmath12 are functions on @xmath13 .
when convenient , we set @xmath14 . the spectral gap of @xmath15 with respect to @xmath5 is defined as @xmath16 let @xmath17 be a matrix given by @xmath18 for @xmath19 and @xmath20 obviously , @xmath21 is the smallest non - zero eigenvalue of @xmath17 .
undirected paths equipped with measures @xmath5 are closely related to birth and death chains .
a birth and death chain on @xmath22 with birth rate @xmath23 , death rate @xmath24 and holding rate @xmath25 is a markov chain with transition matrix @xmath26 given by @xmath27 where @xmath28 and @xmath29 . under the assumption of irreducibility ,
that is , @xmath30 for @xmath31 , @xmath26 has a unique stationary distribution @xmath9 given by @xmath32 , where @xmath33 is the positive constant such that @xmath34 .
the smallest non - zero eigenvalue of @xmath35 is exactly the spectral gap of the path on @xmath36 with measures @xmath5 , where @xmath37 for @xmath31 .
note that if @xmath38 is the constant function of value @xmath39 and @xmath40 is a minimizer for @xmath21 , then @xmath41 is an eigenvector of @xmath17 .
this implies that any minimizer @xmath40 for @xmath21 satisfying @xmath42 satisfies the euler - lagrange equation , @xmath43\nu(i-1,i)+[\psi(i)-\psi(i+1)]\nu(i , i+1),\ ] ] for all @xmath44 . assuming the connectedness of @xmath15 ( i.e. , the superdiagonal and subdiagonal entries of @xmath17 are positive )
, the rank of @xmath45 is at least @xmath46 .
this implies that all eigenvalues of @xmath17 are simple .
see lemma [ l - mat ] for an illustration .
observe that , by ( [ eq - el ] ) , any non - trivial eigenvector of @xmath17 has mean @xmath47 under @xmath9 .
this implies that all minimizers for the spectral gap are of the form @xmath48 , where @xmath49 are constants and @xmath40 is a nontrivial solution of ( [ eq - el ] ) . in 2009 , miclo obtained implicitly the following result .
[ t - miclo ] ( * ? ? ?
* proposition 1 ) if @xmath40 is a minimizer for @xmath21 , then @xmath40 must be monotonic , that is , either @xmath50 for all @xmath51 or @xmath52 for all @xmath51 . one aim of this paper is to provide a scheme to compute the spectrum of @xmath17 , in particular , the spectral gap .
based on miclo s observation , it is natural to consider the following algorithm .
@xmath53\nu(i-1,i ) -\lambda_k\pi(i)\psi_k(i)\}^+}{\nu(i , i+1)},\\ & \quad\text{for } 1\le i < n,\,\text{where } t^+=\max\{t,0\},\\ & 3.\,\lambda_{k+1}=\frac{\mathcal{e}_\nu(\psi_k,\psi_k)}{\textnormal{var}_{\pi}(\psi_k)}. \end{aligned}\ ] ] the following theorems discuss the behavior of @xmath54 .
[ t - main1 ] referring to ( a1 ) , if @xmath55 , then @xmath56 for all @xmath57 .
if @xmath58 , then the sequence @xmath59 satisfies * if @xmath60 , then @xmath56 for all @xmath61 . * if @xmath62 , then @xmath63 for @xmath57 . *
set @xmath64 .
then , @xmath65 and @xmath66 .
[ t - main2 ] referring to _ theorem [ t - main1 ] _ , there is a constant @xmath67 independent of the choice of @xmath68 such that @xmath69 for all @xmath57 . by theorem [ t - main2 ]
, we know that the sequence @xmath54 generated in ( [ alg - sp ] ) converges to the spectral gap exponentially but the rate @xmath70 is undetermined . the following alternative scheme is based on using more information on the spectral gap and will provide convergence at a constant rate .
@xmath71\nu(i-1,i ) -\lambda_k\pi(i)\psi_k(i)\}^+}{\nu(i , i+1)},\\ & \quad\text{for } 1\le i < n,\,\text{where } t^+=\max\{t,0\},\\ & 3.\,\begin{cases}l_{k+1}=l_k,\,u_{k+1}=\lambda_k&\text{if } \pi(\psi_k)>0\\ l_{k+1}=\lambda_k,\,u_{k+1}=u_k&\text{if } \pi(\psi_k)<0\\l_{k+1}=u_{k+1}=\lambda_k&\text{if } \pi(\psi_k)=0\end{cases}. \end{aligned}\ ] ] [ t - main3 ] referring to ( [ alg - sp2 ] ) , it holds true that @xmath72 in theorem [ t - main3 ] , the convergence to the spectral gap is exponentially fast with explicit rate , @xmath73 .
see remark [ r - ini ] for a discussion on the choice of @xmath74 and @xmath75 . for higher
order spectra , miclo has a detailed description of the shape of eigenvectors in @xcite and this will motivate the definition of similar algorithms for every eigenvalue in spectrum .
see ( [ alg - evi2 ] ) and theorem [ t - di ] for a generalization of ( [ alg - sp2 ] ) and theorem [ t - local ] for a localized version of theorem [ t - main2 ] .
the spectral gap is an important parameter in the quantitative analysis of markov chains .
the cutoff phenomenon , a sharp phase transition phenomenon for markov chains , was introduced by aldous and diaconis in early 1980s .
it is of interest in many applications .
a heuristic conjecture proposed by peres in 2004 says that the cutoff exists if and only if the product of the spectral gap and the mixing time tends to infinity .
assuming reversibility , this has been proved to hold for @xmath76-convergence with @xmath77 in @xcite .
for the @xmath78-convergence , ding _
@xcite prove this conjecture for continuous time birth and death chains . in order to use peres conjecture in practice ,
the orders of the magnitudes of spectral gap and mixing time are required .
the second aspect of this paper is to derive a theoretical lower bound on the spectral gap using only the birth and death rates .
this lower bound is obtained using the same idea used to analyze the above algorithm . for estimates on the mixing time of birth and death chains
, we refer the readers to the recent work @xcite by chen and saloff - coste . for illustration
, we consider several examples of specific interest and show that the lower bound provided here is in fact of the correct order in these examples .
this article is organized as follows . in section 2 ,
the algorithms in ( [ alg - sp])-([alg - sp2 ] ) are explored and proofs for theorems [ t - main1]-[t - main3 ] are given . in section 3 , the spectrum of @xmath17 is discussed further and , based on miclo s work @xcite , algorithm ( [ alg - sp2 ] ) is generalized to any specified eigenvalue of @xmath17 .
our method is applicable for paths of infinite length ( one - sided ) and this is described in section 4 . for illustration , we consider some metropolis chains and display numerical results of algorithm ( [ alg - sp2 ] ) in section 5 . in section 6 , we focus on uniform measures with bottlenecks and determine the correct order of the spectral gap using the theory in sections 2 - 3 .
it is worthwhile to remark that the assumptions in section 6 can be relaxed using the comparison technique in @xcite . as the work in this paper
can also be regarded as a stochastic counterpart of theory of finite jacobi matrices , we would like to refer the readers to @xcite for a complementary perspective .
this section is devoted to proving theorems [ t - main1]-[t - main3 ] .
first , we prove theorem [ t - miclo ] in the following form .
[ l - polish ] let @xmath79 and @xmath40 be a non - constant function on @xmath13 .
suppose @xmath80 solves ( [ eq - el ] ) and @xmath40 is monotonic .
then , @xmath40 is strictly monotonic , that is , either @xmath81 for @xmath51 or @xmath82 for @xmath51 .
obviously , ( [ eq - el ] ) implies that @xmath42 . without loss of generality
, it suffices to consider the case when @xmath83 and @xmath84 .
since @xmath40 is non - constant and @xmath85 , we have @xmath86 and @xmath87 .
note that if there are @xmath88 such that @xmath89 , @xmath90 and @xmath91 for @xmath92 , then ( [ eq - el ] ) yields @xmath93\nu(i-1,i)+[\psi(i)-\psi(i+1)]\nu(i , i+1)>0\ ] ] and @xmath94\nu(j-1,j)+[\psi(j)-\psi(j+1)]\nu(j , j+1)<0,\ ] ] a contradiction .
thus , @xmath40 is strictly increasing .
we note the following corollary .
[ c - polish ] let @xmath80 be a pair satisfying ( [ eq - el ] ) .
then , @xmath95 if and only if @xmath40 is monotonic .
one direction is obvious from theorem [ t - miclo ] . for the other direction , assume that @xmath40 is monotonic and let @xmath96 be a minimizer for @xmath21 with @xmath97 .
since @xmath80 and @xmath98 are solutions to ( [ eq - el ] ) , one has @xmath99 by lemma [ l - polish ] , @xmath40 and @xmath96 are strictly monotonic and this implies @xmath100 . as a consequence of the above equations , we have @xmath95 .
the following proposition is the key to theorem [ t - main1 ] .
[ p - main1 ] suppose that @xmath80 satisfies @xmath79 , @xmath83 and , for @xmath51 , @xmath101\nu(i-1,i)-\lambda\pi(i)\psi(i)\}^+}{\nu(i , i+1)},\ ] ] where @xmath102 .
then , the following are equivalent . * @xmath103 .
* @xmath42 . * @xmath95 .
furthermore , if @xmath58 , then any of the above is equivalent to * @xmath104 [ r - n=2 ] for @xmath55 , it is an easy exercise to show that @xmath105 . by following the formula in ( [ eq - psigen ] ) , one has @xmath106 $ ] , which leads to @xmath107 .
set @xmath108 and @xmath109 . since @xmath83 and @xmath79 , @xmath86 and @xmath110 is nonempty .
according to ( [ eq - psigen ] ) , @xmath40 is non - decreasing .
note that if @xmath111 , then @xmath112 and @xmath113 .
this implies @xmath40 is strictly increasing on @xmath114 and , for @xmath115 , @xmath116\nu(i , i+1)+[\psi(i)-\psi(i-1)]\nu(i-1,i).\ ] ] multiplying @xmath117 on both sides and summing over all @xmath118 in @xmath110 yields @xmath119 ^ 2\nu(i , i+1)\notag\\ & \qquad+\psi(i_0)[\psi(i_0)-\psi(i_0 + 1)]\nu(i_0,i_0 + 1)\notag\\ & = \mathcal{e}_\nu(\psi,\psi)+\psi(i_0 + 1)[\psi(i_0)-\psi(i_0 + 1)]\nu(i_0,i_0 + 1)\notag\\ & = \mathcal{e}_\nu(\psi,\psi)+\lambda\psi(n)\sum_{i=1}^{i_0}\psi(i)\pi(i).\notag\end{aligned}\ ] ] this is equivalent to @xmath120,\ ] ] which proves ( 1)@xmath121(2 ) .
if @xmath95 , then @xmath40 is an eigenvector for @xmath17 associated to @xmath21 .
this proves ( 3)@xmath122(2 ) . for ( 2)@xmath122(3 ) , assume that @xmath42 . in this case , @xmath40 must be strictly increasing .
otherwise , @xmath123 for @xmath124 and , according to ( [ eq - psigen ] ) , this implies @xmath125 ^ 2\nu(i , i+1)=\mathcal{e}(\psi,\psi),\ ] ] which contradicts ( 1 ) .
as @xmath40 is strictly increasing and @xmath42 , @xmath80 solves ( [ eq - el ] ) . by corollary [ c - polish ] , @xmath95 .
to finish the proof , it remains to show ( 4)@xmath122(3 ) ( ( 3)@xmath122(4 ) is obvious from the equivalence among ( 1 ) , ( 2 ) and ( 3 ) ) .
assume that @xmath126 .
by lemma [ l - polish ] , @xmath40 is strictly monotonic and this implies , for @xmath51 , @xmath116\nu(i , i+1)+[\psi(i)-\psi(i-1)]\nu(i-1,i).\ ] ] as @xmath40 is a minimizer for @xmath21 , one has , for @xmath44 , @xmath127=[\psi(i)-\psi(i+1)]\nu(i , i+1 ) + [ \psi(i)-\psi(i-1)]\nu(i-1,i).\ ] ] if @xmath128 , the comparison of both systems yields @xmath129 as @xmath58 , @xmath130 , a contradiction !
this forces @xmath95 , as desired .
the following is a simple corollary of proposition [ p - main1 ] , which plays an important role in proving theorem [ t - main3 ] .
[ c - t ] let @xmath58 . for @xmath79 ,
let @xmath131 be the vector generated by ( [ eq - psigen ] ) with @xmath132 .
then , @xmath133 for @xmath79 and @xmath128 . without loss of generality , we fix @xmath134 for all @xmath79 . set @xmath135 . to prove this corollary
, it suffices to show that @xmath136 for @xmath79 , define @xmath137 . by ( [ eq - enu ] )
, one has @xmath138}{\text{var}_\pi(\phi_\lambda)}.\ ] ] since @xmath131 is non - constant , @xmath139 .
this implies @xmath140 for @xmath141 . for @xmath142 , set @xmath143 .
by proposition [ p - main1 ] , @xmath144 if and only if @xmath95 . by the continuity of @xmath145 ,
this implies either @xmath146 or @xmath147 . in the case @xmath146
, one has @xmath148 for @xmath149 .
as @xmath150 is bounded , @xmath151 is convergent with limit @xmath152 and this yields @xmath153 = \frac{\widetilde{\lambda}t(\widetilde{\lambda})[\pi(\phi_{\widetilde{\lambda } } ) -\phi_{\widetilde{\lambda}}(n)]}{\text{var}_\pi(\phi_{\widetilde{\lambda}})}>0,\ ] ] a contradiction .
hence , @xmath154 for @xmath142 .
the proof for @xmath55 is obvious from a direct computation and we deal with the case @xmath58 , here . by the equivalence of proposition [ p - main1 ] ( 3)-(4 ) , if @xmath60 , then @xmath56 for all @xmath57 . if @xmath62 , then @xmath155 for @xmath57 .
note that @xmath59 solves the system in ( [ eq - psigen ] ) . by ( [ eq - enu ] ) , this implies @xmath156}{\text{var}_\pi(\psi_k ) } , \quad\forall k\ge 0.\ ] ] the strict monotonicity of @xmath54 in ( 2 ) comes immediately from corollary [ c - t ] . in ( 3 ) , the continuity of ( [ eq - psigen ] ) in @xmath157 implies that @xmath158 is a solution to ( [ eq - psigen ] ) and @xmath159 . by proposition [ p - main1 ] , @xmath160 and @xmath66 , as desired .
recall the notation in the proof of corollary [ c - t ] : for @xmath79 , let @xmath131 be the function defined by ( [ eq - psigen ] ) and @xmath161 . by ( [ eq - enu ] ) and corollary [ c - t ] , @xmath162 for @xmath142 .
as @xmath163 is bounded , theorem [ t - main2 ] follows from lemma [ l - conv ] .
immediate from corollary [ c - t ] . in the end of this section , we use the following proposition to find how the shape of the function @xmath40 in ( [ eq - psigen ] ) evolves with @xmath157 . in proposition
[ p - shape ] , we set @xmath164 when @xmath40 is given by ( [ eq - psigen ] ) .
it is easy to see from ( [ eq - psigen ] ) that @xmath131 is strictly increasing before some constant , say @xmath165 , and then stays constant equal to @xmath166 after @xmath167 .
the proposition shows how the constant @xmath168 evolves .
[ p - shape ] for @xmath79 , let @xmath131 be the function generated by ( [ eq - psigen ] ) with @xmath134 and , for @xmath44 , set @xmath169 . for @xmath51 , let @xmath170 @xmath171 and let @xmath172 be the smallest root of @xmath173 .
then , * @xmath174 . * @xmath175 for @xmath176 and @xmath177 , where @xmath178 . * @xmath179 for @xmath180 . in particular , @xmath181 for @xmath182 and @xmath183 for @xmath184 with @xmath185 .
by lemma [ l - matt ] , @xmath186 and , for @xmath185 , @xmath187 where @xmath188 . note that if @xmath189 for some @xmath185 , then @xmath190\nu(j-1,j ) -\lambda\pi(j)\phi_\lambda(j)}{\nu(j , j+1)},\quad\forall 1\le j\le i.\ ] ] this implies @xmath191 multiplying @xmath192 and adding up @xmath193 yields @xmath194 from the above discussion , we conclude that if @xmath189 , then @xmath195 when @xmath196 , ( [ eq - ail2 ] ) implies @xmath197 for @xmath198 . by the continuity of @xmath199 and @xmath200 , if there is some @xmath198 such that @xmath189 , then @xmath201 for @xmath198 . as a consequence of ( [ eq - ta ] ) with @xmath202
, this will imply @xmath181 for @xmath198 .
hence , it remains to show that @xmath189 for some @xmath198 . to see this , according to corollary [ c - t ] , one can choose a constant @xmath203 such that @xmath204 . since @xmath205 is non - decreasing in @xmath118 , we obtain @xmath206 , as desired .
this proves @xmath181 for @xmath198 .
in particular , @xmath207 for @xmath208 . by corollary [ c - t ] , we have @xmath209 .
this proves proposition [ p - shape ] ( 1 ) .
next , observe that , for @xmath210 , @xmath211 by ( [ eq - phil ] ) , it is easy to see that @xmath212\nu(i , i+1)=-\lambda t_i(\lambda)>0 $ ] and @xmath213\nu(i+1,i+2)\notag\\ = & \left\{[\phi_\lambda(i+1)-\phi_\lambda(i)]\nu(i , i+1 ) -\lambda\pi(i+1)\phi_\lambda(i+1)\right\}^+\notag\\ = & \{-\lambda t_{i+1}(\lambda)\}^+=0.\notag\end{aligned}\ ] ] this proves proposition [ p - shape ] ( 2 ) . to prove proposition [ p - shape ]
( 3 ) , we use ( 1 ) to derive @xmath214 using ( [ eq - phil ] ) , this implies @xmath179 .
the last part of proposition [ p - shape ] follows easily from ( [ eq - ail2 ] ) and the fact that @xmath215 [ r - ini ] in proposition [ p - shape ] , if @xmath216 $ ] , then @xmath217 for @xmath218 .
note that , for @xmath219 , @xmath220 and @xmath221 by ( [ eq - l ] ) , this leads to @xmath222 $ ] for @xmath219 . in the case
@xmath55 , it is clear that @xmath223=\nu(1,2)[\pi(1)^{-1}+\pi(2)^{-1}]=\lambda^g_{\pi,\nu}$ ] .
in this section , we generalize the algorithms ( [ alg - sp ] ) and ( [ alg - sp2 ] ) so that they can be applied for the computation to any specified eigenvalue .
recall that @xmath15 is a graph with vertex set @xmath224 and edge set @xmath225 . given two positive measures @xmath5 on @xmath6 with @xmath7 ,
let @xmath17 be a @xmath0-by-@xmath0 matrix defined in the introduction and given by @xmath226/\pi(i)&\text{if } j = i\\0&\text{if } |i - j|>1\end{cases}.\ ] ] since @xmath8 is positive everywhere and @xmath17 is tridiagonal , all eigenvalues of @xmath17 have algebraic multiplicity @xmath39 . throughout this section ,
let @xmath227 denote the eigenvalues of @xmath17 with associated @xmath228-normalized eigenvectors @xmath229 . clearly , @xmath230 , @xmath231 and , for @xmath232 , @xmath233\nu(k-1,k)+[\zeta_i(k)-\zeta_i(k+1)]\nu(k , k+1).\ ] ] let @xmath185 .
as @xmath234 is non - constant , it is clear that @xmath235 and @xmath236 .
moreover , if @xmath237 for some @xmath238 , then @xmath239 and @xmath240 .
gantmacher and krein @xcite showed that there are exactly @xmath118 sign changes for @xmath234 with @xmath44 .
miclo @xcite gives a detailed description on the shape of @xmath234 as follows .
[ t - miclo2 ] for @xmath185 , let @xmath234 be an eigenvector associated to the @xmath118th smallest non - zero eigenvalue of the matrix in ( [ eq - m ] ) with @xmath241 . then , there are @xmath242 with @xmath243 such that @xmath234 is strictly increasing on @xmath244 $ ] for odd @xmath245 and is strictly decreasing on @xmath244 $ ] for even @xmath245 , and @xmath246 for @xmath247 . in the following ,
we make some analysis related to the euler - lagrange equations in ( [ eq - el2 ] ) .
[ def - type ] fix @xmath248 and let @xmath249 be a function on @xmath250 . for @xmath185 , @xmath249 is called `` type @xmath118 '' if there are @xmath251 satisfying @xmath243 such that * @xmath249 is strictly monotonic on @xmath244 $ ] for @xmath252 .
* @xmath253[f(a_{j+1})-f(a_{j+1}+1)]<0 $ ] for @xmath247 .
* @xmath254 , for @xmath247 , and @xmath255 , for @xmath256 .
the points @xmath257 will be called `` peak - valley points '' in this paper .
note that the difference between definition [ def - type ] and theorem [ t - miclo2 ] is the requirement @xmath258 , instead of @xmath259 . by theorem [ t - miclo2 ] , any eigenvector associated to the @xmath118th smallest non - zero eigenvalue of the matrix in ( [ eq - m ] ) must be of type @xmath118 with @xmath259 .
[ def - xil ] let @xmath5 be positive measures on @xmath6 with @xmath7 .
for @xmath260 , let @xmath261 be a function on @xmath250 defined by @xmath262 and , for @xmath263 , @xmath264\nu(k-1,k ) -\lambda\pi(k)\xi_\lambda(k)}{\nu(k , k+1)}.\ ] ] [ r - xi1 ] note that @xmath265 and , for @xmath266 , @xmath261 is strictly decreasing and of type @xmath39 . for @xmath79 , if @xmath267 , then @xmath268 and this implies @xmath269 .
similarly , if @xmath270 , then @xmath271 and @xmath272 .
thus , @xmath261 must be of type @xmath118 for some @xmath185 .
[ l - xi ] for @xmath79 , let @xmath261 be the function in definition [ def - xil ] .
suppose that @xmath261 is of type @xmath118 with @xmath185 .
* if @xmath273 , then there is @xmath274 such that @xmath275 is of type @xmath118 for @xmath276 . *
if @xmath277 , then there is @xmath274 such that @xmath275 is of type @xmath278 and @xmath279 is of type @xmath118 for @xmath280 .
let @xmath257 be the peak - valley points of @xmath261 . by the continuity of @xmath261 in @xmath157 and remark [ r - xi1 ]
, one can choose @xmath274 such that , for @xmath281 , @xmath275 remains strictly monotonic on @xmath244 $ ] for @xmath282 and @xmath283 [ \xi_{\lambda+\delta}(a_{j+1}+1)-\xi_{\lambda+\delta}(a_{j+1})]>0,\ ] ] for @xmath247 . in ( 1 ) , @xmath259 .
fix @xmath281 and set @xmath284 , @xmath285 . for @xmath286 , set @xmath287[\xi_{\lambda+\delta}(b_j)-\xi_{\lambda+\delta}(a_{j+1})]<0\\ b_j'=a_{j+1}'=a_{j+1 } & \text{if } [ \xi_{\lambda+\delta}(b_j-1)-\xi_{\lambda+\delta}(b_j)][\xi_{\lambda+\delta}(b_j)-\xi_{\lambda+\delta}(a_{j+1})]>0\\ b_j'=b_j,\,a_{j+1}'=a_{j+1}&\text{if } [ \xi_{\lambda+\delta}(b_j-1)-\xi_{\lambda+\delta}(b_j)][\xi_{\lambda+\delta}(b_j)-\xi_{\lambda+\delta}(a_{j+1})]=0 \end{cases}.\ ] ] clearly , @xmath275 is of type @xmath118 with peak - valley points @xmath288 .
this proves lemma [ l - xi ] ( 1 ) . for part ( 2 )
, we consider @xmath289 and @xmath290 . by similar argument as before , one can choose @xmath274 such that the restriction of @xmath275 to @xmath291 is of type @xmath118 for @xmath281 .
to finish the proof , it remains to compare @xmath292 and @xmath293 .
recall that @xmath294 as in the proof for proposition [ p - shape ] . using a similar reasoning as for ( [ eq - ta ] )
, one shows that @xmath181 for @xmath51 , where @xmath295 is the matrix in ( [ eq - ail ] ) .
this implies that the non - zero eigenvalues of @xmath17 , say @xmath296 , are the roots of @xmath297 . as a consequence of lemma [ l - matt ] , @xmath298 has exactly @xmath299 distinct roots , say @xmath300 , and they satisfy the interlacing property @xmath301 for @xmath302 .
note that @xmath303 and @xmath304 tend to infinity as @xmath305 tends to infinity .
this leads to the fact that if @xmath298 and @xmath306 , then @xmath307 is strictly decreasing in a neighborhood of @xmath157 . if @xmath298 and @xmath308 , then @xmath307 is strictly increasing in a neighborhood of @xmath157 . back to the proof of ( 2 ) .
suppose that @xmath309 . by remark [ r - xi1 ]
, it is easy to check that @xmath310 and @xmath311 or , equivalently , @xmath298 and @xmath306 . according to the conclusion in the previous paragraph
, we can find @xmath274 such that @xmath307 is strictly decreasing on @xmath312 , which yields @xmath313 this gives the desired property in lemma [ l - xi ] ( 2 ) .
the other case , @xmath314 , can be proved in the same way and we omit the details .
the following proposition characterizes the shape of @xmath261 for @xmath79 .
[ p - xi ] for @xmath79 , let @xmath261 be the function in _ definition [ def - xil]_. let @xmath315 be non - zero eigenvalues of @xmath17 in ( [ eq - m ] ) and @xmath316 be zeros of @xmath303 , where @xmath317 is the matrix in ( [ eq - ail ] ) .
then , * @xmath301 , for @xmath302 . *
@xmath261 is of type @xmath245 for @xmath318 $ ] and @xmath319 , where @xmath320 and @xmath321 .
\(1 ) is immediate from lemma [ l - matt ] . for ( 2 ) , note that @xmath322 is an eigenvalue of the submatix of @xmath17 obtained by removing the @xmath0th row and column .
this implies @xmath323 for @xmath324 and @xmath273 for @xmath79 and @xmath325 . by lemma [ l - xi ]
, @xmath326 is of type @xmath118 for @xmath327 . given @xmath79 , the above proposition provides a simple criterion to determine to which of the intervals @xmath328 $ ] @xmath157 belongs to , that is , the type of @xmath261 .
however , knowing the type of @xmath261 is not sufficient to determine whether @xmath157 is bigger or smaller than @xmath329 .
we need the following remark .
[ r - xil ] using the same argument as the proof of proposition [ p - shape ] , one can show that @xmath330 , where @xmath331 is the matrix in ( [ eq - ail ] ) . clearly , @xmath332 has zeros @xmath296 and tends to minus infinity as @xmath157 tends to minus infinity
this implies that @xmath333 , for @xmath334 , and @xmath335 for @xmath336 , where @xmath337 . as a consequence of proposition [ p - xi ] and remark [ r - xil ] ,
we obtain the following dichotomy algorithm , which is a generalization of ( [ alg - sp2 ] ) .
let @xmath185 .
@xmath338 [ t - di ] referring to ( [ alg - evi2 ] ) , @xmath339 immediate from proposition [ p - xi ] and remark [ r - xil ] . proposition [ p - xi ] ( 2 ) bounds the eigenvalues using the shape of @xmath261 generated from one end point .
we now introduce some other criteria to bound eigenvalues using the shape of @xmath261 from either boundary point .
those results will be used to prove theorem [ t - lower ] .
[ p - twosided ] for @xmath79 , let @xmath261 be the function in definition [ def - xil ] and @xmath340 be a function given by @xmath341\nu(k , k+1)-\lambda\pi(k)\widetilde{\xi}_\lambda(k)}{\nu(k-1,k)},\ ] ] for @xmath342 with @xmath343 .
let @xmath344 be eigenvalues of @xmath17 in ( [ eq - m ] ) and let @xmath345 be the restriction of @xmath249 to a subset @xmath346 of @xmath13 .
suppose @xmath347 .
* if @xmath348 is of type @xmath118 with @xmath349 and @xmath350 is of type @xmath245 with @xmath351 , then @xmath352 . * if @xmath348 is of type @xmath118 with @xmath353 and @xmath350 is of type @xmath245 with @xmath354 , then @xmath355 .
* if @xmath348 is of type @xmath118 with @xmath349 and @xmath350 is of type @xmath245 with @xmath354 , then @xmath356 . by proposition [ p - xi ]
, @xmath357 is a polynomial of degree
@xmath46 satisfying @xmath358 this implies that there are @xmath359 , @xmath360 , such that @xmath361 for @xmath362 and @xmath363 with @xmath364 and @xmath365 .
the proofs for ( 1)-(3 ) in proposition [ p - twosided ] are similar and we deal with ( 1 ) only . by the euler - lagrange equations in ( [ eq - el2 ] ) ,
it is easy to see that , for @xmath366 , @xmath367 and @xmath368 are eigenvectors of @xmath17 in ( [ eq - m ] ) associated with @xmath369 , which implies @xmath370 .
first , assume that @xmath371 .
by proposition [ p - xi ] , @xmath372 is of type at least @xmath118 and @xmath373 is of type at least @xmath245 .
this implies that the patching of @xmath372 and @xmath374 , which equals to @xmath375 , is of type at least @xmath376 .
this is a contradiction .
next , assume that @xmath377 . by proposition [ p - xi ] , we may choose @xmath378 ( resp .
@xmath379 ) such that @xmath380 ( resp .
@xmath350 ) changes the type at @xmath381 ( resp .
@xmath382 ) . if @xmath383 , then a similar reasoning as before implies that @xmath384 is of type at most @xmath385 , a contradiction . if @xmath386 , then exactly one of @xmath387 and @xmath388 does not change its type .
this implies that the gluing point @xmath389 can not be a local extremum and , thus , the patching function is of type at most @xmath385 , another contradiction ! according to the discussion in the first paragraph of this proof , if @xmath390 , then none of @xmath387 and @xmath388 changes type nor , of course , the sign at @xmath389 .
consequently , we obtain @xmath391 , which contradicts the fact @xmath392 .
[ p - signchange ] for @xmath79 and @xmath393 , let @xmath394 be the @xmath395th sign change of @xmath261 defined by @xmath396 and @xmath397 , where @xmath398 .
then , for @xmath399 , @xmath400 for all @xmath393
. let @xmath393 .
if @xmath401 , then it is clear that @xmath400 .
suppose that @xmath402 . obviously , @xmath403 is of type @xmath395 .
referring to ( [ eq - ail ] ) , let @xmath404 be the roots of @xmath405 and @xmath406 be roots of @xmath407 .
according to the first paragraph of the proof for proposition [ p - twosided ] , there are @xmath408 with @xmath409 such that @xmath410 for @xmath411 and @xmath409 , where @xmath412 . since @xmath413 , one has @xmath414 . as it is assumed that @xmath415 , if @xmath416
, then @xmath417 is of type at least @xmath418 and , consequently , @xmath419 .
if @xmath420 , then @xmath417 is type @xmath395 and @xmath421 .
this implies @xmath422 , as desired . motivated by theorem [ t - miclo2 ]
, we introduce another scheme generalizing ( [ eq - psigen ] ) to bound the other eigenvalues of @xmath17 from below .
[ def - xii ] for @xmath79 , let @xmath261 be a function in definition [ def - xil ] . if @xmath261 is of type @xmath118 , @xmath185 , with peak - valley points @xmath251 , then define @xmath423 and set @xmath424 for @xmath425 .
[ r - xi2 ] for @xmath79 , if @xmath261 is of type @xmath118 , then @xmath426 is of type @xmath245 for @xmath427 .
moreover , for @xmath428 , @xmath429\nu(k-1,k)-\lambda\pi(k)\xi^{(j)}_\lambda(k)}{\nu(k , k+1)}\notag\\ & = \xi^{(j)}_\lambda(k)-\frac{\lambda[\pi(1)\xi^{(j)}_\lambda(1)+\cdots+\pi(k)\xi^{(j)}_\lambda(k)]}{\nu(k , k+1)},\notag\end{aligned}\ ] ] and , for @xmath430 , @xmath431\nu(k-1,k)-\lambda\pi(k)\xi^{(j)}_\lambda(k))}{\nu(k , k+1)},\ ] ] where @xmath432 if @xmath245 is odd , and @xmath433 if @xmath245 is even . note that @xmath434 is exactly @xmath131 in proposition [ p - shape ] .
thereafter , let @xmath435 and @xmath436 be functions on @xmath437 defined by @xmath438 where @xmath261 and @xmath439 are functions in definitions [ def - xil]-[def - xii ] .
[ r - lli ] note that @xmath440 . by a similar reasoning as in the proof for ( [ eq - enu ] ) , one can show that , for @xmath79 , @xmath441 } { \textnormal{var}_\pi(\xi_\lambda)},\quad \mathcal{l}^{(i)}(\lambda)=\lambda+\frac{\lambda\pi(\xi^{(i)}_\lambda)[\pi(\xi^{(i)}_\lambda ) -\xi^{(i)}_\lambda(n)]}{\textnormal{var}_\pi(\xi^{(i)}_\lambda)}.\ ] ] from proposition [ p - xi ] , it followss immediately that @xmath442 for @xmath443 $ ] . to explore further @xmath435 and @xmath436
, we need more information of @xmath332 , @xmath444 , @xmath445 and @xmath446 .
[ l - xi2 ]
let @xmath261 be the function in _ definition [ def - xil ] _ and @xmath447 be constants in _ proposition [ p - xi]_. then , @xmath448 has @xmath46 distinct roots , say @xmath449 , which satisfy @xmath450 and @xmath451 for @xmath177 .
furthermore , @xmath452 for @xmath453 and @xmath454 for @xmath455 , with @xmath456 and @xmath457 . set @xmath458 .
according to definition [ def - xil ] , @xmath459 is a polynomial of degree @xmath46 and satisfies @xmath460 .
note that @xmath461 for @xmath462 .
if @xmath118 is odd , then @xmath463 .
this implies @xmath464 and , hence , @xmath465 .
similarly , if @xmath118 is even , then @xmath466 . by lemma [ l - xi ] and proposition [ p - xi ] , if @xmath467 with odd @xmath118 , then @xmath468 is of type @xmath118 with @xmath323 .
this implies @xmath469 and @xmath470 , which yields @xmath471 .
similarly , one can show that @xmath472 if @xmath118 is even .
[ r - xii ] we consider the sign of @xmath444 and @xmath446 in this remark . by proposition [ p - xi ] , @xmath473 for @xmath474 .
if @xmath475 with @xmath177 , then @xmath261 is of type @xmath245 with @xmath476 .
fix @xmath177 and set @xmath477 .
clearly , @xmath478 for @xmath475 .
observe that , for @xmath475 with odd @xmath118 , @xmath479 , which implies @xmath480 and @xmath481 .
a similar reasoning for the case of even @xmath118 gives @xmath482 and @xmath483 .
consequently , we obtain @xmath484>0,\ ] ] for @xmath475 and @xmath177 .
note that , by proposition [ p - xi ] , @xmath473 for @xmath474 .
in addition with remark [ r - xil ] , lemma [ l - xi2 ] and the continuity of @xmath439 , the first inequality of ( [ eq - xii ] ) holds for @xmath485 and the second inequalities of ( [ eq - xii ] ) hold for @xmath486 . according to lemma [ l - xi2 ] and remark [ r - xii ]
, we derive a generalized version of proposition [ p - main1 ] in the following .
[ p - lli ] let @xmath58 and @xmath185 .
for @xmath79 , let @xmath487 be the functions in _ definition [ def - xil ] _ and @xmath488 be the constants in _ lemma [ l - xi2]_. * for @xmath486 , the following are equivalent . * * @xmath489 . * * @xmath490 . * * @xmath491 . * for @xmath492 ,
the following are equivalent .
* * @xmath493 . * * @xmath461 . *
* @xmath491 .
the proof for proposition [ p - lli ] ( 2 ) is similar to the proof for proposition [ p - lli ] ( 1 ) and we deal only with the latter . by lemma [ l - xi2 ] and remark [ r - xii ] , one has @xmath494\begin{cases}<0&\text{for } \lambda>\lambda^g_i\\>0&\text{for } \beta_{i-1}<\lambda<\lambda^g_i\end{cases}.\ ] ] this proves the equivalence of ( 1 - 1 ) and ( 1 - 2 ) . under the assumption of ( 1 - 2 ) and using remark [ r - xil ]
, one has @xmath474 .
this implies @xmath473 is an eigenvector for @xmath17 with associated eigenvalue @xmath157 . as @xmath495 $ ]
, it must be the case @xmath491 .
this gives ( 1 - 3 ) , while ( 1 - 3)@xmath122(1 - 2 ) is obvious and omitted .
it is worthwhile to note that if ( 1 - 1 ) and ( 2 - 1 ) of proposition [ p - lli ] are removed , then the equivalence in ( 1 ) holds for @xmath496 and the equivalence in ( 2 ) holds for @xmath497 . once @xmath498 is known , we can determine @xmath329 using the sign of @xmath444 .
see theorem [ t - xi ] for details .
note that condition ( 4 ) of proposition [ p - main1 ] is not included in proposition [ p - lli ] .
in fact , the equivalence may fail , that is , there may exist some @xmath499 such that @xmath500 .
see example [ ex - ehrenfest ] for a counterexample . as proposition [ p - lli ] focuses on the characterization of zeros of @xmath501 , the following theorem concerns the sign of @xmath501 .
[ t - xi ] let @xmath502 be the constants in _ proposition [ p - xi ] _ and _ lemma [ l - xi2 ] _ , and @xmath435 be the function in ( [ eq - lli ] ) .
then , @xmath503 are fixed points of @xmath435 and , for @xmath177 , * @xmath504 for @xmath505 .
* @xmath506 for @xmath507 . * @xmath508 for @xmath509 .
immediate from lemma [ l - xi2 ] and remarks [ r - lli]-[r - xii ] . by theorem [ t - xi ] , we obtain a lower bound on any specified eigenvalue of @xmath17 .
[ c - lli ] let @xmath185 and @xmath510 .
consider the sequence @xmath511 with @xmath512 and set @xmath513 where @xmath514 .
then , @xmath515 .
it is not clear yet whether the sequence @xmath516 in corollary [ c - lli ] is convergent , even locally .
this subject will be discussed in the next subsection .
now , we establish some relations between the roots of @xmath173 and the shape of @xmath439 . this is a generalization of proposition [ p - shape ] .
[ p - shape2 ] for @xmath185 , let @xmath295 be the matrix in ( [ eq - ail ] ) , @xmath517 be zeros of @xmath173 and set @xmath518 . referring to the notation in _ proposition [ p - xi ]
_ , it holds true that , for @xmath185 , * @xmath519 . * @xmath520 for @xmath521 and @xmath522 . * @xmath523 for @xmath524 and @xmath525 .
the order in ( 1 ) is a simple application of lemma [ l - mat ] . for ( 2 ) ,
fix @xmath185 and set @xmath526 and @xmath527 , @xmath528 .
clearly , @xmath529 .
we use the notation @xmath530 to denote the restriction of @xmath261 to a set @xmath531 .
suppose that @xmath118 is odd . by remark [ r - xi2 ] ,
@xmath473 on @xmath532 and @xmath533 is of type @xmath118 with @xmath534 by lemma [ l - xi](1 ) , if @xmath535 , then there is @xmath274 such that , for @xmath536 , @xmath537 is of type @xmath118 and @xmath538 this implies @xmath539 for @xmath281 . by lemma [ l - xi](2 ) ,
if @xmath540 , then there is @xmath274 such that , for @xmath541 , @xmath542 is of type @xmath118 with @xmath543 and , for @xmath544 , @xmath542 is of type @xmath278 with @xmath538 this yields @xmath539 for @xmath544 and @xmath545 for @xmath541 .
the proof for the case of even @xmath118 is similar and we conclude from the above that @xmath546 is a non - increasing and right - continuous function taking values on @xmath547 .
let @xmath548 be the discontinuous points of @xmath546 such that @xmath549 for @xmath550 . as a consequence of the above discussion
, @xmath551 is of type @xmath118 with @xmath552 and this implies @xmath553 .
that means @xmath554 is a root of @xmath555 for @xmath556 . by proposition [ p - xi ] and the second equality in ( 1 ) , @xmath557 for @xmath558 and , thus , @xmath559 for @xmath560 . as a consequence of the interlacing relationship @xmath561
, it must be @xmath562 for @xmath550 .
this finishes the proof .
[ r - shape ] for @xmath185 , @xmath563 are also non - zero eigenvalues of the @xmath564 principal submatrix of ( [ eq - m ] ) indexed by @xmath565 .
[ r - shape2 ] in fact , by proposition [ p - shape ] , @xmath566 for @xmath567 , which is better than proposition [ p - shape2](3 ) .
this subsection is dedicated to the local convergence of @xmath435 in ( [ eq - lli ] ) .
let @xmath568 be the constants in proposition [ p - xi ] and lemma [ l - xi2 ] .
as before , let @xmath569 denote the @xmath228-normalized eigenvectors of @xmath17 associated with @xmath570 .
clearly , @xmath571 and @xmath572 , where @xmath573 for @xmath363 .
note that @xmath574 is a polynomial of degree @xmath46 and satisfies @xmath575 for @xmath576 .
this implies @xmath577 for all @xmath185 .
moreover , by multiplying ( [ eq - el2 ] ) with @xmath578 and summing up @xmath395 , we obtain @xmath579 . in the same spirit
, one can show that @xmath580 $ ] using definition [ def - xil ] . putting both equations together yields @xmath581 as a consequence of remark [ r - lli ] , this gives @xmath582 for @xmath583 .
the next proposition follows immediately from the second equation in ( [ eq - rhoi1 ] ) and ( [ eq - rhoi2 ] ) .
[ p - zeta ] let @xmath296 be the non - zero eigenvalues of @xmath17 in ( [ eq - m ] ) and @xmath584 be the corresponding @xmath228-normalized eigenvectors . then , @xmath585 set @xmath586 . by theorem [ t - xi ] ,
@xmath587 are zeros of @xmath588 , which is a polynomial of degree @xmath299 .
this implies @xmath589 where @xmath590 .
putting this back to @xmath435 yields @xmath591 for @xmath592 .
p - l ] let @xmath435 be the function in ( [ eq - lli ] ) , @xmath329 be the eigenvalue of @xmath17 and @xmath488 be the constant in _ lemma [ l - xi2]_. let @xmath593 with @xmath185 . then , for @xmath594 , * if @xmath595 , then there is @xmath596 such that @xmath435 is strictly increasing on @xmath597 and strictly decreasing on @xmath598 . *
if @xmath599 , then there is @xmath600 such that @xmath435 is strictly increasing on @xmath601 and strictly increasing on @xmath602 . * if @xmath603 , then @xmath435 is strictly increasing on @xmath604 . using ( [ eq - lsp ] ) and ( [ eq - lsp2 ] ) , one can show that @xmath605 and @xmath606=2d_i.\ ] ] to prove ( 1 ) and ( 2 ) , it suffices to show that if @xmath607 for some @xmath596 , then @xmath608 is a local minimum of @xmath435 , and if @xmath609 for some @xmath600 , then @xmath610 is a local maximum of @xmath435 .
we discuss the first case , whereas the second case is similar and is omitted .
recall that @xmath611 . as @xmath608 is a critical point for @xmath435 ,
one has @xmath612 .
this implies @xmath613}{2(u'(\tau))^3}>0,\ ] ] where the last inequality uses the fact that @xmath614 , for @xmath505 , and @xmath615 ^ 2<0.\ ] ] this proves ( 1 ) and ( 2 ) . to see ( 3 )
, we assume that @xmath603 .
computations show that @xmath616\notag\\ = & ( \lambda-\lambda^g_i)^2\left[\sum_{j=1,j\ne i}^{n-1}\frac{1}{(\lambda^g_j-\lambda)(\lambda^g_j-\lambda^g_i)}-\sum_{j=1}^{n-1}\frac{1}{(\beta_j-\lambda)(\beta_j-\lambda^g_i)}\right]<0,\notag\end{aligned}\ ] ] for @xmath617 , where the last inequality uses the fact that @xmath618 for @xmath427 and @xmath619 for @xmath476 . by theorem [ t - xi ] , this implies @xmath620 for @xmath505 and @xmath621 for @xmath622 .
the desired property comes immediate from the discussion in the previous paragraph .
note that @xmath623 and @xmath624 . using the same proof as above
, this implies that @xmath625 is strictly increasing on @xmath626 .
moreover , by ( [ eq - lsp ] ) , one may compute @xmath627 this implies @xmath625 is strictly decreasing on @xmath628 and @xmath629 the following local convergence is a simple corollary of theorem [ t - xi ] and proposition [ p - l ] .
[ t - local ] let @xmath630 and set @xmath631 for @xmath512 .
then , there is @xmath274 such that the sequence @xmath632 is monotonic and converges to @xmath329 for @xmath633 and @xmath185 .
we use the following examples to illustrate the different cases in proposition [ p - l ] .
[ ex - sym ] let @xmath634 .
a simple random walk on @xmath250 with reflecting probability @xmath635 at the boundary is a birth and death chain with transition matrix given by @xmath636 for @xmath637 .
it is easy to see that the uniform probability is the stationary distribution of @xmath26 . in the setting of graph
, we have @xmath638 and @xmath639 .
one may apply the method in @xcite to obtain the following spectral information .
@xmath640 see , e.g. , ( * ? ? ?
* section 7 ) . by ( [ eq - l ] ) , we get @xmath641 clearly , @xmath623 and @xmath624 . if @xmath0 is even , then @xmath642 .
[ ex - ehrenfest ] an ehrenfest chain on @xmath643 is a markov chain with transition matrix @xmath26 given by @xmath644 and @xmath645 for @xmath646 .
the associated stationary distribution is the unbiased binomial distribution on @xmath13 , that is , @xmath647 for @xmath648 . to the ehrenfest chain , the measure @xmath8 is defined by @xmath649 for @xmath646 . using the group representation for the binary group @xmath650
, one may compute @xmath651 plugging this back into ( [ eq - l ] ) yields @xmath652 this example points out the possibility of different signs in @xmath653 including @xmath47 . in this subsection
, we give a new proof of a result , theorem [ t - sep ] , which deals with convergence in separation distance for birth and death chains .
let @xmath654 be a birth and death chain with transition matrix @xmath26 given by ( [ eq - bdc ] ) .
in the continuous time setting , we consider the process @xmath655 , where @xmath656 is a poisson process with parameter @xmath39 independent of @xmath657 .
given the initial distribution @xmath658 , which is the distribution of @xmath659 , the distributions of @xmath657 and @xmath660 are respectively @xmath661 and @xmath662 , where @xmath663 .
briefly , we write @xmath664 .
it is well - known that if @xmath26 is irreducible , then @xmath665 converges to @xmath9 as @xmath666 .
if @xmath26 is irreducible and @xmath667 for some @xmath118 , then @xmath661 converges to @xmath9 as @xmath668 . concerning the convergence , we consider the separations of @xmath669 with respect to @xmath9 , which are defined by @xmath670 the following theorem is from @xcite . [ t - sep ]
let @xmath26 be an irreducible birth and death chain on @xmath36 with eigenvalues @xmath671 .
* for the discrete time chain , if @xmath672 for all @xmath31 , then @xmath673 * for the continuous time chain , it holds true that @xmath674 diaconis and fill @xcite introduce the concept of dual chain to express the separations in theorem [ t - sep ] as the probability of the first passage time .
brown and shao @xcite characterize the first passage time using the eigenvalues of @xmath26 for a special class of continuous time markov chains including birth and death chains .
the idea in @xcite is also applicable for discrete time chains and this leads to the formula above . see @xcite for further discussions . here , we use proposition [ p - zeta ] and lemma [ l - bdc ] to prove this result directly .
[ l - bdc ] let @xmath26 be the transition matrix in ( [ eq - bdc ] ) with stationary distribution @xmath9 .
suppose that @xmath658 is a probability distribution satisfying @xmath675 for all @xmath363 .
* for the discrete time chain , if @xmath672 for all @xmath31 , then @xmath676 for all @xmath31 and @xmath677 .
* for the continuous time chain , @xmath678 for all @xmath31 and @xmath679 .
note that ( 2 ) follows from ( 1 ) if we write @xmath680 . for the proof of ( 1 ) , observe that @xmath681 by induction , if @xmath676 for @xmath31 , then @xmath682 lemma [ l - bdc ] is also developed in @xcite in which it is shown that , for any non - negative function @xmath249 , @xmath683 is non - decreasing if @xmath249 is non - decreasing for all @xmath677 .
consider the adjoint chain @xmath684 of @xmath26 in @xmath228 . as
birth and death chains are reversible , one has @xmath685 . using the identity @xmath686 , it is easy to see that the above proof is consistent with the proof in @xcite .
assume that @xmath26 is irreducible and let @xmath671 be the eigenvalues of @xmath35 with @xmath228-normalized eigenvector @xmath687 .
by lemma [ l - bdc ] , if @xmath658 satisfies @xmath688 for @xmath31 , then @xmath689 where @xmath690 .
if @xmath26 satisfies @xmath672 for all @xmath31 , then @xmath691 by proposition [ p - zeta ] , setting @xmath658 to be one of the dirac measure @xmath692 leads to the desired identities .
in this section , the graph @xmath2 under consideration is infinite with @xmath693 and @xmath694 . as before , let @xmath5 be positive measures on @xmath6 satisfying @xmath7 .
the dirichlet form and the variance are defined in a similar way as in the introduction and the spectral gap of @xmath15 with respect to @xmath5 is given by @xmath695 for @xmath696 , let @xmath697 be the subgraph of @xmath15 with @xmath698 , @xmath699 and let @xmath700 be normalized restrictions of @xmath5 to @xmath701 .
that is , @xmath702 , @xmath703 with @xmath704 $ ] . as
before , let @xmath17 be an infinite matrix indexed by @xmath13 and defined by @xmath705 clearly , @xmath706 is the principal submatrix of @xmath17 indexed by @xmath707 .
[ l - gapinf ] referring to the above setting , @xmath708 for @xmath634 and @xmath709 .
briefly , we write @xmath157 for @xmath21 and @xmath710 for @xmath711 . note that @xmath710 is the smallest non - zero eigenvalue of the principal submatrix of @xmath17 indexed by @xmath707 . as a consequence of proposition [ p - shape2](1 ) and remark [ r - shape ] , @xmath712 .
for @xmath634 , let @xmath713 be a minimizer for @xmath710 and define @xmath714 for @xmath336 . clearly , one has @xmath715 and @xmath716 .
this implies @xmath717 for @xmath696 .
let @xmath718 .
note that it remains to show @xmath719 . for @xmath274
, choose a function @xmath249 on @xmath13 such that @xmath720 with @xmath721 . for @xmath722 , we choose @xmath723 such that @xmath724 and @xmath725 , where @xmath726 , the restriction of @xmath249 to @xmath727 .
this implies @xmath728 letting @xmath729 and then @xmath730 yields @xmath731 , as desired .
silver @xcite contains a discussion of the ( weak * ) convergence of the spectral measure for @xmath732 to the spectral measure for @xmath15 in a very general setting .
lemma [ l - gapinf ] can also be proved using theorem 4.3.4 in @xcite .
[ p - shapeinf ] for @xmath79 , let @xmath134 and @xmath733\nu(i-1,i)-\lambda\pi(i)\phi_\lambda(i)\}^+}{\nu(i , i+1)},\,\forall i\ge 1.\ ] ] set @xmath734 and @xmath735 for @xmath696 .
* for @xmath736 and @xmath737 , @xmath738 .
* for @xmath739 $ ] , @xmath740 for all @xmath336 .
immediate from proposition [ p - shape2 ] and remarks [ r - shape]-[r - shape2 ] . by proposition [ p - shapeinf ]
, one may generate a dichotomy algorithm for @xmath21 using the shape of @xmath131 .
see ( [ alg - evi2 ] ) .
the following theorem extends theorem [ t - miclo ] to infinite paths .
[ t - elinf ] if @xmath85 and @xmath107 for some function @xmath40 on @xmath13 with @xmath42 , then @xmath40 is strictly monotonic and satisfies @xmath741\nu(i , i+1)+[\psi(i)-\psi(i-1)]\nu(i-1,i),\,\,\forall i\ge 1.\ ] ] [ t - gapinf ] for @xmath79 , let @xmath131 be the function in proposition [ p - shapeinf ] and set @xmath742 . then , * @xmath743 for @xmath744 . * @xmath745 as @xmath746 for @xmath744 .
let @xmath142 .
by lemma [ l - gapinf ] , @xmath747 for some @xmath736 . by proposition [ p - shapeinf ]
( 1 ) , one has @xmath738 . as in ( [ eq - enu ] ) , we obtain @xmath748}{\text{var}_\pi(\phi_\lambda)},\quad \sum_{j=1}^i\phi_\lambda(j)\pi(j)\ge 0.\ ] ] this leads to @xmath749 and @xmath750 , which implies @xmath751 .
that means @xmath163 has no fixed point on @xmath752 .
the lower bound of ( 1 ) follows immediately from theorem [ t - elinf ] . for ( 2 ) ,
set @xmath753 . as a consequence of ( 1 ) , @xmath163 is continuous on @xmath752 . if @xmath754 , then @xmath755 is a fixed point of @xmath163 , a contradiction !
hence , @xmath160 .
in this section , we illustrate the algorithm ( [ alg - sp2 ] ) on a specific metropolis chain .
the metropolis algorithm introduced by metropolis _ et al . _ in 1953 is a widely used construction that produces a markov chain with a given stationary distribution @xmath9 .
let @xmath9 be a positive probability measure on @xmath13 and @xmath26 be an irreducible markov transition matrix on @xmath13 . for simplicity , we assume that @xmath756 for all @xmath757 .
the metropolis chain evolves in the following way .
given the initial state @xmath758 , select a state , say @xmath759 , according to @xmath760 and compute the ratio @xmath761 .
if @xmath762 , then move to @xmath759 .
if @xmath763 , then flip a coin with probability @xmath764 on heads and move to @xmath759 if the head appears .
if the coin lands on tails , stay at @xmath758 .
accordingly , if @xmath765 is the transition matrix of the metropolis chain , then @xmath766 it is easy to check @xmath767 .
as @xmath26 is irreducible , @xmath765 is irreducible .
moreover , if @xmath9 is not uniform , then @xmath768 for some @xmath769 .
this implies that @xmath765 is aperiodic and , consequently , @xmath770 and @xmath771 as @xmath666 . for further information on metropolis chains ,
see @xcite and the references therein . for @xmath248 ,
let @xmath697 be a graph with @xmath772 and @xmath773 .
suppose that @xmath774 is the transition matrix of the simple random walk on @xmath775 , that is , @xmath776 and @xmath777 for all @xmath778 .
for @xmath779 , let @xmath780 be probabilities on @xmath772 given by @xmath781 where @xmath782 and @xmath783 are normalizing constants .
it is easy to compute that @xmath784 where @xmath785 the metropolis chains , @xmath786 and @xmath787 , for @xmath788 and @xmath789 based on the simple random walk @xmath774 have transition matrices given by @xmath790 and @xmath791\\\frac{i^a}{2(i+1)^a}&\text{if } j = i-1,i\in[1,n]\\\frac{(i+1)^a - i^a}{2(i+1)^a}&\text{if } j = i , i\notin\{0,n\}\\1-\frac{n^a}{2(n+1)^a}&\text{if } i = j = n\end{cases}\ ] ] and @xmath792\\\frac{(n - i)^a}{2(n - i+1)^a}&\text{if } j = i+1,i\in[0,n-1]\\\frac{(n - i+1)^a-(n - i)^a}{2(n - i+1)^a}&\text{if } j = i\ne 0\\1-\frac{n^a}{(n+1)^a}&\text{if }
i = j=0\end{cases}.\ ] ] saloff - coste @xcite discussed the above chains and obtained the correct order of the spectral gaps .
let @xmath793 denote the spectral gaps of @xmath794 . referring to the recent work in @xcite
, one has @xmath795 where @xmath796 is any of @xmath797 and @xmath798 , and @xmath799 and @xmath800 [ t - metropolis ] let @xmath793 be spectral gaps for @xmath794 .
then , @xmath801 and @xmath802 where @xmath803 .
the bound for @xmath804 follows immediately from the fact @xmath805 for @xmath806 , note that @xmath807 taking @xmath808 yields the upper bound . for the lower bound , we write @xmath809 for @xmath810 ,
it is clear that @xmath811 observe that , for @xmath779 , @xmath812 where @xmath813 it is clear that , for @xmath810 , @xmath814 and this leads to @xmath815 summarizing all above gives the desired lower bound . in theorem [ t - metropolis ] in order from the top @xmath816 and @xmath817 .
the right most point corresponds to a path of length @xmath818.,width=302 ] .these numbers denote @xmath819 in theorem [ t - metropolis ] . [ cols="^,^,^,^,^,^",options="header " , ] [ r - sym ] comparing with ( * ? ? ?
* theorem 9.5 ) , the bounds for @xmath804 given in theorem [ t - metropolis ] have a similar lower bound and an improved upper bound by a multiple of about @xmath820 . for @xmath806 , observe that @xmath821 where @xmath822 recall the constant @xmath823 in the proof of theorem [ t - metropolis ] .
note that @xmath824 and , for @xmath779 , @xmath825 and @xmath58 , @xmath826 where the last inequality is obtained by considering the subcases @xmath827 and @xmath828 .
the above computation also applies for @xmath829 and @xmath830 . in the same spirit
, one can show that @xmath831 .
this yields @xmath832 hence , we have @xmath833 . as a consequence of ( [ eq - checkna ] ) and ( [ eq - cina ] )
, we obtain that , uniformly for @xmath779 , @xmath834 where @xmath835 and @xmath836 for @xmath825 .
[ r - sym2 ] note that the lower bound in theorem [ t - lower ] provides the correct order of the spectral gap for the chain @xmath786 uniformly in @xmath837 but not for @xmath787 . for instance , if @xmath837 grows with @xmath0 , say @xmath838 , then theorem [ t - lower ] implies @xmath839 , while ( [ eq - cna ] ) gives @xmath840 .
consider the chain in theorem [ t - metropolis ] .
a numerical experiment of algorithm ( [ alg - sp2 ] ) is implemented and the data is collected in figure [ fig1 ] and table [ tab1 ] .
one may conjecture that @xmath841 as @xmath746 , where @xmath842 is a constant depending on @xmath837 .
in this section , we discuss some examples of special interests and show how the theory developed in the previous sections can be used to bound the spectral gap . in the first subsection , we develop a lower bound on the spectral gap in a very general setting using the theory in section 3 . in the second subsection , we focuses on the case of one bottleneck , where a precise estimation on the spectral gap is presented .
those computations are based on the theoretical work in section 2 . in the third subsection
, we consider the case of multiple bottlenecks in which the exact order of the spectral gap is determined for some special classes of chains . in
what follows , we will use the notation @xmath843 to represent the summation @xmath844 for any measure @xmath9 on @xmath13 and any set @xmath845 . given two sequences of positive reals @xmath846
, we write @xmath847 if @xmath848 is bounded . if @xmath847 and @xmath849 , we write @xmath850 . if @xmath851 , we write @xmath852 . in this subsection , we give a lower bound on the spectral gap in the general case .
[ t - lower ] let @xmath2 be a graph with vertex set @xmath643 and edge set @xmath853 .
let @xmath5 be positive measures on @xmath6 with @xmath7 .
then , @xmath854)}{\nu(j , j+1)}\right)^{-1}\wedge\left(\sum_{j = i+1}^{n } \frac{\pi([j , n])}{\nu(j-1,j)}\right)^{-1}\right\},\ ] ] where @xmath855 .
[ r - lower ] let @xmath531 be the lower of the spectral gap in theorem [ t - lower ] . note that , for any positive reals , @xmath856 .
using this fact , it is easy to see that @xmath857 , where @xmath858)}{\nu(j , j+1)}+\sum_{j = i+1}^{n } \frac{\pi([j , n])}{\nu(j-1,j)}\right)^{-1}.\ ] ] in particular , if @xmath167 is the median of @xmath9 , that is , @xmath859)\ge 1/2 $ ] and @xmath860)\ge 1/2 $ ] , then @xmath861)}{\nu(j , j+1)}+\sum_{j = i_0 + 1}^{n } \frac{\pi([j , n])}{\nu(j-1,j)}\right)^{-1}.\ ] ] let @xmath654 be an irreducible birth and death chain on @xmath36 with birth rate @xmath23 , death rate @xmath24 and holding rate @xmath25 as in ( [ eq - bdc ] )
. for @xmath862 , set @xmath863 as the first passage time to state @xmath118 . by the strong markov property ,
the expected hitting time to @xmath118 started at @xmath47 can be expressed as @xmath864)}{p_j\pi(j)},\quad\mathbb{e}_n\tau_i=\sum_{j = i+1}^n\frac{\pi([j , n])}{q_j\pi(j)},\ ] ] where @xmath9 is the stationary distribution of @xmath654 .
let @xmath157 be the spectral gap for @xmath654 .
then , @xmath95 , where @xmath15 is the path with vertex set @xmath865 and @xmath866 for @xmath31 .
the conclusion of theorem [ t - lower ] can be written as @xmath867 .
the lower bound in theorem [ t - lower ] is not necessary the right order of the spectral gap .
see remark [ r - sym2 ] . for @xmath79 ,
let @xmath261 be the function in definition [ def - xil ] .
that is , @xmath868 and , for @xmath869 , @xmath870\nu(i , i+1)=[\xi_\lambda(i)-\xi_\lambda(i-1)]\nu(i-1,i)-\lambda\pi(i)\xi_\lambda(i).\ ] ] inductively , one can show that if @xmath871)/\nu(j , j+1)]$ ] , then @xmath872)/\nu(i , i+1),\\-1\le\xi_\lambda(i+1)\le -1+\lambda\sum_{j=1}^{i}[\pi([0,j])/\nu(j , j+1)]<0,\end{cases}\ ] ] for @xmath873 .
one may do a similar computation from the other end point and , by proposition [ p - twosided ] , this implies @xmath874)}{\nu(j , j+1)},\sum_{j=1}^{n-\ell}\frac{\pi([n - j+1,n])}{\nu(n - j , n - j+1)}\right\}.\ ] ] taking the minimum over @xmath875 gives the desired inequality . for @xmath248 ,
let @xmath697 be the path on @xmath36 and set @xmath876 and @xmath877 with @xmath878 . using feller s method in ( * ? ? ?
* chapter xvi.3 ) , one can show that the eigenvalues of @xmath706 are @xmath879 for @xmath862 .
[ t - bottleneck ] for @xmath248 , let @xmath880 , @xmath881 and set @xmath876 , @xmath882 then , the spectral gap are bounded by @xmath883 in particular , @xmath884 . the lower bound is immediate from theorem [ t - lower ] by choosing @xmath808 in the computation of the maximum . for the upper bound , we set @xmath885 and let @xmath886 be the function on @xmath887 defined by @xmath888 and , for @xmath889 , @xmath890\nu_{n - x_n}(i-1,i)-2\lambda_{n - x_n}\pi_{n - x_n}(i)f_n(i)}{\nu_{n - x_n}(i , i+1)}.\ ] ] by proposition [ p - main1 ] , @xmath891 and @xmath892 .
let @xmath893 be the function on @xmath775 defined by @xmath894 for @xmath895 and @xmath896 for @xmath897 .
a direct computation shows that @xmath898 and @xmath899 this implies @xmath900 . on the other hand , using the test function , @xmath901 for @xmath897 and @xmath902 for @xmath903 , one has @xmath904\le \epsilon_n / x_n$ ] .
this finishes the proof .
the next theorem has a detailed description on the coefficient of the spectral gap .
the proof is based on section 3 , particularly proposition [ p - shape2 ] and remark [ r - shape2 ] , and is given in the appendix .
[ t - bottleneck2 ] for @xmath248 , let @xmath905 be as in theorem [ t - bottleneck ] .
suppose @xmath906 $ ] and @xmath907 $ ] . * if @xmath908 and @xmath909 , then @xmath910 .
* if @xmath908 and @xmath911 $ ] , then @xmath912 , where @xmath531 is the unique positive solution of the following equation . @xmath913}=0.\ ] ] * if @xmath914 , then @xmath915 . in this subsection
, we consider paths with multiple bottlenecks .
as before , @xmath697 with @xmath916 and @xmath917 .
let @xmath395 be a positive integer and @xmath918 be a @xmath395-vector satisfying @xmath919 and @xmath920 and @xmath921 for @xmath922 .
let @xmath923 be a vector with positive entries and @xmath924 be the measure on @xmath925 given by @xmath926 [ t - kbottleneck ] let @xmath697 be the path on @xmath865 . for @xmath927 , let @xmath928 be the uniform probability on @xmath775 and @xmath924 be the measure on @xmath925 given by ( [ eq - kbottleneck ] ) .
then , @xmath929 where @xmath930 and @xmath931 [ r - kbottleneck ] observe that , in theorem [ t - kbottleneck ] , @xmath932 and @xmath933 we first prove the upper bound .
let @xmath934 be a function on @xmath36 satisfying @xmath935 for @xmath936 and @xmath937 be a function on @xmath938 obtained by identifying points @xmath939 and @xmath940 for @xmath936 .
by setting @xmath937 as a minimizer for @xmath941 with @xmath942 , we obtain @xmath943 to see the other upper bound , let @xmath944 be the function on @xmath775 satisfying @xmath945 for @xmath946 and @xmath947 for @xmath948 .
computations show that @xmath949 , @xmath950 for @xmath951 , and @xmath952 .
set @xmath953 .
as a consequence of the above discussion , we obtain @xmath954 taking @xmath955 for @xmath956 and @xmath957 otherwise gives the bound @xmath958 . the lower bound is immediate from theorem [ t - lower ] and remark [ r - lower ] .
finally , we discuss some special cases illustrating theorem [ t - kbottleneck ] .
( 160,80)(-10,-20 ) ( 20,0)(275,0)2(-115,10)(15,0)4 ( 20,0)(200,0)2(-60,10)(10,0)2 ( 10,0)(110,0)2(-30,10)(15,0)5 ( 50,10)(10,0)4 ( 10,0)(-105,10)(3,0)5(1,0)2 ( -90,10)(1,0)15 ( -75,10)(3,0)5(1,0)2 ( -20,10)(3,0)5(1,0)2 ( -5,10)(3,0)5(1,0)2 ( 10,10)(1,0)15 ( 25,10)(3,0)5(1,0)2 ( 90,10)(1,0)15 ( 105,10)(3,0)5(1,0)2 ( 120,10)(1,0)15 ( 135,10)(3,0)5(1,0)2 ( -10,0)(190,10)(1,0)15 ( 205,10)(3,0)5(1,0)2 ( 220,10)(1,0)15 ( 20,0)(245,0)2(-90,0)(-1,0)25(-65,0)(1,0)25(-85,-4)@xmath959 ( 20,0)(185,0)2(-65,-10)(-1,0)50(-30,-10)(1,0)50(-58,-14)@xmath960 ( 25,40)(-1,-2)10 ( 105,40)(1,-2)10 ( 28,45)most weak edges [ t - unifp ] for @xmath248 , let @xmath876 and @xmath961 be the measure in ( [ eq - kbottleneck ] ) with @xmath962 bottlenecks satisfying @xmath963 .
suppose there are @xmath964 , @xmath965 and @xmath966 such that @xmath967 is bounded and , for @xmath968 , @xmath969 .
then , @xmath970 it is easy to get the lower bound from theorem [ t - kbottleneck ] , while the upper bound is the minimum of @xmath958 over all connected components of @xmath971 and @xmath972 .
see figure [ fig2 ] for a reference on the bottlenecks .
the following are immediate corollaries of theorems [ t - kbottleneck]-[t - unifp ] .
[ c - finitebtnk ] referring to _ theorem [ t - unifp ] _ ,
if @xmath962 is bounded , then @xmath973 [ c - farawaybtnk ] referring to _ theorem [ t - unifp ] _
, if @xmath963 and there are @xmath965 and @xmath966 such that @xmath974 for @xmath975 , then @xmath976 [ c - unifgap ] referring to _ theorem [ t - unifp ] _ , if @xmath977 and @xmath978 with @xmath979 , then @xmath980 note that the assumption of the uniformity of @xmath9 and @xmath8 , except at the bottlenecks , can be relaxed by using a comparison argument .
we start with an elementary lemma .
[ l - conv ] let @xmath779 and @xmath981 be a continuous function satisfying @xmath982 and @xmath983 for @xmath984 . for @xmath985 , set @xmath986 .
then , @xmath987 and @xmath988 for @xmath989 . moreover , if @xmath249 is bounded on @xmath990 , then @xmath991 for @xmath989 and @xmath992 with @xmath993 .
[ l - matt ] let @xmath994 be sequences of reals with @xmath995 and @xmath996 . for @xmath248 and @xmath997
, let @xmath998 then , there are @xmath0 distinct real roots for @xmath999 , say @xmath1000 , and @xmath1001 furthermore , if @xmath1002 and @xmath1003 , then @xmath1004 for all @xmath248 .
[ l - mat ] fix @xmath1005 and , for @xmath1006 , let @xmath1007 be reals with @xmath995 and @xmath1008 . consider the following matrix @xmath1009 then , the eigenvalues of @xmath765 are distinct reals and independent of @xmath1010 . furthermore , if @xmath1002 and @xmath1003 , then all eigenvalues of @xmath765 are positive .
let @xmath1011 be diagonal matrices with @xmath1012 , @xmath1013 and @xmath1014 for @xmath1015 .
one can show that @xmath1016 since @xmath1017 is independent of the choice of @xmath1010 , the eigenvalues of @xmath765 are independent of @xmath1010 .
note that @xmath1018 is hermitian .
this implies that the eigenvalues of @xmath765 are all real .
as @xmath765 is tridiagonal with non - zero entries in the superdiagonal , the rank of @xmath1019 is either @xmath46 or @xmath0 .
this implies that the eigenvalues of @xmath765 are all distinct .
next , assume that @xmath1002 and @xmath1003 .
let @xmath1020 be the leading @xmath1021 principal matrices of @xmath1018 .
by induction , one can prove that @xmath1022 , where @xmath1023 and @xmath1024 for @xmath1025 . by the assumption at the beginning of this paragraph , @xmath1026 for all @xmath1025 and @xmath1027 for all @xmath44 .
as the leading principal matrices have positive determinants , @xmath1028 is positive definite .
this proves that all eigenvalues of @xmath765 are positive .
we prove this lemma by induction . for @xmath1029 ,
it is clear that @xmath1030 is the root for @xmath1031 . for @xmath55 , note
that @xmath1032 is a quadratic function that tends to infinity as @xmath1033 . since @xmath1034 , the polynomial , @xmath1032 , has two real roots , say @xmath1035 , satisfying @xmath1036
now , we assume that , for some @xmath248 , @xmath1037 and @xmath1038 have reals roots @xmath1039 and @xmath1040 satisfying @xmath1041 for @xmath44 .
clearly , @xmath1042 as @xmath1043 .
this implies @xmath1044 observe that @xmath1045 . replacing @xmath1046 with @xmath1047 yields @xmath1048
this proves that @xmath1049 has @xmath1050 distinct real roots with the desired interlacing property . for the second part ,
assume that @xmath1002 and @xmath1003 for all @xmath336 . for @xmath1029 , it is obvious that @xmath1051 .
suppose @xmath1004 .
according to the first part , we have @xmath1052 . by lemma [ l - mat ] , @xmath1053 , which implies @xmath1054 .
as it is known that @xmath1055 for @xmath1056 , it must be the case @xmath1057 .
otherwise , there will be another root for @xmath1038 between @xmath1058 and @xmath47 , which is a contradiction . for convenience , we set @xmath1059 for @xmath1060 and let @xmath295 be the @xmath118-by-@xmath118 tridiagonal matrix with entries @xmath1061 for @xmath1062 and @xmath1063 .
for @xmath252 , let @xmath1064 be the matrix equal to @xmath1065 except the @xmath1066-entry , which is defined by @xmath1067 .
by remark [ r - shape ] , @xmath1068 is the smallest root of @xmath1069 and @xmath1070 are roots of @xmath1071 .
note that , for @xmath1072 , @xmath1073 where @xmath1074 , @xmath1075 and @xmath1076 to prove this theorem , one has to determine the sign of @xmath1077 .
let @xmath1078 with @xmath1079 . as @xmath746 , @xmath1080 where @xmath1081 is uniform for @xmath44 .
note that @xmath1082 .
this implies @xmath1083 by a similar reasoning , one can prove that @xmath1084 for bounded @xmath245 .
this shows that , for @xmath1085 and @xmath1086 , @xmath1087 next , we compute @xmath1088 with @xmath1089 and @xmath1090 . note that , for @xmath0 large enough , @xmath1091 calculus shows that @xmath1092 and @xmath1093 observe that , as @xmath746 , @xmath1094 putting this back into ( [ eq - rj ] ) implies @xmath1095 we consider the following two cases . * case 1 : @xmath1096 .
* in this case , theorem [ t - bottleneck ] implies that @xmath1097 .
we assume further that @xmath1098 and @xmath1099 with @xmath1100 and @xmath1101 $ ] .
let @xmath1089 . replacing @xmath1102 with @xmath1103 in ( [ eq - rjl ] ) and with @xmath1104 in ( [ eq - rj2 ] ) yields that , for @xmath909 , @xmath1105 and , for @xmath911 $ ] , @xmath1106 where @xmath1107}$ ] .
this proves ( 1 ) and ( 2 ) .
laurent miclo .
monotonicity of the extremal functions for one - dimensional inequalities of logarithmic sobolev type . in _
sminaire de probabilits xlii _ , volume 1979 of _ lecture notes in math .
_ , pages 103130 .
springer , berlin , 2009 .
l. saloff - coste .
simple examples of the use of nash inequalities for finite markov chains . in _ stochastic geometry ( toulouse , 1996 ) _ , volume 80 of _ monogr .
_ , pages 365400 .
chapman & hall / crc , boca raton , fl , 1999 . | we consider the spectrum of birth and death chains on a @xmath0-path .
an iterative scheme is proposed to compute any eigenvalue with exponential convergence rate independent of @xmath0 .
this allows one to determine the whole spectrum in order @xmath1 elementary operations . using the same idea
, we also provide a lower bound on the spectral gap , which is of the correct order on some classes of examples . partially supported by nsc grant nsc100 - 2115-m-009 - 003-my2 and ncts , taiwan ] partially supported by nsf grant dms-1004771 ] |
last years electron transport in molecular transistors became a hot topic of experimental and theoretical investigations in nanoelectronics ( see e.g. @xcite ) . from experimental point of view
it is a real challenge to place a single molecule in a gap between electric leads and to repeatedly measure electric current as a function of bias and gate voltages .
being in a gap the molecule may form chemical bonds with one of metallic electrodes and then a considerable charge transfer from the electrode to the molecule takes place . in this case
one can consider the trapped molecule as a part of metallic electrode and the corresponding device does not function as a single electron transistor ( set ) .
much more interesting situation is the case when the trapped molecule is more or less isolated from the leads and preserves its electronic structure . in a stable state at zero gate voltage
the molecule is electrically neutral and the chemical potential of the leads lies inside the gap between homo ( highest occupied molecular orbital ) and lumo ( lowest unoccupied molecular orbital ) states .
this structure demonstrates coulomb blockade phenomenon @xcite and coulomb blockade oscillations of conductance as a function of gate voltage ( see review papers in @xcite and references therein ) .
in other words a molecule trapped in a potential well between the leads behaves as a quantum dot and the corresponding device exhibits the properties of set .
the new features in a charge transport through molecular transistors as compared to the well - studied semiconducting set appear due to `` movable '' character of the molecule trapped in potential well ( the middle electrode of the molecular transistor ) .
two qualitatively new effects were predicted for molecular transistors : ( i ) vibron - assisted electron tunnelling ( see e.g. @xcite ) and , ( ii ) electron shuttling @xcite ( see also review @xcite ) .
vibron(phonon)-assisted electron tunnelling is induced by the interaction of charge density on the dot with local phonon modes ( vibrons ) which describe low - energy excitations of the molecule in a potential well .
this interaction leads to satellite peaks ( side bands ) and unusual temperature dependence of peak conductance in resonant electron tunnelling @xcite . for strong electron - vibron interaction the exponential narrowing of level width and as a result strong suppression of electron transport ( polaronic blockade )
was predicted @xcite .
the effect of electron shuttling appears at finite bias voltages when additionally to electron - vibron interaction one takes into account coordinate dependence of electron tunnelling amplitude @xcite .
recent years carbon nanotubes are considered as the most promising candidates for basic element of future nanoelectronics .
both @xmath0-based and carbon nanotube - based molecular transistors were already realized in experiment @xcite .
the low - energy features of i - v characteristics measured in experiment with @xmath0-based molecular transistor @xcite can be theoretically explained by the effects of vibron - assisted tunnelling @xcite .
it is well known that in single - wall carbon nanotubes ( swnt ) electron - electron interaction is strong and the electron transport in swnt quantum wires is described by luttinger liquid theory .
resonant electron tunnelling through a quantum dot weakly coupled to luttinger liquid leads for the first time was studied in ref.@xcite were a new temperature scaling of maximum conductance was predicted : @xmath1 with interaction dependent exponent ( g is the luttinger liquid correlation parameter ) . in this paper
we generalize the results of refs.@xcite to the case when a quantum dot with vibrational degrees of freedom is coupled to luttinger liquid quantum wires .
the experimental realization of our model system could be , for instance , @xmath0-based molecular transistors with swnt quantum wires . in our model
electron - electron and electron - phonon interactions can be of arbitrary strength while electron tunnelling amplitudes are assumed to be small ( that is the vibrating quantum dot is weakly coupled to quantum wires ) .
we will use master equation approach to evaluate the average current and noise power . for noninteracting electrons
this approximation is valid for temperatures @xmath2 , where @xmath3 is the bare level width . for interacting electrons the validity of this approach ( perturbation theory on @xmath3 ) for high - t regime of electron transport
was proved for @xmath4 ( strong interaction ) @xcite and when @xmath5 ( weak interaction ) @xcite .
we found that at low temperatures : @xmath6 ( @xmath7 is the characteristic energy of vibrons ) the peak conductance scales with temperature accordingly to furusaki prediction @xcite : @xmath8 ( @xmath9 is the luttinger liquid cutoff energy ) .
the influence of electron - phonon interaction in low - t region results in renormalization of bare level width : @xmath10 , where @xmath11 is the dimensionless constant of electron - phonon interaction . in the intermediate temperature region : @xmath12 , ( @xmath13 ) , furusaki scaling is changed to @xmath14 and at high temperatures when all inelastic channels for electron tunnelling are open we again recovered furusaki scaling with nonrenormalized level width ( @xmath3 ) .
for nonlinear regime of electron tunnelling we showed that zero - bias peak in differential conductance , presenting elastic tunnelling , is suppressed by coulomb correlations in the leads .
this is manifestation of the kane - fisher effect @xcite . when interaction is moderately strong ( @xmath15 ) the dependence of differential conductance on bias voltage is non - monotonous due to the presence of satellite peaks .
for @xmath16 the zero - bias peak can be even more suppressed than the satellite peaks , which dominate in this case .
this is the manifestation of the interplay between the luttinger liquid effects in the leads and the electron - phonon coupling in the dot . for strong interaction @xmath4 satellites
are also suppressed and the differential conductance at low temperatures ( @xmath17 ) scales as @xmath18 .
this scaling coincides with the furusaki prediction , where temperature is replaced by the driving voltage ( @xmath19 ) which becomes the relevant energy scale for @xmath20 .
it means that the influence of vibrons on the resonant electron tunnelling through a vibrating quantum dot can be observed only for weak or medium strong interaction ( @xmath21 ) in the leads .
the hamiltonian of our system ( vibrating quantum dot weakly coupled to luttinger liquid leads , ( see fig.1 ) consists of three parts @xmath22 here @xmath23 describes quantum wires adiabatically connected to electron reservoirs .
quantum wires ( left - l and right - r ) are supposed equal and modelled by luttinger liquid hamiltonians with equal luttinger liquid parameters @xmath24 : @xmath25 ( see e.g.@xcite ) @xmath26 here @xmath27(@xmath28 ) are the creation ( annihilation ) operators of bosons which describe the charge density fluctuations propagating in the leads with velocity @xmath29 .
these operators satisfy canonical bosonic commutation relations @xmath30=\delta ( k - k')$ ] . in what follows
we consider for simplicity the case of spinless electrons .
the hamiltonian of vibrating single level quantum dot takes the form ( see e.g.@xcite ) @xmath31 where @xmath32 is the energy of electron level on the dot , @xmath7 is the energy of vibrons , @xmath33 is the electron - vibron interaction energy , @xmath34(@xmath35 ) and @xmath36(@xmath37 ) are fermionic ( @xmath35 ) and bosonic ( @xmath37 ) creation ( annihilation ) operators with canonical commutation relations @xmath38 , @xmath39=1 $ ] .
the tunnelling hamiltonian is given by standard expression @xmath40 where @xmath41 is the electron tunnelling amplitude and @xmath42 , @xmath43 is the annihilation operator of electron at the end point of l(r)-electrode .
this operator could be written in a `` bosonised '' form ( according to @xcite ) @xmath44 \ , \label{5}\ ] ] here @xmath45 is a short - distance cutoff of the order of the reciprocal of the fermi wave number @xmath46 and @xmath47 is the interaction parameter in the `` fermionic '' form of the luttinger liquid hamiltonian ( 2 ) , it defines the luttinger liquid parameter @xmath48 which is varied between @xmath49 and @xmath50 : the case @xmath51 describes the `` noninteracting '' ( fermi - liquid ) leads , than in the case @xmath52 the interaction in the leads goes to infinity . ) to the luttinger liquid leads ( @xmath53 ) with the chemical potentials @xmath54 ( @xmath55 is the driving voltage ) .
all the energies are counted from the fermi energy , which chosen to be zero .
electrons tunnel from one lead to another by hopping on and off the dot level with the energy @xmath56 ( elastic channel ) and due to electron - vibron coupling they can emit or absorb vibrons ( vibron - assisted tunnelling ) .
inelastic channels are represented as side - levels with energies @xmath57 .
the position of the dot levels with respect to the fermi energy can be uniformly shifted by applying voltage @xmath58 to the `` gate '' electrode.,width=8,height=8 ] hamiltonian ( 3 ) is `` diagonalized '' to @xmath59 by the unitary transformation ( see e.g. @xcite ) @xmath60 , where @xmath61 , @xmath62 and the dimensionless parameter @xmath63 characterizes electron - vibron coupling .
the unitary transformation results in : ( i ) the shift of fermionic level ( polaronic shift ) @xmath64 and , ( ii ) the replacement of tunnelling amplitude in ( 3 ) @xmath65 .
the model eqs.([1])-([5])can not be solved exactly and one needs to exploit certain approximations to go further .
we will use `` master equation '' approximation ( see e.g. @xcite ) to evaluate the average current and noise power in our model .
it is in this approximation that average current separately for the model with interacting leads @xcite and for vibrating quantum dot with noninteracting leads @xcite was calculated earlier .
master ( rate ) equation approach exploits such quantities as the probability for electron to occupy dot level and the transition rates .
it neglects quantum interference in electron tunnelling and therefore describes only the regime of sequential electron tunnelling which is valid when the width of electron level @xmath66 .
in other words , in our case `` master equation '' approach is equivalent to the lowest order of perturbation theory in @xmath3 . for interacting electrons the validity of master equation approach for high - t regime of resonant electron tunnelling can be justified for strong repulsive interaction @xmath4 @xcite .
it is correct also for weak interaction @xmath5 as one can check by comparing the results of ref.@xcite and ref .
@xcite , where resonant tunnelling through a double - barrier luttinger liquid was considered for weak electron - electron interaction .
notice , that the results @xcite of exact solution known for @xmath67 , where a mapping to free - fermion theory can be used @xcite , do not agree with the high - t scaling of @xmath68 @xcite extrapolated to this special point @xmath67 .
the free - fermion scaling @xmath69 found for @xmath67 ( master equation approach predicts @xmath70-independent value @xcite ) could be a special feature of this exactly solvable case .
we will assume that beyond the close vicinity to @xmath67 the master equation approach for high - t behavior of conductance is a reasonable approximation .
in master equation approach the average current through a single level quantum dot expressed in terms of transition rates takes the form @xmath71 where @xmath72 is the rate of electron tunnelling from the dot to right ( left ) electrode , @xmath73 describes the reverse process and @xmath74 , @xmath75 @xmath76 . to evaluate these rates in our approach
we will use fermi `` golden rule '' ( quantum mechanical perturbation theory ) for tunnelling hamiltonian obtained from eq.(4 ) after the unitary transformation : @xmath77 @xmath78 the standard calculation procedure results in the following expressions for tunnelling rates @xmath79 @xmath80 where @xmath81 is the bias voltage and @xmath43 .
notice that in the perturbation calculation on the bare level width @xmath82 , we neglect the level width in the green function of the dot level .
besides , in this approximation averages over bosonic and fermionic operators in formulas for tunnelling rates are factorized and , thus , the averages @xmath83 over bosonic variables @xmath84 \label{10}\ ] ] can be calculated with the quadratic hamiltonian @xmath85 . in what follows
we will assume that vibrons are characterized by equilibrium distribution function : @xmath86^{-1}$ ] .
averages @xmath87 over fermionic operators in eqs.(8),(9 ) are calculated with the luttinger liquid hamiltonian ( 2 ) . the corresponding correlation functions in eqs.(8),(9 )
are well known in the literature ( see e.g.@xcite ) @xmath88\exp[-ilw_{0}(t+i\hbar/2t)]\right\ } \nonumber\\ % \label{11}\end{aligned}\ ] ] @xmath89\right\}^{-1/g } \ . \label{12}\ ] ] here @xmath90 is a modified bessel function , @xmath91 is a ultraviolet cutoff energy , @xmath48 is the luttinger liquid correlation parameter . by putting correlation functions ( 11),(12 ) in eqs.(8),(9 ) and evaluating time integrals we get the following equations for tunnelling rates @xmath92 and @xmath93 , ( @xmath43 ) @xmath94}{\gamma(1/g ) } \nonumber\\ \cdot\left(\frac{2\pi t}{\lambda}\right)^{1/g-1}\sum_{l=-\infty}^{\infty } i_{l}\left[\frac{\lambda^{2}}{\sinh\left(\hbar w_{0}/2t\right)}\right ] \nonumber\\ \cdot\left|\gamma\left(\frac{1}{2g}+i\frac{(\pm\delta_{j}-\hbar w_{0}l)}{2\pi t}\right)\right|^{2 } \nonumber\\ % \label{13}\end{aligned}\ ] ] where @xmath95 is the partial level width ( see , for example,@xcite ) @xmath96 ( @xmath43 ) , @xmath97 , @xmath98 ; here @xmath99 is gamma function . at
first we consider different limiting cases when it is possible to obtain simple analytical expressions for the average current ( 6 ) .
notice , that electric current depends on the gate voltage @xmath100 through the corresponding dependence of level energy @xmath101 .
it is convenient for the further analysis to choose the value of gate voltage at which the current at low bias is maximum as : @xmath102 .
in what follows we also put @xmath103 . for noninteracting leads @xmath104 and noninteracting quantum dot @xmath105
it is easy to derive from eqs.(6),(13 ) the well - known formula for the maximum ( resonant ) current at temperatures @xmath106 @xmath107 where @xmath108 is the effective level width .
it is evident that at high voltages : @xmath109 the current through a single level dot does not depend on the bias voltage and its value is totally determined by the effective level width @xmath110 .
for a vibrating quantum dot @xmath111 weakly coupled to noninteracting leads @xmath104 our approach reproduces the results of ref.@xcite . in the temperature region
we are interesting in @xmath112 the general formulae derived in @xcite can be rewritten in a more clear and compact form . in particular , by using for @xmath51 in eq.(11 ) the well - known representation for gamma function ( see e.g.@xcite ) @xmath113 it is easy to obtain the following expression for the maximum ( peak ) conductance @xmath114 where @xmath115 is the standard resonance conductance of a single - level quantum dot at @xmath106 .
the dimensionless function @xmath116 takes the form @xmath117\nonumber\\ \cdot\left\{i_{0}(z(x))+2\cdot \sum_{l=1}^{\infty } \frac{i_{l}(z(x))}{\cosh^{2}(lx/2)}\right\ } \ , \label{18}\end{aligned}\ ] ] here @xmath118}$ ] and @xmath119 . at low temperature region @xmath120 , when there are no thermally activated vibrons in the dot @xmath121 only the first term in the brackets contribute to the sum and : @xmath122
we see , that zero - point fluctuations of the dot position result in renormalization of the level width @xmath123 . for strong electron - vibron coupling this phenomenon
( polaronic narrowing of level width ) leads to polaronic ( franck - condon ) blockade of electron transport through vibrating quantum dot @xcite .
the temperature behavior of peak conductance ( 16 ) was considered in ref.@xcite .
now we will study the general case when interacting quantum dot @xmath111 is connected to interacting leads @xmath124 .
analytical expressions for conductance in this case can be obtained in the limits of low @xmath125 and high @xmath126 temperatures . at low temperatures
the main contribution to the sum over @xmath127 in eq.(13 ) comes from elastic transition @xmath128 .
all inelastic channels @xmath129 are exponentially suppressed for @xmath130 . at @xmath131
the peak conductance takes the form @xmath132
\left(\frac{\pi t}{\lambda}\right)^{1/g-1 } \ . \label{19}\ ] ] we see that at low temperatures conductance scales with temperature according to furusaki s prediction @xcite : @xmath133 . the influence of electron - vibron coupling results in multiplicative renormalization of bare level width @xmath134 .
at high temperatures : @xmath135 one can use the well known asymptotic expansion for bessel function @xmath136 , which can be used in summation eqs.([18 ] ) , ( 13 ) until @xmath137 .
besides , in this temperature region the summation in eq.(13 ) , can be replaced by integration and the corresponding integral can be taken exactly @xmath138 this allows us to derive the following expression for the temperature dependence of peak conductance in the intermediate temperature region @xmath139 , @xmath140 @xmath141 \left(\frac{\pi t}{\lambda}\right)^{1/g-1 } \ . \label{21}\ ] ] notice that in the considered temperature region the polaronic blockade is already partially lifted @xmath142 at @xmath143 and conductance scales with temperature as @xmath144 . at last , at temperatures
@xmath145 when all inelastic channels for electron transport are open , the polaronic blockade is totally lifted @xcite and we reproduce again furusaki scaling .
it is clear from our asymptotic formulae ( 19),(21 ) that both in low- and in high- temperature regions the contributions of electron - electron and electron - vibron interactions to the conductance are factorized . in general case
these contributions are not factorized , as one can see from eqs.(8),(9 ) and from eq.(13 ) for tunnelling rates , and we can expect interplay of kane - fisher effect and the effect of phonon(vibron)-assisted tunnelling . to see this interplay we consider nonlinear ( differential ) conductance @xmath146 .
it is well known that kane - fisher effect is pronounced for the energies close to the fermi energy . for differential
conductance it means that the zero - bias ( elastic ) resonance peak is suppressed with the increase of electron - electron interaction , while satellite peaks are less affected by the interaction .
when electron - electron interaction is weak or moderately strong @xmath147 the dependence of differential conductance on the bias voltage ( for @xmath148 ) is not a monotonous function due to the presence of satellite peaks ( see figs.2,3 ) . ) as a function of driving voltage ( in the units of @xmath7 ) for the case @xmath16 .
here we put @xmath149 ; @xmath150 ; @xmath151 ; and tune the level energy to the resonant position @xmath152 ( @xmath153 ) .
solid line corresponds to the case of noninteracting leads : @xmath51 .
dot line corresponds to value @xmath154 , while dash - dot line - to @xmath155 .
figure shows how zero - bias ( elastic ) resonance peak is gradually suppressed with the increase of electron - electron correlations ( decrease of luttinger liquid parameter @xmath48 ) while the satellite peaks survive until @xmath16.,width=8,height=6 ] ) as a function of driving voltage ( in the units of @xmath7 ) for the case @xmath4 . here
@xmath149 ; @xmath156 ; @xmath157 ; and @xmath32 are the same as on fig.2 .
solid line corresponds to the case of noninteracting leads : @xmath51 ; dot line corresponds to value @xmath158 ; dash - dot line - to @xmath159 .
one can see from the figure that for @xmath4 the resonance - like behavior of differential conductance disappears and conductance scales as a power - law of the bias voltage.,width=8,height=6 ] the resonant behavior disappears for strong interaction @xmath4 ( figs.2,3 ) , when at low temperatures @xmath160 differential conductance scales with bias voltage as @xmath161 in accordance with the luttinger liquid prediction for nonlinear electron transport through a single - level quantum dot .
for instance , if we put @xmath162 and tune the level energy to the resonance point @xmath152 -(``resonant '' location of the level in the presence of `` polaronic '' shift ) , we obtain for the differential conductance @xmath163 the following expression for @xmath164 @xmath165\cdot \left(\frac{ev}{2\lambda}\right)^{n-2 } \ , \label{22}\ ] ] where @xmath134 .
one can readily see that expression ( 22 ) reproduces furusaki temperature scaling eq.(19 ) when @xmath19 is replaced by @xmath70 .
analogous interplay of kane - fisher and polaronic effects one can see on figs.4,5 , where differential conductance is plotted as a function of level energy @xmath166 ( or , equivalently , as a function of gate voltage ) . for noninteracting leads
@xmath104 the resonance conductance peaks correspond to the level positions at @xmath167 ( in our plot we put : @xmath168 ) .
this elastic resonance peak is suppressed by electron - electron interaction in the leads @xmath124 .
the dependence @xmath169 for weak and moderately strong interaction still reveals resonance structure with 4 satellites in our case ( see fig.4 ) .
the inelastic resonance peaks disappear at @xmath4 and maximum of differential conductance corresponds at @xmath170 to the level position at @xmath102 , that is exactly in the middle between chemical potentials of left and right electrodes ( see fig.5 ) . ) as a function of level energy @xmath166 , counted from the fermi energy for the case @xmath16 .
solid line corresponds to the case of noninteracting leads : @xmath51 ; dot line corresponds to value @xmath154 ; dash - dot line - to @xmath155 .
the bias voltage @xmath171 is sufficiently high to excite vibrons and to support electron transport through inelastic channels .
all parameters are the same as for fig.2.,width=8,height=6 ] ) as a function of level energy @xmath166 , counted from the fermi energy for the case @xmath4 .
solid line corresponds to the case of noninteracting leads : @xmath51 ; dot line corresponds to value @xmath158 ; dash - dot line - to @xmath159 .
the bias voltage @xmath171 is sufficiently high to excite vibrons and to support electron transport through inelastic channels .
all parameters are the same as for fig.3.,width=8,height=6 ] it is important to stress here once more that for moderately strong electron - electron interaction in the leads the inelastic tunnelling can dominate in electron transport .
one can see from figs.2 - 5 that there is region of coupling constants when the first satellite peak is higher than the `` main '' ( zero - bias ) resonant peak , which corresponds to elastic ( @xmath128 ) tunnelling channel .
it is the most significant prediction , we have made in this paper .
the knowledge of tunnelling rates eqs.(8,9 ) allows us to evaluate not only the average current eq.(6 ) but the noise power as well .
we will follow the method developed in refs .
@xcite where quantum noise was calculated for resonant electron transport through a quantum dot weakly coupled to noninteracting electrodes .
the noise power is defined ( see e.g. @xcite ) as @xmath172 where @xmath173 ( @xmath174 is the average current ) .
the noise defined in eq.(23 ) , in the case of sequential tunnelling through a quantum dot , can be expressed in terms of tunnelling rates . for a single level quantum dot
this formula for low frequency noise @xmath175 takes the form @xmath176 here the average current @xmath174 is determined by eq.(6 ) .
the noise power eq.(24 ) depends on temperature and bias voltage @xmath177 and contains both thermal ( jonson - nyquist ) noise @xmath178 ( @xmath179 is the conductance ) and the non - equilibrium ( shot ) noise @xmath180 .
since the thermal noise is totally determined by temperature dependence of conductance , we will study in what follows only shot noise and fano factor @xmath181 .
in particular , fano factor in our case can be represented as follows @xmath182 for noninteracting leads @xmath104 and noninteracting quantum dot @xmath183 one readily gets from eqs.(13),(24 ) a simple expression for the `` full '' noise @xmath184 of a single electron transistor ( set ) . on resonance @xmath102 and at temperatures @xmath185
one finds @xmath186\nonumber\\ + \frac{e^{2}\gamma}{\hbar}\left[\frac{\exp(-ev/4t)}{\cosh^{2}(ev/4 t ) } \right]\ . \label{26}\end{aligned}\ ] ] from eq.(26 ) in the limit @xmath187 we obtain @xmath188 , where @xmath189 is the thermal noise . in the opposite case
@xmath190 we rederive the well - known formulae for the shot - noise and the fano factor of a single level quantum dot @xcite @xmath191 these formulae ( 26),(27 ) can be also re - derived from the general expression for the full noise of noninteracting electrons ( see e.g. , eq.(61 ) in ref.@xcite ) @xmath192\nonumber\\ + \frac{e^{2}}{\hbar}\int d\varepsilon t_{t}(\varepsilon)\left[1-t_{t}(\varepsilon)\right](f_{l}-f_{r})^{2}\ , \label{28}\end{aligned}\ ] ] where @xmath193 is the transmission coefficient and @xmath194 + 1\right\}^{-1}$ ] is the equilibrium distribution function of electrons in the leads ( @xmath195 is the chemical potential ; @xmath43 ) . in the case of single level quantum dot
@xmath193 takes the form breit - wigner tunnelling probability @xmath196 where @xmath197 . for a weak tunnelling when @xmath95 are the smallest energy scales in the problem the lorentzian shape of the breit - wigner resonance shrinks to @xmath198-function @xmath199 with the help of eqs.(28),(30 ) for the resonance condition @xmath102 we easily re - derive eqs.(26 ) .
( notice , that in sequential tunnelling approach the tunnelling transitions through the left and right barriers are assumed to be weak and uncorrelated .
therefore we can safely neglect @xmath200-term in eq.(28 ) ) .
it is evident from eqs.(25),(27 ) that for noninteracting electrons the fano factor is sub - poissonian @xmath201 and @xmath202 approaches @xmath50 for strongly asymmetric junction @xmath203 and for @xmath190 .
the master equation approach we have used in our analysis holds when electron tunnelling amplitudes are small . for noninteracting electrons
this assumption is satisfied when electron energies are far from the resonant energy level , i.e. @xmath204 .
the differential shot noise in this case as a function of bias voltage or gate voltage behaves similarly to the differential conductance .
notice however that due to different dependence on temperature the shot noise unlike the thermal one even in sequential tunnelling regime ( @xmath205 ) can not be expressed in terms of conductance . ) as function of the level energy @xmath166 in the nonlinear transport regime ( when @xmath171 ) for the case @xmath4 .
solid line corresponds to the case of noninteracting leads @xmath51 ; dot line corresponds to value @xmath158 ; dash - dot line - to @xmath159 .
other parameters are the same as on fig.5.,width=8,height=6 ] as the function of the level energy @xmath166 in the nonlinear transport regime ( when @xmath171 ) which corresponds to figs.5,6 .
solid line corresponds to the case of noninteracting leads @xmath51 ; dot line corresponds to value @xmath158 ; dash - dot line - to @xmath159 .
all other parameters are the same as on figs.5,6.,width=8,height=6 ] by comparing fig.6 with fig.5 , one can see that the above similarity is preserved for interacting electrons ( @xmath206 ) as well . the corresponding fano factor which is the `` shot noise / current '' ratio and thus is less sensitive to the details of tunnelling process , for strong electron - electron interaction exhibits a simple behavior ( see fig.7 ) .
it dips ( @xmath207 ) at symmetric ( with respect to chemical potentials of the leads ) position of the dot level .
outside this region @xmath208 ( poissonian noise ) .
the width of the dip decreases with the increase of interaction ( see fig.7 ) .
we considered the influence of interaction on transport properties of molecular transistor which was modelled as a vibrating single - level quantum dot weakly coupled to the luttinger liquid leads .
we found interesting interplay between polaronic and luttinger liquid effects in our system .
in particular it was shown that for weak or moderately strong interaction ( @xmath209 ) the differential conductance demonstrates resonance - like behavior and for moderately strong interaction inelastic channels can even dominate in electron transport through a vibrating quantum dot . for strong interaction ( @xmath170 )
the resonant character of vibron - assisted tunnelling disappears and the differential conductance scales as a power law on temperature ( linear regime @xmath210 ) or on bias voltage ( nonlinear regime @xmath190 ) .
the authors would like to thank s.i.kulinich for valuable discussions .
this work was partly supported by the joint grant of the ministries of education and science in ukraine and israel and by the grant `` effects of electronic , magnetic and elastic properties in strongly inhomogeneous nanostructures '' of the national academy of sciences of ukraine .
a.nitzan and m.a.ratner , _ science _ * 300 * , 1384 ( 2003 ) .
m.galperin , m.a.ratner , a.nitzan , _ j.phys.:cond.matt .
_ * 19 * , 103201 ( 2007 ) .
r.i.shekhter , _ zh.eksp.teor.fiz . _ * 68 * , 623 ( 1975 ) . i.o.kulik and r.i.shekhter , _ zh.eksp.teor.fiz . _ * 63 * , 1400 ( 1972 ) . , edited by h.grabert and m.h.devoret , nato asi ser.b , vol.294 , plenum press , n.y . , ( 1992 ) .
l.i.glazman , r.i.shekhter,_zh.eksp.teor.fiz._ * 94 * , 292 ( 1988 ) [ _ sov.phys.jetp _ * 67 * , 163 ( 1988 ) ] .
s.braig , k.flensberg , _
phys.rev.b _ * 68 * , 205324 ( 2003 ) .
l.y.gorelik , a.isacsson , m.v.voinova , b.kasemo , r.i.shekhter , m.jonson , _ phys.rev.lett . _ * 80 * , 4526 ( 1998 ) .
r.i.shekhter , l.y.gorelik , m.jonson , y.m.galperin , v.m.vinokur , _ j.comput.theor.nanosci . _ * 4 * , 860 ( 2007 ) .
u.lundin , r.m.mckenzie , _
phys.rev.b _ * 66 * , 075303 ( 2002 ) .
j.koch , f. von oppen , a.v.andreev , _
phys.rev.b_ , * 74 * , 205438 ( 2006 ) .
h.park , j.park , a.k.l.lim , e.h.anderson , a.p.alivisatos , p.l.mceuen , _ nature _ * 407 * , 57 ( 2000 ) . h.w.ch.postma,t.teepen , z.yao , m.grifoni , c.dekker , _ science _ * 239 * , 76 ( 2001 ) .
a.furusaki , _ phys.rev.b _ * 57 * , 7141 ( 1998 ) .
c.l.kane , m.p.a.fisher , _ phys.rev.b _ * 46 * , 15233 ( 1992 ) . yu.v.nazarov and l.i.glazman , _ phys.rev.lett . _ * 91 * , 126804 ( 2003 ) .
g.d.mahan , _ many - particle physics _ , plenum press , new york ( 1990 ) .
a.komnik and a.o.gogolin , _ phys.rev.lett .
_ * 90 * , 246403 ( 2003 ) . "
i.s.gradshtein , and i.m.ryzhik , _ tables of integrals , series and products _ , academic press , ny ( 1965 ) .
i.v.krive , r.ferone , r.i.shekhter , m.jonson , p.utko , j.nygard , _ new j.phys._ , ( 2008 ) to be published .
s.hershfield , j.h.davies , p.hyldgaard , c.j.stanton , j.w.wilkins , _ phys.rev.b _ * 47 * , 1967 ( 1993 ) .
i.djuric , b.dong , h.l.cui , _ j.appl.phys . _ * 99 * , 063710 ( 2006 ) .
y.m.blanter , m.buttiker , _ shot noise in mesoscopic conductors _ , _ phys.rep . _ * 336 * , pp.1 - 166 , ( 2000 ) . | electron transport through a single - level quantum dot weakly coupled to luttinger liquid leads is considered in the master equation approach .
it is shown that for a weak or moderately strong interaction the differential conductance demonstrates resonant - like behavior as a function of bias and gate voltages .
the inelastic channels associated with vibron - assisted electron tunnelling can even dominate electron transport for a certain region of interaction strength . in the limit of strong interaction resonant behavior
disappears and the differential conductance scales as a power low on temperature ( linear regime ) or on bias voltage ( nonlinear regime ) . |
analysis of data collected by the fermi - lat experiment has revealed two large circular structures near the galactic centre , above and below the galactic plane the so - called fermi bubbles @xcite .
the approximate edges of the fermi bubble regions are shown in figure [ fig : fb_shape ] .
these structures are characterised by gamma - ray emission with a hard @xmath0 spectrum and a constant intensity over the whole emission region .
signals from roughly the fermi bubble regions were also observed in the microwave band by wmap @xcite and , recently , in the radio - wave band @xcite .
moreover , the edges correlate with the x - ray emission measured by rosat @xcite .
several proposed models explaining the emission include hadronic mechanisms , in which gamma rays together with neutrinos are produced by the collisions of cosmic - ray protons with interstellar matter @xcite . others which include leptonic mechanisms or dark matter decay would produce lower neutrino emission or none at all @xcite .
the observation of a neutrino signal from the fermi bubble regions would play a unique role in discriminating between models .
the properties of the hypothesised neutrino emission are described in section [ s : fermi ] .
an overview of the antares neutrino detector is given in section [ s : antares ] and the neutrino event reconstruction is described in section [ s : reconstruction ] .
the search for neutrino emission is performed by comparing the number of events in the fermi bubble regions to the number found in similar off - zone regions ( section [ s : offzones ] ) .
the event selection optimisation is based on a simulation of the expected signal as described in section [ s : simulation ] .
the selected events are presented in section [ s : results ] together with the significance and the upper limit on the neutrino flux from the fermi bubbles .
the estimated photon flux in the energy range 1100 gev covered by the fermi - lat detector from the fermi bubble regions is @xcite : @xmath1 assuming a hadronic model in which the gamma - ray and neutrino fluxes arise from the decay of neutral and charged pions respectively , the @xmath2 and @xmath3 fluxes are proportional to the gamma - ray flux with proportionality coefficients of 0.211 and 0.195 respectively @xcite . with this assumption and using ( [ f :
gamma ] ) the expected neutrino flux is : @xmath4 @xmath5 the neutrino flux , as well as the gamma - ray flux , is expected to have an exponential energy cutoff , so the extrapolation of ( [ f : fb_flux ] ) towards higher energies can be represented by : @xmath6 the cutoff is determined by the primary protons which have a suggested cutoff @xmath7 in the range from 1 pev to 10 pev @xcite .
the corresponding neutrino - energy cutoff may be estimated by assuming that the energy transferred from @xmath8 to @xmath9 derives from the fraction of energy going into charged pions ( @xmath10 ) which is then distributed over four leptons in the pion decay .
thus : @xmath11 which gives a range from 50 tev to 500 tev for @xmath12 .
the antares telescope is a deep - sea cherenkov detector which is located 40 km from toulon ( france ) , at a latitude of @xmath13 n and a mooring depth of 2475 m. at a mooring depth of 2475 m. the energy and direction of incident neutrinos are measured by detecting the cherenkov light produced in water from muons originating in the charged - current interactions of @xmath14 and @xmath15 .
the light is detected with a three - dimensional array of twelve detection lines comprising 885 optical modules , each containing a 10 inch pmt .
more details on the detector construction , its positioning system and the time calibration can be found in @xcite .
the antares detector started data - taking with the first 5 lines installed in 2007 .
the construction of the detector was completed , with installation of the last two lines , in may 2008 .
the apparatus has been operating continuously ever since .
its main goal is the detection of neutrinos produced by the cosmic sources .
muons and neutrinos created in cosmic - ray induced atmospheric showers provide the two main background components for the search for cosmic neutrinos .
although the more than 2 km of water above the detector acts as a partial shield against the atmospheric muons , the downgoing atmospheric muon background at these depths is still bigger than the expected signal .
therefore , the search for cosmic signal concentrates on upgoing events which corresponds to neutrinos which have crossed the earth . also , the optical modules are oriented downwards at @xmath16 to favour the detection of upgoing particles .
the antares neutrino telescope has an excellent visibility by means of the upgoing neutrinos to the galactic centre region and to the fermi bubbles .
since atmospheric neutrinos may traverse the earth and lead to upgoing tracks in the detector , any signal from the fermi bubbles would be inferred by observing a significant statistical excess over the background .
the signal - to - noise ratio can be improved by rejecting low - energy neutrino events , as the spectrum of the atmospheric neutrinos is steeper than the expected source spectrum .
the track of a muon passing through the detector is reconstructed using the arrival time of the photons together with the positions and orientations of the photomultipliers .
details of the tracking algorithm are given in @xcite .
only events reconstructed as upgoing have been selected .
in addition , cuts on the reconstruction quality parameters have been applied in order to reject downgoing atmospheric muon events that are incorrectly reconstructed as upgoing tracks .
these parameters are the quality @xmath17 of the track fit , which is derived from the track fit likelihood , and the uncertainty @xmath18 of the reconstructed track direction .
the choice of the cut on @xmath17 fixes the amount of background from misreconstructed atmospheric muons in the neutrino sample .
neutrino simulations for an @xmath0 neutrino spectrum have yielded a median angular resolution on the neutrino direction of less than @xmath19 for events with @xmath20 and @xmath21 @xcite .
shower - like events are identified by using a second tracking algorithm with @xmath22-like fit , assuming the hypothesis of a relativistic muon ( @xmath23 ) and that of a shower - like event ( @xmath24 ) @xcite .
events with better point - like fit ( @xmath25 ) have been excluded from the analysis . in this analysis
the energy of the muons entered or born in the detector was estimated using artificial neural networks , which are produced using a machine learning algorithm which derives the dependence between a set of observables and the energy estimate in a semi - parametric way @xcite .
the parameters used include the number of detected photons , and the total deposited charge .
the median resolution for @xmath26 is about 0.3 for muons with an energy of 10 tev .
the reconstructed energy @xmath27 is used to reject the atmospheric neutrino background while @xmath17 is used mostly to reject atmospheric muons .
the choice of cuts on @xmath17 and @xmath27 in this work is discussed in section [ s : mrf ] .
a signal from the combined fermi bubble regions is searched for by comparing the number of selected events from the area of both bubbles ( on - zone ) to that of similar regions with no expected signal ( off - zones ) .
the simplified shape of each fermi bubble as used in this analysis is shown in figure [ fig : fb_shape ] .
off - zones are defined as fixed regions in equatorial coordinates which have identical size and shape as the on - zone but have no overlap with it . in local coordinates ,
such off - zones have the same , sidereal - day periodicity as the on - zone and span the same fraction of the sky , but with some fixed delay in time .
the size of the fermi bubbles allows at maximum three non - overlapping off - zones to be selected .
the on - zone and three off - zones are shown in figure [ fig : fb_offzones_eq ] together with the sky visibility .
the visibility of each point on the sky is the fraction of the sidereal day during which it is below the horizon at the antares site ( in order to produce upgoing events in the detector ) .
the average visibility of the fermi bubbles is 0.68 ( 0.57 for the northern bubble and 0.80 for the southern bubble ) and it is the same for the off - zones . slightly changing detector efficiency with time and gaps in the data acquisition
can produce differences in the number of background events between the on - zone and the three off - zones . in order to test for such an effect , firstly
, the number of events in the off - zones is extracted from the data for various cuts ( @xmath28 , @xmath29 ) and the difference in the event numbers between each pair of off - zones is calculated .
this difference is compared with the statistical uncertainty and no excess is seen beyond the expected statistical fluctuations .
secondly , the number of events in the on - zone together with the average number of events in the three off - zones is tested using the simulated atmospheric background and the difference is found to be within the expectation from the statistical uncertainty .
it can be concluded , therefore , that this effect is negligible . showing the fermi bubble regions ( on - zone ) shaded area in the centre .
the regions corresponding to the three off - zones are also depicted .
the colour fill represents the visibility of the sky at the antares site .
the maximum on the colour scale corresponds to a 24 h per day visibility.,scaledwidth=80.0% ]
the analysis adopts a blind strategy in which the cut optimisation is performed using simulated data for the signal and the background .
the main quantities used to discriminate between the cosmic neutrino candidate events and the background from misreconstructed atmospheric muons and from atmospheric neutrinos are the tracking quality parameter @xmath17 and the reconstructed muon energy @xmath27 . [ s : simulation ] the simulation chain for antares is described in @xcite . for the expected signal from the fermi bubbles , the @xmath2 and @xmath3 fluxes according to section [ s : fermi ]
are assumed , using four different cutoffs @xmath12 : no cutoff ( @xmath30 ) , 500 tev , 100 tev and 50 tev .
atmospheric neutrinos are simulated using the model from the bartol group @xcite which does not include the decay of charmed particles .
data in the period from may 2008 , when the detector started to operate in its complete configuration , until december 2011 are used .
the total livetime selected for this analysis amounts to 806 days .
figure [ fig : fb_lambda ] shows the distribution of data and simulated events as a function of the parameter @xmath17 for events arriving from the three off - zones . here the events with at least 10 detected photons and the angular error estimate @xmath31 are selected .
the requirement on the number of photons removes most of the low - energy background events .
the angular error condition is necessary in order to ensure a high angular resolution to avoid events originating from an off - zone region being associated with the signal region and vice versa . for the upgoing events arriving from the three off - zones : data ( black crosses ) , 68% confidence area given by the total background simulation ( grey area ) , @xmath32 ( blue filled circles ) , @xmath33 ( pink empty circles ) ; bin - ratio of the data to the total background simulation ( bottom).,scaledwidth=70.0% ] at @xmath34 the main background component changes from the misreconstructed atmospheric muons to the upgoing atmospheric neutrino events as seen in figure [ fig : fb_lambda ] .
the flux of atmospheric neutrinos in the simulation is 23% lower than observed in the data .
this is well within the systematic uncertainty on the atmospheric neutrino flux and the atmospheric flux from the simulations was scaled accordingly in the following analysis .
a comparison of the energy estimator for data and for atmospheric neutrino simulation is shown in figure [ fig : fb_eann ] for the same event selection but with a stricter cut @xmath35 to remove most of the misreconstructed atmospheric muons .
the reconstructed energy of all simulated events has been shifted , @xmath36 , in order to improve the agreement between data and simulations .
this is within the estimated uncertainty of the optical module efficiency and the water absorption length ( * ? ? ?
* figure 4.24 ) .
distribution of the events arriving from the three off - zones with @xmath35 : data ( black crosses ) , 68% confidence area for the total background from simulation ( grey area ) , @xmath32 ( blue filled circles ) , @xmath37 ( pink empty circles ) , expected signal from the fermi bubbles according to ( [ f : a_flux][f : fb_flux2 ] ) without neutrino energy cutoff ( green dotted area ) and with 50 tev energy cutoff ( green dashed area ) .
the expected signal was scaled by a factor of 3 to allow easy comparison with the total off - zone distribution.,scaledwidth=70.0% ] the final event selection is optimised by minimising the average upper limit on the flux : @xmath38 where @xmath39 is the number of events simulated with the flux @xmath40 from ( [ f : fb_flux2 ] ) .
the method uses an approach following feldman & cousins @xcite to calculate signal upper limits with 90% confidence level , @xmath41 , for a known number of simulated background events @xmath42 .
this best average upper limit in the case of no discovery represents the sensitivity of the detector to the fermi bubbles flux @xcite . using ( [ f : fb_flux2 ] ) the average upper limit on the flux coefficient @xmath43
can be defined as : @xmath44 table [ table : mrf ] reports the optimal cuts ( @xmath28 , @xmath29 ) obtained for the four chosen cutoff energies ( @xmath45 , 500 , 100 , 50 tev ) of the neutrino source spectrum and the corresponding value of the average upper limit on the flux coefficient @xmath46 .
additionally , the optimal cuts for @xmath47 tev are applied for the other neutrino - energy cutoffs and the values @xmath48 are reported for comparison . as the obtained values @xmath49 and @xmath48 for each cutoff are similar , the 100 tev cuts are chosen for the final event selection . [
table : mrf ] .optimisation results for each cutoff of the neutrino energy spectrum .
average upper limits on the flux coefficient @xmath49 are presented in units of @xmath50 .
numbers with a star indicate the cut used for the @xmath51 calculation presented in the last row of the table . [
cols="<,>,>,>,>",options="header " , ]
high - energy neutrino emission from the region of the fermi bubbles has been searched for using data from the antares detector .
an analysis of the 20082011 antares data yielded a 1.2 @xmath52 excess of events in the fermi bubble regions , compatible with the no - signal hypothesis .
for the optimistic case of no energy cutoff in the flux , the upper limit is within a factor of three of a prediction from the purely hadronic model based on the measured gamma - ray flux .
the sensitivity will improve as more data is accumulated ( more than 65% gain in the sensitivity is expected once 20122016 data is added to the analysis ) . the next generation km3net neutrino telescope will provide more than an order of magnitude improvement in sensitivity @xcite .
the authors acknowledge the financial support of the funding agencies : centre national de la recherche scientifique ( cnrs ) , commissariat lnergie atomique et aux nergies alternatives ( cea ) , agence national de la recherche ( anr ) , commission europenne ( feder fund and marie curie program ) , rgion alsace ( contrat cper ) , rgion provence - alpes - cte dazur , dpartement du var and ville de la seyne - sur - mer , france ; bundesministerium fr bildung und forschung ( bmbf ) , germany ; istituto nazionale di fisica nucleare ( infn ) , italy ; ministerio de ciencia e innovacin ( micinn ) , prometeo of generalitat valenciana and multidark , spain ; agence de loriental , morocco ; stichting voor fundamenteel onderzoek der materie ( fom ) , nederlandse organisatie voor wetenschappelijk onderzoek ( nwo ) , the netherlands ; national authority for scientific research ( ancs - uefiscdi ) , romania ; council of the president of the russian federation for young scientists and leading scientific schools supporting grants , russia
. technical support of ifremer , aim and foselev marine for the sea operation and the cc - in2p3 for the computing facilities is acknowledged . 00 m. su , t. slatyer , d. finkbeiner , `` giant gamma - ray bubbles from fermi - lat : agn activity or bipolar galactic wind ? '' , apj * 724 * ( 2010 ) 1044 .
g. dobler `` a last look at the microwave haze / bubbles with wmap '' , apj * 750 * ( 2012 ) 17 .
e. carretti et al . , `` giant magnetized outflows from the centre of the milky way '' , nature * 493 * ( 2013 ) 66 .
s. l. snowden et al . , `` rosat survey diffuse x - ray background maps .
ii . '' , apj * 485 * ( 1997 ) 125 .
r. crocker , f. aharonian `` the fermi bubbles : giant , multi - billion - year - old reservoirs of galactic center cosmic rays '' , phys
* 106 * ( 2011 ) 101102 .
b. c. lacki , `` the fermi bubbles as starburst wind termination shocks '' , arxiv:1304.6137 ( 2013 ) . s. thoudam , `` fermi bubble @xmath53-rays as a result of diffusive injection of galactic cosmic rays '' , arxiv:1304.6972 ( 2013 ) .
d. o. chernyshov et al . , `` particle acceleration and the origin of gamma - ray emission from fermi bubbles '' , icrc2011 proceedings , arxiv:1109.2619 ( 2011 ) .
p. mertsch , s. sarkar .
`` fermi gamma - ray bubbles from stochastic acceleration of electrons . '' , phys .
lett . * 107 * ( 2011 ) 091101 .
g. dobler , i. cholis , n. weiner , `` the fermi gamma - ray haze from dark matter annihilations and anisotropic diffusion '' , apj , * 741 * ( 2011 ) 25 .
f. l. villante , f. vissani , `` how precisely neutrino emission from supernova remnants can be constrained by gamma ray observations ? '' , phys .
d * 78 * , ( 2008 ) 103007 .
antares collaboration , j.a .
aguilar et al . , `` antares : the first undersea neutrino telescope '' , nuclear inst . and methods in physics research , a * 656 * ( 2011 ) 11 - 38 .
antares collaboration , s. adrin - martnez et al .
, `` the positioning system of the antares neutrino telescope '' , jinst * 7 * ( 2012 ) t08002 .
antares collaboration , j.a .
aguilar et al . , `` time calibration of the antares neutrino telescope '' , astropart . phys . * 34 * ( 2011 ) 539 .
antares collaboration , s. adrin - martnez et al .
, `` search for cosmic neutrino point sources with four year data of the antares telescope '' , apj * 760 * ( 2012 ) 53 .
antares collaboration , j.a .
aguilar et al . , `` a fast algorithm for muon track reconstruction and its application to the antares neutrino telescope '' , astropart
* 34 * ( 2011 ) 652 - 662 .
j. schnabel , `` muon energy reconstruction in the antares detector '' , nuclear inst . and methods in physics research , a , * 725 * ( 2013 ) 106 - 109 .
j. brunner , ed .
e. de wolf , `` antares simulation tools '' , vlvnt workshop ( amsterdam ) , http://www.vlvnt.nl/proceedings.pdf , ( 2003 ) . bartol group , v. argawal et al .
, `` atmospheric neutrino flux above 1 gev '' , phys .
d , * 53 * ( 1996 ) 1314 - 1323 .
d. palioselitis , `` measurement of the atmospheric neutrino energy spectru '' , phd thesis work , http://antares.in2p3.fr/publications/thesis/2012/phdthesis_palioselitis.pdf , ( 2012 ) .
g. j. feldman , r. d. cousins , `` unified approach to the classical statistical analysis of small signals '' , phys .
d , * 57 * ( 1998 ) 3873 - 3889 .
g. c. hill , k. rawlins , `` unbiased cut selection for optimal upper limits in neutrino detectors : the model rejection potential technique '' , astropart . phys . * 19 * ( 2003 ) 393 - 402 . c. g. s. costa , `` the prompt lepton cookbook '' , astropart .
* 16 * ( 2001 ) 193 .
a. martin , m. ryskin , a. stasto , `` prompt neutrinos from atmospheric @xmath54 and @xmath55 production and the gluon at very small @xmath56 '' , acta phys .
b * 34 * ( 2003 ) 3273 .
r. enberg , m. reno , i. sarcevic , `` prompt neutrino fluxes from atmospheric charm '' , phys .
d * 78 * ( 2008 ) 043005 .
t. li , y. ma , `` analysis methods for results in gamma - ray astronomy '' , apj * 272 * ( 1983 ) 317 - 323 .
j. conrad et al . ,
`` including systematic uncertainties in confidence interval construction for poisson statistics '' , phys .
d. * 67 * 012002 ( 2003 ) .
km3net collaboration , s. adrin - martnez et al .
, `` detection potential of the km3net detector for high - energy neutrinos from the fermi bubbles '' , astropart .
phys . , * 42 * ( 2013 ) 7 - 14 . c. lunardini , s. razzaque , `` high energy neutrinos from the fermi bubbles '' , phys .
, * 108 * ( 2012 ) 221102 . i. cholis , `` searching for the high energy neutrino counterpart of the fermi bubbles signal or from dark matter annihilation '' , arxiv:1206.1607 ( 2012 ) . | analysis of the fermi - lat data has revealed two extended structures above and below the galactic centre emitting gamma rays with a hard spectrum , the so - called fermi bubbles .
hadronic models attempting to explain the origin of the fermi bubbles predict the emission of high - energy neutrinos and gamma rays with similar fluxes .
the antares detector , a neutrino telescope located in the mediterranean sea , has a good visibility to the fermi bubble regions . using data collected from 2008 to 2011 no statistically significant excess of events is observed and therefore upper limits on the neutrino flux in tev range from the fermi bubbles are derived for various assumed energy cutoffs of the source .
fermi bubbles , antares , neutrino |
throughout history , people have attempted to understand the universe by dividing matter into smaller and smaller pieces .
this approach has proven extremely fruitful : successively smaller distance scales have revealed successively simpler and more fundamental structures . over the last century , the fundamental building blocks of nature have been reduced from atoms to electrons , protons and neutrons , to most recently , the `` elementary '' particles that make up the @xmath1 standard model . today , a great deal of research is devoted to finding even more fundamental building blocks - such as superstrings .
this entire approach is based on the idea of reductionism - the idea that the fundamental nature of particles is revealed by dividing them into smaller pieces .
but reductionism is not always useful or appropriate .
for example , in condensed matter physics there are particles , such as phonons , that are collective excitations involving many atoms .
these particles are `` emergent phenomena '' that can not be meaningfully divided into smaller pieces . instead , we understand them by finding the _ mechanism _ that is responsible for their emergence . in the case of phonons , for example , this mechanism is symmetry breaking.@xcite this suggests alternate line of inquiry .
could the elementary particles in the standard model be analogous to phonons ?
that is , could they be collective modes of some `` structure '' that we mistake for empty space ?
recent work suggests that they might be .
@xcite this work has revealed the existence of new and exotic phases of matter whose collective excitations are gauge bosons and fermions .
the microscopic degrees of freedom in these models are spins on a lattice - purely local , bosonic objects with local interactions .
there is no trace of gauge boson or fermion degrees of freedom in the underlying lattice model .
the gauge bosons and fermions are thus emergent phenomena - a result of the collective behavior of many spins .
what is the mechanism responsible for their emergence ? in these exotic phases , the spins organize into a special pattern - a particular kind of entangled ground state , which we call a `` string - net condensed '' state .
a string - net condensed state is a spin state where the spins organize into large string - like objects ( or more generally networks of strings ) .
the strings then form a quantum string liquid ( see fig . [ stringnets ] ) .
this kind of ground state naturally gives rise to gauge bosons and fermions .
the gauge bosons correspond to fluctuations in the strings - the collective motions of the strings that fill the space .
@xcite the fermions correspond to endpoints of the strings - that is , defects in the string liquid where a string ends in empty space .
@xcite what makes the string - net picture particularly compelling is that the gauge bosons and fermions naturally emerge _
together_. they are just different aspects of the same underlying structure .
therefore , if we believe that the vacuum is such a string - net condensate then the presence of gauge interactions and fermi statistics in the standard model is no longer mysterious .
string - net condensation explains what gauge bosons and fermions are , why they exist , and why they appear together .
@xcite the general theory of string - net condensation was worked out in .
one of the main results in that paper was a series of exactly soluble models realizing all possible string - net condensates .
these models are quite general and can realize gauge bosons with any gauge group
. however , they are also complicated when discussed in full generality , and did not provide an explicit example of the most physically relevant case - a model realizing gauge bosons and fermions in ( 3 + 1 ) dimensions . in this paper , we attempt to remedy this problem .
we demonstrate the string - net picture of ( 3 + 1)d emerging gauge bosons and fermions with concrete lattice models .
we describe a rotor model on the cubic lattice that produces both @xmath0 gauge bosons and fermions .
the fermions can be gapped excitations ( as in an insulator ) or gapless ( as in a fermi liquid )
. they can also behave like massless dirac fermions . in this case
, the low energy physics of the rotor model is identical to massless quantum electrodynamics ( qed ) .
the rotor model can then be viewed as a `` quantum ether '' : a medium that gives rise to both photons and electrons .
in addition , the rotor model is closely related to @xmath0 lattice gauge theory coupled to a higgs field .
it demonstrates that a simple modification or `` twist '' can change the higgs boson into a fermion .
while this is not the first lattice bosonic model with emergent massless gauge bosons and massless dirac fermions @xcite , it has two characteristics which distinguish it from previous examples .
first , the mapping between the rotor model and qed is essentially exact , and does not require a large @xmath2 limit or similar approximation .
second , the rotor model is a special case of a general construction , @xcite , unlike the other models which were in some sense , discovered by accident .
it therefore provides a deeper understanding of emergent fermions and gauge bosons .
in addition to its relevance to high energy physics , this understanding may prove useful to condensed matter physics , particularly the search for phases of matter with exotic low energy behavior .
the paper is organized as follows : we begin with a `` warm - up '' calculation in section ii - a rotor model with emergent photons and _ bosonic _ charges .
this model is closely related to @xmath0 lattice gauge theory coupled to a higgs field .
then , in section iii , we show that the rotor model can be modified in a natural way by adding a phase factor or `` twist '' to a term in the hamiltonian .
this modified or `` twisted '' rotor model has ( massive ) fermionic charges . in section
iv , we describe further modifications which give rise to gapless fermi liquids and massless dirac fermions .
in this section , we present a warm - up example - a rotor model with emergent photons and _ bosonic _ charges .
this model is closely related to @xmath0 lattice gauge theory coupled to a higgs field . while a number of other similar models @xcite have been analyzed previously , this model has the advantage of being quasi - exactly soluble , and generalizing easily to the fermionic case .
the string - net picture of the ground state is also particularly evident in this case . ) .
the term @xmath3 acts on the six `` legs '' of @xmath4 - that is , the six rotors adjacent to @xmath4 , drawn above as filled dots .
the term @xmath5 acts on the four rotors , labeled by @xmath6 , along the boundary of the plaquette @xmath7 . ]
the model is a quantum rotor model with rotors on the links of a cubic lattice .
each rotor can be viewed as a particle moving on a circle .
the position of the particle is given by the angle @xmath8 , and the angular momentum of the particle by @xmath9 , where @xmath10 is the moment of inertia .
the hamiltonian of the rotor model is given by ( see fig . [ cubbos]a)@xcite @xmath11 where @xmath12 labels the vertices , @xmath13 labels the links and @xmath7 labels the plaquettes of the cubic lattice .
the `` legs of @xmath4 '' are the six links that are attached to the vertex @xmath4 , and 1 , 2 , 3 , 4 label the four links of the plaquette @xmath7 ( see fig . [ cubbos ] ) .
@xmath14 are the raising / lowering operators of @xmath15 and @xmath16 . , while all the other rotors ( depicted as empty dots ) have @xmath17 . ] in this section , we show that the rotor model exhibits string - net condensation in the regime @xmath18 .
in particular , we show that the rotors organize into effective extended objects , namely `` string - nets '' , and the ground state is a quantum liquid of these string - nets .
the string - net condensed ground state has two types of excitations - string collective motions ( which will correspond to photons ) and string endpoints ( which will correspond to charges ) .
the first step is to notice that the @xmath19 operators commute with each other as well as the other terms in the hamiltonian .
thus , we can label all the eigenstates of @xmath20 by their eigenvalues under @xmath19 : @xmath21 . the quantum number @xmath22 - which we will call the `` charge '' on site @xmath4 - takes values in the integers .
clearly the lowest energy charge configuration is the charge-@xmath23 configuration @xmath24 .
other charge configurations cost an additional energy @xmath25 .
hence , in the limit @xmath26 , the low energy physics is completely contained in the charge-@xmath23 sector .
we would like to enumerate the states in the charge-@xmath23 sector .
it is natural to describe these states in terms of extended objects - in particular , string - nets .
the simplest state has @xmath27 for every rotor . in the string - net language , we think of this state as the vacuum
. another state can be obtained by alternately increasing or decreasing @xmath15 by @xmath28 along a loop ( see fig .
[ cubcstr ] ) .
we think of this state as the vacuum with a single ( closed ) string .
other states can be constructed by repeating this process .
one finds that the most general charge-@xmath23 state can have many strings and that the strings can overlap to form networks of strings or `` string - nets . ''
what is the action of the hamiltonian ( [ strneth ] ) on these charge-@xmath23 states ? if we think about the states using the string - net language , we see that the @xmath29-term penalizes strings for being long .
it therefore corresponds to a string tension . on the other hand
, the effect of the @xmath30 operator is to either create a small loop of small string ( if applied to the vacuum ) or to deform the existing strings ( if applied to a more complicated state ) .
thus the @xmath31-term generates string `` hopping '' or string fluctuations .
one can think of it as a string kinetic energy .
there are two regimes to consider . when @xmath32 , the string tension dominates and the ground state contains almost no strings .
the ground state is a `` normal '' state . when @xmath33 , the string kinetic energy dominates and the ground state is a superposition of many large closed strings @xcite .
the ground state is thus `` string - net condensed . ''
we expect a quantum phase transition between these two states at some @xmath34 of order @xmath28 .
we expect that the excitations above the ground state have very different properties in these two regimes . in the string - net condensed case , there are two types of excitations : low energy excitations with charge-@xmath23 and high energy excitations with nonzero charge . the low energy excitations can be constructed from linear combinations of string - net states .
these excitations can be thought of as collective motions of closed strings .
the high energy , charged , excitations can be constructed from linear combinations of string - net configurations with _
open _ strings .
for example , if one takes the vacuum state @xmath17 and alternately increases and decreases @xmath15 along an open path @xmath35 , the resulting state has nonzero charge at the two ends of @xmath35 .
this is true quite generally : nonzero charge configurations are made up of open string - nets - with the charge located at the endpoints of the open strings .
we can think of the string endpoints as point defects in the condensate .
the higher energy excitations are composed out of these point defects .
the defects carry a quantum number - charge - which is measured by the operator @xmath19 . in this section
, we examine the rotor model more quantitatively .
we show that the low energy physics of the rotor model ( [ strneth ] ) is described by compact @xmath0 gauge theory coupled to infinitely massive charges .
the two types of excitations discussed in the previous section have a natural interpretation in terms of the gauge theory .
the collective motions of the strings give rise to gapless excitations and behave like photons , while the point defects ( the endpoints of the strings ) are gapped and behave like charged particles . in the gauge theory language ,
the strength of the interaction between the photons ( the gauge field ) and the charges is characterized by a `` fine structure constant '' @xmath36 .
the `` fine structure constant '' also characterizes the strength of quantum fluctuations of the gauge field .
when @xmath37 , the quantum fluctuations are so large that the gauge theory is in the confining phase and there are no gauge bosons at low energies .
the gauge bosons ( such as photons ) can exist at low energies only in the deconfined phase when @xmath36 is small . in the following
, we will show that the `` fine structure constant '' @xmath36 of the gauge theory is of order @xmath38 .
the `` speed of light '' @xmath39 is of order @xmath40 where @xmath41 is the lattice constant .
this result provides insight into the phase transition between the normal and string - net condensed states : the normal state corresponds to the large @xmath36 confining phase of the gauge theory , while the string - net condensed state corresponds to the small @xmath36 deconfined phase .
the string - net condensation transition is the usual confinement - deconfinement transition from gauge theory .
the simplest way to derive these results is to map ( [ strneth ] ) onto a lattice gauge theory hamiltonian . consider the following hamiltonian for compact @xmath0 gauge field coupled to a charge-@xmath28 higgs field
@xmath42 : @xmath43 here , @xmath44 , @xmath45 is an alternative notation for links @xmath13 , and plaquettes @xmath7 in the cubic lattice .
the operator @xmath46 is the integer valued electric field , canonically conjugate to the vector potential : @xmath47 = i$ ] .
similarly , @xmath48 is the occupation number operator canonically conjugate to the higgs field : @xmath49 = i$ ] . in lattice gauge theory ,
one is interested in the properties of this hamiltonian within the gauge invariant subspace where gauss law holds : @xmath50 our claim is that within this gauge invariant subspace , @xmath51 is mathematically equivalent to @xmath20 . to see this , note that the electric field operators and the occupation number operators all commute with each other .
so the lattice gauge theory model has a complete basis consisting of states @xmath52 with electric field @xmath53 , and occupation number @xmath48 , @xmath54 similarly , the rotor model @xmath20 has a complete basis consisting of states @xmath55 with angular momentum @xmath56 . on can map the rotor model hilbert space onto the gauge invariant subspace of the lattice gauge theory by mapping the basis state @xmath55 onto the basis state @xmath57 given by @xmath58 ( here @xmath13 is the link connecting sites @xmath4 and @xmath59 ) .
clearly , this correspondence maps the operators @xmath48 and @xmath46 onto the operators @xmath19 and @xmath60 .
one can also show that this correspondence maps @xmath61 onto @xmath62 and @xmath63 onto @xmath64 .
examining the two hamiltonians , we conclude that the gauge theory _ in the special case @xmath65 _ can be mapped onto the rotor model .
the physics of the rotor model ( [ strneth ] ) is therefore completely equivalent to a @xmath0 gauge field coupled to bosonic charges .
the fine structure constant @xmath36 and speed of light @xmath39 can be computed using the lattice gauge theory hamiltonian ( [ lgth ] ) ) . @xcite as for the mass of the charges , the rotor model ( [ strneth ] ) corresponds to the special case where @xmath66 - the case where the charges are infinitely massive and have no dynamics . in the previous section
, we discovered that the point defect excitations have an infinite mass .
this is not surprising , since the operator @xmath19 commutes with the hamiltonian ( [ strneth ] ) , and hence charge configurations are completely static .
we would like to modify @xmath67 so that the point defects ( or `` charges '' ) do have dynamics . to do this ,
we need to add a term , @xmath68 , to the hamiltonian that does nt commute with @xmath19 .
such a term will make the charges hop from site to site .
we expect that any @xmath68 will give rise to the small qualitative behavior ( e.g. same particle statistics ) .
we therefore consider the simplest hopping term : @xmath69 it s easy to see that this term makes the charges hop from site to site just as in a nearest neighbor tight binding model .
in fact , if we examine the mapping , ( [ lgtmap ] ) , we see that @xmath68 corresponds exactly with the nearest neighbor hopping term in lattice gauge theory , @xmath70 .
therefore , the low energy physics of the rotor hamiltonian @xmath71 is equivalent to that of a @xmath0 gauge field coupled to charges with _ finite _ mass . in this section ,
we compute the statistics of the charges and show that they are bosons .
the main purpose of this computation is to facilitate comparison with the `` twisted '' rotor model .
( indeed , the fact that the charges are bosons follows immediately from the exact mapping to lattice gauge theory , ) .
it is easiest to perform the computation in the case @xmath72 . in this case ,
@xmath30 commutes with the hamiltonian @xmath73 .
we can therefore divide the hilbert space into different sectors @xmath74 corresponding to different flux configurations @xmath75 .
a complete basis for each sector can be obtained by listing all the different charge configurations @xmath21 .
the action of the hamiltonian @xmath73 on a charge configuration @xmath21 is simple .
the first part of the hamiltonian , @xmath67 , does nt affect the charge configuration at all , while the second part , @xmath68 acts in two ways : it either creates two charges at neighboring sites , or it makes a charge hop from one site to another . thus , within each sector , @xmath76 is simply a hopping hamiltonian on the cubic lattice .
the hopping operators @xmath77 make the charges hop from site @xmath4 to site @xmath59 , where @xmath78 are the two endpoints of @xmath13 .
to compute the statistics of the charges , we use the statistical hopping operator algebra described in .
we note that the hopping operators satisfy the algebra @xmath79 for any @xmath80 incident to some vertex @xmath4 . according to , this is the bosonic hopping operator algebra .
we conclude that the charged particles are bosons .
in this section , we present another rotor model which is very closely related to the model from the previous section , ( [ strneth ] ) .
the hamiltonian only differs from @xmath67 by a `` twist '' - an additional phase factor in the @xmath30 term . despite the apparent similarity
, we will see that this `` twisted '' rotor model gives rise to emergent photons and _ fermionic _ charges .
the relationship between the twisted and untwisted models is a special case of a general relationship between fermionic and bosonic string - net condensates .
@xcite it is part of a systematic and general construction , unlike previous examples of emergent photons and fermions @xcite , which were in some sense discovered by accident .
cubic lattice onto the @xmath81 plane .
( b ) an example of a curve @xmath82 with a framing @xmath83 .
the links along the string are labeled @xmath84 .
the links which cross @xmath83 are the filled dots labeled by @xmath85 . the corresponding twisted string operator @xmath86 is given by @xmath87 ] in order to motivate the twisted rotor model
, we first discuss the `` twisted string operator . ''
we begin with the original `` untwisted '' rotor model @xmath20 .
notice that the operator @xmath88 is a special case of a general string operator @xmath89 that can be associated with any curve @xmath90 in the cubic lattice : @xmath91 @xmath30 corresponds to the case where @xmath82 is the boundary of a plaquette @xmath92 .
this `` untwisted '' string operator has an important property .
when @xmath82 is a closed curve , @xmath89 commutes with all charge operators and all other closed string operators : @xmath93 = [ w(c'),w(c ) ] = 0\ ] ] this property is essential for the mapping to lattice gauge theory ( [ lgtmap ] ) to hold .
it implies that the @xmath89 can be simultaneously diagonalized within a given charge sector .
the simultaneous eigenstates can then be interpreted as different flux configurations , and the @xmath89 can be interpreted as wilson loop operators measuring the flux through a curve @xmath82 .
it turns out that there is another string operator - which we call the `` twisted string operator '' - that also has this property .
the twisted string operator @xmath94 is similar to the usual string operator @xmath89 except that it contains an additional phase factor that depends on rotors on the `` legs '' of @xmath82 - that is , rotors on the links incident to vertices @xmath95 along the curve @xmath82 . to define @xmath94
, one must first choose a projection of the @xmath96 lattice onto a @xmath81 plane .
any projection will work ; in this paper , we choose the particular projection illustrated in fig .
[ twstr](a ) .
the twisted string operator is then defined by @xmath97 here , @xmath98 is the number of times that the link @xmath13 crosses the curve @xmath83 , where @xmath83 is some `` framing '' of @xmath82 .
a `` framing '' is a curve @xmath83 drawn next to @xmath82 , but not exactly on it ( see fig . [ twstr](b ) ) .
the choice of framing , @xmath83 , is not unique , so whenever we discuss @xmath99 , we will need to specify a particular choice of framing .
one can check directly that both of the equalities in hold for the twisted string @xmath99 .
alternatively , one can refer to the general argument given in .
term in the twisted rotor model ( [ twistedh ] ) . just like any other twisted string operator , the term @xmath100 acts on the four rotors , labeled by @xmath6 , along the boundary of the plaquette @xmath7 , and the @xmath101 rotors , labeled by @xmath102 , that cross the framing curve . ] the twisted model is obtained by modifying the @xmath30 term in ( [ strneth ] ) . instead of using a @xmath30 term based on the usual string operator , that is , @xmath103
, we use a modified @xmath30 term based on the twisted string operator : @xmath104 .
we use a framing obtained by taking the plaquette boundary @xmath92 and shifting it up and to the left ( see fig .
[ cubferm ] ) .
the result is the twisted rotor hamiltonian @xmath105 where the above explicit definition of @xmath106 applies to plaquettes in the @xmath107 plane ( see fig .
[ cubferm ] ) .
( the definition for plaquettes in the @xmath108 and @xmath109 planes is similar and can be obtained using @xmath104 together with the above framing convention ) .
note that that ( [ twistedh ] ) is very similar to the original rotor model ( [ strneth ] ) ( and hence also very similar to lattice gauge theory ) .
the @xmath106 term alternately increases and decreases @xmath15 along the plaquette boundary @xmath92 , just like the @xmath30 term in the model ( [ strneth ] ) .
the only difference is that here the amplitude for this process is not always @xmath28 .
the amplitude is @xmath110 , depending on the rotors on the legs of the plaquette boundary .
the twisted rotor model can be analyzed in the same way as the untwisted model ( sec .
[ strnetpict ] ) .
the charge operators @xmath19 commute with the hamiltonian , so all the states are associated with a charge configuration @xmath21 .
low energy states have charge-@xmath23 and can be thought of as string - net states .
they can be enumerated in the same way as in the untwisted model . just as the untwisted closed - string states could be obtained by repeatedly applying different @xmath30 s to the vacuum state ( the state with @xmath111 ) ,
the twisted closed - string states can be obtained by repeatedly applying different @xmath106 s to the vacuum state . just as before , the @xmath29 term can be interpreted as a string tension , and the @xmath31 term as a string kinetic energy . when @xmath33 , string - net condensation occurs and the resulting ground state has two types of excitations : low energy string collective motions , and higher energy defects in the condensate .
more quantitatively , one can show that the low energy physics of the twisted rotor model is described by a compact @xmath0 gauge field coupled to infinitely massive charges - just like the untwisted model .
in fact , the twisted and untwisted rotor hamiltonians ( [ twistedh ] ) , ( [ strneth ] ) are completely equivalent and can be mapped onto one another . to see this , note that the operators @xmath19 and @xmath106 commute with each other and can therefore be simultaneously diagonalized .
let the simultaneous eigenstates be denoted @xmath112 where @xmath113 and the @xmath22 , @xmath114 are integers and real numbers , respectively .
the states @xmath112 form a complete basis for the twisted rotor model .
similarly , we can simultaneously diagonalize @xmath19 and @xmath30 , to construct a complete basis @xmath115 for the untwisted model .
note that the matrix elements of @xmath116 in the twisted basis are the same as those of @xmath117 in the untwisted basis .
furthermore the two sets of operators , @xmath118 and @xmath119 , satisfy the same commutation relations .
thus , @xmath120 must also have the same matrix elements in the two models .
we conclude that the two hamiltonians , ( [ twistedh ] ) and ( [ strneth ] ) , have exactly the same matrix elements ( in their respective bases ) .
the two models are therefore equivalent .
as before , we would like to modify the hamiltonian @xmath121 so that the point defect excitations ( or `` charges '' ) have a finite mass . to accomplish this , we need to add a term @xmath122 that does nt commute with @xmath19 .
such a term will make the charges hop from site to site .
we expect that any hopping term will give rise to the same qualitative behavior ( e.g. same particle statistics ) .
however , the resulting hamiltonian is easier to analyze if we pick a hopping term which commutes with @xmath106 .
while in the untwisted case the simplest hopping term happened to have this property , in the case at hand we have to work a little harder .
we notice that the untwisted hopping term @xmath68 is simply @xmath123 where @xmath124 is the shortest open string operator ( the string consists of a single link @xmath13 ) .
by analogy we choose the twisted hopping term to be @xmath125 with a framing convention given by drawing the framing curve just down and to the right of the link @xmath13 .
the result is ( see fig .
[ cubhop ] ) : @xmath126 where @xmath127 ( one can check that @xmath128 does in fact commute with @xmath106 - either by explicit calculation or using the general properties of twisted string operators ) .
@xcite note that the term @xmath122 causes the charges to hop from sites to neighboring sites .
the low energy physics of @xmath129 is therefore described by a compact @xmath0 gauge field coupled to charges with _ finite _ mass
. however , these charges are not higgs bosons as in the untwisted case .
indeed , with the addition of charge dynamics , the two hamiltonians @xmath129 , @xmath130 , are no longer equivalent .
the two models are only equivalent in the special case where @xmath131 and the charges have infinite mass .
as soon as the charges can hop , the mapping between the two models breaks down , as does the mapping to lattice gauge theory .
the reason for this is that the charged particles in @xmath129 are not bosons , but _
fermions _ - as we will see in the next section . .
the @xmath132 are defined as twisted string operators @xmath133 , with the framing curve drawn just below and to the right of @xmath13 , as shown above .
the result is @xmath134 when @xmath13 is the link labeled @xmath135 pointing in the @xmath136 direction , @xmath137 , when @xmath13 is the link labeled @xmath138 pointing in the @xmath139 direction , and @xmath140 when @xmath13 is the link labeled @xmath28 pointing in the @xmath141 direction . ] in this section , we compute the statistics of the charges in the twisted rotor model .
we use the same technique as in sec.[statbos ] . as before
, we perform the computation in the case @xmath72 .
since @xmath142 commutes with the hamiltonian @xmath143 , we can divide the hilbert space into different sectors @xmath74 corresponding to different flux configurations @xmath144 .
the states within each sector can be labeled by their charge configuration @xmath21 .
the action of the hamiltonian @xmath143 on a charge configuration @xmath21 is simple .
the first part of the hamiltonian , @xmath121 , does nt affect the charge configuration at all , while the second part , @xmath122 acts in two ways : it either creates two charges at neighboring sites , or it makes a charge hop from one site to another . thus , within each sector , @xmath145 is simply a hopping hamiltonian on the cubic lattice .
the hopping operators @xmath128 make the charges hop from site @xmath4 to site @xmath59 , where @xmath78 are the two endpoints of @xmath13 . to compute the statistics of the charges , we use the statistical hopping operator algebra @xcite , just as in the untwisted case .
we find that the hopping operators satisfy the same algebra but with a @xmath146 sign .
that is , @xmath147 for any @xmath80 incident to some vertex @xmath4 .
this sign makes all the difference .
the above algebra is not the bosonic hopping operator algebra , but rather the _ fermionic _ hopping algebra .
we conclude that in the twisted rotor model the charged particles are fermions .
one might expect that , by analogy with the untwisted case , there is a mapping to fermionic lattice gauge theory .
however , no such mapping exists .
the reason is that , at the lattice scale , these fermions behave differently from the usual fermions in a tight binding model .
there are two key differences .
first , in a standard tight binding model of spinless fermions , each site can be occupied by @xmath23 or @xmath28 fermion .
however , in our case the charge @xmath19 can be any integer from @xmath148 to @xmath149 .
thus , a given site can be occupied by arbitrarily many fermions .
second , in a standard tight binding model , the fermions are totally noninteracting in the zero coupling limit @xmath150 .
however , in our case the the fermions do interact , even in this limit .
for example , the hopping term @xmath68 allows positive and negatively charged fermions to annihilate each other if they occupy neighboring sites .
neither of these differences should affect the physics of the twisted rotor model in the low energy limit , if @xmath25 is large .
in that case , the low energy physics of the rotor model is equivalent to an insulator : it is described by two species of gapped fermions with opposite charges coupled to a @xmath0 gauge field .
however , these microscopic differences could be important when @xmath25 becomes small enough that a fermi surface develops .
unfortunately , in order to construct massless dirac fermions , we need to consider that limit .
therefore , in order to make our calculation well - controlled , we need to modify the twisted rotor model so that the fermions are equivalent to tight binding model fermions , _ even at the lattice scale_. this is the subject of the next section .
in this section , we modify the twisted rotor model @xmath151 so that the fermions are exactly equivalent to tight binding model fermions .
we then show that this more well - controlled model can give rise to massless dirac fermions .
it can also give rise to fermi liquid behavior .
there are many ways to make the fermions in @xmath151 behave like the usual fermions from a tight binding model . here
we will only describe one of them .
first , we divide the vertices of the cubic lattice into @xmath135 classes : `` @xmath152 vertices '' , `` @xmath153 vertices '' and `` mixed vertices . ''
a vertex is called an @xmath152 vertex if the three components @xmath154 are all even , an @xmath153 vertex if the three components @xmath154 are all odd , and a mixed vertex otherwise .
we then modify the @xmath25 term in ( [ strneth ] ) to @xmath155 throughout our discussion we will take the @xmath156 limit . in this limit , @xmath19 must be @xmath157 on the @xmath152 vertices , @xmath158 on the @xmath153 vertices , and @xmath23 on the mixed vertices . in other words ,
the @xmath0 charges can only live on the @xmath152 or @xmath153 vertices , and these two sublattices can only contain unit positive and negative charges respectively .
next , we modify the hopping hamiltonian @xmath159 to @xmath160 where the sum runs over pairs of neighboring links @xmath161 which are collinear and can be joined together to form a single line segment of length @xmath138
. this hopping term has the property that charges only move within the @xmath152 and @xmath153 sublattices .
putting this all together , and including a term which will correspond to a chemical potential @xmath162 , we arrive at the following rotor model : @xmath163 we begin by considering the case where @xmath164 . in this case
@xmath165 is exactly soluble .
indeed , the basis states @xmath166 are eigenstates of @xmath165 with an energy @xmath167 as we mentioned earlier , we will take the @xmath168 limit . in this limit ,
the only low energy states are those with @xmath169 in the @xmath152 sublattice , @xmath170 in the @xmath153 sublattice and @xmath171 everywhere else .
if we restrict to this low energy subspace , the energy @xmath172 is simply given by @xmath173 where @xmath174 is the total number of charged particles .
now , consider the case @xmath175 .
the @xmath176 term generates nearest neighbor hopping in the @xmath152 and @xmath153 sublattices .
if we let @xmath35 denote the projection onto the low energy subspace , then the action of the hopping hamiltonian ( [ ffhop ] ) within this subspace can be written as @xmath177 where @xmath178 runs over the unit vectors @xmath179 and the @xmath180 are hopping operators that make the particles hop from site @xmath181 to site @xmath4 .
they are defined by @xmath182 where @xmath13 is the link connecting @xmath183 , @xmath184 is the link connecting @xmath185 , and we use the upper ( lower ) signs for the @xmath152 ( @xmath153 ) sublattices . the low energy effective hamiltonian can then be written as @xmath186 this hamiltonian describes two types of hardcore charged particles with charges @xmath110 hopping on two different sublattices .
we can completely characterize this hamiltonian by investigating two physical properties : the statistics of the charges , and the gauge flux through each plaquette @xmath7 . to determine the statistics of the charges , we use the statistical hopping operator algebra .
@xcite we notice that the hopping operators satisfy the relation @xmath187 for any @xmath188 adjacent to @xmath4 . according to @xcite , this is the fermionic hopping operator algebra , so the particles are fermions .
to determine the flux that the particles see , we compute the product of hopping operators around a @xmath189 plaquette @xmath190 in the @xmath152 or @xmath153 sublattice .
we find that ( up to some signs having to do with orientation conventions ) , @xmath191 here the sum on the right hand side runs over the @xmath192 plaquettes in the original lattice that are contained in the ( doubled ) plaquette @xmath190 .
also , @xmath193 denotes the occupation number of site @xmath4 .
this relation implies that the particles see a flux @xmath114 through each plaquette @xmath7 .
the hopping operators @xmath194 are completely characterized by the two algebraic relations ( [ ferhop]),([ferflux ] ) .
any collection of hardcore hopping operators that satisfy these relations will give rise to a hamiltonian equivalent to ( [ heff ] ) .
a particularly simple collection of hopping operators satisfying these relations can be constructed from fermionic operators @xmath195 with @xmath196 the hopping operators are given by @xmath197 where the upper ( lower ) sign applies to the @xmath152 ( @xmath153 ) sublattice and @xmath198 are phases defined on each link satisfying @xmath199 for any closed curve @xmath82 .
in other words , @xmath200 is a vector potential for the flux configuration @xmath114 .
rewriting the low energy effective hamiltonian in terms of these fermionic operators gives @xmath201 note that this is nothing other than the standard tight binding hamiltonian for fermions hopping in fixed flux configurations @xmath114 .
if we now allow @xmath202 , then the flux configurations acquire dynamics .
the hamiltonian ( [ hfree ] ) becomes modified by the addition of an electric energy term @xmath203 , where @xmath46 is canonically conjugate to @xmath204 : @xmath205 = i$ ] .
this term is precisely what one needs for compact lattice @xmath0 gauge theory .
we conclude that in the general case @xmath206 , the twisted rotor model ( [ hff ] ) is mathematically equivalent to a tight binding model with @xmath110 charged fermions coupled to a compact @xmath0 gauge field .
the hamiltonian @xmath207 naturally gives rise to fermi liquid behavior .
this is easiest to understand in the regime @xmath208 . in this case
, the minimum energy flux configuration is the configuration @xmath209 .
the flux @xmath114 fluctuates about this minima giving rise to photon - like excitations .
the fermion dispersion in the background flux configuration is given by @xmath210 the ground state is obtained by filling all the negative energy levels . clearly , the model behaves differently for different values of chemical potential @xmath162 . if @xmath211 , then there are no negative energy levels .
all the energy levels are empty in the ground state , and the system is an insulator . in this case ,
the rotor model ( [ hff ] ) describes gapped fermions with a gap @xmath212 coupled to a compact @xmath0 gauge field .
the fermions come in two species ( corresponding to the two sublattices , @xmath152 and @xmath153 ) with charges @xmath213 and @xmath214 , respectively .
hence , just like the original twisted rotor model @xmath145 , the physics is equivalent to that of a @xmath0 gauge field coupled to two species of gapped fermions with opposite charges . on the other hand , if @xmath215 , then the result is a fermi liquid - or more precisely two fermi liquids corresponding to the two species of fermions .
these two fermi liquids are both coupled to a compact @xmath0 gauge field .
the model ( [ hfree ] ) can also give rise to massless dirac fermions .
indeed , massless dirac fermions can occur whenever the fermion band structure has nodal points .
this happens naturally in a @xmath216-flux configuration where each plaquette contains a flux of @xmath216 . in our case , the relevant plaquettes are not plaquettes @xmath7 in the original cubic lattice , but rather `` doubled '' @xmath189 plaquettes in the @xmath152 and @xmath153 sublattices . one way to ensure that the flux through these `` doubled '' plaquettes is @xmath216 , is to introduce a spatial dependence into the parameter @xmath176 .
that is replace @xmath176 by @xmath217 where @xmath218 on the @xmath152 sublattice and @xmath219 on the @xmath153 sublattice .
the result is that becomes modified so that each fermion sees an effective flux of @xmath220 through each doubled plaquette @xmath190 .
the minimum energy flux configuration is still @xmath209 , but now the fermions see an effective flux of @xmath216 through each doubled plaquette .
the fermion dispersion in this background is given by @xmath138 pairs of degenerate bands with energies @xmath221 and brilluoin zone , @xmath222 .
the ground state of @xmath223 is again obtained by filling all the levels with negative energies . if @xmath224 , then the fermi surface consists of @xmath138 nodal points with linear dispersion - at @xmath225 .
the low energy fermionic excitations are concentrated near these @xmath138 nodes .
together , the two nodes give rise to one four - component dirac fermion in the continuum limit .
since the bands are doubly degenerate , and there are two species of fermions ( corresponding to the @xmath152 and @xmath153 sublattice ) , there are a total of @xmath192 massless four component dirac fermions . we conclude that the low energy physics of the rotor model @xmath165 with @xmath224 , @xmath226 , @xmath227 , and @xmath176 given by the above formula , is equivalent to qed with four species of massless dirac fermions .
thus , @xmath165 is a local bosonic model that gives rise to both photons and massless fermions .
it can be viewed as a realization of a quantum ether - an ether that produces not only light , but also electrons !
we would like to point out that the speed of light @xmath228 and the speed of the massless dirac fermion @xmath229 are not the same in our rotor model .
however , we can tune the value of @xmath176 to make @xmath230 . in this case the low energy effective theory will have lorentz invariance . the reason that we do not get a lorentz invariant low energy
effective theory naturally is that our approach is based on hamiltonian formalism where space and time are treated very differently .
if one uses a path integral formalism and discretizes space and time in the same way , lorentz invariance can emerge naturally in a low energy effective theory .
another difficulty with the above model is that the emergent electrons are massless .
however , this problem can also be overcome : when bosonic models contain emergent non - abelian gauge bosons , the models can produce electrons with a _
finite _ mass and such a mass can be much less than the cut - off scale without fine tuning any parameters .
in this paper we have explicitly demonstrated the string - net picture of ( 3 + 1)d emerging gauge bosons and fermions .
we presented a rotor model on the cubic lattice with a string - net condensed ground state , and excitations which behave just like massless photons and massless dirac fermions .
we saw that the model was closely related to @xmath0 lattice gauge theory coupled to a higgs field .
the only real difference was a phase factor or `` twist '' in the hamiltonian .
surprisingly , this `` twist '' was enough to turn the higgs boson into a fermion .
this model is part of a much more general construction @xcite that can produce gauge bosons with any gauge group . in particular
, one should be able to construct a model that gives rise to @xmath231 gauge bosons and massless dirac fermions - that is , qcd . these models could have applications to lattice gauge theory .
they may make it possible to simulate qed with electrons and qcd with quarks without ever introducing quark or electron degrees of freedom on the sites . only the link variables ( representing the gauge field )
would be necessary in such a simulation . from a high energy point of view
, the string - net picture of the vacuum is quite appealing .
it explains why the standard model looks the way it does - that is , why nature chooses such peculiar things as gauge bosons and fermions to describe itself .
in addition , it unifies the mysterious gauge symmetries and anticommuting fields into a single underlying structure : a string - net condensate .
but can we actually construct a string - condensed local bosonic model that produces the entire standard model ?
we are close , but not quite there . in terms of elementary particles
we can produce photons , gluons , leptons and quarks , but we do not know how to produce neutrinos or @xmath232 gauge bosons .
the problem with the neutrinos and the @xmath232 gauge bosons is the famous chiral - fermion problem .
@xcite neutrinos are chiral fermions and the @xmath232 gauge bosons couple chirally to other fermions . at the moment
, we do not know how to obtain chiral fermions and chiral gauge theories from _ any _ local lattice model , much less a local bosonic model . from a condensed matter point of view , the above model shows how a simple spin system can give rise to emergent fermions and gauge bosons .
extended objects can give rise to both of these phenomena , as long as their condensate has an appropriate `` twist . ''
this may provide intuition in the search for new and exotic phases of matter - phases beyond the scope of landay s theory of symmetry breaking .
the discovery of a real material with string - net condensation would represent a breakthrough , particularly a material containing excitations which behave just like the photons and electrons in our vacuum . | we give an example of a purely bosonic model a rotor model on the 3d cubic lattice whose low energy excitations behave like massless @xmath0 gauge bosons and massless dirac fermions .
this model can be viewed as a `` quantum ether '' : a medium that gives rise to both photons and electrons .
it illustrates a general mechanism for the emergence of gauge bosons and fermions known as `` string - net condensation . ''
other , more complex , string - net condensed models can have excitations that behave like gluons , quarks and other particles in the standard model .
this suggests that photons , electrons and other elementary particles may have a unified origin : string - net condensation in our vacuum . |
the nearest molecular cloud complex to the sun ( distance @xmath0 65 pc ) consists of clouds 11 , 12 , and 13 from the catalog of magnani et al .
( 1985 ) and is located at ( l , b ) @xmath0 ( 159.4,@xmath534.3 ) .
this complex of clouds ( which we will refer to as mbm12 ) was first identified by lynds ( 1962 ) and appears as objects l1453-l1454 , l1457 , l1458 in her catalog of dark nebulae .
the mass of the entire complex is estimated to be @xmath0 30200 m@xmath6 based on radio maps of the region in @xmath7co , @xmath8co and c@xmath9o ( pound et al .
1990 ; zimmermann & ungerechts 1990 ) .
recently , there has been much interest in understanding the origin of many isolated t tauri stars ( tts ) and isolated regions of star - formation .
for example , within @xmath0 100 pc from the sun there are at least two additional regions of recent star - formation : the tw hydrae association ( distance @xmath0 50 pc ; e.g , kastner et al .
1997 ; webb et al . 1999 ) and the @xmath10 chamaeleontis region ( distance @xmath0 97 pc ; mamajek et al .
both of these star - forming regions appear to be isolated in that they do not appear to be associated with any molecular gas .
in addition , both are comprised mainly of `` weak - line '' tts equivalent widths , w(h@xmath11 ) @xmath4 @xmath510 and `` classical '' tts ( ctts ) to be tts with w(h@xmath11 ) @xmath12 @xmath510 where the negative sign denotes emission ] .
in contrast , most of the tts in mbm12 are ctts which are still associated with their parent molecular cloud .
in addition to the above isolated star - forming regions , tts have been found outside of the central cloud core regions in many nearby star - forming cloud complexes ( see references in feigelson 1996 ) .
several theories exist to explain how tts can separate from their parent molecular clouds either by dynamical interactions ( sterzik & durisen 1995 ) or by high - velocity cloud impacts ( @xcite ) .
feigelson ( 1996 ) also suggests that some of these tts may form in small turbulent cloudlets that dissipate after forming a few tts .
since the tts in mbm12 appear to still be in the cloud in which they formed , we know they have not been ejected from some other more distant star - forming region .
therefore mbm12 may be an example of one of the cloudlets proposed by feigelson ( 1996 ) .
moriarity - schieven et al .
( 1997 ) argue that mbm12 has recently been compressed by a shock associated with its interface with the local bubble .
this shock may also have recently triggered the star - formation currently observed in mbm12 ( cf .
elmegreen 1993 ) .
alternatively ballesteros - paredes et al . ( 1999 ) suggest that mbm12 may be an example of a star - forming molecular cloud that formed via large scale streams in the interstellar medium .
mbm12 is different from most other high - latitude clouds at @xmath13 @xmath4 30@xmath14 in terms of its higher extinction and its star formation capability ( e.g. , hearty et al .
based on co observations and star counts , the peak extinction in the cloud is @xmath3 @xmath0 5 mag ( duerr & craine 1982a ; magnani et al . 1985 ; pound et al .
1990 ; zimmermann & ungerechts 1990 ) . however , molecular clouds are clumpy and it is possible that some small dense cores with @xmath3 @xmath4 5 mag were not resolved in previous molecular line and extinction surveys .
for example , zuckerman et al .
( 1992 ) estimate @xmath3 @xmath0 11.5 mag through the cloud , along the line of sight to the eclipsing cataclysmic variable h0253 + 193 located behind the cloud and we estimate @xmath3 @xmath0 8.48.9 along the line of sight to a g9 star located on the far side of the cloud ( sect .
[ cafos ] ) although there is evidence for gravitationally bound cores in mbm12 , the entire cloud does not seem to be bound by gravity or pressure ( pound et al .
1990 ; zimmermann & ungerechts 1990 ) .
therefore , the cloud is likely a short - lived , transient , object similar to other high latitude clouds which have estimated lifetimes of a few million years based on the sound crossing time of the clouds ( @xcite ) .
if this is the case , mbm12 will dissipate in a few million years and leave behind an association similar to the tw hydrae association that does not appear to be associated with any molecular material .
previous searches for tts in mbm12 have made use of h@xmath11 , infrared , and x - ray observations .
the previously known tts in mbm12 are listed in table [ previous ] with their coordinates , spectral types , apparent magnitudes , and selected references .
we include the star s18 in the list even though downes & keyes ( 1988 ) point out that it could be an me star rather than a t tauri star since our observations confirm that it is a ctts .
the previously known and new tts stars identified in this study are plotted in fig . [ iras ] with an iras 100 @xmath15 m contour that shows the extent of the cloud .
.previously known t tauri stars in mbm12 [ cols="^,^,^,^,^,^ " , ] @xmath16 herbig ( 1977 ) measured a radial velocity of @xmath17 km s@xmath18 for this object .
we obtained high resolution spectra of two of the t tauri stars in mbm12 with foces at the calar alto 2.2-m telescope in august 1998 .
the spectra for these stars ( rxj0255.4 + 2005 and lkh@xmath11264 , see fig . 3 . ) allow us to estimate their radial velocities and confirm the w(li ) measurements of our low resolution spectra presented in sect .
[ cafos ] .
determinations of radial velocity , rv , and projected rotational velocity , vsin@xmath19 , have been obtained by means of cross correlation analysis of the stellar spectra with those of radial velocity and rotational standard stars , treated in analogous way . given the large spectral range covered by the foces spectra , the cross correlation of the target and template stars was performed after rebinning the spectra to a logarithmic wavelength scale , in order to eliminate the dependence of doppler shift on the wavelength . moreover , only parts of the spectra free of emission lines and/or not affected by telluric absorption lines have been used .
therefore , the nai d , and h@xmath11 lines as well as wavelengths longer than about 7000 have been excluded from the cross - correlation analysis .
the result of the cross - correlation is a correlation peak which can be fitted with a gaussian curve .
the parameters of the gaussian , center position and full - width at half - maximum ( fwhm ) are directly related to rv and vsin@xmath19 , respectively .
the method of the correlation has been fully described by queloz ( 1994 ) , and soderblom et al .
more details about the calibration procedure can be found in appendix a of covino et al .
( 1997 ) .
the radial velocities we measured for the two mbm12 tts listed in table [ velocity ] are similar to that of the molecular gas ( zimmermann & ungerechts 1990 ; pound et al .
radial velocity measurements have not yet been made for the fainter stars .
nevertheless , the superposition of the tts on the cloud and the similar radial velocities of at least two of the stars with the gas are strong evidence to support that the tts are associated with the cloud .
since both ctts and wtts are typically @xmath0 10@xmath2010@xmath21 times more luminous in the x - ray region of the spectrum than average ( i.e. , older ) low - mass stars ( damiani et al .
1995 ) , we made use of the _ rosat _ pointed and the rass observations of mbm12 to identify previously unknown tts in the cloud .
the 25 ks _ rosat _ pspc pointed observation ( sequence number 900138 ) was centered at ( ra , dec ) @xmath0 ( 2:57:04.8,+19:50:24 ) .
although they were originally discovered by other means , all of the previously known tts in the central region of mbm12 were also detected with _
rosat_. since the extent of the molecular gas is not known ( in particular for the mbm13 region ) and tts can sometimes be displaced several parsecs from their parent clouds , we also searched in the rass database in a @xmath0 25 deg@xmath22 region around mbm12 .
details about _ rosat _ and its pspc detector can be found in trmper ( 1983 ) and briel & pfeffermann ( 1995 ) , respectively . the rass broad - band image of the region investigated around mbm12 and the _ rosat _ pointed observation
are displayed in fig .
[ xrayfig ] .
the x - ray source search was conducted in different _ rosat _ standard bands " , defined as follows : broad " = 0.082.0 kev ; soft " = 0.080.4 kev ; hard " = 0.52.0 kev ; hard1 " = 0.50.9 kev ; hard2 " = 0.92.0 kev .
we identified all of the x - ray sources above a maximum likelihood exp(@xmath5ml ) .
] threshold of 7.4 in both the rass and the _ rosat _ pspc pointed observations of mbm12 .
in addition , we selected only those x - ray sources above a count rate threshold of @xmath0 0.03 cts s@xmath18 in the rass observation and a count rate threshold of 0.0013 cts s@xmath18 in the _ rosat _ pspc pointed observation .
the one previously known tts candidate , s18 , near the cloud mbm13 was detected in the rass with a ml = 6.4 ( i.e. , below our threshold ) , however , since our optical spectroscopic observations confirm that it is a tts we include it in our study .
we identified 49 x - ray sources in the _ rosat _ pspc pointed observation of mbm12 ( including all of the previously known tts in the central region of the cloud ) and 28 x - ray sources detected in the rass ( including s18 ) in the regions displayed in fig .
[ xrayfig ] .
three stars were detected both on the rass and in the pointed observation .
we list the sources detected in the _ rosat _ pspc pointed observation and in the rass in table 4 .
we include the _ rosat _ source name , the x - ray source coordinates , the maximum likelihood for existence for each source , the broad - band count rates , the x - ray hardness ratios @xmath23 and @xmath24 ( as defined in neuhuser et al .
1995 ) , the apparent visual magnitude taken from the guide star catalog ( magnitudes for the fainter sources indicated with a `` : '' are estimated from the digitized sky survey images ) , and the broad - band x - ray to optical flux ratio .
we also list the spectral type and the h@xmath11 and lithium equivalent widths of the sources that have been observed spectroscopically and comments collected from our search through the simbad and ned databases concerning the objects . assuming a mean x - ray count - rate - to - flux conversion factor of 1.1 @xmath25 10@xmath26 erg cts@xmath18 @xmath27 , which we derive from x - ray spectral fits of the tts in sect .
[ xlf ] , if the cloud is at a distance of 65 pc , the limiting luminosities of the observations are @xmath28 erg s@xmath18 and @xmath29 erg s@xmath18 for the rass and _ rosat _ pointed observations , respectively .
therefore , these observations are sufficient to detect most of the wtts in the cloud since the threshold is below the x - ray faintest stars in the wtts x - ray luminosity function ( e.g. , neuhuser et al .
1995 ) . although the rass observation of mbm12 in not sensitive enough to detect all of the ctts in the cloud , the objective prism survey by stephenson ( 1986 ) identified all of the h@xmath11 emission sources in this region down to a visual magnitude threshold of @xmath0 13.5 .
since this limiting magnitude corresponds to the early m spectral types in mbm12 , the current population of tts in mbm12 presented in this paper should be complete for all earlier spectral types .
since mbm12 is at relatively high galactic latitude , many of the 81 x - ray sources we identified are extragalactic .
therefore , we used the x - ray to optical flux ratios ( see table 4 ) to remove extragalactic sources from our list of candidates ( cf .
hearty et al . 1999 ) .
all sources which have log(@xmath30 ) @xmath4 0.0 are considered to be extragalactic and those with log(@xmath30 ) @xmath12 0.0 are considered to be stellar objects , some of which could be pms .
we also searched the literature to remove cataloged non - pms stars from our list of candidates .
finally we were left with a list of x - ray sources identified in the rass and _ rosat _ pointed observations of the cloud which have stellar optical counterparts that may be pms stars . however
, many of these stars may be other types of x - ray active stars ( e.g. , rs cvn and dme stars ) and nearby main sequence stars ( which may not be intrinsically x - ray bright , but are near enough so that their x - ray flux is large ) that are more difficult to separate from pms stars by x - ray observations alone .
therefore , follow - up spectral observations are necessary to identify which x - ray sources are t tauri stars .
in order to complete the census of the tts population of mbm12 we require follow - up observations . since lithium is burned quickly in convective stars , a measurement of w(li ) along with a knowledge of the spectral type of a star can be a reliable indicator of youth .
therefore we obtained broad - band , low - resolution , optical spectra of the x - ray emitting tts candidates to determine spectral types and measure the equivalent width of the h@xmath11 emission and 6708 absorption lines .
the spectra were obtained from october 911 , 1998 with the calar alto faint object spectrograph ( cafos ) at the 2.2-m telescope at calar alto , spain .
the 24@xmath15 m pixels of the site-1d 2048@xmath252048 chip with the g-100 grism provided a reciprocal dispersion of @xmath0 2.1 pixel@xmath18 . the resolving power , @xmath31 = @xmath32 @xmath0 1000 , derived from the measurement of the fwhm ( fwhm @xmath0 6.4 ) of several well isolated emission lines of the comparison spectra is sufficient to resolve the lithium absorption line in t tauri stars . the wavelength range @xmath0 4900 to 7800 was chosen to detect two indicators of possible youth ( h@xmath11 emission and @xmath336708 absorption ) and to determine spectral types .
all spectra were given an initial inspection at the telescope .
if a particular star showed signs of youth or the integration produced fewer than @xmath0 1000 cts pixel@xmath18 , at least one additional integration was performed .
the results of the spectroscopic observations of the tts in mbm12 are summarized in table [ eqw ] .
we list the name of the star ; the coordinates for the optical source ; the spectral type ; the log of the effective temperature , log@xmath34 , assuming luminosity class v and using the spectral type - effective temperature relation of @xcite ; apparent magnitude , @xmath35 , taken from the guide star catalog ; the equivalent width of h@xmath11 , w(h@xmath11 ) ; both the low and high resolution ( when available ) measurements of w(li ) ; the veiling corrected w(li ) ( cf .
strom et al . 1989 ) ; and the derived lithium abundance based on the non - lte curves of growth of pavlenko & magazz ( 1996 ) assuming log@xmath36=4.5 .
the estimated error for the low - resolution w(li ) measurements is @xmath0 @xmath37 90 m based on the correlation with the three stars for which we have high resolution measurements . @c@c@c@c@c@c@c@c@c@c@c@ star & ra & dec & spt & log@xmath34 & @xmath35 & w(h@xmath11)@xmath38 & w(li ) &
w(li ) hi - res & w(li ) deveiled & logn(li ) + & [ 2000 ] & [ 2000 ] & & & [ mag ] & [ ] & [ m ] & [ m ] & [ m ] & + + hd 17332 & 02:47:27.3 & + 19:22:24 & g1v & 3.769 & 6.8 & 2.96 & 190 & & & 3.2 + rxj0255.3 + 1915 & 02:55:16.5 & + 19:15:02 & f9 & 3.785 & 10.4 & 3.75 & 170 & & & 3.3 + + rxj0255.4 + 2005 & 02:55:25.7 & + 20:04:53 & k6 & 3.631 & 12.2 & @xmath51.26 & 380 & @xmath39 & 462 & 2.7 + lkh@xmath11262 & 02:56:07.9 & + 20:03:25 & m0 & 3.584 & 14.6 & @xmath532.1 & 290 & & 412 & 2.0 + lkh@xmath11263 & 02:56:08.4 & + 20:03:39 & m4 & 3.517 & 14.6 & @xmath532.9 & 380 & & 543 & 1.6 + lkh@xmath11264 & 02:56:37.5 & + 20:05:38 & k5 & 3.644 & 12.5 & @xmath558.9 & 490 & @xmath40 & 836 & 3.8 + e02553 + 2018 & 02:58:11.2 & + 20:30:04 & k4 & 3.657 & 12.3 & @xmath51.6 & 620 & @xmath41 & 499 & 3.1 + rxj0258.3 + 1947 & 02:58:15.9 & + 19:47:17 & m5 & 3.501 & 15.0 : & @xmath524.5 & 580 & & 783 & 1.8 + s18 & 03:02:21.1 & + 17:10:35 & m3 & 3.532 & 13.5 & @xmath579.0 & 310 & & 552 & 1.8 + rxj0306.5 + 1921 & 03:06:33.1 & + 19:21:52 & k1 & 3.698 & 11.4 & filled & 350 & & & 3.1 + @xmath42 a negative sign denotes emission .
@xmath43 the high resolution measurement of w(li ) for this star is taken from martn et al .
( 1994 ) .
the optical spectra of the tts in mbm12 are displayed in fig .
[ ttsspectra ] .
the stars which show strong h@xmath11 emission are also scaled by an appropriate factor to display the emission line .
the spectra of the two stars we classify as young main sequence stars which still show lithium are displayed in fig .
[ zams ] .
in addition to confirming that the star s18 is a ctts with strong h@xmath11 emission and absorption , we identified 3 previously unknown tts in mbm12 . in order to estimate the relative age of the mbm12 stars with lithium
we plot them in an w(li ) vs. t@xmath44 diagram ( fig .
[ lithtemp ] ) along with stars from taurus ( age @xmath0 a few myr ) , the tw hydrae association ( age @xmath0 10 myr ) , the @xmath10 chamaeleontis cluster ( age @xmath0 218 myr ) , ic 2602 ( age @xmath0 30 myr ) , and the pleiades ( age @xmath0 100 myr ) .
in addition , we plot isoabundance lines for the non - lte curves of growth of pavlenko & magazz ( 1996 ) for log@xmath36=4.5 stars and the isochrones for the non - rotating lithium depletion model of pinsonneault et al .
the positions of the mbm12 stars in the diagram indicates they are young objects with ages much less than that of the pleiades or ic 2602 .
although the relative ages between the stars in mbm12 , the tw hydrae association , and the @xmath10 chamaeleontis cluster , can not be discerned in fig .
[ lithtemp ] , since most of the tts in mbm12 are ctts which are still associated with their parent molecular cloud the tts in mbm12 must be younger than those in the tw hydrae association or the @xmath10 chamaeleontis cluster which are comprised mainly of wtts not associated with any molecular cloud ( i.e. , the tts in mbm12 have ages @xmath12 10 myr ) .
although the two f and g spectral type stars in which we detected lithium ( hd 17332 and rxj0255.3 + 1915 ) are located above the pleiades in the w(li ) vs. t@xmath44 diagram , since they both show h@xmath11 absorption stronger than any similar spectral type stars in ic 2602 ( e.g. , randich et al .
1997 ) , they are probably older than 30 myr .
thus , we list them as main sequence stars in table [ eqw ] .
covino et al .
( 1997 ) have shown that low - resolution spectra tend to overestimate w(li ) in intermediate spectral types , therefore we probably over estimated the w(li ) for these two stars .
the tts in mbm12 are clearly lithium - rich relative to the stars in the pleiades .
however , current age dependent stellar population models predict that there should be a population of young stars with ages @xmath12 150 myr distributed across the sky .
therefore we compare the density of young x - ray sources detected in mbm12 with the age dependent stellar population model of guillout et al .
( 1996 ) to find out if we are really seeing an excess of young x - ray sources in the direction of mbm12 . in the galactic latitude range of 40@xmath14 @xmath4 @xmath45b@xmath45 @xmath4 30@xmath14 guillout et al .
( 1996 ) predict there should be 0.61.0 stars deg@xmath46 and 0.140.20 stars deg@xmath46 above an x - ray count rate threshold of 0.0013 cts s@xmath18 and 0.03 cts s@xmath18 , respectively , which have ages @xmath12 150 myr .
therefore , we expect to detect @xmath0 1.93.1 young stars in this age group in the _ rosat _ pointed observation and 3.55 stars in this age group in the rass observation . since we observed several young stars which probably have ages @xmath12 150 myr but
are not associated with mbm12 ( i.e. , the two intermediate spectral type stars which have not yet depleted their lithium and the 3 me and 3 ke stars listed in table 4 which have depleted their lithium but show h@xmath11 emission ) , there are a sufficient number of x - ray active stars in this region to account for the numbers predicted by guillout et al .
therefore , the tts we observe represent an excess of x - ray active young stars associated with mbm12 .
in addition to the x - ray selected t tauri star candidates , we also observed the reddest star from a list of stars compiled by duerr & craine ( 1982b ) which are along the line of sight to mbm12 and have v - i colors redder than 2.5 mag .
the optical spectrum of this star , which we will call dc48 , indicates it is a g9 star .
since duerr & craine ( 1982b ) measured @xmath35 = 18.7 and v - i = 5.6 mag , it corresponds to a main sequence star with @xmath3 @xmath0 8.9 mag at a distance of @xmath0 63 pc or a giant star with @xmath3 @xmath0 8.4 mag at a distance of @xmath0 950 pc .
the spectrum of the highly reddened ( @xmath3 @xmath0 8.48.9 ) g9 star , dc48 , is displayed in fig .
we tested all of the tts for x - ray variability using the methods described in @xcite .
the only t tauri star which showed x - ray variability is the newly identified star rxj0255.5 + 2005 that was detected both in the rass and in the _ rosat _ pointed observation and flared during the pointed observation ( see the light curve displayed in fig .
[ lc ] ) . the peak x - ray count rate during the flare increased by more than a factor of 6 from the pre - flare count rate .
although we do no have a sufficient number of counts ( @xmath0 1000 counts for the non - flare phase and @xmath0 500 counts for the flare phase ) for a detailed study of the evolution of the coronal temperature during the flare , we performed a rough spectral fit using a 2 temperature raymond - smith model ( raymond & smith 1977 ) including a photoelectric absorption term using the morrison & mccammon ( 1983 ) cross sections .
we fit the data for 3 time intervals : the pre - flare phase , the flare , and the post - flare phase fig .
[ xspecflare ] .
both temperature components increased during the flare and remained high throughout the post - flare phase .
the parameters derived from the x - ray spectral fits are listed in table [ xraylum ] ( see sect . [ xlf ] for a description of the table columns ) .
since the two temperature components are not well constrained by the spectral fit during the flare , these estimates should be viewed as a lower limits .
the results of the spectral fits are consistent with the type of coronal heating seen in high signal - to - noise x - ray spectra of other flaring wtts ( e.g. , tsuboi et al .
although our x - ray spectra do not have sufficiently high signal - to - noise for a detailed comparison of x - ray spectral models , we performed a spectral fit using a 2 temperature raymond - smith model including a photoelectric absorption term as described in sect .
[ xrayvar ] for the sources with at least 100 counts . for the sources with fewer than 100 counts we calculate the x - ray flux using an x - ray count rate to flux conversion factor of 1.1 @xmath25 10@xmath26 erg cts@xmath18 @xmath27 which is the mean conversion factor derived from the spectra for which we performed spectral fits .
we list the total _ rosat _ broad band ( 0.082.0 kev ) counts and count rates for the tts in mbm12 and the derived interstellar+circumstellar absorption cross sections and plasma temperatures for the spectra in which we performed spectral fits in table [ xraylum ] .
the x - ray luminosities assume a distance of 65 pc .
since the binary lkh@xmath11262/263 was not spatially resolved with the pspc we fit the combined x - ray spectrum to estimate the combined x - ray luminosity but we divide that value in half to generate the x - ray luminosity function .
@l@c@c@c@c@c@c@c@c@c@c@ star & & & + & counts & rate & counts & rate & n@xmath47/10@xmath48 & @xmath49 & @xmath50 & @xmath51/dof & @xmath52/10@xmath53 & log@xmath54 + & & [ cts s@xmath18 ] & & [ cts s@xmath18 ] & [ @xmath27 ] & [ kev ] & [ kev ] & & [ erg s@xmath18 @xmath27 ] & [ erg s@xmath18 ] + rxj0255.4 + 2005 + `` '' total & @xmath57 & @xmath58 & @xmath59 & @xmath60 & 0.86 & 0.89 & 0.08 & 39.7/34 & 7.81 & 29.60 + `` '' pre - flare & & & @xmath61 & @xmath62 & 0.49 & 0.88 & 0.11 & 9.2/9 & 4.12 & 29.32 + `` '' flare & & & @xmath63 & @xmath64 & 0.18 & 1.69 & 0.30 & 9.5/6 & 20.1 & 30.01 + `` '' post - flare & & & @xmath65 & @xmath66 & 0.27 & 1.14 & 0.24 & 10.7/15 & 6.96 & 29.55 + lkh@xmath11262/263 & & & @xmath67 & @xmath68 & 2.65 & 0.97 & 0.16 & 6.29/8 & 2.01 & 29.01 + lkh@xmath11264 & & & @xmath69 & @xmath70 & & & & & 0.47 & 28.39 + e02553 + 2018 & @xmath71 & @xmath72 & @xmath73 & @xmath74 & 8.21 & 1.05 & 0.10 & 16.5/16 & 5.03 & 29.41 + rxj0258.3 + 1947 & & & @xmath75 & @xmath76 & 8.42 & 1.14 & 0.13 & 2.63/2 & 0.86 & 28.64 + s18 & @xmath77 & @xmath78 & & & & & & & 3.30 & 29.22 + rxj0306.5 + 1921 & @xmath79 & @xmath78 & & & & & & & 3.30 & 29.22 + in order to compare the derived x - ray luminosity function for the tts in mbm12 with other flux limited x - ray luminosity functions we used the asurv rev . 1.2 package ( isobe & feigelson 1990 ; lavalley et al .
1992 ) , which implements the methods presented in feigelson & nelson ( 1985 ) .
although the currently known tts in mbm12 are all x - ray detected , the luminosity functions of other , more distant , star forming regions include upper limits . the derived x - ray luminosity function is displayed in fig .
[ lumfunc ] with the x - ray luminosity function for the tts in the l1495e cloud in taurus which ( like mbm12 ) was observed in a deep ( 33 ks ) _ rosat _ pspc pointed observation ( strom & strom 1994 ) .
the _ rosat _ pointed observation of l1495e is @xmath0 20 times more sensitive than previous observations with the _ einstein _ satellite .
strom & strom ( 1994 ) used this observation to show that the x - ray luminosity of tts extends to fainter luminosities than were observed with einstein .
we have re - reduced the pointed observation of l1495e in a way analogous to that of mbm12 .
the x - ray luminosity function we derive for l1495e ( 1 ) includes only the k and m spectral type tts , ( 2 ) includes 6 upper limits , ( 3 ) assumes an x - ray to optical flux conversion factor of 1.1 @xmath25 10@xmath26 erg cts@xmath18 @xmath27 , and ( 4 ) assumes a distance of 140 pc . the x - ray luminosity functions in mbm12 and l1495e agree well : in mbm12 the log@xmath80 = [email protected] erg s@xmath18and log@xmath81 = 28.7 erg s@xmath18 ; in l1495e log@xmath80 = [email protected] erg s@xmath18 and log@xmath81 = 28.9 erg s@xmath18 .
however , we note that the mbm12 x - ray luminosity function has a lower high - luminosity limit and a higher low - luminosity limit than the l1495e x - ray luminosity function . therefore , although the pointed observation of mbm12 is more sensitive than the pointed observation l1495e ( because mbm12 is much closer ) our follow - up observations of the tts in mbm12 may be incomplete for sources fainter than @xmath35 @xmath0 15.5 mag .
in addition , since we know that one of our x - ray sources , s18 , is detected but below our threshold for follow - up observations , there may be other , fainter , x - ray emitting tts in mbm12 with spectral types later than @xmath0 m2 ( i.e. , the spectral type of s18 ) that will be discovered in more sensitive follow - up observations .
the discrepancy at the high luminosity end of the x - ray luminosity function may also be explained if the distance to the tts in mbm12 is larger than 65 pc .
although an increased distance is allowed by the recent _ hipparcos _ results it should be confirmed with further observations .
although mbm12 is not a prolific star - forming cloud when compared to nearby giant molecular clouds it is the nearest star - forming cloud to the sun and offers a unique opportunity to study the star - formation process within a molecular cloud at high sensitivity .
we have presented follow - up observations of x - ray stars identified in the region of the mbm12 complex .
these observations have doubled the number of confirmed tts in this region . since the _ rosat _ pspc pointed observation of the central region of the cloud was sensitive enough to detect all of the previously known tts in the cloud , we believe the list of 5 ctts and 3 wtts presented in table [ eqw ] to be a nearly complete census of the tts in mbm12 for spectral types earlier than @xmath0 m2 . assuming a mean mass of @xmath0 0.6 m@xmath6 for the 8 currently known tts in mbm12 and a cloud mass of 30200 m@xmath6 ( pound et al .
1990 ; zimmermann & ungerechts 1990 ) the star - formation efficiency of mbm12 is @xmath0 224% . since the currently known tts population in mbm12 is incomplete only for the lower mass objects , unless there are a huge number of these objects yet to be discovered in the cloud , this estimate of the star - formation efficiency will not change significantly .
although there is still a large uncertainty in the mass of the cloud the estimated star - formation efficiencies are consistent with that expected from clouds with masses on the order of 100 m@xmath6 ( elmegreen & efremov 1997 ) . by comparing the strengths of the h@xmath11 emission and @xmath336708 absorption lines of the tts in mbm12 with those found in other young clusters , we place an upper limit on the age of the stars in mbm12 @xmath0 10 myr . by comparing the x - ray luminosity function of the tts in mbm12 with that of the tts in l1495e
we predict that there are more young , low - mass , stars to be discovered in mbm12 and the assumed distance to the cloud may have to be increased .
although this prediction agrees with the recently revised distance estimate to the cloud ( @xmath0 @xmath2 pc ) based on results of the _ hipparcos _ satellite , it should be confirmed with future observations .
we have also identified a reddened g9 star behind the cloud with @xmath3 @xmath0 8.48.9 mag .
therefore , there are at least two lines of sight through the cloud that show larger extinctions ( @xmath3 @xmath4 5 mag ) than previously thought for this cloud .
this higher extinction explains why mbm12 is capable of star - formation while most other high - latitude cloud are not .
we wish to thank patrick guillout for helpful discussions about the expected population of young x - ray active stars located at high galactic latitude and loris magnani for insightful comments concerning this paper .
we also thank an anonymous referee for suggestions which enable us to put firmer constraints on the age of the tts in mbm12 .
project is supported by the max - planck - gesellschaft and germany s federal government ( bmbf / dlr ) .
th is grateful for a stipendium from the max - planck - gesellschaft for support of this research .
rn acknowledges a grant from the deutsche forschungsgemeinschaft ( dfg schwerpunktprogramm `` physics of star formation '' ) | we present the _ rosat _ pspc pointed and _ rosat _ all - sky survey ( rass ) observations and the results of our low and high spectral resolution optical follow - up observations of the t tauri stars ( tts ) and x - ray selected t tauri star candidates in the region of the high galactic latitude dark cloud mbm12 ( l1453-l1454 , l1457 , l1458 ) .
previous observations have revealed 3 `` classical '' t tauri stars and 1 `` weak - line '' t tauri star along the line of sight to the cloud . because of the proximity of the cloud to the sun ,
all of the previously known tts along this line of sight were detected in the 25 ks _ rosat _ pspc pointed observation of the cloud .
we conducted follow - up optical spectroscopy at the 2.2-meter telescope at calar alto to look for signatures of youth in additional x - ray selected t tauri star candidates .
these observations allowed us to confirm the existence of 4 additional tts associated with the cloud and at least 2 young main sequence stars that are not associated with the cloud and place an upper limit on the age of the tts in mbm12 @xmath0 10 myr .
the distance to mbm12 has been revised from the previous estimate of @xmath1 pc to @xmath2 pc based on results of the _ hipparcos _ satellite . at this distance mbm12
is the nearest known molecular cloud to the sun with recent star formation .
we estimate a star - formation efficiency for the cloud of 224% . we have also identified a reddened g9 star behind the cloud with @xmath3 @xmath0 8.48.9 mag
. therefore , there are at least two lines of sight through the cloud that show larger extinctions ( @xmath3 @xmath4 5 mag ) than previously thought for this cloud .
this higher extinction explains why mbm12 is capable of star - formation while most other high - latitude clouds are not . |
estimation of covariance matrices is a crucial component of many signal processing algorithms [ 1 - 4 ] . in many applications ,
there is a limited number of snapshots and the sample covariance matrix can not yield the desired estimation accuracy .
this covariance matrix estimation error significantly degrades the performance of such algorithms . in some applications ,
the true covariance matrix has a specific structure .
for example , the array covariance matrix of a linear array with equally spaced antenna elements is a toeplitz matrix when the sources are uncorrelated [ 5 , 6 ] .
moreover , in some applications [ 4 , 8 ] , the structure of the problem suggests that the underlying true covariance matrix is the kronecker product of two valid covariance matrices [ 4 , 7 ] .
this side information can be leveraged in covariance matrix estimation to improve the estimation quality .
for instance , in [ 5 ] a weighted least square estimator for covariance matrices with toeplitz structures was proposed and it was shown that the resulting covariance matrix can enhance the performance of angle estimation algorithms , such as multiple signals classification ( music ) [ 13 ] . in [ 8 ] , covariance matrices with kronecker structure are investigated and a maximum likelihood based algorithm is introduced .
in addition , the structure of covariance matrices has been exploited in various doa estimation algorithms , such as the linear structure in [ 9 ] , and the diagonal structure for the covariance matrix of uncorrelated signals in [ 10 ] .
recently some research works have focused on the application of sparse signal processing in doa estimation based on the sparse representation of the array covariance matrix .
for example , [ 11 ] proposes the idea that the eigenvectors of the array covariance matrix have a sparse representation over a dictionary constructed from the steering vectors . in
[ 12 , 14 ] , it is shown that when the received signals are uncorrelated , the array covariance matrix has a sparse representation over a dictionary constructed using the atoms , i.e. the correlation vectors .
a similar idea is proposed in [ 15 ] , with the difference that the proposed method does not require choosing a hyper - parameter . in this paper , we focus on the estimation of array covariance matrices with linear structure .
first , we show that when the sources are uncorrelated , the array covariance matrix has a linear structure implying that all possible array covariance matrices can be described by a specific subspace . based on this idea ,
a subspace - based covariance matrix estimator is proposed as a solution to a semi - definite convex optimization problem .
furthermore , we propose a nearly optimal closed - form solution for the proposed covariance matrix estimator .
our results show that the proposed method can noticeably improve the covariance matrix estimation quality .
moreover , the closed - form solution is shown to closely approach the optimal performance .
the system model under consideration is a narrowband array system with @xmath0 antennas .
all the signals are assumed to be narrowband with the same center frequency and impinge on the array from the far field .
the baseband array output can be expressed as @xmath1 where @xmath2 is the @xmath3 array output vector , @xmath4 is the number of the received signals , @xmath5 is the @xmath6 signal , @xmath7 is the elevation and azimuth arrival angle of the @xmath6 signal , @xmath8 is the baseband array response to @xmath6 signal and * n*(t ) is the noise vector . the baseband array response , @xmath8 , is called the `` steering vector '' [ 13 ] . if the received signals are uncorrelated , the covariance matrix can be written as @xmath9 where @xmath10 represents the power of the @xmath6 signal , @xmath11 is the noise variance and @xmath12 is the identity matrix .
we define the `` correlation vector '' which belongs to direction @xmath13 as follows @xmath14 where @xmath15 is a linear transformation that converts its matrix argument to a vector by stacking the columns of the matrix on top of one another .
consequently , the covariance matrix can be rewritten as @xmath16 therefore , @xmath17 is a linear combination of the correlation vectors of the received signals . according to ( 4 )
, @xmath17 lies in the subspace of the correlation vectors .
hence , if we build the subspace spanned by all possible correlation vectors@xmath18 , then @xmath17 completely lies in this subspace . for many array structures ,
the matrix @xmath19 inherits some symmetry properties .
accordingly , the correlation vectors can not span an @xmath20 dimensional space .
for example , when the incoming signals are uncorrelated , the covariance matrix of a uniform linear array is a toeplitz matrix [ 5 ] .
it is easy to show that all the @xmath21 toeplitz matrices can be described by a @xmath22 dimensional space .
the subspace of the correlation vectors @xmath23 can be obtained by constructing a positive definite matrix @xmath24 where ( 5 ) is an element - wise integral .
based on ( 5 ) , the subspace dimension of the correlation vectors @xmath25 is equal to the number of non - zero eigenvalues of the matrix @xmath26 .
consequently , the subspace of the correlation vectors can be constructed using the eigenvectors which correspond to the non - zero eigenvalues .
1 shows the eigenvalues of @xmath26 for a square planar array with 16 elements ( the horizontal and vertical space between the elements is half a wavelength ) .
one can observe that the number of non - zero eigenvalues is equal to 49 .
therefore , for this array , the subspace of the correlation vectors can be constructed from the 49 eigenvectors corresponding to the non - zero eigenvalues .
note that for a 16-element linear array , we observe 31 non - zeros eigenvalues because the covariance matrix is a toeplitz matrix[5 ] .
for some array structures such as circular array , we may not observe zero eigenvalues but our investigation has shown that the subspace of the correlation vectors can be effectively approximated using the dominant eigenvectors ( the eigenvectors corresponding to the dominant eigenvalues ) . therefore , if we construct the matrix @xmath27 whose columns form a basis for the correlation vectors subspace
, we can rewrite the covariance matrix as @xmath28 hence , we can choose the columns of @xmath27 as the eigenvectors corresponding to the non - zero eigenvalues ( or the dominant eigenvectors ) . by imposing the linear structure constraint ( 6 ) to the covariance matrix estimation problem , we can significantly improve the estimation quality . some works
have studied covariance matrices with linear structures .
for example , a weighted least - square estimator was proposed in [ 5 ] based on the linear structure for toeplitz covariance matrices .
however , the toeplitz structure is restricted to linear arrays and the resulting matrix is not guaranteed to be positive definite .
based on ( 4 ) and ( 6 ) , the estimated covariance matrix should lie in the subspace spanned by the columns of @xmath27 .
we are going to estimate @xmath29 which is defined as @xmath30 based on the previous discussion , we propose the following optimization problem @xmath31 where @xmath32 is the sample covariance matrix and @xmath33 is the number of time samples . the matrix @xmath34 is the projection matrix on the subspace that is orthogonal to the subspace of the correlations vectors . as such , the first constraint in
( 8) ensures that the resulting matrix lies in the correlations vectors subspace .
the second constraint guarantees that the resulting matrix is positive definite . note that , ( 8) is a convex optimization problem and can be solved using standard tools from convex optimization .
the proposed method imposes the linear structure using the subspace constraint .
if the covariance matrix is toeplitz , the subspace constraint enforces the resulting matrix to be toeplitz .
however , the proposed algorithm is not limited to toeplitz structures and can be used for any linear structure . the sample covariance matrix ( 9 )
can be expressed as @xmath35 the second term on the right hand side of ( 10 ) , @xmath36 , is the unwanted part ( estimation error ) which tends to zero if we have an infinite number of snapshots .
the estimation error has some random behavior and can lie anywhere in the entire space . since the first constraint in ( 8) enforces the estimated matrix to lie in the correlation vectors subspace , it is expected to eliminate the component of estimation error which is not in this subspace .
the dimension of the correlation vectors subspace is typically smaller than the entire space dimension @xmath37 .
for example , for a 30-element uniform linear array , the dimension of the correlation vectors subspace is equal to 59 ; while the entire space dimension is 900 .
thus , it is conceivable that the proposed method could yield a much better estimation performance in comparison to the sample covariance matrix ( 9 ) . the proposed optimization problem ( 8) is an @xmath37 dimensional optimization problem .
therefore , it may be hard to solve for large arrays . in this section
, we derive a closed form near optimal solution which makes our method easy for practical implementation . according to ( 4 ) , the covariance matrix should be in the correlation vectors subspace .
we define @xmath38 and @xmath39 as follows @xmath40 thus , @xmath41 is orthogonal to the correlation vectors subspace and @xmath42 contains the desired part
. therefore , we rewrite ( 8) as @xmath43 in the proposed estimator ( 8) , we placed the first constraint to suppress the estimation error which does not lie in the correlation vectors subspace . in ( 11 ) , we project the sample covariance matrix to the correlation vectors subspace .
thus , we have eliminated the estimation error which does not lie in the correlation vectors subspace .
accordingly , we simplify ( 13 ) as follows @xmath44 which has a simple closed form solution @xmath45 where @xmath46 is the number of positive eigenvalues of @xmath42 , @xmath47 are the positive eigenvalues and @xmath48 are their corresponding eigenvectors .
actually , we break the primary optimization problem ( 8) into two optimization problems .
first , we find a matrix in the correlation vectors subspace which is most close to the sample covariance matrix and the resulting matrix is @xmath42 .
in the second step , we find the closest positive semi - definite matrix to @xmath42 and the resulting matrix is given in ( 15 ) .
in this section , we provide some simulation results to illustrate the performance of the proposed approach .
the examples provided include doa estimation and subspace estimation , which underscores the flexibility of the proposed covariance matrix estimation approach for a broad range of applications .
all the curves are based on the average of 500 independent runs .
assume a uniform linear array with @xmath49 omnidirectional sensors spaced half a wavelength apart .
for this array the correlation vectors subspace is a 19 dimensional space since the covariance matrix is toeplitz .
the additive noise is modeled as a complex gaussian zero - mean spatially and temporally white process with identical variances in each array sensor . in this experiment
, we compare the performance of music when used with the sample covariance matrix and with the proposed covariance matrix estimation method .
we also compare its performance with the sparse covariance matrix representation method [ 14 , 12 ] and the sparse iterative covariance - based estimation approach ( spice ) [ 15 ] .
we consider two uncorrelated sources located at @xmath50 and @xmath51 ( @xmath52 is the direction orthogonal to the array line ) and both sources are transmitted with the same power .
2 shows the probability of resolution ( the probability that the algorithm can distinguish these two sources ) versus the number of snapshots for one fixed sensor with @xmath53 db .
it is clear that using the proposed method leads to significant improvement in performance in comparison to using the sample covariance matrix .
spice [ 15 ] is an iterative algorithm , which is based on the sparse representation of the array covariance matrix and requires one matrix inversion in each iteration .
one can see that this algorithm fails when we use 20 iterations , however , performs well with 1000 iterations .
nevertheless , for practical purposes it is generally computationally prohibitive to perform 1000 matrix inversion operations .
in addition , one can observe that the proposed near optimal solution to ( 8) yields a close performance to the optimal solution .
3 displays the probability of resolution against @xmath54 for a fixed training size @xmath55 snapshots .
roughly , the music algorithm based on the proposed method is 7 db better than the music algorithm based on the sample covariance matrix . in summary ,
the proposed method yields notable and promising performance even with a small number of snapshots and at low snr regimes .
furthermore , it is easily implementable using the proposed closed - form solution , which consists of a matrix multiplication and eigen - decomposition .
, scaledwidth=50.0% ] the estimation of the subspace of the received signals is an important task in many signal processing algorithms .
for example , in the eigen - space based beamforming algorithm [ 3 ] , the subspace of the received signals is used to make the beamformer robust against the steering vector mismatch . in the music algorithm ,
the subspace of the received signals is used to obtain the noise subspace [ 13 ] .
the subspace of the received signals is usually estimated using the dominant eigenvectors of the estimated covariance matrix . in this simulation , we consider three uncorrelated sources located at @xmath56 , @xmath52 and @xmath57 and the sources are received with same signal to noise ratio . to investigate the accuracy of the subspace estimation , we define the distance between two subspaces as follows [ 16 ] : + given two matrices @xmath58,@xmath59 , the distance between the subspaces spanned by the columns of @xmath58 and @xmath60 is defined as @xmath61 where @xmath62 and @xmath63 are orthonormal bases of the spaces @xmath64 and @xmath65 , respectively .
similarly , @xmath66 is an orthonormal basis for the subspace which is orthogonal to @xmath64 and @xmath67 is an orthonormal basis for the subspace which is orthogonal to @xmath65 .
in addition , @xmath68 denotes the spectral norm of matrix @xmath69 .
4 displays the distance between the true signals subspace and the estimated one as a function of the number of snapshots for @xmath70 db .
we construct the signal subspace using the first three eigenvectors .
one can observe that the proposed method exhibits a better rate of convergence .
in addition , the performance of the closed - form solution closely approaches the optimal solution .
in this paper , a subspace method for array covariance matrix estimation was proposed .
we have shown that when the received signals are uncorrelated , the covariance matrix lies in the subspace of the correlation vectors .
based on this idea , we posed the estimation problem as a convex optimization problem and enforced a subspace constraint .
in addition , a near optimal closed - form solution for the proposed optimization problem was derived .
a number of numerical examples demonstrated the notable performance of the proposed approach and its applicability to a wide range of signal processing problems , including but not limited to , doa estimation and subspace estimation .
in contrast to some of the existing approaches , which suffer from drastic performance degradation with limited data and at low snr regimes , the proposed method showed very graceful degradation in such settings .
s. a. vorobyov , _ principles of minimum variance robust adaptive beamforming design _ , elsevier signal processing , invited paper , special issue : advances in sensor array processing , vol .
3264 - 3277 , dec .
k. yu , m. bengtsson , b. ottersten , d. mcnamara , and p. karlsson , _ modeling of wide - band mimo radio channels based on nlos indoor measurements _ , ieee transactions on vehicular technology , vol . 53 , no . 8 , pp
. 655665 , may 2004 .
h. li , p. stoica , j. li , _ computationally efficient maximum likelihood estimation of structured covariance matrices _ , ieee transactions on signal processing , vol .
1314 1323 , may 1999 .
m. rahmani , m. h. bastani , _ robust and rapid converging adaptive beamforming via a subspace method for the signal - plus - interferences covariance matrix estimation _ , iet signal processing , vol . 8 , issue 5 , pp .
507 520 , july 2014 .
j. c. de munck , h. m. huizenga , l. j. waldorp , and r. m. heethaar , _ estimating stationary dipoles from meg / eeg data contaminated with spatially and temporally correlated background noise _
, ieee transactions on signal processing , vol .
50 , no . 7 , pp . 15651572 , jul .
k. werner , m. jansson , p. stoica , _ on estimation of covariance matrices with kronecker product structure _ , ieee transactions on signal processing , vol .
478 491 , feb . 2008 .
b. ottersten , p. stoica , and r. roy , _ covariance matching estimation techniques for array signal processing applications _ , digital signal processing , vol . 8 , pp . 185210 , 1998 . b. gransson , m. jansson and b. ottersten , _ spatial and temporal frequency estimation of uncorrelated signals using subspace fitting _ , in proc .
of 8th ieee signal processing workshop on statistical signal and array processing , corfu , greece , jun .
1996 , pp .
d. malioutov , m. etin , a. willsky , _ a sparse signal reconstruction perspective for source localization with sensor arrays _ ,
ieee trans signal process .
53 , pp . 30103022 , 2005 . l. blanco1 , m. njar , _ sparse covariance fitting for direction of arrival estimation _ , eurasip journal on advances in signal processing , 2012 h.l vantrees , _ optimum array processing .part iv of detection , estimation , and modulation theory _ , wiley,2002 j. s. picard , a. j. weiss , _ direction finding of multiple emitters by spatial sparsity and linear programming _ , 9th international symposium on communications and information technology , pp . 1258 - 1262 , 2009 .
p. stoica , p. babu , j. li , _ spice : a sparse covariance - based estimation method for array processing _ , ieee trans signal process .
629 - 638 , 2011 .
p. jain , p. netrapalli , and s. sanghavi , _ low - rank matrix completion using alternating minimization _ , arxiv preprint arxiv:1212.0467 , 2012 . | this paper introduces a subspace method for the estimation of an array covariance matrix .
it is shown that when the received signals are uncorrelated , the true array covariance matrices lie in a specific subspace whose dimension is typically much smaller than the dimension of the full space . based on this idea ,
a subspace based covariance matrix estimator is proposed .
the estimator is obtained as a solution to a semi - definite convex optimization problem .
while the optimization problem has no closed - form solution , a nearly optimal closed - form solution is proposed making it easy to implement . in comparison to the conventional approaches , the proposed method yields higher estimation accuracy because it eliminates the estimation error which does not lie in the subspace of the true covariance matrices .
the numerical examples indicate that the proposed covariance matrix estimator can significantly improve the estimation quality of the covariance matrix .
shell : bare demo of ieeetran.cls for journals covariance matrix estimation , subspace method , array signal processing , semidefinite optimization . |
a number of studies have reported an anti - correlation between fractional linear polarization and total intensity flux density for extragalactic 1.4 ghz sources ; faint sources were found to be more highly polarized . as a result , the euclidean - normalised differential number - counts of polarized sources have been observed to flatten at linearly polarized flux densities @xmath4 @xmath5 1 mjy to levels greater than those expected from convolving the known total intensity source counts with plausible distributions for fractional polarization @xcite .
the flattening suggests that faint polarized sources may exhibit more highly ordered magnetic fields than bright sources , or may instead suggest the emergence of an unexpected faint population . the anti - correlation trend for fractional linear polarization
is not observed at higher frequencies ( @xmath6 ghz ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) . to investigate possible explanations for the fractional polarization trend seen in previous studies , we have produced the second data release of the australia telescope large area survey ( atlas dr2 ) as described in paper i @xcite of this two paper series .
atlas dr2 comprises reprocessed and new 1.4 ghz observations with the australia telescope compact array ( atca ) about the _ chandra _ deep field - south ( cdf - s ; galactic coordinates @xmath7 , @xmath8 ; * ? ? ?
* ) and european large area _
infrared space observatory _ survey - south 1 ( elais - s1 ; @xmath9 , @xmath10 ; * ? ? ?
* ) regions in total intensity , linear polarization , and circular polarization .
the mosaicked multi - pointing survey areas for atlas dr2 are 3.626 deg@xmath11 and 2.766 deg@xmath11 for the cdf - s and elais - s1 regions , respectively , imaged at approximately @xmath12 resolution .
typical source detection thresholds are 200 @xmath1jy in total intensity and polarization . in paper
i we presented our data reduction and analysis prescriptions for atlas dr2 .
we presented a catalogue of components ( discrete regions of radio emission ) comprising 2416 detections in total intensity and 172 independent detections in linear polarization .
no components were detected in circular polarization .
we presented a catalogue of 2221 sources ( groups of physically associated radio components ; grouping scheme based on total intensity properties alone , as described below ) , of which 130 were found to exhibit linearly polarized emission .
we described procedures to account for instrumental and observational effects , including spatial variations in each of image sensitivity , bandwidth smearing with a non - circular beam , and instrumental polarization leakage , clean bias , the division between peak and integrated flux densities for unresolved and resolved components , and noise biases in both total intensity and linear polarization .
analytic correction schemes were developed to account for incompleteness in differential component number counts due to resolution and eddington biases .
we cross - identified and classified sources according to two schemes , summarized as follows . in the first scheme ,
described in 6.1 of paper i , we grouped total intensity radio components into sources , associated these with infrared sources from the _ spitzer _ wide - area infrared extragalactic survey ( swire ; * ? ? ? * ) and optical sources from @xcite , then classified them according to whether their energetics were likely to be driven by an active galactic nucleus ( agn ) , star formation ( sf ) within a star - forming galaxy ( sfg ) , or a radio star . due to the limited angular resolution of the atlas data , in paper i we adopted the term _ lobe _ to describe both jets and lobes in sources with radio double or triple morphologies .
the term _ core _ was similarly defined in a generic manner to indicate the central component in a radio triple source . under this terminology , a core does not indicate a compact , flat - spectrum region of emission ; restarted agn jets or lobes may contribute or even dominate the emission observed in the regions we have designated as cores .
agns were identified using four selection criteria : radio morphologies , 24 @xmath1 m to 1.4 ghz flux density ratios , mid - infrared colours , and optical spectral characteristics .
sfgs and stars were identified solely by their optical spectra . of the 2221 atlas
dr2 sources , 1169 were classified as agns , 126 as sfgs , and 4 as radio stars . we note that our classification system was biased in favour of agns . as a result ,
the atlas dr2 data are in general unsuited for statistical comparisons between star formation and agn activity . in the second scheme ,
described in 6.2 of paper i , we associated linearly polarized components , or polarization upper limits , with total intensity counterparts . in most cases it was possible to match a single linearly polarized component with a single total intensity component , forming a one - to - one match .
in other cases this was not possible , due to ambiguities posed by the blending of adjacent components ; for example , a polarized component situated mid - way between two closely - separated total intensity components . in these cases , we formed group associations to avoid biasing measurements of fractional polarization .
we classified the polarization
total intensity associations according to the following scheme , which we designed to account for differing ( de-)polarized morphologies ( see paper i for graphical examples ) : * _ type 0 _ a one - to - one or group association identified as a lobe of a double or triple radio source .
both lobes of the source are clearly polarized , having linearly polarized flux densities within a factor of 3 .
( the ratio between lobe total intensity flux densities was found to be within a factor of 3 for all double or triple atlas dr2 sources . ) * _ types 1/2 _ a one - to - one or group association identified as a lobe of a double or triple radio source that does not meet the criteria for type 0 . a lobe classified as type 1 indicates that the ratio of polarized flux densities between lobes is greater than 3 . a lobe classified as type 2 indicates that the opposing lobe is undetected in polarization and that the polarization ratio may be less than 3 , in which case it is possible that more sensitive observations may lead to re - classification as type 0 .
sources with lobes classified as type 1 exhibit asymmetric depolarization in a manner qualitatively consistent with the laing - garrington effect @xcite , where one lobe appears more fractionally polarized than the opposite lobe . *
_ type 3 _ a group association representing a source , involving a linearly polarized component situated midway between two total intensity components .
it is not clear whether such associations represent two polarized lobes , a polarized lobe adjacent to a depolarized lobe , or a polarized core . * _ type 4 _ an unclassified one - to - one or group association representing a source . *
_ type 5 _ a one - to - one association clearly identified as the core of a triple radio source ( where outer lobes are clearly distinct from the core ) . *
_ type 6 _ a source comprising two type 0 associations , or a group association representing a non - depolarized double or triple radio source where blended total intensity and linear polarization components have prevented clear subdivision into two type 0 associations . * _ type 7 _ a source comprising one or two type 1 associations . * _ type 8 _ a source comprising one type 2 association . * _ type 9 _ an unpolarized component or source . in this
work ( paper ii ) we present the key observational results from atlas dr2 , with particular focus on the nature of faint polarized sources .
this paper is organised as follows . in [ ch5:secres ]
we present the atlas dr2 source diagnostics resulting from infrared and optical cross - identifications and classifications , diagnostics resulting from polarization total intensity cross - identifications and classifications , differential component number - counts , and our model for the distribution of fractional polarization . in [ ch5:secdisc ]
we compare the atlas dr2 differential counts in both total intensity and linear polarization to those from other 1.4 ghz surveys , and discuss asymmetric depolarization of classical double radio sources .
we present our conclusions in [ ch5:secconc ] .
this paper follows the notation introduced in paper i. we typically denote flux density by @xmath13 , but split into @xmath14 for total intensity and @xmath4 for linearly polarized flux density when needed for clarity .
in the following sections we present diagnostics of atlas dr2 sources resulting from the infrared and optical cross - identification and classification schemes described in 6.1 of paper i ( summarised in [ sec:1 ] of this work ) .
we focus on three parameter spaces formed by comparing flux densities between different wavelength bands : radio to mid - infrared , mid - infrared colours , and radio to far - infrared . in fig .
[ ch5:fig : rnir ] we compare the total intensity 1.4 ghz radio to 3.6 @xmath1 m mid - infrared flux densities for all 2221 atlas dr2 sources , taking into account infrared upper bounds for the 298 radio sources without detected infrared counterparts . the bottom - right panel of fig .
[ ch5:fig : rnir ] indicates that the atlas sources classified as stars or sfgs typically exhibit radio flux densities @xmath5 1 mjy .
the paucity of atlas sources with @xmath15 @xmath5 0.1 mjy and star or sfg classifications likely represents a selection bias , in which only those sources with relatively bright optical counterparts could be classified spectroscopically .
the top - left panel highlights all 130 polarized atlas sources , 12 of which are represented by upper bounds .
the paucity of polarized sources with @xmath16 @xmath5 1 mjy is due to the limited sensitivity of our linear polarization data ; fractional polarization trends will be presented in [ ch5:secresidentm ] . in fig .
[ ch5:fig : nircc ] we present mid - infrared colour - colour diagrams in which the irac flux density ratios @xmath17 and @xmath18 have been compared for atlas dr2 sources . of the 2221 atlas sources , 988 were detected in all four irac bands , while 935 were detected in only 2 or 3 bands ; the remaining 298 sources were not detected in any band , and have not been shown in fig .
[ ch5:fig : nircc ] . regarding the 130 polarized atlas sources ,
55 were detected in all four irac bands , 63 were detected in only 2 or 3 bands , and 12 were not detected in any irac band ; thus 118 polarized sources are indicated in fig .
[ ch5:fig : nircc ] . the dotted lines in each panel of fig .
[ ch5:fig : nircc ] represent the divisions identified through simulations by @xcite . by considering the effects of redshift evolution on the observed mid - infrared colours of three general source classes with spectral characteristics dominated by old - population ( 10 gyr ) starlight , polycyclic aromatic hydrocarbon ( pah ) emission , or a power - law continuum , @xcite identified four regions that could be used to preferentially select different source classes at different redshifts within the @xmath19 range simulated .
region 1 was found to preferentially select sources with spectra dominated by continuum emission , likely produced by dust tori associated with agns @xcite , over the full redshift range .
region 2 was found to preferentially select pah - dominated sources , indicative of intense star formation , over the full redshift range .
region 3 was found to preferentially select sources with spectra dominated by direct stellar light , but only for sources with @xmath20 @xmath5 0.4 . for increasing redshifts , region 3
was found to comprise a mixture of stellar- and pah - dominated sources as the latter migrated from region 2 .
however , beyond @xmath20 @xmath21 1.6 , region 3 was found to be largely void of sources following the migrations of both stellar- and pah - dominated sources towards region 4 .
region 4 was found to be largely void of sources for @xmath20 @xmath5 0.4 . for increasing redshifts ,
pah - dominated sources were found to migrate from region 2 into region 4 .
for @xmath20 @xmath21 0.9 , stellar - dominated sources from region 3 were also found to migrate into region 4 .
@xcite found that at all redshifts , sources dominated by pah emission were located slightly within the boundaries of region 1 . in order to classify as agns only those sources most likely to be such in fig .
[ ch5:fig : nircc ] , we constructed the restricted locus indicated by the dashed lines ; we label this region 1r . in paper
i we defined this locus ( following * ? ? ?
* ) as the union of @xmath22>0 $ ] , @xmath23>0 $ ] , and @xmath22 < 11\log_{{\scriptscriptstyle}10}[s_{5.8\,\mu{\textrm}{\scriptsize m}}/s_{3.6\,\mu{\textrm}{\scriptsize m}}]/9 + 0.3 $ ] .
continuum - dominated sources are expected to exhibit power - law spectra , given by the dot - dashed locus in each panel . as noted by @xcite , the spectra of continuum - dominated sources
are only expected to exhibit blue irac colours for largely unobscured agns , in cases for which their rest - frame optical wavelengths are redshifted into the mid - infrared band for sources at @xmath24 .
thus the atlas sources with blue irac colours in fig .
[ ch5:fig : nircc ] are unlikely to be represented by continuum - dominated sources as defined by @xcite .
however , this does not imply that a source observed with blue irac colours at @xmath25 can not be an agn , because sources with mid - infrared spectra dominated by old stellar light may yet exhibit stronger signs of agn activity at other wavelengths .
the bottom - left panel of fig .
[ ch5:fig : nircc ] indicates that atlas sources classified as agns are predominantly located in regions 1r and 3 .
the sources classified as agns in region 2 perhaps suggest combinations of star formation and agn activity , or perhaps misclassifications due to the largely statistical nature of our classification system .
the upper bounds classified as agns in the top - right panel are consistent with the observed distribution of agns presented in the bottom - left panel .
these upper bounds suggest that additional agns are situated in region 4 , though likely in proportion with the additional agns remaining undetected in regions 1 and 3 .
the bottom - right panel indicates that atlas sources classified as sfgs are predominantly located in region 2 , as well as between the boundaries of regions 1 and 1r , as expected .
a small number of atlas sources classified as sfgs are located in regions 1r and 3 , consistent with the migratory paths of pah - dominated sources .
the upper bounds classified as sfgs in the top - right panel are consistent with the majority of sfgs being located in region 2 .
all 4 atlas sources classified as stars are located in region 3 .
the polarized atlas sources detected in all four irac bands follow the distribution of agns , situated predominantly in regions 1 and 3 in almost equal proportions .
the upper bounds for polarized sources presented in the top - right panel are consistent with this finding .
the lack of polarized sources in region 2 suggests that the polarized atlas sources observed in region 3 are unlikely to be sfgs with rest - frame colours located in region 2 ( i.e. if sfgs are migrating from region 2 to region 3 with redshift , then a trail of sources would be expected ) .
instead , we find two concentrations of polarized atlas sources , highly coincident with the regions of parameter space identified by @xcite in which starlight- and continuum - dominated sources were preferentially located .
thus we find that the radio emission from polarized atlas sources is most likely powered by agns , where the active nuclei are embedded within host galaxies with mid - infrared spectra dominated by old - population stellar light ( blue irac colours ) or continuum likely produced by dusty tori ( red irac colours ) .
this finding is in general agreement with the results from the elais - north 1 ( elais - n1 ) region presented by both @xcite and @xcite , but with the following two notable exceptions .
first , both these works identified radio sources ( both polarized and unpolarized ) that were concentrated in region 3 about @xmath26 , @xmath27 , well beyond the parameter space typically occupied by the three generic source classes investigated by @xcite .
@xcite reported that these sources were associated with elliptical galaxies dominated by old - population starlight .
however , fig .
11 from @xcite indicates that these sources are located within a region of parameter space occupied by individual stars .
it is not clear why the irac colours of so many of the radio sources presented by @xcite and @xcite were found to occupy this region of parameter space , though it is possible that their selection of isophotal flux densities for unresolved infrared sources may have biased their colour ratios ( aperture values are more appropriate for point sources ) . and second , unlike these previous works , we do not find any polarized atlas sources in which the radio emission is likely to be powered by star formation ( i.e. we do not see any polarized sources in region 2 ; cf . * ? ? ?
* ) ; we can not conclude that any polarized sources have infrared colours suggestive of significant pah emission ( cf . * ? ? ?
the fractional polarization properties of atlas agns and sfgs are described in [ ch5:secresidentm ] and modelled in [ ch5:secressubpi ] . in fig .
[ ch5:fig : rfir ] we compare the total intensity 1.4 ghz radio to 24 @xmath1 m infrared flux densities for all atlas dr2 sources , taking into account infrared upper bounds for all radio sources without detected infrared counterparts .
as noted in paper i , we use 24 @xmath1 m flux density as a proxy for far - infrared ( fir ) flux density .
the bottom - left panel indicates that atlas sources classified as agns are prevalent both away from and on the fir - radio correlation ( frc ) .
the presence of a substantial number of agns below the dashed line demonstrates the value of using multiple diagnostic criteria to classify sources ; only sources above the dashed line have been classified as agns using the radio to far - infrared diagnostic .
the bottom - right panel indicates that , as expected , atlas sources classified as sfgs typically cluster along the frc and have radio flux densities @xmath5 1 mjy
. however , a small number of sources classified as sfgs ( and stars ) are observed with upper bounds clearly located within the agn parameter space .
the top - left panel highlights all 130 polarized atlas sources , indicating that each of these was classified as an agn .
no polarized stars or sfgs were detected in our data . we now present diagnostics of components , groups , and sources in atlas dr2 , resulting from the linear polarization@xmath28total intensity cross - identification and classification procedures described in 6.2 of paper i ( summarised in
[ sec:1 ] of this work ) . in this section
we focus on a number of parameter spaces in which we detail relationships between the polarized flux densities , fractional polarizations , classifications , and angular sizes of sources and their constituents . in fig .
[ ch5:fig : fracpolraw ] we plot the polarized flux densities and fractional polarizations for all atlas dr2 components , groups , and sources versus their total intensity flux densities , taking into account polarization upper limits .
the fractional polarization uncertainties displayed in the lower - left panel were estimated following standard error propagation as @xmath29 fig .
[ ch5:fig : fracpolraw ] shows that the polarization upper limits for components and sources are distributed almost identically , the reason being that the majority of unpolarized sources comprise a single component ( relevant statistics are detailed toward the end of this section ) .
regarding polarization detections , we find that all components , groups , and sources exhibit @xmath30 .
this finding is in contrast to the data from other 1.4 ghz polarization surveys .
@xcite found that 1% ( 381/38454 ) of polarized sources in the nrao vla sky survey ( nvss ) exhibited @xmath31 . @xcite and
@xcite found that 10% ( 8/83 ) and 7% ( 10/136 ) of polarized sources in the elais - n1 field exhibited @xmath31 , respectively .
@xcite found that 10% ( 84/869 ) of polarized sources throughout the two australia telescope low - brightness survey ( atlbs ) fields
exhibited @xmath31 .
if a population of extragalactic sources with high 1.4 ghz fractional polarizations were to exist , then it would be unexpected for such sources to be detected in the surveys above [ with @xmath32 full - width at half - maximum ( fwhm ) beam sizes ] yet undetected in this work ( with @xmath3 fwhm beam size ) , because the former are more susceptible to both beam and bandwidth depolarization .
instead , we attribute the lack of atlas sources with @xmath33 to our careful treatment of local ( rather than global ) root mean square ( rms ) noise estimates , in particular to our employment of blobcat s flood - fill technique for extracting polarized flux densities ( see @xcite for details regarding biases introduced through gaussian fitting ) , and to our statistical classifications of unresolved and resolved components .
we found through testing that components with abnormally high levels of fractional polarization ( up to and even beyond 100% ) could be obtained if the features above were not taken into account . in fig .
[ ch5:fig : fracpolclass ] we plot the polarized flux densities and fractional polarizations for all atlas dr2 sources only , indicating their infrared / optical classifications .
panels in the left column highlight polarized sources with infrared counterparts detected in all four irac bands , or otherwise .
we split those detected in all four bands into sources located within or just beyond region 3 in the lower - right panel of fig .
[ ch5:fig : nircc ] [ i.e. polarized sources with @xmath34 and those located within or just beyond region 1r [ i.e. @xmath35 .
we find no clear distinction between the fractional polarization properties of sources with blue ( region 3 ) or red ( region 1r ) mid - infrared colours .
it is possible that the region 3 polarized sources exhibit a larger dispersion in fractional polarization than the region 1r polarized sources ( compare range of observed fractional polarizations in lower - left panel of fig .
[ ch5:fig : fracpolclass ] ) , though given our sample size this marginal effect may be attributed to sampling variance . using data from @xcite , @xcite found that polarized sources in region 3 were more highly polarized than those in region 1r ; the atlas data do not support this result .
the distributions of upper limits presented in the right - column panels of fig .
[ ch5:fig : fracpolclass ] indicate that all sources with @xmath36 have been classified as agns .
the fractional polarization upper limits for sources classified as sfgs are not particularly restrictive , as their total intensity flux densities are typically @xmath5 1 mjy .
characteristic @xmath37 levels for the sub - millijansky sfg population are @xmath5 @xmath38 . focusing on the lower - left panel of fig .
[ ch5:fig : fracpolclass ] , we note that a general observational consequence of the rising distribution of fractional polarization upper limits with decreasing total intensity flux density is that the mean or median fractional polarization of _ detected _ polarized sources will always _ appear _ to increase with decreasing flux density .
this increase represents a selection bias ; it is not possible to detect low levels of fractional polarization for the faintest total intensity sources .
any changes to the underlying distribution of fractional polarization with decreasing total intensity flux density will be masked , and thus dominated , by this selection bias .
therefore , it is not possible to investigate the distribution of fractional polarization at faint flux densities without accounting for polarization non - detections .
recently , 1.4 ghz polarimetric studies of the elais - n1 field @xcite and atlbs fields @xcite concluded that their observational data demonstrated an anti - correlation between fractional polarization and total intensity flux density .
these studies found that sources with @xmath14 @xmath5 @xmath39 mjy were more highly polarized than stronger sources .
however , @xcite did not account for polarization upper limits , leading to their misinterpretation of selection bias as an indication of true anti - correlation .
@xcite accounted for selection bias using monte carlo analysis , effectively incorporating polarization upper limits .
@xcite accounted for selection bias by comparing samples of sources in bins of polarized flux density rather than total flux density , at sufficient polarized flux densities to neglect upper limits .
however , the findings of increased fractional polarization at faint total flux densities by @xcite and @xcite appear to be reliant on the increasing number of sources observed with @xmath33 at these faint levels .
for example , both studies reported extreme sources with @xmath40 , but only at faint total intensities .
both @xcite and @xcite found that @xmath41 of polarized sources with linearly polarized flux densities @xmath42 mjy ( i.e. a significant proportion of these sources ) exhibited @xmath33 , while no sources with such high levels of fractional polarization were found for @xmath43 mjy .
as described earlier , the @xmath33 sources ( and perhaps many with lower @xmath37 ) are likely to reflect rms noise estimation and source extraction errors .
the analytic form assumed by @xcite for the distribution of fractional polarization ( which will be described in [ ch5:secressubpi ] ) may have also contributed to their conclusion regarding increased fractional polarization ; spurious conclusions may be obtained if the observed fractional polarization data do not follow the assumed analytic form of the fit .
the arguments above suggest that existing evidence for an anti - correlation between fractional polarization and total flux density may not be robust .
similar to the studies above , earlier works by and @xcite concluded that nvss @xcite sources exhibited an anti - correlation between fractional linear polarization and total intensity flux density .
these analyses in effect incorporated polarization upper bounds ( though not upper _ limits _ ; see * ? ? ?
* ) because @xcite recorded a linearly polarized flux density for each nvss source , regardless of the statistical significance of the polarization measurement . to determine the significance of their findings and thus form a conclusion regarding evidence for anti - correlation , which we use to justify our own fractional polarization model presented in
[ ch5:secressubpi ] , we need to examine their works in more detail . and @xcite presented fractional polarization distributions for steep- and flat / inverted - spectrum nvss sources in four flux density intervals : @xmath44 , @xmath45 , @xmath46 , and @xmath47 mjy .
their distributions are remarkably consistent for @xmath48 , exhibiting a log - normal form with approximately equal dispersion and a peak at @xmath49 .
a separate component with a peak at @xmath50 is also present in each distribution , representing sources with polarization dominated by instrumental leakage .
we observe that the dispersions of their distributions broaden with increasing flux density , solely due to broadening at @xmath36 .
found that the median fractional polarization was larger for the @xmath44 mjy data than for the @xmath47 mjy data , for both steep- and flat - spectrum sources .
@xcite found the same result but for steep - spectrum sources only .
these results were essentially based on the lack of sources with @xmath51 in the @xmath44 mjy data when compared with the increased presence of such sources at higher flux densities ; a proportional increase in the number of sources with @xmath48 for decreasing flux density was not observed .
however , the presence of sources with @xmath51 ( i.e. less than the typical leakage level of @xmath50 ) , and more generally the slight changes in distribution shape observed for @xmath36 between different flux density intervals , may be more appropriately explained by the influence of noise on polarized flux densities rather than by variation in the underlying distribution of fractional polarization . to demonstrate
, we first note that the expectation value of @xmath4 for an unpolarized nvss source is given by the mean of a rayleigh distribution ( i.e. a ricean distribution with no underlying polarized signal ) , which is @xmath52 mjy for @xmath53 mjy .
this value is also characteristic of the expected observed polarized flux density for a source with true underlying polarized signal @xmath54 @xmath5 @xmath55 ( e.g. see the upper panel of fig . 1 from @xcite ) .
thus , a tail of sources with true polarization @xmath54 @xmath5 0.29 mjy will appear in the fractional polarization distribution at @xmath37 @xmath21 0.18% for total intensity sources with @xmath56 mjy , and at @xmath37 @xmath5 0.05% for @xmath57 mjy .
these estimates are consistent with the distributions presented by and @xcite ; a tail of sources with @xmath51 was observed for the @xmath47 mjy data but not for the @xmath44 mjy data .
we therefore conclude that the results presented by and @xcite do not demonstrate a statistically significant anti - correlation between fractional linear polarization and total intensity flux density .
furthermore , we note that the fractional polarization distributions presented in these works are likely to overestimate the population of sources with @xmath36 , even for the @xmath44 mjy data , for the following two reasons .
first , all catalogued nvss measurements of polarized flux density were debiased using a modified version of the expectation value for a ricean distribution @xcite .
this debiasing scheme is known to impart a significant overcorrection ( i.e. negative bias ) at low snr ( e.g. see ; the relevant scheme is labelled in reference to its application by @xcite ) .
thus measurements of fractional polarization obtained using the nvss catalogue are likely to be negatively biased . and
second , raw polarization measurements for nvss sources were obtained by interpolation at the total intensity centre position .
therefore , polarized flux densities were underestimated for each source in which the spatial peak of polarized emission was located in an adjacent pixel to the total intensity peak .
both of these effects could have been largely mitigated by obtaining polarization upper limits for sources , rather than upper bounds .
returning to the lower - left panel of fig .
[ ch5:fig : fracpolclass ] , we find that the maximum level of fractional polarization exhibited by atlas sources does not appear to be correlated with total intensity flux density .
the maximum level appears to be limited to @xmath37 @xmath5 20% for @xmath14 @xmath21 1 mjy , which becomes a strict limit for @xmath58 mjy when accounting for the presence of all upper limits .
furthermore , we find @xmath37 @xmath21 0.4% for sources with @xmath14 @xmath21 10 mjy , where sources exhibiting higher levels of fractional polarization significantly outnumber those potentially exhibiting @xmath59 as indicated by the upper limits .
in paper i we found that 138 of the total 172 catalogued linearly polarized components exhibited a clear one - to - one match with individual total intensity components .
the remaining 34 polarized components required grouping in order to be associated with total intensity counterparts . of the one - to - one associations , we classified 58 as type 0 , 4 as type 1 , 25 as type 2 , 48 as type 4 , and 3 as type 5 .
all 3 sources containing type 5 core associations were found to exhibit unpolarized lobes . of the group associations comprising a total of 34 polarized components , 2 groups were classified as type 0 , 14 as type 3 , 1 as type 4 , and 8 as type 6 . there were 29 sources classified as type 6 , 2 as type 7 , and 25 as type 8 .
these classifications are catalogued in appendix b of paper i. in fig .
[ ch5:fig : fracpoltypes ] we indicate the polarization@xmath28total intensity classifications for all polarized atlas dr2 components , groups , and sources . in the lower - left panel we plot the levels of fractional polarization exhibited by classical double or triple radio sources ( types 68 ) and their individual lobes ( types 02 , respectively ) .
we find that sources classified as type 6 , which comprise pairs of roughly equally polarized type 0 lobes , are located throughout most of the populated parameter space .
we find that type 7 sources , which comprise pairs of type 1 lobes where one is clearly less polarized than the other , appear to occupy the same parameter space populated by type 6 sources .
a selection bias against identifying type 0/1 lobes , and thus type 6/7 sources , is present within the diagonal region of parameter space populated by polarization upper limits ( for visual clarity these limits are not shown in fig .
[ ch5:fig : fracpoltypes ] ; see fig .
[ ch5:fig : fracpolraw ] ) .
type 2 lobes and their parent type 8 sources , which represent ambiguous cases in which it is not possible to differentiate between types 0/1 or 6/7 , are largely confined to this diagonal region .
given the observed prevalence of type 6 sources compared with type 7 , it seems likely that more sensitive observations would result in a majority of type 8 sources being reclassified as type 6 . from the lower - right panel of fig .
[ ch5:fig : fracpoltypes ] we find that sources classified as type 3 , which exhibit a single polarized component situated midway between two total intensity components , appear to populate the same region of parameter space occupied by type 8 sources .
similarly , associations classified as type 5 , which represent cores of triple radio sources , as well as the remaining unclassified sources denoted by type 4 , also appear to be concentrated within the diagonal region of parameter space populated by upper limits .
we note that many of the type 4 associations are likely to represent individual type 0 or type 1 lobes of as - yet unassociated multi - component sources , having been erroneously assigned to single - component sources in our catalogue ( note 6.1 of paper i ; statistics regarding polarized multi - component sources are presented below ) .
we find that type 5 associations occupy a parameter space consistent with type 6 and type 7 sources . as the latter represent average polarization properties for dual - lobed radio sources , it is possible that type 5 associations also represent dual - lobed structures but with small angular sizes , such as compact steep - spectrum ( css ) sources @xcite .
curiously , we found that each of the 3 sources with type 5 cores was found to exhibit unpolarized outer radio lobes .
it is possible that the type 5 cores represent restarted agn activity and that the outer lobes are unpolarized because any large - scale magnetic fields within them have dissipated over time since their production during an earlier distinct phase of agn activity .
for example , we may be seeing sources similar to the double - double radio galaxy j1835@xmath60620 , though at an earlier stage of evolution where the inner lobes have not yet separated into two separate lobes ( note that fractional polarization levels for the inner lobes of j1835@xmath60620 are higher than for the outer lobes ) . in fig .
[ ch5:fig : fracpoltheta ] we plot polarized flux density and fractional polarization versus largest angular size ( las ) for all polarized atlas dr2 sources , highlighted according to morphology and infrared colour .
the las for a single - component source is given by its total intensity deconvolved angular size or size upper limit , while the las for a multi - component source is given by the maximum angular separation between its constituent total intensity components .
for visual clarity we plot sources with polarization upper limits separately in fig .
[ ch5:fig : fracpolthetauls ] , also highlighted according to morphology and infrared colour .
note that the apparent anti - correlations between fractional polarization upper limits and angular size upper limits for single - component sources throughout fig .
[ ch5:fig : fracpolthetauls ] are spurious ; the restrictiveness of both types of upper limits are intrinsically anti - correlated with total intensity flux density . in fig .
[ ch5:fig : fracpoltheta2 ] we again plot polarized flux density and fractional polarization versus las for all polarized sources , but now highlighted according to the polarization@xmath28total intensity classification scheme from 6.2 of paper i. for reference , we note that 1 subtends a linear scale of 1.8 , 3.3 , 6.1 , 8.0 , and 8.5 kpc at redshifts 0.1 , 0.2 , 0.5 , 1.0 , and 2.0 , respectively @xcite , assuming a @xmath61cdm cosmology with parameters @xmath62 km s@xmath63 mpc@xmath63 , @xmath64 , and @xmath65 . following an evolutionary relationship for galaxy sizes given by @xmath66^{-\frac{1}{2 } } \;,\ ] ] and assuming that a typical galaxy has size @xmath67 kpc at @xmath68 ( e.g. * ? ? ?
* ) , the corresponding sizes of typical galaxies at the redshifts above are approximately @xcite , may cause observed angular sizes of extended sources to be smaller than true sizes , due to faint source edges . ] 19 , 18 , 16 , 12 , and 7 kpc , respectively , or 10 , 56 , 26 , 15 , and 08 , respectively .
we summarise our findings from figs .
[ ch5:fig : fracpoltheta][ch5:fig : fracpoltheta2 ] as follows . of the 130 ( 2091 )
polarized ( unpolarized ) sources catalogued in atlas dr2 and presented in fig . [ ch5:fig : fracpoltheta ] ( fig .
[ ch5:fig : fracpolthetauls ] ) , 81 ( 74 ) comprise multiple components in total intensity , 40 ( 140 ) comprise a single resolved component in total intensity , and 9 ( 1877 ) comprise a single unresolved component in total intensity .
we note that while components observed in linear polarization in atlas dr2 are typically unresolved ( only 29/172 or 17% of polarized components are resolved ; see 5 of paper i ) , 121/130 or 93% of sources exhibiting polarized emission are resolved in total intensity .
these statistics support the findings by @xcite that polarized 1.4 ghz sources tend to have structure at arcsecond scales and that , as a consequence , their polarized emission is unlikely to be beamed .
combined with our earlier classification from fig .
[ ch5:fig : rfir ] of all polarized atlas sources as agns , and our interpretation from fig .
[ ch5:fig : fracpoltypes ] that most or all polarized components are associated with agn jets or lobes ( rather than cores ) , the statistics above demonstrate that ( sub-)millijansky polarized sources tend to be extended jet- or lobe - dominated active radio galaxies .
this conclusion is supported by the finding from @xcite that polarized sources tend to have steep spectra , which are characteristic of lobes . in fig .
[ ch5:fig : fracpolthetauls ] we find that atlas dr2 sources typically have las @xmath5 10 , suggesting that most sources are located at @xmath20 @xmath21 0.2 .
this is consistent with the preliminary redshift distributions presented by @xcite and @xcite for atlas dr1 sources ( see also discussion of radio source redshift distribution by * ? ? ?
focusing on the panels in the lower - left corners of fig .
[ ch5:fig : fracpoltheta ] and fig .
[ ch5:fig : fracpolthetauls ] , we find that single- and multi - component sources are distributed approximately equally in fractional polarization space ; their fractional polarization upper limits are not restrictive enough to identify any possible underlying trends . however , having found above that polarized sources are likely to represent lobed galaxies , it is perhaps surprising that we do not find a clear correlation between fractional polarization and las due to beam depolarization .
given the @xmath3 resolution of atlas , in general a classical double radio source with dual polarized lobes should exhibit greater fractional polarization than a similar source with smaller las that is observed as a single - component source . a likely explanation may be that a significant number of the polarized single - component sources indicated in fig .
[ ch5:fig : fracpoltheta ] are actually individual lobes of as - yet unassociated multi - component sources ( see 6.1 of paper i ) .
note that all single - component sources in fig .
[ ch5:fig : fracpoltheta ] are classified as type 4 in fig .
[ ch5:fig : fracpoltheta2 ] .
another potential explanation may be that for dual - lobed sources with small angular size observed as single - component sources , asymmetric depolarization between the lobes @xcite could result in overall source fractional polarization levels similar to those of type 7 sources ( see fig .
[ ch5:fig : fracpoltheta2 ] ) , rather than resulting in significantly beam - depolarized ( and thus perhaps unpolarized ) sources overall .
the upper limits presented in the left column of fig . [ ch5:fig : fracpolthetauls ] do not reveal any clear underlying trends within or between source classes .
the multi - component sources classified as sfgs in fig .
[ ch5:fig : fracpolthetauls ] , which are also shown located within the agn parameter space in the lower - right panel of fig .
[ ch5:fig : rfir ] , require future study .
these may represent composite sources exhibiting both agn and sfg characteristics , for example similar to the ultra - luminous infrared galaxy f001837111 investigated by @xcite or the more general classes of post - starburst quasars ( e.g. * ? ? ?
focusing on the right column of fig .
[ ch5:fig : fracpoltheta ] , we do not find any angular size distinctions between polarized sources based on their infrared colours .
furthermore , we find no underlying trends within the associated upper limit data from the right column of fig .
[ ch5:fig : fracpolthetauls ] . focusing on fig .
[ ch5:fig : fracpoltheta2 ] , we find that type 6 sources typically extend to greater angular sizes than type 7 sources , though a larger sample size with proportionally fewer type 8 classifications is required to confirm this finding .
we also find that each of the 3 polarized cores classified as type 5 are resolved , and that they populate the same region of parameter space as type 4 sources .
we present euclidean - normalised differential number - counts derived from the atlas dr2 total intensity and linear polarization component catalogues in fig .
[ ch5:fig : countsi ] and fig .
[ ch5:fig : countsl ] , respectively , and in tabulated form in appendix a. counts for each bin have been plotted and tabulated at the expected average flux density , which we denote by @xmath69 , as given by equation ( 19 ) from .
this value takes into account the number - count slope and becomes important when assigning flux densities for bins containing few counts or with large widths in flux density space ; @xmath69 only equals the bin geometric mean when @xmath70 , where @xmath71 is the slope of the differential number counts @xmath72 .
bin widths for all total intensity counts were selected to be a factor of 0.07 dex for @xmath73 mjy , 0.13 dex for @xmath74 mjy , and 0.2 dex otherwise . in linear polarization ,
bin widths were selected to be a factor of 0.16 dex for @xmath75 mjy , and 0.3 dex otherwise .
we removed all bins containing components with visibility area corrections @xmath76 , so as to prevent the number - counts from being dominated by the few components detected in the most sensitive and potentially least - representative regions of the atlas images .
( note that we did not remove individual offending components in order to retain the faintest bins , as this would have led to a bias in their resulting number - counts . ) in total intensity this resulted in the removal of the faintest few bins containing @xmath7730 components from each of the cdf - s component- and bin - corrected datasets , and @xmath7720 components from each of the elais - s1 component- and bin - corrected datasets .
the maximum visibility area corrections for any components in the remaining valid cdf - s and elais - s1 bins were @xmath78 and @xmath79 , respectively .
in linear polarization , the maximum visibility area corrections for any components in the cdf - s and elais - s1 datasets were @xmath80 and @xmath81 , respectively .
as a result , we did not remove any bins in linear polarization .
resolution and eddington bias corrections were calculated in 7 of paper i. the former was designed to correct for incompleteness to resolved components with low surface brightness , and for the redistibution of counts between bins resulting from systematic undervaluation of flux densities for components classified as unresolved .
the latter was designed to correct for the redistribution of counts between bins due to random measurement errors in the presence of a non - uniformly distributed component population .
these bias corrections were calculated in paper i by assuming that the true underlying differential number counts in total intensity were given by the sixth - order empirical fit to the phoenix and first surveys presented by @xcite .
this fit , which we denote h03 , is given by @xmath82 = \sum_{j=0}^{6 } a_{j } \left [ \log\left ( \frac{i}{{\textrm}{mjy } } \right)\right]^{j}\,,\ ] ] with @xmath83 , @xmath84 , @xmath85 , @xmath86 , @xmath87 , @xmath88 , and @xmath89 . to illustrate the potential boosting effects of an exaggerated population of faint components , paper
i also defined a modified h03 distribution , denoted h03 m , in which a euclidean slope was inserted between 30@xmath28300@xmath1jy , @xmath90 for bias corrections in linear polarization , we modelled the true underlying differential number counts @xmath91 by convolving the total intensity h03 distribution from equation ( [ ch4:eqn : h03 ] ) with a probability distribution for fractional linear polarization @xmath92 , which we denote @xmath93 .
the @xmath94 distribution is presented in equation ( [ ch5:eqn : fracpol ] ) in [ ch5:secressubpi ] .
the atlas dr2 component counts extend down to a flux density of approximately 140 @xmath1jy in both total intensity and linear polarization .
the brightest flux density bins are sparsely sampled because the atlas survey areas are not large enough to include significant numbers of increasingly rare bright components . in both fig .
[ ch5:fig : countsi ] and fig .
[ ch5:fig : countsl ] we find that the number - counts from the two separate atlas fields are consistent within the errors over their full observed flux density ranges .
the impacts of the combined resolution and eddington bias corrections on the number - counts appear to be relatively minor . in total intensity
, the two corrections largely cancel each other out , while in linear polarization the resolution bias corrections dominate . in both total intensity and linear polarization ,
the combined corrections affect the underlying visibility area corrected counts by a factor of @xmath5 0.5 , and do not affect the counts for @xmath13 @xmath21 3 mjy .
we find that differences between the two independent eddington bias correction schemes are largely negligible for both the total intensity and linear polarization number - counts , providing confidence in these approaches . in fig .
[ ch5:fig : countsi ] we find that the total intensity counts closely follow the h03 model within a factor of @xmath95 , though the atlas counts may begin to systematically drop below the h03 model for @xmath13 @xmath5 @xmath96 mjy .
it is likely that the drop is caused by residual incompleteness in our resolution bias corrections , in turn caused by uncertainties regarding our assumed true angular size distribution for @xmath97 as discussed in 7.1 of paper i. however , we note that if we assume that the model presented in fig .
19 of paper i is the best representation of the true angular size distribution ( without any flux density scaling ) , then the faintest bins at @xmath98 @xmath1jy only require an additional correction factor of at most approximately @xmath6030% .
the faintest bins are therefore consistent with the h03 model .
as we do not find any systematic divergence between the atlas total intensity counts and the h03 model at the faintest flux densities ( when accounting for the suspected residual resolution bias described above ) , we confirm that the h03 model is suitable for predicting 1.4 ghz component counts ( and source counts as described below ) down to at least @xmath0 @xmath1jy in surveys with resolution fwhm @xmath3 . should we have found a systematic divergence , it would have indicated that our predicted eddington bias corrections were unrealistic , and that in turn the h03 model underpinning these corrections formed an increasingly poor representation of the true number - counts for decreasing flux density . under this hypothetical situation
, an iterative approach would have been required in order to correctly identify an input true number - count model so as to bring about convergence with the fully corrected observed counts . in 7.2 of paper i we predicted the levels of eddington bias that would be present within the observed atlas counts if the true counts were given by the h03 or h03 m models [ the latter model contains a larger population of components with @xmath99 mjy than the former ; see equation ( [ ch4:eqn : h03 m ] ) ] .
we predicted that the h03 m model would induce significantly greater eddington bias at @xmath99 mjy than the h03 model ( see fig . 23 in paper
i ) . therefore
, if the h03 model was used to predict the observed eddington bias when in fact the h03 m model best represented the true counts , then the observed counts would exhibit significant positive residual eddington bias ; if vice versa , the residual bias would be negative . given that we do not observe a systematic rise ( or fall ) at faint flux densities in the fully corrected atlas counts ( again accounting for the suspected residual resolution bias described above ) , we conclude that the h03 m model is not supported by the atlas data .
we note that the resolution bias corrections applied in this work are practically insensitive to changes between the h03 and h03 m models .
this is because for any given flux density bin , the resolution bias corrections are unaffected by the assumed form of the number - count distribution at fainter flux densities .
therefore , assuming that our resolution bias corrections are appropriate to begin with , we can focus on eddington bias alone in order to draw the conclusions described above . below a flux density of @xmath100 mjy
, we expect the atlas total intensity component counts to be dominated by single - component sources , with negligible contributions from components within multi - component sources .
while we are unable to explicitly quantify this expectation given present data , we note that conservatively @xmath101 of all 2416 atlas components reside within multi - component sources ( this fraction takes into account the number of components estimated to reside within as - yet unassociated multi - component sources ; see 6.1 of paper i ) .
we expect that most of these multi - component sources represent frii sources , which are known to dominate the source counts at flux densities @xmath21 10 mjy and which diminish significantly below @xmath102 mjy ( e.g. * ? ? ?
* ) . at sub - mjy levels , radio sources in general
are expected to have angular sizes @xmath103 ; these are likely to be observed as single - component sources in atlas .
therefore , we conclude that the atlas component counts may act as a suitable proxy for source counts at sub - mjy levels .
we note that our characterisation of the faint component / source population using the h03 model in this work has relied on the similar resolutions of the atlas and phoenix surveys . should these resolutions have differed significantly , so too would have the properties of their observed components .
@xcite obtained their model by using a sixth - order fit to the observed component counts from the phoenix survey , supplemented at @xmath104 mjy by source counts from the first survey @xcite .
the h03 model was thus intended to characterise source counts at all flux densities , despite being derived from a component catalogue at faint flux densities . for @xmath104 mjy ,
the atlas total intensity component counts follow the h03 model and thus the first source counts .
we explain this correspondence as follows by first presenting results that examine how source and component counts are expected to differ .
given that frii sources dominate the source counts above @xmath2 mjy and that these sources are likely to comprise multiple components within a survey such as atlas , we expect the differential counts for sources to rise and extend to brighter flux densities than those for components . to roughly illustrate this behaviour and examine the difference between source and component counts in general , we considered an idealised scenario in which all sources were assumed to comprise two identical components , each with half the flux density of their parent . for illustrative purposes we assumed that the component count distribution was given by the h03 model . to derive the idealised differential source counts , we integrated the differential component counts to obtain integral component counts , divided these integral counts by two , doubled the flux density scale , and differentiated . for completeness , we also derived differential source counts in linear polarization by following a similar procedure , where the relevant differential component counts were assumed to follow the @xmath93 model .
we present the resulting total intensity and linear polarization source counts in fig .
[ ch5:fig : ratiosc ] .
we find that the predicted source counts remain within @xmath105 of the component counts across the flux density ranges probed by the atlas data in total intensity ( @xmath14 @xmath5 1 jy ) and linear polarization ( @xmath4 @xmath5 100 mjy ) .
( separately , while not shown , we note that the integral counts for both components and sources within our rudimentary model are very similar , for both total intensity and linear polarization . )
as expected , at bright flux densities the component counts drop below the source counts , though these drops occur at brighter flux densities than relevant to the atlas data .
note that in reality , the differential source and component counts are likely to overlap more closely than presented in fig .
[ ch5:fig : ratiosc ] because of the presence of single - component sources .
thus we conclude that for surveys with resolution fwhm @xmath3 similar to phoenix and atlas , the h03 model may be used to characterize both component and source counts in total intensity for @xmath13 @xmath5 1 jy .
we conjecture that , as modelled above , the h03 model characterises component rather than source counts at all flux densities , including at @xmath106 jy . to justify this claim
, we note that components in the first survey were only grouped into multi - component sources if they were located within 50@xcite . from fig .
[ ch5:fig : fracpoltheta ] of this work we can see that a cutoff of 50 is likely to be too small to capture sources with the most widely - separated components , which are also likely to be the brightest sources .
in addition , flux densities for extended first components are likely to be underestimated due to insensitivity to extended emission . therefore , the first source counts are likely to be deficient at the brightest flux densities
. incidentally , the first source counts and thus the h03 model appear to form a suitable hybrid distribution for describing component counts at all flux densities in surveys with resolution fwhm @xmath3 such as atlas .
we may therefore conclude that the @xmath93 model is suitable for characterising component counts in linear polarization at all flux densities , not just at @xmath4 @xmath5 100 mjy where differences between polarized component and source counts are likely to diminish as shown in fig .
[ ch5:fig : ratiosc ] .
if the h03 model were to better represent source counts rather than component counts at @xmath107 jy , then the polarized counts resulting from convolution with @xmath108 would reside ambiguously between a component and source count distribution for @xmath4 @xmath21 5 mjy .
thus it would be inappropriate to estimate integral component or source counts from the @xmath93 ( or indeed h03 ) model ; this point is relevant to results presented shortly . in fig .
[ ch5:fig : countsl ] we find that the atlas linear polarization component counts steadily decline with decreasing flux density , as generally predicted by all four models displayed in the background .
the solid curve displays our assumed true component count model , namely @xmath93 , which we used to derive the corrections for resolution and eddington bias .
the fully corrected atlas counts closely follow this model within statistical error , indicating consistency between the model , the corrections , and the observational data .
each of the four background models in fig .
[ ch5:fig : countsl ] were calculated by convolving the h03 model with a fractional polarization distribution .
we note that these convolutions are only appropriate because , as described above , the h03 model appears to appropriately characterise the total intensity component counts at all flux densities relevant to atlas . in [ ch5:secressubpi ] we describe each of the fractional polarization distributions underlying the four background models , and compare their abilities to predict the atlas polarized counts and polarization data in general .
the number of polarized components expected per square degree at or brighter than a given flux density , as constrained by the observed atlas component counts , can be estimated by integrating the @xmath93 polarized count distribution ( the solid curve in fig .
[ ch5:fig : countsl ] ) .
the resulting integral component counts are displayed in fig .
[ ch5:fig:17 ] .
we estimate that the sky density of polarized components for @xmath109 @xmath1jy is 30 deg@xmath110 , for @xmath111 @xmath1jy it is 50 deg@xmath110 , and for @xmath112 @xmath1jy it is 90 deg@xmath110 . if we make the rudimentary assumption described earlier regarding fig .
[ ch5:fig : ratiosc ] that every polarized component belongs to a dual - component source with double the flux density , we can estimate the integral source count distribution ; this is displayed alongside the integral component count distribution in fig .
[ ch5:fig:17 ] .
we thus estimate that the sky density of polarized sources for @xmath109 @xmath1jy is @xmath113 deg@xmath110 , and for @xmath111 @xmath1jy it is @xmath114 deg@xmath110 .
we expect that these integral source count estimates are accurate to within 10% , even if a more suitable model incorporating polarized single - component sources is utilised . in this section
we present a model to describe the distribution of fractional polarization for agn sources and their components / groups observed at 1.4 ghz in surveys with resolution fwhm @xmath21 10 , as constrained by the atlas dr2 data .
there appears to be a significant overlap between the fractional polarization properties of all classification types representing both components / groups and sources in fig .
[ ch5:fig : fracpoltypes ] . taking into account the presence of upper limits
( see fig .
[ ch5:fig : fracpolraw ] ) , we find that typical levels of fractional polarization are concentrated between 0.4% and 20% , regardless of whether the focus is on sources or on their constituent components / groups . given this apparent overlap , we assume for simplicity that the distribution of fractional polarization for both components / groups and sources can be modelled using the same pdf , which we denote by @xmath94 . before presenting our model for this distribution
, we note three caveats .
first , following our conclusions presented in [ ch5:secresidentm ] regarding potential correlation of the distribution of fractional polarization with total flux density , we assume that @xmath94 is independent of total intensity flux density
. this assumption may not be suitable for @xmath14 @xmath5 @xmath39 mjy for which our atlas data become sparse .
second , our model for @xmath94 may only be relevant for surveys with resolution fwhm @xmath21 10 .
surveys with finer resolution may encounter less beam depolarization across components , and thus recover higher average levels of fractional polarization ( in 5 of paper i we found that @xmath115 of polarized atlas components were resolved ) .
we note that surveys with coarser resolution will incur increased blending between components within multi - component sources , resulting in a greater number of low-@xmath37 sources than observed for atlas due to enhanced beam depolarization .
and third , given that all polarized components in atlas dr2 are associated with agns , we restrict our model for @xmath94 to the characterisation of agns , rather than the characterisation of all radio sources including sfgs and individual stars .
we do not attempt to differentiate between different types of agns or their components within our model , i.e. fri / frii / radio quiet / core / lobe .
we discuss fractional polarization levels for sfgs in [ ch5:secdiscsfg ] .
we modelled @xmath94 by qualitatively fitting two independent sets of atlas data : ( i ) the fractional polarizations of components , groups , and sources displayed in fig .
[ ch5:fig : fracpolraw ] , importantly taking into account upper limits , and ( ii ) the differential number - counts for polarized components displayed in fig .
[ ch5:fig : countsl ] .
we obtained a concordance fit to these data by modelling @xmath94 using a log - normal distribution , @xmath116^{{\scriptscriptstyle}2 } } { 2 \sigma_{{\scriptscriptstyle}10}^{{\scriptscriptstyle}2}}\bigg\ } \,,\ ] ] where the parameters @xmath117 and @xmath118 are the median fractional polarization and scale parameter , respectively , given by best - fit values @xmath119 and @xmath120 .
the fit given by equation ( [ ch5:eqn : fracpol ] ) is consistent with the result obtained by analysing the fractional polarization data alone , using the product - limit estimator @xcite as implemented within the survival package in the r environment .
the mean level of fractional polarization for the distribution in equation ( [ ch5:eqn : fracpol ] ) is given by @xmath121 , which equates to @xmath122 .
for values of @xmath117 or @xmath118 larger than the best - fit values above , we found that the @xmath93 model predicted differential counts in excess of the observed atlas counts . for smaller values ,
the predicted counts were deficient . we plot equation ( [ ch5:eqn : fracpol ] ) in fig .
[ ch5:fig : pimodels ] . for comparison
we also plot the 1.4 ghz fractional polarization distributions proposed by @xcite , @xcite , and @xcite . for clarity
we explicitly document each of these distributions , as follows .
@xcite investigated the distribution of fractional polarization for nvss sources with @xmath123 mjy , which they fit using the following quasi log - normal form , @xmath124^{{\scriptscriptstyle}2 } } { 2 \sigma_{{\textrm}{\tiny b04}}^{{\scriptscriptstyle}2}}\bigg\ } \,,\ ] ] where @xmath125 @xmath126 and where @xmath127 .
the median and mean fractional polarization levels of the @xmath128 distribution are 2.1% and 3.3% , respectively .
similarly , @xcite investigated the distribution of fractional polarization for nvss sources with @xmath129 mjy , which they fit using the following monotonic form , @xmath130^{{\scriptscriptstyle}-1 } + b_{{\textrm}{\tiny t04}}\right\}\ ] ] where @xmath131 , @xmath132 and where we have included a correction factor of 1.32 to ensure that the distribution is normalised .
the median and mean fractional polarization levels of the @xmath133 distribution are 2.1% and 2.7% , respectively .
@xcite fit the distribution of fractional polarization for sources with @xmath134 mjy in the elais - n1 field by modifying a gram - charlier series of type a ( e.g. * ? ? ?
* ) , resulting in the following monotonic form , @xmath135\bigg\ } & \\ \hspace{4 cm } \textrm{if $ i<30$~mjy}\\ f_{{\textrm}{\tiny b04}}\left(\pi\right ) & \\ \hspace{4 cm } \textrm{if $ i\ge30$~mjy}\,,\\ \end{array } \right.\ ] ] where @xmath136 , @xmath137 , and where we have included a correction factor of 11.06 to ensure that the distribution is normalised . for @xmath138
mjy , @xcite found that the elais - n1 data were consistent with the @xmath128 distribution from equation ( [ ch5:eqn : fracpolb04 ] ) .
the median and mean fractional polarization levels of the @xmath139 distribution for @xmath134 mjy are 4.8% and 6.0% , respectively .
the four curves presented in fig . [ ch5:fig : pimodels ] are replicated in figs .
[ ch5:fig : fracpolraw][ch5:fig : fracpoltheta ] and fig .
[ ch5:fig : fracpoltheta2 ] .
the four curves are also presented in fig.s [ ch5:fig : countsl ] and [ ch5:fig : countsl2 ] following convolution with the h03 differential count model . in fig .
[ ch5:fig : countsl2 ] we find that the fractional polarization distributions proposed by @xcite , @xcite , and @xcite are in general agreement with the observed atlas polarized number counts .
the models predict polarized counts that are within a factor of 5 of each other , and they all pass within a few standard errors of the atlas data points . however , we find that these three distributions are incompatible with the observed distribution of fractional polarization for atlas components , groups , sources , and in particular upper limits as presented in fig .
[ ch5:fig : fracpolraw ] .
the extended tails below @xmath36 for the distributions proposed by @xcite and @xcite are likely to reflect the various systematic biases we described earlier in [ ch5:secresidentm ] regarding the nvss data . polarized flux densities for nvss sources
were recorded regardless of whether or not the measurements met statistical criteria for formal detection . if upper limits were calculated for the nvss data following a similar procedure to that described for the atlas data in 6.2 of paper i , then we suspect that far fewer detections strictly implying @xmath36 would have been made .
we note that the @xmath108 model proposed in this work peaks at @xmath140 , which is consistent with the nvss data for @xmath48 from and @xcite . the extended tail below @xmath36 in the @xcite model reflects their assumption that the distribution peaks at @xmath141 and declines monotonically with increasing @xmath37 .
the atlas dr2 data do not support this assumption . as noted earlier
, a caveat of the @xmath108 model is that it may not be suitable for @xmath14 @xmath5 @xmath39 mjy , because the upper limits presented in fig .
[ ch5:fig : fracpolraw ] do not constrain the behaviour of the true fractional polarization distribution for low values of @xmath37 . however , given that the maximum level of fractional polarization exhibited by atlas components and sources appears to be limited to @xmath37 @xmath5 @xmath142 , and given that this limit appears to be uncorrelated with flux density down to at least @xmath100 mjy ( see comments regarding fig .
[ ch5:fig : fracpolclass ] in [ ch5:secresidentm ] ) , we may draw tentative conclusions regarding the true distribution of fractional polarization for @xmath143 @xmath5 @xmath14 @xmath5 @xmath39 mjy .
the atlas dr2 data are consistent with 3 general alternatives .
first , the @xmath108 distribution may remain unchanged for @xmath144 mjy .
second , for decreasing @xmath14 , the mean of @xmath108 may decrease while its dispersion increases so as to maintain an approximately constant level of fractional polarization for outliers with large @xmath37 . and
third , for decreasing @xmath14 , the mean of @xmath108 may increase while its dispersion decreases .
more sensitive observations are required to distinguish between these alternatives .
in fig . [ ch5:fig : ratiosc ] of [ ch5:secressubcnts ] we demonstrated that differences between differential number - counts of components and sources within a survey such as atlas are likely to be negligible below @xmath100 jy in total intensity , and below @xmath0 mjy in linear polarization .
we may therefore directly compare the atlas dr2 component counts with source counts from the literature in both total intensity and linear polarization .
we present these comparisons in the following two sections . in fig .
[ ch5:fig : countsi2 ] we compare the atlas dr2 bin - corrected total intensity component counts ( from fig . [ ch5:fig : countsi ] or tabulated data from appendix a ) with source counts from other 1.4 ghz surveys of comparable sensitivity .
these include the b1301@xmath603034 field @xcite , the hdf - n , lockman hole , and elais - n2 fields @xcite , the ssa13 field @xcite , the j1046@xmath605901 field ( * ? ? ?
* hereafter om08 ) , a revised survey of the lockman hole @xcite , the goods - n field @xcite , the cdf - s field observed with the vla @xcite , and the atlbs fields with counts at @xmath145 mjy @xcite and deeper counts @xcite . at high flux densities
the source counts are dominated by luminous radio galaxies and quasars .
the flattening of the source counts below 1 mjy is produced by the emerging dominance of a population of sources comprised of radio - quiet agns ( agns lacking significant jets and dominated in the radio band by non - thermal emission ) , low - power radio - loud agns , and star forming galaxies @xcite .
the extent to which the source counts flatten is somewhat controversial because counts from deep surveys appear to exhibit a large degree of scatter , for example as seen in fig .
[ ch5:fig : countsi2 ] where there is a factor of 2 variation in the counts below 1 mjy .
measurements at 3 ghz from the absolute radiometer for cosmology , astrophysics , and diffuse emission ( arcade ) 2 balloon - borne experiment have indicated a temperature for the radio background about five times that previously expected from known populations of radio sources @xcite , which if not due to a residual calibration error @xcite suggest the presence of a new population of faint ( @xmath146 @xmath1jy at 1.4 ghz ) or diffuse ( few mpc in extent ) extragalactic sources @xcite .
some studies have attributed the large scatter in the faint counts to cosmic variance , namely to intrinsic differences between survey fields caused by large scale structure ( e.g. * ? ? ?
* ) . however , significant differences in the counts for fields observed in separate studies , such as the goods - n field ( located within the hdf - n field ) or the lockman hole ( see fig . [
ch5:fig : countsi2 ] ) , indicate that data processing and calibration errors may be entirely responsible for the scatter ( e.g. * ? ? ?
by considering the consistent power - law form of the angular correlation function for both nvss and first sources obtained by @xcite and , @xcite estimated the cosmic variance for millijansky radio sources to be @xmath147 where @xmath148 is the survey area in square degrees .
the total variance for each source count bin containing @xmath149 sources is then given by @xmath150 , which includes poisson chance .
for a survey with @xmath151 deg@xmath11 and @xmath152 , cosmic variance contributes @xmath153 to the total rms uncertainty for each bin ( this is consistent with a similar estimate presented by @xcite and a more detailed analysis by @xcite ) .
the clustering behaviour of sub - millijansky sources is likely to be similar to that of millijansky sources , or perhaps even less clustered @xcite , in which case the cosmic variance contribution estimated above represents a conservative upper limit .
the error bars for many of the faintest counts in fig .
[ ch5:fig : countsi2 ] require enlargement by factors much larger than @xmath154 to become consistent with each other within a few standard errors .
our experience in constructing source counts for atlas suggests to us that there are a large number of data processing procedures that , if not carefully implemented , could easily give rise to significant systematic biases of order the observed scatter in the faint counts .
we therefore agree with previous conclusions in the literature that the observed scatter in the sub - millijansky counts is likely to be significantly affected by data processing differences between surveys .
the atlas data support the h03 model down to @xmath0 @xmath1jy and rule out any flattening above this level ; flattening similar to the h03 m model is ruled out by a lack of residual eddington bias .
however , the dr2 data are not sensitive enough to support or refute the general trend of flattening reported by deeper surveys .
recently , @xcite used the probability of deflection technique [ @xmath155 ; @xcite ] and a spectral index conversion to investigate the behaviour of the 1.4 ghz source counts at @xmath156 @xmath1jy within a confusion - limited observation of the om08 j1046@xmath605901 field at 3 ghz . by combining the results from a similar @xmath155 analysis performed by mc85
, @xcite ruled out any flattening or an upturn in the 1.4 ghz euclidean counts between 2 @xmath1jy and 100 @xmath1jy , such as that reported by om08 or proposed to account for the arcade 2 results @xcite . in fig .
[ ch5:fig : countsl2 ] we compare the atlas dr2 bin - corrected linear polarization component counts ( from fig . [ ch5:fig : countsl ] or tabulated data from appendix a ) with the 1.4 ghz polarized source counts from the nvss @xcite and the elais - n1 field ( @xcite ; deeper counts from @xcite ) .
the atlas dr2 counts improve upon the @xcite study by a factor of @xmath157 in sensitivity .
the observed number - counts from the @xcite , @xcite , and @xcite studies are in general agreement with the atlas counts , though the atlas data do not exhibit flattening at faint levels that might otherwise lead to suggestions of increasing levels of fractional polarization with decreasing flux density or perhaps the emergence of a new source population .
the flattening of the data from these studies are unlikely to be real , but rather probably reflective of spurious populations of sources with abnormally high levels of fractional polarization as described earlier in [ ch5:secresidentm ] .
this explains the difficulty encountered by @xcite in attempting to model the flattening .
similarly , despite the apparent consistency between the observed counts and the various predicted curves displayed in fig .
[ ch5:fig : countsl2 ] , in [ ch5:secressubpi ]
we found that the fractional polarization distributions presented by @xcite , @xcite , and @xcite were inconsistent with the atlas data and therefore unlikely to be suitable for population modelling .
the flux density range over which the @xmath93 model extends in fig .
[ ch5:fig : countsl2 ] corresponds to the brightest regions of the total intensity counts , in which luminous radio galaxies and quasars dominate .
this is consistent with our independent conclusion from [ ch5:secresidentm ] that the polarized sources contributing to the atlas counts tend to be fri / ii radio galaxies , and with our earlier findings regarding the infrared colours of polarized sources from [ ch5:secres:2 ] .
to fully confirm this picture , luminosity functions for polarized sources of different classifications need to be constructed ( e.g. fri / frii / radio - quiet ) , which can then be compared with theory ( e.g. * ? ? ?
very recently , while we were finalising this manuscript for resubmission , @xcite published a similar study of faint polarized sources .
@xcite presented 1.4 ghz linearly polarized integral ( not differential ) source counts from the goods - n field , observed with the vla at @xmath158 resolution .
their polarized counts extend to 20 @xmath1jy , an order of magnitude deeper than our atlas dr2 results .
qualitatively , their results are consistent with our main finding that the fractional polarization levels of faint sources are not anti - correlated with total flux density .
quantitatively , however , their results are discrepant with ours . @xcite
predict that surveys with @xmath159 resolution will observe a polarized source density of 22 deg@xmath110 for @xmath111 @xmath1jy ; this is a factor of 2 lower than the directly observed counts presented in this work .
explain that it is difficult to directly compare polarized source counts from surveys with @xmath158 and @xmath159 resolutions , but they consider a factor of 2 difference to be optimistically large ( this factor was used to form their @xmath159 resolution prediction ) .
given that @xcite use peak surface brightness measurements as a proxy for flux density irrespective of source angular size , that they do not attempt to correct their data for effects such as resolution bias , and that they do not present total intensity counts with similar processing as for their linear polarization counts , it is difficult to assess the robustness of their results here .
detailed assessment is beyond the scope of this work , and is better suited to future studies when results from other deep polarization surveys or modelling efforts become available .
we did not detect any polarized sfgs in this work .
the fractional polarization upper limits for individual sfgs presented in the lower - right panel of fig . [ ch5:fig : fracpolclass ] indicate that characteristic @xmath37 levels for the sub - millijansky ( @xmath14 @xmath21 100 @xmath1jy ) sfg population are likely to be typically less than @xmath160 . given that sfgs are only expected to begin contributing significantly to the total intensity source counts at @xmath14 @xmath5 100 @xmath1jy ( e.g. see fig .
[ ch5:fig : fracpolclass ] ; see also * ? ? ?
* ; * ? ? ?
* ) , the limit above indicates that the @xmath93 model is unlikely to be affected by the presence of sfgs unless @xmath4 @xmath5 60 @xmath1jy .
our limit of @xmath161 is consistent with the fractional polarization distribution for 1.4 ghz sfgs predicted by @xcite ; see panel ( b ) of fig . 6 from their work . @xcite modelled the integrated polarized emission of spiral galaxies , finding typical fractional polarization levels of @xmath162 with overall mean level @xmath163 .
the @xmath93 number - counts predicted in this work are therefore likely to represent an upper limit to the true polarized number - counts at @xmath1jy levels , due to the diminished mean level of fractional polarization for faint radio sources with respect to the @xmath108 model .
more sensitive observations are required to detect polarized emission from faint sfgs and to quantify their polarization properties .
@xcite and @xcite discovered that double radio sources depolarize less rapidly with increasing wavelength on the side with the brighter ( or only ) radio jet than on the opposite side , providing strong evidence that the apparent one - sided nature of jets in otherwise symmetric radio galaxies and quasars is caused by relativistic beaming .
this ` laing - garrington ' effect is typically interpreted as being caused by orientation - induced path - length differences through a foreground , turbulent , magnetised intragroup or intracluster medium which surrounds the entire radio source , where the approaching side is seen through less of this medium . however ,
this interpretation is not unambiguous . the asymmetric depolarization effect may be contaminated or even dominated by depolarization internal to the lobes @xcite , a sheath mixing layer at the interface where relativistic and thermal plasmas meet @xcite , draping of undisturbed intracluster magnetic fields over the surface of a lobe expanding subsonically @xcite or supersonically @xcite , or by intrinsic asymmetries in local environment which act separately or in addition to orientation - induced depolarization @xcite . in reality
, it is likely that each of the mechanisms above may contribute , requiring a ` unification scheme ' to predict which will dominate for any given source .
for example , @xcite describes an emerging picture that differentiates between properties expected for fri and frii sources , and that includes an inner depolarization region associated with shells of dense thermal plasma around the radio jets in addition to the undisturbed intergalactic medium surrounding the source .
however , this picture does not yet include variations on the general orientation - induced depolarization effect due to source environment asymmetries , such as the correlation between lobe properties and optical line emission described by @xcite . if the laing - garrington effect is caused predominantly by source orientation , rather than asymmetries in source environment
, then we expect the fractional surface density of sources exhibiting asymmetric depolarization in a volume - limited sample to approximately relate to the fraction of randomly - oriented sources with @xmath13 @xmath21 10 mjy that are pointed towards earth .
we justify this expected relationship by noting that frii sources are dominated by un - beamed lobe emission rather than jet emission which may be beamed , frii sources dominate fri sources in flux - limited samples for @xmath13 @xmath21 10 mjy @xcite , and the median redshift for frii sources is @xmath164 with relatively small scatter @xcite such that a flux - limited sample may crudely approximate a volume - limited sample . for a jet lying within an angle @xmath165 to the line of sight , the fraction of randomly - oriented sources pointed towards earth is @xmath166 .
orientation schemes predict that the transition from quasars ( typically beamed ) to radio galaxies ( typically not beamed ) is expected to occur at @xmath167 @xcite .
the critical angle to induce asymmetric depolarization in a double radio source ( quasar or radio galaxy ) is likely to be similar ( e.g. * ? ? ?
* ) ; here we estimate this angle as ranging between @xmath168 and @xmath169 , implying fractional surface densities of @xmath170 amongst the general double radio source population .
this is , of course , a crude model , not least because intracluster magnetic field strengths at @xmath171 ( i.e. the expected median redshift for double radio sources ; * ? ? ?
* ) are only expected to be a few percent of their @xmath68 values .
a magnetised cluster atmosphere is clearly a prerequisite for depolarization , though a separate depolarizing medium close to the radio jets as described by @xcite may null this point .
as far as we are aware , no census of asymmetric depolarization has been performed for radio sources in a blind survey ; studies to date have typically compiled samples of targeted observations ( e.g. * ? ? ?
* ; * ? ? ?
the atlas data are suitable for this purpose .
we note that it is difficult to estimate what the fractional surface density of asymmetrically depolarized sources might be if environmental asymmetries were to dominate , rather than orientation ( significant merger activity is certainly expected in clusters ; e.g. * ? ? ?
therefore , we simply focus on whether the atlas data are consistent with an orientation scheme or not . to avoid selection effects relating to visibility area and the detectability of sources with low fractional polarization
, we selected only the 40 atlas sources with total intensity @xmath172 mjy .
we expect that each of these sources is a radio - loud agn with dual - lobe structure dominated by un - beamed lobe emission ( though not all need exhibit multiple components due to viewing angle and source size ) .
the breakdown of these 40 sources according to the polarization@xmath28total intensity classification scheme ( see [ sec:1 ] ) is as follows .
there was 1 polarized source classified as type 3 ( midway polarized ) , though it is unclear if this demonstrates a depolarization asymmetry or not .
there were 7 polarized single - component sources classified as type 4 . in an attempt to account for the possibility that many of these ` unclassified ' sources represent as - yet unassociated lobes of multi - component sources ( see comments in [ ch5:secresidentm ] ) , we assumed that perhaps only 1 of the detected type 4 sources was likely to truly represent a polarized single - component source .
we have interpreted this as a dual - lobed asymmetrically - depolarized source with jet axis close to the line of sight , such that only a single component is effectively seen .
there were 15 polarized dual - lobed sources classified as type 6 ( no asymmetric depolarization ) .
there were 2 polarized dual - lobed sources classified as type 7 ( clear indication of asymmetric depolarization ) ; for reference , these sources were displayed earlier in fig .
17 of paper i. there were 4 polarized dual - lobed sources classified as type 8 ( unclear whether asymmetrically depolarized or not ) .
we assumed that approximately one quarter of these sources ( i.e. 1 source ) would likely demonstrate asymmetric depolarization if more sensitive observations were obtained .
this assumption is consistent with the finding from fig .
[ ch5:fig : fracpoltheta2 ] that most type 8 sources have lass reflective of type 6 sources , rather than the smaller lass observed for the type 7 sources .
finally , our sample included 11 unpolarized sources ( type 9 ) , each with fractional polarization upper limits below 1% ( see fig .
[ ch5:fig : fracpolclass ] ) .
we do not interpret these sources as being asymmetrically depolarized .
we note that while it is possible that some of the 11 unpolarized sources represent unassociated lobes of multi - component sources , at least some of them must be truly isolated , single - component sources with @xmath36 .
for example , the brightest unpolarized source displayed in fig .
[ ch5:fig : fracpolclass ] is the source c3 , which is barely - resolved in atlas dr2 ( deconvolved angular size @xmath173 ) and has a @xmath174 ghz spectral index of @xmath175 .
this source is therefore consistent with identification as a cso @xcite ; csos are known to exhibit flat radio spectra @xcite and low fractional polarization due to strong depolarization @xcite .
the statistics above suggest that between @xmath176 and @xmath177 of the 40 atlas dr2 sources in our flux density limited sample exhibit depolarization asymmetry , i.e. , @xmath178 .
this fraction falls within the theoretical range estimated above , demonstrating that the laing - garrington effect appears consistent with orientation dependence , at least within the rudimentary confines of our analysis .
future high - resolution polarization studies are clearly required to form more robust conclusions .
in this work we have presented results and discussion for atlas dr2 .
our key results are summarised as follows .
for convenience we use the term ` millijansky ' loosely below to indicate flux densities in the range @xmath179 mjy .
a. radio emission from polarized millijansky sources is most likely powered by agns , where the active nuclei are embedded within host galaxies with mid - infrared spectra dominated by old - population ( 10 gyr ) starlight or continuum produced by dusty tori .
we find no evidence for polarized sfgs or individual stars to the sensitivity limits of our data - all polarized atlas sources are classified as agns .
b. the atlas data indicate that fractional polarization levels for sources with starlight - dominated mid - infrared hosts and those with continuum - dominated mid - infrared hosts are similar . c. the morphologies and angular sizes of polarized atlas components and sources are consistent with the interpretation that polarized emission in millijansky sources originates from the jets or lobes of extended agns , where coherent large - scale magnetic fields are likely to be present .
we find that the majority of polarized atlas sources are resolved in total intensity , even though the majority of components in linear polarization are unresolved .
this is consistent with the interpretation that large - scale magnetic fields that do not completely beam depolarize are present in these sources , despite the relatively poor resolutions of the atlas data .
d. we do not find any components or sources with fractional polarization levels greater than 24% , in contrast with previous studies of faint polarized sources .
we attribute this finding to our improved data analysis procedures .
e. the atlas data are consistent with a distribution of fractional polarization at 1.4 ghz that is independent of flux density down to @xmath180 mjy , and perhaps even down to 1 mjy when considering the upper envelope of the distribution .
this result is in contrast to the findings from previous deep 1.4 ghz polarization surveys ( with the very recent exception of * ? ? ?
* ) , and is consistent with results at higher frequencies ( @xmath6 ghz ) . the anti - correlation observed in previous 1.4 ghz studies is due to two effects : a selection bias , and spurious high fractional polarization detections .
both of these effects can become more prevalent at faint total flux densities .
we find that components and sources can be characterised using the same distribution of fractional linear polarization , with a median level of 4% .
we have presented a new lognormal model to describe the distribution of fractional polarization for 1.4 ghz components and sources , specific to agns , in surveys with resolution fwhms @xmath3 .
f. no polarized sfgs were detected in atlas dr2 down to the linear polarization detection threshold of @xmath181 @xmath1jy .
the atlas data constrain typical fractional polarization levels for the @xmath14 @xmath21 100 @xmath1jy sfg population to be @xmath161 .
g. differences between differential number - counts of components and of sources in 1.4 ghz surveys with resolution fwhm @xmath3 are not likely to be significant ( @xmath5 20% ) at millijansky levels . h. the atlas total intensity differential source counts do not exhibit any unexpected flattening down to the survey limit @xmath182jy . i. the atlas linearly polarized differential component counts do not exhibit any flattening below @xmath100 mjy , unlike previous findings which have led to suggestions of increasing levels of fractional polarization with decreasing flux density or the emergence of a new source population .
the polarized counts down to @xmath0 @xmath1jy are consistent with being drawn from the total intensity counts at flux densities where luminous fr - type radio galaxies and quasars dominate .
j. constrained by the atlas data , we estimate that the surface density of linearly polarized components in a 1.4 ghz survey with resolution fwhm @xmath3 is 50 deg@xmath110 for @xmath183 @xmath1jy , and 90 deg@xmath110 for @xmath184 @xmath1jy .
we estimate that the surface density for polarized sources is @xmath185 deg@xmath110 for @xmath186 @xmath1jy , assuming that most polarized components belong to dual - component sources ( e.g. fr - type ) at these flux densities .
k. we find that the statistics of atlas sources exhibiting asymmetric depolarization are consistent with the interpretation that the laing - garrington effect is due predominantly to source orientation within a surrounding magnetoionic medium . to our knowledge
, this work represents the first attempt to investigate asymmetric depolarization in a blind survey .
we thank walter max - moerbeck for insightful discussions .
we thank the anonymous referee for helpful comments that led to the improvement of this manuscript . c. a.
h. acknowledges the support of an australian postgraduate award , a csiro oce scholarship , and a jansky fellowship from the national radio astronomy observatory .
b. m. g. and r. p. n. acknowledge the support of the australian research council centre of excellence for all - sky astrophysics ( caastro ) , through project number ce110001020 . the australia telescope compact array is part of the australia telescope national facility which is funded by the commonwealth of australia for operation as a national facility managed by csiro .
this paper includes archived data obtained through the australia telescope online archive ( http://atoa.atnf.csiro.au ) . 99 arshakian t. g. , beck r. , 2011 , mnras , 418 , 2336 appleton p. n. , et al . , 2004 , apjs , 154 , 147 augusto p. , gonzalez - serrano j. i. , perez - fournon i. , wilkinson p. n. , 2006 , mnras , 368 , 1411 battye r. a. , browne i. w. a. , peel m. w. , jackson n. j. , dickinson c. , 2011 , mnras , 413 , 132 banfield j. k. , george s. j. , taylor a. r. , stil j. m. , kothes r. , scott d. , 2011 , apj , 733 , 69 barthel p. d. , 1989 , apj , 336 , 606 beck r. , gaensler b. m. , 2004 , newar , 48 , 1289 bicknell g. v. , cameron r. a. , gingold r. a. , 1990 , apj , 357 , 373 biggs a. d. , ivison r. j. , 2006 , mnras , 371 , 963 blake c. , wall j. , 2002 , mnras , 337 , 993 bondi m. , et al . , 2003 , a&a , 403 , 857 burn b. j. , 1966 , mnras , 133 , 67 cales s. l. , et al . , 2011 , apj , 741 , 106 condon j. j. , 2007 , aspc , 380 , 189 condon j. j. , et al . , 2012 , apj , 758 , 23 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 condon j. j. , 1988 , gera.book , 641 cotton w. d. , et al . , 2003 , pasa , 20 , 12 de zotti g. , massardi m. , negrello m. , wall j. , 2010 , a&arv , 18 , 1 dolag k. , bartelmann m. , lesch h. , 2002 , a&a , 387 , 383 dursi l. j. , pfrommer c. , 2008 , apj , 677 , 993 ensslin t. a. , vogt c. , clarke t. e. , taylor g. b. , 2003 , apj , 597 , 870 feain i. j. , et al . , 2009 , apj , 707 , 114 ferguson h. c. , et al .
, 2004 , apj , 600 , l107 fernini i. , 2001 , aj , 122 , 83 fine s. , shanks t. , nikoloudakis n. , sawangwit u. , 2011 , mnras , 418 , 2251 fixsen d. j. , et al . , 2011 , apj , 734 , 5 fomalont e. b. , kellermann k. i. , cowie l. l. , capak p. , barger a. j. , partridge r. b. , windhorst r. a. , richards e. a. , 2006 , apjs , 167 , 103 garrington s. t. , conway r. g. , 1991 , mnras , 250 , 198 garrington s. t. , leahy j. p. , conway r. g. , laing r. a. , 1988 , natur , 331 , 147 gendre m. a. , wall j. v. , 2008 , mnras , 390 , 819 ginzburg v. l. , syrovatskii s. i. , 1965 , ara&a , 3 , 297 gopal - krishna , wiita p. j. , 2004 , arxiv : astro - ph/0409761 grant j. k. , taylor a. r. , stil j. m. , landecker t. l. , kothes r. , ransom r. r. , scott d. , 2010 , apj , 714 , 1689 guidetti d. , laing r. a. , bridle a. h. , parma p. , gregorini l. , 2011 , mnras , 413 , 2525 guidetti d. , laing r. a. , croston j. h. , bridle a. h. , parma p. , 2012 , mnras , 423 , 1335 hales c. a. , gaensler b. m. , norris r. p. , middelberg e. , 2012a , mnras , 424 , 2160 hales c. a. , murphy t. , curran j. r. , middelberg e. , gaensler b. m. , norris r. p. , 2012b ,
mnras , 425 , 979 hales c. a. , et al . ,
2014 ( paper i ) , mnras , xxx , xxx heywood i. , jarvis m. j. , condon j. j. , 2013 , mnras , 432 , 2625 holder g. p. , 2014 ,
apj , 780 , 112 hopkins a. m. , afonso j. , chan b. , cram l. e. , georgakakis a. , mobasher b. , 2003 , aj , 125 , 465 huarte - espinosa m. , krause m. , alexander p. , 2011
, mnras , 418 , 1621 ibar e. , ivison r. j. , biggs a. d. , lal d. v. , best p. n. , green d. a. , 2009 , mnras , 397 , 281 ishwara - chandra c. h. , saikia d. j. , kapahi v. k. , mccarthy p. j. , 1998 , mnras , 300 , 269 jarvis m. j. , rawlings s. , 2004 , newar , 48 , 1173 johnson r. a. , leahy j. p. , garrington s. t. , 1995 , mnras , 273 , 877 kaplan e. l. , meier p. , 1958
, journal of the american statistical association , 53 , 457 kashyap v. l. , van dyk d. a. , connors a. , freeman p. e. , siemiginowska a. , xu j. , zezas a. , 2010 , apj , 719 , 900 kharb p. , odea c. p. , baum s. a. , daly r. a. , mory m. p. , donahue m. , guerra e. j. , 2008 , apjs , 174 , 74 kuchar p. , enlin t. a. , 2011 , a&a , 529 , a13 laing r. a. , 1988 , natur , 331 , 149 laing r. a. , bridle a. h. , parma p. , murgia m. , 2008 , mnras , 391 , 521 lara l. , mrquez i. , cotton w. d. , feretti l. , giovannini g. , marcaide j. m. , venturi t. , 1999 , a&a , 348 , 699 liu r. , pooley g. , 1991a , mnras , 249 , 343 liu r. , pooley g. , 1991b , mnras , 253 , 669 lonsdale c. j. , et al . , 2003 , pasp , 115 , 897 mao m. y. , et al . , 2012 , mnras , 426 , 3334 massardi m. , bonaldi a. , negrello m. , ricciardi s. , raccanelli a. , de zotti g. , 2010 , mnras , 404 , 532 massardi m. , et al .
, 2013 , mnras , 436 , 2915 mccarthy p. j. , van breugel w. , kapahi v. k. , 1991 , apj , 371 , 478 mesa d. , baccigalupi c. , de zotti g. , gregorini l. , mack k .- h . , vigotti m. , klein u. , 2002 , a&a , 396 , 463 middelberg e. , et al . , 2008 , aj , 135 , 1276 mitchell k. j. , condon j. j. , 1985 , aj , 90 , 1957 morrison g. e. , owen f. n. , dickinson m. , ivison r. j. , ibar e. , 2010 , apjs , 188 , 178 negrello m. , magliocchetti m. , de zotti g. , 2006 , mnras , 368 , 935 nenkova m. , sirocky m. m. , ivezi . , elitzur m. , 2008 , apj , 685 , 147 norris r. p. , et al . , 2006 ,
aj , 132 , 2409 norris r. p. , lenc e. , roy a. l. , spoon h. , 2012 , mnras , 422 , 1453 odea c. p. , 1998 ,
pasp , 110 , 493 osullivan s. , stil j. , taylor a. r. , ricci r. , grant j. k. , shorten k. , 2008 , in pos , proc .
vlbi network symp . for radio astron . ,
osullivan s. p. , et al .
, 2013 , apj , 764 , 162 overzier r. a. , rttgering h. j. a. , rengelink r. b. , wilman r. j. , 2003 , a&a , 405 , 53 owen f. n. , morrison g. e. , 2008 , aj , 136 , 1889 padovani p. , miller n. , kellermann k. i. , mainieri v. , rosati p. , tozzi p. , 2011
, apj , 740 , 20 pier e. a. , krolik j. h. , 1993 , apj , 418 , 673 readhead a. c. s. , taylor g. b. , pearson t. j. , wilkinson p. n. , 1996 , apj , 460 , 634 regener v. h. , 1951 , phys .
, 84 , 161l rudnick l. , blundell k. m. , 2004 , in reiprich t. , kempner j. , soker n. , eds , proc .
of the riddle of cooling flows in galaxies and clusters of galaxies , e37 rudnick l. , owen f. n. , 2014 , apj , in press ( arxiv:1402.3637 ) sajina a. , lacy m. , scott d. , 2005 , apj , 621 , 256 sajina a. , partridge b. , evans t. , stefl s. , vechik n. , myers s. , dicker s. , korngut p. , 2011 , apj , 732 , 45 sandage a. , lubin l. m. , 2001 , aj , 121 , 2271 scheuer p. a. g. , 1957 , pcps , 53 , 764 seiffert m. , et al . , 2011 , apj , 734 , 6 serkowski k. , 1958 , aca , 8 , 135 seymour n. , mchardy i. m. , gunn k. f. , 2004 , mnras , 352 , 131 shi h. , liang h. , han j. l. , hunstead r. w. , 2010 , mnras , 409 , 821 simmons j. f. l. , stewart b. g. , 1985 , a&a , 142 , 100 simpson c. , et al .
, 2006 , mnras , 372 , 741 singal j. , stawarz . ,
lawrence a. , petrosian v. , 2010 , mnras , 409 , 1172 smoli v. , et al . , 2008 , apjs , 177 , 14 sokoloff d. d. , bykov a. a. , shukurov a. , berkhuijsen e. m. , beck r. , poezd a. d. , 1998 , mnras , 299 , 189 stawarz . , et al . , 2013 , apj , 766 , 48 stepanov r. , arshakian t. g. , beck r. , frick p. , krause m. , 2008 , a&a , 480 , 45 stil j. m. , krause m. , beck r. , taylor a. r. , 2009 , apj , 693 , 1392 subrahmanyan r. , ekers r. d. , saripalli l. , sadler e. m. , 2010 , mnras , 402 , 2792 subrahmanyan r. , cowsik r. , 2013 , apj , 776 , 42 tasitsiomi a. , kravtsov a. v. , gottlber s. , klypin a. a. , 2004 , apj , 607 , 125 taylor a. r. , et al .
, 2007 , apj , 666 , 201 thorat k. , subrahmanyan r. , saripalli l. , ekers r. d. , 2013 , apj , 762 , 16 tolman r. c. , 1930 , pnas , 16 , 511 tribble p. c. , 1992 , mnras , 256 , 281 tribble p. c. , 1991 , mnras , 250 , 726 tucci m. , martnez - gonzlez e. , toffolatti l. , gonzlez - nuevo j. , de zotti g. , 2004 , mnras , 349 , 1267 tucci m. , toffolatti l. , 2012 , adast , 2012 , 624987 urry c. m. , padovani p. , 1995 ,
pasp , 107 , 803 van der marel r. p. , franx m. , 1993 , apj , 407 , 525 vernstrom t. , scott d. , wall j. v. , 2011 , mnras , 415 , 3641 white r. l. , becker r. h. , helfand d. j. , gregg m. d. , 1997 , apj , 475 , 479 wilman r. j. , et al .
, 2008 , mnras , 388 , 1335 windhorst r. a. , van heerde g. m. , katgert p. , 1984
, a&as , 58 , 1 wright e. l. , 2006 , pasp , 118 , 1711 xu h. , li h. , collins d. c. , li s. , norman m. l. , 2009 , apj , 698 , l14 zinn p .- c .
, middelberg e. , norris r. p. , hales c. a. , mao m. y. , randall k. e. , 2012 , a&a , 544 , a38
this appendix presents 1.4 ghz euclidean - normalised differential number - counts , in tabulated form , derived from the atlas dr2 total intensity and linear polarization component catalogues from appendix a of paper i. the tabulated results have been organised as follows according to emission type , eddington bias correction scheme , and atlas field .
tables [ tbl : countsicdfsie ] and [ tbl : countsielaisie ] ( each with 7 columns ) present total intensity number - counts for the cdf - s and elais - s1 fields , respectively , using ` component - corrected ' data whereby individual components were deboosted prior to the application of visibility area and resolution bias corrections .
similarly , tables [ tbl : countsicdfs ] and [ tbl : countsielais ] ( each with 9 columns ) present total intensity number - counts for the two atlas fields , but now using ` bin - corrected ' data whereby non - deboosted components were corrected for visibility area , resolution bias , and eddington bias . tables [ tbl : countslcdfsie ] and [ tbl : countslelaisie ] ( each with 11 columns ) present linear polarization number - counts for the cdf - s and elais - s1 fields , respectively , using component - corrected data
. tables [ tbl : countslcdfs ] and [ tbl : countslelais ] ( each with 17 columns ) present bin - corrected linear polarization number - counts for the cdf - s and elais - s1 fields , respectively
. columns for the tables above are organised as follows .
tables describing component - corrected total intensity data give for each bin the flux density range ( @xmath187 ) , expected average flux density ( @xmath69 ) , raw number of deboosted components ( @xmath188 ) , effective number of deboosted components following visibility area correction only ( @xmath189 ) , effective number of deboosted components following both visibility area and resolution bias corrections ( @xmath190 ) , euclidean - normalised counts following visibility area correction only ( @xmath191 ) , and euclidean - normalised counts following both visibility area and resolution bias corrections ( @xmath192 ) .
columns for tables describing bin - corrected total intensity data are similar , but without the superscript d which indicates use of deboosted component data .
the bin - corrected tables contain two additional columns : effective number of components following combined visibility area , resolution bias , and eddington bias corrections ( @xmath193 ) , and an associated column for their euclidean - normalised counts ( @xmath194 ) .
the tables describing linear polarization data are similar to those for total intensity data , but with additional columns cataloguing the effective number of components or euclidean - normalised counts resulting from the resolution bias corrections associated with the lower ( -r ) or upper ( + r ) bounds described in 7.1 of paper i and displayed in fig .
22 of paper i. thus the additional columns have been assigned descriptors with superscripts v - r , v+r , v - re , and v+re .
errors associated with the fully corrected euclidean - normalised counts in the figures and tables above are @xmath195 poissonian and were calculated following @xcite . | this is the second of two papers describing the second data release ( dr2 ) of the australia telescope large area survey ( atlas ) at 1.4 ghz . in paper
i we detailed our data reduction and analysis procedures , and presented catalogues of components ( discrete regions of radio emission ) and sources ( groups of physically associated radio components ) . in this paper
we present our key observational results .
we find that the 1.4 ghz euclidean normalised differential number counts for atlas components exhibit monotonic declines in both total intensity and linear polarization from millijansky levels down to the survey limit of @xmath0 @xmath1jy .
we discuss the parameter space in which component counts may suitably proxy source counts .
we do not detect any components or sources with fractional polarization levels greater than 24% .
the atlas data are consistent with a lognormal distribution of fractional polarization with median level 4% that is independent of flux density down to total intensity @xmath2 mjy and perhaps even 1 mjy .
each of these findings are in contrast to previous studies ; we attribute these new results to improved data analysis procedures .
we find that polarized emission from 1.4 ghz millijansky sources originates from the jets or lobes of extended sources that are powered by an active galactic nucleus , consistent with previous findings in the literature .
we provide estimates for the sky density of linearly polarized components and sources in 1.4 ghz surveys with @xmath3 resolution .
[ firstpage ] polarization radio continuum : galaxies surveys . |
new accelerators like cebaf , mami - b and cosy will investigate hadron observables at a scale intermediate to the low - energy region where hadron phenomena mainly reflect the underlying principles of chiral symmetry and its dynamical breakdown and to the high momentum region where the `` strong '' interaction has the appearance of being a perturbation on free - moving quarks and gluons .
a reasonable theoretical description of intermediate energy physics therefore has to satisfy at least the following requirements : formulated in terms of quarks and gluons it has to account for a mechanism of dynamical chiral symmetry breaking ( dcsb ) and identify the related goldstone bosons in the measured spectrum of quark bound states .
in addition , all the correlation functions have to transform properly under the renormalization group of quantum chromo dynamics ( qcd ) , _ i.e. _ at large momenta one should recover the correct anomalous dimensions .
only if these requirements are fulfilled the theoretical framework is fast in those grounds of the theory of strong interaction that are well established .
furthermore , it is desirable to formulate a microscopic picture of the cause of confinement . in this context growing interest
has been focused on the euclidean dyson
schwinger equations ( dse s ) as a tool for developping and applying non - perturbative methods in quantum field theories , a recent review on this subject is given in ref.@xcite .
realistic models of qcd derived through truncation of the dse tower are _ applicable in the whole euclidean momentum space _ and , among others , have the following features : quark condensation gives rise to dcsb , the chiral goldstone bosons are identified with the lowest mass pseudoscalar quark - antiquark bound states , perturbative qcd is matched at large momenta , and a study of the analytic structure of the quark propagator provides some insight into the mechanism of confinement .
the dse s include the bethe - salpeter equations ( bse s ) for bound states , especially the ones describing physical mesons @xcite .
the bse s are coupled to the dse s for the quark and gluon propagators , _
i.e. _ the bse s need as input at least the full renormalized quark and gluon propagators which for calculating the meson mass spectrum , for instance have to be known at complex euclidean momenta far away from the real axis . in principle
, the quark and gluon propagators can be determined from their dse s . as these are coupled to the infinite tower of dse s
one is not able to solve the equations exactly .
possible approximation schemes are based on the truncation of the dse tower : some @xmath1-point functions like the gluon propagator and the quark - gluon vertex are parametrized choosing _ anstze _ such that certain requirements are fulfilled .
important constraints are imposed by the discrete and continuous symmetries of qcd most of which are formulated in ward or slavnov - taylor identities .
additional input may be taken from hadron phenomenology . in this paper
we investigate the analytic structure of the quark propagator using the euclidean dse for the quark self energy in rainbow approximation where the full quark - gluon vertex is replaced by the perturbative one .
the main ingredient for our study is a color diagonal gluon propagator whose transverse part vanishes at zero four momentum : @xmath2 herein @xmath3 denotes the ( spacelike ) gluon momentum and @xmath0 is an energy scale to be determined .
this form of the gluon propagator is motivated by very distinct considerations .
one is based on the complete elimination of gribov copies within a lattice formulation of non - abelian gauge theories@xcite : landau gauge is fixed uniquely with the help of a ` thermodynamic parameter ' .
the resulting functional integral is dominated by field configurations on the gribov horizon .
this implies that the gluon propagator vanishes for small momenta @xmath3 as @xmath4 .
in a couple of different studies a gluon propagator of the form ( [ stingl1 ] ) is obtained by considering a generalized perturbative expansion of the seven superficially divergent proper vertices of qcd which are allowed to depend non - analytically on the coupling constant @xcite .
furthermore , the field strength approach ( fsa ) to qcd supports the form ( [ stingl1 ] ) for being the dominant part of the infrared quark - quark interaction@xcite ; however , the color structure of the fsa gluon propagator is not diagonal . at last
, some recent lattice calculations obtained a gluon propagator which allow a fit of the form ( [ stingl1 ] ) ( but which do not rule out a fit using a standard massive particle propagator)@xcite .
as the gluon propagator ( [ stingl1 ] ) has no poles for timelike momenta it may be argued that it describes confined gluons .
instead , the gluon propagator has poles on the imaginary axis , @xmath5 , and therefore the gluon ` decays ' after some time @xmath6 .
this can be interpreted as the gluons being screened . on the other hand ,
the gluon propagator ( [ stingl1 ] ) is smaller than the perturbative one for all momenta : @xmath7 .
this has drastic consequences for dcsb as we will show : for the values of @xmath0 that are suggested by the lattice studies of ref .
@xcite chiral symmetry will be realized in the nambu - goldstone mode for unrealistic large coupling constants only .
we also find that quarks are not confined in this model , at least for a large range of model parameters @xmath0 .
a similar study was undertaken by hawes et al.@xcite . in this work the gluon propagator ( [ stingl1 ] ) and
a non - trivial quark - gluon vertex has been used to calculate the quark propagator in the spacelike region .
their results are similar to ours , especially they find unconfined quarks .
in contrast to our work where the analytic continuation to timelike momenta is performed explicitly and a pole signaling free massive quasi quarks as asymptotic states is found , hawes et al . use a test for confinement employing a fourier transformation of the quark schwinger functions .
the large time behavior of these fourier transforms indicate asymptotic states of massive deconfined quarks . in sec . 2 we present our model dse for the quark self energy
we also discuss the _ ansatz _ eq .
( [ stingl1 ] ) for the gluon propagator in more detail . in sec . 3 we present the numerical solution for the quark self energy at real euclidean momenta and discuss the dynamical breaking of chiral symmetry . in sec . 4
we describe the analytic continuation of the quark self energy with special emphasis on its pole structure and the wick rotation from euclidean to minkowski space . in sec .
5 we conclude .
our starting point is the dse for the qcd quark two - point function in euclidean momentum space which can be written in terms of the quark self energy @xmath8 .
after renormalization it reads @xcite : @xmath9 with @xmath10 for three colors and @xmath11 denoting the qcd coupling constant .
@xmath12 is the current quark mass .
@xmath13 , @xmath14 and @xmath15 are renormalization constants depending on a regularization parameter @xmath16 and getting fixed at the renormalization point @xmath17 . the three green s functions that appear on the r.h.s . of eq .
( [ sderen ] ) are * 1 . * the quark propagator , @xmath18 with @xmath19 and @xmath20 denoting the scalar and vector parts of the quark self energy @xmath21 , respectively , and @xmath22 and @xmath23 the field renormalization function and the mass function , respectively ; * 2 . *
the gluon propagator @xmath24 where only the transverse piece , @xmath25 , is dressed due to vacuum polarization and * 3 . * the quark - gluon vertex @xmath26 .
of course , the gauge parameter @xmath27 appearing in the gluon propagator ( [ defgluon ] ) is a function of the renormalization point @xmath28 , _
i.e. _ any redefinition of the renormalization point leads to a new gauge .
the only gauge which is not affected by renormalization is the fix point @xmath29 , _ i.e. _ landau gauge @xcite . in our studies
we therefore concentrate on this case . for studying eq .
( [ sderen ] ) one has to put in the full gluon propagator and the full quark - gluon vertex ; an exact investigation of the quark dse ( [ sderen ] ) therefore requires solving their dse s , too .
to do so the gluon 3 and 4-point functions , the ghost self energy , the quark - quark scattering kernel , and so on , have to be known .
each of them again fulfills its own dse .
we do not address the problem of solving all these dse s simultaneously but instead truncate the infinite tower of dse s by making _
anstze _ for both the gluon propagator as well as the quark - gluon vertex . neglecting ghost contributions to the gluon vacuum polarization the transverse part of the gluon propagator ( [ defgluon ] ) , @xmath30 ,
is closely related to the running coupling constant of qcd , @xmath31 : @xmath32 in addition , @xmath33 in landau gauge ; this is a result of the ward identity that connects the quark - gluon vertex to the inverse of the quark propagator .
taking into account one - loop corrections to the qcd perturbation expansion one possible smooth interpolation between ir and uv behavior is the following _ ansatz _ for the gluon propagator : @xmath34 with @xmath35 the qcd scale parameter and @xmath36 the coupling strength which is related to the anomalous dimension of the quark mass , @xmath37@xcite : @xmath38 .
however , in our numerical studies we will treat @xmath36 as an adjustable parameter .
furthermore , we set @xmath39 to ensure that the whole strength of the infrared ( ir ) quark - quark interaction is carried by the function @xmath40 . in the following
we will call the second factor on the r.h.s.of eq .
( [ uvgluon ] ) ` ultraviolet ( uv ) improvement ' .
note that in eq .
( [ uvgluon ] ) the four - momentum @xmath3 is assumed to be spacelike .
we like to mention that its continuation to large complex momenta is by no means fixed by the behavior in the asymptotic spacelike region .
for example one could replace @xmath41 without altering the spacelike uv behavior .
indeed , both expressions are quite similar even for small spacelike four - momenta . on the other hand ,
their analytic properties are very different .
while the function on the left hand side has its cut on the timelike axis , , the one on the right hand side has cuts for . of course , one can use a two- or a three - loop order running coupling of qcd instead of the uv - improvement in eq .
( [ uvgluon ] ) .
nevertheless , even these can not provide information about the structure of the gluon propagator at small spacelike momenta ( @xmath42 2 gev ) .
hence , for proceeding a parametrization of @xmath43 is unavoidable . on the parametrization of the ir part of the gluon propagator
a great deal of work has been done during the last years .
several parametrizations have been investigated and have supported the opinion that quark confinement is closely related to infrared slavery or at least to a strong finite quark - quark interaction in the infrared momentum regime . nevertheless , as remarked in the introduction , there
exist very distinct studies @xcite , @xcite ( using landau gauge ) and @xcite that propose @xmath44 with a dynamical mass scale @xmath0 assumed to be proportional to @xmath45 .
in addition , refs .
@xcite even argue that @xmath46 .
this propagator has no klln
lehmann representation @xcite indicating that the propagation of a free quasi gluon is not possible . as the strength of the quark - quark interaction vanishes at small momenta
( corresponding to large distances in coordinate space ) the solution of the quark dse ( [ sderen ] ) with the gluon propagator ( [ stingl ] ) as input would exhibit a new mechanism for quark confinement _
if any_. for simplicity we use the rainbow approximation , _ i.e. _ we formally replace the dressed by the perturbative quark - gluon vertex in eq .
( [ sderen ] ) : @xmath47 in this approximation the solutions of the bethe - salpeter equations ( bse s ) for color - singlet quark - antiquark bound states show the correct low energy behavior as formulated in current algebra and partial conserved axial current ( pcac ) relations .
for instance , the quark bound states carrying the quantum numbers of the pseudoscalar mesons can be identified with the goldstone bosons related to the dynamical breakdown of the @xmath48 chiral symmetry of massless qcd @xcite ; in case of massive qcd the pion mass is determined by a generalized gell - mann oakes renner relation .
the rainbow and the separable approximation @xcite are the only known approximations that imply the foregoing consistent formulation of quark dse and bse for mesons .
furthermore , the rainbow approximation ( [ rainbow ] ) is equivalent to the stationary phase approximation of the global color - symmetry model ( gcm ) @xcite , an abelianized version of qcd with quark fields being transformed to collective fields composed of quarks and anti - quarks and carrying the quantum numbers of mesonic fields . after inserting eqs .
( [ stingl ] ) and ( [ rainbow ] ) ( together with or without eq .
( [ uvgluon ] ) ) into eq .
( [ sderen ] ) , using eq .
( [ z1=z2 ] ) and fixing the renormalization point , @xmath28 , such that vector and scalar part of the quark self energy obey the renormalization conditions @xmath49 with constants @xmath50 and @xmath51 being fixed later , we get two integral equations which couple the functions @xmath20 and @xmath19 that now depend on the momentum @xmath52 , the renormalization point @xmath28 , and the regularization parameter @xmath53 : @xmath54 ( s,\mu^2;\lambda^2_{\rm uv})}{1-j_a [ a , m ] ( \mu^2,\mu^2;\lambda^2_{\rm uv})}\ ; , \label{a } \\
b(s,\mu^2;\lambda^2_{\rm uv } ) & = & \beta + \alpha\ , \frac{j_b [ a , m ] ( s,\mu^2;\lambda^2_{\rm uv})-j_b [ a , m ] ( \mu^2,\mu^2;\lambda^2_{\rm uv } ) } { 1-j_a ( \mu^2,\mu^2,\lambda^2_{\rm uv } ) } \label{b}\end{aligned}\ ] ] with @xmath55 ( s,\mu^2;\lambda^2_{\rm uv } ) } \nonumber\\ & = & \frac{\lambda}{6\pi^3}\int_0^{\lambda^2_{\rm uv } } \frac{dr}{a(r,\mu^2;\lambda^2_{\rm uv})[r + m^2(r,\mu^2;\lambda^2_{\rm uv } ) ] } \int_{-1}^1 dx i_a ( x;r , s)\;\xi ( x;r , s ) \label{ja } \ ; , \\[.5ex ] \lefteqn{j_b [ a , m ] ( s,\mu^2;\lambda^2_{\rm uv})}\nonumber\\ & = & \frac{\lambda}{6\pi^3}\int_0^{\lambda^2_{\rm uv } } \frac{dr m(r,\mu^2;\lambda^2_{\rm uv})}{a(r,\mu^2;\lambda^2_{\rm uv})[r + m^2(r,\mu^2;\lambda^2_{\rm uv } ) ] } \int_{-1}^1 dx i_b ( x;r , s)\;\xi ( x;r , s)\label{jb}\end{aligned}\ ] ] where @xmath56^{-1 } & \mbox{with uv improvement , eq.~(\ref{uvgluon } ) } \ , . \end{array}\right . \label{uvimprovement}\end{aligned}\ ] ] the functions @xmath57 and @xmath58 are defined in the appendix , eqs .
( [ ia ] ) and ( [ ib ] ) , where also the analytic expressions for the angle integrals ( @xmath59-integrals ) in case of @xmath60 are given ( eqs .
( [ iaex ] ) and ( [ ibex ] ) ) . the mass function in eqs .
( [ a ] ) and ( [ b ] ) reads : @xmath61 + \alpha\left[j_b(s,\mu^2;\lambda^2_{\rm uv } ) - j_b(\mu^2,\mu^2;\lambda^2_{\rm uv})\right]}{\alpha\left [ 1-j_a ( s,\mu^2,\lambda^2_{\rm uv})\right]}\ ; .
\label{mass}\end{aligned}\ ] ] to make contact with perturbative qcd one has to formulate an appropiate renormalization condition ensuring the asymptotic freedom of quarks .
it is convenient to choose the renormalization point as large as possible and set @xmath62 at this point .
the largest available momentum is ; therefore we set : if not stated otherwise .
numerically , it turns out that @xmath63 ( \mu^2,\mu^2;\lambda^2_{\rm uv})$ ] is very small for @xmath28 being large enough .
actually , in the limit it vanishes exactly .
hence , solving the dse ( [ a ] ) and ( [ b ] ) for the renormalized quark self energy is ( within numerical accuracy ) identical to solving the equations @xmath64 ( s,\mu^2;\lambda^2_{\rm uv } ) \ ; , \label{a0}\\ b(s,\mu^2;\lambda^2_{\rm uv } ) & = & \widehat{m}(\mu^2 ) + j_{b } [ a , m ] ( s,\mu^2;\lambda^2_{\rm uv } ) \ ; , \label{b0}\end{aligned}\ ] ] when imposing the renormalization condition @xmath65 @xmath66 is a running current mass @xmath67 ( \mu^2,\mu^2;\lambda^2_{\rm uv } ) \ ; .\ ] ] we like to remark that in case of the dse system ( [ a0],[b0 ] ) this condition is sufficient because the approximation sets the renormalization constant .
for the system ( [ a],[b ] ) @xmath68 has to be fixed dynamically and hence two constraints , eqs .
( [ rena ] ) and ( [ renb ] ) , are needed . at large spacelike momenta @xmath69 the ir mass scale @xmath0 in the gluon propagator ( [ stingl ] )
can be neglected and the integral equations ( [ a ] ) and ( [ b ] ) with ( _ i.e. _ no uv improvement ) can be converted into differential equations @xcite .
as for large @xmath69 the integral @xmath63 ( s,\mu^2,\lambda_{\rm uv}^2)$ ] becomes very small the vector part of the quark self energy is approximatly equal to 1 and hence .
the scalar part @xmath19 is then determined by the non - linear differential equation @xmath70 with the boundary condition formulated in eqs .
( [ renb],[renbuv ] ) .
if there exists a bounded solution one can drop the term @xmath71 in the denominator and one gets the asymptotic solution @xmath72 \right\ } \label{mtail}\end{aligned}\ ] ] with @xmath73 .
@xmath74 is ( up to a trivial color factor ) the critical coupling of quantum electro dynamics ( qed ) . implying the uv boundary condition ( [ renbuv ] ) one gets @xmath75 the parameter @xmath76 is uniquely fixed by the ir boundary condition _
i.e. _ by the solution of the integral equation ( [ b ] ) which determines the slope of @xmath19 at the renormalization point @xmath28 .
in case of the uv improved model as defined by the gluon propagator ( [ uvgluon ] ) the asymptotic behavior of the mass function @xmath77 differs qualitatively from eq .
( [ mtail ] ) .
an analysis based on the renormalization group yields @xcite : @xmath78^{-\lambda/4\pi^2}}{\left[\log\left(s/ \lambda_{\rm qcd}^2 \right ) \right ] ^{1-\lambda/4\pi^2 } } \ , .\end{aligned}\ ] ] compared to the hadronic mass scale @xmath79 the up and down quark current masses @xmath12 are negligible . in the limit of zero current masses , @xmath80 ,
the qcd lagrangian is invariant under chiral transformations .
the wigner - weyl realization of the vacuum state corresponds to the trivial solution of eqs .
( [ a ] ) and ( [ b ] ) : @xmath81 in qed in landau gauge this trivial solution _ maximizes _ the cjt action @xcite whose stationary phase condition is identical to the quark dse ( [ sderen ] ) @xcite .
this statement holds also true for abelianized qcd as long as @xmath82 .
hence , a vacuum configuration with @xmath83 will be dynamically favored if @xmath82 .
this may be tested by evaluating the cjt action at the stationary points .
a nontrivial solution of eqs . ( [ a ] ) and ( [ b ] ) ( in the case @xmath80 ) indicates the dynamical breaking of chiral symmetry , _ i.e. _ the corresponding vacuum state is realized in the nambu - goldstone mode . from the definition of the quark condensate @xmath84 \stackrel{m=0}{= } -12 \int \frac{d^4p}{(2\pi)^4 } \frac{b(p^2)}{a^2(p^2 ) p^2 + b^2(p^2 ) } \label{quarkcond}\end{aligned}\ ] ] it is obvious that a non - vanishing scalar part of the self energy , @xmath85 , leads necessarily to a non - vanishing condensate @xmath86 .
the scalar part of the quark self energy at zero momentum , @xmath87 , or , correspondingly , the mass value @xmath88 is therefore an order parameter of dcsb . the analytic continuation of eqs .
( [ a ] ) and ( [ b ] ) to complex momenta is very important in the studies of dse s for the following reasons : * 1 . * a quark propagator whose mass function obeys the relation @xmath89 and whose renormalization function @xmath22 is non - vanishing at @xmath90 describes a massive particle which should be detectable in some appropriate experiment .
hence , one way to obtain information about quark confinement is the examination of the pole structure of the quark propagator after continuing real euclidean to imaginary momenta , _
i.e. _ @xmath69 to @xmath91 : if there does not exist any momentum @xmath92 for which eq .
( [ mc ] ) is fulfilled , there is no klln - lehmann representation for the quark propagator and the described quasi - quark is confined .
* 2 . * in order to solve the quark bound state equations ( _ i.e. _ the bse s for mesons or the faddeev equations for baryons ) information about the quark self energy in a large domain of the complex momentum plane is needed . while external momenta ( _ e.g. _ the momentum of a meson close to its mass - shell ) can be timelike the loop momentum occurring in the bse is spacelike . hence , there appear combinations of spacelike and timelike momenta as arguments of the quark self energy functions @xmath20 and @xmath19 ; the analytic continuation of these functions to complex momenta is therefore unavoidable if one wants to describe hadrons others than pions .
studying the analytic properties of the quark self energy provides some insight into the connection of the euclidean formulation of qcd and qcd in minkowski space .
recently it has been shown @xcite that in models containing both gluon and quark confinement the wick rotation can not be performed navely indicating thereby the close relation between confinement and singularities or branch cuts appearing in the quark propagator at some complex euclidean momenta .
we have solved eqs .
( [ a ] ) and ( [ b ] ) by iteration with fixed cutoff @xmath93 .
different parametrizations and grid densities of the discretized momentum interval @xmath94 $ ] have been used .
all the results we report herein do not depend on changes of the grid parametrization or further increasements of the grid density . to check the accuracy of the numerical procedure we first study the asymptotic behavior of the dse - model without uv - improvement , _
i.e. _ we set in eqs .
( [ a ] ) and ( [ b ] ) . at large renormalization points
we fix the mass function arbitrarily and fit a tail of the form ( [ mtail ] ) in the momentum region allowing all parameters @xmath76 , @xmath95 and @xmath96 to get varied . in fig . 1
the function @xmath97 is shown for @xmath98 , @xmath99 , @xmath100 and @xmath101 together with an asymptotic fit ( [ mtail ] ) with and @xmath102 a comparison with the calculated values and shows high accuracy for the numerical solution of eqs .
( [ a ] ) and ( [ b ] ) . for ( and )
the value of the mass function at zero momentum increases linearily with increasing renormalization point , @xmath103 a scaling behavior that is well known from studies of four - dimensional qed @xcite .
the vector part of the quark self energy , @xmath20 , tends toward unity while the scalar part @xmath19 admits an almost constant value of order @xmath104 up to . obviously , the dse - model defined by eqs .
( [ stingl ] ) and ( [ rainbow ] ) is just an ir regularised version of four - dimensional qed for @xmath105 . modifying the asymptotic behavior of the gluon exchange by taking into account
the renormalization group ( rg ) improved loop - corrections , cf.eq .
( [ uvgluon ] ) , introduces a third mass scale : @xmath79 .
hence , no scaling comparable to that of eq .
( [ scaling ] ) occurs anymore . instead
, the quark self energy functions @xmath20 and @xmath19 show the proper rg transformation properties . for several parameter pairs @xmath106
we have studied the large momentum behavior of the quark mass function @xmath23 and have found a very good agreement with an asymptotic analysis based on the rg .
this is illustrated in fig . 2 where the calculated mass function for , , and is shown and an asymptotic tail of the form @xmath107^{d-1}\end{aligned}\ ] ] with @xmath108 is fitted .
the coupling constant @xmath36 extracted from this uv tail coincides within 5% with the value inserted into the dse ( ) demonstrating good numerical accuracy .
( note that @xmath109 is the exponent of a logarithmic term . )
the quark condensate in the model characterized by the above parameter set can be calculated from eq .
( [ andim ] ) together with eq .
( [ paraas ] ) ; its value is @xmath110 we have used mev @xcite . the same result is obtained calculating the trace of the quark propagator , eq .
( [ quarkcond ] ) : . in fig .
3 the order parameter @xmath88 is shown as a function of the coupling constant @xmath36 . for vanishing current mass , , we have studied the dependence on the renormalisation point @xmath104 ( here set equal to the cutoff @xmath93 ) choosing the ir mass scale @xmath0 equal to @xmath79 as suggested by refs .
. the critical behavior of the order parameter , of the solution in the chiral limit gets softened if the current quark mass is non - vanishing , _ i.e. _ we find the well - known behavior of an order parameter of dcsb .
while increasing the regularization parameter @xmath93 as well as the renormalization point @xmath104 the critical coupling @xmath96 decreases slightly .
the extrapolation @xmath111 yields a value of the critical coupling of @xmath112 . as in case of @xmath113
the vector part of the quark self energy , @xmath20 , is quite close to unity , the nambu - goldstone phase of dcsb is dynamically favored .
the critical exponent extracted from the data is @xmath114 .
this value is almost identical to @xmath115 hawes _ et al . _
have derived in their study of the dse system ( [ a0],[b0 ] ) inserting several _ dressed _ quark - gluon vertices@xcite .
the critical coupling strongly depends on the ir scale @xmath0 : small values of @xmath116 lead to small , large ones to large critical coupling constants as one might expect . for @xmath117
the ir scale that yields a critical coupling identical to the coupling suggested by the qcd asymptotics , @xmath118 ( @xmath119 ) , is @xmath120 .
therefore only for very small ir mass parameters @xmath121 mev dcsb takes place . in fig .
4 the mass functions for different ir parameters @xmath0 are shown for @xmath122 .
the parameter @xmath0 is varied by two orders of magnitude ( for @xmath123 no dcsb occurs ) .
it is remarkable that neither the strength of dcsb as `` measured '' by the order parameter @xmath88 nor the momentum range that contributes dominantly to the mechanism of dcsb depend on the ir scale @xmath0 introduced by the effective gluon exchange , _
i.e. _ the scales of dcsb and the gluon mass scale @xmath0 are not or only loosely connected to each other if a realistic coupling constant @xmath36 is used .
to study quark confinement we will analyze the quark propagator for timelike momenta . therefore the euclidean quark self energy has to be continued from spacelike to timelike momenta .
correspondingly , the variable @xmath52 has to be continued to negative values , @xmath124 .
( [ a ] ) and ( [ b ] ) can be understood as the defining equations for the vector and scalar parts of the quark self energy , @xmath20 and @xmath19 , respectively .
therefore the analytic continuation can be performed by smoothly changing the external momenta @xmath69 while keeping the loop momenta @xmath125 spacelike .
thus one generates the quark self energy as a function of complex momenta by simply using complex @xmath69 in eqs .
( [ a ] ) and ( [ b ] ) .
obviously , @xmath126 and @xmath127 and hence solving the dse in the upper half of the complex plane is sufficient .
note that the functions @xmath128 and @xmath129 inserted into the right hand side of eqs .
( [ a ] ) and ( [ b ] ) are still those determined self - consistently on the spacelike momentum axis .
first we generate the functions @xmath20 and @xmath19 on a small strip including the positive real axis .
there we find that @xmath20 and @xmath19 change continuously .
no singularities appear and therefore the analytical continuation can be performed patching open sets towards timelike momenta , .
problems arise caused by the poles of the angular integrals .
those are kinematical singularities that appear for where with .
they are integrable as can be seen from the analytic expressions ( [ iaex ] ) and ( [ ibex ] ) of the appendix , for instance , or from a laurent series expansion of the kernels ( [ rena ] ) and ( [ renb ] ) .
nevertheless , the numerical results for the domain have been quite unstable against a change of numerical parameters .
hence we restrict our discussion on results for only .
both functions , @xmath20 and @xmath19 , are entire on this domain .
the identification of a mass pole in the quark propagator fullfilling condition ( [ mc ] ) is only possible if .
this implies that in the case of dcsb our numerical method of identifying such a pole is only reliable for ir mass scale parameters @xmath0 larger than the created dynamical mass @xmath88 . in this restricted parameter space
we find that the mass shell condition ( [ mc ] ) for the quark is fulfilled for both models , with and without uv improvement .
_ i.e. _ close to the momentum @xmath130 quarks propagate like stable particles with constituent mass @xmath131 . for parameter values we are not able to make any statements about the existence of a quark mass pole
. in figs .
5a and 5b the mass function @xmath77 in the uv - improved model ( [ uvgluon ] ) for @xmath132 and @xmath133 , resp . , are shown .
the parameters used are such that the kinematical singularities are far away from the timelike momentum region considered here .
one recognizes that the mass condition ( [ mc ] ) is fullfilled and that the pole mass @xmath134 is very well approximated by the dynamical mass @xmath88 .
our result of quark _ deconfinement _ ( obtained for such values of the ir parameter @xmath0 where our numerical results are stable ) is in full agreement with the results of hawes _ et al .
_ they have investigated the scalar part of the quark schwinger function , @xmath135 , analyzed its asymptotic behavior for large euclidean times @xmath136 , and extracted the mass @xmath137 of a stable asymptotic fermion state through the relation @xcite @xmath138 = -m_{\rm as}\ , .\ ] ] @xmath137 is found to be ( positive ) finite and of the order of @xmath79 . while we are using the rainbow approximation ( [ rainbow ] ) for the quark - gluon vertex hawes _
_ employ non - trivial vertices .
these are constructed in a way that they obey the ward - takahashi identity , carry no kinematical singularities , go versus the free vertex for free fields , and ensure the propagators transforming properly under the landau - khalatnikov transformations @xcite .
in addition , one of the vertices used in ref .
@xcite leads to multiplicative renormalizability @xcite .
our study confirms the result of unconfined quarks obtained for an ir vanishing quark quark interaction @xcite . we may also deduce that in this model the form of the quark
gluon vertex is of minor importance ( at least , as long as it is free of singularities ) . in order to study the analytic properties of the quark propagator we employ cauchy
s integral theorem in the following way .
suppose that the kernels @xmath139 of the integrals @xmath140 and @xmath141 ( cf .
( [ ja ] ) and ( [ jb ] ) ) are analytic . then the integral of @xmath142 over a closed contour in the complex momentum plane has to vanish .
due to the renormalization condition ( [ mass ] ) the functions @xmath142 fall off faster than @xmath143 for large @xmath125 where .
therefore one obtains @xmath144 where the difference between the left hand side and the right hand side vanishes for .
this relation implies that the iteration of eqs .
( [ a ] ) and ( [ b ] ) along a rotated axis should lead to the same values @xmath145 and @xmath87 as the iteration along the spacelike axis as long as the kernels @xmath142 are analytic .
this provides us with an analyticity test for the functions @xmath142 .
the gluon propagator ( [ stingl ] ) is symmetric under the exchange of spacelike and timelike momenta : @xmath146 even though the uv - improved version of this gluon exchange does not carry the symmetry ( [ stinglsym ] ) an uv - improvement with the replacement ( [ replacement ] ) does .
as the spacelike properties ( such as dynamical chiral symmetry breaking ) are only slightly affected by the replacement ( [ replacement ] ) we believe our symmetry considerations being quite general and we concentrate on the uv - unimproved form ( [ stingl ] ) for the gluon propagator alone . in the following we study the approximate dse system ( [ a0],[b0 ] ) and we drop the dependencies on regularization parameter @xmath93 and renormalization point @xmath104 . assuming @xmath147 and @xmath148 being analytic within the whole complex momentum plane the iterations of eqs .
( [ a0 ] ) and ( [ b0 ] ) along the timelike axis @xmath149 should lead to the same values @xmath145 and @xmath87 as the iteration along the spacelike axis .
obviously , @xmath150 are solutions of eqs .
( [ zeita ] ) and ( [ zeitb ] ) if @xmath20 and @xmath19 are solutions of eqs .
( [ a0 ] ) and ( [ b0 ] ) . in the nambu - goldstone realization of the vacuum ,
, the solutions ( [ zeitloes ] ) do not obey the analyticity condition and .
therefore the kernels @xmath147 and @xmath148 are _ not _ analytic in the whole complex momentum plane in case of chiral symmetry being broken dynamically .
note that a `` usual '' mass pole of the quark propagator is sufficient to explain the behavior ( [ zeitloes ] ) .
the functions @xmath147 and @xmath148 are products of the vector or scalar part of the quark propagator and the angular integral over euclidean projections of the gluon propagator .
we will argue in the following that the non - analyticity of both functions are caused by the quark propagator . for doing this we study the wigner
weyl phase of the system , . in that case
the self consistency equation for reads : @xmath151 with @xmath152 and @xmath153 being the real and the imaginary part of , respectively .
it is straightforward to proof the following symmetries @xmath154 hence it is sufficient to analyze the integral equation ( [ atriv ] ) for only . in fig .
6 the function @xmath155 is shown for and various angles @xmath156 .
we have used a small cut - off , . however , the solutions do not depend on a further increasement of @xmath157 as the chiral symmetric solutions do not display the scaling behavior ( [ scaling ] ) as in the case of the nambu - goldstone phase
. the numerical error in fig .
6 is about 1% to 5% ; it is largest close to the imaginary axis .
it is obvious that for all the angles @xmath156 the real parts @xmath152 at zero momentum ( ) are identical ; the imaginary parts which are not shown vanish for small momenta within the numerical accuracy .
for the iteration of eq .
( [ atriv ] ) does not converge due to the ( integrable ) singularities of the gluon propagator .
nevertheless there is no signature of any further non - analyticity close to the imaginary axis .
we therefore conclude that there is a strong indication for the vector part of the quark self energy being analytic .
then the quark propagator has only one pole , namely the one at . in the nambu - goldstone phase , , the integral kernels @xmath147 and @xmath148 from eqs .
( [ fa ] ) and ( [ fb ] ) are not holomorphic on the whole complex momentum plane either .
there has to be a pole contribution from the contour integration in eq .
( [ ana ] ) . as the angular integrals over @xmath57 and @xmath58 do not carry a pole singularity the scalar and the vector parts of the quark propagator have at least one pole .
if this singularity is not lying on the real axis but within the complex plane the wick rotation can not be performed navely . if it is placed on the timelike axis the nave wick rotation is allowed as long as the causal behavior of the quark propagator is suitably determined .
we have not found any singularities except the kinematical ones and the mass pole on the real axis .
even though we are not able to draw compulsory conclusions this numerical result strongly indicates that the quark propagator is analytic within the whole complex momentum plane except on the real timelike axis .
this statement is also supported by the analysis within different dse models @xcite where it is found that poles of the quark propagators are lying quite close to the real axis .
in addition , the analyticity condition ( [ zeitloes ] ) is quite easily understood from a physical point of view . under the action of the wick rotation
the euclidean mass term becomes times the corresponding minkowski term , a kinetic fermionic term keeps `` unmodified '' . in terms of the quark propagator functions
this reads : @xmath158 where denotes the euclidean and the corresponding minkowski four momentum squared close to a mass pole of the quark propagator . in minkowski space
we therefore just recover one of the `` solutions '' ( [ zeitloes ] ) which is not surprising . summarizing this section
, there are strong indications that the euclidean quark propagator of the dse model specified by the gluon exchange ( [ stingl ] ) can be continued back to minkowski space . within a small region around the mass pole ( [ mc ] )
the minkowskian quark propagator has a klln - lehmann representation and it therefore resembles the propagator of a massive stable fermion . for a restricted set of parameters we have been able to verify this numerically .
due to the numerical problems associated with the kinematical singularities we are not really able to prove this statement rigorously for all values of the parameter @xmath0 .
nevertheless , it is the most plausible conclusion drawn from the properties of the functions @xmath147 and @xmath148 as discussed in this section .
we have investigated the euclidean dyson - schwinger equation ( dse ) for the quark self energy for an infrared vanishing gluon propagator in rainbow approximation .
we have found that this model is not appropriate for describing hadron phenomenology due to several reasons .
first , dynamical chiral symmetry breaking ( dcsb ) occurs either for unphysically large coupling constants only or for a small ir scale , @xmath159 .
the strength and the scale of dcsb are then independent of @xmath0 : the spontaneous breakdown of chiral symmetry is mainly driven by the term added to obtain the correct uv behavior and _ not _ by the infrared part .
in addition , continuation to timelike momenta reveals that the quark propagator possesses a pole . as our numerical method
is only reliable for ir mass scale parameters @xmath0 larger than the created dynamical quark mass ( see the discussion in subsection [ subsecmass ] ) this pole can be unambigously identified for such values of @xmath0 only . to be explicit , we do not claim that there is no pole , we only state that within our method it can not be found even if it exists .
however , the occurrence of a mass pole in the propagator can be related to the propagation of a stable particle .
therefore it is clear that quarks are not confined in this model , at least for a large range of the ir parameter @xmath0 .
this is in contradiction to results obtained in ref .
there heavy quarkonia have been studied using a non - relativistic reduction of the gluon propagator ( [ stingl ] ) together with a quark propagator and a quark - gluon vertex which result from the dse studies of ref .
. however , in this work the quark - quark interaction ( [ stingl ] ) has been replaced by a coulomb potential .
the resulting equation for the quarkonium bound states is just a lippmann - schwinger equation for hydrogen with the exception that the non trivial quark propagators of ref .
@xcite have been used .
these propagators are then probably the reason for the `` confining mechanism '' described in ref .
. stated otherwise , the unusual behavior of the quarks have been put in `` by hand '' . such a mechanism is not supported by our calculations . instead , our results are in accordance with those of hawes et al .
@xcite who have studied the quark propagator for spacelike momenta only and have extracted the quark mass from the large time behavior of the schwinger functions .
additionally , they have used non - trivial quark - gluon vertices .
the good quantitative agreement of our results with those of ref .
@xcite shows that the rainbow approximation is a good approximation for this model interaction .
this can be understood from the fact that the influence of a non - trivial quark - gluon vertex is concentrated in the infrared which for the studied case is suppressed by the infrared vanishing gluon exchange .
even though a study of the analyticity of the quark propagator is hampered by numerical difficulties we have found convincing evidence for the quark propagator being analytic within the whole complex euclidean momentum plane except the timelike ( negative real ) axis .
this has two consequences : first , it is possible to perform the nave wick rotation .
second , and more important , the model does _ not _ lead to quark confinement .
we take this opportunity to express our gratitude to prof .
h. reinhardt for his support and his interest in this work .
we thank dr . c. d. roberts for numerous helpful discussions .
ab got benefit from the courses and meetings of the graduiertenkolleg `` hadrons and nuclei '' ( dfg contract mu 705/3 ) ; therefore he especially thanks prof .
h. mther for organising these meetings .
he also gratefully acknowledges financial support by the studienstiftung des deutschen volkes .
this work has also been partly supported by cosy under contract 41170833 .
the angular integrands in eqs .
( [ a ] ) and ( [ b ] ) are : @xmath160 for @xmath161 in eqs .
( [ a ] ) and ( [ b ] ) ( _ i.e. _ no uv improvement for the quark - quark interaction ) the x - integral can be evaluated exactly : @xmath162 with @xmath163 [email protected] , + [email protected] c.d
. roberts and a.g .
williams , _ dyson - schwinger equations and their application to hadronic physics _ , progress in particle and nuclear physics , vol .
33 , pp .
477 - 575 , ed .
a. fler , pergamon press , oxford , 1994 .
baryons are described by relativistic faddeev equations .
d. zwanziger , nucl .
b * 364 * , 127 ( 1991 ) .
u. hbel _ et al .
_ , z. phys . a atomic nuclei * 336 * , 423 and 435 ( 1990 )
+ m. stingl , phys . rev
. d * 34 * , 3863 ( 1986 ) ; erratum _ ibid . d _ * 36 * , 651 ( 1987 ) .
+ m. stingl , _ a systematic extended iterative solution for quantum chromdynamics _ , universitt mnster preprint ms - tpi-94 - 13
. l. v. smekal , ph.d .
thesis , tbingen university , 1995 . c. bernard , c. parrinello , and
a. soni , phys .
d * 49 * , 1585 ( 1994 ) .
hawes , c.d .
roberts , and a. g. williams , phys .
d * 49 * , 4683 ( 1994 ) .
we use negative semidefinite euclidean metric , @xmath164 , @xmath165 @xmath166 . for abbreviation
, we nevertheless define the square of a momentum as .
all euclidean dirac matrices , @xmath167 , are antihermitean and obey the anticommutation relations @xmath168 .
w. marciano and h. pagels , phys .
rep . * 36 * , 137 ( 1978 ) .
l. p. pascual and r. tarrach , _ qcd : renormalization for the practitioner _ , lecture notes in physics * 194 * , springer
verlag , berlin - heidelberg - new york , 1984 .
r. delbourgo and m. d. scadron , j. phys .
g * 5 * , 1631 ( 1979 ) . c. d. roberts , private communication .
r. t. cahill and c. d. roberts , phys .
d * 32 * , 2419 ( 1985 ) .
h. d. politzer , nucl .
b * 117 * , 397 ( 1976 ) .
f. j. yndurin , _ the theory of quark and gluon interactions _ , springer
verlag , berlin - heidelberg - new york , 1993 .
j. d. bjorken and s. d. drell , _ relativistic quantum fields _ , mcgraw hill , new york , 1965 .
j. m. cornwall , r. jackiw , and e. tomboulis , phys .
d * 10 * , 2428 ( 1974 ) . c. d. roberts and r. t. cahill , phys .
d * 33 * 1755 ( 1986 ) .
d. atkinson and p. w. johnson , phys .
d * 35 * 1943 ( 1987 ) .
r. fukuda and t. kugo , nucl .
* b117 * , 250 ( 1976 ) . s. j. stainsby and r. t. cahill , phys .
a * 146 * , 467 ( 1990 ) ; int .
j. mod .
phys . a * 7 * 7541 ( 1992 )
. + p. maris and h. holties , int .
j. mod .
phys . a * 7 * , 5369 ( 1992 ) .
particle data group , phys .
d * 50 * , 1173 ( 1994 ) .
l. d. landau and i. m. khalatnikov , sov .
jetp * 2 * , 69 ( 1956 ) .
d. c. curtis and m. r. pennington , phys .
d * 42 * , 4165 ( 1990 ) .
m. becker _
et al . _ ,
b * 267 * , 261 ( 1991 ) . | the dyson - schwinger equation for the quark self energy is solved in rainbow approximation using an infrared ( ir ) vanishing gluon propagator that introduces an ir mass scale @xmath0 .
there exists a @xmath0 dependent critical coupling indicating the spontaneous breakdown of chiral symmetry . if one chooses realistic qcd coupling constants the strength and the scale of spontaneous chiral symmetry breaking decouple from the ir scale for small @xmath0 while for large @xmath0 no dynamical chiral symmetry breaking occurs . at timelike momenta
the quark propagator possesses a pole , at least for a large range of the parameter @xmath0 .
therefore it is suggestive that quarks are not confined in this model for all values of @xmath0 .
furthermore , we argue that the quark propagator is analytic within the whole complex momentum plane except on the timelike axis .
hence the nave wick rotation is allowed . * to appear in phys .
rev .
d * |
different carbon - based structures are considered as promising electrode material for field emission ( fe ) cathodes . in particular ,
the field emission properties of single - walled ( swnt ) and multi - walled ( mwnt ) carbon nanotubes @xcite as well as graphite films @xcite are presently under intensive experimental and theoretical investigations . in experiment , many factors such as inhomogeneities at the cathode surface , surface contamination ( surface adsorbates and oxides ) , local electric fields and barriers , electronic structure of cathode , etc . can drastically change fe results @xcite .
in addition , these factors vary from one experiment to another thus markedly complicating the theoretical description .
nevertheless , the electronic characteristics of cathodes should be equally manifested in different experiments .
for this reason , the effect of electronic structure on the emission features of cathodes is of definite interest . for swnts
this problem was studied numerically in @xcite by using an approach which can be called as a method of _ independent channels_. in this paper , we present a rather simple modification of this method to study analytically the influence of the 3d band structure on the field emission current ( fec ) of crystalline graphite . as an illustration
, we consider the case of swnt where our approach allows us to reproduce the fe results obtained in @xcite .
as is well - known , the electronic structure near the fermi energy of the crystalline graphite markedly depends on the weak interlayer interaction ( see , e.g. , @xcite ) .
accordingly , the fec in this case should be sensitive to the specific electronic structure . in order to show this
, we consider the simplest possible modification of graphite ( hypothetical @xmath0 stacking ) where the three - dimensional energy spectrum was calculated analytically in @xcite .
two possible orientations of the applied electric field ( along and normal to the graphite layers ) are of our interest .
the emitted current density can be written as @xcite @xmath1 here the field emission is directed along the @xmath2-axis , @xmath3 is the electric charge , @xmath4 the planck constant , @xmath5 the energy , @xmath6 momentum , @xmath7^{-1}$ ] the fermi - dirac distribution function , @xmath8 the transmission probability of an electron through a potential barrier , and @xmath9 the group velocity .
the integrals are over the first brillouin zone with account of the positivity of @xmath10 .
notice that in most cases the using of infinite limits in integrals is a good approximation . for parabolic bands
@xmath11 , and this relation is widely used in deriving the well - known fowler - nordheim equation . in the case of carbon nanotubes ,
two important differences from the generally accepted consideration should be taken into account .
first , an open swnt has a finite small radius which results in quantization of momentum . in this case
, the corresponding integrals in ( [ main ] ) transform into sums .
for example , choosing the axes like shown in fig .
[ figure1 ] is the circumference of the nanotube , @xmath12 is a thickness of the graphite layer.,width=151 ] one has @xmath13 where @xmath14 , and @xmath15 is the circumference of the nanotube , @xmath12 the thickness of the graphite layer . for a swnt there exists only a single layer in the @xmath16-direction and , accordingly , there is exactly one term in the sum for @xmath17 .
the number of terms for @xmath18 depends on the tube circumference @xmath15 .
second , the energy near the fermi level for a single graphite layer ( graphene ) is approximated by @xmath19 , where @xmath20 is the fermi velocity @xcite .
the electrons move in the @xmath21 plane , so that the energy does not depend on @xmath22 . since @xmath23 is quantized the energy @xmath24 turns out to be divided into a set of channels with @xmath25 where @xmath26 takes integer values .
therefore , the current density ( [ main ] ) takes the form @xmath27 as is known , the dispersion relation for carbon nanotubes depends on their chirality ( see , e.g. , @xcite ) . for a chiral vector @xmath28
it can be written as @xmath29^{1/2}. \label{eheh}\ ] ] generally , there are two symmetric curves with a gap @xmath30 . however , at certain values of @xmath31 , @xmath32 and @xmath26 the gap turns out to be zero and one gets the linear dispersion relation . therefore , at fixed @xmath26 there exist one metallic branch and a set of semiconducting branches for a swnt with a given chirality . as stated above , the condition @xmath33 imposes restrictions on the limits of integral in ( [ off ] ) .
in addition , two approximations will be used .
first , we consider the zero - temperature limit when the fermi - dirac distribution becomes the step function .
second , we suggest that the transmission probability is given by the wkb approximation ( see , e.g. , @xcite ) in the form @xmath34\right\ } = b\exp(d\varepsilon)\ ] ] where @xmath35 , @xmath36 , @xmath37 is the electric field , @xmath38 the work function , @xmath39 the dielectric constant , and we denoted @xmath40 and @xmath41 for convenience .
the functions @xmath42 and @xmath43 describe a deviation of the barrier from the triangle form due to image effects and can be approximated by @xcite @xmath44 now we are able to calculate the @xmath26th term in the sum ( [ off ] ) . for the metallic branch the integration in ( [ off ] ) spreads from @xmath45 to @xmath46 .
one obtains @xmath47 for semiconducting branches the range of integration in ( [ off ] ) is ( @xmath45 , @xmath48 ) and @xmath49 notice that the dispersion relation enters ( [ eee ] ) only through the gap .
this agrees with the well - known fact that the group velocity and the density of states are canceled in the one - dimensional case @xcite .
the sum over all branches in ( [ off ] ) gives the total fec .
[ fig : tube ] shows the calculated emission current density , which is a current divided by the circumference of a nanotube .
the parameter set is @xmath50 ev , @xmath51 m / s , @xmath52 , @xmath53 ev@xmath54 m@xmath55.,width=226 ] for @xmath56 swnt the circumference is defined as @xmath57 with @xmath58 being the lattice constant .
in fact , the main contribution to the sum in ( [ off ] ) comes from the first few terms corresponding to branches close to the fermi level .
this is due to the exponential dependence of the fec on the gap . for metallic nanotubes ,
the leading term is @xmath59 , so that @xmath60 . in the case of semiconducting nanotubes
, the leading contribution comes from the @xmath26th term with the smaller gap in ( [ eee ] ) and , therefore , @xmath61 $ ] .
a similar behavior was found numerically in @xcite .
moreover , comparing our results in fig . [
fig : tube ] with the exact numerical calculations in @xcite one can find out a good qualitative agreement .
notice that the quantitative difference is also not great and varies from a few to ten percent depending on the diameter of the nanotube .
in comparison with @xcite our points in fig .
[ fig : tube ] are situated slightly lower for metallic nanotubes and slightly higher for semiconducting nanotubes .
this difference can be explained by at least two reasons .
first , we have used the simplified expression for the tunneling probability in ( [ dd ] ) where the image effects were approximated in a standard way ( see , e.g. , @xcite ) .
second , as distinct from @xcite we consider the zero - temperature limit .
in this section , we study a case of noninteracting graphite layers .
the layers are oriented as shown in fig .
[ fig : orient ] . to calculate the fec
we will use the method of independent channels described in the previous section .
namely , let us consider the 2d graphite lattice with the born - von karman boundary conditions applied in the @xmath62 direction .
this gives the natural quantization conditions . for layers of a large ( infinite )
size the sum in ( [ off ] ) can be replaced by the integral @xmath63 , so that finally one obtains @xmath64 where the relation @xmath65 is taken into account .
it is interesting to mention that this result is very similar to the fowler - nordheim formula @xmath66 indeed , the exponents are exactly the same and the preexponential factors differ only slightly .
what is important , the @xmath67-dependence is equal in both cases . for the interlayer distance @xmath68
one can estimate @xmath69 .
generally , there are known three possible configurations of crystalline graphite : @xmath70 stacking sequence of hexagonal layers ( bernal structure ) , rhombohedral @xmath71 stacking , and @xmath72 stacking when layers of carbon atoms are located directly on top of each other @xcite .
the @xmath0 stacking is called hypothetical because it has not been observed yet in crystalline graphite . however , this configuration is expected in disordered or pregraphitic carbon @xcite . in this paper , we consider the model of @xmath0 stacking which is the simplest one and allows us to study the effect of interlayer interaction on the emission properties analytically .
let us consider the interacting graphite layers oriented parallelly to the electric field ( see fig .
[ fig : orient ] ) . in the framework of the @xmath0 model
the interaction modifies the 3d band structure near the fermi level which can be written in a simple analytical form @xcite @xmath73 here the @xmath74 parameter represents the interaction between first - neighboring atoms in a layer , @xmath75 is related to the interaction between two atoms of the same projection on the @xmath21 plane , from two neighboring layers , and @xmath76 is the interlayer spacing . the influence of other parameters @xmath77 and @xmath78 introduced in @xcite is suggested to be negligible and only linear terms in @xmath79 perturbation expansion are taken into account .
actually , the analysis in @xcite shows that the maximum effect of the next - to - leading term ( breaking the cylindrical symmetry ) is of the order of five percent .
the upper sign in ( [ 3ds ] ) corresponds to the conduction band , and the lower sign corresponds to the valence band , @xmath80 . in compliance with ( [ 3ds ] )
[ fig : fermi1 ] represents the fermi surface of the @xmath0 graphite .
the fermi surface is composed of a hole pocket ( the valence band , sigh `` minus '' in ( [ 3ds ] ) ) and two half pockets of electrons ( sign `` plus '' in ( [ 3ds ] ) ) . in our case , the emission occurs along the z - axis .
generally , the possible values of the momentum of emitting electrons with respect to the fermi surface can be collected into five different groups .
we call them _ independent channels_. solid lines in fig .
[ fig : fermi1 ] indicate five possible types of independent channels : ( 1 ) for an intermediate electron - hole region , ( 2,3 ) for holes , and ( 4,5 ) for electrons . for finite - size layers ,
quantization of momentum in the @xmath81 plane occurs . in this case
, the spectrum is written as @xmath82 where @xmath83 are integer , and @xmath84 lies in the region @xmath85 .
the total fec is a sum of all channels .
let us consider these contributions separately .
the hole region is defined as @xmath86 . as is shown in fig .
[ fig : fermi1 ] there are two types of channels for the hole region and the channel 1 can be considered as an intermediate case .
[ holechannel ] shows all possible one - dimensional dispersion relations for this case .
the channel 3 crosses the fermi surface at two points while the channel 2 does not cross the fermi surface . for each channel @xmath87 .
the integration here is over the occupied states , and @xmath88 is the distance from the fermi level to the extremum point of the branch ( see fig .
[ holechannel ] ) .
for the channel 1 @xmath89 , and the current density is found to be equal to @xmath59 .
notice that there are only two channels of this type . analytically , the channel 2 is defined as @xmath90 , where @xmath91 . for this channel
one obtains @xmath92 here @xmath93 takes the form @xmath94 , which can be easily found from ( [ 3dsij ] ) . replacing the sum @xmath95 by the integral one gets @xmath96 d p_y = \frac{2eb}{h^2d^2 c\upsilon_f}.\ ] ] for the channel 3 one has @xmath97 and @xmath98 .
like before , one obtains @xmath99 the electron region is defined as @xmath100 and @xmath101 or , taken into account a periodicity of the brillouin zone , @xmath102 .
there are two kinds of channels in the electron region with spectra shown in fig .
[ electronch ] .
the channel 4 does not cross the fermi surface and , therefore , the current density turns out to be zero . for the channel 5 one has @xmath103 .
this channel contains occupied states below the fermi level and , hence , there is nonzero contribution to the fec .
one gets @xmath104,\ ] ] where @xmath105\}d
p_y \nonumber\\ = \frac{2 eb[4 d \alpha_1/\pi + \bigl(\textbf{i}_0(2 d\alpha_1 ) - \textbf{l}_0(2 d \alpha_1 ) - 1\bigr ) ] } { d^2 c \upsilon_f h^2},&\end{aligned}\ ] ] where @xmath106 is the modified struve function , and @xmath107 is the modified bessel function .
the total current density is found to be @xmath108}{h^2d^2 c \upsilon_f } .\ ] ] notice that @xmath109 reduces to ( [ g - j ] ) for @xmath110 .
let us consider interacting graphite layers placed normally to the electric field ( see fig .
[ fig : orient ] ) .
this situation differs markedly from the previous case .
let us denote @xmath111 in the dispersion relation in ( [ 3ds ] ) .
quantization of momentum results in the replacement @xmath112 .
there are only two types of channels in this case ( see fig .
[ channelsnormal ] ) .
-direction.,width=226 ] as before , let us consider them separately .
the hole region is defined by @xmath113 .
there are two kinds of channels in the hole region with spectra shown in fig .
[ fig : holech ] .
the channel 1 is defined by @xmath114 .
the current density reads @xmath115,\ ] ] where @xmath116 .
one obtains @xmath117\ } 2 \pi \rho d \rho \\
\nonumber = \frac{4 \pi e b}{(d h)^3 \upsilon_f^2}\left(\frac{(2\alpha_1 d)^2}{2 } - 1 + \exp(- 2\alpha_1 d)(2\alpha_1 d + 1 ) \right).\end{aligned}\ ] ] the channel 2 is defined by @xmath118 , and @xmath119 finally , @xmath120
2\pi\rho d\rho \\
\nonumber = \frac{4 \pi e b(2\alpha_1 d + 1)}{(d h)^3 \upsilon_f^2 } \exp(-2\alpha_1 d)(\exp(2\alpha_1 d ) - 1).\end{aligned}\ ] ] in the electron region @xmath121 .
there are also two kinds of channels in this region with spectra shown in fig .
[ fig : electronch ] .
the channel 3 is equivalent to the channel 5 in the previous section .
it is defined by @xmath114 .
the current is @xmath104.\ ] ] here @xmath122 ( see fig . [
fig : electronch ] and ( [ 3dsij ] ) . in this case
, @xmath123\ } 2 \pi \rho d \rho\\ = \frac{4 \pi e b}{(d h)^3 \upsilon_f^2}\left(\frac{(2\alpha_1 d)^2}{2 } - 2\alpha_1 d + \exp(2 \alpha_1 d ) - 1 \right).\end{aligned}\ ] ] the channel 4 is defined by @xmath118 . since it does not cross the fermi surface the current density is equal to zero .
the total current density is the sum of all channels @xmath124.\ ] ]
[ fig : jf ] shows the calculated current densities as functions of the applied electric field . for comparison ,
the fowler - nordheim curve is drawn . ) and normally ( @xmath125 ) to the electric field with @xmath126 ev , @xmath127 ev . for comparison the fawler - nordheim curve ( @xmath128 ) is shown.,width=226 ] the most important difference comes from preexponential factors .
as is known , the fowler - nordheim theory gives @xmath129 at all @xmath37 . in accordance with ( [ result ] )
the preexponential factor has a different field - dependence . at small
@xmath37 one obtains @xmath130 .
when @xmath37 increases ( which means @xmath131 ) the current density comes to @xmath132 and , finally , @xmath133 as was shown in section [ nonint ] .
indeed , at large @xmath16 the difference @xmath134 tends to @xmath135 while at small @xmath16 it tends to @xmath136 @xcite .
this is clearly seen in fig .
[ fig : all ] where the comparative curves are demonstrated .
for @xmath125 we have a similar behavior . according to ( [ result2 ] ) ,
@xmath137 at small @xmath37 , and @xmath138 at large @xmath37 , so that @xmath139 tends to a constant with increasing @xmath37 .
one can conclude that the bigger is the electric field the lesser is the role of the interlayer interaction .
the anisotropy of the emission from the 3d graphite is also shown in fig .
[ fig : all ] . as is seen , @xmath140 , that is almost a constant in the considered interval of @xmath37 .
therefore , we obtain three times increase in fec when graphite layers are oriented in parallel with the electric field . is almost a constant in the considered interval of @xmath37 .
, width=6 ] it is interesting to discuss the dependence of fec from the parameter @xmath75 which characterizes the interlayer interaction .
it was found in @xcite that this parameter is very sensitive to the interlayer distance . based on their results
one can approximate @xmath141 where @xmath142 and @xmath75 is measured in ev .
, @xmath143 for two field orientations .
@xmath125 is markedly more sensitive to @xmath16 than @xmath144.,width=226 ] as is seen from fig .
[ interl ] , there is a strong dependence of the fec from the interlayer distance .
the less is this distance the more is the emission current .
this is valid for both orientations
. it would be interesting to check this finding in experiments with graphite crystals under pressure .
notice that this result follows from the fact that the dos at the fermi level ( which is of the most importance in the emission process ) is determined by @xmath75 ( see @xcite ) . as is seen from fig . , [ interl ] , @xmath125 is more sensitive to @xmath16 than @xmath144 .
moreover , for @xmath145 one has @xmath146 , which follows from the fact that the movement of electrons between layers is suppressed in the absence of the interlayer interaction .
in conclusion , we have found that the band structure of the 3d graphite has a marked impact on the field emission current .
experimentally , the field emission from carbon materials was studied in @xcite .
unfortunately , the polycrystalline carbon films used in experiment can not be properly described in the framework of our approach because for this purpose we have to consider a mixture between different crystalline structures .
in fact , the fermi surface of the @xmath147 structure of graphite is found to be much more complex and , in particular , it does not possess the cylindrical symmetry ( see , e.g. , @xcite ) . in this case , our approach should be markedly modified .
besides , many additional factors like the presence of a diamond - like phase on the surface of samples and the absence of any information about the local electric fields do not allow us to clarify the role of the band structure in this case .
therefore , specific emission experiments with graphite single crystals at different orientations of the electric field would be of evident interest . this work has been supported by the russian foundation for basic research under grant no .
05 - 02 - 17721
jonge b n and bonard j 2004 _ phil .
r. soc . lond .
a _ * 362 * 2239 obraztsov a n , volkov a p and pavlovskii i yu 1998 _ jetp letters _ * 68 * 59 modinos a 1984 _ field - , thermionic - and secondary electron emission spectroscopy _ ( new york : plenum press ) shi - dong l and xu n s 2003 _ app .
lett . _ * 83 *
1213 shi - dong l , huang , n y , deng s z and xu n s 2004 _ app .
lett . _ * 85 * 813 xinghui l , changchun z and yukui l 2003 _ physica b _ * 344 * 243 wallace p r 1947 _ phys . rev . _ ,
* 71 * 622 slonczewski j c 1958 _ phys .
rev . _ * 109 * 272 mcclure j w 1964 _ ibm .
_ * 8 * 255 charlier j c , michenaud j p , gonze x and vigneron j p 1991 _ phys .
b _ * 44 * 13237 saito r , fujita m , dresselhaus g , and dresselhaus m s _ app .
lett . _ * 60 * 2204 gadzuk j w and plummer e w 1973 _ rev .
* 45 * 487 edgcombe c j 2002 _ phil . mag .
b _ * 82 * 1009 hawkes p w and kasper e 1989 _ principles of electron optic vol . 2 _ ( london : academic press ) abramowitz m , stegun i 1984 _ handbook of mathematical function _ ( new york : dover publication ) * 12.2.6*. charlier j c , michenaud j p and gonze x 1992 _ phys . rev .
b _ * 46 * 4531 | the field emission of crystalline @xmath0 graphite is studied within a simple analytical approach with account of the exact dispersion relation near the fermi level .
the emission current is calculated for two crystal orientations with respect to the applied electric field .
it is found that the exponent of the fowler - nordheim equation remains the same while the preexponential factor is markedly modified . for both field directions
, the linear field dependence is found in weak fields and the standard quadratic fowler - nordheim behavior takes place in strong fields . a strong dependence of the emission current from the interlayer distance is observed . as an illustration of the method the known case of a single - walled carbon nanotube
is considered . |
a particular interest has been devoted during the last decade to the quantum algebra @xmath2 @xcite .
this algebra is generated by three operators @xmath3 and @xmath4 , also named the @xmath0-angular momentum components .
they have the following commutation relations : @xmath5~=~\pm~l_\pm,\ ] ] @xmath6~=~\left[2~l_0\right],\ ] ] where the quantity in square brackets is defined as @xmath7~=~{q^n - q^{-n}\over q - q^{-1}}.\ ] ] in the most general case the deformation parameter @xmath0 is an arbitrary complex number and the physicist considers it as a phenomenological parameter @xcite . when @xmath8 , the quantum algebra @xmath2 , which defines a @xmath0-analogue of the angular momentum , reduces to the lie algebra @xmath9 of the ordinary angular momentum .
it is therefore interesting to investigate @xmath0-analogues of dynamical systems and to look for new effects when @xmath10 .
this has been first achieved for the harmonic oscillator by using algebraic methods , as e.g. in refs .
taking , for example , @xmath11 with @xmath12 a real , positive quantity , one can find that the distance between subsequent levels of the @xmath0-harmonic oscillator decreases when the excitation increases .
this is a desired property in describing rotational bands of deformed nuclei @xcite .
however the accidental degeneracy of the harmonic oscillator persists in this treatment .
another , more appealing way to introduce @xmath0-analogues of simple dynamical systems , is through deriving a @xmath0-deformed schrdinger equation . in this vein
several attempts have been made for the harmonic oscillator , as for example in refs.@xcite , for an attractive coulomb potential @xcite or for both potentials @xcite .
this procedure leads to the removal of the accidental degeneracy whenever it exists . here
we follow the approach of ref .
@xcite where a @xmath0-deformed schrdinger equation has been derived for a general central potential and the exact solution for the particular cases of the coulomb and the harmonic oscillator potentials have been obtained . the crucial starting point in ref .
@xcite was the search for a hermitian realization of the position , momentum and angular momentum operators , all behaving as _ vectors _ with respect to @xmath2 algebra .
this allowed the construction of an angular momentum operator entering the expression of the hamiltonian .
its components are different from the generators of the @xmath2 algebra . in the case of central potentials ( spinless particles )
the eigenfunctions of the @xmath0-deformed angular momentum have been derived as @xmath0-deformed spherical harmonics and then closed expressions for the eigenvalues of the @xmath0-deformed schrdinger equation have obtained as a function of @xmath0 .
this study is devoted to the derivation of a @xmath0-deformed spin - orbit coupling , consistent with the approach of ref .
there an angular momentum @xmath13 ( @xmath14 ) has been defined as a @xmath0-_vector _ with respect to the @xmath2 algebra ( [ def1])-([def2 ] ) . by analogy , here we introduce a spin operator @xmath15 which is a @xmath0-_vector _ in the algebra of @xmath16 analogue to ( [ def1])-([def2 ] ) .
next a hermitian spin operator @xmath17 is constructed as a @xmath0-_vector _ in the coproduct algebra of the total angular momentum @xmath18 by using a unitary version of the universal @xmath19 matrix .
the spin - orbit interaction is defined as a @xmath0-scalar hermitian operator in the space of @xmath18 and its matrix elements are calculated exactly for the representations @xmath1 . in previous applications of the @xmath0-deformed algebras to physical systems , as for example ref .
@xcite , the spin - orbit coupling is derived in a different way , based on a boson realization of the @xmath20 algebra @xcite . there
the spin operator does not form a vector in the coproduct algebra . accordingly
the eigenvalues of the spin - orbit operator are different from ours . in the next section
we summarize the findings of ref . @xcite . in section
iii a @xmath0-analogue of the spin - orbit coupling is derived . in section
iv we calculate numerically the spectra of the @xmath0-harmonic oscillator and the @xmath0-coulomb potentials without and with a spin - orbit contribution .
physical implications are discussed .
we stress that we do not aim at a particular fit of the deformation parameter to describe some particular system but at modelling physical systems through @xmath2 algebra .
the last section is devoted to some closing remarks .
in this section we follow closely ref .
the hamiltonian entering the @xmath0-deformed schrdinger equation is @xmath21 here and in the following we shall take @xmath22 the eigenfunctions of this hamiltonian are @xmath23 where @xmath24 are the normalized @xmath0-spherical harmonics ( 56 ) and ( 57 ) of ref .
@xcite , depending of the deformation parameter @xmath0 and @xmath25 .
they are related to @xmath0-hypergeometric functions @xcite .
the function @xmath26 satisfies the following radial equation @xmath27 + v_0(r)\right\ } r^lu_l(r ) = e_{n\ell}~r^lu_l(r)~,\ ] ] where @xmath28 is the non - negative solution of @xmath29}{\left[2\right ] } \frac{\left[2 \ell + 2\right]}{\left[2\right ] } + c_{\ell}^2 - c_{\ell}~,\ ] ] with @xmath30}~.\ ] ] it then follows that for the coulomb potential @xmath31 the eigenvalue is @xmath32 and for the harmonic oscillator potential @xmath33 the eigenvalue is @xmath34 @xmath35 being in both cases the radial quantum number . the spectrum is degenerate with respect to the magnetic quantum number @xmath36 but the accidental degeneracy typical for the undeformed equation is removed both for the coulomb and the harmonic oscillator potentials when @xmath10 . from eq .
( [ cl ] ) it follows that for @xmath37 one has @xmath38 .
thus for @xmath37 the only non - negative solution of ( [ llplus1 ] ) is @xmath39 , for all deformations . as a consequence ,
the @xmath37 levels are independent of the deformation parameter both for the harmonic oscillator and the coulomb potentials .
the centrifugal barrier disappears and taking @xmath40 one reobtains the free particle case , as for undeformed equations . for @xmath41
it is useful to distinguish between two different types of deformation parameter : @xmath42 in this case one can easily prove that @xmath43 so that eq .
( [ llplus1 ] ) has real solutions .
therefore to each non - zero @xmath44 corresponds a positive @xmath28 which is no more an integer .
we found it interesting to use real @xmath0 for the coulomb potential , as shown in sec .
the other case is : @xmath45 in this case for small values of @xmath12 one can find numerically that real positive values of @xmath28 exist .
this case is applicable to the harmonic oscillator potential , because it leads to interesting analogies of its spectrum with a known case in nuclear physics , as discussed in sec .
now the hamiltonian ( 4 ) contains a potential of the form @xmath46 where @xmath47 is the central potential from the previous section , @xmath48 the spin - orbit operator and @xmath49 a function which vanishes when @xmath50 . in atomic or nuclear physics the spin - orbit operator is the ordinary scalar product between the spin and angular momentum . in the deformed case considered here we aim at introducing a similar definition .
however there are inherent differences due to the more complex nature of the q - deformed vector operators , as explained below .
by analogy to the q - angular momentum @xmath51 one can define a spin operator @xmath16 through relations similar to ( 1 - 3 ) .
the operators @xmath51 and @xmath16 satisfy the hermiticity relations @xmath52 however the situation is different from the @xmath9 case because neither @xmath51 nor @xmath16 form a vector with respect to their @xmath2 algebra . in a @xmath2 algebra a q - vector of components @xmath53 , is defined through the relations @xcite @xmath54}v_{i\pm1}\nonumber\\ & & [ l_0,v_i]=iv_i\end{aligned}\ ] ] where one takes @xmath55 = 0 whenever it appears .
but as in ref .
@xcite , instead of @xmath51 we have to use @xmath56 defined as @xmath57}}~q^{-l_0}~l_\pm,\ ] ] @xmath58}~\left(q~l_+~l_-~-~q^{-1}~l_-~l_+\right).\ ] ] these quantities form a vector in the @xmath2 algebra , i.e. satisfy relations the ( [ r4 ] ) as it can be easily checked . by analogy to ( [ lapm ] ) and ( [ lazero ] )
we introduce a vector of components @xmath15 in the @xmath2 algebra having @xmath16 as generators @xmath59}}~q^{-s_0}~s_\pm,\ ] ] @xmath60}~\left(q~s_+~s_-~-~q^{-1}~s_-~s_+\right).\ ] ] in the space generated by @xmath16 the quantities @xmath51 are scalars and vice versa , which implies that @xmath61 = 0.\ ] ] in dealing with the spin - orbit operator we have to also introduce the coproduct algebra of @xmath51 and @xmath16 .
the generators @xmath62 of this algebra are defined as @xmath63 @xmath64 one can directly prove that they satisfy commutation relations of type ( [ def1 ] ) and ( [ def2 ] ) .
one can also prove that @xmath56 are the components of a vector in the coproduct algebra which means that they satisfy relations analogous to ( [ r4 ] ) with @xmath62 instead of @xmath65 .
on the other hand @xmath15 do not fulfil such relations .
however , instead of @xmath15 one can introduce another vector @xmath66 satisfying relations of type ( [ r4 ] ) with @xmath62 instead of @xmath65 .
this can be achieved by using the universal @xmath19 matrix .
in fact we need both the @xmath19 matrix and its conjugate @xcite .
the latter is here denoted by @xmath67 .
the @xmath19 matrix or its conjugate has the property that it replaces @xmath0 by @xmath68 in definition ( [ jpm ] ) i.e. one has @xmath69 and @xmath70 the operator ( [ jzero ] ) remains unchanged or in other words @xmath71 = 0 , \,\,\,\,\
[ { \mathcal r},~j_0 ] = 0.\ ] ] we found it convenient to use the @xmath19 matrix as defined in ref . @xcite . for @xmath72
it contains two terms only @xmath73 where @xmath74 one can check that the expression ( [ rq ] ) satisfies ( [ r8 ] ) .
the conjugate @xmath67 of @xmath19 takes the form @xmath75 and it satisfies equation ( [ r8prim ] ) . using ( [ rq ] ) and ( [ roverq ] ) one defines @xcite @xmath76 and @xmath77 the operator ( [ sigq ] ) with @xmath78 form a vector in the space of the coproduct algebra . the proof is given in appendix a. in a similar way one can prove that the operator ( [ sigoverq ] ) is also a vector in the coproduct algebra . note that none of the above operators is hermitian but each @xmath79-component of one is related to the corresponding component of the other through the relation @xmath80 relating operators associated with @xmath19 and @xmath67 . to overcome the lack of hermiticity
one can make use of the unitary matrix @xmath81 introduced in ref .
@xcite as @xmath82 where @xmath83 is a normalization factor
. with the help of @xmath81 one can define the vector @xmath84 the components of which are hermitian operators , i.e. satisfy the relation @xmath85 now we can define a hermitian spin - orbit operator as @xmath86 where the scalar product between the @xmath0-vectors @xmath87 and @xmath88 is defined as in ref .
@xcite @xmath89 with an implied summation over @xmath79 . using ( [ r15 ] )
one can rewrite ( [ r18 ] ) as @xmath90 let us consider the first term in the right hand side of ( [ r19 ] ) where @xmath81 is replaced by its definition ( [ r17 ] ) @xmath91 here we look for example at the term @xmath92 which can be rewritten by inserting the identity @xmath93 in front of @xmath94 and also using the property @xmath95 this gives @xmath96 in a similar way the other term of ( [ r22 ] ) becomes @xmath97 where we have used @xmath98 and @xmath99 thus @xmath100.\ ] ] one can see that in the above relation the vectors @xmath101 and @xmath88 are next to each other as it should be in a @xmath0-scalar product . for the second term of ( [ r19 ] ) we have @xmath102 or using @xmath103 in the manner explained above , we get @xmath104.\ ] ]
thus the spin - orbit interaction takes the form : @xmath105 .
\nonumber\\ + \vec{\lambda}\vec{\sigma}({\mathcal r})~(1+q~r^+r)~ + ~\vec{\lambda}\vec{\sigma}(r)~(1+{1\over q } { \mathcal r}^+{\mathcal r } ) ] % % % \right]\end{aligned}\ ] ] i.e. it contains the operators : @xmath106 these are scalars because they commute with @xmath107 . in particular , for the last two operators , the commutation with @xmath108 follows directly from ( [ com ] ) . to prove the commutation with @xmath109 we have to use eqs .
( [ r8 ] ) and ( [ r8prim ] ) .
for example in the case of @xmath110 we have @xmath111 where after the second equality sign alternative forms of eqs .
( [ r28 ] ) have been used .
we can obtain the expectation value of @xmath48 for states of total angular momentum @xmath112 by calculating the expectation values of the scalars ( [ scalars ] ) .
the simplest way is to use the state of maximum weight with @xmath113 . for @xmath114
this state reads @xmath115 where @xmath116 are defined by eqs .
( 56 ) and ( 57 ) of ref .
@xcite and @xmath117 with @xmath118 is the @xmath119 spinor . in this case
one can show that the last two operators of the list ( [ scalars ] ) have the following expectation values : @xmath120 for @xmath121 and @xmath113 the wave function is : @xmath122}}\left(\sqrt{[2\ell]\over q}~y_{\ell \ell}\chi_{-1/2}-q^l~y_{\ell , \ell-1}\chi_{1/2}\right)~.\ ] ] in this case the last two operators of ( [ scalars ] ) have the following expectation values : @xmath123 both for @xmath124 and @xmath125 the proof is similar to that given in the appendix b for the other scalars of ( [ scalars ] ) .
using all these expectation values in the case where @xmath124 one can easily show that the expectation value of @xmath48 is : @xmath126}{\left[2\right]^2}~ \frac{q^{l+5/2}+q^{-l-5/2}}{q^{l+1/2}+q^{-l-1/2}}~. % { \begin{tabular}[t]{c } \longrightarrow\\ % q \rightarrow 1 \end{tabular } \frac{\ell}{2 } % \label{eplus}\ ] ] in a similar but somewhat longer way the following expectation value of @xmath48 is obtained for @xmath121 : @xmath127\over[2]^2}~{q^{l-3/2}+q^{-l+3/2}\over q^{l+1/2}+q^{-l-1/2}}\ ] ] the proof of eqs .
( [ eplus ] ) and ( [ eminus ] ) is given in appendix b. in the limit @xmath128 @xmath129 and @xmath130 recover the expectation values of the non - deformed spin - orbit coupling @xmath131 namely @xmath132 for @xmath133 and @xmath134 for @xmath135 respectively .
in fig . 1 we represent the eigenvalues ( [ coul ] ) of the coulomb potential as a function of @xmath12 ( real ) , when @xmath136 ( eq . ( [ qreal ] ) ) .
one can see that every @xmath137 increases with @xmath12 when @xmath41 , the reason being that one has @xmath138 when one chooses @xmath0 to be real .
therefore at a given @xmath10 one has @xmath139 these inequalities are similar to those satisfied by the eigenvalues of the klein - gordon equation for which one has @xmath140 for fixed @xmath141 @xcite .
one expects similar inequaltities to be also satisfied by the eigenvalues of the spinless bethe - salpeter ( or herbst ) equation for a particle in an attractive coulomb potential @xcite .
in fact as long as @xmath142 where @xmath143 is the charge and @xmath49 the fine structure constant the expansion of the eigenvalues of the herbst equation coincides with that of the klein - gordon equation @xcite . thus the results shown in fig .
1 suggest that the splitting found for @xmath10 can simulate relativistic kinematic effect . in fig .
2 the eigenvalues ( [ ho ] ) of the harmonic oscillator potential are plotted as a function of @xmath12 , where @xmath12 and the deformation parameter are related by eq .
( [ qcom ] ) .
this choice is based on the fact that it implies @xmath144 so that in the interval @xmath145 the @xmath0-deformed spectrum satisfies inequalities as @xmath146 which correspond to a potential the form of which is between a harmonic oscillator and a square well potential . in nuclear physics
@xcite the standard form is the woods - saxon potential @xmath147^{-1}~ , \end{array}\ ] ] depending on three parameters @xmath148 , @xmath149 and @xmath150 . in the limit @xmath151 one approaches a square well potential of radius @xmath149 and depth @xmath148 . the bound spectrum of a potential of type ( [ ws ] ) satisfies the inequalities ( [ woods ] ) ( see figs . 2 - 23 of ref .
@xcite ) .
next we add the spin - orbit contribution .
to single out the role of @xmath48 here we choose @xmath49 to be a constant . in fig
3 we plot @xmath152 as a function of @xmath12 , where @xmath12 is related to @xmath0 by eq .
( [ qreal ] ) .
the levels are labelled by @xmath153 where @xmath44 is the value of the angular momentum at @xmath8 and @xmath154 . with
@xmath155 one always has @xmath133 levels above the @xmath156 levels due to eqs .
( [ eplus ] ) and ( [ eminus ] ) . for convenience
we choose @xmath49 = 0.001 .
we therefore see that the energies increase with increasing @xmath157 for fixed @xmath44 and increasing @xmath35 or @xmath44 for fixed @xmath157 .
such a pattern corresponds to solutions of the dirac equation for a coulomb potential plus a perturbation which removes the two - fold degeneracy of the eigenvalues for a coulomb field . in ref .
@xcite it has been shown that for a dirac particle moving in a purely attractive potential the level sequence is @xmath158 @xmath159 which is here the case when @xmath160 for the first and when @xmath161 for both rows of inequalities respectively .
such sequences are expected for alkaline atoms . in a similar way we add the spin - orbit coupling ( [ eplus ] ) and ( [ eminus ] ) to @xmath162 of eq .
( [ ho ] ) and in fig .
4 we plot @xmath163 as a function of @xmath12 , where @xmath12 is related to @xmath0 via eq .
( [ qcom ] ) . for the sake of the discussion here
we choose @xmath49 = - 0.1 the addition of a spin - orbit coupling to @xmath162 brings us a picture even closer to the single particle spectra encountered in nuclear physics .
provided @xmath49 is negative the level sequence of fig .
4 is similar to that of the neutron single particle spectrum ( see e.g. figs . 2 - 30 of ref .
@xcite ) . also hartree - fock calculations based on effective density dependent nucleon - nucleon interactions @xcite give a similar spectrum .
we have constructed a @xmath0-analogue of the spin - orbit coupling for being used in a @xmath0-deformed schrdinger equation previously derived for a central potential .
the spin - orbit coupling is a @xmath0-scalar product between the angular momentum @xmath56 and the spin operator @xmath66 both defined a @xmath0-vectors in the coproduct algebra of the generators @xmath18 .
the spin operator has been obtained with the help of a hermitian form of the universal @xmath19 matrix .
accordingly , our result is new and entirely different from previous work on the spin - orbit coupling .
numerically we have shown that the @xmath0-deformed schrdinger equation for a spinless particle in a coulomb field has a spectrum which simulates relativistic effects .
the removal of the accidental degeneracy by a real deformation of type @xmath164 with @xmath165 leads to a level sequence similar to that of the klein - gordon or of the herbst equations . with the addition of a spin - orbit coupling
the level sequence is close to that of alkaline atoms .
the @xmath0-deformed schrdinger equation for a spinless particle in a harmonic oscillator potential has a spectrum similar to that of the bound spectrum of an woods - saxon potential .
the deformation is of type @xmath166 with @xmath12 real and positive .
the addition of a spin - orbit coupling leads to a spectrum similar to single particle spectra of nuclei .
it would be interesting to pursue this study in a more quantitative way .
in this appendix we prove that the operators @xmath167 with @xmath168 form a @xmath0-vector in the coproduct algebra of @xmath62 defined by ( [ jpm ] ) and ( [ jzero ] ) .
a vector is an irreducible tensor of rank @xmath169 .
the proof given below is valid for any @xmath170 .
let us consider a @xmath0-tensor @xmath171 which is irreducible in the space generated by @xmath16 .
by definition it must obey the relations
@xcite : @xmath172 = \mu~u^{\lambda}_{\mu}~,\ ] ] @xmath173[\lambda\pm\mu+1]}~u^{\lambda}_{\mu}~.\ ] ] the operator @xmath15 defined by ( [ sipm ] ) and ( [ sizero ] ) is an example of @xmath171 with @xmath169 . in the composite system of the coproduct algebra of @xmath62 a tensor @xmath174 defined by @xmath175 is irreducible if it satisfies relations analogous to ( [ we0 ] ) and ( [ we1 ] ) but with @xmath18 instead of @xmath16 .
suppose @xmath174 satisfies such relations .
below we show that they are compatible with ( [ we0 ] ) and ( [ we1 ] ) .
the validity of @xmath176 = \mu~w^{\lambda}_{\mu}~,\ ] ] is immediate due to the independence of @xmath108 of @xmath0 , see eq.([jzero ] ) .
using ( [ jpm ] ) the analogue of ( [ we1 ] ) becomes @xmath177[\lambda\pm\mu+1]}~r^{-1}u^{\lambda}_{\mu\pm1}r~,\end{aligned}\ ] ] for @xmath174 defined by ( [ a2 ] ) .
we multiply the above equation by @xmath19 on the left and by @xmath178 on the right and use eq .
( [ r8 ] ) to shift the @xmath19 from the left to the right of @xmath179 . using the identity @xmath180
we obtain @xmath181[\lambda\pm\mu+1]}~u^{\lambda}_{\mu\pm1}~.\end{aligned}\ ] ] next we use @xmath182 which is a consequence of ( [ we0 ] ) and @xmath183 which is a consequence of ( [ comm ] ) .
these relations help to cancel out two of the four terms in the left hand side of ( [ a4 ] ) .
the resulting equation is ( [ we1 ] ) which proves that ( [ a3 ] ) is correct . identifying @xmath174 with @xmath17 i.e. setting @xmath169 in ( [ we0p ] ) and ( [ a3 ] ) we obtain equations of type ( [ r4 ] ) for @xmath17 i.e.
we prove that @xmath17 is a @xmath0-vector in the coproduct algebra @xmath62 .
in this appendix we show how the formulae ( [ eplus ] ) and ( [ eminus ] ) can be obtained . for this purpose
we need the expectation values of the scalars ( [ scalars ] ) . in order to calculate explicitly the expectation value of the first and third scalar products from the list ( [ scalars ] ) we need the operators @xmath184 . which can be obtained by introducing eq .
( [ rq ] ) in eq .
( [ sigq ] ) .
this gives : @xmath185\lambda\lambda_{-1}\sigma_1~,\nonumber\\ & & \sigma_{-1}(r)=q^{2l_0}\sigma_{-1}-[2]\lambda~ q^{l_0 } \lambda_{-1}q^{l_0}\sigma_0+[2]\lambda^2 q^{l_0}\lambda^2_{-1 } q^{l_0}\sigma_1~.\end{aligned}\ ] ] to calculate the expectation value of the second and fourth scalar products ( [ scalars ] ) we need @xmath186\lambda~ q^{l_0 } \lambda_{1}q^{l_0}\sigma_0+[2]\lambda^2 q^{l_0}\lambda^2_{1 } q^{l_0}\sigma_{-1}~,\nonumber\\ & & \sigma_0({\mathcal r})=\sigma_0-[2]\lambda\lambda_{1}\sigma_{-1}~,\nonumber\\ & & \sigma_{-1}({\mathcal r})=q^{-2l_0}\sigma_{-1}~,\end{aligned}\ ] ] which have been derived from the formulae ( [ roverq ] ) and ( [ sigoverq ] ) .
for the purpose of this appendix , as an example , we first calculate the expectation value of the third scalar product from the list ( [ scalars ] ) .
this is @xmath187 from this expression only the first and second terms bring a nonvanishing contribution to the expectation value when @xmath188 .
looking at the expression of @xmath189 above we see that only the second term contributes so that @xmath190 has a nonvanishing contribution due to @xmath191}{q}~\lambda \lambda_1~q^{l_0}\lambda_{-1}q^{l_0}\sigma_0.\ ] ] using the definition ( [ lapm ] ) one can rewrite this operator as @xmath192 at this stage we need the relation @xmath193~ [ \ell - m+1]~y_{\ell m}(q , x_0,\varphi).\ ] ] for the particular case of @xmath194 we have @xmath195 ~y_{\ell \ell}(q , x_0,\varphi).\ ] ] the relation ( [ 47 ] ) has a spin counterpart @xmath196~ [ s - m_s+1]~\chi_{m_s}.\ ] ] together with ( [ sizero ] ) this gives @xmath197}~s_+~s_-~\chi_{1/2 } = \frac{q}{[2]}~~\chi_{1/2}.\ ] ] altogether we get @xmath198}{[2]}~ \psi_{\ell+1/2,\ell+1/2}.\ ] ] according to ( [ sigr ] ) the nonzero contribution of @xmath199 acting on @xmath200 comes from @xmath201 . using @xmath202 as defined by ( [ lazero ] ) and the relation ( [ 47 ] ) we get @xmath203 } [ 2 \ell]~ y_{\ell \ell}~.\ ] ] together with ( [ sigma_0 ] ) this gives @xmath204 ^ 2}~[2 \ell ] ~ \psi_{\ell+1/2,\ell+1/2}~.\ ] ] the addition of ( [ term1 ] ) and ( [ term2 ] ) leads to the following expectation value : @xmath205 ^ 2}[2l]~.\ ] ] in the same representation , i. e. @xmath188 the expectation value of @xmath206 is even easier to obtain inasmuch as only the term @xmath207 contributes . using the result ( [ term2 ] ) one gets @xmath208 ^ 2}[2l]~.\ ] ] in a similar manner as above we obtain @xmath209 ^ 2}[2l]~,\end{aligned}\ ] ] and @xmath210 ^ 2}[2l]~.\ ] ] by using the expectation values ( [ aplus1])-([aplus4 ] ) , together with ( [ rcrucerp ] ) and ( [ rcrucer ] ) one can calculate the expectation value of ( [ operator ] ) which leads straightforwardly to ( [ eplus ] ) . for the representation @xmath125 , in a similar but longer way
one obtains @xmath211 ^ 2}[2l+2]~,\nonumber\\ { \langle \vec{\sigma}({\mathcal r } ) \vec{\lambda } \rangle}_{\ell-1/2 } = { \langle \vec{\lambda}\vec{\sigma}(r ) \rangle}_{\ell-1/2 } = -{q^{-2}\over[2]^2}[2l+2]~.\end{aligned}\ ] ] the relations ( [ rcrucerp ] ) , ( [ rcrucer ] ) and ( [ aminus ] ) lead to the expectation value ( [ eminus ] ) .
kulish p. p. and reshetikin n. yu .
1981 zapiski semenarov lomi * 101 * 101 sklyanin e. k. 1982 funct .
* 16 * 262 jimbo m. 1986 lett .
phys . * 11 * 247 biedenharn l. c. 1989 j. phys .
a : math . gen .
* 22 * l873 macfarlane a. j. 1989 j. phys .
a : math . gen . * 22 * 4581 see e.g. bonatsos d. , drenska s. b. , raychev p. p. , roussev r. p. and smirnov yu .
f. 1991 j. phys .
g : nucl . part .
* 17 * l67 and references therein ; for a recent review on applications of quantum algebras to nuclear physics see bonatsos d. and daskaloyanis c 1999 progr .
* 43 * 337 minahan j. a. 1990 mod .
phys . lett . *
a5 * 2625 li you - quan and sheng zheng - mao 1992 j. phys .
a : math . gen . * 25 * 6779 carow - watamura u. and watamura s. 1994 int . j. mod .
phys . * a9 * 3989 xing - chang song and li liao 1992 j. phys .
* 25 * 623 irac - astaud m. 1996 lett .
phys . * 36 * 169 papp e. 1996 j. phys .
a : math . gen . * 29 * 1795 m. micu 1999 j. phys
. a : math . gen . * 32 * 7765 raychev p. p. , rousev r. p. , lo iudice n. and terziev p.a .
, 1998 j. phys .
g : nucl . part . phys . * 24 * 1931 raychev p. p. , rousev r. p. , terziev p.a .
bonatsos d. and lo iudice n. 1996 j. phys .
a : math . gen . *
29 * 6939 andrews g. , askey r. and roy r. _ special functions _
, cambridge university press , 1999 curtright t. l. , ghandour g. i. and zachos c. k. 1991 j. math .
phys . * 32 * 676 rittenberg v. and scheunert m. 1992 , j. math .
phys . * 33 * 436 quesne c. 1993 j. phys .
a : math . gen .
* 26 * l299 grosse h. , martin a. and stubbe j. 1991 phys .
b255 * 563 grosse h. , martin a. and stubbe j. 1994 j. math . phys .
* 35 * 3805 see e.g. lucha w and schberl f. f , 1997 phys
* a56 * 139 , martin a. 1997 _ quark confinement and the hadron spectrum ii _ eds .
brambilla n. and prosperi g. m. world scientific , singapore p.187 martin a , talk given at the workshop _ critical stability of quantum few - body systems _ , ect * trento , february 3 - 14 1997 , unpublished bohr a. and mottelson b. r. 1969 _ nuclear structure _ , benjamin new york , vol .
1 , chapter 2 , sec . 4 | = 0.50 cm based on the tensor method , a @xmath0-analogue of the spin - orbit coupling is introduced in a @xmath0-deformed schr " odinger equation , previously derived for a central potential .
analytic expressions for the matrix elements of the representations @xmath1 are derived .
the spectra of the harmonic oscillator and the coulomb potential are calculated numerically as a function of the deformation parameter , without and with the spin - orbit coupling .
the harmonic oscillator spectrum presents strong analogies with the bound spectrum of an woods - saxon potential customarily used in nuclear physics .
the coulomb spectrum simulates relativistic effects .
the addition of the spin - orbit coupling reinforces this picture . |
ferdinand magellan s expedition was the first that completed the circumnavigation of our globe during 1519 - 1522 , after discovering the _ strait of magellan _ between the atlantic and pacific ocean in search for a westward route to the `` spice islands '' ( indonesia ) , and thus gave us a first @xmath7 view of our planet earth .
five centuries later , nasa has sent two spacecraft of the stereo mission on circumsolar orbits , which reached in 2011 vantage points on opposite sides of the sun that give us a first @xmath7 view of our central star .
both discovery missions are of similar importance for geographic and heliographic charting , and the scientific results of both missions rely on geometric triangulation .
the twin stereo / a(head ) and b(ehind ) spacecraft ( kaiser et al .
2008 ) , launched on 2006 october 26 , started to separate at end of january 2007 by a lunar swingby and became injected into a heliocentric orbit , one propagating `` ahead '' and the other `` behind '' the earth , increasing the spacecraft separation angle ( measured from sun center ) progressively by about @xmath8 per year .
the two spacecraft reached the largest separation angle of @xmath9 on 2011 february 6 .
a stereo secchi cor1-a / b intercalibration was executed at @xmath9 separation ( thompson et al .
thus , we are now in the possession of imaging data from the two stereo / euvi instruments ( howard et al . 2008 ; wlser et al .
2004 ) that cover the whole range from smallest to largest stereoscopic angles and can evaluate the entire angular range over which stereoscopic triangulation is feasible .
it was anticipated that small angles in the order of @xmath10 should be most favorable , similar to the stereoscopic depth perception by eye , while large stereoscopic angles that are provided in the later phase of the mission would be more suitable for tomographic 3d reconstruction .
the first stereoscopic triangulations using the stereo spacecraft have been performed for coronal loops in active regions , observed on 2007 may 9 with a separation angle of @xmath11 ( aschwanden et al .
2008 ) and observed on 2007 june 8 with @xmath12 ( feng et al .
further stereoscopic triangulations have been applied to oscillating loops observed on 2007 june 26 with a stereoscopic angle of @xmath13 ( aschwanden 2009 ) , to polar plumes observed on 2007 apr 7 with @xmath14 ( feng et al .
2009 ) , to an erupting filament observed on 2007 may 19 with @xmath15 ( liewer et al .
2009 ) , to an erupting prominence observed on 2007 may 9 with @xmath16 ( bemporad 2009 ) , and to a rotating , erupting , quiescent polar crown prominence observed on 2007 june 5 - 6 with @xmath17 ( thompson 2011 ) .
thus , all published stereoscopic triangulations have been performed within a typical ( small ) stereoscopic angular range of @xmath18 , as it was available during the initial first months of the stereo mission .
the largest stereoscopic angle used for triangualtion of coronal loops was used for active region 10978 , observed on 2007 december 11 , with a spacecraft separation of @xmath19 ( aschwanden and sandman 2010 ; sandman and aschwanden 2011 ) , which produced results with similar accuracy as those obtained from smaller stereoscopic angles .
so there exists also an intermediate rangle of aspect angles that can be used for stereoscopic triangulation . however , nothing is known whether stereoscopy is also feasible at large angles , say in the range of @xmath20 , and how the accuracy of 3d reconstruction depends on the aspect angle , in which range the stereoscopic correspondence problem is intractable , and whether stereoscopy at a maximum angle near @xmath21 is equally feasible as for @xmath22 for optically thin structures ( as it is the case in soft x - ray and euv wavelengths ) , due to the @xmath9 symmetry of line - of - sight intersections . in this study
we are going to explore stereoscopic triangulation of coronal loops in the entire range of @xmath23 and quantify the accuracy and quality of the results as a function of the aspect angle .
observations and data analysis are reported in section 2 , while a discussion of the results is given in section 3 , with conclusions in section 4 . indicated approximately at the beginning of the years , ranging from @xmath24 in april 2007 to @xmath25 in february 2011.,scaledwidth=100.0% ]
we select stereo observations at spacecraft separation angles with increments of @xmath26 over the range of @xmath27 to @xmath25 , which corresponds to time intervals of about a year during the past mission lifetime 20072011 .
a geometric sketch of the spacecraft positions stereo / a+b relative to the earth - sun axis is shown in fig .
1 . additional constraints in the selection are : ( i ) the presence of a relatively large prominent active region ; ( ii ) a position in the field - of - view of both spacecraft ( since the mutual coverage overlap drops progressively from @xmath9 initially to @xmath28 during the first 4 years of the mission ) ; ( iii ) a time near the central meridian passage of an active region viewed from earth ( to minimize confusion by foreshortening ) ; and ( iii ) the availability of both stereo / euvi / a+b and calibrated soho / mdi data .
the selection of 5 datasets is listed in table 1 , which includes the following active regions : ( 1 ) noaa 10953 observed on 2007 april 30 ( also described in derosa et al . 2009 ; sandman et al .
2009 , aschwanden and sandman 2010 ; sandman and aschwanden 2011 , aschwanden et al .
2012 ) , ( 2 ) noaa region 10978 observed on 2007 december 11 ( also described in aschwanden and sandman 2010 , aschwanden et al .
2012 , and subject to an ongoing study by alex engell and aad van ballegooijen , private communication ) , ( 3 ) noaa 11010 observed on 2009 jan 12 , ( 4 ) noaa 11032 observed on 2009 nov 21 , and ( 5 ) noaa 11127 observed on 2010 nov 23 .
this selection covers spacecraft separation angles of @xmath29 , and @xmath1 . for each of the 5 datasets we stacked the images during a time interval of 2030 minutes in order to increase the signal - to - noise ratio of the euvi images .
during the first 3 years ( 2006 - 2008 ) the nominal cadence of 171 images was 150 s , which yields 8 stacked images per 20 minute interval . later in the mission
, the highest cadence was chosen for the 195 wavelength , but dropped from 150 s to 300 s due to the reduced telemetry rate at larger spacecraft distances , which yields 612 stacked images per 30 minute interval . in one case ( 2010 nov 23 ) the cadence in euvi / a and b are not equal , either due to data loss or different telemetry priorities ( see time intervals and number of stacked images in table 1 ) .
the solar rotation during the time interval of stacked image sequences was removed to first order by shifting the images by an amount corresponding to the rotation rate at the extracted subimage centers .
.data selection of 5 active regions observed with stereo / euvi and soho / mdi .
[ cols= " < , < , < , < , > , > " , ] the first case is active region noaa 10953 ( fig .
2 ) , where we display the same 100 loop segments that have been triangulated in an earlier study ( aschwanden and sandman 2010 ) .
the spacecraft separation angle is @xmath30 and the almost identical direction of the line - of - sights of both stereo / a and b spacecraft makes it easy to identify the corresponding loops in a and b , and thus the triangulation is very reliable .
note that the height range where discernable loops can be traced in the highpass - filtered images is about @xmath31 solar radii ( or @xmath32 mm ) , which is commensurable with the hydrostatic density scale height expected for a temperature of @xmath33 mk that corresponds to the peak sensitivity of the euvi 171 filter .
this is particularly well seen in the side view shown in the bottom left panel in fig . 2 .
a measurement of the mean misalignment angle averaged over 10 positions of the 100 reconstructed loops with the local magnetic potential field shows a value of @xmath34 ( table 2 ) , similar to earlier work ( aschwanden and sandman 2010 ; sandman and aschwanden 2011 ) .
however , forward - fitting of a nonlinear force - free field model reduces the misalignment to @xmath35 , which implies that this active region is slightly nonpotential .
the remaining misalignment is attributed to at least two reasons , partially to inadequate parameterization of the force - free field model , and partially to stereoscopic measurement errors @xmath36 due to misidentified loop correspondences and limited spatial resolution .
an empirical estimate of the stereoscopic error was devised in aschwanden and sandman ( 2010 ) , based on the statistical non - parallelity of closely - spaced triangulated loop 3d trajectories , which yielded for this case a value of @xmath37 . in summary
, we find that this active region is very suitable for stereoscopy , allows to discern a large number ( 100 ) of loops , minimizes the stereoscopic correspondence problem due to the small ( @xmath30 ) spacecraft separation angle , displays a moderate misalignment angle and stereoscopic measurement error ( @xmath38 ) .
this well - defined case will serve as a reference for stereoscopy at larger angles . .
the images are highpass - filtered to enhance loop structures ( middle left and right panels ) .
a near - simultaneous soho / mdi magnetogram is shown ( bottom right ) , overlaid with the stereoscopically triangulated loops ( blue curves ) and magnetic field lines computed with a nonlinear force - free model ( red curves ) , viewed from the direction of earth or soho / mdi ( bottom right ) , and rotated by @xmath39 to the north ( bottom left).,scaledwidth=100.0% ] .
a soho / mdi magnetogram is shown ( bottom right ) , overlaid with the stereoscopically triangulated loops ( blue curves ) and magnetic field lines computed with a nonlinear force - free model ( red curves).,scaledwidth=100.0% ] .
a soho / mdi magnetogram is shown ( bottom right ) , overlaid with the stereoscopically triangulated loops ( blue curves ) and magnetic field lines computed with a nonlinear force - free model ( red curves).,scaledwidth=100.0% ] .
a soho / mdi magnetogram is shown , overlaid with the stereoscopically triangulated loops ( blue curves ) and magnetic field lines computed with a potential field model .
magnetic field lines have a footpoint threshold of @xmath40 g.,scaledwidth=100.0% ] .
a soho / mdi magnetogram is shown , overlaid with the stereoscopically triangulated loops ( blue curves ) and magnetic field lines computed with a potential field model .
magnetic field lines have a footpoint threshold of @xmath40 g.,scaledwidth=100.0% ] the second case is active region noaa 10978 ( fig .
3 ) , observed on 2007 dec 11 with a spacecraft separation angle of @xmath19 .
note that the views from euvi / a and b appear already to be significantly different with regard to the orientation of the triangulated loops , as seen from a distinctly different aspect angle .
a set of 52 coronal loops were stereoscopically triangulated in this region ( aschwanden and sandman 2010 ) , a mean misalignment angle of @xmath41 is found for a potential field model , and a reduced value of @xmath42 is found for the force - free model ( table 2 ) , while a stereoscopic error of @xmath43 is estimated ( aschwanden and sandman 2010 ) .
thus , the quality of stereoscopic triangulation ( as well as the degree of non - potentiality ) is similar to the first active region , although we performed stereoscopy with a 7 times larger spacecraft separation angle ( @xmath19 ) than before ( @xmath30 ) . apparently , stereoscopy is still easy at such angles , partially helped by the fact that the active region is located near the central meridian ( @xmath44 ) for both spacecraft , which provides an unobstructed view from top down , so that the peripheral loops of the active region do not overarch the core of the active region , where the bright reticulated moss pattern ( berger et al .
1999 ) makes it almost impossible to discern faint loops in the highpass - filtered images .
the top - down view provides also an optimum aspect angle to disentangle closely - spaced loops , which is an important criterion in the stereosopic correspondence identification .
the third case is active region noaa 11010 ( fig
. 4 ) , observed on 2009 jan 12 with a near - orthogonal spacecraft separation angle of @xmath45 . due to the quadrature of the spacecraft , only a sector of @xmath46 east and west of the central meridian ( viewed from earth )
is jointly visible by both spacecraft .
this particular active region is seen at a @xmath8 angle by both stereo / a and stereo / b .
this symmetric view is the optimum condition to discern a large number of inclined loop segments and to identify the stereoscopic correspondence .
we triangulate some 20 loop segments , which appear almost mirrored in the stereo / a and b image due to the east - west symmetry of the magnetic dipole .
a mean misalignment angle of @xmath47 with the potential field model is found , and a reduced value of @xmath48 with the force - free field model .
an estimate of the statistical ( non - parallelity ) stereoscopic error is not possible due to the small number of triangulated loops .
thus , we conclude that stereoscopy is still possible in quadrature . mathematically , the orthogonal projections should yield the most accurate 3d coordinates of a curvi - linear structure , but in practice , confusion of multiple structures with near - aligned projections can cause a disentangling problem in the stereoscopic correspondence identification at this intermediate angle . the fourth case is active region noaa 11032 ( fig
. 5 ) , observed on 2009 nov 21 with a large spacecraft separation angle of @xmath49 .
stereo / a sees the active region near the east limb from an almost side - on perspective , while stereo / b sees a similar mirror image near the west limb , where confusion near the limb makes the stereoscopic correspondence identification more difficult .
we trace some 15 loop segments , but do not succeed in pinning down a larger number of loops , partially because this active region is small and does not exhibit numerous bright loops , and partially because of increasing confusion problems near the limb .
we searched for larger active regions over several months around this time , but were not successful due to a dearth of solar activity during this time .
we find a misalignment angle of @xmath50 for the potential field , and @xmath51 for the force - free field model , which is still comparable with the previous active regions triangulated at smaller sterescopic angles .
thus , stereoscopy seems to be still feasible at such large stereoscopic angles .
the last case is active region noaa 11127 ( fig .
6 ) , observed on 2010 nov 23 with a very large spacecraft separation angle of @xmath52 , only two months before the two stereo spacecraft pass the largest separation point . at this point , the common field - of - view that is overlapping from stereo / a and b is only the central meridian zone seen from earth ( or the opposite meridian behind the sun ) .
stereo / a observes active region noaa 11127 at its east limb , while stereo / b sees it at its west limb , so both spacecraft see only the vertical structure of the active region from a side view ( see fig . 6 top ) .
this particular configuration is very unfavorable for stereoscopy .
although the vertical structure in altitude can be measured very accurately , the uncertainty in horizontal direction in longitude is very large and suffers moreover the sign ambiguity of positive or negative longitude difference with respect to the limb seen from earth .
consequently , we have reliable information on the altitude and latitude of loops , while the longitude is essentially ill - defined . in order to reduce the large scatter in the measurement of @xmath53-coordinates along a loop , introduced by the near - infinite amplification of parallax uncertainties tangentially at the limb , we restrict the general solution of geometric 3d triangulation to planar loops , by applying a linear regression fit of the @xmath54 coordinates .
the example in fig .
6 shows that we can trace some ( 5 ) loops in the plane of the sky and have no problem in identifying the stereoscopic counterparts in both stereo / a and b images , but the stereoscopic triangulation is ill - defined at this singularity of the sign change in the parallax effect .
the misalignment between the three loop directions and the potential field is @xmath55 , and for the force - free field model is @xmath56 , which indicates that the orientation of the loop planes is less reliably determined .
stereoscopic triangulation brakes down at this singularity of separation angles at @xmath57 , although the stereoscopic correspondence problem is very much reduced for the `` mirror images '' , similar to the near - identical images at small separation angles @xmath58 .
we are discussing now the pro s and con s of stereoscopy at small and large aspect angles , which includes quantitative estimates of the formal error of stereoscopic triangulation ( section 3.1 ) , the stereoscopic correspondence and confusion problem ( section 3.2 ) , and the statistical probability of stereoscopable active regions during the full duration of the stereo mission ( section 3.3 ) , all as a function of the stereoscopic aspect angle ( or spacecraft separation angle @xmath59 in the case of the stereo mission ) .
stereoscopic triangulation involves a parallax angle around the normal of the epipolar plane . for the stereo mission
, the epipolar plane intersects the sun center and the two spacecraft a and b positions , which are separated mostly in east - west direction .
no parallax effect occurs when the loop axis coincides with the epipolar plane , i.e. , when the loop axis points in east - west direction .
thus , the accuracy of stereoscopic triangulation depends most sensitively on this orientation angle @xmath60 , which we define as the angle between the loop direction and the normal of the epipolar plane ( i.e. , approximately the y - axis of a solar image in north - south direction ) .
if the position of a loop centroid can be determined with an accuracy of a half pixel size @xmath61 , the dependence of the stereoscopic error on the orientation angle @xmath60 is then ( aschwanden et al .
2008 ) , @xmath62 thus , for a highly inclined loop that has an eastern and western footpoint at the same latitude , the stereoscopic error is minimal near the loop footpoints ( pointing in north - south direction ) and at maximum near the loop apex ( pointing in east - west direction ) .
in addition , the accuracy of stereoscopic triangulation depends also on the aspect angle ( or spacecraft separation angle @xmath59 ) , which can be quantified with an error trapezoid as shown in fig .
if the half separation angle @xmath63 is defined symmetrically to the earth - sun axis ( z - axis ) , the uncertainty in z - direction is @xmath64 and is in the x - direction ( in the epipolar plane ) @xmath65/2)$ ] . including also a half pixel - size error @xmath66 in the y - direction
, we have then a combined error for the 3d position of a triangulated point as , @xmath67 } } } \ .\ ] ] this positional error is symmetric for small @xmath59 and large stereoscopic angles @xmath68 , and has a minimum at an orthogonal angle of @xmath69 .
the error is largest in z - direction for small spacecraft separation angles , while it is largest in x - direction for separation angles near @xmath70 . .
the uncertainties @xmath71 in x - direction and spacecraft in z - direction depend on the pixel width @xmath72 and half aspect angle @xmath63.,scaledwidth=50.0% ] to compare the relative importance of the two discussed sources of errors we can evaluate the parameters that increase the individual errors by a factor of two .
this is obtained when the orientation angle of a loop segment ( with respect to the east - west direction ) increases from @xmath73 to @xmath74 , or if the spacecraft separation angle changes from the optimum angle @xmath75 to @xmath76 ( or @xmath77 , respectively ) . if stereoscopy at small angles of @xmath78 is attempted , the positional error is about ten - fold ( corresponding to @xmath79 ) , compared with the optimum angle at @xmath75 ( corresponding to @xmath80 ) . for stereo / euvi with a pixel size of @xmath81 mm , this amounts to an accuracy range of @xmath82 mm .
the previous considerations are valid for isolated loops that can be unambiguously disentangled in an active region , in both the stereo / a and b images .
however , this is rarely the case . in crowded parts of active regions , the correspondence of a particular loop in image
a with the identical loop in image b can often not properly be identified .
we call this confusion problem also the _ stereoscopic correspondence problem _ , which appears in every stereoscopic tie - point triangulation method . in order to quantify this source of error
, we have to consider the area density of loops and their relative orientation . a top - down view of an active region , e.g. , as seen for small stereoscopic angles by both spacecraft for an active region near disk center ( e.g. , fig .
2 ) , generally allows a better separation of individual loops , because only the lowest density scale height is detected ( due to hydrostatic gravitational stratification ) , and neighbored loop segments do not obstruct each other due to the foreshortening projection effect near the footpoints .
in contrast ,
every active region seen near the limb , shows many loops at different longitudes , but at similar latitudes , cospatially on top of each other , which represents the most severe confusion problem .
thus , we can essentially quantify the degree of confusion by the loop number density per pixel , which approximately scales with the inverse cosine - function of the center - to - limb angle due to foreshortening .
in other words , we can define a quality factor @xmath83 for identifying the stereoscopic correspondence @xmath84 of properly disentangled coronal loops , which drops from @xmath85 at disk center to @xmath86 at the limb , where @xmath87 is the center - to - limb angle measured from sun center , @xmath88 of stereoscopic triangulation as a function of the spacecraft separation angle @xmath59 , which is a function of the accuracy @xmath89 of triangulated stereoscopic positions and the stereoscopic correspondence quality factor @xmath84 .
the best quality ( within a factor of 2 ) occurs in the range of @xmath90.,scaledwidth=90.0% ] the orbits of the stereo mission reduce the overlapping area on the solar surface that can be jointly viewed by both spacecraft a and b linearly with increasing spacecraft separation angle , so that the center - to - limb distance @xmath87 of an active region located on the central meridian ( viewed from earth ) is related to the spacecraft separation angle by @xmath91 increasing linearly with the separation angle from @xmath92 at the beginning of the mission to @xmath93 at maximum spacecraft separation angle .
the location of an active region at the central meridian provides the best view for both spacecraft , because an asymmetric location would move the active region closer to the limb for one of the spacecraft , and thus would increase the degree of confusion , as we verified by triangulating a number of asymmetric cases .
thus , we can express the quality factor of stereoscopic correspondence @xmath84 ( eq .
7 ) by the spacecraft separation angle ( eq . 8) and obtain the relationship , @xmath94 defining a quality factor @xmath95 for stereoscopic triangulation by combining the stereoscopic correspondence quality @xmath84 ( eq . 9 ) with the accuracy @xmath89 of stereoscopic positions ( eq . 6 ) , which we may define by the normalized inverse error ( i.e. , @xmath96 ) , we obtain @xmath97 } } } \ .\ ] ] we plot the functional dependence of this stereoscopic quality factor @xmath98 together with their underlying factors @xmath99 and @xmath100 in fig . 8 and
obtain now an asymmetric function of time ( or spacecraft separation angle ) that favors smaller stereoscopic angles .
the stereoscopic quality factor is most favorable ( within a factor of 2 ) in the range of @xmath90 , which corresponds to the mission phase between august 2007 and november 2009 .
the same optimum range will repeat again at the backside of the sun 5 years later between august 2012 and november 2014 . from our analysis of 5 active regions spread over the entire spacecraft separation angle range we find acceptable results regarding triangulation accuracy in the range of separation angles of @xmath101 and @xmath102 ( based on acceptable misalignment angles of @xmath103 , which coincides with the predicted optimum range of @xmath2 , while stereoscopy definitely brakes down at @xmath104 , as predicted by theory ( fig .
8 and eq .
10 ) . during the solar cycle ( thin solid curve ) , the spacecraft separation angle @xmath105 ( dotted curve ) , the stereoscopic accuracy @xmath106 ( dashed curve ) , and the expected number of stereoscopable active regions @xmath107 ( curve with grey area ) as a function of time during a full 16-year mission cycle of the stereo mission.,scaledwidth=90.0% ] there are different factors that affect the quality or feasibility of solar stereoscopy , such as ( i ) the availability of large active regions ( which varies statistically as a function of solar cycle ) , ( ii ) the simultaneous viewing by both spacecraft stereo / a and b ( which depends on the spacecraft separation angle ) , ( iii ) the geometric foreshortening that affects the stereoscopic correspondence problem ( which depends on the center - to - limb distance for each spacecraft view ) , and ( iv ) the time of the central meridian passage of the active region for a viewer from earth ( which determines the symmetry of views for both spacecraft , where minimum confusion occurs in the stereoscopic correspondence identification ) .
all but the first factor depend on the spacecraft separation angle @xmath105 , which is a specific function of time for the stereo mission ( with a complete cycle of 16 years ) . in order to assess the science return of the stereo mission or future missions with stereoscopic capabilities , it is instructive to quantify the statistical probability of acceptable stereoscopic results as a function of spacecraft separation angle or time .
we already quantified the quality of stereoscopy @xmath98 as a function of the stereoscopic angle in eq .
let us define the number probability @xmath108 of existing active regions at a given time @xmath109 with a squared sinusoidal modulation during the solar cycle , @xmath110 where @xmath111 is the maximum number of active regions existing on the total solar surface during the maximum of the solar cycle , @xmath112 is the time of the solar minimum ( e.g. , @xmath113 ) , and @xmath114 yrs the current average solar cycle length .
the second effect is the overlapping area on the solar surface that is simultaneously seen by both spacecraft stereo / a and b , which decreases linearly with the spacecraft separation angle from 50% at @xmath115 ( with @xmath116 at the start of spacecraft separation ) to 0% at @xmath117 ( with @xmath118 at maximum separation ) , and then increases linearly again for the next quarter phase of a mission cycle . if we fold the variation @xmath108 of the solar cycle ( eq .
11 ) with the triangular stereoscopic overlap area variation @xmath106 together , we obtain a statistical probability for the number of stereoscopically triangulable active regions .
however , the number of accurate stereoscopical triangulations scales with the quality factor @xmath98 ( eq . 10 ) , where the spacecraft separation angle @xmath105 is a piece - wise linear ( triangular ) function of time @xmath109 according to the spacecraft orbit .
essentially we are assuming that the probability of successfull stereoscopic triangulations at a given time scales with the quality factor or feasibility of accurate stereoscopy at this time .
so , we obtain a combined probability of stereoscopically triangulable active regions of , @xmath119 \ .\ ] ] in fig .
9 we show this combined statistical probability of feasible stereoscopy in terms of the expected number of active regions for a full mission cycle of 16 years , from 2006 to 2022 .
it shows that the best periods for solar stereoscopy are during 2012 - 2014 , 2016 - 2017 , and 2021 - 2023 .
after the stereo mission reached for the first time a full @xmath7 view of the sun this year ( 2007 feb 6 ) , the two stereo a and b spacecraft covered also for the first time the complete range of stereoscopic viewing angles from @xmath58 to @xmath70 .
we explored the feasibility of stereoscopic triangulation for coronal loops in the entire angular range by selecting 5 datasets with viewing angles at @xmath29 and @xmath120 . because previous efforts for solar stereoscopy covered only a range of small stereoscopic angles ( @xmath121 ) , we had to generalize the stereoscopic triangulation code for large angles up to @xmath122 .
we find that stereoscopy of coronal loops is feasible with good accuracy for cases in the range @xmath123 , a range that is also theoretically predicted by taking into account the triangulation errors due to finite spatial resolution and confusion in the stereoscopic correspondence identification in image pairs , which is hampered by projection effects and foreshortening for viewing angles near the limb .
accurate stereoscopy ( within a factor of 2 of the best possible accuracy ) is predicted for a spacecraft separation angle range of @xmath124 .
based on this model we predict that the best periods for stereoscopic 3d reconstruction during a full 16-year strereo mission cycle occur during 2012 - 2014 , 2016 - 2017 , and 2021 - 2023 , taking the variation in the number of active regions during the solar cycle into account also .
why is the accuracy of stereoscopic 3d reconstruction so important ?
solar stereoscopy has the potential to quantify the coronal magnetic field independently of conventional 2d magnetogram and 3d vector magnetograph extrapolation methods , and thus serves as an important arbiter in testing theoretical models of magnetic potential fields , linear force - free field models ( lfff ) , and nonlinear force - free field models ( nlfff ) .
a benchmark test of a dozen of nlfff codes has been compared with stereoscopic 3d reconstruction of coronal loops and a mismatch in the 3d misalignment angle of @xmath125 has been identified ( derosa et al .
2009 ) , which is attributed partially to the non - force - freeness of the photospheric magnetic field , and partially to insufficient constraints of the boundary conditions of the extrapolation codes .
empirical estimates of the error of stereoscopic triangulation based on the non - parallelity of loops in close proximity has yielded uncertainties of @xmath126 .
thus the residual difference in the misalignment is attributed to either the non - potentiality of the magnetic field ( in the case of potential field models ) , or to the non - force - freeness of the photospheric field ( for nlfff models ) .
we calculated also magnetic potential fields here for all stereoscopically triangulated active regions and found mean misalignment angles of @xmath127 , which improved to @xmath128 for a nonlinear force - free model , which testifies the reliability of stereoscopic reconstruction for the first time over a large angular range .
the only case where stereoscopy clearly fails is found for an extremely large separation angle of ( @xmath129 ) , which is also reflected in the largest deviation of misalignment angles found ( @xmath130 , @xmath131 ) . based on these positive results of stereoscopic accuracy over an extended angular range from small to large spacecraft separation angles
we anticipate that 3d reconstruction of coronal loops by stereoscopic triangulation will continue to play an important role in testing theoretical magnetic field models for the future phases of the stereo mission , especially since stereoscopy of a single image pair does not require a high cadence and telemetry rate at large distances behind the sun .
this work is supported by the nasa stereo mission under nrl contract n00173 - 02-c-2035 .
the stereo/ secchi data used here are produced by an international consortium of the naval research laboratory ( usa ) , lockheed martin solar and astrophysics lab ( usa ) , nasa goddard space flight center ( usa ) , rutherford appleton laboratory ( uk ) , university of birmingham ( uk ) , max - planck - institut fr sonnensystemforschung ( germany ) , centre spatiale de lige ( belgium ) , institut doptique thorique et applique ( france ) , institute dastrophysique spatiale ( france ) . the usa institutions were funded by nasa ; the uk institutions by the science & technology facility council ( which used to be the particle physics and astronomy research council , pparc ) ; the german institutions by deutsches zentrum fr luft- und raumfahrt e.v .
( dlr ) ; the belgian institutions by belgian science policy office ; the french institutions by centre national detudes spatiales ( cnes ) , and the centre national de la recherche scientifique ( cnrs ) . the nrl effort was also supported by the usaf space test program and the office of naval research .
# 1 [ aschwanden , m.j . ,
wlser , j.p . , nitta , n. , lemen , j. 2008 , , 827 . ]
[ aschwanden , m.j .
2009 , , 31 . ]
[ aschwanden , m.j . and
sandman , a.w .
2010 , , 723 . ]
[ aschwanden , m.j .
2012 , sol.phys .
( subm . ) , _ a nonlinear force - free magnetic field approximation suitable for fast forward - fitting to coronal loops . i. theory _ , http : www.lmsal.com/~aschwand / eprints/2012_fff1.pdf ] [ aschwanden , m.j . and malanushenko , a .
2012 , sol.phys .
( subm ) , _ a nonlinear force - free magnetic field approximation suitable for fast forward - fitting to coronal loops .
numeric code and tests _ , http://www.lmsal.com/~aschwand/eprints/2012_fff2.pdf ] [ aschwanden , m.j . , wuelser , j .-
p . , nitta , n.v . ,
lemen , j.r . , schrijver , c.j . ,
derosa , m. , and malanushenko , a. 2012 , apj ( subm ) , _ first 3d reconstructions of coronal loops with the stereo a and b spacecraft : iv .
magnetic field modeling with twisted force - free fields _ , http://www.lmsal.com/~aschwand/eprints/2012_stereo4.pdf , http://www.lmsal.com/~aschwand/movies/stereo_fff_movies ] [ bemporad , a. 2009 , , 298 . ]
[ berger , t.e . ,
depontieu , b. , fletcher , l. , schrijver , c.j . ,
tarbell , t.d . , and title , a.m. 1999 , , 409 . ] [ derosa , m.l . ,
schrijver , c.j . ,
barnes , g. , leka , k.d . , lites , b.w .
, aschwanden , m.j . , amari , t. , canou , a. , mctiernan , j.m . ,
regnier , s. , thalmann , j. , valori , g. , wheatland , m.s . ,
wiegelmann , t. , cheung , m.c.m . ,
conlon , p.a . ,
fuhrmann , m. , inhester , b. , and tadesse , t. 2009 , , 1780 . ]
[ feng , l. , inhester , b. , solanki , s. , wiegelmann , t. , podlipnik , b. , howard , r.a . , and wlser , j.p .
2007 , , l205 . ]
[ feng , l. , inhester , b. , solanki , s.k . , wilhelm , k. , wiegelmann , t. , podlipnik , b. , howard , r.a . ,
plunkett , s.p . ,
wlser , j.p . , and
gan , w.q .
2009 , , 292 . ]
[ howard , r.a . , howard , r.a . , moses , j.d . ,
vourlidas , a. , newmark , j.s . ,
socker , d.g . ,
plunkett , s.p . ,
korendyke , c.m . ,
cook , j.w . , hurley , a. , davila , j.m . and 36 co - authors , 2008 , , 67 . ]
[ inhester , b. 2006 , arxiv e - print : astro - ph/0612649 . ]
[ kaiser , m.l . ,
kucera , t.a . ,
davila , j.m .
cyr , o.c .
, guhathakurta , m. , and christian , e. 2008 , , 5 . ]
[ liewer , p.c . , dejong , e.m . ,
hall , j.r . , howard , r.a . , thompson , w.t . , culhane , j.l . , bone , l. , van driel - gesztelyi , l. 2009 , , 57 . ]
[ sandman , a. , aschwanden , m.j . ,
derosa , m. , wlser , j.p . , and alexander , d. 2009 , , 1 . ]
[ sandman , a.w .
and aschwanden , m.j .
2011 , , 503 . ]
[ thompson , w.t .
2006 , , 791 . ]
[ thompson , w.t .
2011 , , 1138 . ]
[ thompson , w.t . ,
davila , j.m .
, st . cyr , o.c . , and reginald , n.l .
2011 , , 215 . ]
[ wlser , j.p .
, lemen , j.r .
, tarbell , t.d . ,
wolfson , c.j . ,
cannon , j.c . ,
carpenter , b.a . ,
duncan , d.w . ,
gradwohl , g.s . ,
meyer , s.b . ,
moore , a.s . , and 24 co - authors , 2004 , spie * 5171 * , 111 . ] | we performed for the first time stereoscopic triangulation of coronal loops in active regions over the entire range of spacecraft separation angles ( @xmath0 , and @xmath1 ) .
the accuracy of stereoscopic correlation depends mostly on the viewing angle with respect to the solar surface for each spacecraft , which affects the stereoscopic correspondence identification of loops in image pairs . from a simple theoretical model we predict
an optimum range of @xmath2 , which is also experimentally confirmed .
the best accuracy is generally obtained when an active region passes the central meridian ( viewed from earth ) , which yields a symmetric view for both stereo spacecraft and causes minimum horizontal foreshortening .
for the extended angular range of @xmath3 we find a mean 3d misalignment angle of @xmath4 of stereoscopically triangulated loops with magnetic potential field models , and @xmath5 for a force - free field model , which is partly caused by stereoscopic uncertainties @xmath6 .
we predict optimum conditions for solar stereoscopy during the time intervals of 20122014 , 20162017 , and 20212023 . |
serial double quantum dots are ideal systems to investigate various quantum mechanical effects such as molecular binding@xcite or coherent dynamics@xcite between the constituent dots .
furthermore , they are considered as an implementation of a charge@xcite or spin qubit.@xcite elaborate experimental techniques were developed to control and characterize double - dot structures,@xcite and many information about the system can be deduced from the electric conductance through the device.@xcite recent experiments include the measurements of quantum mechanical level repulsion due to interdot coupling@xcite as well as due to external magnetic fields,@xcite the detection of molecular states in a double dot dimer,@xcite and the observation of coherent time evolution of the dot states.@xcite transport through serial double dots , as depicted in fig .
[ fig : model ] , inherently visualizes the basic quantum mechanical concept of coherent superposition of charge states.@xcite the states that are coupled to the left and right lead , the localized states in the left and right dot , respectively , are no energy eigenstates of the double dot .
this leads to oscillations of the electron in the double dot as it was shown in recent experiments.@xcite to account for this internal dynamics , descriptions using classical rates only , are insufficient , which is why approaches including non - diagonal density matrix elements for the double dot have been developed.@xcite . schematic energy profile for a double dot coupled in series to two reservoirs .
each reservoir is coupled to the dot of the corresponding side by the coupling strength @xmath0 .
the inter - dot coupling is determined by @xmath1 .
the energies of the two dot states are characterized by the mean energy @xmath2 and their relative distance @xmath3 . ] in this paper , we propose to use a serial double quantum dot to probe another consequence of quantum mechanics : the energy level renormalization of the quantum dot levels due to tunnel coupling to a reservoir . this idea is based on two properties of a serial dot system .
first , the left and right dot levels are tunnel - coupled to different reservoirs . since the level renormalization is a function of the level energy , the lead chemical potential , and the tunnel coupling , the energy shift of the left and right dot levels is , in general , different .
second , the conductance through the double dot is very sensitive to the difference of the energy levels .
it shows a resonant behavior with the width given by the tunnel couplings,@xcite which can be much smaller than the temperature .
this sharpness of the resonance makes the conductance a valuable experimental tool , for example to measure the shell structure of quantum dots.@xcite it is well known@xcite that tunnel coupling to reservoirs renormalizes the energy levels . in single - dot geometries
such an energy renormalization is only accessible in transport of higher order in the tunnel coupling strength .
as we will show below , this is different for the serial double dot geometry , for which renormalization effects are visible in the conductance already in the limit of weak dot - lead coupling , described by transport to first order ( sequential tunneling ) in the tunnel - coupling strength @xmath4 .
the paper is organized as follows : in sec .
[ sec : model ] we present the model hamiltonian for the double dot.@xcite in sec . [ sec : kineq ] we derive the stationary density matrix and the @xmath5current for arbitrary bias voltages . in sec .
[ sec : discussion ] , we discuss how renormalization effects appear in the @xmath5current through the double dot .
we draw our conclusions in sec . [
sec : conclusions ] , followed by some technical notes in the appendices . in app .
[ app : diagramaticrules ] we make explanatory notes of the diagrammatic technique , and in app . [ app : isospin ] we give an illustrative reformulation of the master equation in terms of a pseudo spin .
we consider a double quantum dot , contacted in series , which is described by the hamiltonian@xcite @xmath6 the first part of the hamiltonian describes the electric contacts on the left @xmath7 and right @xmath8 side .
these contacts are modeled by large reservoirs of noninteracting electrons @xmath9 . here
@xmath10 denote the annihilation and creation operators for electrons in the reservoir @xmath11 with spin @xmath12 .
the reservoirs are assumed to be in equilibrium , so that they can be characterized by the fermi distribution @xmath13 .
an applied bias voltage @xmath14 is modeled by different chemical potentials in the left and right contact @xmath15 .
the second part of the hamiltonian , @xmath16 , describes two dots , containing one electronic level each , which are coupled by the coulomb interaction : @xmath17 here , @xmath18 and @xmath19 are the occupation number operators for dot @xmath20 with spin @xmath21 , where @xmath22 being the annihilation and creation operators of an electron on dot @xmath23 with spin @xmath21 .
each dot consists of a single electronic level at the energy @xmath24 measured relative to the equilibrium chemical potential of the leads .
we parameterize the levels by their average energy @xmath25 and their difference @xmath26 , so that @xmath27 .
double occupation of one individual dot is associated with the intradot charging energy @xmath28 .
simultaneous occupation of the both dots with one electron each costs the interdot charging energy @xmath29 with @xmath30 .
states with three or more electrons in the double dot are not considered in the following .
the remaining eigenstates of @xmath16 , then , are : both dots empty @xmath31 , one electron with spin @xmath12 in the left @xmath32 or right dot @xmath33 , and one electron in each dot @xmath34 .
we assume that the intra - dot charging energy always exceeds the lead fermi energies .
therefore the states with two electrons in the same dot @xmath35 and @xmath36 will have a vanishing occupation probability .
however , these states will appear as intermediate ( virtual ) states in our calculation , providing a natural high - energy cut - off . the third part @xmath37 of the hamiltonian eq .
( [ hamiltonian ] ) describes both , tunneling between the two dots , @xmath38 , as well as tunneling between dots and leads , @xmath39 , @xmath40 due to the serial geometry , an electron from the right ( left ) reservoir can only tunnel to the right ( left ) dot .
the tunnel coupling of reservoir @xmath41 to the corresponding dot is characterized by the coupling strength @xmath42 .
we consider only spin conserving tunneling processes , and assume flat bands in the reservoirs , which yields energy independent couplings @xmath0 .
furthermore , we choose the interdot tunnel coupling amplitude @xmath43 as a positive , real parameter , which can be always achieved by a proper gauge transformation .
in the following section we calculate the stationary reduced density matrix @xmath44 for the double dot system and the @xmath5current through the system .
the reduced density matrix of the double - dot is obtained from the density matrix of the whole system by integrating out the reservoir degrees of freedom .
the liouville equation for the reduced density matrix then has the following structure : @xmath45 + [ h_{\rm \d},\bm \rho_{\rm st } ] + \hat{\s}\bm \rho_{\rm st}\ , .
\label{master}\end{aligned}\ ] ] the first two parts represent the internal dynamics on the double dot , which depends on the level separation @xmath46 and the interdot coupling @xmath43 . the third part of eq .
( [ master ] ) accounts for the tunnel coupling between double dot and external reservoirs .
the fourth order tensor @xmath47 contains imaginary and real parts , associated with particle transfer processes and with tunnel induced energy renormalization of the dot levels , respectively .
the latter has been neglected in previous works.@xcite we calculate @xmath47 using a real - time diagrammatic approach@xcite as explained in app .
[ app : diagramaticrules ] .
also alternative methods are available such as bloch - redfield theory.@xcite in the following we concentrate on the limit of weak tunnel coupling between double dot and leads .
therefore , we calculate @xmath47 to lowest order in the tunnel - coupling strength @xmath48 , which defines the so - called sequential - tunneling approximation .
this approximation implies that all tunneling events are independent from each other , which is fulfilled for @xmath49 . since the correlations generated in the bath during a tunnel process decay on the time scale @xmath50,@xcite ( this follows from the dependence of the tunneling line in fig .
[ fig : contour]-[fig : diagexam ] on its extension in time ) , while the average time between consecutive tunneling events is given by the inverse of the coupling strength @xmath51 , higher order , coherent tunneling events are suppressed by the condition @xmath49 and may be neglected .
the energy eigenstates of the double dot subsystem @xmath52 are the bonding and anti - bonding states with energies @xmath53 where @xmath54 denotes their energy splitting .
this identifies @xmath55 as frequency of the charge oscillations,@xcite and @xmath43 as minimum distance between the bonding and anti - bonding eigenstates as function of the left and right energy level.@xcite
if the splitting exceeds the intrinsic broadening of the levels , @xmath56 , then the internal oscillations are fast , and transport through the double - dot system takes place through two separate incoherent levels . in this case , off - diagonal matrix elements of the stationary density matrix vanish , which can be seen from the expansion of the liouville equation @xmath57 , where @xmath58 denotes the matrix element @xmath59 of the reduced density .
the more interesting transport regime is in the opposite limit , @xmath60 , where the external coupling strongly modifies the internal dynamics , which is captured by the off - diagonal elements of the reduced density matrix.@xcite combined with the validity condition for sequential tunneling , i.e. @xmath61 this implies @xmath62 , i.e. , internal oscillations are slow as compared to the time scale for the correlations during a tunneling event . as a consequence ,
the localized states @xmath32 and @xmath33 can be used as eigenstates of the double dot in the calculation of @xmath47 , which facilitates the interpretation of the dynamics .
technically , the condition @xmath63 means that for a consistent theory , we do not only expand @xmath47 in eq .
( [ master ] ) to first order in @xmath64 , but also have to expand it to zeroth order in @xmath55 .
this is accomplished by replacing the energies @xmath65 arising in the calculation of @xmath47 by the mean level energy @xmath66 .
( therefore our formulas only contain the fermi functions at the average single particle level @xmath67 , while energies of the order of the interdot tunneling or the level separation are smeared by temperature ) .
it is worth to point out , that by using the localized states as basis of the @xmath47 ( i.e. by calculating the transition rates in the localized basis ) one is automatically limited to the regime @xmath68.@xcite the technical details of how to calculate @xmath47 are described in app .
[ app : diagramaticrules ] .
the master eq .
( [ master ] ) then must be solved under the constraint of probability normalization @xmath69=1 $ ] .
the stationary current @xmath70 is given by the time derivative of the expectation value of the total number of electrons in either the left or the right lead .
for the lowest - order expansion used in the present context , the current can alternatively be written in the form@xcite @xmath71\rangle= -\frac{e}{\hbar}\d { \rm i m } ( \sum_\s p^{\l\s}_{\r\s})\quad,\end{aligned}\ ] ] where @xmath72 denotes the imaginary part . in app .
[ app : isospin ] we give an analytical solution for the current as function of bias voltage and gate voltages . in the following we discuss our results . instead of working with an off - diagonal density matrix
one may switch to a pseudo spin representation of the problem
. then this double - dot transport problem shows similarities to the system of a quantum dot connected to ferromagnetic leads.@xcite this will be discussed in app .
[ app : isospin ] .
the stationary current takes the form @xmath73 the numerical factors @xmath74 and @xmath75 ( the explicit form is given in app .
[ app : isospin ] ) depend only on the tunnel coupling constants @xmath76 , @xmath77 , and @xmath1 as well as on the fermi distribution functions @xmath78 and @xmath79 of the left and right lead , but not on the level energy difference @xmath3 .
the current in eq .
( [ current ] ) shows the well - known@xcite lorentzian dependence on the energy separation @xmath80 between left and right dot level .
however the energy separation is affected by the renormalization of the bare localized levels : @xmath81 this is the central statement of the present paper .
the energy shift @xmath82 of the energy level in dot @xmath41 , caused by the external tunnel coupling , is given by @xmath83 with @xmath84 here , @xmath85 denotes the real part , @xmath86 is the digamma function , and @xmath87 the leads chemical potentials .
we want to emphasize that this energy renormalization is not due to a capacitative but rather due to the tunnel coupling to the reservoirs .
furthermore , it vanishes in the noninteracting case @xmath88 . the intradot charging energy @xmath28 ( which we usually treat as infinite to avoid double occupation of one dot ) serves as a natural cut off for the energy renormalization in eq .
( [ renorm ] ) .
this is the reason why we allowed the intermediate states @xmath89 in app .
[ app : diagramaticrules ] to occupy these states .
the energy shift of the localized levels is proportional to the tunnel coupling strength and depends on the dot level positions relative to the fermi energy .
the renormalized level separation as function of the bias voltage is plotted in fig . [
fig : i - v]a .
the renormalized level separation @xmath80 reaches a ( local ) extremum each time , when the fermi energy of a lead becomes resonant with the energy needed for single ( @xmath90 ) or double occupation ( @xmath91 ) .
upper part : renormalized level spacing @xmath80 ( solid line ) between the electronic levels in the left and right dot as function of the transport voltage @xmath14 .
@xmath80 is extremal , when the chemical potential of a lead aligns with the energy needed for either single ( @xmath2 ) or double occupation ( @xmath92 ) .
lower part : current - voltage characteristics for bare ( dashed line ) and renormalized level spacing ( solid line ) .
renormalization of energy levels leads to an asymmetric current - voltage characteristic .
the current increases ( decreases ) whenever the level spacing is reduced ( increased ) with respect to the bare value .
plot parameters are : @xmath93 , @xmath94 , @xmath95 , and @xmath96 . ]
[ fig : i - v]b shows the current as function of the transport voltage taking the level shift into account ( solid line ) . by neglecting the level shifts ( dashed line in fig .
[ fig : i - v]b ) , the current shows a typical coulomb staircase .
the steps occur when a lead chemical potential aligns with an electronic level in the double dot .
since the bare energy level separation @xmath46 as well as the interdot tunneling @xmath43 shall be of the order of or smaller @xmath97 and we consider @xmath98 , the different single particle states are not resolved as individual steps in the @xmath99 staircase .
the tunnel induced renormalization leads to additional features on the staircase ( solid line in fig . [
fig : i - v]b ) . whenever the magnitude of the renormalized level spacing grows ( drops )
the current decreases ( increases ) .
this leads to a suppression or an enhancement of the current around the steps of the @xmath99 characteristic , leading to regions of negative differential conductance .
the width of these feature is of the order of the charging energy and can exceed temperature and coupling strength significantly . neglecting renormalization effects and assuming symmetric coupling to the reservoirs ( @xmath100 ) , the current through the double dot is an odd function of the transport voltage
( see dashed line in fig .
[ fig : i - v]b ) .
this is no longer the case when renormalization is taken into account ( see solid line in fig . [
fig : i - v]b ) .
the reason for this asymmetry is that even though the change of asymmetry , @xmath101 , caused by level renormalization is antisymmetric with respect to the bias voltage , this in not true for the total asymmetry @xmath102 due to the non - vanishing bare splitting @xmath3 ( see fig . [
fig : i - v]a ) . a comparable asymmetry in transport through two coupled dots
was recently observed by ishibashi _
et al_.@xcite and theoretically described by fransson _
et al_.@xcite however , a negative differential conductance feature can not be uniquely linked to such renormalization effects .
due to interface capacities the level positions in the left and right dot are always affected by the transport voltage in real experiments.@xcite to exclude the effect of interface capacities , we propose a different experiment : measuring the current @xmath103 at a constant transport voltage as function of the left and right gate voltages on the dots .
the resulting stability diagram is plotted in fig .
[ fig : e - e]a .
elastic sequential tunneling from the left to the right dot is possible if @xmath104 .
furthermore electron transport from the left to the right reservoir takes only place if the dot level for single ( @xmath2 ) or double occupation ( @xmath92 ) is located in the bias voltage window .
therefore the current resonance forms two stripes in the regions @xmath105 and @xmath106 .
away from the current stripes the occupation number of the left and right dot @xmath107 is fixed , and no current can cross the structure . for a detailed discussion on stability diagrams for transport through double dots we refer to the review of van der wiel .
_ et al_.@xcite upper panel : stability diagram @xmath103 of the current through the double dot in the nonlinear transport regime .
well inside the areas , separated by the black line , the occupation of the individual dots is fixed to the written values @xmath107 .
elastic sequential current can cross the structure for @xmath108 and either @xmath105 or @xmath109 resulting in two current stripes .
lower panel : gray scale plot of the current as function of the average single particle energy @xmath2 and the bare level separation @xmath110 .
the different renormalization of left and right level shifts the current maxima by @xmath111 ( dashed black - white line ) where @xmath112 .
this leads to a tilting of the current stripes relative to each other .
relevant plot parameters are @xmath113 , @xmath114 , @xmath115 , @xmath116 , and interdot charging energy @xmath117 . ] in the absence of renormalization effects , the current stripes would exactly coincide with the condition @xmath118 . by plotting the current as function of the mean level position @xmath25 and the relative energy difference @xmath119
, one would therefore expect a straight horizontal line .
instead , the maximum of the current follows the renormalization shift , where the condition @xmath112 is fulfilled , see fig .
[ fig : e - e]b .
the shift of the resonance is of order @xmath64 as shown in eq .
( [ renorm ] ) and can be small on the scale of bias voltage or temperature . the width of the current maxima in the stability diagram in fig .
[ fig : e - e ] is not determined by temperature but rather by the dominant coupling strength@xcite @xmath120 .
therefore the resonance width is sharp enough to be able to measure the renormalization of energy levels if @xmath121 as used in fig .
[ fig : e - e ] . in the nonlinear transport regime @xmath122 depends on @xmath2 and therefore the current stripes in fig .
[ fig : e - e ] are bent and tilted against each other .
this dependence can be used as a stringent experimental prove of the renormalization of energy levels .
due to internal cross capacities , always appearing in real experiments , the gate voltage of one dot is a linear function of the gate voltage of the other dot .
therefore the stability diagram @xmath123 as plotted in fig .
[ fig : e - e]a would experience a linear shear transformation
. however straight ( parallel ) lines stay straight ( parallel ) . thus
, cross capacities can not mimic the bending due to renormalization effects .
in real experiments in addition to the resonant current stripes explained here , further features can arise due to inelastic processes , cotunneling , or due to excited levels within the bias voltage window.@xcite these effects mainly lead to features within the triangles below the current strips in fig .
[ fig : e - e]a and are expected not to interfere with our presented results .
finally we compare our result obtained for the stationary current eq .
( [ currentapp ] ) with previous theoretical works . for this we set the fermi functions to @xmath124 and @xmath125 .
this simplifies the current to : @xmath126 neglecting renormalization effects ( setting @xmath127 ) , this equation reproduces eq .
( 4.19 ) in the paper by gurvitz,@xcite . choosing the voltages such
that the dot structure can also be doubly occupied , i.e. @xmath128 and @xmath129 one obtains eq .
( 4.18 ) of ref . .
several publications assume , that if the lead fermi energies are far away from the electronic states of the dots , then the principal value integrals ( eq . [ prinval ] ) , leading to the renormalization , can be neglected .
however the energy shifts are relevant on an energy scale given by the charging energy @xmath29 , as shown in fig .
[ fig : i - v]a .
therefore the assumption , that one can neglect renormalization effects and still exclude states with more than one electron occupying the double dot is not justified .
if a quantum dot is connected to a reservoirs , the tunnel coupling causes an energy renormalization of the electronic states .
we derived the conductance of a double dot connected in series to external reservoirs for general bias voltages and temperatures , taking into account these energy renormalizations .
we have shown , that the conductance of such a double dot structure is affected by the energy level shifts already in a lowest order expansion in the tunnel coupling strength , due to its high sensitivity on the relative detuning of energy levels .
therefore we propose to use a double - dot system as detector for these energy renormalization effects .
we present experimental consequences of the renormalization in the current - voltage characteristics and in the stability diagram for the double dot in the nonlinear transport regime . in the current - voltage characteristics we find prominent negative differential conductances in voltage windows of the order of the charging energy . in the stability diagram of the double dot
, we found that the current stripes arising as function of the gate voltages for left and right dot are tilted against each other and do not lie on a straight line , as it is the case when energy renormalization is neglected .
we showed that the tilting of the current stripes is resolvable even in the sequential tunneling regime ( i.e. for @xmath130 ) as long as the interdot tunneling , @xmath43 is of the same order or smaller than the external coupling @xmath131 .
we thank s. debald and b. kubala , for fruitful discussions .
this work was supported by the deutsche forschungsgemeinschaft via sfb 508 and under the emmy - noether program , through sfb 491 and grk 726 .
with the definition @xmath132 , the master equation eq .
( [ master ] ) can be written as : @xmath133 { \left|\chi_2\right\rangle } + { \left\langle\chi_1\right|}[h_{\rm \d},\rho_{\rm st } ] { \left|\chi_2\right\rangle}\notag\\ & & + \sum_{\chi_3,\chi_4 } \s^{\chi_1 \chi_3}_{\chi_2 \chi_4 } p^{\chi_3}_{\chi_4}\ , .
\label{app : master}\end{aligned}\ ] ] in the following we show , how we calculate the tensor @xmath134 , where @xmath135 are the localized eigenstates of @xmath16 , including the spin degree of freedom .
we apply a diagrammatic technique , where each class of tunneling processes can be represented by a diagram .
its general derivation can be found in ref . .
recently this technique was applied to anderson - like hamiltonians to investigate spin - valve effects,@xcite or signatures of the excitation spectrum in the coulomb blockade.@xcite within this approach , the tensor @xmath136 is represented as block diagram , which is a part of the keldysh time contour as shown in fig .
[ fig : contour ] .
the upper and lower line of the keldysh time contour @xmath137 represent the propagation of the double dot system forward and backward in time .
they connect the matrix element characterized by the labels on the left side with the matrix element characterized by the labels on the right side . in the sequential tunneling approximation
all transitions are allowed where a single electron first leaves and then reenters the double dot or vice versa .
the two tunnel hamiltonians are represented by vertices on the propagators .
these vertices are connected by the contraction of the lead fermi operators ( indicated by a dashed line ) .
each line is characterized by its energy @xmath138 , the spin @xmath21 of the transfered electron , as well as the corresponding reservoir label @xmath139 .
a vertex with an outgoing ( incoming ) tunneling line represents an electron leaving ( entering ) the double dot on the specified side @xmath41 .
all possible transitions in lowest order in the external coupling @xmath140 belong to one of the eight diagrams depicted in fig .
[ fig : alldiag ] .
sketch of the structure of a diagram .
the upper ( lower ) horizontal line denotes the forward ( backward ) propagator of the double dot system .
the keldysh time contour is labeled by @xmath137 , while the real time runs from left to right . ] all topologically different diagrams contributing to the tensor @xmath134 calculated in first order in the external coupling @xmath140 .
labeling of the eigenstates at the four corners and of the tunneling line like in first diagram .
@xmath89 labels an intermediate charge state of the double dot . ]
@xmath134 is given by the sum of all diagrams with the corresponding eigenstates at the four corners , see fig .
[ fig : alldiag ] .
the number of relevant diagrams is limited by spin and particle number conservation as well as to the serial system geometry . the rules to evaluate these diagrams in lowest order are : 1 .
draw the upper and lower time contour .
add two tunnel vertices in any topological different way .
the relevant criteria are the upper and lower contour , and the time ordering of the vertices on the real axes , not only on the kelysh time contour .
assign to each free segment of the contour a state of the double dot and the corresponding energy .
for bubble diagrams like in the lower row of fig .
[ fig : alldiag ] , an intermediate state @xmath89 participates .
the two vertices are connected by a tunnel line .
each tunnel line is labeled with the energy of the tunneling electron @xmath141 , its reservoir label @xmath41 and its spin @xmath21 .
spin and reservoir label of the tunneling electron are uniquely determined by the eigenstates involved in the tunneling processes .
3 . assign to each diagram the resolvent @xmath142 where @xmath143 is the difference between energies belonging to left going lines and energies belonging to right going lines ( the tunneling line as well as the propagators ) .
4 . the tunneling line connecting two vertices and labeled by the reservoir index @xmath41 gives rise to the factor @xmath144 here , the fermi function @xmath145)$ ] corresponds to a tunneling line that is backward directed in the keldysh time ordering ( compare fig .
[ fig : contour ] ) , and @xmath146 corresponds to a tunneling line forward directed in the keldysh time ordering .
each diagram gets a prefactor @xmath147 where @xmath148 is the number of vertices on the backward propagator .
( this leads to an @xmath149 for the diagrams in the upper row of fig . [
fig : alldiag ] . ) 6 .
sum over possible internal eigenstates @xmath89 and integrate over the energy @xmath138 of the tunneling electron .
relevant diagrams contributing to two specific entries of @xmath47 , in a lowest order expansion in @xmath140 .
every diagram corresponding to a specific entry is labeled by the same eigenstates at its four corners . ] in the parameter regime we are interested in , the following relations hold : @xmath150 .
therefore the energy difference between the single particle states is not resolved by the fermi functions in the reservoir , so that we have to approximate the eigenenergies of @xmath151 , @xmath32 , @xmath33 , @xmath152 by @xmath153 . while we exclude a double occupation of a single dot for the initial or final states by setting @xmath154 we allow the intermediate state @xmath89 to be in such a state .
these states have the eigenenergy @xmath155 . in fig .
[ fig : diagexam ] , we show as examples the diagrammatic expansion of the tensor elements @xmath156 and @xmath157 .
@xmath156 is purely imaginary and its magnitude has the meaning of a transition rate for a tunneling - in process starting from the empty double dot and resulting in a single electron with spin @xmath21 sitting in the left dot .
in contrast , @xmath157 also has a real part which renormalizes the energy levels .
calculated in lowest order in @xmath64 , each element of the tensor @xmath158 can be expressed by terms of the form : @xmath159 where @xmath160 and @xmath161 are either @xmath162 or @xmath163 . in this notation ,
the algebraic expression for @xmath157 is : @xmath164 where , within this appendix , we allow for an arbitrary spin degeneracy @xmath165 . since @xmath154 the imaginary part of the last row vanishes , however this is not the case for the real part , which causes the level renormalization .
the real part of the diagrams is determined by the principal values of the integrals in eq .
( [ prinval2 ] ) and can be expressed as a sum over digamma functions , see eq .
( [ renorm ] ) . since the hamiltonian given in eq .
( [ hamiltonian ] ) is independent of the orientation of the spin , each spin - realization of a charge state is equally probable .
we can therefore define @xmath166 , @xmath167 , and @xmath168 .
furthermore the stationary density matrix is diagonal in spin and particle number .
thus the reduced density matrix @xmath169 describing the double dot is given by the @xmath170 matrix @xmath171 the diagonal elements of the density matrix are the probabilities to find the double dot empty @xmath172 , the left @xmath173 or right dot @xmath174 singly occupied , or the two dots simultaneously occupied by one electron @xmath175 .
superpositions of the two single occupied states are possible @xmath176 .
one can define an effective tensor for @xmath47 , that only depends on the orbital part of the matrix elements ( denoted in the following formula by @xmath177 , @xmath178 , @xmath179 , @xmath180 ) and no longer on the spin variables .
the new tensor elements are defined by : @xmath181 here @xmath23 labels any possible spin - realization for the initial states , @xmath182 , and @xmath183 for the final states @xmath184 .
( due to spin degeneracies the two particle states are four fold degenerate , and the left and right states are each two - fold degenerate . )
the tunnel tensor @xmath136 is independent of the spin - realization @xmath23 .
the spin degeneracy appears only as a prefactor , but does not change the functional form of the elements .
for example , @xmath185 describing the transition from @xmath186 to @xmath187 is twice as big for spin - degenerate electrons as for spin - less fermions . on the other hand @xmath188 describing the loss term of @xmath187
is the same for spin - degenerate or spin - less fermions since @xmath189 .
this treatment of the spin allows a general solution of the problem including both , the case of spin polarized electrons and the case of spin degenerate electrons . for the interested reader ,
we specify the degeneracy of fermions in the further appendix by the variable @xmath190 : @xmath191 for electrons , @xmath192 for spin - less fermions .
instead of working with off - diagonal density matrix elements , we can switch to a pseudo spin representation . as any two level system , the @xmath193 hermitian submatrix of the singly occupied states in eq .
( [ dotdm ] ) can be treated as @xmath194 representation of a pseudo spin bloch vector @xmath195 . for a complete set of variables ,
we further introduce @xmath196 as the probability of a singly - occupied double dot .
such a pseudo spin representation is often used in the quantum information community.@xcite with this change of variables , the dynamics of the double dot system can be mapped on the motion of a spin in an external magnetic field .
this is in close analogy to the dynamics of a quantum dot connected to ferromagnetic leads.@xcite due to the serial geometry the external tunneling affects only the z - direction of the pseudo spin and the left and right contacts couple with a different sign to @xmath197 .
this is captured by the definitions @xmath198 and @xmath199 , which can be understood as pseudo - spin magnetizations of the leads . with this definitions
the occupation probabilities obey the following master equations : @xmath200 in equilibrium ( @xmath201 ) the diagonal matrix elements are given by the boltzmann statistics @xmath202/z,\ , p_2=g_\s^2 \exp[-(\bar{e}+u)/k_{\rm b}t]/z,\ , z = p_0+p_1+p_2 $ ] and the accumulation term as well as all components of the pseudo spin vanish .
the dynamics of the single particle state is described by a bloch - like equation : @xmath203\notag\\ \left(\frac{d { \bf i}}{dt}\right)_{\rm rel.}&= & \frac{1}{2}\sum_r \frac{\g_r}{\hbar } \left(f_r^-(\bar{e})+g_\s f_r(\bar{e}+u)\right ) { \bf i}\notag\end{aligned}\ ] ] three different terms can be identified in the bloch equation .
the term @xmath204 describes the accumulation of pseudo spin in z - direction due to the serial external coupling .
the third term looks like a rotation of the pseudo spin around a fictitious magnetic field @xmath207 , where @xmath80 denotes the renormalized level separation @xmath208)-({\rm l}\rightarrow{\rm r})\,,\quad\end{aligned}\ ] ] where @xmath85 denotes the real part .
the cauchy principal value integrals are defined in eq .
( [ prinval2 ] ) .
this third term describes coherent oscillations inside the double dot which mix the accumulated spin in z - direction with the other components .
the interdot tunneling characterized by @xmath43 leads to a precession of the isospin around the x - axes , while the energy separation between the dot levels results in a rotation around the z - axes .
it is important to note that the renormalized level separation between the dots changes due to the external coupling and it is not given by the bare level separation @xmath46 . in the following we outline the close analogy between the transport through a serial double dot described here and the dynamics in a spin valve described in ref . .
[ fig : spin ] sketches a spin valve , realized by a single level quantum dot placed between anti - aligned ferromagnets . relating the pseudo spin @xmath209 , in the present work with the real spin @xmath210 , in such a spin valve
, one can perform the following mapping .
the serial setup for the double dot system corresponds to the anti - aligned magnetization of the contacts in the spin valve .
furthermore the interdot tunneling translates to a transverse magnetic field in the single dot , while the level separation @xmath46 corresponds to the magnetic field component along the magnetization of the contacts .
finally the renormalization of the energy levels discussed here was introduced in the spin valve as an exchange field leading to the hanle effect.@xcite sketch of quantum dot spin valve . a single level quantum dot is connected to two ferromagnetic reservoirs with antiparallel magnetization .
the spin precesses around an external field with a component transverse ( @xmath211 ) and along ( @xmath212 ) the magnetization of the leads .
@xmath212 is modified by an exchange field arising due to the external coupling .
this exchange field is manifest in the transport properties of the spin valve . ]
the system of master equations can be solved analytically and the current as function of bias voltage and gate voltages has the following form : @xmath215 with the factors @xmath216 here @xmath217 denotes the opposite of @xmath41 and we use the abbreviations @xmath218 , @xmath219 , @xmath220 , @xmath221 , as well as the approximation @xmath222 .
j. m. elzerman , r. hanson , j. s. greidanus , vanbeveren , l.h.willems , s. de franceschi , l. m. k. vandersypen , s. tarucha , and l. p. kouwenhoven , phys . rev .
b * 67*,161308(r ) ( 2003 ) ; j. r. petta , a. c. johnson , c. m. marcus , m. p. hanson , and a. c. gossard , phys . rev . lett . *
93 * , 186802 ( 2004 ) .
j. knig , h. schoeller , and g. schn , phys .
* 76 * , 1715 ( 1996 ) ; j. knig , j. schmid , h. schoeller , and g. schn , phys .
b * 54 * , 16820 ( 1996 ) ; h. schoeller , in _ mesoscopic electron transport _ , edited by l.l . sohn , l.p .
kouwenhoven , and g. schn ( kluwer , dordrecht , 1997 ) ; j. knig , _ quantum fluctuations in the single - electron transistor _
( shaker , aachen , 1999 ) . | we study electron transport through double quantum dots in series .
the tunnel coupling of the discrete dot levels to external leads causes a shift of their energy .
this energy renormalization affects the transport characteristics even in the limit of weak dot - lead coupling , when sequential transport dominates .
we propose an experimental setup which reveals the renormalization effects in either the current - voltage characteristics or in the stability diagram . |
the search for the higgs boson and , hence , for the origin of electroweak symmetry breaking and fermion mass generation , remains one of the premier tasks of present and future high energy physics experiments .
fits to precision electroweak ( ew ) data have for some time suggested a relatively small higgs boson mass , of order 100 gev @xcite .
this is one of the reasons why the search for an intermediate mass higgs boson is particularly important @xcite . for the intermediate mass range ,
most of the literature has focussed on higgs boson production via gluon fusion @xcite and @xmath7 @xcite or @xmath8 @xcite associated production .
cross sections for standard model ( sm ) higgs boson production at the lhc are well - known @xcite , and while production via gluon fusion has the largest cross section by almost one order of magnitude , there are substantial qcd backgrounds .
a search for the very clean four - lepton signature from @xmath9 decay can find a higgs boson in the mass region @xmath10 gev , but due to the small branching fraction of this mode very large integrated luminosities , up to 100 fb @xmath5 or more , are required .
one can search for @xmath11 via @xmath12 decays with much lower luminosity @xcite , but with lower signal - to - background ratios .
the second largest production cross section is predicted for weak - boson fusion ( wbf ) , @xmath13 .
these events contain additional information in their observable quark jets .
techniques like forward jet tagging @xcite can then be exploited to significantly reduce the backgrounds .
wbf and gluon fusion nicely complement each other : together they allow for a measurement of the @xmath14 coupling ratio .
another feature of the wbf signal is the lack of color exchange between the initial - state quarks .
color coherence between initial- and final - state gluon bremsstrahlung leads to suppressed hadron production in the central region , between the two tagging - jet candidates of the signal @xcite .
this is in contrast to most background processes , which typically involve color exchange in the @xmath15-channel and thus lead to enhanced hadronic activity between the tagging jets .
we exploit these features , via a veto of soft jet activity in the central region @xcite .
while some attention has been given to intermediate - mass @xmath16 searches at the lhc in the framework of gluon fusion @xcite , production via weak boson fusion for the same decay mode has not yet been discussed in the literature .
thus , we provide a first analysis of intermediate - mass @xmath17 at the lhc ( and of the main physics and reducible backgrounds ) which demonstrates the feasibility of higgs boson detection in this channel , with very low luminosity . @xmath16
event characteristics are analyzed for dual leptonic decays to @xmath18 only , to avoid backgrounds from @xmath19 .
our analysis is a parton - level monte carlo study , using full tree - level matrix elements for the wbf higgs signal and the various backgrounds . in section [ sec : calc ]
we describe our calculational tools , the methods employed in the simulation of the various processes , and important parameters .
extra minijet activity is simulated by adding the emission of one extra parton to the basic signal and background processes .
generically we call the basic signal process ( with its two forward tagging jets ) and the corresponding background calculations `` 2-jet '' processes , and refer to the simulations with one extra parton as `` 3-jet '' processes . in section [ sec : analysis ] , using the 2-jet programs for the backgrounds , we demonstrate forward jet tagging , a @xmath20 veto and other important cuts which combine to yield an @xmath212/1 to 1/2 signal - to - background ( s / b ) ratio , depending on the higgs mass . in section [ sec : minijet ] we analyze the different minijet patterns in signal and background , using both the truncated shower approximation ( tsa ) @xcite to regulate the cross sections , and the gluon exponentiation model to estimate the minijet multiplicity @xcite . by exploiting the two most important characteristics of the extra radiation , its angular distribution and its hardness ,
the qcd backgrounds can be suppressed substantially by a veto on extra central jet emission . within the tsa and exponentiation models ,
probabilities are estimated for vetoing signal and background events , and are combined with the production cross sections of the previous section to predict signal and background rates in table [ summary ] .
these rates demonstrate the feasibility of extracting a very low background @xmath16 signal at the lhc .
our signal selection is not necessarily optimized yet .
the variables we identify for cuts are the most distinctive , but deserve a multivariate analysis with detector simulation .
we do construct an additional variable in section [ sec : disc ] which is not used for cuts , but rather can be used to extract the higgs boson mass from the final event sample .
we simulate @xmath22 collisions at the cern lhc , @xmath23 tev .
all signal and background cross sections are determined in terms of full tree level matrix elements for the contributing subprocesses and are discussed in more detail below . for all our numerical results we have chosen @xmath24 , @xmath25 gev , and @xmath26 , which translates into @xmath27 gev and @xmath28 when using the tree - level relations between these input parameters .
this value of @xmath29 is somewhat lower than the current world average of @xmath30 gev .
however , this difference has negligible effects on all cross sections , e.g. the @xmath31 signal cross section varies by about @xmath32 between these two @xmath1 mass values .
the tree level relations between the input parameters are kept in order to guarantee electroweak gauge invariance of all amplitudes . for all qcd effects , the running of the strong coupling constant
is evaluated at one - loop order , with @xmath33 .
we employ cteq4l parton distribution functions @xcite throughout . unless otherwise noted the factorization scale is chosen as @xmath34 min(@xmath35 ) of the defined jets .
the signal can be described , at lowest order , by two single - feynman - diagram processes , @xmath37 , _
i.e. _ @xmath38 and @xmath39 fusion where the weak bosons are emitted from the incoming quarks @xcite . because of the small higgs boson width in the mass range of interest
, these events can reliably be simulated in the narrow width approximation . from previous studies of @xmath40
@xcite and @xmath41 @xcite decays in weak boson fusion we know several features of the signal , which can be exploited here also : the centrally produced higgs boson tends to yield central decay products ( in this case @xmath42 ) , and the two quarks enter the detector at large rapidity compared to the @xmath1 s and with transverse momenta in the 20 to 100 gev range , thus leading to two observable forward tagging jets . for the study of a central jet veto
, we utilize the results of previous studies where we simulated the emission of at least one extra parton @xcite .
this was achieved by calculating the cross sections for the process @xmath43 , _
i.e. _ weak boson fusion with radiation of an additional gluon , and all crossing related processes .
an important additional tool for distinguishing the @xmath44 signal from various backgrounds is the anti - correlation of the @xmath1 spins , as pointed out in ref .
this is due to the preservation of angular momentum in the decay of the spin-0 higgs boson . of course
, we can observe only the angular distributions of the charged decay leptons , but this is sufficient .
the decay rate is proportional to @xmath45 . in the rest frame of the higgs boson ,
in which the @xmath46 or @xmath47 pairs are emitted back - to - back for @xmath42 production at threshold , this product is a maximum for the charged leptons being emitted parallel .
this characteristic is preserved and even enhanced when boosted to the lab frame , as the higgs boson in weak boson fusion is typically emitted with @xmath48 gev . given the h decay signature , the main physics background to our @xmath50 signal arises from @xmath51 production , due to the large top production cross section at the lhc and because the branching ratio @xmath52 is essentially @xmath53 . the basic process we consider is @xmath54 , which can be either @xmath55- or @xmath56-initiated , with the former strongly dominating at the lhc .
qcd corrections to this lead to additional real parton emission , _
i.e. _ to @xmath57 events .
relevant subprocesses are [ qcd_tt ] g q t |t q , g |q t |t |q , q @xmath58 events can be obtained similarly . for the case of no additional partons ,
the @xmath20 s from the decaying top quarks may be identified as the tagging jets . in this case , calculating the cross section for @xmath57 where the @xmath20 s are explicitly identified as the tagging jets serves to estimate the effect of additional soft parton emission , _
i.e. _ minijet activity in the central detector ; this is described in detail in sec . [ sec : minijet ] . at the same time
, we can identify a distinctly different , perturbative region of phase space , where the final - state light quark or gluon gives rise to one tagging jet , and one of the two decay @xmath20 s is identified as the other tagging jet . in this case
, @xmath58 may be used to estimate minijet activity for the hard process @xmath59 .
finally , there is a third distinct region of phase space , for the perturbative hard process @xmath60 , where the final state light quarks or gluons are the two tagging jets .
thus , the `` @xmath61 '' and `` @xmath62 '' calculations serve a dual purpose : to obtain the cross sections for the contribution of the perturbative processes where light quark or gluon jets lie in the region of phase space where they are experimentally identified as far - forward / backward tagging jets ; and to estimate the additional qcd radiation patterns for the next - lower - order perturbative @xmath51 process .
the @xmath63 and @xmath64 matrix elements were constructed using madgraph @xcite , while the @xmath62 matrix elements are from ref .
@xcite .
decays of the top quarks and @xmath1 s are included in the matrix elements ; however , while the @xmath1 s are allowed to be off - shell , the top quarks are required to be on - shell .
energy loss from @xmath65 is included to generate more accurate @xmath66 distributions . in all cases ,
the factorization scale is chosen as @xmath34 min(@xmath67 ) of the massless partons / top quarks .
the overall strong coupling constant factors are taken as @xmath68 , where the product runs over all light quarks , gluons and top quarks ; _ i.e. _ the transverse momentum of each additional parton is taken as the relevant scale for its production , irrespective of the hardness of the underlying scattering event .
this procedure guarantees that the same @xmath69 factors are used for the hard part of a @xmath51 event , independent of the number of additional minijets , and at the same time the small scales relevant for soft - parton emission are implemented .
the next obvious background arises from real - emission qcd corrections to @xmath42 production . for @xmath71 events
these background processes include @xcite [ qcd_ww ] q g q g w^+ w^- , q q q q w^+ w^- , which are dominated by @xmath15-channel gluon exchange , and all crossing related processes , such as q |q g g w^+ w^- , g g q |q w^+ w^- .
we call these processes collectively the `` qcd @xmath72 '' background
. we do not calculate cross sections for the corresponding @xmath73-jet processes , but instead follow the results of our analysis of the radiation patterns of qcd @xmath74 processes , detailed in sec .
[ sec : minijet ] , and apply those results here to estimate minijet veto probabilities . the factorization scale is chosen as for the higgs boson signal .
the strong coupling constant factor is taken as @xmath75 , _
i.e. _ , the transverse momentum of each additional parton is taken as the relevant scale for its production .
variation of the scales by a factor 2 or @xmath76 reveals scale uncertainties of @xmath77 , however , which emphasizes the need for experimental input or nlo calculations .
the @xmath38 background lacks the marked anti - correlation of @xmath1 spins seen in the signal . as a result
the momenta of the charged decay leptons will be more widely separated than in @xmath16 events .
these backgrounds arise from @xmath42 bremsstrahlung in quark(anti)quark scattering via @xmath15-channel electroweak boson exchange , with subsequent decay @xmath78 : qq qq w^+w^- [ eq : ew_ww ] navely , this ew background may be thought of as suppressed compared to the analogous qcd process in eq .
( [ qcd_ww ] ) .
however , it includes electroweak boson fusion , @xmath79 via @xmath80- or @xmath15-channel @xmath81-exchange or via @xmath82 4-point vertices , which has a momentum and color structure identical to the signal .
thus , it can not easily be suppressed via cuts .
the matrix elements for these processes were constructed using madgraph @xcite .
we include charged - current ( cc ) and neutral - current ( nc ) processes , but discard s - channel ew boson and t - channel quark exchange processes as their contribution was found to be @xmath83 only , while adding significantly to the cpu time needed for the calculation . in general , for the regions of phase space containing far - forward and -backward tagging jets , s - channel processes are severely suppressed .
we refer collectively to these processes as the `` ew @xmath72 '' background .
both @xmath1 s are allowed to be off - shell , and all off - resonance graphs are included . in addition
, the higgs boson graphs must be included to make the calculation well - behaved at large @xmath1-pair invariant masses .
however , these graphs include our signal processes and might lead to double counting .
thus , we set @xmath84 to 60 gev in the ew @xmath72 background to remove their contribution .
a clean separation of the higgs boson signal and the ew @xmath72 background is possible because interference effects between the two are negligible for the higgs boson mass range of interest .
again we will need an estimate of additional gluon radiation patterns .
this was first done for ew processes in ref .
@xcite , but for different cuts on the hard process , and again for ew @xmath85 processes in ref .
we reanalyze the ew @xmath85 case in sec .
[ sec : minijet ] and directly apply the resulting minijet emission probabilities here .
the ew @xmath85 and ew @xmath72 backgrounds are quite similar kinematically , which justifies the use of the same veto probabilities for central jets .
the leptonic decay of @xmath87 s provides a source of electrons , muons and neutrinos which can be misidentified as @xmath1 decays .
thus , we need to study real - emission qcd corrections to the drell - yan process @xmath88 . for @xmath89 events
these background processes include @xcite [ qcd_tau ] q g q g ^+ ^- , q q q q ^+ ^- , which are dominated by @xmath15-channel gluon exchange , and all crossing - related processes , such as q |q g g ^+ ^- , g g q |q ^+ ^- .
all interference effects between virtual photon and @xmath90-exchange are included .
we call these processes collectively the `` qcd @xmath85 '' background .
the cross sections for the corresponding @xmath91-jet processes , which we need for our modeling of minijet activity in the qcd @xmath85 background , have been calculated in refs .
similar to the treatment of the signal processes , we use a parton - level monte - carlo program based on the work of ref .
@xcite to model the qcd @xmath85 and @xmath92 backgrounds . from our study of @xmath41 in weak boson fusion @xcite , we know that the ew ( t - channel weak boson exchange ) cross section will be comparable to the qcd cross section in the phase space region of interest .
thus , we consider those processes separately , in a similar manner as for the ew @xmath72 contribution .
we use the results of ref .
@xcite for modeling the ew @xmath85 background .
the dual leptonic decays of the @xmath87 s are simulated by multiplying the @xmath93 cross section by a branching ratio factor of @xmath94 and by implementing collinear tau decays with helicity correlations included as in our previous analysis of @xmath41 @xcite .
the qcd processes discussed above lead to steeply falling jet transverse momentum distributions . as a result , finite detector resolution can have a sizable effect on cross sections .
these resolution effects are taken into account via gaussian smearing of the energies of jets/@xmath20 s and charged leptons .
we use = 5.2 e .009 , for jets ( with individual terms added in quadrature ) , based on atlas expectations @xcite . for charged leptons
we use = 2% .
in addition , finite detector resolution leads to fake missing - transverse - momentum in events with hard jets .
an atlas analysis @xcite showed that these effects are well parameterized by a gaussian distribution of the components of the fake missing transverse momentum vector , @xmath95 , with resolution ( p_x , p_y ) = 0.46 , for each component . in our calculations ,
these fake missing transverse momentum vectors are added linearly to the neutrino momenta .
the @xmath96 dual leptonic decay signal is characterized by two forward jets and the @xmath1 decay leptons ( @xmath97 ) . before discussing background levels and further details like minijet radiation patterns ,
we need to identify the search region for these hard @xmath98 events .
the task is identical to the higgs searches in @xmath99 which were considered previously @xcite . we can thus adopt the strategy of these earlier analyses and start out by discussing a basic level of cuts on the @xmath100 signal
throughout this section we assume a higgs mass of @xmath101 gev , but we do not optimize cuts for this mass .
the minimum acceptance requirements ensure that the two jets and two charged leptons are observed inside the detector ( within the hadronic and electromagnetic calorimeters , respectively ) , and are well - separated from each other : [ eq : basic ] & p_t_j 20 gev , |_j| 5.0 , r_jj 0.7 , + & p_t _ 20 gev , |_| 2.5 , r_j 0.7 .
a feature of the qcd @xmath72 background is the generally higher rapidity of the @xmath1 s as compared to the higgs signal : weak boson bremsstrahlung occurs at small angles with respect to the parent quarks , producing @xmath1 s forward of the jets .
thus , we also require both @xmath102 s to lie between the jets with a separation in pseudorapidity @xmath103 , and the jets to occupy opposite hemispheres : [ eq : lepcen ] _
j , min + 0.7 < _ _ 1,2 < _ j , max - 0.7 , _ j_1 _
j_2 < 0 finally , to reach the starting point for our consideration of the signal and various backgrounds , a wide separation in pseudorapidity is required between the two forward tagging jets , [ eq : gap ] _ tags = |_j_1-_j_2| 4.4 , leaving a gap of at least 3 units of pseudorapidity in which the @xmath102 s can be observed .
this technique to separate weak boson scattering from various backgrounds is well - established @xcite , in particular for heavy higgs boson searches .
line 1 of table [ ww_data ] shows the effect of these cuts on the signal and backgrounds for a sm higgs boson of mass @xmath104 gev .
overall , about @xmath105 of all @xmath106 events generated in weak boson fusion are accepted by the cuts of eqs .
( [ eq : basic]-[eq : gap ] ) ( for @xmath104 gev ) . somewhat surprisingly , the ew @xmath72 background rate reaches 2/3 of the qcd @xmath72 background rate already at this level .
this can be explained by the contribution from @xmath107 exchange and fusion processes which can produce central @xmath1 pairs and are therefore kinematically similar to the signal .
this signal - like component remains after the forward jet tagging cuts .
as is readily seen from the first line of table [ ww_data ] , the most worrisome background is @xmath1 pairs from @xmath51 production . of the 1080 fb at the basic cuts level ,
12 fb are from @xmath2 , 310 fb are from @xmath61 , and the remaining 760 fb arise from @xmath62 production .
the additional jets ( corresponding to massless partons ) are required to be identified as far forward tagging jets .
the @xmath62 cross section is largest because the @xmath2 pair is not required to have as large an invariant mass as in the first two cases , where one or both @xmath20 s from the decay of the top quarks are required to be the tagging jets . for the events where one or both of the @xmath20 s are not identified as the tagging jets
, they will most frequently lie between the two tagging jets , in the region where we search for the @xmath1 decay leptons .
vetoing events with these additional @xmath20 jets provides a powerful suppression tool to control the top background .
note that this does _ not _ require a @xmath20-tag , merely rejection of any events that have an additional jet , which in this case would be from a hadronically decaying @xmath20 .
we discard all events where a @xmath20 or @xmath108 jet with @xmath109 gev is observed in the gap region between the tagging jets , [ eq : bveto ] p_t_b > 20 gev , _
j , min < _ b < _ j , max .
this leads to a reduction of @xmath61 events by a factor 7 while @xmath62 events are suppressed by a factor 100 .
this results in cross sections of 43 and 7.6 fb , respectively , at the level of the forward tagging cuts of eqs .
( [ eq : basic]-[eq : gap ] ) , which are now comparable to the other individual backgrounds .
this is shown in the second line of table [ ww_data ] .
note that the much higher @xmath20 veto probability for @xmath62 events results in a lower cross section than that for @xmath61 events , an ordering which will remain even after final cuts have been imposed ( see below ) .
qcd processes at hadron colliders typically occur at smaller invariant masses than ew processes , due to the dominance of gluons at small feynman @xmath110 in the incoming protons .
we observe this behavior here , as shown in fig .
[ fig : mjj ] .
the three @xmath51 backgrounds have been combined for clarity , even though their individual distributions are slightly different .
we can thus significantly reduce much of the qcd background by imposing a lower bound on the invariant mass of the tagging jets : [ eq : mjj ] m_jj > 650 gev .
another significant difference is the angular distribution of the charged decay leptons , @xmath111 and @xmath112 , relative to each other . in the case of the higgs signal ,
the @xmath1 spins are anti - correlated , so the leptons are preferentially emitted in the same direction , close to each other .
a significant fraction of the various backgrounds does not have anti - correlated @xmath1 spins .
these differences are demonstrated in fig .
[ fig : angdist ] , which shows the azimuthal ( transverse plane ) opening angle , polar ( lab ) opening angle , and separation in the lego plot .
we exploit these features by establishing the following lepton - pair angular cuts : [ eq : ang ] _ e < 105^ , cos _
e > 0.2 , r_e < 2.2 . it should be noted that while these cuts appear to be very conservative , for higher higgs boson masses the @xmath113 and @xmath114 distribution broadens out to higher values , overlapping the backgrounds more . for @xmath115 gev these cuts are roughly optimized and further tightening would require greater integrated luminosity for discovery at this upper end of the mass range . because of the excellent signal - to - background ratio achieved below , we prefer to work with uniform acceptance cuts , instead of optimizing the cuts for specific higgs boson mass regions .
we also examine the distributions for lepton - pair invariant mass , @xmath116 , and maximum lepton @xmath35 , as shown in fig .
[ fig : mllptl ] for the case @xmath117 gev .
as is readily seen , the qcd backgrounds and ew @xmath72 background prefer significantly higher values for both observables .
thus , in addition to the angular variables , we find it useful to restrict the individual @xmath35 of the leptons , as well as the invariant mass of the pair : [ eq : adv ] m_e < 110 , p_t_e , < 120
. these are particularly effective against the top backgrounds , where the large top mass allows for very high-@xmath35 leptons far from the tagging jets , and against the ew @xmath72 background , where the leptons tend to be well - separated in the lego plot .
again , the cuts are set quite conservatively so as not to bias a lower higgs boson mass .
results after cuts ( [ eq : mjj]-[eq : adv ] ) are shown on the third line of table [ ww_data ] , for the case of a 160 gev higgs boson .
.signal rates @xmath118 for @xmath104 gev and corresponding background cross sections , in @xmath22 collisions at @xmath119 tev .
results are given for various levels of cuts and are labeled by equation numbers discussed in the text . on lines six the minijet veto is included .
line five gives the survival probabilities for each process , with @xmath120 gev .
the expected tagging jet identification efficiency is shown on the last line .
all rates are given in fb .
[ cols="<,^,^,^,^,^,^,^",options="header " , ] the high purity of the signal is made possible because the weak boson fusion process , together with the @xmath121 decay , provides a complex signal with a multitude of characteristics which distinguish it from the various backgrounds .
the basic feature of the @xmath31 signal is the presence of two forward tagging jets inside the acceptance region of the lhc detectors , of sizable @xmath35 , and of dijet invariant mass in the tev range .
typical qcd backgrounds , with isolated charged leptons and two hard jets , are much softer .
in addition , the qcd backgrounds are dominated by @xmath1 bremsstrahlung off forward scattered quarks , which give typically higher - rapidity charged leptons .
in contrast , the ew processes give rise to quite central leptons , and this includes not only the higgs signal but also ew @xmath72 and @xmath85 production , which also proceed via weak boson fusion .
it is this similarity that prevents one from ignoring ew analogs to background qcd processes , which a priori are smaller by two orders of magnitude in total cross section , but after basic cuts remain the same size as their qcd counterparts . for @xmath122 decays ,
lepton angular distributions are extremely useful for reducing the backgrounds even further .
the anti - correlation of @xmath1 spins in @xmath123 decay forces the charged leptons to be preferentially emitted in the same direction , close together in the lego plot .
this happens for a small fraction of the background only .
we have identified the most important distributions for enhancing the signal relative to the background , and set the various cuts conservatively to avoid bias for a certain higgs boson mass range .
there is ample room for improvement of our results via a multivariate analysis of a complete set of signal and background distributions , which we encourage the lhc collaborations to pursue .
additional suppression of the @xmath51 background may be possible with @xmath20 identification and veto in the @xmath124 gev region .
in addition to various invariant mass and angular cuts , we can differentiate between the @xmath1 s of the signal and @xmath125 backgrounds and the real @xmath87 s in the qcd and ew @xmath85 backgrounds .
this is possible because the high energy of the produced @xmath87 s makes their decay products almost collinear .
combined with the substantial @xmath35 of the @xmath86 system this allows for @xmath87-pair mass reconstruction .
the @xmath1 decays do not exhibit this collinearity due to their large mass , thus the angular correlation between the @xmath126 vector and the charged lepton momenta is markedly different .
our real-@xmath87 rejection makes use of these differences and promises to virtually eliminate the @xmath85 backgrounds .
we advocate taking advantage of an additional fundamental characteristic of qcd and ew processes .
color - singlet exchange in the @xmath15-channel , as encountered in higgs boson production by weak boson fusion ( and in the ew @xmath127 background ) , leads to additional soft jet activity which differs strikingly from that expected for the qcd backgrounds in both geometry and hardness : gluon radiation in qcd processes is typically both more central and harder than in wbf processes .
we exploit this radiation , via a veto on events with central minijets of @xmath109 gev , and expect a typical @xmath128 reduction in qcd backgrounds and about a @xmath129 suppression of ew backgrounds , but only about a @xmath130 loss of the signal . beyond the possibility of discovering the higgs boson in the @xmath122 mode , or confirmation of its existence , measuring the cross sections in both weak boson and gluon fusion will be important both as a test of the standard model and as a search for new physics .
for such a measurement , via the analysis outlined in this paper , minijet veto probabilities must be precisely known . for calibration purposes
, one can analyze @xmath127 events at the lhc .
the production rates of the qcd and ew @xmath127 events can be reliably predicted and , thus , the observation of the @xmath131 peak allows for a direct experimental assessment of the minijet veto efficiencies , in a kinematic configuration very similar to the higgs signal .
observation of sm @xmath44 at the lhc is possible for very low integrated luminosities , if the higgs boson lies in the mass range between about 130 and 200 gev .
weak boson fusion at the lhc will be an exciting process to study , for a weakly coupled higgs sector just as much as for strong interactions in the symmetry breaking sector of electroweak interactions .
this research was supported in part by the university of wisconsin research committee with funds granted by the wisconsin alumni research foundation and in part by the u. s. department of energy under contract no .
de - fg02 - 95er40896 . for recent reviews , see e.g. j.l .
rosner , comments nucl .
* 22 * , 205 ( 1998 ) [ hep - ph/9704331 ] ; k.hagiwara , ann .
nucl . part .
1998 , 463 ; w.j .
marciano , hep - ph/9902332 ; and references therein . for recent reviews , see e.g. s. dawson , [ hep - ph/9703387 ] ; m. spira , fortsch .
* 46 * , 203 ( 1998 ) ; and references therein . w. j. marciano and f. e. paige , phys .
lett . * 66 * , 2433 ( 1991 ) ; j. f. gunion , phys .
* b261 * , 510 ( 1991 ) .
a. stange , w. marciano , and s. willenbrock , phys .
* d50 * , 4491 ( 1994 ) , [ hep - ph/9404247 ] ; r. kleiss , z. kunszt , w. j. stirling , phys .
* b253 * , 269 ( 1991 ) ; h. baer , b. bailey , j. f. owens , phys .
* d47 * , 2730 ( 1993 ) . e.w . glover , j. ohnemus and s.s .
willenbrock , phys .
rev . * d37 * , 3193 ( 1988 ) ; v. barger , g. bhattacharya , t. han and b.a .
kniehl , phys .
* d43 * , 779 ( 1991 ) . m. dittmar and h. dreiner , phys .
* d55 * , 167 ( 1997 ) ; and [ hep - ph/9703401 ] . v. barger , r. j. n. phillips , and d. zeppenfeld , phys .
* b346 * , 106 ( 1995 ) .
r. n. cahn , s.d .
ellis , r. kleiss and w.j .
stirling , phys .
rev . * d35 * , 1626 ( 1987 ) ; v. barger , t. han , and r. j. n. phillips , phys . rev .
* d37 * , 2005 ( 1988 ) ; r. kleiss and w. j. stirling , phys .
* 200b * , 193 ( 1988 ) ; d. froideveaux , in _ proceedings of the ecfa large hadron collider workshop _ , aachen , germany , 1990 , edited by g. jarlskog and d. rein ( cern report 90 - 10 , geneva , switzerland , 1990 ) , vol ii , p. 444 ; m. h. seymour , _ ibid _ , p. 557 ; u. baur and e. w. n. glover , nucl .
* b347 * , 12 ( 1990 ) ; phys .
b252 * , 683 ( 1990 ) . v. barger , k. cheung , t. han , and r. j. n. phillips , phys .
rev . * d42 * , 3052 ( 1990 ) ; v. barger _
et al . _ ,
rev . * d44 * , 1426 ( 1991 ) ; v. barger , k. cheung , t. han , and d. zeppenfeld , phys . rev .
* d44 * , 2701 ( 1991 ) ; erratum phys .
rev . * d48 * , 5444 ( 1993 ) ; phys .
rev . * d48 * , 5433 ( 1993 ) ; v. barger _
et al . _ ,
* d46 * , 2028 ( 1992 ) .
d. dicus , j. f. gunion , and r. vega , phys .
* b258 * , 475 ( 1991 ) ; d. dicus , j. f. gunion , l. h. orr , and r. vega , nucl .
b377 * , 31 ( 1991 ) . y. l. dokshitzer , v. a. khoze , and s. troian , in _ proceedings of the 6th international conference on physics in collisions _ , ( 1986 ) ed .
m. derrick ( world scientific , 1987 ) p.365 ; j. d. bjorken , int .
j. mod .
phys . * a7 * , 4189 ( 1992 ) ; phys .
* d47 * , 101 ( 1993 ) . v. barger and r. j. n. phillips , phys .
lett . * 55 * , 2752 ( 1985 ) ; h. baer , v. barger , h. goldberg , and r. j. n. phillips , phys .
rev . * d37 * , 3152 ( 1988 ) .
d. rainwater , d. summers , and d. zeppenfeld , phys . rev . * d55 * , 5681 ( 1997 ) .
h. l. lai _ et al .
_ , phys .
* d55 * , 1280 ( 1997 ) , [ hep - ph/9606399 ] .
r. cahn and s. dawson , phys .
* 136b * , 196 ( 1984 ) .
d. rainwater and d. zeppenfeld , journal of high energy physics 12 , 005 ( 1997 ) .
d. rainwater , d. zeppenfeld and k. hagiwara , phys .
rev . * d59 * , 014037 ( 1999 ) . d. rainwater , r. szalapski , and d. zeppenfeld , phys . rev .
* d54 * , 6680 ( 1996 ) , [ hep - ph/9605444 ] . t. stelzer and w. f. long , comp .
comm . * 81 * , 357 ( 1994 ) , [ hep - ph/9401258 ] .
a. stange , private communication .
v. barger , t. han , j. ohnemus and d. zeppenfeld , phys .
rev . * d41 * , 2782 ( 1990 ) .
a. duff and d. zeppenfeld , phys .
rev . * d50 * , 3204 ( 1994 ) ; k. iordanidis and d. zeppenfeld , phys .
rev . * d57 * , 3072 ( 1998 ) , [ hep - ph/9709506 ] . s. d. ellis , r. kleiss , and w. j. stirling , phys .
* 154b * , 435 ( 1985 ) ; r. kleiss and w. j. stirling , nucl .
* b262 * , 235 ( 1985 ) ; phys .
* 180b * , 171 ( 1986 ) ; j. f. gunion , z. kunszt , and m. soldate , phys .
163b * , 389 ( 1985 ) ; erratum , phys .
* 168b * , 427 ( 1986 ) ; j. f. gunion and m. soldate , phys .
* d34 * , 826 ( 1986 ) ; r. k. ellis and r. j. gonsalves , in _ proc . of the workshop on super high energy physics _ , eugene , or ( 1985 ) , ed .
d. e. soper , p. 287 .
k. hagiwara and d. zeppenfeld , nucl .
b313 * , 560 ( 1989 ) .
v. barger , t. han and j. ohnemus , and d. zeppenfeld , phys .
lett . * 62 * , 1971 ( 1989 ) ; phys .
rev . * d40 * , 2888 ( 1989 ) .
f. a. berends _ et al .
_ , phys .
* b224 * , 237 ( 1989 ) . h. chehime and d. zeppenfeld , phys .
rev . * d47 * , 3898 ( 1993 ) .
w. w. armstrong _ et al .
_ , atlas technical proposal , report cern / lhcc/94 - 43 ( 1994 ) . | weak boson fusion promises to be a copious source of intermediate mass standard model higgs bosons at the lhc .
the additional very energetic forward jets in these events provide for powerful background suppression tools .
we analyze the @xmath0 decay mode for a higgs boson mass in the 130 - 200 gev range .
a parton level analysis of the dominant backgrounds ( production of @xmath1 pairs , @xmath2 and @xmath3 in association with jets ) demonstrates that this channel allows the observation of @xmath4 in a virtually background - free environment , yielding a significant higgs boson signal with an integrated luminosity of 5 fb@xmath5 or less .
weak boson fusion achieves a much better signal to background ratio than inclusive @xmath6 and is therefore the most promising search channel in the 130 - 200 gev mass range . |
as usual , we will say that @xmath11 is a convex body if @xmath1 is a convex , compact subset of @xmath8 equal to the closure of its interior .
we say that @xmath1 is origin - symmetric if @xmath12 , where @xmath13 for @xmath14 . for a set @xmath1
we denote by dim@xmath15 its dimension , that is , the dimension of the affine hull of @xmath1 .
we define @xmath16 to be the minkowski sum of @xmath17 .
we will also denote by @xmath18 the @xmath4-dimensional hausdorff measure , and if the body @xmath1 is @xmath4-dimensional we will call @xmath19 the volume of @xmath1 .
finally , let us denote by @xmath20 a hyperplane perpendicular to a unit vector @xmath21 , i.e. @xmath22 we refer to @xcite for general definitions and properties of convex bodies .
the slicing problem of bourgain @xcite is , undoubtedly , one of the major open problems in convex geometry asking if a convex , origin - symmetric body of volume one must have a large ( in volume ) hyperplane section .
more precisely , it asks whether there exists an absolute constant @xmath23 so that for any origin - symmetric convex body @xmath1 in @xmath8 @xmath24 the problem is still open , with the best - to - date estimate of @xmath25 established by klartag @xcite , who improved the previous estimate of bourgain @xcite , we refer to @xcite and @xcite for detailed information and history of the problem .
recently , alexander koldobsky proposed an interesting generalization of the slicing problem @xcite : does there exists an absolute constant @xmath26 so that for every even measure @xmath27 on @xmath8 , with a positive density , and for every origin - symmetric convex body @xmath1 in @xmath8 such that @xmath28 koldobsky was able to solve the above question for a number of special cases of the body @xmath1 and provide a general estimate of @xmath29 .
the most amazing fact here is that the constant @xmath26 in ( [ eq : cont ] ) can be chosen independent of the measure @xmath27 under the assumption that @xmath27 has even positive density .
in addition , koldobsky and the second named author were able to prove in @xcite that @xmath26 is of order @xmath30 if one assumes that the measure @xmath27 is @xmath31-concave .
we note that the assumption of positive density is essential for the above results and ( [ eq : cont ] ) is simply not true if this condition is dropped . indeed , to create a counterexample consider an even measure @xmath27 on @xmath32 uniformly distributed over @xmath33 points on the unit circle ,
then the constant @xmath26 in ( [ eq : cont ] ) will depend on @xmath34 . during the 2013 aim workshop on `` sections of convex bodies '' koldobsky asked
if it is possible to provide a discrete analog of inequality ( [ eq : cont ] ) : let @xmath35 be the standard integer lattice in @xmath8 , @xmath1 be a convex , origin - symmetric body , define @xmath36 , the number of points of @xmath35 in @xmath1 .
1em * question : * _ does there exist a constant @xmath37 such that _ @xmath38 _ for all convex origin - symmetric bodies @xmath11 containing @xmath4 linearly independent lattice points ? _
1em we note here that we require that @xmath1 contains @xmath4 linearly independent lattice points , i.e. , @xmath39 , in order to eliminate the degenerate case of a body ( for example , take a box @xmath40^{d-1}\times [ -20 , 20]$ ] ) whose maximal section contains all lattice points in the body , but whose volume may be taken to 0 by eliminating a dimension .
koldobsky s question is yet another example of an attempt to translate questions and facts from classical convexity to more general settings including discrete geometry .
the properties of sections of convex bodies with respect to the integer lattice were extensively studied in discrete tomography @xcite , where many interesting new properties were proved and a series of exciting open questions were proposed .
it is interesting to note that after translation many questions become quite non - trivial and counterintuitive , and the answer may be quite different from the continuous case .
in addition , finding the relation between the geometry of a convex set and the number of integer points contained in the set is always a non - trivial task .
one can see this , for example , from the history of khinchin s flatness theorem @xcite .
the main goal of this paper is to study koldobsky s question . in section 2
we will show the solution for the @xmath41-dimensional case .
the solution is based on the classical minkowski s first and pick s theorems from the geometry of numbers and gives a general idea of the approach to be used in sections 3 and 4 . in section 3
we apply a discrete version of the theorem of f. john due to t. tao and v. vu @xcite to give a partial answer to koldobsky s question and show that the constant @xmath37 can be chosen independent of the body @xmath1 and as small as @xmath42 .
we start section 4 with a case of unconditional bodies and present a simple proof that in this case @xmath37 can be chosen of order @xmath43 which is best possible .
after , we prove the discrete analog of brunn s theorem and use it to show that the constant @xmath37 , for the general case , can be chosen as small as @xmath44 .
in fact , we prove the slightly more general result that @xmath45 where the maximum is taken over all @xmath7-dimensional linear subspaces @xmath46 .
finally , we also provide a short observation that @xmath47 . * acknowledgment * : we are indebted to alexander koldobsky and fedor nazarov for valuable discussions .
let us start with recalling two classical statements in the geometry of numbers ( see @xcite , theorem 3.28 pg 134 and @xcite , theorem 2.8 pg 38 ) : [ th : m1 ] ( minkowski s first theorem ) let @xmath5 be an origin - symmetric convex body such that @xmath49 then @xmath1 contains at least one non - zero element of @xmath35 .
( pick s theorem ) let @xmath50 be an integral @xmath41-dimensional convex polygon , then @xmath51 where @xmath52 is the area of the polygon , @xmath53 is the number of lattice points in the interior of @xmath50 , and @xmath54 is the number of lattice points on the boundary . here
a polygon is called integral if it can be described as the convex hull of lattice points .
now we will use the above theorems to show that the constant @xmath37 in koldobsky s question can be chosen independently of a convex , origin - symmetric body in @xmath32 .
let @xmath1 be a convex origin - symmetric body in @xmath32 , @xmath55 , then @xmath56 * proof : * let @xmath57 , then by minkowski s theorem , since @xmath58 , there exists a non - zero vector @xmath59 .
then @xmath60 and @xmath61 where @xmath62 is the integer part of @xmath31 , and @xmath63 is the line containing @xmath64 and the origin .
next , consider @xmath65 , i.e. , the convex hull of the integral points inside @xmath1 .
@xmath50 is an integral @xmath41-dimensional convex polytope , and so by pick s theorem we get that @xmath66 using that @xmath67 .
thus @xmath68 since the minmal volume of an origin - symmetric integral convex polygon is at least 2 .
we now have that @xmath69
it is a standard technique to get a first estimate in slicing inequalities , i.e. @xmath70 , by using the classical f. john theorem , @xcite , @xcite , or @xcite , which claims that for every convex origin - symmetric body @xmath11 there exists an ellipsoid @xmath71 such that @xmath72 . in this section
we will use a recent discrete version of f. john s theorem , proved by t. tao and v. vu ( see @xcite ) to prove that the constant @xmath37 in koldobsky s question can be chosen independent of the origin - symmetric convex body @xmath11 .
we first recall the definition of a generalized arithmetic progression ( see @xcite for more details ) : let @xmath73 be an additive group , @xmath74 an @xmath4-tuple of non - negative integers and @xmath75 .
then a generalized symmetric arithmetic progression @xmath76 is a triplet @xmath77 .
in addition , define @xmath78 \cdot { { \boldsymbol v}}= \left\{n_1 { { \boldsymbol v}}_1 + \ldots + n_d { { \boldsymbol v}}_d : n_j \in [ -n_j , n_j ] \cap
\zz \text { for all } 1\leq j \leq d \right\}.\ ] ] the progression is called proper if the map @xmath79 is injective , @xmath80 is called its basis vectors , and @xmath4 its rank .
[ def : progression ] below is a version for @xmath35 of the discrete john theorem from @xcite ( theorem 1.6 there ) : [ th : dj ] let @xmath1 be a convex origin - symmetric body in @xmath8 .
then there exists a symmetric , proper , generalized arithmetic progression @xmath81 , such that @xmath82 and @xmath83 in addition @xmath84 now we are ready to state and prove our first estimate in koldobsky s question and prove that for any origin - symmetric convex body @xmath5 , @xmath39 , @xmath85 to prove ( [ eq : viajohn ] ) we apply the discrete john s theorem to get a symmetric , proper , generalized arithmetic progression @xmath86 as in definition [ def : progression ] . we note that if @xmath87 , then there exists a hyperplane @xmath20 such that @xmath88 and using ( [ eq : size ] ) we get @xmath89 by our assumption @xmath90 we have @xmath91 and so @xmath92 .
thus @xmath93 next we consider the case @xmath94 . without loss of generality ,
take @xmath95 , then define @xmath96 .
application of ( [ eq : size ] ) gives @xmath97 where the last inequality follows from the minimality of @xmath98 and we use ( [ eq : incl ] ) to claim that @xmath99 now we consider the volume covered by our progression .
take a fundamental parallelepiped @xmath100 let @xmath101\cdot { { \boldsymbol v}}$ ] , we notice that @xmath102 indeed from @xmath103 we get that the vertices of @xmath104 belong to @xmath105 for all @xmath106 and thus , by convexity , @xmath107 for all @xmath106 .
next @xmath108 where the last inequality follows from the fact that @xmath109 are independent vectors in @xmath35 and thus @xmath110 . finally , @xmath111
the goal of this section is to improve the estimate provided in section 3 .
we will need to consider counting points intersecting a body with a general lattice , and so we will adapt our notation slightly .
we refer to @xcite , @xcite and @xcite for the general facts and introduction on the properties of the cardinality of intersections of convex bodies and a lattice .
given a lattice @xmath112 we will take @xmath113 and , as before , if the lattice is omitted we will take the lattice to be the standard integer lattice of appropriate dimension .
we begin with the statement of minkowski s second theorem which is an extension of minkowski s first theorem ( theorem [ th : m1 ] above ) and can be found , for example , in @xcite ( theorem 1.2 ) or @xcite ( theorem 3.30 pg 135 ) .
first we recall the definition of successive minima .
note , that it follows directly from the definition that @xmath117 .
in addition , the assumption that @xmath1 contains @xmath4 linearly independent lattice points of @xmath112 implies that @xmath112 has rank @xmath4 and that @xmath118 .
moreover , according to the definition of the successive minima there exists a set of linearly independent vectors from @xmath112 , @xmath119 , such that @xmath120 lies on the boundary of @xmath121 but the interior of @xmath122 does not contain any lattice vectors outside the span of @xmath123 .
the vectors @xmath124 are called a directional basis , and we note that they may not necessarily form a basis of @xmath112 .
next we will study the behavior of constant @xmath37 in the case of unconditional convex bodies .
a set @xmath11 is said to be unconditional if it is symmetric with respect to any coordinate hyperplane , i.e. , @xmath127 , for any @xmath128 and any choice of @xmath129 signs .
* proof : * this result follows from the simple observation that the section of @xmath1 by a coordinate hyperplane @xmath133 is maximal in cardinality among all parallel sections of @xmath1 , i.e. @xmath134 we can see this by considering a point @xmath135 .
let @xmath136 be the reflection of @xmath137 over @xmath133 , i.e. , @xmath138 .
using unconditionality of @xmath1 , we get that @xmath139 and convexity gives us @xmath140 .
hence , the projection of a point in @xmath141 is associated to a point in @xmath142 , which explains ( [ eq : uncon ] ) .
let @xmath143 be the successive minima of @xmath1 with respect to @xmath35 . using an argument similar to the one above one
can show that that the vectors @xmath144 associated with @xmath143 may be taken as a rearrangement of @xmath145 .
we may assume without loss of generality that @xmath146 corresponds to @xmath147 .
so @xmath148 and @xmath149
. thus @xmath150 , where , as before , @xmath151 is a line containing @xmath147 and the origin . using ( [ eq : uncon ] ) , we get @xmath152 by assumption we have @xmath153 and , using @xmath154 , for all @xmath155 , we get @xmath156 finally we use theorem [ th : m2 ] to finish the proof : @xmath157 the cross - polytope @xmath158 of @xmath159 shows that the bound is optimal up to multiplication with constants . the idea of the proof of the above theorem follows from the classical brunn s theorem : the central hyperplane section of a convex origin - symmetric body is maximal in volume among all parallel sections ( see @xcite , @xcite , @xcite ) .
one may notice that , in general , it may not be the case that the maximal hyperplane in cardinality for an origin - symmetric convex body passes through the origin . indeed , see figure 1 below , or consider an example of a cross - polytope @xmath160 , then @xmath161 but a face of @xmath162 contains @xmath4 integer points .
[ lm : pack ] let @xmath165 be a lattice in @xmath8 .
if @xmath166 is an arbitrary bounded set and @xmath167 is a finite non - empty set , then @xmath168 if @xmath169 is a origin - symmetric convex body , then @xmath170 let @xmath172 and assume that @xmath173 ( the statement of the theorem is trivial in the other case ) .
consider a point @xmath174 then @xmath175 moreover , @xmath176-{{\boldsymbol \gamma}}\right)+ \frac{1}{2}\left(\left[k
\cap ( { { \boldsymbol \xi}}^\perp - t { { \boldsymbol \xi}})\right]+{{\boldsymbol \gamma}}\right).\ ] ] thus @xmath177-{{\boldsymbol \gamma}}\right)+ \frac{1}{2}\left(\left[k \cap ( { { \boldsymbol \xi}}^\perp - t { { \boldsymbol \xi}})\right]+{{\boldsymbol \gamma}}\right ) \right ] \cap \gamma'.\ ] ] our goal is to estimate the number of lattice points on the right hand side of the above inclusion . let @xmath178-{{\boldsymbol \gamma}}\right)\ ] ] then , using the symmetry of @xmath1 , we get @xmath179+{{\boldsymbol \gamma}}\right).\ ] ] thus @xmath180 is an origin - symmetric convex body in @xmath20 . next we use ( [ eq:3_10 ] ) from lemma [ lm : pack ] to claim that @xmath181 notice that @xmath182 thus we may use ( [ eq:3_9 ] ) from lemma [ lm : pack ] with @xmath183 , @xmath20 associated with @xmath184 , and @xmath185 to claim that @xmath186 thus we proved that @xmath187 \cap \gamma \bigg ) \\ \ge & 9^{-(d-1 ) } \ # \left ( \left[k \cap ( { { \boldsymbol \xi}}^\perp+t { { \boldsymbol \xi}})-{{\boldsymbol \gamma}}\right]\cap \gamma \right)\end{aligned}\ ] ] but @xmath188 \cap \gamma \right)=\#\left(\left[k \cap ( { { \boldsymbol \xi}}^\perp+t { { \boldsymbol \xi } } ) \right]\cap\gamma \right).\ ] ] [ cor : brunn ] consider a convex , origin - symmetric body @xmath189 , lattice @xmath190 and @xmath7-dimensional lattice subspace @xmath191 ,
i.e. , it contains @xmath7 linearly independent points of @xmath165 , then @xmath192 * proof : * let @xmath193 .
then we may assume @xmath194 and let @xmath195 be the linear space spanned by @xmath191 and @xmath196 .
then @xmath197 and the corollary follows from theorem [ th : brunn ] with @xmath195 associated with @xmath198 , @xmath199 , and @xmath200 .
let @xmath201 be the set of all @xmath202-dimensional linear subspaces containing @xmath202-linearly independent lattice vectors of @xmath35 , i.e. , the set of all @xmath202-dimensional lattice hyperplanes .
the next theorem gives a general bound on the number of integer points in co - dimensional slices .
* proof : * let @xmath207 be the successive minima of the polar body @xmath208 with respect to @xmath35 and let @xmath209 be the associated directional basis .
these vectors are linearly independent and @xmath210 for all @xmath202 .
thus we have @xmath211 let @xmath212 and let @xmath213 be the orthogonal complement of @xmath195 . observe that @xmath214 . since for @xmath215 we have @xmath216 , @xmath217 , we also have @xmath218 , @xmath219 , where @xmath220 is the orthogonal projection onto @xmath195 . in view of we obtain @xmath221 and thus @xmath222 due to our assumption that @xmath1 contains @xmath4-linearly independent lattice points we have that @xmath223 ; otherwise implies @xmath224 for all @xmath225 .
so we conclude by @xmath226 here the last step follows from corollary [ cor : brunn ] , the co - dimensional version of the discrete brunn s theorem .
next minkowski s second theorem ( theorem [ th : m2 ] ) gives the upper bound @xmath227 and so we find @xmath228 hence @xmath229 by the bourgain - milman inequality ( isomorphic version of reverse santal inequality see @xcite or @xcite ) , there exists an absolute constant @xmath230 with @xmath231 and so we get @xmath232 thus together with and we obtain @xmath233 we notice that the methods used in section 3 , i.e. computation via discrete version of the john theorem ( theorem [ th : dj ] from above ) , can also be used to provide a bound for general co - dimensional sections .
but such computation gives the estimate of order @xmath42 which is worse than the one in the above theorem .
observe that theorem [ tm : main ] can be restated for an arbitrary @xmath4-dimensional lattice @xmath165 : let @xmath165 be a lattice in @xmath8 and @xmath5 be an origin - symmetric convex body with @xmath234 . then @xmath235 * proof : * first we notice that if @xmath1 is an unconditional body and @xmath191 is a coordinate subspace of dimension @xmath7 ( i.e. it is spanned by @xmath7 coordinate vectors ) with @xmath237 then @xmath238 must be an unconditional convex body in @xmath239 .
thus , using this property together with the proof of theorem [ th : unconditional ] we get that for any unconditional body @xmath1 and for any coordinate subspace @xmath191 @xmath240 next we follow the steps of the proof of theorem [ tm : main ] and similarly to ( [ eq : bound1 ] ) get @xmath241 finally , we finish the proof using minkoswki s second theorem and the bourgain - milman inequality .
we finish this section with a remark about the relationship between the constant in the original slicing inequality , @xmath23 , and the constant in the discrete version , @xmath37 . using the general idea from @xcite and gauss s lemma on the intersection of a large convex body with a lattice
we will show that @xmath47 .
consider a convex symmetric body @xmath1 and let @xmath250 be such that @xmath251 thus @xmath252 then @xmath253 our goal is to study a section of @xmath1 with a maximal number of points from @xmath35 , if @xmath254 is such a section , then , without loss of generality , we may assume that @xmath255 is a lattice of a full rank @xmath256 .
indeed , if @xmath255 has a rank less then @xmath256 we may rotate @xmath21 to catch @xmath256 linearly independent vectors in @xmath20 , without decreasing the number of integer points in @xmath254 .
now , we may use gauss s lemma ( see for example lemma 3.22 in @xcite ) to claim that for @xmath257 large enough we have @xmath258 @xmath259 which we can rearrange to get the following two equations @xmath260 @xmath261 next , using that @xmath262 and @xmath263 we get @xmath264 then for @xmath265 there is a sufficiently large @xmath266 such that for all @xmath267 @xmath268 so then if @xmath269 we have that @xmath270 for all @xmath4 , @xmath271 , and bodies @xmath1 . which leads us to conclude that @xmath272 .
, _ discrete tomography : determination of finite sets by x - rays , _ trans .
( 1997 ) , 22712295 . ,
uniqueness and complexity in discrete tomography , in : discrete tomography : foundations , algorithms and application , ed .
by g. t. herman and a. kuba , birkhauser , boston .
ma , 1999 , pp . 85113 . , _ a brunn - minkowski inequality for the integer lattice _ , trans .
( 2001 ) , 39954024 . , _ successive minima and lattice points _ , rend .
circ . mat .
palermo ( 2 ) suppl .
70 , part i ( 2002 ) , 377384 . , _
minkowski s successive minima _
, number theory & discrete geometry , 129142 , ramanujan math .
notes ser . , ramanujan math .
mysore , ( 2008 ) . | let @xmath0 be a number of integer lattice points contained in a set @xmath1 . in this paper we prove that for each @xmath2 there exists a constant @xmath3 depending on @xmath4 only , such that for any origin - symmetric convex body @xmath5 containing @xmath4 linearly independent lattice points @xmath6 where the maximum is taken over all @xmath7-dimensional subspaces of @xmath8 .
we also prove that @xmath3 can be chosen asymptotically of order @xmath9 .
in addition , we show that if @xmath1 is an unconditional convex body then @xmath3 can be chosen asymptotically of order @xmath10 . |
let @xmath1 be positive integers .
we denote by @xmath2 the @xmath0-dimensional rectangle of sides @xmath3 , that is , @xmath4 .
a @xmath0-dimensional rectangle @xmath5 is said to be _ tiled _ with _ bricks _
( i.e. , small @xmath0-dimensional rectangles ) @xmath6 if @xmath5 can be filled entirely with copies of @xmath7 , @xmath8 ( rotations allowed ) .
it is known @xcite that rectangle @xmath9 can be tiled with @xmath10 if and only if @xmath11 divides @xmath12 or @xmath13 , @xmath14 divides @xmath12 or @xmath13 and if @xmath15 divides one side of @xmath5 then the other side can be expressed as a nonnegative integer combination of @xmath11 and @xmath14 . in 1995 , fricke @xcite gave the following characterization when @xmath16 ( see also @xcite for a @xmath0-dimensional generalization with @xmath17 ) .
[ kler]@xcite let @xmath18 be positive integers with @xmath19 .
then , @xmath20 can be tiled with @xmath21 and @xmath22 if and only if either @xmath12 and @xmath13 are both multiple of @xmath23 or @xmath12 and @xmath13 are both multiple of @xmath24 or one of the numbers @xmath25 is a multiple of both @xmath23 and @xmath24 and the other can be expressed as a nonnegative integer combination of @xmath23 and @xmath24 .
let us consider the following natural question .
[ qq ] does there exist a function @xmath26 such that if @xmath27 then @xmath20 can be tiled with @xmath10 and @xmath28 for some positive integers @xmath29 and @xmath30 ?
an algebraic result due to barnes @xcite seems to show the existence of such @xmath31 .
however , barnesmethod does not give an explicit lower bound for @xmath31 .
the special case when @xmath32 and @xmath33 was posed in the 1991 william mowell putnam examination ( problem b-3 ) . in this case , klosinski _ et . al . _
@xcite gave a lower bound of @xmath31 .
their method was based on knowledge of the _
frobenius number_. the _ frobenius number _ , denoted by @xmath34 , of a set of relatively prime positive integers @xmath35 , is defined as the largest integer that is not representable as a nonnegative integer combination of @xmath35 .
it is well known that @xmath36 however , to find @xmath34 , for general @xmath0 , is a difficult problem from the computational point of view ; we refer the reader to @xcite for a detailed discussion on the frobenius number .
klosinski _ et . al .
_ used equation ( [ frob2 ] ) , with particular integers @xmath37 and @xmath38 , to show that @xmath20 can be tiled with @xmath39 and @xmath40 if @xmath41 .
.3 cm in this paper , we will use the frobeniuis number in a more general way to show that a @xmath0-dimensional rectangle @xmath5 can be tiled with some set of bricks if the sides of @xmath5 are larger than a certain function ( see theorem [ maint ] ) .
we use then theorem [ maint ] to obtain the following result .
[ cor1 ] let @xmath42 be integers with @xmath43 , @xmath44 and @xmath45 .
then , @xmath20 can be tiled with @xmath46 and @xmath47 if @xmath48 in the case when @xmath49 and @xmath50 , corollary [ cor1 ] implies that @xmath20 can be tiled with @xmath39 and @xmath40 if @xmath51 , improving the lower bound given in @xcite .
we remark that this lower bound is not optimal . in @xcite ,
narayan and schwenk showed that , in this particular case , it is enough to have @xmath52 .
however , their tiling constructions allow rotations of both bricks ( and tilings with more complicated patterns ) which is not the case of corollary [ cor1 ] . we shall also use theorem [ maint ] to prove the following result concerning tilings of squares .
[ cor2 ] let @xmath53 be prime integers .
then , @xmath54 can be tiled with @xmath55 if @xmath56 we finally improve the lower bound given in theorem [ cor2 ] in some special cases .
[ cor3 ] let @xmath57 be an odd integer with @xmath58 and let @xmath59 be a positive integer .
then , @xmath60 can be tiled with @xmath61 and @xmath62 if @xmath63 .
moreover , @xmath60 can be tiled with @xmath61 and @xmath64 if and only if @xmath65 and with @xmath61 and @xmath66 if and only if @xmath67 . a collection of some unpublished work , due to d.a .
klarner , in relation with theorem [ cor3 ] can be found in @xcite .
we need to introduce some notation and definitions .
let @xmath68 where @xmath69 are positive integers .
we will write @xmath70 instead of @xmath71 and @xmath72 instead of @xmath73 .
let @xmath74 be a positive integer for each @xmath75 and each @xmath76 .
let @xmath77 , @xmath78 .
we define the set we denote by @xmath82 the rectangle obtained from @xmath83 by sticking together @xmath84 copies of @xmath85 along the @xmath86-axis , that is , @xmath87 .
finally , we denote by @xmath88 the @xmath89-dimensional rectangle obtained from @xmath83 by setting @xmath90 , that is , @xmath91 . _ proof .
_ we shall use induction on @xmath0 .
for @xmath100 we have that @xmath101 and thus @xmath102 . by definition of the frobenius number ,
any integer @xmath103 is of the form @xmath104 where @xmath105 are nonnegative integers .
thus , the 1-dimensional rectangle @xmath106 ( that is , the interval @xmath107 $ ] ) can be tiled by sticking together @xmath108 ( that is , the interval @xmath109 $ ] ) and @xmath110 ( that is , the interval @xmath111 $ ] ) .
.3 cm we suppose that it is true for @xmath112 and let @xmath74 be a positive integer for each @xmath113 and each @xmath114 with @xmath94 for any @xmath115 , @xmath116 and let @xmath117 , @xmath114 and @xmath118 for all @xmath119 . indeed ,
if we consider the rectagle @xmath120 embedded in @xmath127 with @xmath128 then by replacing each brick @xmath129 used in the tiling of @xmath120 by @xmath130 we obtain a tiling of @xmath125 with bricks @xmath126 .
.3 cm now , since @xmath131 then @xmath132 where each @xmath133 is a nonnegative integer . by the above claim
, @xmath134 can be tiled with bricks @xmath135 for each @xmath136 .
thus , @xmath137 can be tiled with @xmath138 by sticking together bricks @xmath139 along the @xmath140-axis .
\(b ) by induction on @xmath166 . for @xmath167
we have @xmath168 since @xmath169 and @xmath170 for all @xmath171 .
we suppose that it is true for @xmath172 and assume that @xmath173 . by equation ( [ eerr ] )
we have _ proof of theorem [ cor2 ] .
_ let @xmath53 be prime integers , @xmath93 .
we consider theorem [ maint ] , where , for each @xmath76 , we let @xmath184 for all @xmath99 .
then , @xmath54 can be tiled with @xmath55 if @xmath185 [ llls ] let @xmath191 and @xmath84 be positive integers with @xmath192 and such that @xmath193 for some integers @xmath194 and @xmath195 for some integers @xmath196 .
then , @xmath197 and @xmath198 can be tiled with @xmath199 and @xmath200 for any integer @xmath201 .
_ proof of theorem [ cor3 ] .
_ by theorem [ kler ] , we have that @xmath60 can be tiled with @xmath210 and @xmath211 if @xmath212 or @xmath213 .
so , we only need to show that @xmath60 can be tiled with @xmath214 and @xmath62 for any odd integer @xmath63 with @xmath215 or @xmath216 .
we have two cases .
.3 cm case a ) @xmath217 .
let @xmath218 for any integer @xmath201 . since @xmath219 then there exist nonnegative integers @xmath220 and @xmath221 such that @xmath222 .
since @xmath223 then , by proposition [ llls ] , @xmath224 can be tiled with @xmath214 and @xmath62 for any @xmath225 and @xmath226 . now , since @xmath227 for some integer @xmath201
then for @xmath228 we have that @xmath229 and by proposition [ llls ] ( with @xmath230 ) , we have that @xmath231 can be tiled with @xmath214 and @xmath62 for any odd integer @xmath232 with @xmath233 .
.3 cm case b ) @xmath234 .
let @xmath235 for any integer @xmath201 . since @xmath219 then there exist nonnegative integers @xmath220 and @xmath221 such that @xmath222 .
since @xmath223 then , by proposition [ llls ] , @xmath224 can be tiled with @xmath214 and @xmath62 for any @xmath236 and @xmath237 .
now , since @xmath238 for some integer @xmath201 then for @xmath228 we have that @xmath229 and by proposition [ llls ] ( with @xmath230 ) , we have that @xmath231 can be tiled with @xmath214 and @xmath62 for any odd integer @xmath232 with @xmath239 .
.3 cm let us set @xmath240 .
it is clear that @xmath241 and @xmath66 can not be tiled with @xmath214 and @xmath64 . by the above cases
, we have that @xmath60 can be tiled with @xmath214 and @xmath64 if @xmath242 and , by theorem [ kler ] , @xmath60 can be tiled with @xmath210 and @xmath211 if @xmath212 or @xmath213 .
this leave us the cases when @xmath243 and @xmath244 . if @xmath245 is trivial .
@xmath246 can be tiled with @xmath214 and @xmath64 since , by tha above case b , the result is true for any odd integer @xmath247 and @xmath226 .
finally , @xmath248 can be tiled as it is illustrated in figure [ fig2 ] .
let us set @xmath249 .
it is clear that @xmath241 , @xmath64 and @xmath246 can not be tiled with @xmath214 and @xmath66 . by the above cases ,
we have that @xmath60 can be tiled with @xmath214 and @xmath66 if @xmath250 and , by theorem [ kler ] , @xmath60 can be tiled with @xmath210 and @xmath211 if @xmath212 or @xmath213 .
this leave us the cases when @xmath251 and @xmath252 . if @xmath253 is trivial . @xmath248 and
@xmath254 both can be tiled since , by the above case a , the result is true for any odd integer @xmath255 with @xmath233 .
finally , @xmath256 can be tiled as it is illustrated in figure [ fig3 ] . | in this paper , we give some sufficient conditions for a @xmath0-dimensional rectangle to be tiled with a set of bricks .
these conditions are obtained by using the so - called frobenius number . |
blazars form one of the most energetically extreme classes of active galactic nuclei ( agn ) .
blazars can be observed in all wavelengths , ranging from radio all the way up to @xmath0-rays .
their spectral energy distribution ( sed ) is characterized by two broad non - thermal components , one from radio through optical , uv , or even x - rays , and a high - energy component from x - rays to @xmath0-rays .
in addition to spanning across all observable frequencies , blazars are also highly variable across the electromagnetic spectrum , with timescales ranging down to just a few minutes at the highest energies .
there are two fundamentally different approaches to model the seds and variability of blazars , generally referred to as leptonic and hadronic models ( see , e.g. , * ? ? ?
* for a review of blazar models ) . in the case of leptonic models , where leptons are the primary source of radiation , synchrotron , synchrotron self - compton ( ssc ) , and external - compton ( ec ) radiation mechanisms
are employed to explain the blazar sed ( see , e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
the focus of the present study is also on a leptonic model . in hadronic models ,
the low - energy sed component is still produced by synchrotron emission from relativistic electrons , while the high - energy component is dominated by the radiative output from ultrarelativistic protons , through photo - pion induced cascades and proton synchrotron emission ( e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
one aspect common to all blazar models is a relativistic jet oriented at a small angle with respect to our line of sight , resulting in relativistic doppler boosting and the shortening of observed variability time scales . given computational limitations , the complex physical processes in relativistic jets can , realistically , only be evaluated with certain simplifying approximations . in order to facilitate analytical as well as numerical calculations ,
the two most common approximations employed in blazar jet models are to assume that the magnetic ( * b * ) field is randomly oriented and tangled , and that the lepton momentum distribution is isotropic in the comoving frame of the high - energy emission region .
these two assumptions greatly simplify the evaluation of the synchrotron and compton emission by eliminating various integrals over the interaction and scattering angles . however , there is increasing evidence @xcite for a fairly well defined helical * b*-field structure within agn jets .
these observations also suggest a spine - sheath geometry for agn jets .
the differential velocity profiles within the jet is expected to create anisotropies in the particle distributions .
it is therefore important to explore jet models where we can not only simulate an ordered * b*-field , but also study the resulting radiation behaviour with anisotropic lepton distributions .
the standard approach to diagnosing the magnetic field properties is via synchrotron polarization .
if the underlying distribution of emitting electrons is a power - law with power - law index @xmath1 , the maximum degree of synchrotron polarization is given by : @xmath2 where @xmath3 and @xmath4 are the synchrotron power per unit frequency in directions perpendicular and parallel to the projection of the magnetic field on the plane of the sky .
using equation [ eq : synchpol ] we can see that for a power - law index of @xmath5 , the degree of polarization can be as high as 75% .
it is therefore possible to estimate the magnetic field orientation based on polarization measurements , but an estimate of the field strength usually requires the consideration of flux and spectral properties of the synchrotron emission .
furthermore , polarization measurements are notoriously difficult ( and even barely feasible at frequencies higher than optical ) , and may often not give realistic results due to faraday rotation and depolarization along the line of sight .
in this work , we are interested in taking a complementary approach to estimating the magnetic field orientation where the difference in observed flux levels of the spectrum can give an estimate of how the magnetic orientation may be changing .
ratio shows how the b - field estimate can vary depending on the electron power - law index and whether one assumes an isotropic b - field or a specific pitch angle.,scaledwidth=45.0% ] the principle behind this approach can be demonstrated when one compares the * b*-field estimates based on a power - law distribution of electrons with an arbitrary power - law index , @xmath1 , and pitch angle , @xmath6 .
the comparison of synchrotron emission coefficients for a power - law distribution of electrons with and without pitch - angle ( @xmath6 ) dependence gives us a measure of how the estimated magnetic field strength can differ .
the emission coefficients can be found in @xcite and are given by ( in the units of @xmath7 ) : @xmath8 and @xmath9 where @xmath10 is the electron distribution power - law normalization .
the above two expressions can be solved for the magnetic field to obtain : @xmath11 this gives an estimate of how , for a given luminosity , the * b*-field estimates can differ depending upon whether we assume an isotropic pitch angle approximation or a given pitch angle ( which , in the case of relativistic electrons , is equal to the angle between the magnetic field and the line of sight ) .
the above relation is only applicable in the optically thin regime .
we can see in figure [ fig : bestimate ] that depending on the pitch angle assumption , and the electron distribution power - law index , the @xmath12 fraction can range from 0 ( there is negligible synchrotron emission along an ordered magnetic field ) to @xmath13 .
because the compton emissivity is approximately isotropic for an isotropic distribution of electrons , the ratio @xmath14 will change with the pitch angle .
it is therefore important to see how the overall synchrotron and synchrotron - self compton spectra differ with well ordered magnetic fields .
one point worth noting is that in our set - up the lower limit on orientation is limited by the @xmath15 approximation ( see section [ sec : synch ] ) ; a magnetic field perfectly aligned with the observing direction will give zero output .
however , in a more rigorous treatment , the lower limit on the minimum angle , @xmath6 , for the orientation will be determined by the relativistic beaming characteristic of synchrotron emission along an electron s direction of motion into a cone of opening angle @xmath16 , where @xmath0 is the electron lorentz factor . in the case of optical frequencies and magnetic fields of @xmath17 g , @xmath18 will be the order of @xmath19 , while in the x - ray regime @xmath20 . using the relation in equation [ eqn : bestimate ] it is possible to estimate the effects on the estimates between this value of @xmath18 and an isotropic magnetic field . for a power - law index of 3
, @xmath18 gives @xmath12 value of 0.0039 .
the following section briefly outlines our model , including the synchrotron radiation and compton scattering treatments followed as well as the numerical techniques used to implement them .
@xmath21{figure2.epsi } \\
\includegraphics[width=0.35\textwidth]{figure3.epsi}\\ \end{array}$ ] , the viewing angle with respect to the volume / jet is incorporated into doppler boosting calculations .
, title="fig:",scaledwidth=40.0% ] + in our model the basic volume structure is a cubic cell .
this allows the model to be modular and build an arbitrarily large volume with any desired anisotropies .
each cell contains a magnetic field plus electron and photon distributions .
the magnetic field can have an arbitrary orientation and strength in each cell .
this means that the overall volume can be modelled to contain a completely uniform , partially anisotropic , or pseudo - random * b*-field .
the purpose of the present work is to isolate the effects of the degree of order and orientation of the magnetic field on the emerging synchrotron emission .
therefore , we choose the simplest conceivable approach concerning the electron distribution , and do not take electron cooling into account .
this means , we only focus on static electron distributions which do not evolve due to energy losses . in future work
we aim to include self - consistent cooling effects which would also allow us to probe how the pitch - angle dependence of the synchrotron cooling would give rise to different electron distributions in different cells , depending on the magnetic field set - up . the directional information in the electron and photon distributions and the is with respect to the cell . in the case of electrons
the distribution is a function of energy and two angles with respect to the cell ( see figure [ fig : eleb ] ) .
this gives us the ability to create anisotropies in the electron distributions as well by either having preferential direction for the electrons or by setting up the electron distributions differently in various cells .
for both electrons and photons , the distributions energy grids ( lorentz factor @xmath0 for electrons and frequency @xmath22 for the photons ) are calculated using logarithmic binning .
therefore each distribution is modelled using a 3 dimensional array with the dimensions of @xmath23 $ ] .
figures [ fig : eleb ] ( right ) illustrates how various angles with respect to the cell are defined .
angles @xmath24 and @xmath25 run from 0 to @xmath26 and 0 to 2@xmath26 respectively . in figure
[ fig : volgeo ] we can see how the overall volume can be constructed from individual cells . for a given viewing angle , the emission from the visible outer layer of cells is combined to produce an overall spectrum from an effectively larger volume .
the simulation currently transfers , from one cell to another , only the photons . in order to achieve this
, we need to calculate which of the six cubic faces a given photon direction will intersect . to calculate this
, we assume that all the photons are produced in the center of the cell , and then trace photon paths in any given direction towards the nearest boundary .
even though our simulation considers a static situation , the transfer and radiative feedback between different cells requires an inherent time - dependence in the code .
the time step for our radiation transfer approach is the light crossing time across a single cell , which is equal to the time it takes for the photons to travel from one cell to another . at the end of each time step and
depending on the physical processes being modelled , the photon distribution is modified and passed to the appropriate neighbour . when being passed to a neighbour
the entire photon distribution is passed .
therefore at the end of a time - step each cell s ( intrinsic ) photon distribution is emptied into six neighbouring cells , unless it is a boundary cell .
the six incoming ( transiting ) photon distributions are stored until the start of the following time - step when they are combined to form a single intrinsic photon distribution again .
the physical processes are then carried out on this single photon distribution .
synchrotron radiation is calculated first and the photons added to the intrinsic distribution .
compton scattering is carried out after the synchrotron radiation . at this point
we reach the end of a time - step and the process of transferring photon distributions to neighbouring cells begins again .
the observed photon distribution originates from the boundary cells .
the photon distributions emerging from visible faces of the boundary cells are combined to create a single observed photon distribution .
this process of combining the photon distributions from the boundary cells in effect treats the whole multi - cell structure like a single cubic / cuboid structure .
here we highlight the key points of the synchrotron radiation treatment that we follow .
a more in - depth analysis and details can be found in @xcite .
the synchrotron emissivity per electron , @xmath27 , is given by : @xmath28 where f(x ) is given by : @xmath29 @xmath30 , where @xmath31 is the critical frequency given by @xmath32c .
@xmath6 , the pitch angle , is calculated using spherical trigonometry : @xmath33 the synchrotron emission coefficient is given by : @xmath34 the numerical bessel function integration in equation [ eqn : bessel ] can be time consuming .
however , some fast routines to perform this integration are given by @xcite which we modified for our precision and computer language . in a full treatment of the synchrotron radiation the emitted photons
are distributed within a solid angle ( @xmath35 ) about the pitch angle @xmath6 . however ,
for our purposes we assume the emitted photons travel in the same direction as the emitting electrons .
a detailed calculation of synchrotron self - absorption can be found @xcite .
the absorption coefficient when recast in terms of electron lorentz factors , @xmath0 , instead of @xmath36 , can be written as : @xmath37 the photons produced via synchrotron radiation are added to the intrinsic photon distribution of the cell .
the photons received from neighbouring cells are added to the intrinsic photon distribution prior to calculating the synchrotron spectrum .
therefore the photons passing through any cell are also synchrotron self - absorbed .
the emission and absorption coefficients are used to calculate the total spectrum , @xmath38 where @xmath39 is the optical depth and _ l _ is the size of the emission zone / cell .
to @xmath40 off electrons @xmath41,scaledwidth=40.0% ] in the limit @xmath42 , and in the electron rest frame , the incident photon travels in nearly the opposite direction to the electron .
this is due to photon aberration : @xmath43 when @xmath44 , @xmath45 , we are in the head - on approximation regime , which we employ to greatly simplify compton cross section calculations .
that is , we can assume that the scattered photon solid angle , @xmath46 , is well approximated by the electron solid angle @xmath47 . when the differential compton ( klein - nishina ) cross section is integrated over @xmath46 , we get @xcite : @xmath48 where h is a heaviside function , @xmath49 , and the compton kernel is given by : @xmath50 and @xmath51 .
the compton cross section can then be used in the emission coefficient formula to obtain the comptonized spectrum .
the head - on approximation simplifies the emission coefficient calculation by eliminating two integrals from the compton emissivity treatment without the approximation .
the following relation can be used to obtain the number of interacting photons : @xmath52 the interacting photons are a combination of photons originating from synchrotron radiation and the photons received from neighbouring cells . at the start of a time - step
, the photon distributions received from the neighbouring cells are combined , while preserving the direction information , to form the intrinsic photon distribution .
synchrotron photons are also added to the intrinsic photon distribution .
the total photon distribution is then used in the compton emissivity relation to obtain the compton spectrum in the head - on approximation , given by : @xmath53 after the intrinsic photon distribution has been compton scattered it is redistributed based on change in energy and direction .
the redistributed photon distribution is then used to work out which neighbouring cells receive which proportion of the distribution . the overall spectrum is obtained by combining the synchrotron and compton spectra . as it stands in our model , synchrotron emission is the only source of photons which are then compton scattered by the same population of electrons .
the resulting photon spectrum is given as a function of two cell angles @xmath24 and @xmath25 ( see figure [ fig : eleb ] ) .
although not included in the present study , it is straight forward to include external compton effects by adding the external photon field to the photon distributions .
once we have a spectrum , @xmath54 , the flux can be calculated using : @xmath55 for an emission zone with an area @xmath56 and luminosity distance @xmath57 . for a viewing angle
@xmath58 the doppler factor @xmath59 is given by @xmath60^{-1 } \ .
\label{eqn : doppler}\end{aligned}\ ] ] the ` @xmath61 ' corresponds to either an approaching or a receding component of the jet . in the case of blazars the observed emission is strongly dominated by the approaching jet , boosted with the doppler factor @xmath62 . also , any given frequency , @xmath22 , in the emission region rest frame will be shifted by a factor of : @xmath63 we follow the @xcite formulation to calculate the luminosity distance based on the redshift of an object .
photons emitted at an angle @xmath64 in the cell rest frame will appear at an angle @xmath65 due to angle aberration , which can be expressed as follows : @xmath66
( @xmath67 in jet frame ) . the spectra shown results of different b - field configurations , including when it is randomly oriented in different cells .
the plot shows @xmath68 values ( @xmath69 , unless randomly oriented ) .
the red dot - dot - dashed curve shows the corresponding calculation based on the angle averaged emissivity and spherical geometry , using the code of @xcite .
see figure [ fig : eleb ] for details on various angles.,scaledwidth=45.0% ] , @xmath69 .
see figure [ fig : eleb ] for details on various angles.,scaledwidth=45.0% ] in order to study the effects of the * b*-field orientation on the synchrotron and synchrotron self - compton spectra , we set up two scenarios . in the first set up the magnetic field
is uni - directional in all the cells ( 27 in total ) and in the second scenario the magnetic field is randomly oriented in each of the cells .
these set - ups are likely to be the two extreme scenarios for a jet .
evidence points to the * b*-field being semi - ordered in agn jets ; for example , helical @xcite . in all the presented cases ,
the electron distribution is a power - law and distributed uniformly over the angles @xmath70 and @xmath71 .
figure [ fig : b_angles_ssc ] shows synchrotron self - compton spectra with different configurations .
various simulation parameters can be found in table [ tab : figparams ] .
the spectra are for a fixed viewing angle . changing the orientation has a significant impact on the observed spectrum .
there is a large difference in flux values depending on the orientation with respect to the observing direction .
the minimum flux levels are observed at an orientation along the viewing angle while the maximum flux levels are observed when the is perpendicular to the line of sight .
the figure also demonstrates the fact that only the synchrotron spectrum component is heavily affected by the orientation .
the compton scattered component of the spectrum is almost independent of the orientation .
the main reason for this is that the photon distribution anisotropies introduced by the orientation are lost when scattering off an isotropic distribution of electrons .
the small variations that remain are due to the @xmath72 factor present in compton emissivity calculations .
the line of sight and photon anisotropies therefore affect the extent to which the compton spectrum is boosted .
additional anisotropies are introduced by the discretization of the photon and the electron angular distributions .
figure [ fig : b_angles_ssc ] also shows a comparison with another ssc calculation @xcite which assumes a randomly oriented magnetic field .
this simulation is set up with identical parameters to the ones outlined in table [ tab : figparams ] , except it uses a spherical volume instead of a cube ; the total volumes are identical , therefore the sphere has a radius of @xmath73 cm .
we can see that the synchrotron components are in good agreement .
however , the compton component in the @xcite calculation is much higher .
the inverse compton to synchrotron ratio @xmath74 differs by a factor @xmath750.68 between the two calculations .
this is most likely due to differing geometries . for a sphere ,
the average photon escape time is @xmath76 , whereas in our cubic set up , due to the way radiation transport between cells is treated ( see section [ sec : volume ] ) , a photon takes , on average , @xmath77 to escape the region ( @xmath78 is the width of an individual cell ) . since the flux ratio @xmath79 and the volume - averaged radiation energy density @xmath80 is proportional to the photon escape time scale , the longer photon escape time scale in the spherical geometry results in a larger ssc flux .
figures [ fig : view_angles_ssc ] shows the effects of the viewing angle on the spectrum .
it shows seds for a fixed azimuthal angle , but different viewing angles with a single configuration . as before ,
there are normalization differences between the seds when comparing uniform and randomly oriented , but the main factor in this case is the variation in the doppler boosting due different viewing angles . for the fit parameters.,scaledwidth=45.0% ] for the fit parameters.,scaledwidth=45.0% ] for the fit parameters.,scaledwidth=45.0% ] markarian 421 was the first extragalactic source to be detected in tev energies , hence making it an extensively studied source .
we present spectral fits to _ xmm - newton _ and _ veritas _ data presented in @xcite .
we refer the reader to the above paper for the details on data reduction .
figures [ fig : mrk421_0402 ] , [ fig : mrk421_0802 ] , and [ fig : mrk421_1802 ] show fits to mrk 421 data using our code .
our aim in the present paper is to explore the impact of the orientation on the fit parameters , especially the estimates . in figure
[ fig : mrk421_0402 ] we can see that the synchrotron component is best fit for a particular orientation ( see table [ tab : mrkparams ] for fit parameters ) .
the gamma - ray data , however , is fit well with all the orientations .
this is due the fact that the self - compton spectrum is not affected much by the orientation ( see the discussion in the previous section ) .
there is a significant difference in the synchrotron peak when comparing various orientations .
the spectra in figure [ fig : mrk421_0402 ] show that at a good fit is achieved when the is oriented at @xmath81 with a strength of @xmath82 g. however , the fit shown is not unique .
we can see in figures [ fig : mrk421_0802 ] and [ fig : mrk421_1802 ] that for identical parameters , except the strength , the best - fit orientation is very different . in one case
a pseudo - random provides the best fit with @xmath83 g , while for @xmath84 g , a orientation of @xmath85 provides the best fit .
therefore good fits can be achieved with different strengths at different orientations .
the main point here is the fact that it is possible to over- or underestimate the strength when assuming it to be randomly oriented . in the cases presented here ,
the strength ranges from 0.18 g to 0.25 g for very similar fits to the data , but with different magnetic - field orientations .
therefore it is possible to overestimate the magnetic field strength by at least @xmath86 if a particular orientation , whether uniform or tangled , is assumed .
if the were pointed closely aligned with the line of sight , much higher values will be necessary to obtain similar fits ( see discussion in section [ subsec : bestimate ] ) .
we also note that the bulk lorentz factor used in the fits are lower than the values obtained by some authors for fitting mrk 421 data ( e.g. see * ? ? ?
the value of the bulk lorentz factor values used in our fits is likely to be on the lower end of the limits imposed by pair opacity arguments @xcite .
however , the main point of this paper is not the determination of actual best - fit values for mrk 421 ( which would not be realistic due to our neglect of cooling effects anyway ) , but to demonstrate the orientation - dependent magnetic - field degeneracy in the course of blazar sed fitting .
in this paper we have presented first results from a new relativistic jet radiation transfer code that we are currently developing . here
we take the full angular dependence into account when modelling synchrotron and synchrotron self - compton processes .
we are able to model the at arbitrary orientations and study its impact on the resulting spectra .
we have seen that the orientation plays an important role on the normalization of the synchrotron spectrum . using fits to markarian 421 data ,
we have shown how the orientation can mislead into over / under - estimating its strength .
any future work should therefore be mindful of the fact that the underlying assumption about the orientation will play a considerable role in the errors associated with the magnetic field strength estimates . | we report on the development of a numerical code to calculate the angle - dependent synchrotron + synchrotron self - compton radiation from relativistic jet sources with partially ordered magnetic fields and anisotropic particle distributions . using a multi - zone radiation transfer approach , we can simulate magnetic - field configurations ranging from perfectly ordered ( unidirectional ) to randomly oriented ( tangled ) . we demonstrate that synchrotron self - compton model fits to the spectral energy distributions ( seds ) of extragalactic jet sources may be possible with a wide range of magnetic - field values , depending on their orientation with respect to the jet axis and the observer .
this is illustrated with the example of a spectral fit to the sed of mrk 421 from multiwavelength observations in 2006 , where acceptable fits are possible with magnetic - field values varying within a range of an order of magnitude for different degrees of b - field alignment and orientation . |
many problems of stellar dynamics deal with self - gravitating systems which are in dynamical equilibrium , but slowly evolve due to two - body relaxation or some other factor , such as a massive black hole or the diffusion of chaotic orbits .
the most general method of studying these systems is a direct @xmath0-bodysimulation , however , in many cases it turns out to be too computationally expensive .
alternative methods , such as fokker planck , gaseous , or monte carlo models , have historically been developed mostly for spherical star clusters . in this paper
we present a formulation of the monte carlo method suitable for non - spherical stellar systems .
the paper is organized as follows .
section [ sec : overview ] reviews the existing simulation methods and outlines the motivation for the proposed new formulation ; section [ sec : relaxation ] presents the theoretical background of two - body relaxation theory ; section [ sec : raga ] discusses the implementation of the non - spherical monte carlo code and section [ sec : tests ] presents the results of test simulations .
section [ sec : conclusions ] lists possible applications of the new method and sums up .
the development of monte carlo methods for simulation of star clusters started in early 1970s with two different approaches , pioneered by spitzer and hnon . in the original formulation of @xcite , the motion of test stars in a spherically symmetric potential
was followed numerically on the dynamical timescale , and perturbations to the velocity was computed assuming a maxwellian distribution of background stars ( scatterers ) , with the mean density and velocity dispersion computed in 25 radial bins by averaging over 40 stars in each bin ; thus , the test stars were also used for determining the smoothed properties of the field stars . to speed up computation , dependence of velocity diffusion coefficients on the velocity of the test star
was ignored ( the values corresponded to the average thermal velocity ) ; this simplification was lifted in @xcite . since perturbations to each star
s velocity are independent of each other , the global conservation of energy is not guaranteed ; thus a correction is applied after each timestep which cancels the residual fluctuations .
this method became known as the `` princeton '' monte carlo code @xcite . in another variant of this method ,
@xcite turned to using the diffusion coefficients in energy @xmath1 and angular momentum @xmath2 , averaged over the radial period of the test star .
this approach was subsequently developed by @xcite to study the steady - state solution for the distribution of stars around a massive black hole : the potential was assumed to be dominated by the point mass , the diffusion coefficients in @xmath1 and @xmath2 were computed self - consistently from the distribution function @xmath3 , which was then adjusted iteratively until convergence .
the capture of low angular momentum stars by the black hole was also taken into account , which necessitated a rather complex scheme for choosing the timestep : it was determined by the relaxation time but also required not to miss a potentially disruptive periapsis passage near the black hole .
it also had an ingenious scheme for particle cloning ( mass refinement ) to allow for better sampling of phase - space close to the black hole .
subsequent papers extended the method to self - consistent ( rather than point - mass - dominated ) potentials @xcite and to evolutionary simulation including the heating by the black hole , core collapse , and evaporation @xcite .
this approach has been dubbed the `` cornell '' code @xcite .
more recently , @xcite and @xcite have used this formulation to study the dynamics around massive black holes . at the same time
, @xcite introduced another variant of monte carlo method , in which pairs of stars are interacting directly ( see also * ? ? ? * ) .
unlike the conventional @xmath0-bodysimulations , these pairwise interactions are computed only between particles that are adjacent in radius . for each pair of interacting particles ,
their relative velocity is changed by an amount which reproduces statistically the effect of many individual encounters during the same interval of time .
the timestep is chosen to be a fraction of the relaxation time @xmath4 , instead of a fraction of the dynamical time @xmath5 .
after each timestep , the stars are assigned new positions ( or , rather , radii , since the system is assumed to be spherically symmetric ) .
this method was subsequently improved by @xcite , who included a variable timestep ( proportional to the radius - dependent @xmath4 ) , correction of velocities due to the changes in potential after recomputing new positions of particles , continuous stellar mass spectrum , and shock heating due to passages of the globular cluster through the galactic disc . @xcite introduced many other physical ingredients such as stellar evolution , primordial binaries ( also studied by * ? ? ?
* ) and cross - sections for three- and four - body interactions , and stellar collisions .
all presently used codes follow the hnon s approach . since late 1990s , two groups @xcite have been developing sophisticated codes including much additional physics beyond two - body relaxation : parametrized single and binary stellar evolution @xcite , direct integration of few - body encounters @xcite , accurate treatment of escapers @xcite .
the present versions of these codes are described in @xcite and @xcite . in these codes
, the number of particles in the simulation equals the number of stars in the system under study , which facilitates a correct proportion between various dynamical processes .
a third code of the same family was developed by @xcite for studying dense galactic nuclei , featuring accurate treatment of loss - cone effects ( including a timestep adjustment algorithm similar to that of shapiro ) , and a model for physical collisions based on a large library of smooth particle hydrodynamics ( sph ) simulations .
table [ tab : compare_mc ] compares the features of various monte carlo methods . [ cols="<,<,<,<,<,<,<",options="header " , ] in parallel with the monte carlo codes , the approach based on direct integration of the fokker
planck equation using finite - difference schemes was developed by @xcite , and later by @xcite and @xcite for spherical systems .
however , it seems to be impractical to extend it beyond two - integral axisymmetric case @xcite , as the method relies on the explicit knowledge of integrals of motion .
another related method is the gaseous model , in which the relaxation is treated using a conductive approximation @xcite , and which can be combined with monte carlo treatment of stellar binaries @xcite .
this approach also was developed in the spherical case only . with the advent of special - purpose hardware in 1990s
, it became possible to perform direct @xmath0-bodysimulations of globular clusters with more than @xmath6 stars @xcite , and a wide range of physics may be added to the dynamical evolution ( e.g. * ? ? ? * ) .
these simulations are also not restricted to any particular geometry , but are very computationally expensive and , as we will show , still practically unsuitable for some classes of problems , in which collisional relaxation should be rather small compared to collisionless effects arising from non - spherical mass distribution .
a number of more esoteric approaches have been proposed to combine the flexibility of collisional direct @xmath0-bodysimulations with fokker
planck ( e.g. * ? ? ?
* ) , spherical - harmonic expansion ( e.g. * ? ? ?
* ) , self - similar dynamic renormalization @xcite , or tree - code @xcite , none of which apparently gained substantial popularity . from the side of collisionless simulations of nearly - equilibrium systems ,
the most relevant for this study are the spherical - harmonic methods aarseth1967,cluttonbrock1973,vanalbada1977,hernquistostriker1992,meironlhbs2014 . in this approach ,
the smooth potential of a stellar system is represented as a sum of angular harmonics , with the radial variation of the expansion coefficients being either an explicit function of radius , or another sum over several basis functions .
the coefficients of expansion are computed from the spatial distribution of particles , and the equations of motion of particles are governed by the gradients of this smooth potential . in all existing implementations , however , the timestep for particle motion , which is necessarily a small fraction of dynamical time , is also used for updating the coefficients of expansion , thereby imposing random fluctuations on them which effectively create numerical noise comparable to that of more direct @xmath0-body methods @xcite . on the other hand , using some sort of temporal smoothing for the expansion coefficients might help to reduce the unwanted relaxation considerably below the discreteness limit , while retaining the ability to follow the slow evolution of a non - spherical system .
a similar idea was recently used by @xcite in the context of evolution of the population of globular clusters in the galaxy . the idea to marry the benefits of collisionless expansion codes and collisional monte carlo approach has led us to a new formulation of the monte carlo method that avoids the restriction to spherical symmetry while retaining the ability to model the two - body relaxation rather faithfully .
in essence , it is a successor to the spitzer s variant of monte carlo method , with the orbits of test stars followed in real space in a smooth potential represented by a suitable expansion , and perturbations are applied to particle velocities in accordance with local diffusion coefficients .
the new method is dubbed the ragacode , which stands for `` relaxation in arbitrary geometry '' , but also alludes to slowly developing musical themes in the classical indian tradition .
in addition to this method , we have also implemented a variant of spherical isotropic fokker
planck code , similar to that of @xcite , and an orbit - averaged spherical isotropic monte carlo code , a simplified version of the method of @xcite , mainly for the purpose of testing the main code .
below , we present a complete mathematical description and test simulations .
in this section we review the standard two - body relaxation theory as used in our code , referring to ( * ? ? ? * chapter 5 ) for a more complete description . as in most previous studies ( the notable exception being monte carlo codes based on the hnon s approach ) , we consider scattering of test particles by an isotropic spherically symmetric population of background particles , described by the mass distribution function @xmath7 , where @xmath8 is the energy per unit mass . the scattering is described in terms of local ( position - dependent ) velocity drift and diffusion coefficients ( e.g. * ? ? ?
* equations 5.23 , 5.55 ) : [ eq : dc_vel ] vv _ & = -(1 + ) i_1/2 , + v^2 _ & = ( i_0 + i_3/2 ) , + v^2_&= ( 2 i_0 + 3 i_1/2 - i_3/2 ) , where @xmath9 and @xmath10 are masses of the test and field stars , correspondingly , and i_0 & _ e^0
def(e ) , + i_n/2 & _ ( r)^e def(e ) ( ) ^n/2 , + & 16 ^ 2g^2m_= 16 ^ 2 g^2 m _ ( n_^-1 ) .
[ eq : perturbation_term ] these coefficients represent mean and mean - squared changes in velocity per unit time . in the last equation ,
the term in brackets , or its appropriate generalization for a multimass case , is the only one that depends on @xmath11 ( for a given combination of @xmath12 ) . in the monte carlo code
, we may assign the amplitude of perturbations at will , adjusting this term to a desired number of stars in the target system , which needs not be related to the number of particles in the simulation . in hnon s formulation , the particles were called `` superstars '' , and their masses were a fixed multiple of actual stellar mass ; contemporary codes usually have 1:1 correspondence between the particle and star mass , which facilitates the introduction of additional physical processes such as binary single star scattering cross - section . in our approach
, we do not require a fixed proportionality coefficient between the particle and star masses
particles are just mass tracers and not actual stars , and the relaxation is determined by the smooth distribution function and not by discrete encounters . in the rest of this section , we focus on the isotropic spherically symmetric case , which is used in the auxiliary codes , while the treatment of relaxation in the main ragacode relies only on the velocity diffusion coefficients ( [ eq : dc_vel ] ) , with some secondary routines using the orbit - averaged energy diffusion coefficient . even in the spherical case , it is not necessary that the distribution of stars is isotropic : @xcite worked in a two - dimensional @xmath13 phase - space , but retained the isotropic background approximation . local drift and diffusion coefficients in energy are [ eq : dc_energy_local ] e & = vv _ + v^2
= i_0 - i_1/2 , + e^2 & = v^2 v^2 _ = v^2 ( i_0 + i_3/2 ) .
the isotropic fokker
planck equation describing the relaxation in energy has two forms .
the first is more convenient for monte carlo simulations , the second ( flux - conservative ) is more suitable for solving the fokker
planck equation on a grid .
define @xmath14 to be the mass density of stars per unit energy , where the density of states @xmath15 @xmath16 is the apocentre radius of a radial orbit with energy @xmath1 ( so that @xmath17 ) , @xmath18 is the angular momentum of a circular orbit with the given energy , and @xmath19 is the period of a radial orbit ( time needed to complete one oscillation in the radial direction ) . ignoring the time dependence of potential , the fokker
planck equation reads @xmath20 with @xmath21 being the averaged values of corresponding quantities over the phase volume accessible to the orbit . for the spherical case ,
these averages are given by @xmath22 the calculation of averaged coefficients for energy is made easier by introduction of a few auxiliary functions : @xmath23 [ eq : k ] k_1(e ) & _
e^0 def(e ) , + k_g(e ) & _ ( 0)^e def(e ) g(e ) , + k_h(e ) & _ ( 0)^e def(e ) h(e ) . the function
@xmath24 measures the mass of stars having energy below @xmath1 , while @xmath25 does the same for kinetic energy ( up to a factor @xmath26 ) .
these three functions , together with @xmath27 and its derivative @xmath28 , can be tabulated for the given combination of @xmath29 and @xmath30 and cheaply interpolated to obtain the drift and diffusion coefficients : [ eq : dc_energy_av ] e _ & = , + e^2 _ & = 2/g(e ) .
having reviewed the theory of two - body relaxation , we now describe the implementation of the new monte carlo code . in the present form ,
it is hardly suitable for realistic dynamical simulations of star clusters , lacking many sophistications found in other existing codes .
we assume a population of identical single stars , neglect the dynamical influence of binaries and stellar evolution , and do not consider external tidal forces that would lead to escape of stars from the systems .
most of these ingredients are not difficult to add ; the purpose of this paper is to show the feasibility and benefits of non - spherical dynamical monte carlo modelling .
the main advantage of position - dependent diffusion coefficients in velocity is that one may apply them to orbits of arbitrary shape , not restricted to spherical symmetry . for a very general and flexible representation of the potential , we use two variants of spherical - harmonic expansions implemented in the publicly available smile software @xcite : basis - set and spline expansions . in both cases , the angular dependence of the potential
is given by spherical harmonics , while for the radial part either a finite sum over a particular set of basis functions with adjustable coefficients is used , or the radial dependence of each spherical harmonic is represented with a spline function .
this representation typically uses @xmath31 radial terms , and the order of the angular expansion @xmath32 is sufficient for moderately flattened systems ( with major to minor axis ratio @xmath33 ) .
we refer to the appendix of the above paper for more details . throughout this section
, we denote the actual non - spherical potential in which the particles move as @xmath34 , and its associated density as @xmath35 , while the quantities from an equivalent spherical system , approximating the actual density profile ( see below ) , are without tildes .
the evolution of the @xmath0-bodysystem is followed through a series of `` episodes '' intervals of time @xmath36 during which all particles move along their orbits independently from each other ( thus the computation of orbits is trivially parallelized ) . at the end of an episode ,
the global state of the system ( the potential and the diffusion coefficients ) is updated using the orbits of particles during the episode : each orbit is sampled with @xmath37 points ( position and velocity of the given particle at regular intervals of time ) . if @xmath38 , this increases the effective number of particles used in recomputing the potential and distribution function , reducing the discreteness noise .
the motion of particles in the smooth potential of the entire system is computed using one of the ode integrators from smile : a standard eighth - order runge kutta method dop853 @xcite , or several other methods from the odeint package @xcite .
after each timestep , the perturbations to the velocity are computed as [ eq : vel_changes ] v _ & = v _ t + _ 1 , + v_&= _ 2 , where @xmath39 are two independent random numbers with standard normal distribution , @xmath40 is the timestep adjusted internally in the ode integrator , and the diffusion coefficients are given by ( [ eq : dc_vel ] ) .
while there are more sophisticated methods for dealing with stochastic differential equations ( e.g. * ? ? ?
* ) , we used the simplest explicit order 0.5 method for the stochastic part in combination with a high - order method for the deterministic part , which enables to follow the unperturbed trajectories with a great accuracy .
we have checked that the choice of the integration method and the timestep criterion for the ode integrator do not affect the statistical properties of accumulated changes in energy and angular momentum after a given interval of time @xmath41 .
typically , the ode integrator places several tens of timesteps and reaches the energy conservation error of better than @xmath42 per @xmath5 for an unperturbed orbit . a similar approach ( interleaving the evolution in the smooth field with the two - body perturbations )
was used by @xcite for studying the effect of noise on the behaviour of near - resonant orbits in spiral galaxies , and by @xcite for simulating the tidal mass - loss from galactic satellites ( they used diffusion coefficient computed under the approximation of locally maxwellian velocity distribution ) .
the treatment of relaxation relies on the local diffusion coefficients ( [ eq : dc_vel ] ) which are computed using an isotropic spherically symmetric equivalent of the system under study .
this approximation ( in particular , assumption of isotropy of velocities of background stars ) is typical in the relaxation theory , however it could break down in a strongly non - spherical or anisotropic system .
a modification of the present approach could be adopted for a rotating stellar system , assuming isotropic velocity distribution in the corotating frame @xcite ; we have not implemented it here . at the beginning of the simulation ,
the equivalent spherical system is constructed by averaging the density profile @xmath43 of the actual model over angles @xmath44 , retaining only the radial dependence of mass profile @xmath45 .
then the associated spherically symmetric potential @xmath30 is computed , along with the isotropic distribution function @xmath29 from the eddington equation . later
in the course of simulation , both the spherically symmetric mass profile which produces the associated potential , and the distribution function , are updated directly from the particle orbits ( using points sampled during the episode ) .
both the mass profile and the distribution function are constructed using a penalized spline smoothing approach @xcite , similar to the one employed in the mkspherical program from smile . for
the latter , we first compute @xmath46 from the positions and velocities of sample points , using the spherically symmetric potential @xmath30 , then smooth it , and finally @xmath29 is obtained by dividing the smoothed @xmath47 by the density of states @xmath28 ( equation [ eq : ded ] ) , again from the equivalent spherical system . after calculating the distribution function @xmath29 , we compute the functions @xmath48 , @xmath49 that enter the definition of diffusion coefficients ( [ eq : dc_vel ] ) .
they depend on the energy of the test star @xmath1 and the ( spherical ) potential at the given position @xmath30 , and we store the pre - computed functions on a grid in @xmath50 space . in the course of orbit integration , the actual values of these functions are efficiently obtained from two - dimensional interpolation . for the given position and velocity of the particle in the actual ( non - spherical ) potential , the value of potential @xmath51 and energy @xmath1 used as the arguments of these functions are taken from the spherically symmetric potential @xmath30 at the given position .
thus , the actual potential @xmath52 , responsible for the regular motion , and the diffusion coefficients for the stochastic perturbations , are computed using slightly different methods
one for the actual system , the other for its spherical equivalent .
the small incoherence amounts to the approximation of spherical isotropic scattering background , as used in most previous studies , and is believed not to cause substantial distortions to the dynamics .
note that the test stars themselves do not need to be isotropic in velocity this assumption is only used for the background stars .
in contrast with the hnon s formulation of monte carlo method , in which the energy is conserved by pairwise interactions ( but not by the potential update ; see @xcite for an amendment ) , the spitzer and shapiro s variants do not have this property intrinsically : each particle randomly walks in energy independently from others . to correct for this , at the end of the episode
we compute the accumulated energy error and distribute it between particles , in proportion with their average diffusion coefficient @xmath53 ( [ eq : dc_energy_av ] ) during this episode .
this is slightly different from the correction method employed by @xcite and primarily applies the correction to those particles that have experienced the largest diffusion .
we compensate the energy error by correcting the particle velocity at the end of the episode , changing its magnitude ( but not direction ) by a necessary amount .
the true non - spherical potential used to compute particle motion is also updated at the end of an episode , using the same sampling points ( @xmath37 per orbit ) collected during the episode .
as already mentioned , @xmath38 reduces discreteness noise in the potential expansion coefficients ; furthermore , if @xmath54 for most of the orbits in the system , each particle completes many periods during one episode , thus again smoothing out fluctuations .
this is , in essence , the `` temporal smoothing '' proposed by @xcite but apparently never used before .
( note that there exist simulation methods that rely on `` orbit orbit '' , as opposed to `` particle particle '' interactions , which are used for stars on near - keplerian orbits around a massive black hole @xcite ; these approaches hardly can be generalized for arbitrary potentials not dominated by a single point mass ) .
the energies of particles also need to be corrected after reinitialization of the potential , to account for the time dependence of the potential .
we adopt the method used by @xcite and @xcite , which states that the energy correction for a given particle is @xmath55,\ ] ] where @xmath56 and @xmath57 are particle positions at the beginning and end of the episode , and @xmath58 is the difference between the updated potential and the old one ( used during the episode ) . in these papers , this correction could be applied directly to the particle s energy , while in our case we are again forced to attribute it to the kinetic energy only .
more specifically , the updated velocity after the correction is related to the `` new '' velocity ( at the end of the episode but before the correction ) by 2 + _ ( r_i , ) & = 2 + _ ( r_i , ) + e_i , + v_i,^2-v_i,^2 & = ( r_i , ) - ( r_i , ) . as shown by @xcite
, this correction ensures the conservation of total energy of the system @xmath59 $ ] in his case . indeed , & e_-e _ = + & _ i m_i = + & _ i m_i - [ eq : sumenew ] + & _ i m_i + [ eq : sumeold ] + & _
[ eq : difpot ] the first two terms ( [ eq : sumenew ] ) and ( [ eq : sumeold ] ) represent the sum of energies of all particles , which is conserved by the relaxation step followed by the cancellation of fluctuations described above . the last term ( [ eq : difpot ] ) also should tend to zero in the continuum limit ( * ? ? ?
. however , his proof is valid only if the updated potential is computed from the positions of particles at the end of the episode ; if we use @xmath60 sampling points , this is no longer true .
therefore , we compute the last term explicitly , and cancel the total energy error @xmath61 by distributing it between all particles . in all these correction steps , we can only attribute the energy error to the kinetic energy by changing the magnitude of particle velocities .
this might introduce some bias , as the energy excess / deficit is attributed entirely to the kinetic energy ( and furthermore , if the energy needs to be subtracted from a particle happened to be around its turning point , the velocity may be too small to allow it , in which case it remains undercorrected ) , but it is the simplest practical way of cancelling the energy errors . we do not apply a similar correction to angular momentum fluctuations , as they remain small ( @xmath62 ) and do not have a preferred sign ( however , we only considered systems with zero total angular momentum and can not be sure that there will be no secular drift in angular momentum if it was non - zero initially ) . for the two auxiliary methods used for comparison spherical isotropic fokker planck and monte carlo codes there is no need to follow orbits in space , only the evolution of distribution function @xmath29 and its associated potential
density pair @xmath63 . in the finite - difference fokker
planck scheme , @xmath29 is sampled on a non - uniform grid in @xmath1 , and a flux - conservative implicit @xcite scheme is used ( see * ? ? ? * for a comparison of numerical methods ) . in the monte carlo scheme ,
the distribution function is sampled by discrete particles with energies @xmath64 , and during each episode , each particle performs one or more monte carlo steps with timestep @xmath40 , according to @xmath65 with @xmath66 being a random number with standard normal distribution , and the diffusion coefficients given by ( [ eq : dc_energy_av ] ) .
the timestep is assigned so that the expected change in energy does not exceed @xmath67 , with the tolerance parameter @xmath68 .
when all particles have completed the episode , a new distribution function is computed in the same way as in the full ragacode ( i.e. , using penalized spline smoothing ) .
the spherical potential is updated after the new distribution function has been computed , by using the following relation for the density : @xmath69 and then the poisson equation for the potential .
followed by recomputation of the potential , the distribution function must be changed adiabatically , which is easiest to achieve by expressing it in terms of the phase volume @xmath70 ( [ eq : phasevol ] ) instead of @xmath1 , and then transforming back using the updated potential . as the equation ( [ eq : density_from_df ] ) contains the unknown potential itself ,
it should be applied iteratively until convergence , while keeping @xmath71 constant at each iteration while @xmath1 changes . in practice
, we found that for the fokker
planck method one iteration is sufficient , provided that timestep for the update is small enough ; for the spherical monte carlo code , we perform several iterations to reduce the impact of fluctuations of potential at origin , where the number of particles is small . for a simulation of a deep collapse ,
the accumulated energy error is @xmath72 .
@xmath73 @xmath74 in this section , we describe several test problems for the new monte carlo method .
first , we demonstrate that temporal smoothing does help to reduce energy exchange between particles due to fluctuations of the potential to a negligible level , compared with the typical two - body relaxation rates .
then , we perform two standard tests : the core collapse of a plummer sphere , and the growth of a bahcall wolf cusp around a massive black hole . finally , we consider the shape evolution of a triaxial model with a black hole . in this test
we consider the relaxation rate of a spherical plummer model , evolved with different methods : @xmath0-bodysimulation with a direct - summation code , self - consistent field ( scf ) method , and the ragacode with relaxation turned off .
the goal is to demonstrate that temporal smoothing does substantially reduce the energy and angular momentum relaxation rate , compared to more direct simulation methods .
we take an @xmath75 plummer model with total mass and scale radius both equal to 1 @xmath0-bodyunits , and evolve it for @xmath76 time units , or roughly @xmath77 of the half - mass relaxation time . to measure the relaxation rate , we record the changes in energy and angular momentum of individual particles , average them over particles in each of 100 bins sorted in energy , and fit a linear regression to the squared difference between initial and current values of @xmath1 and @xmath2 as functions of time ( see @xcite for a somewhat different method of estimating the relaxation rate ) . the coefficient of this regression represents the diffusion coefficient @xmath78 ( [ eq : dc_energy_av ] ) and a similarly computed coefficient for @xmath2 .
we have checked that the growth of @xmath79 and @xmath80 is indeed close to linear in time , with occasional fluctuations . for the conventional @xmath0-bodysimulation we use the gpu - accelerated direct - summation code @xmath81grape@xcite with the sapporo library @xcite .
figure [ fig : energydiff ] demonstrates that the theoretically computed diffusion rates agree very well with the measured values from the direct @xmath0-bodysimulation without softening , using the standard value of the coulomb logarithm @xmath82 @xcite . in collisionless simulations , softening is used to reduce the graininess of the potential ; we have run another simulation with @xmath83 , which is close to the optimal value for this @xmath0 @xcite and reduces the relaxation rate by a factor of few @xcite .
the other , `` indirect '' @xmath0-bodysimulation method that we used was the scf method of @xcite , employed in two regimes : in the first case we used all expansion coefficients ( @xmath84 radial and @xmath85 angular terms ) , in the second retained only the non - zero terms for a triaxially symmetric model ( that is , cosine terms with even @xmath86 and @xmath9 ) .
figure [ fig : energydiff ] shows that the scf method demonstrates a several times lower rate of diffusion than a direct @xmath0-bodysimulation , when using all coefficients , and a further factor of few lower rate for a model with imposed triaxial symmetry .
this is not unexpected , given that the potential in the scf method is fairly smooth , but the reduction of relaxation rate is limited by the fluctuations in the potential arising from frequent updates in the coefficients , as the update interval is equal to the timestep of equations of motion ( taken to be a small fraction ( @xmath87 ) of the dynamical time in centre ) , and is comparable to the reduction due to softening in a direct @xmath0-bodysimulation @xcite .
it can further be improved by a factor of few by using a carefully constructed basis set @xcite . on the other hand ,
if we allow for less frequent updates in the potential while retaining the high accuracy in integrating the equations of motion , then the relaxation rate may be reduced even further , as shown by the simulations of ragacode with longer update intervals ( we checked that running it with the same timestep as the scf code produced identical results to the latter ) . increasing the number of sampling points @xmath37 for each particle per episode reduces the fluctuations even further .
overall , for this model we attained a factor of hundred reduction in the relaxation rate , limited only by the update frequency : if the system needs to be simulated for a time substantially shorter than its relaxation time ( or the time for any other effect to change its structure significantly ) , then the potential update may be switched off altogether , entirely eliminating this source of unwanted fluctuations . on the other hand ,
the necessary level of relaxation is readily restored by adding the stochastic two - body perturbation term to the equations of motion .
we have checked that this produced essentially the same total relaxation as the direct @xmath0-bodysimulation , if the amplitude of perturbation term in equation ( [ eq : perturbation_term ] ) was assigned accordingly , using the same values of @xmath0 and @xmath88 .
@xmath89 self - gravitating systems are known to have negative specific heat and exhibit the phenomenon of core collapse ( * ? ? ?
* chapter 18 ) . the easiest and probably most studied example is that of a plummer sphere composed of equal - mass particles , for which various studies based on isotropic fokker
planck method have found the core collapse time to be @xmath90 initial half - mass relaxation times @xmath91 ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
as discussed in the latter paper , a constant value of coulomb logarithm overestimates the relaxation rate in the centre at later stages of core collapse , as the effective number of stars in the core decreases ; anisotropic models also tend to have longer collapse times ( e.g. * ? ? ?
* ) . for this test
, we set up an @xmath92 plummer model in the virial units ( with the scale radius set to @xmath93 ) . in the calibration @xmath0-bodyrun , performed by the same code @xmath81grape ,
the moment of collapse corresponds to @xmath94 time units , in agreement with other @xmath0-bodystudies @xcite . with the standard choice of coulomb logarithm @xmath95 this corresponds to @xmath96 ; for the purpose of comparison with the isotropic fokker planck and monte carlo simulations we have used a smaller value @xmath97 , which brings the collapse time in this particular simulation into better agreement with other methods .
next we have run spherical isotropic fokker
planck and monte carlo codes , as well as the full ragacode .
the fokker
planck simulation was taken as reference , with the time until core collapse being @xmath98 , in excellent agreement with other studies .
figure [ fig : core_collapse_comparison ] shows the evolution of various quantities ( lagrangian radii , central density and potential , and velocity anisotropy ) in different simulations .
overall , the agreement between various methods is fairly good , at least until the central density increases by a factor of @xmath99 ; closer to the time of collapse , we do not expect either method to be reliable without taking into account the binary formation and heating and other phenomena beyond two - body relaxation . by the end of the simulation ,
the accumulated energy error was around 2% .
the density profile around a point mass ( massive black hole ) has a steady - state power - law solution of the fokker
planck equation , known as the @xcite cusp : @xmath100 for a single - component star cluster .
dynamical models starting from different initial conditions tend to develop the cusp at radii smaller than @xmath101 , where the influence radius of the black hole @xmath102 contains the mass in stars equal to twice the black hole mass .
this has been observed both in fokker
planck models ( e.g. * ? ? ?
* ; * ? ? ?
* ) and in @xmath0-bodysimulations @xcite .
as the relaxation time in the newtonian potential of the central point mass is proportional to @xmath103 , if the initial density profile was shallower than @xmath104 , then the cusp grows from outside in .
we have set up a spherical @xcite model with @xmath105 , @xmath106 and a central black hole with mass @xmath107 of the total mass in stars , drawing particle positions and velocities from a self - consistent isotropic distribution function ( e.g. * ? ? ?
* ) , computed numerically from the eddington s formula , using the mkspherical program from smile . then , we evolved the model until it developed the steady - state bahcall wolf profile .
the @xmath0-bodysimulation used a version of code with chain regularization ( @xmath81grapech , * ? ? ? * ) . in the fokker
planck model we adopted a zero - flux boundary condition at the black hole .
figure [ fig : bahcall_wolf ] shows the gradual evolution of the distribution function towards the @xmath108 solution . the agreement between fokker
planck , monte carlo and @xmath0-bodysimulations is again quite good .
@xmath109 up to now we have considered spherical systems , to facilitate comparison between various methods .
we now turn to the unique feature of ragacode , namely the ability to simulate systems of arbitrary geometry . for this test , we take a triaxial @xmath105 dehnen model with axis ratio of @xmath110 and a central black hole with mass @xmath111 of the total model mass .
this was one of the test models for the smilecode for @xcite modelling described in ( * ? ? ? * section 7.1 ) .
the simulation was conducted with the @xmath81grapech code using @xmath75 particles .
it was found that the model evolved towards a somewhat less flattened and less triaxial shape over the timescale of simulation ( 100 time units ) .
such evolution is not unexpected in light of previous studies ( e.g. * ? ? ?
* ; * ? ? ?
* ) , although later papers suggested that the evolution may be not as rapid as found earlier @xcite .
the driving force behind this shape evolution is thought to be the scattering of chaotic orbits by the central point mass : this would let them more uniformly populate the equipotential surface , which is typically rounder than the equidensity surface .
however , the diffusion of chaotic orbits may be greatly facilitated by the graininess of potential @xcite , and very little has been explored on this topic .
we performed simulations of the same system as studied in @xcite with the monte carlo code , in several regimes , using a combination of two options : ( i ) without two - body relaxation or with the stochastic perturbation equivalent to the relaxation rate of an @xmath75 system , and ( ii ) using a fixed initial potential , or updating the potential every 10 time units ( for a total simulation time of 100 time units ) .
we used the iterative method e1 of @xcite for computing the axis ratios of our models as functions of radius ( the same method was used in the previous paper ) .
figure [ fig : shape_evolution ] shows the evolution of shape for our four runs , together with the one from the @xmath0-bodysimulation .
clearly , in the absence of relaxation the shape does not substantially evolve , regardless of whether we update the potential or keep it fixed .
on the other hand , inclusion of relaxation dramatically accelerates the shape evolution in the central parts of the model , bringing it in good agreement with the results of @xmath0-bodysimulation .
this experiment suggests that the evolution of shape can be at least partially attributed to the discreteness noise which accelerates the chaotic diffusion .
the substantial reduction of unwanted collisional relaxation offered by the presented monte carlo scheme offers new avenues in exploring the interplay between discreteness and chaos , enabling a more robust study of chaotic diffusion and its effect on the galaxy shape ( e.g. * ? ? ?
we have reviewed the existing methods for simulating the evolution of stellar systems driven by the two - body relaxation , and proposed a new variant of monte carlo method suitable for studying non - spherical systems .
it combines the flexible representation of the smooth average potential in terms of spherical - harmonic expansion ( similar to the scf method of * ? ? ?
* ) with the spitzer s approach to the description of two - body relaxation in terms of local ( position - dependent ) velocity diffusion coefficients .
the orbits of particles are thus evolved on a dynamical timescale , with the two - body interaction between them mediated by the diffusion coefficients computed from a smooth , nearly - stationary distribution function ( in a manner similar to the shapiro s variant of the monte carlo method , but without orbit - averaging ) .
we have shown that the method reproduces some standard evolutionary models , and has a substantially reduced artificial relaxation rate ( related to random fluctuations in the potential expansion coefficients ) compared to the scf method .
the wall - clock computation time of the monte carlo code was within one hour for all simulations discussed in this paper ( using a typical multi - core desktop ) , while some of the @xmath0-bodysimulations took a few days using high - end gpus .
the ragacode is made publicly available at http://td.lpi.ru/~eugvas/raga/ ; additionally , its inclusion into the amuse framework @xcite is underway .
* the monte carlo method ( in this and other variants , with the possible exception of spitzer s original formulation ) is not suitable for systems which are not in dynamical equilibrium . * the fluctuations in velocities ( and , hence , energies ) of particles are simulated independently from each other , which means that at the end of the monte carlo episode the total energy has , in general , a non - zero accumulated deviation .
it is corrected by distributing this energy error between all particles , in proportion to their time - averaged energy diffusion coefficient , but the correction is applied to the magnitudes of velocity only .
this could in principle bias the dynamics somewhat , but at least avoids much larger errors which occur without any such correction . the total angular momentum of the system is not conserved , but its fluctuations due to discreteness noise are rather small for a reasonable particle number . *
the calculations assumed that all stars have the same mass .
this is quite easy to generalize , by allowing each simulation particle to carry a given `` token '' dynamical mass ( which enters the expression ( [ eq : dc_vel]a ) for the drift coefficient ) , and this mass needs not be related to the actual amount of mass that this particle contributes to the total potential . in other words , we generalize hnon s concept of `` superstars '' by completely separating the notions of dynamical and tracer mass . likewise
, stellar evolution may be accounted for by allowing this token mass to change with time .
we note that for all simulations in this paper , we scaled the diffusion coefficients in such a way as to model a system with the number of stars @xmath0 being the same as the number of particles in the model , but this was done only to facilitate comparison with @xmath0-bodysimulations and is not a restriction of the code .
* we ignored primordial and dynamically formed binaries and their contribution to the energy budget of the system , and did not consider the process of escape of stars .
this could be implemented in a similar way to other state - of - the - art codes ( e.g. * ? ? ?
* ; * ? ? ?
* stellar collisions in dense galactic nuclei may be accounted for by a scheme similar to @xcite and @xcite .
* the discrete nature of mass tracers makes it difficult to simulate a system with high density contrast without resorting to mass refinement schemes .
fortunately , in our implementation , the mass of a particle can be set in an arbitrary way , for instance , creating initial conditions with higher mass resolution where necessary ( e.g. * ? ? ?
* ; * ? ? ?
* ) . however , if the evolution time is substantially longer than the central relaxation time , particles will tend to mix in energy , erasing the effect of mass refinement . to combat this , an adaptive `` creation annihilation scheme '' , such as that employed by @xcite and @xcite , could be added to the algorithm .
however , this mixing would also presumably drive the system towards spherical symmetry , so that the benefits of the arbitrary - geometry code would be irrelevant ; for systems with longer relaxation times ( such as galactic nuclei ) the initial mass refinement should suffice . *
the diffusion coefficients are computed under the approximation of a spherically symmetric isotropic distribution function of background stars .
this is perhaps the most fundamental limitation , and it means that we may reliably simulate only systems that are not too flattened and not too far from isotropy . it is known that in stellar systems that are at least partially rotationally supported , the two - body relaxation proceeds faster as the velocity dispersion is lower ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
. however , it is possible to adapt the computation of diffusion coefficients for a distribution function that is isotropic in the rotating frame @xcite . * similarly , we did not take into account the processes that are not described by standard two - body relaxation theory , such as resonant relaxation in the vicinity of a massive black hole @xcite or non - gaussian character of energy diffusion at @xmath112 @xcite .
the proper account of these processes is hindered by the fact that they are not simply described by uncorrelated random walk , and require more sophisticated statistical models ( e.g. * ? ? ?
the error introduced by neglecting these effects depends on the question being addressed .
for instance , the total rate of capture of stars by a massive black hole is rather weakly influenced by resonant relaxation , as shown by @xcite as well as by our own comparison of direct @xmath0-bodysimulations with the fokker
planck models @xcite , because the bulk of captured stars come from larger energies that those for which the resonant relaxation is effective . on the other hand
, it surely is important for stars very close to the black hole , as are relativistic effects ( e.g. * ? ? ?
* ) , also ignored in this study .
the possibility of simulating collisional relaxation for stellar systems with arbitrary shape opens up a number of opportunities , especially for studies of elliptical galaxies and galactic nuclei which are otherwise inaccessible to direct @xmath0-bodysimulations with present - day computers : * noise is known to enhance the efficiency of chaotic diffusion , especially in systems with a rich population of sticky chaotic orbits @xcite , such as triaxial dehnen models @xcite , and has been proposed to improve the phase - space coverage of chaotic orbits in the construction of schwarzschild models @xcite . however , very little is known of the implications of noise for the secular evolution of triaxial galaxies which may or may not , depending on their orbital structure evolve noticeably away from triaxiality during the hubble time @xcite .
* non - spherical galactic nuclei have been proposed as a way to increase the rate of star captures by a supermassive black hole @xcite , if the triaxiality can persist for the hubble time . on the other hand , scattering of chaotic orbits by the black hole
might destroy or reduce the triaxiality @xcite , and the collisional relaxation increases the rate of diffusion of stars into the black hole even in the axisymmetric potential @xcite .
the evolution of non - spherical black hole nuclei , including the loss of stars into the black hole and changes in the galaxy shape , is difficult to follow by conventional @xmath0-bodysimulations because of very low relaxation rates in actual galaxies , compared to what can be achieved in the direct simulations .
this topic is explored with the new monte carlo method in a separate paper @xcite . * likewise , the dynamics of binary supermassive black holes is substantially changed in a non - spherical system @xcite , although an accurate treatment of collisional relaxation in the non - spherical case is even more difficult for a binary black hole @xcite . * after implementing mass - dependent velocity drift coefficient , it becomes possible to study dynamical friction of not too massive objects ( heavier than field stars , but much lighter than the total mass of the model ) in non - spherical galaxies ( e.g. * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ) in a more self - consistent way ( including possible feedback on the galaxy shape ) .
a related idea was recently explored by @xcite , although their generic machinery of basis - set expansion was only applied for the spherical case .
likewise , collisional evolution and mass segregation in galactic nuclei has been mostly studied in the spherical case @xcite ; only a few studies have considered non - spherical nuclei , resulting , for instance , from galactic mergers ( e.g. * ? ? ? * ) or globular cluster inspirals ( e.g. * ? ? ?
* accurate treatment of escape of stars from globular clusters in a realistic tidal field of a galaxy is non - trivial @xcite , and it is quite tricky to implement it in a spherical monte carlo code @xcite .
more generally , non - spherical globular clusters may present other interesting phenomena to study ( e.g. * ? ? ?
* ) . * by modifying the expressions for diffusion coefficients using a suitable definition of background distribution function of stars ( for instance , lifting the assumption of isotropy ) , it will be possible to study rotating clusters .
i warmly thank douglas heggie and mirek giersz for detailed comments on the early version of the manuscript , and am grateful to the anonymous referee for helpful remarks that improved the presentation .
this work was partly supported by the national aeronautics and space administration under grant no .
nnx13ag92 g . | we have developed a novel monte carlo method for simulating the dynamical evolution of stellar systems in arbitrary geometry .
the orbits of stars are followed in a smooth potential represented by a basis - set expansion and perturbed after each timestep using local velocity diffusion coefficients from the standard two - body relaxation theory .
the potential and diffusion coefficients are updated after an interval of time that is a small fraction of the relaxation time , but may be longer than the dynamical time . thus our approach is a bridge between the spitzer s formulation of the monte carlo method and the temporally smoothed self - consistent field method .
the primary advantages are the ability to follow the secular evolution of shape of the stellar system , and the possibility of scaling the amount of two - body relaxation to the necessary value , unrelated to the actual number of particles in the simulation .
possible future applications of this approach in galaxy dynamics include the problem of consumption of stars by a massive black hole in a non - spherical galactic nucleus , evolution of binary supermassive black holes , and the influence of chaos on the shape of galaxies , while for globular clusters it may be used for studying the influence of rotation .
[ firstpage ] galaxies : structure galaxies : kinematics and dynamics globular clusters : general methods : numerical |
the two dimensional o(n ) non - linear @xmath0 model is important for its asymptotic freedom and dynamical generation of a mass gap .
classically the model is defined by the action @xmath1 where the real scalar fields satisfy the non - linear constraint @xmath2 . regarding the model as a classical spin system
, @xmath3 plays the role of the temperature ; large @xmath3 encourages fluctuations of the fields , while small @xmath3 discourages them .
the asymptotic freedom of the model , first shown in @xcite , implies not only the validity of perturbation theory at short distances but also the generation of a mass gap due to large field fluctuations at long distances .
the purpose of this paper is to apply the method of the exact renormalization group ( erg ) to renormalize the model consistently with a momentum cutoff .
the perturbative renormalization of the model is done usually with the dimensional regularization @xcite .
its advantage is the manifest o(n ) invariance , but an external magnetic field ( mass term ) must be introduced to avoid ir divergences .
compared with the dimensional regularization , the regularization with a momentum cutoff is physically more appealing , but it is technically more complicated ; the o(n ) invariance is not manifest , and a nave sharp momentum cutoff , inconsistent with shifts of loop momenta , can not be used beyond 1-loop .
we can overcome the technical difficulties using the formulation of field theory via erg differential equations @xcite .
for a general perturbative construction of theories with continuous symmetry , we refer the reader to a recent review article @xcite , and in this paper we give only the minimum background necessary for our purposes .
erg was first applied to the two dimensional o(n ) non - linear @xmath0 model by becchi @xcite ; we aim to simplify and complete his analysis .
in particular , we give a perturbative algorithm for constructing the wilson action of the model with a finite momentum cutoff @xmath4 .
the wilson action results from an integration of fields with momenta larger than @xmath4 , and it is free from ir divergences without an external magnetic field . throughout the paper
we use the euclid metric and the following notation for momentum integrals : @xmath5 a short summary of this paper has appeared in sect .
6.4 of @xcite .
we regularize the model using a uv momentum cutoff @xmath6 . the bare action is given by @xmath7 where the subscript @xmath8 , running from @xmath9 to @xmath10 , is summed over .
the interaction part is given by @xmath11\end{aligned}\ ] ] where we denote @xmath12 .
@xmath13 are functions of @xmath14 and depend logarithmically on the cutoff @xmath6 .
@xmath15 is the most general interaction action allowed by the manifest o(n@xmath161 ) invariance and perturbative renormalizability in the absence of any dimensionful parameters .
the propagator given by the free part of ( [ bare action ] ) is proportional to the smooth cutoff function @xmath17 . by choosing @xmath18 such that 1 .
@xmath18 is a positive and non - increasing function of @xmath19 , 2 .
@xmath20 for @xmath21 , 3 .
@xmath18 damps rapidly ( faster than @xmath22 ) as @xmath23 , we can regularize the uv divergences of the model .
the renormalization functions @xmath24 must be fine tuned , first for renormalizability , and then for the o(n ) invariance .
the wilson action with a finite momentum cutoff @xmath4 has two parts : @xmath25 the free part @xmath26 gives the propagator with a finite momentum cutoff @xmath4 : @xmath27 the interaction part of the wilson action is defined by @xmath28 \right ] \equiv \int [ d\phi']\ , { \nonumber}\\ & & \quad \times \exp \left [ - \frac{1}{2 } \int_p \frac{p^2}{{k \left ( p/\lambda_0 \right ) } - { k \left ( p/\lambda \right ) } } \phi'_i ( -p ) \phi'_i ( p ) + s_{i , b } [ \phi + \phi ' ] \right]\\ & & \qquad = \exp \left [ \frac{1}{2 } \int_p \frac{{k \left ( p/\lambda_0 \right)}-{k \left ( p/\lambda \right)}}{p^2 } \frac{\delta^2}{\delta \phi_i ( p ) \delta \phi_i ( -p)}\right ] \cdot \exp \left [ s_{i , b } [ \phi ] \right]{\nonumber}\end{aligned}\ ] ] alternatively , we can define @xmath29 by the differential equation @xcite @xmath30 and the initial condition @xmath31 for a fixed @xmath4 , we expand @xmath29 up to two derivatives to obtain @xmath32 \ , + \cdots \label{asymp}\end{aligned}\ ] ] where the dotted part contains four or more derivatives .
@xmath33 are functions of @xmath14 , and they can be expanded as @xmath34 the taylor coefficients depend logarithmically on the cutoff @xmath4 . we have chosen the ratio of @xmath4 to an arbitrary renormalization scale @xmath35 as the argument of the logarithm . the initial condition ( [ initial ] ) gives @xmath36 the renormalization functions @xmath24 are determined so that @xmath37 exists for any finite @xmath4 . using the bphz renormalization scheme adapted to the wilson action @xcite
, we can choose @xmath38 & @xmath39 as any functions . as will be explained in the next section , the o(n )
invariance constrains the choice of @xmath38 & @xmath39 .
alternatively , we can construct the continuum limit ( [ continuum limit ] ) directly without starting from a bare action .
we demand that the dotted part of ( [ asymp ] ) is multiplied by the inverse powers of @xmath4 . for
given @xmath38 & @xmath39 , the differential equation ( [ diffeq ] ) uniquely determines @xmath40 and the dotted part of ( [ asymp ] ) @xcite .
this is the preferred approach we adopt in the rest of the paper . in [ basic ] ,
we summarize the basic properties of the correlation functions calculated with @xmath41 .
the wilson action is determined uniquely in terms of @xmath38 & @xmath39 . for the o(n )
symmetry , we must choose @xmath42 & @xmath39 appropriately .
in this and the following two sections , we aim to complete the analysis of becchi given in sect .
6 of @xcite .
the wilson action has manifest o(n@xmath161 ) invariance . to insure the full o(n )
invariance , we must demand the invariance of the action under the following infinitesimal transformation : @xmath43}(p ) \label{deltaphi}\ ] ] where @xmath44 is an infinitesimal constant , and @xmath45}$ ] is the composite operator for the @xmath46-th component of the o(n ) vector , whose @xmath8-th component is proportional to @xmath47 .
more precisely , @xmath45}$ ] is defined by the erg differential equation @xmath48}(p){\nonumber}\\ & & \quad = \int_q \frac{\delta ( q/\lambda)}{q^2 } { \left\lbrace}\frac{\delta { s_{i,\lambda}}}{\delta \phi_i ( q ) } \frac{\delta}{\delta \phi_i ( q ) } + \frac{1}{2 } \frac{\delta^2}{\delta \phi_i ( q ) \delta \phi_i ( -q ) } { \right\rbrace}{[\phi_n]}(p)\label{phindiffeq}\end{aligned}\ ] ] and the derivative expansion @xmath49}(p ) = p \left(\ln \lambda/\mu ; \phi ( x)^2/2\right ) + \cdots\ ] ] where the dotted part , proportional to the inverse powers of @xmath4 , contains derivatives of @xmath50 .
@xmath45}$ ] is parameterized by a function @xmath51 which is arbitrary as far as perturbative renormalizability of @xmath45}$ ] is concerned .
( the composite operators , the concept of which was first introduced in sect . 5 of @xcite , can be considered as infinitesimal deformations of the wilson action , and they satisfy the same linear erg differential equation as ( [ phindiffeq ] ) .
a composite operator vanishes identically if the leading part in the derivative expansion , the part multiplied by the non - negative powers of @xmath4 , vanishes . for more details , see sect .
4 of @xcite . )
following becchi @xcite , we now define the ward - takahashi ( wt ) composite operator for ( [ deltaphi ] ) by @xmath52 { \nonumber}\\ & = & { \epsilon}_i \int_p { k \left ( p/\lambda \right ) } \left [ \frac{\delta { s_\lambda}}{\delta \phi_i ( p ) } { [ \phi_n]}(p ) + \frac{\delta { [ \phi_n]}(p)}{\delta \phi_i ( p ) } \right ] \label{wtoperator}\end{aligned}\ ] ] this satisfies the same erg linear differential equation as ( [ phindiffeq ] ) .
the wt identity @xmath53 is the `` quantum '' invariance of the wilson action under ( [ deltaphi ] ) , whereby the non - trivial jacobian of ( [ deltaphi ] ) is taken into account .
concrete loop calculations show that the coefficient function @xmath54 , corresponding to the quadratically divergent potential in the bare action , is non - vanishing .
but its non - invariance under ( [ deltaphi ] ) is cancelled by the jacobian . taking the correlation of @xmath55 with the elementary fields ,
we obtain the usual wt identity from ( [ wt - wt ] ) : @xmath56 where the renormalized correlation function @xmath57}(p_j ) \cdots \phi_{i_n } ( p_n ) \right\rangle}_{{s_\lambda}}\end{aligned}\ ] ] is independent of the cutoff @xmath4 .
( this @xmath4 independence is a consequence of the differential equations ( [ diffeq ] , [ phindiffeq ] ) .
see , for example , sect.4.1 of @xcite for more explanations . ) for @xmath41 to satisfy ( [ wt - wt ] ) , we must fine tune not only @xmath58 & @xmath59 but also @xmath60 . in the next two sections
, we will show the possibility of such fine tuning .
we expand @xmath61 , etc . , in the number of loops .
we use a superscript @xmath62 to denote the @xmath63-loop level : @xmath64}(p ) = \sum_{l=0}^\infty { [ \phi_n]}^{(l ) } ( p),\quad \sigma_\lambda = \sum_{l=0}^\infty \sigma_\lambda^{(l)},\\ a\left(\ln \lambda/\mu ; \phi^2/2\right ) = \sum_{l=0}^\infty a^{(l ) } \left(\ln \lambda/\mu ; \phi^2/2\right),\,\cdots\\ a_i \left(\ln \lambda/\mu\right ) = \sum_{l=0}^\infty a_i^{(l ) } \left(\ln \lambda/\mu\right),\,\cdots\\ a\left(0 ; \phi^2/2\right ) = \sum_{l=0}^\infty a^{(l ) } \left(\phi^2/2\right),\,\cdots \end{array}\ ] ] in this section , we show how to tune @xmath65 to satisfy @xmath66 . the leading part of the derivative expansion of @xmath67 is given by the classical action : @xmath68 @xmath69 is independent of @xmath4 , and we can write @xmath70\label{tree - s } \end{aligned}\ ] ] likewise , the derivative expansion of @xmath45}^{(0)}$ ] gives @xmath71}^{(0 ) } ( p ) = p^{(0 ) } \left(\phi^2/2\right ) + \cdots \label{tree - p}\ ] ] as a convention , we can choose @xmath72 at tree level , the wt identity gives @xmath73}^{(0 ) } ( p ) = 0\ ] ] the derivative expansion gives @xmath74 substituting ( [ tree - s ] ) into the above , we obtain @xmath75{\nonumber}\end{aligned}\ ] ] where the prime denotes a derivative with respect to @xmath14 .
for @xmath76 to vanish , we must satisfy the following three equations : @xmath77 from ( [ tree1 ] ) and ( [ tree2 ] ) , we get @xmath78 where we write @xmath79 for short . since @xmath80 , we obtain @xmath81 ; hence using ( [ tree - normalization ] ) we obtain @xmath82 thus , ( [ tree1 ] ) gives @xmath83 using ( [ tree - normalization ] ) , we obtain @xmath84\ ] ] finally , ( [ tree3 ] ) gives @xmath85 this is solved by @xmath86 where @xmath87 is an arbitrary constant .
hence , we obtain @xmath88 the constant @xmath87 may be chosen either positive or negative .
if we choose a positive @xmath89 , then we obtain @xmath90 appropriate for the classical o(n ) non - linear @xmath0 model .
if we choose a negative @xmath91 instead , we obtain @xmath92 appropriate for the classical o(n@xmath161,1 ) non - linear @xmath0 model .
we make the first choice . to summarize ,
we have obtained @xmath93 where @xmath3 is an arbitrary * positive * coupling constant .
the corresponding classical action is given by the familiar expression @xmath94\ ] ]
let us now assume that we have determined @xmath41 and @xmath45}$ ] up to @xmath63-loop level ( @xmath95 ) such that @xmath96 under this induction hypothesis , we wish to determine @xmath97 and @xmath45}^{(l+1)}$ ] ( or equivalently @xmath98 ) such that @xmath99 note that @xmath55 is a composite operator , satisfying the same erg differential equation as ( [ phindiffeq ] ) . applying the loop expansion and using the induction hypothesis , we find @xmath100 calling the leading part of the derivative expansion of @xmath101 by @xmath102
, we obtain @xmath103 since @xmath104 for @xmath105 .
hence , @xmath102 is independent of @xmath4 .
thus , we obtain @xmath106 & = & { \epsilon}_i \int d^2 x\ , \phi_i \left [ \partial_\mu \phi_j \partial_\mu \phi_j \cdot s_1 \left(\phi^2/2\right ) \right.{\nonumber}\\ & & \left.\ , + \phi_j \partial^2 \phi_j
\cdot s_2 \left(\phi^2/2\right ) + ( \phi_j \partial_\mu \phi_j)^2 \cdot s_3 \left(\phi^2/2\right ) \right]\end{aligned}\ ] ] where @xmath107 are functions of @xmath14 , independent of @xmath108 .
the definition ( [ wt - wt ] ) of @xmath55 gives the decomposition @xmath109 where @xmath110}^{(0 ) } ( p ) + \frac{\delta { s_\lambda}^{(0)}}{\delta \phi_i ( p ) } { [ \phi_n]}^{(l+1 ) } ( p)\right]\\ \sigma_\lambda^{(l+1),u } & = & { \epsilon}_i \int_p { k \left ( p/\lambda \right ) } \left [ \sum_{k=1}^{l } \frac{\delta { s_\lambda}^{(k)}}{\delta \phi_i ( p ) } { [ \phi_n]}^{(l+1-k ) } ( p ) + \frac{\delta { [ \phi_n]}^{(l ) } ( p)}{\delta \phi_i ( p ) } \right]\end{aligned}\ ] ] only @xmath111 depends on @xmath112 , and @xmath113 are determined by @xmath41 and @xmath45}$ ] up to @xmath63-loop .
therefore , the functions @xmath114 are given as the sum @xmath115 where @xmath116 are linear in @xmath112 , and @xmath117 are determined by the lower loop functions .
we obtain explicitly @xmath118 there is no relation among the @xmath119 s .
thus , whatever @xmath120 s are , we can solve the equations @xmath121 using ( [ tree - results ] ) , the solution is obtained explicitly as follows : @xmath122\\ & & p^{(l+1 ) } ( x ) = \sqrt{1 - 2 g x } \bigg [ p^{(l+1 ) } ( 0 ) + \int_0^x dy\ , \bigg\lbrace 2 a^{(l+1 ) ' } ( y ) { \nonumber}\\ & & \qquad\qquad\qquad - \frac{2 g}{1 - 2 g y } b^{(l+1 ) } ( y ) + \frac{- 2 u_1 ( y ) + u_2 ( y)}{\sqrt{1 - 2 g y } } \bigg\rbrace \bigg ] \end{aligned}\ ] ] note that @xmath123 are left undetermined as constants of integration .
this is expected , since @xmath124 normalizes the coupling @xmath3 , @xmath125 normalizes the field @xmath126 , and @xmath127 normalizes the composite operator @xmath45}$ ] .
for example , we can adopt the convention @xcite @xmath128 analogous to the minimal subtraction for dimensional regularization @xcite .
this concludes our inductive construction of the o(n ) non - linear @xmath0 model .
let us give explicitly the 1-loop corrections to 2- and 4-point vertices in the wilson action .
( a little more details are given in [ loop - calculations ] . ) for the 2-point vertex , we find @xmath129 and for the 4-point vertex , we find @xmath130 we also find the 2- and 4-point vertices for the composite operator @xmath45}(p)$ ] : @xmath131 we have fixed the @xmath4 independent part of @xmath132 , @xmath133 , and @xmath134 using the convention ( [ norma1 ] , [ normb0 ] , [ normp0 ] ) .
the two constants in @xmath135 and @xmath136 are left undetermined by the erg differential equations .
they are determined by the wt identity . calculating @xmath137}^{(0 ) } ( p)}{\delta \phi_i ( p)}\ ] ] only up to cubic in fields and up to two derivatives , we obtain @xmath138 where @xmath139 hence , we obtain @xmath140 in [ beta ] we explain how to obtain the beta function of @xmath3 and anomalous dimension of @xmath47 in the erg approach .
the above 1-loop results reproduce the well known results first obtained in @xcite : @xmath141
in this paper we have applied the erg formulation of quantum field theory for the perturbative construction of the two - dimensional non - linear @xmath0 model .
the model is parameterized by three renormalization functions @xmath142 , and we have shown how to tune these by imposing the wt identity ( [ wt - wt ] ) .
only short - distance physics can be explored perturbatively , and long - distance physics needs non - trivial approximations , such as @xmath143 . for the @xmath143 expansions it is common to linearize the o(n ) symmetry using an auxiliary field ; it would be interesting to extend the erg formulation to accommodate the auxiliary field .
the correlation functions of a wilson action @xmath41 are dependent on the cutoff @xmath4 . using the inverse of the cutoff function ,
however , we can easily construct @xmath4 independent correlation functions : @xmath144 for @xmath145 .
the @xmath4 independence of these correlation functions is a consequence of the erg differential equation ( [ diffeq ] ) .
see sect . 2 of @xcite for more details .
similarly , given a composite operator @xmath146 that satisfies the same linear erg differential equation as ( [ phindiffeq ] ) , we can construct @xmath4 independent correlation functions by @xmath147 see sect . 4 of @xcite for more details .
the interaction part of the classical action is given by @xmath148 hence , @xmath149 ) ] thus , for the graph in , we obtain the feynman rule @xmath150 as the simplest example , we consider the 1-loop contribution to the two - point vertex given by the feynman graph in . :
1-loop correction to the two - point vertex with momentum @xmath151 the erg differential equation gives @xmath152\end{aligned}\ ] ] where we have used @xmath153 hence , integrating this over @xmath4 , we obtain @xmath154 this gives @xmath155 where we have used the normalization condition @xmath156 .
as another example , let us consider the 1-loop contribution to the 1-point vertex of the jacobian : @xmath157}^{(0 ) } ( q)}{\delta \phi_i ( q ) } \label{1loop - jacobian}\ ] ] now , the leading part of the derivative expansion of @xmath45}^{(0)}$ ] is given by @xmath49}^{(0 ) } ( p ) = p^{(0 ) } \left(\phi ( x)^2/2\right ) + \cdots\ ] ] where @xmath158 hence , we obtain @xmath159 let us denote the @xmath160-point vertex @xmath161 for @xmath45}$ ] by .
then , the one - point vertex for the 1-loop jacobian is given by and calculated as @xmath162 -point vertex for @xmath45}(p)$ ] ] ) ] this cancels the contribution of the @xmath163 term to @xmath164 .
the derivation of the mass independent beta functions and anomalous dimensions in the erg formalism has been discussed in @xcite and @xcite .
the wilson action @xmath41 for a different choice of @xmath35 satisfies the same erg differential equation ( [ diffeq ] ) .
hence , @xmath165 is a composite operator satisfying the erg differential equation @xmath166 @xmath167 has the correlation functions @xmath168 expanding @xmath167 up to two derivatives , we obtain @xmath169 + \cdots\end{aligned}\ ] ] where @xmath170 especially at @xmath171 , the coefficient of @xmath172 is @xmath173 and that of @xmath174 is @xmath175 since @xmath167 is an infinitesimal change of the wilson action , it has two degrees of freedom , corresponding to the infinitesimal variation of @xmath3 and that of the normalization of @xmath47 . thus , we can construct two composite operators : 1 .
@xmath176 that generates an infinitesimal change of @xmath3 : @xmath177 the correlation functions of @xmath176 are given by @xmath178 2 .
@xmath179 that generates an infinitesimal renormalization of @xmath47 : @xmath180 @xmath179 counts the number of fields : @xmath181 @xmath167 must be a linear combination of @xmath176 and @xmath179 ; hence @xmath182 where neither @xmath183 nor @xmath184 depends on @xmath4 .
this gives the differential equation @xmath185 hence , @xmath183 is the beta function of @xmath3 , and @xmath184 is the anomalous dimension of @xmath47 .
the derivative expansion gives @xmath186 + \cdots\end{aligned}\ ] ] at @xmath171 , the coefficient of @xmath187 is @xmath188 and the coefficient of @xmath189 is @xmath190 these are consequences of the conventions ( [ norma1 ] , [ normb0 ] ) .
using the interaction part of the action , we can rewrite @xmath191 hence , the derivative expansion gives @xmath192 + \cdots\end{aligned}\ ] ] where @xmath193 up to loop corrections . at @xmath171 ,
the coefficient of @xmath194 is @xmath195 and the coefficient of @xmath189 is @xmath196 comparing the derivative expansion of @xmath167 with those of @xmath176 and @xmath179 , we obtain @xmath183 and @xmath184 as follows : @xmath199 where we have used ( [ og1 ] , [ og2 ] ) . using ( [ a11 ] , [ b01 ] ) , we obtain @xmath200 using ( [ n1 ] , [ n2 ] ) , we also obtain @xmath201 hence , at 1-loop ( [ adot ] , [ bdot ] ) give @xmath202 thus , we obtain ( [ 1loop - betagamma ] ) .
99 a. m. polyakov , phys . lett .
b * 59 * , 79 ( 1975 ) . c. g. bollini and j. j. giambiagi , nuovo cim .
b * 12 * , 20 ( 1972 ) .
g. t hooft and m. j. g. veltman , nucl .
b * 44 * , 189 ( 1972 ) .
k. g. wilson and j. b. kogut , phys .
* 12 * , 75 ( 1974 ) .
j. polchinski , nucl .
b * 231 * , 269 ( 1984 ) .
y. igarashi , k. itoh , and h. sonoda , `` realization of symmetry in the erg approach to quantum field theory , '' arxiv:0909.0327 [ hep - th ] . c. becchi , `` on the construction of renormalized gauge theories using renormalization group techniques , '' arxiv : hep - th/9607188 . m. pernici and m. raciti , nucl .
b * 531 * , 560 ( 1998 ) [ arxiv : hep - th/9803212 ] .
h. sonoda , phys .
d * 67 * , 065011 ( 2003 ) [ arxiv : hep - th/0212302 ] .
h. sonoda , j. phys . a * 40 * , 5733 ( 2007 ) [ arxiv : hep - th/0612294 ] . g. t hooft , nucl .
b * 61 * , 455 ( 1973 ) . | we use the exact renormalization group ( erg ) perturbatively to construct the wilson action for the two - dimensional o(n ) non - linear sigma model
. the construction amounts to regularization of a non - linear symmetry with a momentum cutoff .
a quadratically divergent potential is generated by the momentum cutoff , but its non - invariance is compensated by the jacobian of the non - linear symmetry transformation . |
progress in trapping and cooling of atomic gases @xcite made it possible to coherently couple a bose - einstein condensate to a single cavity mode @xcite .
these experiments pave the way to a new sub - field of amo physics ; _ many - body cavity quantum electrodynamics_. in the ultracold regime , light induced mechanical effects on the matter waves lead to intrinsic non - linearity between the matter and the cavity field @xcite .
in particular , the non - linearity renders novel quantum phase transitions ( qpt ) @xcite , see @xcite .
such non - linearity , due to the quantized motion of the atoms , is absent in the , so called , standard dicke model ( dm ) .
explicitly , the dm describes a gas of @xmath0 non - moving two - level atoms interacting with a single quantized cavity mode @xcite .
the interplay between the field intensity / energy , the free atom energy , and the interaction energy leads to a quantum phase transition ( dqpt ) in the dm .
motivated by the novel phenomena arising from a quantized treatment of atomic motion , it is highly interesting to extend the dm to include atomic motion on a quantum scale , and in particular to analyze how it affects the nature of the dqpt .
it is clear that such generalization of the dm results in new aspects of the system properties .
the atomic motion is directly affected by the shape of the field induced potentials , which in return is determined by the system parameters and the field intensity .
in addition to the terms contributing to the total energy in the regular dm , in this model we have to take into account for the motional energy of the atoms .
this enters in a non - trivial way since the interaction energy depends on the motional states of the atoms .
the dm was first introduced in quantum optics to describe the full collective dynamics of atoms in a high quality cavity .
the dm hamiltonian , given in the rotating wave approximation ( rwa ) , reads @xmath1 here , the boson ladder operators @xmath2 and @xmath3 create and annihilate a photon of the cavity mode , the pauli @xmath4-operators act on atom @xmath5 , @xmath6 , @xmath7 and @xmath8 are the mode and atomic transition frequencies and effective atom - field coupling respectively .
the dm is a well defined mathematical model for any value of its parameters .
note , however , that whether this model describes faithfully some physical situation for any choice of parameters is not guaranteed . for moderate @xmath8 ,
the dm appropriately describes the dynamics of atoms coupled to the cavity field , and it has been thoroughly discussed in terms of collapse - revivals @xcite , squeezing @xcite , entanglement @xcite and state preparation @xcite .
considerable interest , however , has been devoted to the normal - superradiant phase transition @xcite . in the thermodynamic limit ( @xmath9 , @xmath10 )
the system exhibits a non - zero temperature phase transition between the normal phase and the superradiant phase . in the _ normal phase _ , all atoms are in their ground state and the field in vacuum , while the _ superradiant phase _ is characterized by a non - zero field and a macroscopic excitation of the matter .
its critical coupling and critical temperature are @xcite @xmath11 as was shown in @xcite , the critical coupling can be seen as a condition on the atomic density @xmath12 .
interestingly , the dqpt is of second order nature without the rwa , while it is first order if the rwa has been imposed @xcite .
the corrections due to the rwa to various physical observables have been considered @xcite .
more detailed analysis about miscellaneous aspects of the dqpt have been presented in numerous publications .
especially various extensions @xcite as well as approximate methods @xcite concerning the dqpt has been outlined .
recently , c. emary and t. brandes applied the algebraic holstein - primakoff boson representation on the dm .
the method turned out to be very powerful and have since then been applied frequently to the dm @xcite . despite the numerous publications on the dqpt
, the existence of this phase transition was widely discussed .
if the two - level atoms in the dm correspond to atoms in a ground and excited state , and the transition is direct , then quantum mechanics forbids the transition . this can be seen either by realizing the necessity of adding the , so called , @xmath13-term to the hamiltonian , or by employing sum rules to bound the coefficients in the dicke model to the `` trivial '' thermodynamical phase @xcite .
the argumentation of ref .
@xcite can be generalized to a quite general no - go theorem for the dqpt @xcite , but it does not apply if the two level atoms in the dm correspond to atoms in two excited states , such as rydberg states , or if the transition is not direct . one valuable step towards an experimental realization of the dqpt was taken in relation with ref .
@xcite , where typical experimental parameters as well as losses were included .
these authors considered the two levels coupled by a non - resonant raman transition . in such conditions
the atom - field coupling @xmath8 can be tuned more or less independently of the @xmath13 term in an effective two - level model , and one can reach the regime of dqpt .
this paper , however , considers a situation in which atomic motion can be neglected due to high temperatures , _
e.g. _ the standard dqpt .
alternative situation , in which the atomic motion could be neglected , would be to consider quantum dots interacting with a cavity mode @xcite . in this paper
we extend the dm to take into account for atomic motion in a fully quantum mechanical description .
the atomic motion then introduces an additional degree of freedom to the problem , leading to novel appearances of the system phase diagrams .
the gas is assumed dilute such that atom - atom scattering can be neglected , and that the motion is restricted to one dimension due to tight confinement in the remaining two directions via external trapping .
furthermore , the atoms are assumed trapped by the cavity field itself .
consequently , a normal - superradiant qpt is not possible , since for a vanishing field the atomic trapping capability is lost .
however , we may assume a lowest bound of the field such that at least one bound state of the trapping `` potential '' is guaranteed .
this can be achieved by an external pumping of the cavity , which imposes a non - vanishing cavity field .
our research is partly carried out in the adiabatic regime , motivated by the ultracold atoms considered and its justification is numerically verified . in this adiabatic regime , the problem relaxes to solving a 1-d time - independent schrdinger equation .
in particular we study the case of a gaussian mode profile utilizing this adiabatic method . the situation with a standing wave mode profile
is also considered , however using a full numerical rather than adiabatic approach . for a gaussian profile
the number of bound states is crucial for the thermodynamics , and we find great divergences from the regular dm . among these
are the existence of both first and second order qpt s and multiple superradiant phases .
the second , standing wave mode , shows slight similarities to the model of @xcite but the qpt is found to be of second order , and the pt survives for zero temperature and finite @xmath6 opposite to the regular dqpt .
the paper is organized as follows . in the next section we present the generalized dm which includes the motion of the atom .
the adiabatic diagonalization of the single particle hamiltonians utilized for the gaussian mode profile is introduced and the general expression for the partition function given . the following sec .
[ sec3 ] considers the situation of a gaussian mode profile .
we thoroughly discuss the importance of bound states .
section [ sec4 ] instead considers a standing wave mode profile in a fully numerical fashion . in the appendix , however , we derive analytical expressions in the regimes of tight binding which enables us with various asymptotic properties .
last we conclude with a summery in sec . [ sec5 ] .
we consider an gas of @xmath0 ultracold identical two - level atoms , with mass @xmath14 and energy level separation @xmath15 , interacting with a single cavity mode with frequency @xmath6 . for a low temperature gas
we include atomic center - of - mass motion and mode variation . in the dipole and rotating wave approximation
, the extended dm becomes @xmath16\!.\ ] ] here , @xmath17 and @xmath18 the scaled center - of - mass momentum and position of atom @xmath5 respectively , @xmath19 the effective position - dependent atom - field coupling and @xmath20 is the mode volume . throughout the paper
we will use scaled variables such that @xmath21 .
the case of a single atom is given by the generalized jaynes - cummings hamiltonian studied by numerous authors @xcite .
in the thermodynamic limit we let @xmath22 and @xmath23 such that the atomic density is fixed ; @xmath24 .
the partition function reads @xmath25,\ ] ] where @xmath26 and @xmath27 is the scaled temperature and the trace is over the field and atomic degrees of freedom .
it is convenient to perform the trace of the field in terms of glauber s coherent states , @xmath28 . in the thermodynamic limit
one may replace @xmath29 and @xmath30 in the evaluation of the partition function @xcite .
in other words ; in the large atom number limit the photon ladder operators , or more precisely @xmath31 and @xmath32 , can be treated as commuting operators .
using the fact that atomic operators mutually commute between themselves , for example @xmath33=i\delta_{ij}$ ] , the partition function can be written @xmath34\right\}^n,\ ] ] where the integration is over the whole complex @xmath35-plane and @xmath36 in the sense of ultracold atoms as considered here , the kinetic energy of the atoms is assumed smaller than the effective potential energy .
provided that the adiabatic potentials do not cross , it is then legitimized to perform an adiabatic diagonalization of the internal states @xcite . in this regime
, the single particle hamiltonian relaxes to two decoupled adiabatic ones @xmath37 this approximation will be imposed in the next section considering a gaussian mode profile .
however , in the proceeding section dealing with the standing wave mode such an approach is not advocate , since then the curve crossings between the adiabatic potentials break adiabaticity @xcite .
the justification of the adiabatic approximation applied to the gaussian mode profile will be discussed in the end of next section . within this regime ,
the problem has become one of solving for the eigenvalues of two time - independent decoupled schrdinger equations .
the adiabatic hamiltonians depend solely on the norm @xmath38 and in polar coordinates the angle part can therefore be integrate out . by denoting the eigenvalues @xmath39 respectively , where @xmath40 and @xmath41 is a quantum number / numbers that can be either discrete and/or continuous , we get the adiabatic partition function @xmath42\!+\!\mathrm{tr}\big[\mathrm{e}^{-\beta e_n^-(r)}\big]\!\right\}^n.\ ] ] without loss of generality we can choose @xmath43 as it only scales the effective atom - field coupling .
it is worth mentioning that the numerics deal with exponentially large numbers , especially for small temperatures , which restrict the analysis to certain ranges .
a fabry - perot cavity has eigenmodes that are , to a good approximation , gaussian in the transverse and harmonic in the longitudinal direction . assuming an external deep trap in the longitudinal direction and one transverse direction
, we may consider the one dimensional problem in which the atom field coupling has a spatial gaussian shape . as is well known @xcite , and seen from eq .
( [ adham ] ) , only atoms in the `` adiabatic '' internal state corresponding to the hamiltonian @xmath44 will feel an attractive potential , while the others will be scattered away from the cavity field .
we therefore consider only a sub `` quasi '' hilbert space containing the bound states @xmath45 of @xmath46 where @xmath47 is the transverse mode width . in order to proceed in an analytic way
, we make the following approximate ansatz , @xmath48 the unknown constants are determined from the conditions : ( i ) the two functions have the same asymptotic values for @xmath49 , ( ii ) their maximum are the same and ( iii ) they share the same fwhm .
explicitly this yields @xmath50}}}. \end{array}\ ] ] the bound eigenvalues of the hamiltonian @xmath51 are known to be @xcite @xmath52 ^ 2.\ ] ] let us introduce the number of bound states , for a given set of parameters , as @xmath53 and define the function @xmath54 with this , the partition function ( [ partap ] ) , considering only bound states , becomes @xmath55}.\ ] ] by the variable substitution @xmath56 we get @xmath57\big]}.\ ] ] in the thermodynamic limit , this integral is solved by the saddle point method @xcite @xmath58\big]}\right\},\ ] ] where @xmath59 is a constant .
note that @xmath60 has the meaning of scaled field intensity .
one obstacle of the above model already mentioned in the introduction , is the fact that for a shallow potential well the number of bound states will vanish . in this limit
, the cavity field can no longer serve as a trap for the atoms .
consequently the ground state is the one of zero atoms , and we can not have a proper thermodynamic limit @xmath23 .
we therefore add the constrain of a minimum of one bound state is assumed .
this can be met experimentally by including an external driving of the cavity mode , so that the field is non - zero throughout .
thus , the `` normal '' phase will contain a non - zero cavity field which is sustained by the external pumping .
we have numerically checked that this does not introduce any significant changes of the analysis . for the number @xmath53 of bound states ,
we have @xmath61 naturally , @xmath53 depends on the system parameters . as @xmath53 is an integer , a change in the system parameters may bring about jumps between integer numbers of @xmath53
this will cause discontinuities in the function @xmath62 .
letting @xmath63 we find @xmath64 . for small fields , @xmath65 , the potential amplitude @xmath66 vanishes and the bound states cease to exist .
however , for small but non - zero fields , the above inequality may be met for large couplings @xmath8 and widths @xmath47 . of eq .
( [ phi1 ] ) .
the inserted numbers give the couplings @xmath8 and the other dimensionless parameters are @xmath67 , @xmath68 and @xmath69 .
, width=302 ] to study eq .
( [ maxpart ] ) , we analyze the parameter dependence of the function @xmath70.\ ] ] note that @xmath71 is the free energy per particle , given a scaled field intensity @xmath60 .
let us briefly discuss characteristics of @xmath71 before approaching it numerically .
the first part arises from the bare field , and it is energetically favorable to have a vanishing field .
the second part contains the atom - field interactions plus kinetic and potential atomic energies .
the interaction energy enters indirectly into the atomic potential part . increasing the field amplitude deepens the potential well and
therefore lowers the energy , and it is therefore more beneficial to have a large field .
the two terms therefore compete , and in particular , the location of the maximum of @xmath71 depends on the particular system parameters used .
thus , atomic motion , directly related to the shape and depth of the adiabatic potential , is a crucial ingredient for the qpt .
if the smallest possible @xmath60 maximizes the function , the system is said to be in a `` _ normal _ '' phase ( in quotes because the field is still non - zero to guarantee at least one bound state ) , while if a non - minimal @xmath60 is optimal the system is in a _
superradiant _ phase . in the limit of large @xmath60 ,
the second term diverges as @xmath72\sim\sqrt{y}$ ] while the first term goes as @xmath73 , and we conclude that a maximum of @xmath71 is only obtained for a finite @xmath60 . these reflections are numerically verified in fig .
[ fig1 ] showing @xmath71 for four different couplings @xmath8 .
we see that there is a critical coupling @xmath74 for which @xmath75 the system is in a `` normal '' phase and for @xmath76 it is in a superradiant phase . for the potential well ( [ potwell ] ) as function of @xmath6 . in ( a ) gray curve corresponds to @xmath69 and black curve to @xmath77 , while in ( b ) for gray curve @xmath78 and for black curve @xmath68 . in both plots @xmath79 , and in ( a ) @xmath68 and in ( b ) @xmath69 .
all parameters are dimensionless .
, title="fig:",width=302 ] for the potential well ( [ potwell ] ) as function of @xmath6 . in
( a ) gray curve corresponds to @xmath69 and black curve to @xmath77 , while in ( b ) for gray curve @xmath78 and for black curve @xmath68 . in both plots @xmath79 , and in ( a ) @xmath68 and in ( b ) @xmath69 .
all parameters are dimensionless . , title="fig:",width=302 ] in fig .
[ fig2 ] we display the critical coupling @xmath74 as function of @xmath6 while keeping the other parameters fixed . in ( a )
we present two examples for different @xmath27 and in ( b ) two examples for different @xmath47 . for the plots , the minimum @xmath60
is taken so that there is at least one bound state in the well . to the left of the curves the phase is superradiant , while to the right it is `` normal '' .
the structure of the phase diagram is clearly different from the one of the regular dm in which , at zero temperature , @xmath80 . for certain couplings
@xmath81 , the system is always in a superradiant phase independent of @xmath6 .
the location of these resonances are insensitive to the temperature but not to trap width @xmath47 .
the `` sharpness '' of these points makes it possible to have a `` normal''-superadiant-``normal '' qpt by fixing all parameters but the coupling @xmath8 which is varied around @xmath81 .
this novel feature comes about due to the varying number of bound states @xmath53 in the well . for
say @xmath82 and a weak coupling , the system is in the `` normal '' phase with only one bound state . as @xmath8 is increased the system goes through a qpt into a superradiant phase .
a closer numerical study shows that this qpt is caused by the sudden change of going from one to two bound states in the trap . as the coupling
is further increased , the discontinuity that arose from the appearance of a second bound state is of less importance and the system reenters the `` normal '' state .
hence , the presence of a second bound state `` forced '' the system out of the `` normal '' phase by inducing a sudden kink / maximum in the function @xmath71 . when the coupling is increased even further
, the same may happen again when a third bound state is introduced in the trapping potential .
eventually , however , the system enters a superradiant phase without reentering the `` normal '' phase . in order to discuss the character of the qpts we introduce the parameter @xmath83 that maximizes @xmath71 .
thus , @xmath84 is the scaled field intensity where a discontinuity in it indicates a first order qpt and a discontinuity in its first derivative signals a second order qpt . here
@xmath85 is the lower bound which assures at least one bound state . the parameter @xmath84 corresponding to the @xmath77 curve of fig .
[ fig2 ] ( a ) is presented in fig .
we note that the `` normal''-superradiant qpt by increasing @xmath8 is of first order , while the superradiant-``normal '' qpt for growing @xmath8 is of second order . from the figure it is not clear if the second order qpt is really a qpt or a cross - over .
figure [ fig4 ] , showing a slice ( @xmath86 ) of fig .
[ fig3 ] around one singularity @xmath87 , confirms that it is indeed a second order qpt .
thus , contrary to the regular dqpt of normal - superradiance , the qpts can be either of first or second order nature in this extended dm .
figure [ fig3 ] does not only reveal the type of pts of fig .
[ fig2 ] , but also that there is a number of first order qpts between various superradiant phases .
( [ maxy1 ] ) as a function of @xmath8 and @xmath6 . here
the other dimensionless parameters are @xmath79 , @xmath68 and @xmath77 . , width=302 ] of the previous fig .
[ fig2 ] ( b ) as function of @xmath8 when fixing the field frequency @xmath86 and the width @xmath68 .
this figure clearly demonstrates the presence of both first and second order qpts.,width=302 ] figures [ fig3 ] and [ fig4 ] characterize the type of qpt in the @xmath88 phase diagram .
the structure of the qpts of the @xmath89 phase diagram turns out to be equally interesting .
one example of the @xmath89 phase diagram is presented in fig.[fig5 ] ( a ) . as for the @xmath88 diagram ,
both first and second order qpt exist .
figure [ fig5 ] ( b ) , displaying the number of bound states , confirms that the sudden changes ( first order qpts ) are due to additional bound states .
the @xmath89 phase diagrams again display several different superradiant phases . in this case , however , there are both first and second order pts between the superradiant phases .
interestingly , our numeric analysis indicates that the @xmath89 phase diagrams seem to be fairly independent of @xmath7 . , but for the @xmath89 plane ( black curve of fig .
[ fig2 ] ) , while in ( b ) the number @xmath53 of bound states are shown .
note in ( a ) the large number of different suerradiant phases .
, title="fig:",width=302 ] , but for the @xmath89 plane ( black curve of fig .
[ fig2 ] ) , while in ( b ) the number @xmath53 of bound states are shown . note in ( a ) the large number of different suerradiant phases .
, title="fig:",width=302 ] contrary to the regular dm , in the ultracold regime atomic motion plays an important role for the system characteristics .
the confining potential is determined for a given field intensity @xmath60 , and consequently , the atomic density @xmath90 per particle depends as well on @xmath60 . in other words ,
apart from changes in the field intensity and in the atomic inversion , the phase transition will as wll be manifested in the atomic density .
a natural consequence of the coupled system considered here is that the motional state of the atoms are entangled with the internal state .
thus , the phases ( normal and superradiant ) are intrinsically different from the ones of the regular dm . the regular dqpt derives from a competition between the free field energy and the interaction atom - field energy . while the free field is minimized by vacuum , the interaction energy decreases with an increasing field intensity . in the present model
does the kinetic energy contribute to the total energy .
it is thus an interplay between three terms ; free field , atom - field interaction ( containing the bare internal atomic energies ) , and motional energies .
this is not so evident from the free energy per particle ( [ phi1 ] ) , where the kinetic energy is hidden in the second term . to illuminate the importance of the atomic motion we study the atomic inversion @xmath91 defined as the probability for a single atom to be in its excited state @xmath92 minus the probability for it to be in the ground state @xmath93 .
once the adiabatic approximation has been imposed we are left with a single internal state ; the lower adiabatic state @xmath94 . in particular
, the angle @xmath95 depends on the spatial coordinate and the inversion becomes @xmath96=\int dx\left(\sin^2\theta-\cos^2\theta\right)\rho_{at}(x).\ ] ] the above equation clarifies that the atomic motion enters the problem in a non - trivial way . it also shows how the internal atomic properties are taken care of even though in the adiabatic approximation the system properties derives from a single internal state @xmath97 .
we have numerically verified the appearance of the pt in terms sudden changes in @xmath90 ( first order pt ) or @xmath98 ( second order pt ) . in deriving the phase diagrams ,
tight confinement of the atoms in two directions has been assumed .
if only the longitudinal motion is frozen out , one regains an effective two dimensional problem whose eigenvalues are obtained from the schrdinger equation with potential @xmath99 . as in the one dimensional situation studied in this section , @xmath100 possesses a finite number of bound states and one would expect very similar phase diagrams for this two dimensional case as for the one dimensional model .
we conclude this section by analyzing the adiabatic approximation . by a simple rotation , the amplitudes @xmath35 appearing in the single atom hamiltonian ( [ oneatomham ] )
can be taken real . for real @xmath35 , the two last terms of @xmath101
is readily diagonalized by the unitary transformation @xcite @xmath102,\ ] ] where the angle @xmath103 was given right above eq .
( [ atinv ] ) .
momentum transforms as @xmath104 , @xmath105 . due to the spatial dependence of @xmath106 , the transformed hamiltonian is non - diagonal ; @xmath107 . here
, @xmath108 is the adiabatic hamiltonian ( [ adham ] ) and @xmath109 contains the non - adiabatic corrections .
explicitly one finds @xcite @xmath110.\ ] ] the kinetic energy is smaller or of the same order as @xmath111 , which provides a measure of @xmath109 in comparison to the adiabatic hamiltonian @xmath108 . for typical parameters , @xmath112 , @xmath78 , and @xmath113 ,
the terms of @xmath109 are at least one order of magnitude smaller than the terms of @xmath108 , which justifies the use of the adiabatic approximation .
in the previous section we studied how the dqpt was modified due to motion of the atoms in a finite potential well , assuming the atomic motion to be frozen out in the longitudinal and one transverse direction .
here we instead assume the atoms to move freely along the center axis of the fabry - perot cavity while tightly bound in the transverse directions .
the corresponding single atom hamiltonian ( [ oneatomham ] ) reads @xmath114 where @xmath115 is the scaled photon wave number which will be set to unity hereafter , @xmath116 .
this hamiltonian , with the field still quantized , has been considered in several papers , see for example @xcite .
normally @xmath117 , where @xmath118 is the cavity length , so neglecting boundary effects is not a crude approximation @xcite .
the hamiltonian is of the form of a generalized mathiue equation , and hence the spectrum @xmath119 is described by a band index @xmath120 and a quasi momentum @xmath121 extending over the first brillouin zone . due to the internal two - level structure of the atom ,
the brillouin zone is twice the size of what is imposed by the periodicity of the mode @xcite .
clearly , @xmath119 depends on the field amplitude @xmath35 .
the corresponding eigenfunctions are written @xmath122 . for a constant coupling @xmath123 ,
these _ bloch functions _ are simple plane waves giving a constant energy shift independent of system parameters such as @xmath8 , @xmath6 , and @xmath7 . for a standing wave mode coupling on the other hand ,
the bloch functions can not be decoupled from the internal atomic states , and consequently the atomic motion will affect the structure of the phase diagrams as will be demonstrated below . in the previous section we assumed the adiabatic regime and
diagonalized the hamiltonian in its internal degrees of freedom . in this case , the adiabatic potentials @xmath124 cross and adiabaticity breaks down in the range where the qpts occur .
fortunately , the hamiltonian is easily diagonalized numerically by truncation the dimension of the hamiltonian matrix .
we present , however , asymptotic analytical results in the appendix which relies on the adiabatic approximation .
these analytical results enable us to extract the limiting situation of large field amplitudes . furthermore , as a numerical diagonalization directly renders several of the bloch bands we do not restrict the analysis to just the lowest one . however , it turns out that for most of the presented examples only the lowest band contribute to the dynamics due to the low temperatures considered .
exceptions are in the plots of the critical temperature where we indeed go to rather high temperatures and the excited bands become important .
the partition function is written like in the previous section as @xmath125 where @xmath126 is a constant and the free energy per particle @xmath127\ ] ] with @xmath128 here , as above , @xmath129 represent the scaled field intensity .
shown in the appendix , the second part of @xmath130 scales asymptotically as @xmath131 for large @xmath60 .
thus , a maximum of the free energy can only be obtained for finite or zero field intensities @xmath60 .
as function of @xmath6 ( a ) and @xmath7 ( b ) . in ( a ) @xmath79 and in ( b ) @xmath132 .
the numbers to each curve display the dimensionless temperature @xmath27 .
note in particular that for @xmath133 , the critical coupling @xmath134.,title="fig:",width=302 ] as function of @xmath6 ( a ) and @xmath7 ( b ) . in ( a )
@xmath79 and in ( b ) @xmath132 .
the numbers to each curve display the dimensionless temperature @xmath27 .
note in particular that for @xmath133 , the critical coupling @xmath134.,title="fig:",width=302 ] as in sec .
[ sec3 ] , we derive the critical atom - field coupling @xmath74 and temperature @xmath135 . in fig .
[ fig6 ] we show the results of how the critical coupling depends on @xmath6 ( a ) and @xmath7 ( b ) for different temperatures .
the critical temperature as function of @xmath6 and @xmath7 is displayed in fig .
[ fig7 ] , where the inserted numbers indicate the values of the coupling @xmath8 . in both cases , the critical quantities show clear differences compared to the ones of the regular dm , ( [ dickecrit ] ) .
the critical coupling scales as @xmath80 for fixed @xmath7 just like in the regular dm .
however , the @xmath7-dependence is not possessing the same structure as ( [ dickecrit ] ) , and especially for small values on @xmath7 the critical coupling @xmath74 is non - zero .
this was also found in a nearest - neighbor coupling model studied in @xcite .
as function of the system parameters .
the inserted numbers give the value of @xmath8 .
we note that a qpt is possible at zero temperature and finite @xmath6 , and moreover , that for zero @xmath7 the qpt may vanish for small temperatures . in ( a ) @xmath79 while in ( b ) @xmath132 .
, title="fig:",width=302 ] as function of the system parameters .
the inserted numbers give the value of @xmath8 .
we note that a qpt is possible at zero temperature and finite @xmath6 , and moreover , that for zero @xmath7 the qpt may vanish for small temperatures . in ( a ) @xmath79 while in ( b ) @xmath132 .
, title="fig:",width=302 ] in the regular dm , for a fixed @xmath8 and @xmath7 the critical temperature diverges for small @xmath6 and goes to zero for large @xmath6 . here
we note that for high temperatures the regular behavior is regained , while for low temperatures , and in particular zero temperature , a qpt takes place for finite @xmath6 .
fixing @xmath6 instead and vary @xmath7 we get even more surprising results . in the regular dm , there is an upper temperature for which the qpt is lost and at zero temperature a qpt occurs for finite @xmath7 . in our model ,
a similar phase - diagram is obtained for a range of parameters @xmath6 and @xmath8 , but there also exist parameter regimes where no qpt occurs for zero temperature . like in the previous section the nature of the pts
is studied by introducing the scaled field intensity @xmath84 maximizing @xmath130 .
it is found that the qpt s are of second order character in all cases .
in this work we have studied a new regime in the dm .
the atoms are assumed trapped by the cavity field itself and ultracold such that their center - of - mass kinetic energy is of the order of the atom - field interaction .
this calls for a full quantum mechanical treatment of the atomic motion and at the same time take into account for spatial mode variations .
the analysis is motivated from our earlier findings , where we demonstrated that atomic motion greatly affects the system dynamics in many - body cavity qed systems @xcite .
expectedly , we have shown that this is also true for the dm .
the analysis is restricted to considering one dimensional problems , and both the case of a gaussian and a standing wave mode profile were treated .
however , we motivated that similar phenomena are expected also for higher dimensional situations .
in particular , we made evident that the varying number of bound states in the potential formed by a gaussian mode profile induces novel first order qpts .
additionally , we found that for certain couplings @xmath81 , the system is superradiant for any field frequency @xmath6 .
moreover , great differences with the regular dm was also encountered for a standing wave mode profile .
here , we analytically consider the large field asymptotic expressions for the energy per particle @xmath130 in the case of a standing wave mode profile . in this regime
one may utilize the tight binding approximation @xcite to derive the spectrum .
we further assume the adiabatic approximation to be valid and that we can restrict the analysis to the lowest energy band .
the adiabatic potentials are given by @xmath136 where @xmath129 is as before the scaled field intensity .
a convenient base for writing down the periodic hamiltonian in matrix form is to use the wannier states @xmath137 @xcite . the function @xmath138 is localized in the @xmath139th `` well '' of the potentials @xmath124 . by the _ tight binding approximation _ we assume @xmath140 unless @xmath141 or @xmath142 . within the validity regime of this approximation
we may as well replace the wannier functions by gaussian functions @xcite .
the widths of the gaussians are given by approximating the potential wells by harmonic oscillators , giving @xmath143{\pi\sigma^2}}\mathrm{e}^{-\frac{(x - x_j)^2}{2\sigma^2}},\ ] ] where @xmath144 and @xmath145 is the position of the @xmath139th potential well .
to avoid un - physical contributions from the non - orthogonality of the gaussians we impose @xmath146 .
we further introduce the matrix elements @xmath147 where we only consider @xmath148 .
we note that the wannier functions are directly related to the depth of the corresponding potential and therefore their width @xmath149 will also depend on @xmath60 .
this explains the field intensity dependence of @xmath150 .
another important observation is that @xmath151 are localized where @xmath152 are close either to its maximum or its minimum , resulting in different coupling elements @xmath153 , indicated by the @xmath154-superscript . in this notation
we get the lowest band tight binding energy @xmath1552\cos(k ) .
\end{array}\ ] ] the part in front of the cosine function is strictly negative resulting in that the ground state energy is given by @xmath156 .
the kinetic energy integrals of ( [ couplings ] ) are readily solvable , and one finds @xmath157 the potential integrals of ( [ couplings ] ) are not analytically solvable for the given potentials ( [ adpotsw ] ) .
instead we make the same kind of approximation as in section [ sec3 ] @xmath158 and identify @xmath159 within this regime we find @xmath160 we emphasize that the width @xmath161 depends on the field intensity @xmath60 ; @xmath162 the applied approximations are only reliable for @xmath163 @xcite , and it turns out that the qpts occur beyond these approximations .
nonetheless , we may find the asymptotics for the free energy @xmath130 . in the large @xmath60 limit
we find that @xmath164\sim\sqrt{y}$ ] .
consequently , the field intensity @xmath84 will always be finite , regardless of parameter choices .
we have verified numerically the @xmath60 square - root dependence of @xmath164 $ ] for large intensities .
we acknowledge support of the eu ip programme scala , esf pesc programme
qudedis , spanish mec grants ( fis 2005 - 04627 , conslider ingenio 2010 qoit ) .
furthermore , we thank prof .
kazmierz rzaewski for fruitful discussions . j.l . also acknowledges support from the swedish government / vetenskapsrdet and dr .
jon urrestilla for helpful discussions .
s. slama , g. krentz , s. bux , c. zimmermann , and p. w. courteille , phys .
a * 75 * , 063620 ( 2007 ) + f. brennecke , t. donner , s. ritter , t. bourdel , m. khn , and t. esslinger , nature * 450 * , 268 ( 2007 ) + p. truetlein , d. hunger , s. camerer , t. hnsch , and j. reichel , phys
lett . * 99 * , 140403 ( 2007 ) + y. colombe , t. steinmetz , g. dubois , f. linke , d. hunger , and j. reichel , nature * 450 * , 272 ( 2007 ) + f. brennecke , s. ritter , t. donner , and t. esslinger , science * 322 * , 235 ( 2008 ) .
m. lewenstein , a. kubasiak , j. larson , c. menotti , g. morigi , k. osterloh , a. sanpera , in _ atomic physics 20 _ , ed . by c.f .
roos , h. hffner , and r. blatt , ( aip proceedings , melville , 2006 ) , * 869 * , pp .
201 - 211 ; j. larson , b. damski , g. morigi , and m. lewenstein , phys .
rev . lett . * 100 * , 050401 ( 2008 ) ; j. larson , s. fernandez - vidal , g. morigi , and m. lewenstein , new j. phys . * 10 * , 045002 ( 2008 ) ; j. larson , g. morigi , and m. lewenstein , phys .
a. * 78 * , 023815 ( 2008 ) .
t. m. jarrett , c. f. lee , n. f. johnson , phys .
a * 74 * , 121301 ( 2006 ) ; t. c. jarret , c. f. lee , and n. f. johnson , phys .
b * 74 * , 121301(r ) ( 2006 ) ; c. f. lee and n. f. johnson , euro . phys . lett .
* 81 * , 37004 ( 2008 ) ; g. chen , x. wang , j .- q .
liang , and z. d. wang , phys .
a * 78 * , 023634 ( 2008 ) ; s. morrison and a. s. parkins , phys .
* 100 * , 040403 ( 2008 ) ; _ ibid . _ ,
a * 77 * , 043810 ( 2008 ) ; j. larson and j .-
p martikainen , phys .
a * 78 * , 063618 ( 2008 ) ; _ ibid .
_ arxiv:0811.4147 .
m. kozierowski , and s. m. chumakov , phys . rev .
a * 52 * , 4194 ( 1995 ) ; s. m. chumakov , and m. kozierowski , quant .
semiclss . opt .
* 8 * , 775 ( 1996 ) ; h. m. castro - beltran , j. j. sanchez - mondragon , and s. m. chumakov , opt .
comm . * 15 * , 348 ( 1998 ) ; g. ramon , c. brif , and a. mann , phys .
a * 58 * , 2506 ( 1998 ) .
g. doherty , and i. jex , opt
. comm . * 102 * , ( 1993 ) ; s. schneider , and g. j. milburn , phys .
a * 65 * , 042107 ( 2002 ) ; m. c. nemes , k. furuya , g. q. pellegrino , a. c. oliveira , m. reis , and l. sanz , phys .
a * 354 * , 60 ( 2006 ) .
r. gilmore , phys .
a * 55 * , 459 ( 1976 ) ; j. p. provost , f. rocca , g. vallee , and m. sirugae , physica a * 85 * , 2002 ( 1976 ) ; c. c. sun , and c. m. bowden , j. phys .
a * 12 * , 2273 ( 1979 ) ; r. r. puri , s. v. lawande , and s. s. hassan , opt . comm .
* 35 * , 179 ( 1980 ) ; f. pan , t. wang , j. pan , y. f. li , j. p. dranger , phys .
a * 341 * , 94 ( 2005 ) ; y. li , z. d. wang , and c. p. sun , phys .
a * 74 * , 023815 ( 2006 ) ; s. p. lukyanets , and d. a. bevzenko , phys .
a * 74 * , 053803 ( 2006 ) ; g. chen , d. zhao , and z. chen , j. phys .
b * 39 * , 3315 ( 2006 ) ; d. tolkunov , and d. solenov , phys . rev .
b * 75 * , 024402 ( 2007 ) ; h. goto and k. ichimura , phys . rev . a * 77 * , 053811 ( 2008 ) .
j. resken , l. quinoga , and n. f. johnson , euro .
. lett . * 69 * , 8 ( 2005 ) ; g. liberti , f. plastina , and f. piperno , phys . rev .
a * 74 * , 022324 ( 2006 ) ; t. c. jarret , a. oloya - castro , and n. f. johnson , europhyss .
lett . * 77 * , 34001 ( 2006 ) . c. emary , and t. brandes , phys .
. lett . * 90 * , 044101 ( 2003 ) ; _ ibid .
e * 67 * , 066203 ( 2003 ) ; n. lambert , c. emary , and t. brandes , phys .
lett . * 92 * , 073602 ( 2004 ) ; j. vidal , and s. dusuel , euro .
* 74 * , 817 ( 2006 ) ; g. chen , j. q. li , and j. -q .
liang , phys .
a * 74 * , 054101 ( 2006 ) .
b. g. englert , j. schwinger , a. o. barut , and m. o. scully , eourophys . lett . * 14 * , 25 ( 1991 ) ; g. m. meyer , m. o. scully , and h. walther , phys .
a * 56 * , 4142 ( 1997 ) ; t. bastin , and j. martin , phys .
a * 67 * , 053804 ( 2003 ) ; j. larson , j. phys .
b : at . mol .
. phys . * 42 * , 044015 ( 2009 ) .
g. compagno , j. s. peng , and f. persico , phys .
a * 26 * , 2065 ( 1982 ) ; w. ren , and h. j. carmichael , phys .
a * 51 * , 752 ( 1995 ) ; a. vaglica , phys . rev .
a * 52 * , 2319 ( 1995 ) ; c. j. hood , m. s. chapman , t. w. lynn , and h. j. kimble , phys . rev .
a * 80 * , 4157 ( 1998 ) ; j. larson , j. salo , and s. stenholm , phys .
a * 72 * , 013814 ( 2005 ) ; j. larson , phys . rev .
a * 73 * , 013823 ( 2006 ) . | we consider a gas of ultracold two - level atoms confined in a cavity , taking into account for atomic center - of - mass motion and cavity mode variations .
we use the generalized dicke model , and analyze separately the cases of a gaussian , and a standing wave mode shape . owing to the interplay between external motional energies of the atoms and internal atomic and field energies , the phase - diagrams exhibit novel features not encountered in the standard dicke model , such as the existence of first and second order phase transitions between normal and superradiant phases . due to the quantum description of atomic motion , internal and external atomic degrees of freedom
are highly correlated leading to modified normal and superradiant phases . |
control and optimization of quantum systems have been recognized as important issues for many years @xcite and control theory for quantum systems has been developed since the 1980s @xcite .
there has been considerable recent progress in both theory and experiment @xcite . however , despite this progress , there are still many challenges .
most quantum control schemes rely on open - loop control design based on mathematical models of the system to be controlled . however , accurate models are not often not available , especially for manufactured quantum systems such as artificial quantum dot atoms or molecules .
therefore , system identification @xcite is a crucial prerequisite for quantum control . in the quantum information domain ,
procedures for characterization of quantum dynamical maps are often known as quantum - process tomography ( qpt ) @xcite and many schemes have been proposed to identify the unitary ( or completely positive ) processes , for example , standard quantum - process tomography ( sqpt ) @xcite , ancilla - assisted process tomography ( aapt ) @xcite and direct characterization of quantum dynamics ( dcqd ) @xcite . however ,
if control of the system s dynamics is the objective , what we really need to characterize is not a global process but the generators of the dynamical evolution such as the hamiltonian and dissipation operators .
the problem of hamiltonian tomography ( ht ) , though less well - understood , has also begun to be considered recently by a few authors @xcite .
although qpt and ht differ in various regards , both try to infer information about the quantum dynamics from experiments performed on systems , and both can be studied from the point of view of system identification with broad tasks including ( 1 ) experimental design and data gathering , ( 2 ) choice of model sets and model calculation , and ( 3 ) model validation .
recently the quantum system identification problem has been briefly explored from cybernetical point of view , and underlining the important role of experimental design @xcite . in this article
we follow this line of inquiry . throughout the paper , we make the following basic assumptions : ( 1 ) the quantum system can be repeatedly initialized in a ( limited ) set of known states ; ( 2 ) that we can let the system evolve for a desired time @xmath0 ; and ( 3 ) that some projective measurements can be performed on the quantum system .
the main question we are interested in in this context is how the choice of the initialization and measurement affect the amount of information we can acquire about the dynamics of the system . given any a limited range of options for the experimental design , e.g. , a range of measurements we could perform , different choices for the initial states , or different control hamiltonians , how to choose the best experimental design , and what are the theoretical limitations ?
finally , we are interested in efficient ways to extracting the relevant information from noisy experimental data . the paper is organized as follows : in sec .
ii we discuss the model and basic design assumptions .
sec iii deals with the general question of model identifiability in various settings , and in sec iv we compare several different stategies for parameter estimation from a limited set of noisy data from simulated experiments see how they measure up .
to keep the analysis tractable we consider a simple model of a qubit subject to a hamiltonian @xmath1 and a system - bath interaction modelled by a single lindblad operator @xmath2 , i.e. , with system dynamics governed by the master equation @xmath3+\d[v](\rho),\ ] ] where the lindbladian dissipation term is given by @xmath4(\rho ) = v \rho v^\dag - \tfrac{1}{2}(v^\dag v + v v^\dag).\ ] ] we shall further simplify the problem by assuming that @xmath2 is a hermitian operator representing a depolarizing channel or pure phase relaxation in some basis . without loss of generality we can choose the basis so that @xmath2 is diagonal , in fact we can choose @xmath5 with @xmath6 and @xmath7 . under these assumptions
the master equation simplifies @xmath8(\rho ) = \tfrac{\gamma}{2}(\sz\rho\sz-\rho).\ ] ] the control hamiltonian can be expanded with respect to the pauli basis @xmath9 @xmath10 with possibly time - dependent coefficients @xmath11 .
it is convenient to consider a real representation of the system . following the approach in @xcite
we expand @xmath12 with respect to the standard pauli basis for the @xmath13 hermitian matrices @xmath14 where the coefficients are @xmath15 .
similarly expanding the dynamical operators allows us to recast eq .
( [ 3 ] ) in following bloch equation ( @xmath16 ) @xmath17 using this simple model for illustration we subsequently consider the experimental design from three aspects : ( 1 ) initialization procedures , ( 2 ) measurement choice and ( 3 ) hamiltonian design . * ( 1 ) initialization .
* we assume the ability to prepare the system in some initial state @xmath18 with respect to the basis @xmath19 , which coincides with the eigenbasis of @xmath2 .
we can formally represent the initialization procedure by the operator @xmath20 , which is the projector onto the state @xmath21 , with @xmath22 indicating initialization . with these restrictions
the design of the initialization procedure is reduced to the selection of parameter @xmath23 .
note that we assume that we can only prepare one fixed initial state , not a full set of basis states . *
( 2 ) measurement .
* we assume the ability to perform a two - outcome projective measurement @xmath24 where the measurement basis states can be written as [ m+- ] @xmath25 so that the choice of the measurement can be reduced to suitable choice of the parameter @xmath26 , and we shall indicate this by writing @xmath27 . * ( 3 ) hamiltonian . * in practice we may or may not have the freedom to choose the type of hamiltonian but it will be instructive to consider the identification problem for the following three cases : * @xmath28 and @xmath29 . * @xmath30 and @xmath31 . *
@xmath32 and @xmath33 . in this context
the experimental design problem can be reduced to the problem of choosing suitable values @xmath23 and @xmath34 to identify the system parameters @xmath35 and @xmath36 in each of the three cases above .
the problem considered here is similar to that considered in @xcite , in particular we still only allow a single initial state and single fixed measurement , but unlike in @xcite the initial state and the measurement are not assumed to commute with the dephasing operator .
we now consider the identification problem for the dephasing qubit system , concentrating on _ experimental design _
issues for simultaneously discriminating dephasing parameter and hamiltonian parameter of two - level dephasing systems .
in this special case @xmath1 commutes with the dephasing operator @xmath2 . solving the equation @xmath38 with the initial state ( [ iv ] )
gives @xmath39 and applying the binary - outcome projective measurement @xmath27 yields the measurement traces @xmath40 $ ] .
defining @xmath41 , we have @xmath42 from which we can see that we can obtain information about the system parameters @xmath43 and @xmath35 if and only if @xmath44 and @xmath45 , i.e. , if neither the initial state preparation @xmath20 nor the measurement @xmath46 commutes with @xmath1 and @xmath2 .
the measurement traces also yield information about @xmath47 and @xmath26 , i.e. , we can determine the relative angles between the initialization and measurement axis and the fixed hamiltonian / dephasing axis , if they are not known a - priori .
we also see that the visibility is maximized if @xmath48 , which will be the case if the initialization and measurement axis are orthogonal to the joint hamiltonian and dephasing axis .
these results make physical sense . as @xmath49=0 $ ] ,
if the initial state preparation @xmath20 commutes with @xmath1 and @xmath2 then the initial state is a stationary state of the dynamics and the measurement outcome is constant in time @xmath50 . if @xmath51 then the initial state is not stationary and the state follows a spiral path towards the joint hamiltonian and dephasing axis but as both @xmath1 and @xmath2 are proportional to @xmath52 , @xmath53 $ ] is a conserved quantity of the dynamics . if @xmath54 then the measurement commutes with @xmath52 , and as @xmath55 $ ] is a conserved quantity , we again obtain no information about the dynamics from which to identify the system parameters . in this case
, the equation ( [ 7 ] ) is reduced into the following equation @xmath57 and the solution of eq .
( [ 5 - 7c ] ) for the initial state ( [ iv ] ) is @xmath58 where we set @xmath59 and @xmath60.\label{a133}\end{aligned}\ ] ] if @xmath61 then @xmath62 will be purely imaginary and the sine and cosine terms above are replaced by the respective hyperbolic functions .
if @xmath63 , the expression @xmath64 must be analytically continued .
( [ 5 - 8c ] ) implies that we can obtain the estimated value of the following probabilities @xmath65 noting that @xmath66 depends on both @xmath67 and @xmath35 this shows that we can obtain full information about the system parameters if and only if @xmath68 and @xmath69 .
if @xmath70 or @xmath71 then we can still identify @xmath35 but not @xmath67 .
again this makes sense .
@xmath72 for @xmath73 , i.e. , if the initial state is an eigenstate of the hamiltonian . since @xmath49\neq 0 $ ] in this case , eigenstates of @xmath1 are not stationary .
however , since the hamiltonian and dephasing axis are _ orthogonal _ , the initial state remains in a plane orthogonal to the dephasing axis , the @xmath74 plane in our case , following the path @xmath75 .
thus we have @xmath76=0 $ ] for all times , and we can therefore not obtain any information about the hamiltonian parameter @xmath67 , but we can still obtain information about the dephasing parameter @xmath35 . if the hamiltonian and dephasing axis were not orthogonal then we would expect to be able to identify both the hamiltonian and dephasing parameters even if the initial state was an eigenstate of @xmath1 as in this case it would not remain an eigenstate of @xmath1 under the evolution . if @xmath77 but @xmath78 then the measurement commutes with the hamiltonian . transforming to the heisenberg picture , @xmath79 -\tfrac{\gamma}{2 } \d[\sz]m_{\pm}(t ) , \ ] ] and again one can show that @xmath80 remains orthogonal to the dephasing axis and @xmath81=\tr[m_{\pm}\rho(t)]$ ] is independent of the hamiltonian @xmath1 , explaining why we can not obtain any information about @xmath1 in this case . in this case equation ( [ 7 ] ) is reduced into the following equation @xmath83 whose solution for the initial state ( [ iv ] ) is @xmath84 where @xmath85 and @xmath86\\ \phi_{33}^y(t ) & = e^{-\tfrac{\gamma}{2}{t } } [ \cos\widehat{\omega}_{y}t+\tfrac{\gamma}{2\widehat{\omega}_{y } } \sin\widehat{\omega}_{y}t]\end{aligned}\ ] ] this implies that we can obtain the estimated value of the probabilities @xmath87 where the coefficient functions are [ eq : modelc : coeff ] @xmath88 as before , if @xmath89 then @xmath62 will be purely imaginary and the sine and cosine terms above turn into their respective hyperbolic sine and cosine equivalents , and if @xmath90 , the expression @xmath91 must be analytically continued . in this case
it is quite interesting to notice that it is impossible to find such @xmath23 and @xmath34 that @xmath92 and thus that we can identify both model parameters for any choice of the initial state and measurement .
this also makes sense because regardless of the choice of @xmath47 and @xmath26 the initial state in this case is always orthgonal to the hamiltonian axis , and the measurement @xmath46 is always orthogonal to @xmath1 , @xmath93\neq 0 $ ] , there are no conserved quantities and the only stationary state of the system is the completely mixed state .
in this section we explore how to estimate dephasing and hamiltonian parameters from limited noisy measurement data . in an actual experiment we can only estimate the probabilities @xmath94 at a finite number of times @xmath95 by repeatedly initializing the system in some fixed state @xmath96 , and letting it evolve for time @xmath95 before performing the projective measurement @xmath46 .
each single repetition of the experiment yields a binary outcome @xmath97 or @xmath98 and we can estimate the probability @xmath99 by repeating the experiment @xmath100 times and computing the relative frequencies of the respective measurement outcomes @xmath101 , e.g. , @xmath102 . to model noisy experimental data
we could generate @xmath103 by adding a zero - mean white - gaussian noise signal @xmath104 to @xmath105 , i.e. , @xmath106 . by the law of large numbers and iterated - logarithm law @xcite this
gives a gaussian distribution @xmath107 with mean @xmath105 and variance @xmath108 for @xmath109 .
for @xmath100 large this should be a good error model , but for small @xmath100 it may not accurately capture the nature of the projection noise , which follows a poisson distribution .
to more accurately model noisy experimental data when the number of measurement repetitions is relatively small we can simulate the actual experiment by generating @xmath100 random numbers @xmath110 between @xmath111 and @xmath112 , drawn from a uniform distribution , and setting @xmath113 , where @xmath114 is the number of @xmath115 .
one commonly used technique to find frequency components in a noisy time - domain signal is the fourier transform or its discrete version , the discrete or fast fourier transform ( fft ) . in principle
, this allows us to estimate the hamiltonian parameters ( frequencies ) @xmath116 , @xmath117 and @xmath118 @xcite , and the damping rate @xmath35 can be estimated from the lorentzian broadening of the fourier peak @xcite .
when the signal is sparse , i.e. , we only have a relatively small number of sample points , and noisy , however , this approach becomes problematic .
[ fig : exp1 ] shows that we can still identify a peak in the spectrum for sparse noisy signals but the accuracy of the estimate is limited , and accurately estimating the dephasing rate @xmath35 from the broadening of the peak , given the distortion , is very challenging .
alternatively , once an estimated value for @xmath116 has been obtained , @xmath35 can be deduced in other ways . in the simplest case
a , where the evolution is given by eq .
( [ 5 - 10b ] ) , we have @xmath119 which we can rewrite as @xmath120 given @xmath121 , if @xmath47 , @xmath26 and @xmath43 are known , we can therefore in principle calculate @xmath122 and thus @xmath123 .
however , in practice the problem is that @xmath124 and all the other parameters are not known precisely . in this case
we could estimate @xmath35 using the least square method @xcite and the principle of sequential analysis@xcite . to explore this approach we generated simulated noisy data signals @xmath125 with standard deviation @xmath126 according to the procedure above , assuming @xmath127 , @xmath128 , @xmath129 and the sampled period @xmath130s .
if the number of measurements @xmath100 at each time @xmath0 is @xmath131 then the standard deviation of estimated error is @xmath132 .
the original noiseless probability @xmath121 , the noisy signal with standard deviation @xmath133 and the power spectral density are shown in fig.[figp ] . again
the frequency @xmath116 can be easily determined by the peak value in fig .
[ figp](right ) .
we attempt to determine @xmath123 based on @xmath125 using simulated time - series . for a single time series @xmath125 the distribution of @xmath123 around the true value of @xmath134 is shown in fig .
[ figm](left ) .
the estimation is unsatisfactory even if the signal is sampled densely at a high time resolution .
the results can be noticeably improved by averaging over multiple time - series of @xmath125 .
if the mean value of the @xmath35-estimates over several time series is used as the final estimate for @xmath35 , the estimated error of @xmath123 can be remarkably reduced as shown in fig .
[ figm ] ( right ) , which indicates the mean value of @xmath123 with @xmath135 time - series respectively .
the number of simulated time - series of @xmath125 required is determined by the desired estimation accuracy . to achieve a mean value of the distribution of ca .
@xmath134 with standard deviation less than @xmath136 simulations suggest that at least @xmath137 time - series with @xmath131 are necessary , i.e. , about @xmath138 measurements have to be performed at each sample time @xmath0 to ensure that the standard deviation of the estimated error is less than @xmath136 . instead of five time series with @xmath131
we could choose four times series with @xmath139 or two with @xmath140 to reach the nearly same estimation accuracy , as shown in fig .
[ figh ] . in the actual experiment
one can obtain the mean - value sequence of @xmath123 by continually updating @xmath141 at each sampled time @xmath0 .
if the mean - value sequence is found to converge to a certain fixed value and its standard deviation satisfies the requirement , the experiment measurement should be stopped . to estimate @xmath35 by finite measurement data
one will have to identify a suitable time @xmath142 when to terminate the experiment , ideally when the true value of @xmath143 is almost @xmath111 .
a @xmath144 is almost equal to @xmath145 for @xmath146 , the signal after this time will be effectively a pure noise signal . therefore increasing the signal length beyond a certain critical time @xmath147
will not improve the estimation accuracy .
furthermore , we consider the effect of different @xmath116 and @xmath35 on the estimation accuracy .
if the true value of @xmath35 is less than the value of @xmath134 above , the estimation of @xmath116 and @xmath35 will be more accurate ( as shown in fig .
[ figa ] ) as the exponential @xmath148 decays more slowly .
if the value of @xmath116 is reduced , i.e. , the oscillation period is increased while its envelope remains unchanged the accuracy of the estimation is not influenced ( as shown in fig .
[ fige ] ) .
the previous section shows that we can in principle simultaneously identify both dephasing parameter and hamiltonian parameters for simple open systems using fourier and time - series analysis . both fourier and
time - series analysis can deal with very noisy signals but they require rather dense sampling of the signal with is both time - consuming and resource - intensive .
ideally we would like to be able to estimate the parameters by sparse sampling the signal .
one promising approach in this regard is bayesian estimation .
the main idea behind the bayesian approach is to choose the parameters to be determined , here @xmath149 and @xmath35 , to maximize a certain likelihood function @xmath150.\ ] ] where @xmath151 is the measured data and the @xmath152 are the probabilities predicted by the model , which depend on the parameters to be determined , and @xmath153 is the error variance .
following the same approach as in @xcite , we write the signals as a linear combination a small number of basis functions determined by the functional form of the signals . here
the measurement signals @xmath154 can be written as a linear combination of two basis functions @xmath155 the values of the basis functions and the coefficients are given in table [ table1 ] .chosen basis functions and coefficients for bayesian estimation . @xmath156 .
@xmath157 . [ cols="<,^,^,^",options="header " , ] figs [ fig : bayesiana][fig : bayesianc ] show that bayesian analysis allows us to estimate the model parameters not only for case a but for all cases , even if the signal is very noisy , the sampling is sparse and the measurement and initialization choices are not ideal to maximize the visibility or signal - to - noise ratio such as @xmath158 and @xmath159 .
the squeezed peaks indicate that the accuracy of the frequency estimation is much higher than the accuracy of the @xmath35-estimates .
an additional advantage of the bayesian estimation does not require a - priori knowledge of the initialization or measurement angles @xmath47 and @xmath26 .
rather , the estimation procedure provides values for the coefficients of the basis functions , which are related to the parameters @xmath47 and @xmath26 .
we have studied the issue of model identifiability and experiment design for open system dynamics for a dephasing qubit . from the examples in sec .
iii we can derive some general insights about the limits of identifiability and the role of experiment design .
unlike for process tomography , where we require the ability to prepare the system in many different input states , and the ability to measure a complete set of observables , we can in general extract information about both the hamiltonian and the dephasing parameters by repeating a single experiment , initializing the system in single fixed initial state and measuring a fixed basis .
there are certain limitations , however .
we gain no information about the value of the system parameters if the initial state is a _ stationary state _ of the system , or if the measurement is a _ conserved quantity _ , although knowledge of the stationary states or conserved quantities restricts the dynamics and thus provides indirect information about the system
. even if the initial state is not stationary or the measurement is not a conserved quantity , we may fail to obtain information about the hamiltonian parameters is if the operators @xmath1 and @xmath2 are orthogonal and @xmath46 commutes with @xmath1 , for instance .
these limitations also apply to higher dimensional systems , although for such systems additional restrictions on the identifiability of model parameters may arise . if the experiment design is such that the model parameters are identifiable there are various ways to extract the relevant parameters from a set of noisy samples ( time series ) , including fourier analysis , time series analysis and bayesian estimation .
although all of these approaches are in principle able to provide the required information , bayesian estimation appears to be superior to the alternatives , in particular when we are dealing with a limited number of noisy data points .
one reason the bayesian estimation is capable of providing far more accurate estimates for the parameters given the same input data is that it utilizes information about the structure of the signal in the form of the choice of the basis functions we are projecting onto .
this allows us to overcome restrictions on the uncertainties of the parameter estimates imposed by nyquist s law .
we acknowledge funding from the national natural science foundation of china ( grant no 60974037 ) .
sgs acknowledges funding from epsrc arf grant ep / d07192x/1 and hitachi .
we would like to acknowledge daniel k. l. oi . for helpful comments and suggestions .
consider a measurement trace of the form @xmath160 which corresponds directly to the expected result for model a if we set @xmath161 and @xmath162 . if @xmath163 then @xmath164 and @xmath165 . a similar analysis can be done for models b and c. the fourier transform of @xmath166 , where @xmath167 is the heavyside function , is given by @xmath168 .
we are interested in its absolute value @xmath169[\gamma^2+(\omega+\omega_0)^2]}\ ] ] differentiating with respect to @xmath149 and setting the numerator to @xmath111 shows that @xmath170 has extrema for @xmath171 , and in particular we have a maximum at @xmath172 with peak value @xmath173 .
thus , we could in principle estimate both the frequency @xmath174 and dephasing rate @xmath35 from the peak height @xmath175 and position @xmath36 , @xmath176 and @xmath177 , but in practice estimating the height of the peak is delicate and this approach is usually very inaccuate .
we can get a better estimate for @xmath35 using the width of the peak .
let @xmath178 be the ( positive ) frequencies for which @xmath170 assumes half its maximum or @xmath179 .
then the full - width - half - maximum @xmath180 of @xmath170 is @xmath181 or @xmath182\\ & = \left[\sqrt{\omega_*^2 + 2\sqrt{3 } \gamma\sqrt{\omega_*^2+\gamma^2}}-\omega _ * \right]\end{aligned}\ ] ] given the location @xmath36 and half - width @xmath183 of the peak we can solve this equation for @xmath35 @xmath184 where @xmath185 .
thus , in principle we can determine both the frequency and the dephasing rate by estimating the position and width of the fourier peak .
j. b. altepeter , d. branning , e. jeffrey , t. c. wei , p. g. kwiat , r. t. thew , j. l. obrien , m. a. nielsen , and a. g. white , ancilla - assisted quantum process tomography , _ phys .
90 , 193601 ( 2003 ) .
s. g. schirmer and d. k. l. oi , two - qubit hamiltonian tomography by baysean analysis of noisy data , phys .
a. 80 , 022333 ( 2009 ) ; s. g. schirmer and d. k. l. oi , quantum system identification by bayesian analysis of noisy data : beyond hamiltonian tomography , laser physics 20(5 ) , 1203 - 1209 ( 2010 ) jared h. cole , sonia g. schirmer , andrew d. greentree , cameron j. wellard , daniel k. oi , and lloyd c. hollenberg , identifying an experimental two - state hamiltonian to arbitrary accuracy , _ phys .
vol.71 , 062312 ( 2005 ) jared h. cole , andrew d. greentree , daniel k. l. oi , sonia g. schirmer , cameron j. wellard , and lloyd c. l. hollenberg identifying a two - state hamiltonian in the presence of decoherence , _ phys . rev .
vol.73 , 062333 ( 2006 ) m. zhang , s. g. schirmer , h. y. dai , w. zhou and m. lin . experimental design and identifiability of model parameters for quantum systems[c ]
_ joint proceeding of 48th ieee conference on decision and control and 28th chinese control conference_. 2009 : 3827 - 3832 . | the problem of identifiability of model parameters for open quantum systems is considered by investigating two - level dephasing systems .
we discuss under which conditions full information about the hamiltonian and dephasing parameters can be obtained . using simulated experiments several different strategies for extracting model parameters from limited and noisy data
are compared . |
non - sequential double ionization ( nsdi ) in driven two - electron atoms is a prototype process for exploring the electron - electron interaction in systems driven by intense laser fields . as such
, it has attracted a lot of interest @xcite .
most theoretical studies on nsdi are formulated in the framework of the dipole approximation where magnetic field effects are neglected @xcite . however , in the general case that the vector potential @xmath0 depends on both space and time , an electron experiences a lorentz force whose magnetic field component is given by @xmath1 .
magnetic - field effects in the non - relativistic limit are expected to arise when the amplitude of the electron motion due to the magnetic field component of the lorentz force becomes 1 a.u .
, i.e. @xmath21 a.u .
@xcite , with u@xmath3 the ponderomotive energy .
non - dipole effects were previously addressed in theoretical studies of the observed ionization of ne@xmath4 ( @xmath5 ) in ultra - strong fields @xcite , of stabilization @xcite and of high - order harmonic generation @xcite as well as in experimental studies @xcite . in recent studies of single ionization ( si ) ,
the electron momentum distribution along the propagation direction of the laser field was computed using different quantum mechanical approaches @xcite .
the average momentum along the propagation direction of the laser field was found to increase with increasing intensity and therefore increasing @xmath6 @xcite . if magnetic field effects are not accounted for , then , this momentum component averages to zero .
the motivation for these theoretical studies was an experimental observation of the average momentum in the propagation direction of the laser field @xcite
. with non - dipole effects fully accounted for , it was recently shown that the recollision mechanism @xcite is a probe of magnetic field effects in double ionization ( di ) @xcite .
specifically , for intensities around 10@xmath7 w@xmath8 for he at 800 nm ( near - infrared ) and around 10@xmath9 w@xmath8 for xe at 3100 nm ( mid - infrared ) , the average sum of the two electron momenta along the propagation direction of the laser field is found to be unexpectedly large .
it is roughly an order of magnitude larger than twice the average of the respective electron momentum in single ionization .
the propagation direction of the laser field is the same as the direction of the @xmath10 force ( to first order ) . in the current formulation ,
the @xmath11 force acts along the + y - axis .
if magnetic - field effects are not accounted for the average momentum along the propagation direction of the laser field is zero . in single ionization , this electron momentum was shown to increase with increasing intensity both in experimental and theoretical studies @xcite .
unlike single ionization , in double ionization the average sum of the electron momenta along the propagation direction of the laser field was shown in ref .
@xcite to be maximum at intensities smaller than the intensities satisfying the criterion for the onset of magnetic field effects @xmath121 a.u . @xcite . in this work ,
the mechanism underlying the unexpectedly large average sum of the components of the momenta of the two escaping electrons along the propagation direction of the laser field is identified .
namely , the magnetic field in conjunction with the recollision are shown to act as a gate .
this gate selects a subset of the initial tunneling - electron momenta along the propagation direction of the laser field .
only this subset leads to double ionization .
the tunneling electron is the electron that initially tunnels in the field - lowered coulomb potential .
when non - dipole effects are fully accounted for , it is shown that the y - component of the initial tunneling - electron momentum is mostly negative for events leading to double ionization . in the dipole approximation , this initial tunneling - electron momentum that is transverse to the direction of the electric field is symmetric with respect to zero .
the term _ non - dipole recollision - gated ionization _ is adopted to describe ionization resulting from an asymmetric distribution of the initial transverse tunneling - electron momentum due to the combined effect of the recollision and the magnetic field .
_ non - dipole recollision - gated _ ionization is shown to result in an asymmetry in a double ionization observable .
let @xmath13}$ ] denote the angle of the final ( @xmath14 ) momentum of each escaping electron with respect to the propagation axis of the laser field .
the observable in question is p@xmath15p@xmath16-p@xmath17 , where p@xmath16 is the probability of either one of the two electrons to escape with an angle @xmath18 .
p@xmath19 is accessible by kinematically complete experiments . in the dipole approximation ,
when non - dipole effects are accounted for , it is shown that , for @xmath21}$ ] , p@xmath22 and p@xmath19 has considerable values over a wide interval of @xmath18 at lower intensities .
while the former feature is in accord with the effect of the @xmath10 force , the second one is not .
namely , for the intensities considered the @xmath10 force has small magnitude that increases with intensity .
thus , one would expect the distribution p@xmath19 to be very narrowly peaked around @xmath23 , with values increasing with intensity .
ionization in strongly - driven he is studied using a three - dimensional ( 3d ) semiclassical model that fully accounts for the magnetic field during time propagation3d - smnd model .
it was recently developed in ref .
@xcite by extending a previously developed 3d semiclassical model in the framework of the dipole approximation3d - smd model @xcite to fully account for non - dipole effects .
the hamiltonian describing the interaction of the fixed nucleus two - electron atom with the laser field is given by @xmath24 where the vector potential @xmath0 is @xmath25 and @xmath26 , k , e@xmath27 are the frequency , wavenumber and strength of the electric component of the laser field , respectively .
c is the velocity of light and @xmath28 with fwhm the full - width - half - maximum of the laser field .
all coulomb forces are accounted for by setting @xmath29 .
the laser field is linearly polarized . for @xmath30 given by eq .
( [ eqn:2 ] ) , @xmath31 and @xmath32 are along the x- and z - axis , respectively , while the propagation direction of the laser field and the direction of @xmath10 are along the y - axis .
unless otherwise stated , all coulomb forces as well as the electric and the magnetic field are fully accounted for during time propagation . to switch off a coulomb interaction , the appropriate constant
is set equal to zero .
for example , to switch off the interaction of electron 1 with the nucleus , @xmath33 is set equal to zero .
moreover , the coulomb singularity is addressed using regularized coordinates @xcite which were also employed in the 3d - smd model @xcite .
the initial state in the 3d - smnd model entails one electron tunneling through the field - lowered coulomb potential with a non - relativistic quantum tunneling rate given by the ammosov - delone - krainov ( adk ) formula @xcite .
the momentum along the direction of the electric field is zero while the transverse one is given by a gaussian distribution centered around zero @xcite .
the remaining electron is initially described by a microcanonical distribution @xcite .
the tunneling and bound electrons are denoted by electrons 1 and 2 , respectively .
the 3d - smnd model is described in more detail in ref .
[ h ] in double ionization ( black dot - dashed line with open circles ) , twice the average electron momentum @xmath34 in single ionization ( black solid line with circles ) and the ratio @xmath35 ( red dotted line with triangles ) as a function of the intensity of the laser field . ( b ) the average momentum of the tunneling electron @xmath36 ( grey dot - dashed line with open circles ) and the bound electron @xmath37 ( black dot - dashed line with circles ) in double ionization and the ratio @xmath38 ( red dotted line with squares ) as a function of the intensity of the laser field . , title="fig:",scaledwidth=22.0% ] non - dipole effects in double ionization are discussed in the context of he when driven by an 800 nm laser field with 12 fs fwhm and at intensities varying from 1.3@xmath3910@xmath7 w@xmath8 to 3.8@xmath3910@xmath7 w@xmath8 . at these intensities
the ponderomotive energy @xmath40 is 2.86 a.u .
, 4.39 a.u . and 8.35 a.u .
, respectively .
thus , the maximum energy of electron 1 , 3.17@xmath41 is above the energy needed to ionize he@xmath42 . in double ionization
, the ratio @xmath43 is shown in fig .
[ figure1 ] to be maximum and roughly equal to eight at intensities 1.3@xmath3910@xmath7 w@xmath8 and 2@xmath3910@xmath7 w@xmath8 .
@xmath44 is the average sum of the two electron momenta along the propagation direction of the laser field , while @xmath34 is the corresponding average electron momentum in single ionization . in fig .
[ figure1](a ) , @xmath43 is shown to decrease with increasing intensity , for the intensities considered . moreover , in fig .
[ figure1](b ) , it is shown that it is @xmath45 of the tunneling electron that contributes the most to @xmath46 for the intensities considered .
the ratio @xmath38 has surprisingly large values at intensities smaller than the intensities satisfying the criterion for the onset of magnetic - field effects @xmath121 a.u .
@xcite . in contrast
, @xmath34 is found to increase from 0.0035 a.u . at 1.3@xmath3910@xmath7 w@xmath8 to 0.028 a.u . at 4.8@xmath3910@xmath7 w@xmath8 ,
see fig .
[ figure1](a ) and ref .
the small values of the average electron momentum in single ionization and the increase of this average with intensity are in accord with the effect of the @xmath10 force . the latter increases with intensity , since the magnetic field increases .
the change of @xmath34 with intensity has been addressed in several experimental and theoretical studies @xcite .
the mechanism responsible for the large values of @xmath48 and therefore of @xmath49 at smaller intensities of 1.3 - 2@xmath3910@xmath7 w@xmath8 is investigated . as a first step in this direction ,
@xmath50 is expressed as @xmath51 with @xmath52 for electrons 1 and 2 in double ionization .
a similar expression holds in single ionization .
@xmath18 is the angle of the final momentum of electron i with respect to the propagation axis of the laser field , i.e. @xmath53 .
p@xmath54 is the double ionization probability of electron i to ionize with an angle @xmath18 , with p@xmath55 the corresponding single ionization probability .
next , the influence of the magnetic field on @xmath56 and p@xmath54 is investigated .
@xmath57 of the tunnel electron , @xmath58 of the bound electron and @xmath59 are plotted in fig .
[ figure2](a ) , ( c ) and ( e ) , respectively , at intensities of 1.3@xmath3910@xmath7 w@xmath8 , 2@xmath3910@xmath7 w@xmath8 and 3.8@xmath3910@xmath7 w@xmath8 .
it is shown that the magnitude of the average electron momentum as a function of @xmath18 , both in double and single ionization , increases with increasing intensity .
this is evident mostly for @xmath18 around 0@xmath60 and 180@xmath60 .
more importantly , the influence of the magnetic field on the average electron momentum is illustrated at 2@xmath3910@xmath7 w@xmath8 . in fig .
[ figure2](b ) , ( d ) and ( f ) the average y - component of the momentum with the magnetic field switched - on and off ( dipole - approximation ) is plotted .
it is clearly seen that the magnetic field has no influence on any of the average electron momenta considered and that @xmath61 and @xmath62 .
this is expected when the magnetic field is switched - off , since there is no preferred direction of electron escape on the plane that is perpendicular to the polarization direction ( x - axis ) of the laser field .
the effect of the magnetic field on @xmath63 is very small even at a more differential level .
[ ht ] of the tunneling electron , ( c ) @xmath64 of the initially bound electron and ( e ) @xmath34 in single ionization are plotted as a function of @xmath18 at three intensities with the magnetic field switched - on . at 2 @xmath3910@xmath7 w@xmath8 ,
( b ) @xmath65 of the tunneling electron , ( d ) @xmath64 of the bound electron and ( f ) @xmath66 in single ionization are plotted as a function of @xmath18 when the magnetic field is switched - on and off .
@xmath18 is binned in intervals of 18@xmath60 .
@xmath67 denotes a multiplication factor of @xmath68 .
, title="fig:",scaledwidth=45.0% ] [ ht ] of the tunneling electron is plotted as a function of time for six different @xmath18 bins , with @xmath18 binned in intervals of 18@xmath60 .
time zero is set equal to the recollision time for each double - ionized event .
@xmath67 denotes a multiplication factor of @xmath68 .
, title="fig:",scaledwidth=30.0% ] specifically , in fig .
[ figure3 ] , at 2@xmath3910@xmath7 w@xmath8 , @xmath69 of the tunnel electron is plotted as a function of time for three different angles @xmath18 and for the corresponding angles @xmath70 when the magnetic field is switched - on and off .
it is noted that @xmath71 .
the zero of time is set equal to the time of recollision for each doubly ionizing trajectory .
this time is identified by the maximum in the electron pair potential energy .
the dependence of @xmath69 on time is very similar both with the magnetic field switched - on and off , the only difference being a small oscillation due to the magnetic field .
similar results ( not shown ) hold for the bound electron and for all other @xmath18s .
[ ht ] of the tunneling electron , ( c ) p@xmath72 of the initially bound electron and ( e ) p@xmath55 in single ionization are plotted as a function of @xmath18 at three intensities with the magnetic field switched - on . at 2 @xmath3910@xmath7 w@xmath8 ,
( b ) p@xmath73 of the tunneling electron , ( d ) p@xmath72 of the bound electron and ( f ) p@xmath55 in single ionization are plotted as a function of @xmath18 when the magnetic field is switched - on and off .
@xmath18 is binned in intervals of 18@xmath60 .
@xmath67 denotes a multiplication factor of @xmath68 .
, title="fig:",scaledwidth=43.0% ] next , the influence of the magnetic field on p@xmath54 is studied .
p@xmath73 of the tunnel electron and p@xmath72 of the initially bound electron are plotted in fig .
[ figure4](a ) and ( c ) , respectively , at intensities of 1.3@xmath3910@xmath7 w@xmath8 , 2@xmath3910@xmath7 w@xmath8 and 3.8@xmath3910@xmath7 w@xmath8 , while p@xmath55 is plotted in fig .
[ figure4](e ) .
all probability distributions are found to be asymmetric with respect to @xmath74 .
this asymmetry is due to the magnetic field as illustrated in fig .
[ figure4](b ) , ( d ) , ( f ) at 2@xmath3910@xmath7 w@xmath8 where p@xmath73 , p@xmath72 and p@xmath75 are plotted when the magnetic field is switched - on and off .
indeed , it is shown that when the magnetic field is switched - off all distributions are symmetric with respect to @xmath74 .
the latter is expected , since there is no preferred direction of electron escape on the plane that is perpendicular to the x - axis ( polarization direction ) .
moreover , with the magnetic field switched - on , p@xmath76 and p@xmath77 for @xmath21}$ ] .
equivalently p@xmath78p@xmath79-p@xmath80 and p@xmath81p@xmath82-p@xmath83 for @xmath21}$ ] .
this is expected , since , with @xmath10 acting along the + y - axis , an electron is more likely to ionize with a positive rather than a negative y - component of the final momentum .
[ ht ] for the tunneling electron , ( b ) p@xmath85 for the initially bound electron and ( c ) p@xmath86 in single ionization are plotted as a function of @xmath18 at three intensities .
@xmath18 is binned in intervals of 18@xmath60 .
@xmath67 denotes a multiplication factor of @xmath68 .
, title="fig:",scaledwidth=16.0% ] to better illustrate the asymmetry in p@xmath54 and p@xmath55 with respect to @xmath74 , p@xmath87 and p@xmath86 are plotted as a function of @xmath18 in fig . [ figure5 ] at three intensities .
starting with single ionization , @xmath88 is almost zero at 1.3@xmath3910@xmath7 w@xmath8 , while at the higher intensity of 3.8@xmath3910@xmath7 w@xmath8 it is sharply centered around 90@xmath60 reaching roughly 7% , see fig .
[ figure5](c ) .
these features of p@xmath55 are in accord with the effect of the @xmath10 force .
@xmath10 is small for the intensities considered and therefore it has an observable effect mostly when the y - component of the electron momentum is small as well , i.e. for an angle of escape @xmath74 .
in addition , @xmath89 is three times larger for the higher intensity considered compared to the smaller one . as a result @xmath88 has larger values at higher intensities .
next , in double ionization , p@xmath85 of the initially bound electron is plotted in fig .
[ figure5](b ) .
it is shown that p@xmath85 resembles mostly p@xmath90 rather than p@xmath91 .
indeed , p@xmath85 has larger values for higher intensities , as is the case for @xmath88 , reaching roughly 5.5% at 3.8@xmath3910@xmath7 w@xmath8 .
however , different features are found for the distribution @xmath92 of the tunneling electron , see fig .
[ figure5](a ) .
@xmath92 is much wider than @xmath88 .
the larger width of @xmath92 compared to p@xmath90 results in smaller @xmath18 and thus larger @xmath93 ( fig . [ figure2 ] ) having a significant weight in eq .
( [ eq1 ] ) .
this explains why @xmath65 is much larger than @xmath34 . also , for @xmath94}$ ] , @xmath92 has higher values at the smaller intensities of 1.3@xmath3910@xmath7 w@xmath8 and 2@xmath3910@xmath7 w@xmath8 rather than at the highest intensity of 3.8@xmath3910@xmath7 w@xmath84% compared to 2.5% .
it was shown in ref .
@xcite that strong recollisions prevail for strongly - driven he at 800 nm at intensities of 1.3@xmath3910@xmath7 w@xmath8 and 2@xmath3910@xmath7 w@xmath8 , while soft ones prevail at 3.8@xmath3910@xmath7 w@xmath8 .
it then follows that @xmath92 has higher values for strong recollisions .
the times of recollision correspond roughly to zeros of the laser field for strong recollisions and extrema of the laser field for soft ones @xcite .
moreover , the transfer of energy , compared to u@xmath3 , from electron 1 to electron 2 is larger for a strong recollision and smaller for a soft one @xcite .
[ ht ] ) ; ( c1)-(c3 ) at a time just before the time of recollision ( momentum @xmath95 ) and ( d1)-(d3 ) at the asymptotic time ( momentum @xmath96 ) .
@xmath97 are presented in panels ( e1)-(e3 ) .
the black ( grey ) lines correspond to events that have positive ( negative ) y - component of the electron 1 momentum just before the time of recollision .
the red dotted lines correspond to the total distributions of the y - component of the electron 1 momentum .
, title="fig:",scaledwidth=45.0% ] the above features are signatures of strong and soft recollisions both in the presence and in the absence of the magnetic field @xcite . a signature of recollision exclusive to non - dipole effects is the _ non - dipole recollision - gated ionization_. namely , it is found that the magnetic field and the recollision act together as a gate that selects only a subset of transverse initial momenta of the tunneling electron that lead to double ionization .
this is illustrated in fig .
[ figure6](a1 ) and ( a2 ) at an intensity of 2@xmath3910@xmath7 w@xmath8 with the magnetic field switched - on and off , respectively .
the probability distribution @xmath98 of electron 1 to tunnel - ionize with a y - component of the initial momentum equal to @xmath99 is plotted .
@xmath98 is found to be asymmetric when the magnetic field is switched - on while it is symmetric when the magnetic field is switched - off . specifically , it is more likely for electron 1 to tunnel - ionize with a negative rather than a positive y - component of the initial momentum when the magnetic field is switched - on .
the y - component of the electron momentum is parallel to the propagation direction of the laser field and along the lorentz force @xmath10 .
it is found that @xmath100 is symmetric around zero ( not shown ) when the magnetic field is switched - on and off .
this is expected since there is no force due to the laser field acting along the z - axis . when the magnetic field is switched - on , in fig .
[ figure6](a1 ) , it is shown that @xmath98 peaks around small negative values of the momentum of electron 1 .
it is also found that stronger recollisions correspond to larger asymmetry in the distribution @xmath98 of electron 1 .
indeed , @xmath101 , with @xmath102 the time electron 1 tunnel - ionizes , is computed in table i. @xmath103 is found to have larger values ( 30% ) at intensities of 1.3@xmath3910@xmath7 w@xmath8 and 2@xmath3910@xmath7 w@xmath8 compared to a smaller value ( 15% ) at 3.8@xmath3910@xmath7 w@xmath8 , see also fig .
[ figure6](a3 ) .
in contrast , the initially bound electron is found to have a symmetric distribution @xmath104 at the time electron 1 tunnel - ionizes , see table i , and at the time of recollision ( not shown ) . .@xmath103 and @xmath105 in double ionization and @xmath106 in single ionization . [
cols="^,^,^,^",options="header " , ] in single ionization , starting the time propagation from the time electron 1 tunnel - ionizes , the y - component of the escaping electron s momentum was computed both with all coulomb forces switched - off and with all coulomb forces accounted for .
it was shown in ref .
@xcite that the average y - component of the electron momentum is roughly the same for both cases . at the time electron 1 tunnel
- ionizes the y - component of the transverse momentum of electron 1 is roughly symmetric around zero , see @xmath106 in table i. this initial momentum distribution combined with the @xmath11 force give rise to the electron ionizing mostly with @xmath74 or equivalently give rise to a sharply peaked distribution p@xmath90 ( fig .
[ figure5](c ) ) . in double ionization , the full 3d - smnd model was used with all coulomb forces accounted for to propagate the y - component of the momentum of electrons 1 and 2 from the time electron 1 tunnel - ionizes up to the time of recollision . using as initial conditions the momenta of electrons 1 and 2 at the time of recollision ,
the final y - components of the momentum of electrons 1 and 2 are obtained by propagating in time following recollision with all coulomb forces switched - off and with all coulomb forces accounted for .
the average y - component of the momentum of electron 1 is roughly equal in both cases @xcite and the same holds for electron 2 .
it then follows that the decisive time in double ionization is the recollision time .
therefore , to understand the asymmetry in @xmath92 , first , the distribution @xmath107 of the y - component of the electron 1 momentum at the recollision time must be obtained .
this is done by extracting from the full 3d - smnd model the distribution of the y - component of the electron 1 momentum at a time just before the recollision , for instance at t@xmath108 - 1/50 t , with t the period of the laser field .
t@xmath109 is chosen such as to avoid the sharp change of the momenta which occurs at t@xmath110 , see ref .
@xcite . at all intensities considered
, it is found that at the time of recollision a positive over a negative y - component of electron 1 momentum is favored .
this is shown at intensities of 2@xmath3910@xmath7 w@xmath8 and 3.8@xmath3910@xmath7 w@xmath8 in fig .
[ figure6](c1 ) and ( c3 ) , respectively .
in contrast , when the magnetic field is switched - off the distribution @xmath107 is symmetric with respect to zero as illustrated for an intensity of 2@xmath3910@xmath7 w@xmath8 in fig .
[ figure6](c2 ) .
next , the mapping of the initial momentum distribution @xmath111 to the momentum distribution @xmath107 at the time of recollision and finally to the distribution @xmath92 is established . for this mapping to be more clear , first the connection between these distributions in the absence of the magnetic field is obtained at an intensity of 2@xmath3910@xmath7 w@xmath8 . in fig .
[ figure6](c2 ) , the distribution @xmath107 is split in two distributions depending on whether the y - component of the electron 1 momentum is positive @xmath112 ( black line ) or negative @xmath113 ( grey line ) at the recollision time .
both distributions involve the same number of double - ionized events .
then , in fig
. [ figure6](b2 ) , @xmath112 and @xmath113 are mapped back to the corresponding distributions at the time electron 1 tunnel - ionizes .
it is found that the positive ( negative ) y - components of the electron 1 momenta at the time of recollision correspond to mostly negative ( positive ) y - components of the electron 1 momenta at the tunnel - ionization time .
this is expected since at the time of recollision electron 1 returns to the core and thus its momentum must have roughly an opposite sign than its momentum at the tunnel - ionization time .
moreover , it is found that the positive ( negative ) y - components of the electron 1 momenta at the time of recollision correspond to y - components of the electron 1 final momenta that are slightly shifted towards negative ( positive ) values .
it is conjectured that this shift is mainly due to the sign of the y - component of the electron 1 momentum at the time electron 1 tunnel - ionizes .
electron - electron repulsion following recollision was excluded as the reason for this shift .
summarizing , in the dipole approximation mostly negative ( positive ) y - components of the electron 1 momenta at the tunnel - ionization time ( fig .
[ figure6](b2 ) ) map to positive ( negative ) y - components of the electron 1 momenta at the recollision time ( fig . [ figure6](c2 ) ) which in turn map to y - components of the electron 1 final momenta that are slightly shifted towards negative ( positive ) values ( fig . [ figure6](d2 ) ) .
however , as shown in fig .
[ figure6](e2 ) , a slight shift towards positive ( negative ) values of the y - components of the electron 1 final momenta correspond to positive ( negative ) values of @xmath92 .
the values of @xmath92 reduce as @xmath18 decreases from 90@xmath60 to 0@xmath60 .
the reason is that the y - component of the electron 1 momentum at the time of recollision is roughly the same for all bins .
however , as @xmath18 decreases from 90@xmath60 to 0@xmath60 , the y - component of the electron 1 final momentum is larger as a result of the recollision , see fig .
[ figure7 ] .
thus , the double - ionized events registered in bins with smaller @xmath18 lose " memory of the initial momentum resulting in less asymmetry . the same is found to be true when the magnetic field is on ( not shown ) .
the total @xmath92 in fig .
[ figure6](e2 ) adds up to zero .
[ ht ] at the time of recollision and equal to @xmath114 at the asymptotic time for two different @xmath18s at @xmath115 with the magnetic field switched - off .
@xmath18 is binned in intervals of 18@xmath60.,title="fig:",scaledwidth=45.0% ] with the magnetic field switched - on , at 2@xmath3910@xmath7 w@xmath8 , the asymmetric distribution @xmath107 at the time of recollision is again split in two distributions depending on whether the y - component of the electron 1 momentum is positive @xmath112 ( black line ) or negative @xmath113 ( grey line ) at the time of recollision , see fig .
[ figure6](c1 ) . unlike the case
when the magnetic field is switched - off , the distribution @xmath112 involves most of the double - ionized events when the magnetic field is switched - on . @xmath112 and @xmath113 at the time electron 1 tunnel - ionizes map back to mostly negative ( positive ) y - components of the electron 1 momenta , see fig .
[ figure6](b1 ) .
unlike the case when the magnetic field is switched - off , positive y - components of the electron 1 momenta at the time of recollision give rise to y - components of the electron 1 final momenta that are slightly shifted towards positive values , see fig .
[ figure6](d1 ) .
in contrast , as in the case when the magnetic field is switched - off , negative y - components of the electron 1 momenta at the time of recollision give rise to y - components of the electron 1 final momenta that are slightly shifted towards positive values , see fig .
[ figure6](d1 ) . as a result the total @xmath92 in fig .
[ figure6](e1 ) adds up to non - zero values .
it is found that switching - off the magnetic field and the coulomb forces following recollision results in a very similar distribution @xmath92 .
this supports the statement that the initial transverse momentum of electron 1 that results from the combined effect of the recollision and the magnetic field up to the time of recollision solely account for the features of @xmath92 in fig .
[ figure6](e1 ) . a similar analysis at 3.8@xmath3910@xmath7 w@xmath8 is shown in fig . [ figure6](a3)(b3)(c3)(d3)(e3 ) . the values of @xmath92 in fig .
[ figure6](e3 ) at 3.8@xmath3910@xmath7 w@xmath8 are found to be less than the values of @xmath92 in fig .
[ figure6](e1 ) at 2@xmath3910@xmath7 w@xmath8 .
[ ht ] for the two escaping electrons is plotted as a function of @xmath18 .
three intensities are considered and @xmath18 is binned in intervals of 18@xmath60 .
@xmath67 denotes a multiplication factor of @xmath68 .
, title="fig:",scaledwidth=20.0% ] experimentally electron 1 can not be distinguished from electron 2 .
therefore , the probability distributions of electrons 1 and 2 to escape with an angle @xmath18 , p@xmath73 ( fig .
[ figure4](a ) ) and p@xmath72 ( fig .
[ figure4](c ) ) , respectively , are not experimentally accessible .
however , in a kinematically complete experiment , for each double - ionized event , the angle @xmath18 of each ionizing electron can be measured .
then , the probability distribution for any one of the two electrons to ionize with an angle @xmath18 , @xmath116 , can be measured . the distribution @xmath117 is computed and plotted as a function of @xmath18 in fig . [ figure9 ] at 1.3@xmath3910@xmath7 w@xmath8 , 2@xmath3910@xmath7 w@xmath8 and 3.8@xmath3910@xmath7 w@xmath8 . @xmath117 is found to have significant values at smaller intensities over the same wide range of @xmath18s as @xmath92 does .
however , @xmath118 at the smaller intensities .
moreover , at 3.8@xmath3910@xmath7 w@xmath8 , @xmath119 at @xmath120 but has non zero values for a wider range of @xmath18s compared to @xmath121 .
these features are expected since @xmath117 accounts for both the tunneling and the initially bound electron .
the features of @xmath117 still demonstrate that at smaller intensities _ non - dipole recollision - gated ionization _ is in play even though this is more clearly demonstrated by the features of @xmath92 .
in double ionization , when non - dipole effects are accounted for , it was shown that the combined effect of the recollision and the magnetic field result in a distribution of the initial y - component momentum of the tunneling electron which is shifted towards negative values ; negative values in the y - axis are opposite to the propagation direction of the laser field . the term _ non - dipole recollision - gated ionization _ was adopted to describe this effect .
it was shown that the stronger the recollision is the more pronounced the _ non - dipole recollision - gated ionization _ is resulting in a larger number of double - ionized events with negative y - component of electron 1 initial momentum .
the asymmetry in the distribution of the transverse electron 1 momentum at the tunnel - ionization time was shown to give rise to an asymmetry , with respect to @xmath74 , of the probability distribution of electron 1 to escape with an angle @xmath18 .
this asymmetry is found to be significant over a wider range of @xmath18s in double ionization compared to single ionization .
it is this asymmetry over a wide range of @xmath18s that accounts for the large values of the average transverse electron 1 momentum at intensities where strong recollisions prevail .
even though not as pronounced , these features of the probability distribution of the tunneling electron are present in an experimentally accessible observable , namely , the probability distribution for electron 1 or 2 to escape with an angle @xmath18 .
this observable effect of the _ non - dipole recollision - gated ionization _ can be measured by future experiments .
a.e . acknowledges the epsrc grant no .
j0171831 and the use of the computational resources of legion at ucl .
d. b. miloevi ' c , g. g. paulus , d. bauer , and w. becker , j. phys .
b * 39 * , r203 ( 2006 )
. h. r. reiss , phys .
* 101 * , 043002 ( 2008 ) .
h. r. reiss , j. phys .
b. * 47 * , 204006 ( 2014 ) .
p. b. corkum , phys .
lett . * 71 * , 1994 ( 1993 ) .
s. palaniyappan , a. dichiara , e. chowdhury , a. falkowski , g. ongadi , e. l. huskins , and b. c. walker , phys .
* 94 * , 243003 ( 2005 ) .
emelin ad m. yu .
ryabikin , phys .
rev . a * 89 * , 013418 ( 2014 ) . c. c. chiril@xmath122 , n. j. kylstra , and r. m. potvliege and c. j. joachain , phys .
a * 66 * , 063411 ( 2002 ) .
m. w. walser , c. h. keitel , a. scrinzi , and t. brabec , phys .
lett . * 85 * , 5082 ( 2000 ) . c. h. keitel and p. l. knight , phys .
rev . a * 51 * , 1420 ( 1995 ) .
a. ludwig , j. maurer , b. w. mayer , c. r. phillips , l. gallmann , and u. keller , phys .
* 113 * , 243001 ( 2014 ) .
b. wolter , m. g. pullen , m. baudisch , m. sclafani , m. hemmer , a. senftleben , c. d. schr " oter , j. ullrich , r. moshammer , and j. biegert , phys .
x * 5 * , 021034 ( 2015 ) .
s. chelkowski , a. d. bandrauk , and p. b. corkum , phys .
lett . * 113 * , 263005 ( 2014 ) .
s. chelkowski , a. d. bandrauk , and p. b. corkum , phys .
a * 92 * , 051401 ( r ) ( 2015 ) .
a. s. titi and g. w. f. drake , phys .
a * 85 * , 041404(r ) ( 2012 ) .
i. a. ivanov , phys .
rev . a * 91 * 043410 ( 2015 ) . c. t. l. smeenk , l. arissian , b. zhou , a. mysyrowicz , d. m. villeneuve , a. staudte , and p. b. corkum , phys .
lett . * 106 * , 193002 ( 2011 ) .
a. emmanouilidou and t. meltzer , phys .
a * 95 * , 033405 ( 2017 ) .
a. emmanouilidou , phys .
a * 78 * , 023411 ( 2008 ) .
a. emmanouilidou , j. s. parker , l. r. moore , and k. t. taylor , new j. phys .
* 13 * , 043001 ( 2011 ) .
p. kustaanheimo and e. stiefel , j. reine angew . math . * 218 * , 204 ( 1965 ) .
l. d. landau and e. m. lifshitz , _ quantum mechanics _ ( pergamon press , new york , 1977 ) . n. b. delone and v. p. krainov , j. opt .
b * 8 * , 1207 ( 1991 ) .
r. abrines and i. c. percival , proc .
london * 88 * , 861 ( 1966 ) .
a. emmanouilidou and a. staudte , phys .
a * 80 * , 053415 ( 2009 ) . | using a three - dimensional semiclassical model , double ionization for strongly - driven he is studied fully accounting for magnetic field effects .
it was previously found that the average sum of the components of the electron momenta parallel to the propagation direction of the laser field are unexpectedly large at intensities that are smaller than the intensities predicted for magnetic - field effects to arise .
the mechanism responsible for this large sum of the electron momenta is identified .
specifically , it is shown that at these smaller intensities strong recollisions and the magnetic field together act as a gate .
this gate favors more initial tunneling - electron momenta that are opposite to the propagation direction of the laser field .
in contrast , in the absence of non - dipole effects , the initial transverse with respect to the electric field tunneling - electron momentum is symmetric with respect to zero .
this asymmetry in the initial transverse tunneling - electron momentum is shown to give rise to an asymmetry in a double ionization observable . |
a boundary between quantum and classical worlds is rather difficult to draw precisely , while the problem of their distinguishing is getting more important than it was hardly ever before .
this problem is particularly important for the large - scale quantum computer to be constructed because it must be a macroscopic object and exhibit quantum properties at the same time .
the investigations in this field of science are also stimulated by the attempts to describe quantum mechanical phenomena by different kinds of hidden variables models . though these problems have been paid a great attention for many years the common point of view has not been achieved .
the discussions came up with a bang after a recent proposal @xcite of a simple test of checking whether it is possible or not to describe a given set of experimental data by a classical probabilistic model . in quantum mechanics
the state of a system can be identified with a fair probability called tomographic probability distribution or state tomogram ( see , e.g. , the review @xcite ) .
the probability representation of quantum states with continuous variables ( position , momentum ) was introduced in @xcite and that of states with discrete variables ( spin , qudits ) was introduced in @xcite . in the probability representation of quantum mechanics the relation between classical and quantum systems behavior can be studied using the same notion of states expressed in terms of the probability distribution ( tomogram ) in both classical and quantum domains @xcite .
the quantum ( or classical ) nature of a system state can be clarified if in the system the uncertainty relations @xcite for conjugate variables are fulfilled .
also , if the state is quantum , it can be similar to a classical one and there are some studies of the classicality ( see , e.g. , @xcite ) based on properties of the diagonal representation of density operator @xcite ( or _
p_-function @xcite ) . in the works @xcite , quantum or classical properties of a system state were associated with measuring some specific observables such that there exist certain inequalities which hold true in classical domain and are violated in quantum domain .
violation of the inequalities is considered as a quantumness witness of the system state . in this sense ,
the criterion @xcite is similar in its spirit to the bell inequalities @xcite .
the bell inequalities were studied by means of the tomographic probability representation in @xcite .
the aim of our work is to consider the inequalities introduced in @xcite and their properties in classical and quantum domains within the framework of the probability representation of quantum states .
we suppose that such a procedure is necessary while dealing with the quantum probabilistic model based on tomograms .
the paper is organized as follows . in sec .
[ spin - tomogram ] , we are aimed at recalling the tomographic representation of qubit states and observables by employing both ordinary and dual tomographic symbols . in sec .
[ section - quantumness - tests ] , the quantumness test is discussed within the framework of classical and quantum probability descriptions . in sec .
[ example ] , we present a family of observables which can be used to detect quantumness of an arbitrary generally mixed state @xmath0 of a single system . here
we also predict what kind of experiment one should make to test the quantumness of a state specified by its tomogram . in sec .
[ conclusions ] , conclusions and prospects are presented .
apart from being described by the conventional density operator @xmath1 , the state of a qubit is determined thoroughly by its spin tomogram .
the probability distribution function ( spin tomogram ) @xmath2 is nothing else but the probability to obtain @xmath3 ( @xmath4 ) as spin projection on the direction given by the unitary @xmath5 matrix @xmath6 .
this matrix can be considered as a matrix of an irreducible representation of the rotation group depending on two euler angles determining the direction of quantization ( point on the bloch sphere ) .
the relation between @xmath1 and @xmath2 reads @xmath7 where the operator @xmath8 is called the dequantizer operator and @xmath9 is assigned to the set of parameters @xmath10 .
the general procedure to use dequantizer operators was discussed in the context of star - product quantization schiemes in @xcite .
the explicit form of the dequantizer @xmath11 can be obtained readily by exploiting the matrix @xmath6 expressed in terms of the euler angles @xmath12 , @xmath13 , @xmath14 : @xmath15 therefore , taking advantage of @xmath16 , where @xmath17 is the @xmath18 identity matrix and @xmath19 is the third pauli matrix , one can write @xmath20 where @xmath21 and the matrix @xmath22 has the following form @xmath23 if given the spin tomogram @xmath24 it is possible to reconstruct the density operator @xmath1 @xcite .
this reconstruction was shown to take the simple form @xmath25 where the integration implies @xmath26 and the quantizer operator @xmath27 is defined by the formula @xmath28 in quantum mechanics any observable @xmath29 is identified with a hermitian operator @xmath30 . by analogy with the density operator @xmath1 one can introduce the tomographic symbol @xmath31 of the operator @xmath30 . just in the same way we write @xmath32
it is worth noting that both quantizer and dequantizer are operators depending on the set of parameters @xmath9 so it seems possible to swap quantizer with the dequantizer .
substituting the quantizer operator for the dequantizer one and visa versa leads to a so - called dual tomographic symbol @xmath33 @xcite satisfying the following relations : @xmath34 the dual symbol in the tomographic
probability representation turned out to provide the function introduced in @xcite which after averaging with a tomogram yields the mean value of the observable .
let us now express the average value of the observable @xmath29 by means of ordinary and dual tomographic symbols .
indeed , the mean value of @xmath29 reads @xmath35 the formula obtained can be checked immediately for a general case of the density operator @xmath1 and the observable @xmath29 : @xmath36 where @xmath37 , @xmath38 , @xmath39 , @xmath40 are real numbers .
then the spin tomogram @xmath24 is @xmath41 and the dual tomographic symbol @xmath33 reads @xmath42.\ ] ] the direct calculation yields @xmath43 that coincides totally with the quantity @xmath44 computed by using the matrix form of operators ( [ matrix - form - rho - a ] ) .
let us now express the average value of the observable @xmath29 in terms of its possible outcomes @xmath45 and @xmath46 measured through separate experiments .
the numbers @xmath45 and @xmath46 are nothing else but eigenvalues of the operator @xmath30 . consequently there exists a unitary matrix @xmath47 such that the matrix @xmath30 can be factorized as follows @xmath48 the matrix @xmath49 is composed of two columns which are eigenvectors of the operator @xmath30 corresponding to eigenvalues @xmath45 and @xmath46 , respectively . in order to specify the matrix
@xmath47 one can substitute the euler angles @xmath50 for @xmath51 in formula ( [ u - matrix ] ) , i.e. , @xmath52 .
then the dual tomographic symbol of the operator @xmath30 takes the form @xmath53 \bigg\}.\ ] ] now , when @xmath33 is known , it is not impossible to evaluate the integral @xmath54 \nonumber \\ & & \qquad + a_{\downarrow } \left [ \textstyle{1 \over 2 } - \textstyle{1 \over 2}(\rho_{11}-\rho_{22})\cos\theta + \rho_{12}\sin\theta \cos(\zeta-\phi ) \right ] \nonumber \\ & & = w(+\textstyle{1 \over 2},\phi,\theta,\varphi)a_{\uparrow } + w(-\textstyle{1 \over 2},\phi,\theta,\varphi)a_{\downarrow } \nonumber \\ & & = w_{\uparrow}(u_{a})a_{\uparrow } + w_{\downarrow}(u_{a})a_{\downarrow},\end{aligned}\ ] ] that gives the average value of the observable @xmath29 , i.e. , the value of quantity @xmath55 .
here we denoted @xmath56 and @xmath57 by @xmath58 and @xmath59 , respectively .
one can not help mentioning that the same result is achieved by using the definition of the spin tomogram ( [ tomogram - definition ] ) : @xmath60 = { \rm tr}\left[u_{a}\hat{\rho}u_{a}^{\dag}\left(% \begin{array}{cc } a_{\uparrow } & 0 \\ 0 & a_{\downarrow } \\
\end{array}% \right)\right ] \nonumber \\ & & = { \rm tr}\left [ \left(% \begin{array}{cc } \langle \textstyle{1 \over 2 } | u_{a}\hat{\rho}u_{a}^{\dag } | \textstyle{1 \over 2 } \rangle & \langle \textstyle{1 \over 2 } | u_{a}\hat{\rho}u_{a}^{\dag } | -\textstyle{1 \over 2 } \rangle \\
\langle -\textstyle{1 \over 2 } | u_{a}\hat{\rho}u_{a}^{\dag } | \textstyle{1 \over 2 } \rangle & \langle -\textstyle{1 \over 2 } | u_{a}\hat{\rho}u_{a}^{\dag } | -\textstyle{1 \over 2 } \rangle \\ \end{array}% \right ) \left(% \begin{array}{cc } a_{\uparrow } & 0 \\ 0 & a_{\downarrow } \\ \end{array}% \right)\right ] \nonumber \\ & & = \langle \textstyle{1 \over 2 } | u_{a}\hat{\rho}u_{a}^{\dag } -\textstyle{1 \over 2 } | u_{a}\hat{\rho}u_{a}^{\dag } | -\textstyle{1 \over 2 } \rangle a_{\downarrow } \nonumber \\ & & = w_{\uparrow}(u_{a})a_{\uparrow } + w_{\downarrow}(u_{a})a_{\downarrow}.\end{aligned}\ ] ]
in this section , we are going to answer the question whether it is possible or not to describe the system involved by a classical probabilistic model .
the negative reply indicates straightway the quantumness of the system in question . to start , let us consider the case of a single qubit , and then discuss a generalization for an arbitrary qudit system .
in the classical statistical model of a single qubit system the observable @xmath29 is associated with a set of real numbers @xmath61 , where the numbers @xmath45 and @xmath46 are nothing else but possible outcomes of the measurement of @xmath29 . moreover ,
the observable @xmath62 has possible outcomes of the form @xmath63 , i.e. , is in strong relation to @xmath29 .
the states form a simplex of probability distributions @xmath64 , with @xmath65 , @xmath66 , and @xmath67 ( such a simplex is a geometrical treatment of all possible classical states ; geometric interpretation of quantum states is reviewed in @xcite ) .
then the expectation values of observables @xmath29 and @xmath62 read @xmath68 it is worth noting that such a classical system is equivalent to the quantum one , with both the density operator and the operator @xmath69 being of the diagonal form @xmath70 suppose one has two nonnegative observables @xmath29 and @xmath71 such that inequality @xmath72 holds true for all states @xmath64 . that implies the following relations : @xmath73 and @xmath74 .
if this is the case , the average value of observable @xmath62 is necessarily less or equal than the average value of @xmath75 . indeed ,
@xmath76 the mathematical aspect of the classical statistical model is briefly expressed by the implication @xmath77 \le 0 \qquad \rightarrow \qquad { \rm tr } \left [ \left(% \begin{array}{cc } p_{\uparrow } & p_{\uparrow } \\ p_{\downarrow } & p_{\downarrow } \\ \end{array}% \right ) \left(% \begin{array}{cc } a_{\uparrow}^{2 } & a_{\downarrow}^{2 } \\ -b_{\uparrow}^{2 } & -b_{\downarrow}^{2 }
\\ \end{array}% \right ) \right ] \le 0,\ ] ] where we introduced a stochastic matrix with the matrix elements @xmath78 and @xmath79 . in the quantum statistical model the observable @xmath29
is associated with the operator @xmath30 , which in its turn corresponds to the hermitian matrix whose eigenvalues @xmath45 and @xmath46 give possible outcomes of the measurement of @xmath29 .
states are identified with density operators @xmath1 , which are positive hermitian matrices with the trace equal to unity . by using the tomographic representation of quantum states and observables developed in previous section , one can write the average values of observables @xmath29 and @xmath62 as follows : @xmath80 in a way analogues to the classical case , we consider two nonnegative operators @xmath30 and @xmath81 ( i.e. , having nonnegative eigenvalues @xmath82 ) such that the residual operator @xmath83 is nonnegative as well .
the last requirement can be rewritten in the form of the condition @xmath84 \le 0\ ] ] that must be valid for _ all _ tomograms @xmath2 . in ( [ quantum - first - moment ] ) , we introduced a stochastic matrix with the matrix elements @xmath85 and @xmath86 .
in contrast to the classical case , even under these circumstances the average value of the operator @xmath87 does not have to be nonnegative .
therefore , there can exist the tomogram @xmath88 such that @xmath89 > 0.\ ] ] the difference between classical and quantum behaviour of the observable @xmath90 serves as the basis for a simple quantumness test .
* quantumness test*. two observables @xmath29 and @xmath71 are found which averaged values satisfy the inequality @xmath91 for all experimentally accessible states , but for a certain state the condition @xmath92 is violated , then the state involved can not be described by a classical probabilistic model and is surely to be quantum .
this is the reason why the operator @xmath87 was proposed to call quantumness witness @xcite .
the comparison of the expressions ( [ classical - inequalities ] ) and ( [ violation ] ) shows that despite both classical and quantum cases can be treated probabilistically the difference between them occurs due to the classical state being associated with numbers @xmath93 while the quantum one being identified with the functions @xmath58 and @xmath59 taking on the different values at the unitary group elements @xmath94 and @xmath95 .
the generalization of this approach for a qudit state given by its @xmath96 density matrix @xmath0 is evident while dealing with operators @xmath30 , @xmath81 having an effect on the two - dimensional subspace spanned by eigenvectors corresponding to two different eigenvalues of @xmath1 .
the quantumness test described above requires a certain quantumness witness to be found . in the earlier work @xcite , the abundance of quantumness witnesses has been demonstrated for qubit states @xmath97 . in spite of that
, the proposed scheme does not give the explicit shape of the operators @xmath30 and @xmath81 .
in this section , we are going to repair this gap .
first of all , the state @xmath98 can be treated as classical , because its tomogram @xmath99 and coincides with the case of classical probabilities @xmath100 .
let us demonstrate that all the other qubit states are quantum .
indeed , an arbitrary density operator @xmath1 can be reduced to the diagonal form of @xmath101 that becomes maximally mixed ( classical ) if @xmath102 and pure if @xmath103 . in order to find the quantumness witness for all @xmath104
we avoid limitations @xmath105 or @xmath106 imposed in @xcite and @xcite , respectively .
below we introduce a family of the operators @xmath30 and @xmath81 that satisfy the requirements @xmath107 while @xmath87 is a quantumness witness for a particular region of parameters @xmath108 , @xmath109 , @xmath110 . for the diagonal density matrix ( [ rho - diag ] )
one can use @xmath111 the operators @xmath112 , @xmath113 , @xmath114 have nonnegative eigenvalues only while the average value @xmath115 $ ] is negative for all @xmath116 of the range @xmath104 .
figure [ graph-1 ] gives the illustration of this fact for particular values of parameters @xmath108 , @xmath109 , @xmath110 .
the average value @xmath115 $ ] ( expectation value of quantumness witness ) is negative for all @xmath116 of the range @xmath117 ( @xmath118 , @xmath119 , and the values of @xmath110 are depicted in the figure ) . ]
all - tomographic statement of the quantumness test can be formulated as follows .
suppose we have an experimentally measured spin tomogram @xmath2 of a qubit state .
the parameter @xmath116 of the diagonal density matrix @xmath120 can be found as @xmath121 then for the operators @xmath30 and @xmath81 leading to the quantumness witness one can use @xmath122
to conclude we summarize the main results of the paper .
we demonstrated that the criterion of quantumness of a system state found in @xcite can be formulated within the framework of the probability representation of quantum states . in this representation ,
the structure of the criterion is clarified since it is formulated by means of inequalities , where stochastic matrices involved ( see ( [ classical - inequalities ] ) and ( [ violation ] ) ) have different properties . in case of classical states
the stochastic matrix providing inequality has constant matrix elements .
conversely , in case of quantum states the stochastic matrix providing analogous inequality has matrix elements depending on unitary group elements . in view of this , for some values of the unitary group elements the inequality can be violated .
one can extend the analysis of quantumness tests given for qubits to systems with continuous variables , e.g. , for photon quadrature components . | in view of the tomographic probability representation of quantum states , we reconsider the approach to quantumness tests of a single system developed in [ alicki and van ryn 2008 _ j. phys . a : math .
theor . _ * 41 * 062001 ] . for qubits
we introduce a general family of quantumness witnesses which are operators depending on an extra parameter .
spin tomogram and dual spin tomographic symbols are used to study qubit examples and the test inequalities which are shown to satisfy simple relations within the framework of the standard probability theory . |
entangled states are an essential resource for various quantum information processings@xcite . hence , it is required to generate maximally entangled states .
however , for a practical use , it is more essential to guarantee the quality of generated entangled states .
statistical hypothesis testing is a standard method for guaranteeing the quality of industrial products .
therefore , it is much needed to establish the method for statistical testing of maximally entangled states .
quantum state estimation and quantum state tomography are known as the method of identifying the unknown state@xcite .
quantum state tomography @xcite has been recently applied to obtain full information of the @xmath0 density matrix . however , if the purpose is testing of entanglement , it is more economical to concentrate on checking the degree of entanglement .
such a study has been done by tsuda et al @xcite as optimization problems of povm .
however , an implemented quantum measurement can not be regarded as an application of a povm to a single particle system or a multiple application of a povm to single particle systems . in particular , in quantum optics , the following measurement is often realized , which is not described by a povm on a single particle system .
the number of generated particles is probabilistic .
we prepare a filter corresponding to a projection @xmath1 , and detect the number of particle passing through the filter .
if the number of generated particles obeys a poisson distribution , as is mentioned in section [ s2 ] , the number of detected particles obeys another poisson distribution whose average is given by the density and the projection @xmath1 . in this kind of measurements , if any particle is not detected , we can not decide whether a particle is not generated or it is generated but does not pass through the filter .
if we can detect the number of generated particles as well as the number of passing particles , the measurement can be regarded as the multiple application of the povm @xmath2 . in this case , the number of applications of the povm is the variable corresponding to the number of generated particles .
also , we only can detect the empirical distribution .
hence , our obtained information almost discuss by use of the povm @xmath2 . however , if it is impossible to distinguish the two events by some imperfections , it is impossible to reduce the analysis of our obtained information to the analysis of povms .
hence , it is needed to analyze the performance of the estimation and/or the hypothesis testing based on the poisson distribution describing the number of detected particles .
if we discuss the ultimate bound of the accuracy of the estimation and/or the hypothesis testing , we do not have to treat such imperfect measurements .
since several realistic measurements have such imperfections , it is very important to optimize our measurement among such a class of imperfect measurements . in this paper , our measurement
is restricted to the detection of the number of the particle passing through the filter corresponding to a projection @xmath1 .
we apply this formulation to the testing of maximally entangled states on two qubit systems ( two - level systems ) , each of which is spanned by two vectors @xmath3 and @xmath4 .
since the target system is a bipartite system , it is natural to restrict to our measurement to local operations and classical communications ( locc ) . in this paper , for a simple realization , we restrict our measurements to the number of the simultaneous detections at the both parties of the particles passing through the respective filters . we also restrict the total measurement time @xmath5 , and optimize the allocation of the time for each filters at the both parties . as our results , we obtain the following characterizations .
if the average number of the generated particles is known , our choice is counting the coincidence events or the anti - coincidence events .
when the true state is close to the target maximally entangled state @xmath6 ( that is , the fidelity between these is greater than @xmath7 ) , the detection of anti - coincidence events is better than that of coincidence events .
this result implies that the indistinguishability between the coincidence events and the non - generation event loses less information than that between the anti - coincidence events and the non - generation event .
this fact also holds even if we treat this problem taking into account the effect of dark counts . in this discussion , in order to remove the bias concerning the direction of the difference , we assume the equal time allocation among the vectors @xmath8 , which corresponds to the anti - coincidence events , and that among the vectors @xmath9 , which corresponds to the coincidence events , where @xmath10 , @xmath11 , @xmath12 , @xmath13 .
indeed , barbieri et al @xcite proposed to detect the anti - coincidence events for measuring an entanglement witness , they did not prove the superiority of detecting the anti - coincidence events in the framework of mathematical statistics .
however , the average number of the generated particles is usually unknown . in this case
, we can not estimate how close the true state is to the target maximally entangled state from the detection of anti - coincidence events .
hence , we need to count the coincidence events as additional information . in order to resolve this problem
, we usually use the equal allocation between anti - coincidence events and coincidence events in the visibility method , which is a conventional method for checking the entanglement .
however , since we measure the coincidence events and the anti - coincidence events based on one or two bases in this method , there is a bias concerning the direction of the difference . in order to remove this bias ,
we consider the detecting method with the equal time allocation among all vectors @xmath8 and @xmath9 , and call it the modified visibility method . in this paper
, we also examine the detection of the total flux , which can be realized by detecting the particle without the filter .
we optimize the time allocation among these three detections .
we found that the optimal time allocation depends on the fidelity between the true state and the target maximally entangled state .
if our purpose is estimating the fidelity @xmath14 , we can not directly apply the optimal time allocation .
however , the purpose is testing whether the fidelity @xmath14 is greater than the given threshold @xmath15 , the optimal allocation at @xmath15 gives the optimal testing method .
if the fidelity @xmath14 is less than a critical value , the optimal allocation is given by the allocation between the anti - coincidence vectors and the coincidence vectors ( the ratio depends on @xmath14 . ) otherwise , it is given by the allocation only between the anti - coincidence vectors and the total flux .
this fact is valid even if the dark count exists .
if the dark count is greater than a certain value , the optimal time allocation is always given by the allocation between the anti - coincidence vectors and the coincidence vectors .
further , we consider the optimal allocation among anti - coincidence vectors when the average number of generated particles .
the optimal allocation depends on the direction of the difference between the true state and the target state .
since the direction is usually unknown , this optimal allocation dose not seems useful .
however , by adaptively deciding the optimal time allocation , we can apply the optimal time allocation .
we propose to apply this optimal allocation by use of the two - stage method .
further , taking into account the complexity of testing methods and the dark counts , we give a testing procedure of entanglement based on the two - stage method .
in addition , proposed designs of experiments were demonstrated by hayashi et al .
@xcite in two photon pairs generated by spontaneous parametric down conversion ( spdc ) . in this article ,
we reformulate the hypothesis testing to be applicable to the poisson distribution framework , and demonstrate the effectiveness of the optimized time allocation in the entanglement test .
the construction of this article is following .
section [ s2 ] defines the poisson distribution framework and gives the hypothesis scheme for the entanglement .
section [ s3 ] gives the mathematical formulation concerning statistical hypothesis testing .
sections [ s4 ] and [ s5 ] give the fundamental properties of the hypothesis testing : section [ s4 ] introduces the likelihood ratio test and its modification , and section [ s5 ] gives the asymptotic theory of the hypothesis testing .
sections [ s6]-[s9 ] are devoted to the designs of the time allocation between the coincidence and anti - coincidence bases : section [ s6 ] defines the modified visibility method , section [ s7 ] optimize the time allocation , when the total photon flux @xmath16 is unknown , section [ s8 ] gives the results with known @xmath16 , and section [ s9 ] compares the designs in terms of the asymptotic variance .
section [ s10 ] gives further improvement by optimizing the time allocation between the anti - coincidence bases .
appendices give the detail of the proofs used in the optimization .
let @xmath17 be the hilbert space of our interest , and @xmath1 be the projection corresponding to our filter . if we assume generation process on each time to be identical but individual , the total number @xmath18 of generated particles during the time @xmath5 obeys the poisson distribution @xmath19 .
hence , when the density of the true state is @xmath20 , the probability of the number @xmath21 of detected particles is given as @xmath22 in fact , if we treat the fock space generated by @xmath17 instead of the single particle system @xmath17 , this measurement can be described by a povm . however , since this povm dooes not have a simple form , it is suitable to treat this measurement in the form ( [ 5 - 8 - 1 ] ) .
further , if we errorly detect the @xmath23 particles with the probability @xmath24 , the probability of the number @xmath21 of detected particles is equal to @xmath25 this kind of incorrect detection is called dark count .
further , since we consider the bipartite case , i.e. , the case where @xmath26 , we assume that our projection @xmath1 has the separable form @xmath27 . in this paper , under the above assumption
, we discuss the hypothesis testing when the target state is the maximally entangled @xmath28 state while usami et al.@xcite discussed the state estimation under this assumption . here
we measure the degree of entanglement by the fidelity between the generated state and the target state : @xmath29 the purpose of the test is to guarantee that the state is sufficiently close to the maximally entangled state with a certain significance .
that is , we are required to disprove that the fidelity @xmath14 is less than a threshold @xmath15 with a small error probability . in mathematical statistics , this situation is formulated as hypothesis testing ; we introduce the null hypothesis @xmath30 that entanglement is not enough and the alternative @xmath31 that the entanglement is enough : @xmath32 with a threshold @xmath15 .
visibility is an indicator of entanglement commonly used in the experiments , and is calculated as follows : first , a s measurement vector @xmath33 is fixed , then the measurement @xmath34 is performed by rotating b s measurement vector @xmath35 to obtain the maximum and minimum number of the counts , @xmath36 and @xmath37 .
we need to make the measurement with at least two bases of a in order to exclude the possibility of the classical correlation .
we may choose the two bases @xmath38 and @xmath39 as @xmath33 , for example .
finally , the visibility is given by the ratio between @xmath40 and @xmath41 with the respective a s measurement basis @xmath33 .
however , our decision will contain a bias , if we choose only two bases as a s measurement basis @xmath33 .
hence , we can not estimate the fidelity between the target maximally entangled state and the given state in a statistically proper way from the visibility .
since the equation @xmath42 holds , we can estimate the fidelity by measuring the sum of the counts of the following vectors : @xmath43 obeys the poisson distribution with the expectation value @xmath44 , where the measurement time for each vector is @xmath45 .
we call these vectors the coincidence vectors because these correspond to the coincidence events .
however , since the parameter @xmath16 is usually unknown , we need to perform another measurement on different vectors to obtain additional information .
since @xmath46 also holds , we can estimate the fidelity by measuring the sum of the counts of the following vectors : @xmath47 obeys the poisson distribution @xmath48 , where the measurement time for each vector is @xmath49 . combining the two measurements
, we can estimate the fidelity without the knowledge of @xmath16 .
we call these vectors the anti - coincidence vectors because these correspond to the anti - coincidence events
. we can also consider different type of measurement on @xmath16 .
if we prepare our device to detect all photons , i.e. , the case where the projection is @xmath50 , the detected number @xmath51 obeys the distribution @xmath52 ) with the measurement time @xmath53 . we will refer to it as the total flux measurement . in the following ,
we consider the best time allocation for estimation and test on the fidelity , by applying methods of mathematical statistics .
we will assume that @xmath16 is known or estimated from the detected number @xmath51 .
in this section , we review the fundamental knowledge of hypothesis testing for probability distributions@xcite .
suppose that a random variable @xmath54 is distributed according to a probability measure @xmath55 identified by the unknown parameter @xmath56 .
we also assume that the unknown parameter @xmath56 belongs to one of mutually disjoint sets @xmath57 and @xmath58 .
when we want to guarantee that the true parameter @xmath56 belongs to the set @xmath58 with a certain significance , we choose the null hypothesis @xmath30 and the alternative hypothesis @xmath31 as @xmath59 then , our decision method is described by a test , which is described as a function @xmath60 taking values in @xmath61 ; @xmath30 is rejected if @xmath62 is observed , and @xmath30 is not rejected if @xmath63 is observed .
that is , we make our decision only when @xmath62 is observed , and do not otherwise .
this is because the purpose is accepting @xmath31 by rejecting @xmath30 with guaranteeing the quality of our decision , and is not rejecting @xmath31 nor accepting @xmath31 .
therefore , we call the region @xmath64 the rejection region . the test @xmath65 can be defined by the rejection region .
in fact , we choosed the hypothesis that the fidelity is less than the given threshold @xmath66 as the null hypothesis @xmath30 in section [ s2 ] .
this formulation is natural because our purpose is guaranteeing that the fidelity is not less than the given threshold @xmath66 . from theoretical viewpoint ,
we often consider randomized tests , in which we probabilistically make the decision for a given data .
such a test is given by a function @xmath65 mapping to the interval @xmath67 $ ] .
when we observe the data @xmath68 , @xmath30 is rejected with the probability @xmath60 . in the following , we treat randomized tests as well as deterministic tests . in the statistical hypothesis testing , we minimize error probabilities of the test @xmath65 .
there are two types of errors .
the type one error is the case where @xmath30 is rejected though it is true .
the type two error is the converse case , @xmath30 is accepted though it is false .
hence , the type one error probability is given @xmath69 @xmath70 , and the type two error probability is given @xmath71 @xmath72 , where @xmath73 it is in general impossible to minimize both @xmath69 and @xmath71 simultaneously because of a trade - off relation between them . since we make our decision with guaranteeing its quality only when @xmath62 is observed , it is definitively required that the type one error probability @xmath69 is less than a certain constant @xmath74 .
for this reason , we minimize the type two error probability @xmath71 under the condition @xmath75 .
the constant @xmath74 in the condition is called the risk probability , which guarantees the quality of our decision .
if the risk probability is large enough , our decision has less reliability . under this constraint for the risk probability ,
we maximize the probability to reject the hypothesis @xmath30 when the true parameter is @xmath76 .
this probability is given as @xmath69 , and is called the power of @xmath65 .
hence , a test @xmath65 of the risk probability @xmath74 is said to be most powerful ( mp ) at @xmath77 if @xmath78 holds for any test @xmath79 of the risk probability @xmath74 .
then , a test is said to be uniformly most powerful ( ump ) if it is mp at any @xmath77 . in the hypothesis testing , we usually fixed our test before applying it to data .
however , we sometimes focus on the minimum risk probability among tests in a class @xmath80 rejecting the hypothesis @xmath30 with a given data .
this value is called the p - value , which depends on the observed data @xmath68 as well as the subset @xmath57 to be rejected .
in fact , in order to define the p - value , we have to fix a class @xmath81 of tests .
then , for @xmath68 and @xmath57 , p - value is defined as @xmath82 since the p - value expresses the risk for rejecting the hypothesis @xmath30 , hence , this concept is useful for comparison among several designs of experiment .
note that if we are allowed to choose any function @xmath65 as a test , the above minimum is attained by the function @xmath83 : @xmath84 in this case , the p - vale is @xmath85 .
however , the function @xmath83 is unnatural as a test .
hence , we should fix a class of tests to define p - value .
in mathematical statistics , the likelihood ratio tests is often used as a class of standard tests@xcite .
this kind of tests often provide the ump test in some typical cases . when both @xmath57 and @xmath58 consist of single elements as @xmath86 and @xmath87 , the likelihood ratio test @xmath88 is defined as @xmath89 where @xmath90 is a constant , and the ratio @xmath91 is called the likelihood ratio . from the definition ,
any test @xmath65 satisfies @xmath92 when a likelihood ratio test @xmath88 satisfies @xmath93 the test @xmath88 is mp of level @xmath74 .
indeed , when a test @xmath65 satisfies @xmath94 , @xmath95 hence , @xmath96 .
this is known as neyman - pearson s fundamental lemma[multiblock footnote omitted ] .
the likelihood ratio test is generalized to the cases where @xmath57 or @xmath58 has at least two elements as @xmath97 usually , in order to guarantee a small risk probability , the likelihood ratio @xmath90 is choosed as @xmath98 . in cases
where the hypothesis is one - sided , that is , the parameter space @xmath99 is an interval of @xmath100 and the hypothesis is given as @xmath101 we often use so - called interval tests for its optimality under some conditions as well as for its naturalness .
when the likelihood ratio @xmath102 is monotone increasing concerning @xmath68 for any @xmath103 such that @xmath104 , the likelihood ratio is called monotone . in this case , the likelihood ratio test @xmath88 between @xmath105 and @xmath106 is ump of level @xmath107 , where @xmath108 is an arbitrary element satisfying @xmath109 .
indeed , many important examples satisfy this condition .
hence , it is convenient to give its proof here . from the monotonicity , the likelihood ratio test @xmath88 has the form @xmath110 with a threshold value @xmath111 .
since the monotonicity implies @xmath112 for any @xmath113 , it follows from neyman pearson lemma that the likelihood ratio test @xmath88 is mp of level @xmath74 . from ( [ 2 - 14 - 1 ] ) , the likelihood ratio test @xmath88 is also a likelihood ratio test between @xmath105 and @xmath114 , where @xmath115 is another element satisfying @xmath116 .
hence , the test @xmath88 is also mp of level @xmath74 . from the above discussion
, it is suitable to treat p - value based on the class of likelihood ratio tests . in this case , when we observe @xmath111 , the p - value is equal to @xmath117 in mathematical statistics , exponential families are known as a class of typical statistical models@xcite .
a family of probability distributions @xmath118 is called an exponential family when there exists a random variable @xmath68 such that @xmath119 where @xmath120 .
it is known that this class of families includes , for example , the poisson distributions , normal distributions , binomial distributions , etc . in this case , the likelihood ratio @xmath121 is monotone concerning @xmath68 for @xmath122 .
hence , the likelihood ratio test is ump in the hypothesis ( [ eq : hypo - int ] ) .
note that this argument is valid even if we choose a different parameter if the family has a parameter satisfying ( [ 2 - 17 - 3 ] ) .
for example , in the case of the normal distribution @xmath123 , the ump test @xmath124 of the level @xmath74 is given as @xmath125 where @xmath126 the @xmath18-trial binomial distributions @xmath127 are also an exponential family because another parameter @xmath128 satisfies that @xmath129 .
hence , in the case of the @xmath18-trial binomial distribution , the ump test @xmath130 of the level @xmath74 is given as the randomized likelihood ratio test : @xmath131 where @xmath132 is the maximum value @xmath23 satisfying @xmath133 , and @xmath134 is defined as @xmath135 therefore , when @xmath21 is observed , the p - value is @xmath136 .
when @xmath18 is sufficiently large , the distribution @xmath137 can be approximated by the normal distribution with variance @xmath138 .
hence , the ump test @xmath130 of the level @xmath74 is approximately given as @xmath139 the p - value is also approximated to @xmath140 the poisson distributions @xmath141 are also an exponential family because another parameter @xmath142 satisfies @xmath143 .
the ump test @xmath124 of the level @xmath74 is characterized similarly to ( [ 5 - 4 - 1 ] ) .
when the threshold @xmath144 is sufficiently large and the hypothesis is given @xmath145 the ump test @xmath124 of the level @xmath74 is approximately given as @xmath146 the p - value is also approximated to @xmath147 next , we consider testing the following hypothesis in the case of the binomial poisson distribution poi(@xmath148 ) : @xmath149 in this case , as is shown at ( [ 5 - 14 - 1 ] ) and ( [ 5 - 14 - 2 ] ) in section [ s4d ] , the likelihood ratio test @xmath88 is characetrized by the likelihood ratio test of the binomial distributions as @xmath150 hence , it is suitable to employ the likelihood ratio test @xmath151 with the level @xmath74 .
this is because the conditional distribution @xmath152 is equal to the binomial distribution @xmath153 .
therefore , when we observe @xmath154 , the p - value of this class of likelihood ratio tests is equal to @xmath155 .
when the total number @xmath156 is sufficiently large , the test @xmath157 of the level @xmath74 is approximately given as @xmath158 the p - value is also approximated to @xmath159 in the one - parameter case , ump tests can be often characterized by likelihood ratiotests . however , in the multi - parameter case , this type characterization is impossible generally , and the ump test does not always exist . in this case , we have to choose our test among non - ump tests .
one idea is choosing our test among likelihood ratio tests because likelihood ratio tests always exist and we can expect that these tests have good performances .
generally , it is not easy to give an explicit form of the likelihood ratio test .
when the family is a multi - parameter exponential family , the likelihood ratio test has a simple form .
a family of probability distributions @xmath160 is called an @xmath161-parameter exponential family when there exists @xmath161-dimensional random variable @xmath162 such that @xmath163 where @xmath164 . however , this form is not sufficiently simple because its rejection region is given by the a nonlinear constraint .
hence , a test with a simpler form is required . in the following ,
we discuss the likelihood ratio test in the case of multi - nomial poisson distribution .
after this discussion , we propose an alternative test . in an @xmath161-parameter exponential family
, the likelihood ratio test @xmath88 has the form @xmath165 where the divergence @xmath166 is defined as @xmath167 and @xmath168 is defined by @xcite @xmath169 this is because the logarithm of the likelihood function is calculated as @xmath170 in addition , @xmath168 coincides with the mle when @xmath171 is observed .
hence , when @xmath172 , the likelihood ratio test with the ratio @xmath173 is given by the rejection region : @xmath174 in the case of the multi - nomial poisson distributions poi@xmath175 , which is an exponential family , the divergence is calculated as @xmath176 where @xmath177 is the divergence between the multinomial distributions @xmath178 and @xmath179 .
when the hypothesis is given by ( [ eq : hypo - int-2 ] ) and @xmath180 , we have @xmath181 where @xmath55 is the binomial distribution with one observation and @xmath182 is the binomial distribution with @xmath18 observations .
then , the likelihood ratio test is given by the likelihood ratio test of the binomial distributions .
in the following , we treat two hypotheses given as @xmath183 with the condition @xmath184 , using the formula ( [ 5 - 14 - 3 ] ) , and ( [ eq : hypo - int-3 ] ) , we can calculate the likelihood ratio test for a given ratio @xmath90 .
now , we calculate the p - value concerning the class of likelihood ratio tests when we observe the data @xmath185 . when @xmath186 , this p - value is equal to @xmath187 where @xmath188 because the minimum @xmath189 satisfying @xmath190 is @xmath191 .
since the calculation of ( [ 3 - 7 - 7 ] ) is not so easy , we consider its upper bound .
for this purpose , we define the set @xmath192 as @xmath193 where @xmath194 are defined as follows : @xmath195 where @xmath196 and @xmath197 .
note that @xmath194 is a monotone decreasing function of @xmath189 .
as is shown in appendix [ 3 - 6 - 10 ] , @xmath198 then , the p - value concerning likelihood ratio tests is upperly bounded by @xmath199 however , it is difficult to choose the likelihood @xmath90 such that the p - value is equal to a given risk probability @xmath74 because the set @xmath200 is defined by a non - linear constraint . in order to resolve this problem , we propose to modify the likelihood ratio test by using the set @xmath192 instead of the set @xmath200 because @xmath192 is defined by a linear constraint while @xmath200 is by a non - linear constraint . that is , we define the modified test @xmath201 as the test with the rejection region @xmath202 . among this kind of tests
, we can choose the test @xmath203 with the risk probability @xmath74 by choosing @xmath204 in the following way : @xmath205 indeed , the calculation of the probability @xmath206 is easier than that of the probability @xmath207 because of the linearity of the constraint condition of @xmath192 .
next , we calculate the p - value of the set of the modified tests @xmath208 . for an observed data @xmath209 ,
we choose @xmath210 as @xmath211 satisfying @xmath212 the lhs is monotone increasing for @xmath211 because each @xmath213 is monotone decreasing for @xmath211 .
thus , @xmath210 is the maximum @xmath211 such that @xmath214 .
then , the p - value is equal to @xmath215 .
further , the relation ( [ 5 - 2 - 1 ] ) implies @xmath216 .
hence , @xmath217 , which implies @xmath218 .
therefore , the p - value @xmath219 concerning the modified tests @xmath208 is smaller than the upper bound @xmath220 of p - value concerning the likelihood ratio tests .
this test @xmath221 coincides with the likelihood ratio test in the one - parameter case .
assume that the data @xmath222 obeys the identical and independent distribution of the same distribution family @xmath223 and @xmath18 is sufficiently large .
when the true parameter @xmath56 is close to @xmath66 , it is known that the meaningful information for @xmath56 is essentially given as the random variable @xmath224 , where the logarithmic derivative @xmath225 is defined by @xmath226 in this case , the random variable @xmath227 can be approximated by the normal distribution with the expectation value @xmath228 and the variance @xmath229 , where the fisher information @xmath230 is defined as @xmath231 .
hence , the testing problem can be approximated by the testing of this normal distribution family @xcite .
that is , the quality of testing is approximately evaluated by the fisher information @xmath232 at the threshold @xmath66 . in the case of poisson distribution family poi@xmath233
, the parameter @xmath56 can be estimated by @xmath234 .
the asymptotic case corresponds to the case with large @xmath5 . in this case
, fisher information is @xmath235 .
when @xmath54 obeys the unknown poisson distribution family poi@xmath233 , the estimation error @xmath236 is close to the normal distribution with the variance @xmath237 , _
i.e. _ , @xmath238 approaches to the random variables obeying the normal distribution with variance @xmath56 .
that is , fisher information corresponds to the inverse of variance of the estimator .
this approximation can be extended to the multi - parameter case @xmath239 .
similarly , it is known that the testing problem can be approximated by the testing of the normal distribution family with the covariance matrix @xmath240 , where the fisher information matrix @xmath241 is given by @xmath242 when the hypotheses is given by ( [ eq : hypo - int ] ) , the testing problem can be approximated by the testing of the normal distribution family with variance @xmath243 , indeed , the same fact holds for the multinomial poisson distribution family poi@xmath244 .
when the random variable @xmath245 is the @xmath246-th random variable , the random variable @xmath247 converges to the random variable obeying the normal distribution with the variance @xmath248 in distribution : @xmath249 this convergence is compact uniform concerning the parameter @xmath250 . in this case , the fisher information matrix @xmath251 is the diagonal matrix with the diagonal elements @xmath252 .
when our distribution family is given as a subfamily poi@xmath253 , the fisher information matrix is @xmath254 , where @xmath255 .
hence , when the hypotheses is given by ( [ eq : hypo - int-3 ] ) , the testing problem can be approximated by the testing of the normal distribution family with variance @xmath256 in the following , we call this value fisher information .
based on this value , the quality can be compared when we have several testing schemes . in the following , we treat testing of the hypothesis ( [ eq : hypo - int-3 ] ) in the multinomial poisson distribution poi(@xmath250 ) by using normal approximation . in this case , by using @xmath257 defined in ( [ 2 - 6 - 12 ] ) and ( [ 2 - 6 - 11 ] ) , the upper bound ( [ 2 - 6 - 13 ] ) of the p - value concerning the likelihood ratio tests is approximated to @xmath258 because this convergence ( [ 3 - 7 - 1 ] ) is compact uniform concerning the parameter @xmath250 .
letting @xmath259 and @xmath260 , we have @xmath261 where @xmath262 is the convex hull of @xmath263 . as is shown in appendix [ a5 ] , this value is simplified to @xmath264 where @xmath265 where @xmath266 that is , our upper bound of p - value concerning the likelihood ratio tests is given by @xmath267 next , we approximately calculate the test with the risk probability @xmath74 proposed in section[s4d ] .
first , we choose @xmath268 by @xmath269 then , our test is given by the rejection region @xmath270 . using the same discussion ,
the p - value concerning the proposed tests is equal to @xmath271
in the following sections , we apply the discussions in sections [ s3 ] - [ s5 ] to the hypothesis ( [ 5 - 5 - 2 ] ) . that is , we consider how to reject the null hypothesis @xmath272 with a certain risk probability @xmath74 . in the usual visibility
, we usually measure the coincidence events only in the one direction or two directions .
however , in this method , the number of the counts of coincidence events be reflected not only by the fidelity but also by the direction of difference between the true state of target maximally entangled state . in order to remove the bias based on such a direction
, we propose to measure the counts of the coincidence vectors @xmath273 and the given state @xmath20 , using the total number of counts of the coincidence events ( the total count on coincidence event ) @xmath274 and the total number of counts of the anti - coincidence events ( the total count on anti - coincidence events ) @xmath275 obtained by measuring on all the vectors with the time @xmath276 .
when the dark count is negligible , the total count on coincidence events @xmath274 obeys poi@xmath277 , and the count on total anti - coincidence events @xmath275 obeys the distribution poi@xmath278 .
these expectation values @xmath279 and @xmath280 are given as @xmath281 and @xmath282 .
hence , fisher information matrix concerning the parameters @xmath14 and @xmath16 is @xmath283 where the first element corresponds to the parameter @xmath14 and the second one does to the parameter @xmath16 .
then , we can apply the test @xmath284 given in the end of subsection [ 2 - 7 - 5 ] .
that is , based on the ratio @xmath285 , we estimate the fidelity using the ratio @xmath286 as @xmath287 .
based on the discussion in subsection [ 2 - 7 - 4 ] , its variance is asymptotically equal to @xmath288 hence , similarly to the visibility , we can check the fidelity by using this ratio . indeed , when we consider the distribution under the condition that the total count @xmath156 is fixed to @xmath18 , the random variable @xmath275 obeys the binomial distribution with the average value @xmath289 .
hence , we can apply the likelihood ratio test of the binomial distribution . in this case , by the approximation to the normal distribution , the likelihood ratio test with the risk probability @xmath74 is almost equal to the test with the rejection region : @xmath290 concerning the null hypothesis @xmath272 .
the p - value of this kind of tests is @xmath291 .
in this section , we consider the problem of testing the fidelity between the maximally entangled state @xmath292 and the given state @xmath20 by performing three kinds of measurement , coincidence , anti - coincidence , and total flux , with the times @xmath293 and @xmath53 , respectively . when the dark count is negligible , the data @xmath294 obeys the multinomial poisson distribution poi@xmath295 with the assumption that the parameter @xmath16 is unknown . in this problem , it is natural to assume that we can select the time allocation with the constraint for the total time @xmath296 .
the performance of the time allocation @xmath297 can evaluated by the variance ( [ 2 - 16 - 5 ] ) .
the fisher information matrix concerning the parameters @xmath14 and @xmath16 is @xmath298 where the first element corresponds to the parameter @xmath14 and the second one does to the parameter @xmath16 .
then , the asymptotic variance ( [ 2 - 16 - 5 ] ) is calculated as @xmath299 we optimize the time allocation by minimizing the variance ( [ 2 - 16 - 10 ] ) .
we perform the minimization by maximizing the inverse : @xmath300 . applying lemmas [ 2 - 24 - 6 ] and [ 2 - 18 - 6 ] shown in appendix [ app1 ] to the case of @xmath301 , @xmath302 , @xmath303 , @xmath304
, we obtain @xmath305 and @xmath306 then , these relations give the optimal time allocations between ( i ) coincidence and total flux measurements , ( ii ) anti - coincidence and total flux measurements , and ( iii ) coincidence and anti - coincidence measurements , respectively .
the ratio of ( [ 2 - 19 - 3 ] ) to ( [ 2 - 19 - 1 ] ) is equal to @xmath307 as shown in appendix [ 2 - 24 - 10 ] .
that is , the optimal measurement using the coincidence and the anti - coincidence always provides better test than that using the coincidence and the total flux .
hence , we compare ( ii ) with ( iii ) , and obtain @xmath308 where the critical point @xmath309 is defined by @xmath310 the approximated value of the critical point @xmath311 is @xmath312 . the equation ( [ 2 - 24 - 11 ] )
is derived in appendix [ 2 - 24 - 12 ] .
[ ratio1 ] shows the ratio of the optimal fisher information based on the anti - coincidence and total flux measurements to that based on the coincidence and anti - coincidence measurements .
when @xmath313 , the maximum fisher information is attained by @xmath314 , @xmath315 , @xmath316 . otherwise , the maximum is attained by @xmath317 , @xmath318 , @xmath319 . the optimal time allocation shown in fig .
[ ratio1 ] implies that we should measure the counts on the anti - coincidence vectors preferentially over other vectors . .
the measurement time is divided into three periods : coincidence @xmath320 ( plus signs ) , anti - coincidence @xmath321 ( circles ) , and total flux @xmath322 ( squares ) , which are normalized as @xmath323 in the plot.,width=302 ] the optimal asymptotic variance is @xmath324 when the threshold @xmath15 is less than the critical point @xmath311 .
this asymptotic variance is much better than that obtained by the modified visibility method .
the ratio of the optimal asymptotic variance is given by @xmath325 in the following , we give the optimal test of level @xmath74 in the hypothesis testing ( [ eq : hypo ] ) .
assume that the threshold @xmath15 is less than the critical point @xmath311 . in this case
, we can apply testing of the hypothesis ( [ eq : hypo - int-2 ] ) .
first , we measure the count on the coincidence vectors for a period of @xmath326 , to obtain the total count @xmath327 .
then , we measure the count on the anti - coincidence vectors for a period of @xmath328 to obtain the total count @xmath329 .
note that the optimal time allocation depends on the threshold of our hypothesis .
finally , we apply the ump test of @xmath74 of the hypothesis : @xmath330 with the binomial distribution family @xmath331 to the data @xmath274 . in this case , the likelihood ratio test with the risk probability @xmath74 is almost equal to the test with the rejection region : @xmath332 concerning the null hypothesis @xmath272 . the p - value of this kind of tests is @xmath333 .
we can apply a similar testing for @xmath334 .
it is sufficient to replace the time allocation to @xmath335 @xmath336 , @xmath337 . in this case , the likelihood ratio test with the risk probability @xmath74 is almost equal to the test with the rejection region : @xmath338 concerning the null hypothesis @xmath272 .
the p - value of this kind of tests is @xmath339 .
next , we consider the case where the dark count parameter @xmath340 is known but is not negligible , the fisher information matrix is given by @xmath341 hence , from ( [ 2 - 16 - 5 ] ) , the inverse of the minimum variance is equal to @xmath342 then , we apply lemmas [ 2 - 24 - 6 ] and [ 2 - 18 - 6 ] in appendix [ app1 ] to @xmath343 with @xmath344 , @xmath345 , @xmath346 , @xmath347 , and obtain the optimized value : @xmath348 and @xmath349 the ratio of ( [ 2 - 24 - 14 ] ) to ( [ 2 - 24 - 16 ] ) is @xmath350 where the final inequality is derived in appendix [ 2 - 24 - 10 ] .
therefore , the measurement using the coincidence and the anti - coincidence provides better test than that using the coincidence and the total flux , as in the case of @xmath351 .
define @xmath352 and the critical point @xmath353 for the normalized dark count @xmath354 as @xmath355 the parameter @xmath352 is calculated to be @xmath356 .
as shown in appendix [ 2 - 24 - 12 ] , the measurement using the coincidence and the anti - coincidence provides better test than that using the anti - coincidence and the total flux , if the fidelity is smaller than the critical point @xmath353 : @xmath357 the optimal time allocation is given by @xmath314 , @xmath358 , and @xmath359 for @xmath360 , and @xmath361 , @xmath362 , @xmath319 for @xmath363 .
the critical point @xmath353 for optimal time allocation increases with the normalized dark count as illustrated in fig .
[ thresh1 ] . for optimal time allocation as a function of normalized dark counts @xmath364.,width=302 ]
in this section , we consider the case where @xmath16 is known . then , the fisher information is @xmath365 the maximum value is calculated as @xmath366 the above optimization shows that when @xmath367 , the count on anti - coincidence @xmath368 is better than the count on coincidence @xmath369 .
in fact , barbieri _
et al._@xcite measured the sum of the counts on the anti - coincidence vectors @xmath370 to realize the entanglement witness in their experiment . in this case , the variance is @xmath371 .
when we observe the sum of counts on anti - coincidence @xmath275 , the estimated value of @xmath14 is given by @xmath372 , which is the solution of @xmath373 .
the likelihood ratio test with the risk probability @xmath74 can be approximated by the test with the rejection region : @xmath374 concerning the null hypothesis @xmath272 , which is also the ump test .
the p - value of likelihood ratio tests is @xmath375 . when @xmath376 , the optimal time allocation is @xmath377 , @xmath378 .
the fidelity is estimated by @xmath379 .
its variance is @xmath380 .
the likelihood ratio test with the risk probability @xmath74 of the poisson distribution is almost equal to the test with the rejection region : @xmath381 concerning the null hypothesis @xmath272 , which is also the ump test .
the p - value of likelihood ratio tests is @xmath382 .
we compare the asymptotic variances of the following designs for time allocation , when the dark count @xmath340 parameter is zero .
( i ) : : modified visibility : the asymptotic variance is @xmath383 .
( iia ) : : design i ( @xmath16 unknown )
. optimal time allocation between the counts on anti - coincidence and coincidence : the asymptotic variance is @xmath384 .
( iib ) : : design i ( @xmath16 unknown ) , optimal time allocation between the counts on anti - coincidence and the total flux : the asymptotic variance is @xmath385 .
( iiia ) : : design ii ( @xmath16 known ) , estimation from the count on anti - coincidence : the asymptotic variance is @xmath386 .
( iiib ) : : design ii ( @xmath16 known ) , estimation from the count on coincidence : the asymptotic variance is @xmath387 .
[ fig : relent9 ] shows the comparison , where the asymptotic variances in ( iia)-(iiib ) are normalized by the one in ( i ) .
the anti - coincidence measurement provides the best estimation for high ( @xmath388 ) fidelity .
when @xmath16 is unknown , the measurement with the counts on anti - coincidence and the coincidence is better than that with the counts anti - coincidence and the total flux for @xmath389 . for higher fidelity ,
the counts on anti - coincidence and total flux turns to be better , but the difference is small .
the comparison in the previous section shows that the measurement on the anti - coincidence vectors yields a better variance than the measurement on the coincidence vectors , when the fidelity is greater than @xmath7 and the parameters @xmath16 and @xmath340 are known .
we will explore further improvement in the measurement on the anti - coincidence vectors . in the previous sections ,
we allocate an equal time to the measurement on each of the anti - coincidence vectors . here
we minimize the variance by optimizing the time allocation @xmath390 , @xmath391 , @xmath392 , @xmath393 , @xmath394 , and @xmath395 between the anti - coincidence vectors @xmath396 , @xmath397 , @xmath398 , @xmath399 , @xmath400 , and @xmath401 , under the restriction of the total measurement time : @xmath402 . the number of the counts @xmath403 obeys poisson distribution poi(@xmath404 ) with unknown parameter @xmath405 .
then , the fisher information matrix is the diagonal matrix with the diagonal elements @xmath406 since we are interested in the parameter @xmath407 , the variance is given by @xmath408 as mentioned in section [ 2 - 7 - 4 ] . under the restriction of the total measurement time ,
the minimum value of ( [ 2 - 24 - 30 ] ) is @xmath409 which is attained by the optimal time allocation @xmath410 which is called neyman allocation and is used in sampling design@xcite .
the variance with the equal allocation is @xmath411 the inequality ( [ 2 - 25 - 11 ] ) @xmath412 ( [ 2 - 25 - 10 ] ) can be derived from schwartz s inequality of the vectors @xmath413 and @xmath414 .
the equality holds if and only if @xmath415 .
therefore , the neyman allocation has an advantage over the equal allocation , when there is a bias in the parameters @xmath416 . in other words ,
the neyman allocation is effective when the expectation values of the counts on some vectors are larger than those on other vectors .
the optimal time allocation derived above is not applicable in the experiment , because it depends on the unknown parameters @xmath417 @xmath418 @xmath419 @xmath420 @xmath421 and @xmath422 .
in order to resolve this problem , we introduce a two - stage method , where the total measurement time @xmath5 is divided into @xmath423 for the first stage and @xmath424 for the second stage under the condition of @xmath425 . in the first stage , we measure the counts on each vectors for @xmath426 and estimate the expectation value for neyman allocation on measurement time @xmath424 . in the second stage ,
we measure the counts on a vector @xmath427 according to the estimated neyman allocation .
the two - stage method is formulated as follows .
+ ( i ) the measurement time for each vector in the first stage is given by @xmath426 + ( ii ) in the second stage , we measure the counts on a vector @xmath427 with the measurement time @xmath428 defined as @xmath429 where @xmath430 is the observed count in the first stage .
+ ( iii ) define @xmath431 and @xmath432 as @xmath433 where @xmath434 is the number of the counts on @xmath427 for @xmath428 . then , we can estimate the fidelity by @xmath435 .
+ ( iv ) finally , we apply the test @xmath436 given in section [ s4d ] to the two hypotheses given as @xmath437 where @xmath438 and @xmath439 .
we have formulated the hypothesis testing scheme to test the entanglement in the poisson distribution framework .
our statistical method can handle the fluctuation in the experimental data more properly in a realistic setting .
it has been shown that the optimal time allocation improves the test : the measurement time should be allocated preferably to the anti - coincidence vectors .
this test is valid even if the dark count exists .
this design is particularly useful for the experimental test , because the optimal time allocation depends only on the threshold of the test .
we do nt need any further information of the probability distribution and the tested state .
the test can be further improved by optimizing time allocation between the anti - coincidence vectors , when the error from the maximally entangled state is anisotropic .
however , this time allocation requires the expectation values on the counts on coincidence , so that we need to apply the two stage method .
the authors would like to thank professor hiroshi imai of the erato - sorst , qci project for support .
they are grateful to dr .
tohya hiroshima , dr .
yoshiyuki tsuda for useful discussions .
in this section , we maximize the quantities appearing in fisher information . [ 2 - 24 - 6 ] the equation @xmath440 holds and
the maximum value is attained when @xmath441 , @xmath442 .
letting @xmath443 , we have @xmath444 . then , @xmath445 hence , the maximum is attained at @xmath446 , _
@xmath447 and @xmath448 .
thus , @xmath449 [ 2 - 18 - 6 ] the equation @xmath450 holds , and this maximum value is attained when @xmath451 , @xmath452 .
letting @xmath453 , we have @xmath454 and @xmath455 . then , @xmath456 hence , the maximum is attained at @xmath457 , _
@xmath451 and @xmath452 .
thus , @xmath458 further , three - parameter case can be maximized as follows . the maximum value @xmath459 is equal to the maximum among three values @xmath460 , @xmath461 , @xmath462 .
define two parameters @xmath463 and @xmath464 .
then , the range of @xmath68 and @xmath465 forms a convex set . since @xmath466 hence , @xmath467 where @xmath468 , @xmath469 . applying lemma [ 2 - 18 - 6-a ] , we obtain this lemma .
[ 2 - 18 - 6-a ] define the function @xmath470 on a closed convex set @xmath471 .
the maximum value is realized at the boundary @xmath472 .
the condition can be classified to two cases : i ) @xmath473 , ii ) @xmath474 . in the case
i ) , when fix @xmath68 is fixed , @xmath475 .
then , we obtain @xmath476 . in the case
ii ) , when @xmath477 , @xmath478 . hence ,
@xmath479 this maximum is attained at @xmath480 or @xmath481 .
these point belongs to the boundary @xmath472 .
further , @xmath482 .
thus , the proof is completed .
it is sufficient to show @xmath483 by putting @xmath484 , the lhs is evaluated as @xmath485 since @xmath486 , we have @xmath487 further , the function @xmath488 @xmath489)$ ] has the minimum @xmath490 at @xmath491 .
hence , @xmath492 .
it is sufficient to show that @xmath493 if and only if @xmath494 and @xmath495 . by putting @xmath484 , the lhs of ( [ 2 - 24 - 4 ] )
is evaluated as @xmath496 since @xmath486 and @xmath497 , @xmath498 if and only if @xmath499 and @xmath360 .
define @xmath257 by @xmath500 in fact , when @xmath501 , @xmath502 this value is monotone decreasing concerning @xmath503 . when @xmath504 , this value is @xmath505 .
hence , the value @xmath257 coincides with the the value @xmath257 defined by ( [ 2 - 6 - 12 ] ) and ( [ 2 - 6 - 11 ] ) .
thus , the relation ( [ 5 - 2 - 1 ] ) follows from the relation @xmath506 we choose @xmath507 such that @xmath508 .
then , the above inequality follows from lemma [ le-3 - 12 ] in the following way : @xmath509 [ le-3 - 12 ] any real number @xmath510 and any four sequence of positive numbers @xmath511 , @xmath512 , @xmath513 , and @xmath514 satisfy @xmath515 it is sufficient to show @xmath516 the convexity of @xmath517 implies that @xmath518 hence , @xmath519
considering the shape of the graph @xmath520 , we can show that the minimum value @xmath521 can be attained by the boundary of @xmath471 .
hence the boundary of the convex set @xmath262 is included by the union @xmath522 of the lines @xmath523 .
taking the derivative of @xmath524 concerning @xmath5 , we obtain @xmath525 } \frac{t x_i(r)+(1-t)x_j(r ) } { \sqrt{t y_i(r)+(1-t)y_j(r ) } } = z_{i , j}(r).\end{aligned}\ ] ] hence , we obtain ( [ 5 - 5 - 1 ] )
. 99 c.h .
bennett , g. brassard , c. crpeau , r. jozsa , a. peres , and w. k. wootters , _ phys .
lett . _ * 70 * , 1895 ( 1993 ) .
briegel , w. dur , j.i .
cirac , and p. zoller , _ phys .
_ , * 81 * , 5932 ( 1998 ) . c. w. helstrom , _ quantum detection and estimation theory
, academic press ( 1976 ) .
m. barbieri , f. de martini , g. di nepi , p. mataloni , g. m. dariano , and c. macchiavello , _ phys .
_ , * 91 * , 227901 ( 2003 ) .
y. tsuda , k. matsumoto , and m. hayashi .
`` hypothesis testing for a maximally entangled state , '' quant - ph/0504203 .
p. g. kwiat , e. waks , a. g. white , i. appelbaum , and p.h .
eberhard , _ phys .
a _ , * 60 * , 773(r ) ( 1999 ) . k. usami , y. nambu , y. tsuda , k. matsumoto , and k. nakamura , `` accuracy of quantum - state estimation utilizing akaike s information criterion , '' _ phys . rev .
a _ , * 68 * , 022314 ( 2003 ) . | a hypothesis testing scheme for entanglement has been formulated based on the poisson distribution framework instead of the povm framework .
three designs were proposed to test the entangled states in this framework .
the designs were evaluated in terms of the asymptotic variance .
it has been shown that the optimal time allocation between the coincidence and anti - coincidence measurement bases improves the conventional testing method .
the test can be further improved by optimizing the time allocation between the anti - coincidence bases . |
classical and quantum information processing network architectures utilize light ( optical photons ) for the transmission of information over extended distances , ranging from hundreds of meters to hundreds of kilometers @xcite .
the utility of optical photons stems from their weak interaction with the environment , large bandwidth of transmission , and resiliency to thermal noise due to their high frequency ( @xmath0 ) .
acoustic excitations ( phonons ) , though limited in terms of bandwidth and their ability to transmit information farther than a few millimeters , can be delayed and stored for significantly longer times and can interact resonantly with rf - microwave electronic systems @xcite .
this complimentary nature of photons and phonons suggests hybrid phononic - photonic systems as a fruitful avenue of research , where a new class of _ optomechanical _ circuitry could be made to perform a range of tasks out of reach of purely photonic and phononic systems .
a building block of such a hybrid architecture would be elements coherently interfacing optical and acoustic circuits . the optomechanical translator we propose in this paper acts as a chip - scale _ transparent , coherent interface _ between phonons and photons and fulfills a key requirement in such a program . in the quantum realm , systems involving optical , superconducting ,
spin or charge qubits coupled to mechanical degrees of freedom @xcite have been explored .
the recent demonstration of coherent coupling between a superconducting qubit and a mechanical resonance by oconnell , et al .
@xcite , has provided an experimental backing for this vision and is the latest testament to the versatility of mechanics as a connecting element in hybrid quantum systems . in the specific case of phonon - photon state transfer , systems involving trapped atoms , ions , nanospheres @xcite , and mechanically compliant optical cavity structures
@xcite have all been considered . in these past studies ,
the state of an incoming light field is usually mapped onto the motional state of an atom , ion , or macroscopic mirror , through an exact timing of control pulses , turning on and off the interaction between the light and mechanical motion in a precise way . the ability to simultaneously implement phononic and photonic waveguides in optomechanical crystal ( omc ) structures @xcite opens up the opportunity to implement a _ traveling - wave _ phonon - photon translator ( ppt ) .
such a device , operating continuously , connects acoustic and optical waves to each other in a symmetric manner , and allows for on - the - fly conversion between phonons and photons without having to precisely time the information and control pulses . in effect
, the problem of engineering control pulses is converted into a problem of engineering coupling rates .
our proposal for a ppt is motivated strongly by recent work @xcite on radiation pressure effects in micro- and nano - scale mechanical systems @xcite .
furthermore , the concrete realization of a ppt is aided by the considerable advances made in the last decade in the theory , design and engineering of thin - film artificial quasi-2d ( patterned membrane ) crystal structures containing photonic @xcite and phononic @xcite `` band gaps '' .
such systems promise unprecedented control over photons and phonons , and have been separately subject to extensive investigation .
their unification , in the form of omcs which possesses a simultaneous phononic and photonic bandgap @xcite , and in which the interaction between the photons and phonons can be controlled , promises to further expand the capabilities of both photonic and phononic architectures and forms the basis of the proposed ppt implementation .
the outline of this paper is as follows . in sections [ sec : outline ] and
[ sec : analysis ] we introduce and study the ppt system as an abstraction , at first classically and then quantum mechanically . after introducing the basic system , its properties and its scattering matrix , we study the effects of quantum and classical noise on device operation . in section [ sec : implementation ] we design and simulate a possible physical implementation of the system , utilizing recent results in simultaneous phononic - photonic bandgap materials @xcite .
finally , in section [ sec : applications ] , we demonstrate a few possible applications of the ppt . focusing first on `` classical '' applications , we evaluate the performance of the ppt when used for the implementation of an optical delay line and wavelength converter .
finally , we show how such a system could be used in theory to do high fidelity quantum state transfer between optical and superconducting qubits .
the proposed ppt system , shown in fig . [ fig : full_system_diagram ] , consists of a localized mechanical resonance ( @xmath1 ) which couples the two optical resonances ( @xmath2,@xmath3 ) of an optomechanical cavity via radiation pressure . external coupling to and from the mechanical resonance is provided by an acoustic waveguide , while each of the optical resonances are coupled to via separate optical waveguides .
multi - optical - mode optomechanical systems have been proposed and experimentally studied previously in the context of enhancing quantum back - action , reduced lasing threshold , and parametric instabilities @xcite .
here we use a two - moded optical cavity as it allows for the spatial filtering and separation of signal and pump optical beams while reducing the quantum noise in the system , as is explained below . a general description of the radiation - pressure - coupling of the mechanical and optical degrees of freedom in such a structure is as follows .
for each of the two high-@xmath4 optical resonances of the cavity we associate an annihilation operator @xmath5 and a frequency @xmath6 ( @xmath7 ) .
geometric deformation of the optomechanical cavity parameterized by @xmath8 , changes the frequencies of the optical modes by @xmath9 .
the deformation , due to the localized mechanical resonance with annihilation operator @xmath10 and frequency @xmath11 , can be quantized and given by @xmath12 .
there is also a coupling between the two optical cavity modes given by @xmath13 , where for resonant intermodal mechanical coupling the cavity structure must be engineered such that @xmath14 . in a traveling wave ppt device consisting of the two optical cavity resonances and a single mechanical resonance ,
the lower frequency cavity mode ( @xmath2 in this case ) is used as a pump cavity which enables the inter - conversion of phonons in the mechanical resonance ( @xmath1 ) to photons in the second , higher frequency , optical cavity mode ( @xmath3 ) through a two - photon process in which pump photons are either absorbed or emitted as needed .
the signals representing the phonon and photon quanta to be exchanged will thus be contained in @xmath1 and @xmath2 , respectively .
as described , the hamiltonian of this system is , @xmath15 where we have added a classical optical pumping term of electric field amplitude @xmath16 and frequency @xmath17 .
optical pumping is performed through one of the optical waveguides with ( field ) coupling rate to the @xmath2 cavity resonance given by @xmath18 .
in addition to the waveguide loading of each optical resonance ( @xmath19 ) , the total optical loss rate of each cavity mode includes an intrinsic component ( @xmath20 ) of field decay due to radiation , scattering , and absorption .
similarly for the mechanical resonance , we have a field decay rate given by @xmath21 which is a combination of waveguide loading and intrinsic losses .
the constant parts of both @xmath22 and @xmath23 ( @xmath24 , @xmath25 ) can be eliminated by a change of basis and are thus taken to be zero . as discussed below , it is advantageous to choose a cavity structure symmetry in which @xmath26 up to linear order in @xmath27 .
in fact , we can generally assume that the mechanical displacements are small enough to make the linear order the only important term in the interaction .
assuming then a properly chosen cavity symmetry , @xmath28 which yields a simplified hamiltonian @xmath29 treating the system classically and approximately , we can write each intracavity photon and phonon amplitude , and their inputs ( see fig . [ fig : doublecav_sidebands ] ) as a fourier decomposition of a few relevant sidebands : @xmath30 the equations of motion arrived at from the system hamiltonian ( presented generally in the following section ) then become algebraic relations between the @xmath31 and @xmath32 sideband amplitudes . by ignoring the self - coupling term ( @xmath33 ) , pumping on - resonance with cavity mode @xmath2 ( @xmath34 ) , and engineering the optical cavity mode splitting for mechanical resonance ( @xmath35 ) ,
we arrive at classical sideband amplitudes , @xmath36 from here we see that the central sideband amplitude of cavity mode @xmath2 , @xmath37 , is proportional to the sum of a term containing the pump field @xmath38 and terms containing products of the optical and mechanical sideband amplitudes . by increasing @xmath38
the effect of the other sidebands on the pump resonance amplitude can be made negligible , and we assume here and elsewhere in this work that the pump sideband is generally left unaffected by the dynamics of the rest of the system . as desired the optical sideband which contains mechanical information
is @xmath39 since it is the only sideband directly proportional to @xmath40 .
the constant of proportionality between these two terms is seen to contain both @xmath41 and @xmath37 , demonstrating the role of the pump beam in the conversion process .
since coherent information transfer between the optics and mechanics is occurring between @xmath40 and @xmath39 , it is desirable to remove the effects of the lower energy photonic sideband , @xmath42 .
this sideband can be made significantly smaller in magnitude than @xmath39 in the sideband - resolved regime where @xmath43 . a convenient way to visualize all of the processes in the system
is shown in figure [ fig : doublecav_sidebands ] where the photonic sideband amplitudes @xmath44 and @xmath37 are represented as `` energy levels '' , with transitions between them being due to the emission and absorption of phonons . from this approximate analysis
it is clearly suggested that in a sideband resolved optomechanical system , a state - transfer process is possible between the phononic and photonic resonances , and the process is controlled by a pump beam @xcite .
a more in - depth study of the system dynamics required to understand how such processes may be used for traveling wave phonon - photon conversion , and a full investigation of all relevant noise sources required to understand the applicability of such a system to quantum information , is carried out in the following section .
a detailed treatment of the operation of a traveling phonon - photon translator is carried out in this section . at first , the dynamics of the system are simplified while still taking into account the noise processes related to the sideband @xmath42 . in this way one
is left with an effective ` beam - splitter ' hamiltonian , which describes the coherent interaction between the optics and mechanics , while the aforementioned noise processes are accounted for through an effective increase in the thermal bath temperature .
this is followed by a treatment of the traveling - wave problem through a scattering matrix formulation , which provides insight into the role of the intracavity pump photon number ( @xmath45 ) and optimizing the state transfer efficiency .
starting from the hamiltonian in equation ( [ eqn : total_hamiltonian ] ) , a set of heisenberg - langevin equations can be written down ,
@xcite : @xmath46 the input coupling terms as written above include both external waveguide coupling and intrinsic coupling due to lossy channels ( see fig .
[ fig : full_system_diagram ] ) .
separated , the intrinsic ( with subscript @xmath47 ) and extrinsic ( with no subscript ) components look as follows , @xmath48 as the fluctuations in the fields are of primary interest , each heisenberg operator can be rewritten as a fluctuation term around a steady - state value , @xmath49 assuming that the pump beam is driven resonantly with @xmath50 ( @xmath34 ) , the @xmath51-number steady - state values are equal to @xmath52 . for the fluctuation dynamics , with @xmath53 , the resulting equations are : @xmath54 ignoring all the mechanically anti - resonant terms for now , and invoking the rotating wave approximation ( rwa ) valid when @xmath55 and @xmath56 , we arrive at the simplified set of fluctuation equations , @xmath57 [ ss : spont_emission ] by ignoring all of the counter - rotating terms proportional to @xmath58 , we have also neglected the noise processes alluded to previously due to the @xmath42 sideband . of the mechanically anti - resonant terms which have been dropped ,
the terms proportional to @xmath59 ( @xmath60 in equation ( [ eqn : hl_full_a2dot ] ) and @xmath61 in equation ( [ eqn : hl_full_bdot ] ) ) are the largest and most significant in terms of error in the rotating wave approximation .
these terms correspond to @xmath62 + h.c . in the hamiltonian and cause inter - sideband photon scattering between the pump , @xmath37 , and its lower frequency sideband , @xmath42 as shown in figure [ fig : doublecav_sidebands ] .
this inter - sideband scattering process causes emission and absorption of phonons in the mechanical part of the ppt , thus in principle , even when the extrinsic phonon inputs are in the vacuum state , spontaneous scattering of photons from @xmath37 to @xmath42 may populate the mechanical cavity with a phonon .
this effect was studied in refs .
@xcite in the context of quantum limits to optomechanical cooling .
similar to that work , a master equation for the phononic mode with the @xmath63 optical sideband adiabatically eliminated results in an additional _ phononic _ spontaneous emission term given by , @xmath64 where @xmath65 . the master equation for the mechanics , found by tracing over all bath and optical variables , is of the form @xmath66 $ ] , where the terms on the right hand side of the equation are respectively the intrinsic phononic loss , the phonon spontaneous emission , the phonon - waveguide coupling , and the coherent evolution of the system .
the first two terms can be lumped together into an effective intrinsic loss , @xmath67 , @xmath68 and n _ .
hence , by assuming that the intrinsic loss phonon bath is at a modified temperature with occupation number @xmath69 , and changing the intrinsic phonon loss rate to @xmath70 , the spontaneous emission and intrinsic loss noise are lumped into one effective thermal noise liouvillian for the mechanics .
note that it is possible in this model to have @xmath70 negative when @xmath71 ; however , the motional decoherence rate is always positive and given by @xmath72 . in section [ ss : scattering_matrix ] we will see that the optimal state transfer efficiency is given by @xmath73 in this case is in principle @xmath74 as opposed to @xmath75 , and the equations must therefore be solved self - consistently .
this is discussed in section [ ss : spont_mod ] . ]
in which case@xmath76 in the sideband resolved limit . in the case
where @xmath77 ( recall this is the self - coupling radiation pressure term ) , photons scattering from the pump into the lower frequency sideband ( @xmath78 ) can scatter into the @xmath50 cavity mode which is only detuned by @xmath11 .
this is to be compared with the @xmath33 case considered thus far in which the detuning is @xmath79 for scattering into the @xmath80 mode . as such , for the @xmath77 case there will be roughly four times the spontaneous emission noise , with @xmath81 .
a final simplification can be made by neglecting the fluctuations in the strong optical pump of cavity mode @xmath50 .
considering that the fluctuations in the variables are all of the same order , and that @xmath82 always appears as @xmath83 in the equations of motion for @xmath84 and @xmath85 , we can ignore the dynamics of the pump fluctuations in the case where @xmath86 and @xmath87 .
this is the undepleted pump approximation .
adiabatically removing the pump from the dynamics of the system yields a pump - enhanced optomechanical coupling @xmath88 between optical cavity mode @xmath80 and the mechanical resonance @xmath1 .
dropping the subscript from the cavity mode @xmath80 and moving to a rotating reference frame results in the new effective hamiltonian @xcite , @xmath89 the system diagram and symbol corresponding to this simplified model of the ppt are shown in figure [ fig : simplified_system_diagram ] . to understand the properties of the ppt as a waveguide adapter , we begin with a study of its scattering matrix . starting from the effective hamiltonian given in equation ( [ eqn : h_eff_start ] ) ,
the heisenberg - langevin equations of motion for the hamiltonian ( [ eqn : h_eff_start ] ) are written under a markov approximation in the frequency domain , @xmath90 the intrinsic noise terms @xmath91 and @xmath92 are the initial - state boson annihilation operators for the baths , while the extrinsic terms @xmath93 and @xmath94 are annihilation operators for the optical and mechanical guided modes for each respective waveguide . since the effective hamiltonian ( [ eqn : h_eff_start ] ) has been used to derive equations ( [ eqn : freqdom1]-[eqn : freqdom2 ] ) , thus neglecting the counter - rotating terms present in the full system dynamics , the effects of phonon spontaneous emission noise is included separately . following the discussion in section [ ss : spont_emission ] ,
the effective liouvillian ( [ eqn : effective_louivillian ] ) corresponds to replacing @xmath95 with @xmath96 in ( [ eqn : freqdom2 ] ) and using a langevin force @xmath92 satisfying the relations , @xmath97 where @xmath96 and @xmath69 are given in equations ( [ eqn : gamma_prime]-[eqn : n_prime ] ) .
the intrinsic optical noise correlations are only due to vacuum fluctuations and given by @xmath98 . to ensure efficient translation , competing requirements of matching and strong coupling between waveguide and resonator must be satisfied .
this is similar to the problem of designing integrated optical filters using resonators and waveguides @xcite . from the above equations and the input - output boundary condition @xcite
, we arrive at the matrix equation , @xmath99 with scattering and noise matrices @xmath100 the elements of the scattering matrix @xmath101 are @xmath102 similar expressions are also found for the noise scattering matrix elements @xmath103 , with their extrema reported below . in order to obtain efficient conversion
the cavities must be over - coupled to their respective waveguides , ensuring that the phonon ( photon ) has a higher chance of leaking into the waveguide continuum modes than escaping into other loss channels . in this regime , @xmath104 and @xmath105 . in the weak coupling regime , @xmath106 ,
the response of the system exhibits a maximum for @xmath107 and @xmath108 at @xmath109 and a minimum at the same point for @xmath110 and @xmath111 .
in fact , with realistic system parameters , only the weak - coupling regime leads to efficient translation . in strong - coupling ( @xmath112 and @xmath113 ) the photon
is converted to a phonon at rate @xmath114 , and then back to a photon before it has a chance to leave through the much slower phononic loss channel at rate @xmath115 , causing there to be significant reflections and reduced conversion efficiency . to find the optimal value of @xmath114 we consider the extrema given by @xmath116 in the over - coupled approximation and in the case where @xmath117
, it is easy to see that the full translation condition @xmath118 is achievable by setting @xmath114 equal to g^ = [ eqn : matching_condition ] .
this result has a simple interpretation as a matching requirement .
the photonic loss channel viewed from the phononic mode has a loss rate of @xmath119 . matching this to the purely mechanical loss rate of the same phononic mode , @xmath115 , one arrives at @xmath120 .
the same argument can be used for the photonic mode , giving the same result . under this matched condition ,
the linewidth of the translation peak in @xmath121 is simply @xmath122 with intrinsic losses taken into account , either @xmath123 or @xmath124 ( but not both ) can be made exactly 0 by setting @xmath125 or @xmath126 , respectively .
the optimal state transfer condition , however , still occurs for @xmath120 .
the extremal values ( @xmath127 ) of the scattering matrix are in this case are , @xmath128 with corresponding noise matrix elements of , @xmath129 for a set of parameters typical of an optomechanical crystal system , the magnitudes of the scattering matrix elements versus frequency are plotted in figure [ fig : scattering_matrix ] . in these plots
we have assumed resonant optical pumping of the @xmath50 cavity mode and considered several different detuning values @xmath130 .
the normalized optical reflection spectrum ( @xmath131 ) is shown in figure [ fig : scattering_matrix](a ) , in which the broad optical cavity resonance can be seen along with a deeper , narrowband resonance that tunes with @xmath130 .
this narrowband resonance is highlighted in figure [ fig : scattering_matrix](b ) , showing that the optical reflection is nearly completely eliminated on resonance .
photons on resonance , instead of being reflected , are being converted into outgoing phonons as can be seen in resonance peak of @xmath132 shown in figure [ fig : scattering_matrix](d ) .
a similar reflection dip and transmission peak is visible for the phononic reflection ( @xmath111 ) and phonon to photon translation ( @xmath107 ) curves .
it is also notable from figure [ fig : scattering_matrix ] that for small detunings @xmath130 of the system ( @xmath133 ) the resonant scattering matrix elements are only weakly affected and the translation process maintains its efficiency . as a consequence of the counter - rotating terms treated in section [ ss : spont_emission ] , @xmath115 is weakly dependent on @xmath114 . in particular , by making the substitution @xmath134 , where @xmath96 is given by equation ( [ eqn : gamma_prime ] ) , the equation for the optimal @xmath114 becomes @xmath135 , where @xmath96 is itself dependent on @xmath114 .
algebraic manipulations give us the desired value of @xmath114 , @xmath136 simplifying in the sideband - resolved limit to @xmath137 the values for the scattering matrix elements given in equations ( [ eqn : s11]-[n_mat_4 ] ) are also suitably modified by the substitutions @xmath134 and @xmath138 .
the issue of waveguide loss is one that is normally ignored in quantum optical systems where low - loss fiber or free - space links are readily available . on the phononic side of the systems studied here
, the length of the waveguide and its intrinsic losses may be large enough for the waveguide attenuation factor to become important . because of the negative effect of attenuation in applications involving optomechanical delay lines , and quantum state transfer , it is useful to model this loss and see how the scattering matrix elements are altered . a model of the system shown in figure [ fig : waveguide - loss](b ) , where the lossy waveguide is replaced by a single beam splitter .
this is accomplished by first modeling the lossy waveguide as a large number @xmath139 of cascaded beam splitters , each reflecting @xmath140 away from the main beam , with @xmath141 .
this serial array of beam splitters can be combined into one , with reflectivity @xmath142 as @xmath143 .
the relation for @xmath144 can be found by starting from the approximate scattering matrix relation for @xmath145 , @xmath146 and using the beam splitter relations , @xmath147 assuming that the phonon cavity intrinsic loss bath is at the same temperature and uncorrelated with the phonon waveguide intrinsic loss baths , then one finds for translation through the lossy phonon waveguide , @xmath148 where @xmath149 the value of @xmath108 is simply reduced by a factor @xmath150 due to the lossy waveguide . for propagation lengths short relative to the attenuation length of the lossy waveguide , @xmath151 , and
the reduction in translation @xmath152 is small . the added noise due to the waveguide attenuation is contained in @xmath153 , and is also seen to be small for @xmath151 .
in general , the type of noise which is relevant to the ppt depends on the conditions in which it is used .
for example , when used as a bridge between rf - microwave photonics and optics for classical applications at room temperature , the thermal noise affecting the rf signal will be at a level which makes the quantum noise induced by spontaneous pump scattering irrelevant .
on the other hand , when the system is used at cryogenic temperatures for connecting a superconducting circuit qed system to an optical system as described below , the quantum noise of the translation process itself becomes dominant . in what follows we analyze both sources of noise .
the noise power on each of the output waveguide channels can be found by using the scattering matrix formulation described in section [ ss : scattering_matrix ] .
here we evaluate the effects of thermal noise classically .
when the system is used an optical drop filter , in which an optical beam is sent in and the optical reflection is measured , we find : @xmath154 where @xmath155 represents convolution .
assuming that there are no cross - correlations between the various input noise terms , we find the classical spectral density of the noise to be : @xmath156 for a system with mechanical frequency less than @xmath157 ghz , at room temperature ( @xmath158 ) the corresponding spontaneous emission noise is much less than the thermal noise , @xmath159 , and @xmath160 can be ignored . ignoring the quantum noise term for the moment , and evaluating equation ( [ eqn : sout1 ] ) at @xmath127 after substituting in equations
( [ n_mat_1]-[n_mat_4 ] ) , the total thermal noise power on the reflected optical signal is found to be , @xmath161 where @xmath162 is the bandwidth of the ppt . in the classical regime , @xmath163 , and this equation reduces to @xmath164 this result has a simple interpretation . the noise power
@xmath165 is the standard thermal noise input on the phononic side of the ppt .
the ratio @xmath166 is a translation factor , arising from the fact that quantum - limited conversion of phonons to photons causes an increase in energy by the factor of the ratio of their frequencies . finally , the factor @xmath167 is the extraction efficiency of photons from the optical mode to the waveguide .
alternatively , one may define an equivalent optical temperature by setting @xmath168 , yielding @xmath169 , where the @xmath4 s represent the loaded quality factors of the optical and mechanical resonators .
this last expression must be interpreted carefully , only in terms of a power equivalence , as spectrally the noise on the optical side of the ppt has a bandwidth of @xmath170 , while thermal noise radiating from the cavity would have a bandwidth of @xmath171 . to calculate the effective increase in noise brought about by the spontaneous pump scattering and phonon emission process , we start from equation ( [ eqn : sout1 ] ) and use the spontaneous emission contribution @xmath160 to find , @xmath172 the second term in the brackets is due to the spontaneous emission of phonons by the optical pump beam , and is added to the thermal noise exiting the optical side of the ppt . put in terms of an effective contribution to the thermal _ photon
_ occupation number , the spontaneous pump scattering effectively adds , @xmath173 thermal photons to the cavity .
this equivalence is only in terms of total noise power emitted , as spectrally the noise is emitted over the @xmath170 bandwidth of the ppt resonance , not the entire optical cavity resonance as discussed above .
up to this point in the analysis of the ppt , the discussion has been kept as general as possible .
such a system is , however , interesting only insofar as it is realizable , and we attempt here to establish the practicality of a ppt .
building upon recent experimental @xcite and theoretical work @xcite , we provide the design of an optomechanical crystal formed in a silicon microchip that can realize a ppt system with high phonon - photon translation efficiency .
as previously mentioned , optomechanical crystals @xcite are engineered structures in which phonons and photons may be independently routed and their interactions controlled . in order to create a suitable omc structure for the implementation of a ppt device one looks to a crystal lattice providing simultaneous phononic and photonic bandgaps for the guiding and co - localization of phonons and photons @xcite .
we have recently proposed @xcite such an omc system , formed from a silicon - on - insulator wafer and consisting of a patterned thin membrane of silicon .
the proposed `` snowflake '' crystal lattice supports a phononic bandgap in the @xmath174-@xmath157 ghz mechanical frequency band and a photonic bandgap in the @xmath175 nm optical wavelength band .
this quasi-2d crystal structure was also shown to support low - loss photonic and phononic waveguides , optical resonances with radiation - limited @xmath176 co - localized with mechanical resonances of frequency @xmath177 , and a single quanta optomechanical interaction rate of @xmath178 . in this section
we design an example omc implementation of a ppt in the silicon snowflake crystal .
we limit ourselves here to a two - dimensional ( 2d ) crystal involving only 2d maxwell s equations for transverse - electric ( te ) polarized optical waves and in - plane elastic deformations of an infinitely thick slab .
this simplifies the analysis and avoids some of the technical challenges related to achieving high optical @xmath4s in a quasi-2d thin film structure , challenges which have already been studied and met elsewhere @xcite . of the l2 cavity photonic resonance at @xmath179 .
( b - e ) the line defect optical waveguide structure , described in the text , which couples to the l2 cavity photonic resonance . the numerically simulated ( c ) acoustic bandstructure and ( d ) optical bandstructure of the line defect optical waveguide with @xmath180 .
( e ) plot of the out - of - plane component of the magnetic field , @xmath181 , of the guided optical mode at @xmath182-point of the bandstructure .
( f ) the displacement field @xmath183 of the l2 cavity phononic resonance at @xmath184 .
( g - j ) the line defect acoustic waveguide structure , described in the text , which couples to the l2 cavity phononic resonance .
the numerically simulated ( h ) acoustic bandstructure and ( i ) photonic bandstructure of the line defect acoustic waveguide with @xmath185 .
( j ) the magnitude of the mechanical displacement field , @xmath183 , of the guided acoustic mode at the @xmath182-point of the bandstructure .
calculations of the acoustic waveguide bandstructure are done using an fem model @xcite , while for the optical waveguide simulations , a plane - wave - expansion method was utilized @xcite .
[ fig : cavities_and_waveguides ] ] the snowflake crystal consists of a hexagonal lattice of snowflake - shaped holes patterned into silicon as shown in figure [ fig : cavities_and_waveguides ] . the snowflake lattice used here
is characterized by a lattice constant @xmath186 , snowflake radius @xmath187 , and width @xmath188 .
it possesses a full phononic bandgap from @xmath189 to @xmath190 ghz and a photonic pseudo - bandgap for te optical waves from @xmath191 to @xmath192 thz .
defects in the crystal ( features breaking the discrete translational symmetry of the underlying lattice ) can support resonances with frequencies within the optical and mechanical bandgaps , leading to highly localized , strongly interacting resonances @xcite . as an example , by filling an adjacent pair of snowflake - shaped holes in the crystal , a so - called `` l2 '' cavity is formed which supports a localized photonic resonance at a frequency @xmath179 and a phononic resonance at a frequency @xmath184 .
the defect cavity structure and finite - element - method ( fem ) @xcite simulated field profiles of these co - localized modes are shown in figures [ fig : cavities_and_waveguides](a ) and [ fig : cavities_and_waveguides](f ) .
the l2 cavity forms the basis of a more complex double - optical - mode cavity structure with the desired symmetry properties for efficient phonon - to - photon translation . by placing two separate l2 cavities close to one another , at sufficiently small separations , even and odd optical and mechanical super - modes form with split mechanical and optical resonant frequencies . we choose to displace the cavities from each other in the @xmath193 direction ,
as shown in full system diagram of figure [ fig : full_system_only ] .
these super - modes of the coupled cavities are characterized with respect to their vector symmetry about the mirror symmetry transformation @xmath194 .
we denote these super - modes @xmath195 and @xmath196 , where ` @xmath197 ' denotes symmetric and ` @xmath198 ' denotes anti - symmetric symmetry with respect to @xmath199 .
the supermodes can be written approximately as , e_= q_= , where the subscripts @xmath200 label the individual cavity fields . the cavity separation ( 14 rows ) shown in figure [ fig : full_system_only ]
is chosen such that the optical supermode frequency splitting is very nearly identical to the mechanical mode frequency of @xmath184 , as ascertained by fem simulations .
we focus here only on the mechanical mode of odd vector symmetry , @xmath201 , since as we will show below , this is the mode which cross - couples the two optical super - modes to each other .
optomechanical coupling ( or acousto - optic scattering ) arises from the coupling of optical cavity modes under deformations in the geometric structure . in the canonical form of radiation pressure ,
a mechanical deformation in the cavity induces a shift in the resonance frequency of a given cavity mode .
the coefficient describing the cavity mode dispersion with mechanical displacement also quantifies the strength of the radiation pressure force that photons in the cavity mode exert back on the mechanical structure .
more generally , mechanical deformations may couple one optical cavity mode to another .
the self - coupling and inter - modal couplings caused by a mechanical deformation are modeled by the position dependent interaction rates @xmath202 and @xmath203 , respectively , in the hamiltonian of eqn .
( [ eqn : full_total_hamiltonian ] ) .
both types of deformation - dependent optomechanical couplings may be calculated to first order using a variant of the feynman - hellmann perturbation theory , the johnson perturbation theory @xcite , which takes into account moving boundaries in electromagnetic cavities and has been used successfully in the past to model optomechanical crystal cavities @xcite .
the hamiltonian , given to first - order by , & = & _ i _ i + + _ i , j g_i , j ( + ) , is then a generalization of that shown previously in equation ( [ eqn : total_hamiltonian ] ) , with g_i , j & = & , [ eqn : gij ] where @xmath204 , @xmath205 , @xmath206 and @xmath207 are the optical mode electric field , optical mode displacement field , mechanical mode displacement field , and the thickness of the slab or thin - film .
for convenience , in what follows we denote the overlap integral in equation ( [ eqn : gij ] ) as @xmath208 .
since the optical supermodes are nearly degenerate , we replace @xmath209 with either @xmath210 or @xmath211 with little error , and following the notation used previously denote cross - modal coupling as @xmath212 .
for the optical and mechanical modes of a single l2 cavity ( see fig .
[ fig : cavities_and_waveguides](a ) and fig .
[ fig : cavities_and_waveguides](f ) ) the optomechanical self - coupling is numerically calculated using ( [ eqn : gij ] ) to be @xmath213 . then using the symmetry selection rules in the overlap integrals , the only coupling terms involving the @xmath201 mode are the cross - coupling terms @xmath214 .
these terms are calculated to be @xmath215 to good approximation .
for the supermodes of our double l2 cavity system this yields a cross - coupling rate of @xmath216 . a line - defect on an optomechanical crystal acts as a waveguide for light and sound @xcite . in principle
then , the same waveguide may be used to shuttle both the photons and phonons around on an omc microchip . however , due to the different properties of optical and acoustic excitations , in particular , their typically disparate quality factors ( roughly @xmath217 for @xmath218 thz photons and @xmath219 for ghz phonons in silicon ) , the cavity loading requirements may be different for photons and phonons .
as such , it is more convenient to implement two physically separate sets of waveguides , one for optics and the other for mechanics . our chosen line defect for optical waveguiding , shown in fig .
[ fig : cavities_and_waveguides](b ) , consists of a row of removed holes , with the rows above and below shifted toward one another by @xmath220 such that the distance between the centers of the snowflakes across the line defect is @xmath221 .
simulations of this line defect waveguide show that there are no acoustic waveguide bands resonant with the localized mechanical l2 cavity modes of interest ( see fig .
[ fig : cavities_and_waveguides](c ) ) , and therefore that this waveguide will not load the mechanical part of the l2 cavity system .
optically , this line defect has a single optical band crossing the frequencies of the localized optical l2 cavity modes of interest ( see fig .
[ fig : cavities_and_waveguides](d ) ) , providing the single - mode optical waveguiding required for the ppt .
the line defect used for the acoustic waveguide consists of a row of holes which have been reduced in size , as shown in fig .
[ fig : cavities_and_waveguides](g ) ) . by shrinking the size of the central row of snowflake holes by @xmath222 , a single - mode acoustic waveguide is formed which spectrally overlaps the localized mechanical resonances of the l2 cavity ( see fig . [
fig : cavities_and_waveguides](h ) ) , and allows for mechanical coupling to the ppt .
this same line defect waveguide is reflective for optical photons in the band of localized optical resonances of the l2 cavity ( see fig . [
fig : cavities_and_waveguides](i ) ) , and thus is isolated from the optical part of the ppt .
below we discuss how both the optical and mechanical waveguides may be used to load the ppt resonant cavity . ) and ( c ) phononic ( @xmath223 ) poynting vectors of the lower cavity - waveguide structure , illustrating the selective optical loading of the lower waveguide and the selective mechanical loading of the upper waveguide on the cavity.[fig : full_system_only],width=566 ] by bringing the optical waveguide near the l2 cavity , the optical cavity resonance is evanescently coupled to the guided modes of the line - defect , as shown in fig .
[ fig : full_system_only](b ) .
control over this coupling rate is achieved at a coarse level by changing the distance ( number of unit cells ) between the cavity and waveguide .
for the structure considered here , a gap of 8 rows is sufficient to achieve a coupling rate @xmath224 in the desired range .
a fine tuning of the coupling rate is accomplished by adjusting the waveguide width parameter , with a value of @xmath225 resulting in a loaded optical cavity @xmath4-factor of @xmath226 ( the corresponding external waveguide coupling rate is @xmath227 ) . considering that intrinsic optical @xmath4-factors as high as @xmath228
have been achieved in microfabricated thin - film silicon photonic crystal cavities similar to the sort studied here @xcite , the calculated optical waveguide loading should put such a cavity structure well into the over - coupled regime ( @xmath229 ) .
a short section in which @xmath220 is tapered is used to close off the waveguide on one side .
the same design procedure for the acoustic waveguide results in an evanescently coupled waveguide at a distance of only one row from the l2 cavity .
since the acoustic line - defect waveguide does not support bloch modes at the optical cavity frequency , no additional loss is calculated for the optical cavity resonance . in this geometry the mechanical cavity loading
is calculated to be @xmath230 , corresponding to an extrinsic coupling rate @xmath231 .
taking @xmath232 as an achievable intrinsic mechanical @xmath4-factor ( we need a citation to the berkeley group here for si ghz resonators ) , such a loading also puts the mechanical system in the over - coupling regime ( @xmath233 )
. simulations of the above cavity - waveguide couplings are performed using fem @xcite with absorbing boundaries at the ends of the waveguide .
the resulting time - average electromagnetic poynting vector @xmath234 of the optical field leaking from the l2 optical cavity resonance is plotted in fig .
[ fig : full_system_only](b ) , while the mechanical poynting vector @xmath235 ( @xmath236 is the velocity field , and @xmath237 the stress tensor ) of the acoustic waves radiating from the mechanical mode of the l2 cavity is shown in fig .
[ fig : full_system_only](c ) .
it is readily apparent from these two plots that the coupling of the two different waveguides to the l2 cavity act as desired ; the acoustic radiation is coupled only to the phononic waveguide , and the optical radiation is coupled only to the photonic waveguide . in order to individually address and
out - couple from the even and odd symmetry cavity resonances of double - l2-cavity structure used to form the ppt , a pair of waveguides is used for each of the optical and mechanical couplings . as shown in the overall ppt design of figure [ fig : full_system_only](a ) ,
each of the l2 cavities has an acoustic and an optical waveguide coupled to them .
excitation of a pair of waveguides either in or out of phase would thus allow for coupling to the symmetric or anti - symmetric supermodes , respectively , of the double - l2-cavity .
similarly , spatial filtering ( via an integrated directional coupler or waveguide filter for instance ) of the output of a pair of waveguides would allow for the selective read - out of either the symmetric or anti - symmetric cavity modes .
one could in principle utilize spectral filtering to perform the selective mode coupling ; however , with the narrowband nature of the optical and mechanical supermode splittings , spatially independent channels of excitation and read - out may be a preferred option . in summary , the omc ppt as designed couples the symmetric and anti - symmetric optical modes of a double - l2-cavity system via a co - localized anti - symmetric mechanical resonance at frequency @xmath184 . in the notation of section [ sec : analysis ] , the lower frequency symmetric optical mode is the pump mode ( cavity mode @xmath50 ) , the anti - symmetric mode is the signal mode or cavity mode @xmath80 , and the anti - symmetric mechanical resonance is phonon mode @xmath1 .
both optical modes are designed to have a resonant frequency in the near - ir around a frequency of @xmath218 thz , with a frequency splitting engineered to be equal to the mechanical frequency , @xmath238 ghz . the numerically calculated waveguide and optomechanical coupling rates for this system are @xmath239 , with the required number of intracavity pump photons for optimum operation ( @xmath240 ) of such a ppt estimated to be only @xmath241 ( assuming minimal intrinsic losses and @xmath242 , @xmath243 ) .
at the simplest level , the extremely narrow optical response of the ppt , as shown in figure [ fig : scattering_matrix ] , provides the opportunity for design and fabrication of filters with mhz - scale linewidths in the optical domain . by comparison ,
a purely passive optical design would require optical cavities with quality factors of @xmath244 .
more generally , such a scheme demonstrates a promising aspect of optomechanics in the realm of optical information processing . in this section
three examples applications of the ppt are studied in detail . the first two , optical delay lines and wavelength converters are further examples of optical information processing that is of considerable interest in both classical and quantum information processing . the last application , using the ppt to provide optical `` flying qubit '' capability to superconducting microwave quantum systems , is an example of how optomechanics can have a fundamental role in hybrid quantum system engineering .
times slower than photons .
optical filtering is provided by the narrow resonance bandwidth of the mechanical component of the ppt . ] efficient reversible conversion between traveling photons and phonons can be used to realize an optical delay line , as shown schematically in figure [ fig : delay_line ] . in this geometry ,
the acoustic waveguide used to extract phonons from the ppt is terminated abruptly , forming an effective acoustic wave mirror which reflects phonons back toward the ppt after some propagation distance and delay .
resonant photons sent into the optical port of the ppt , will then re - emerge , reflected and delayed by twice the length of the acoustic waveguide .
the delay line functionality comes from the inherent slowness of acoustic waves in comparison to electromagnetic waves ( roughly a factor of @xmath245 for waves in silicon ) . for a similar reason , electro - acoustic piezoelectric structures
are used to create the chip - scale rf - microwave filters found in many compact wireless communication devices @xcite .
the usefulness and bounds on the main characteristics of a ppt - based delay line , i.e. , the total delay possible and the delay - bandwidth product , may be simply estimated without referring to a particular implementation of the system .
the maximum possible delay is given by the lifetime of an excitation on the mechanical side of the system ( cavity and waveguide ) , and is given by @xmath246 , which is limited by material properties of the mechanical system .
the bandwidth of ppt conversion is given by eqn .
( [ eqn : gamma_transfer ] ) , and is the total loss rate seen by the mechanical resonance , @xmath170 .
as such the delay - bandwidth product can at most be ~ , which is approximately twice the acoustic waveguide to mechanical resonance over - coupling ratio @xmath247 in the ppt .
manipulation of the acoustic waves within the delay waveguide itself , before conversion back to optics , may also be envisioned , enabling existing phononic information processing capabilities @xcite to be applied to optical signals . and
@xmath248 are coupled to the same mechanical resonance @xmath84 .
the pump cavities for each system are omitted . ]
figure [ fig : ppt_wavelength_converter](a ) shows the schematic for a ppt based device that can act simultaneously as a narrowband filter , a delay line and a wavelength converter .
it consists of connecting serially by a common acoustic waveguide , two ppt devices operating at different optical but matched mechanical frequencies .
interestingly , if the only goal is to perform photon - to - photon wavelength conversion , one can omit the connecting acoustic waveguide . by placing the two ppts `` on top of each other '' , such that the optical cavities on both ppts are coupled to the same mechanical resonance , as shown in figure [ fig : ppt_wavelength_converter](b ) , optical wavelength conversion can be accomplished .
such a ppt geometry could be realized by either having photonic cavities with multiple modes , or using two photonic cavities coupled to the same mechanical mode .
such photon - to - photon conversion could even be taken to an extreme , allowing , for instance , optical - to - microwave wavelength conversion if one of the photonic cavities is a microwave cavity .
for the simplified wavelength conversion system of fig .
[ fig : ppt_wavelength_converter](b ) , the ppt matching condition of equation ( [ eqn : matching_condition ] ) and noise analysis of section [ ss : noise ] carry over with only minor adjustments .
for the simplified wavelength conversion system the thermal noise is now split between the two optical channels ( @xmath249 and @xmath248 of fig .
[ fig : ppt_wavelength_converter](b ) ) , while the spontaneous emission noise in the system is approximately doubled ( for similar optical cavities ) due to the two uncorrelated spontaneous emission processes occurring from the optical pumping of each individual cavity . in a single element ppt ,
the optimal @xmath114 matches the pure mechanical damping of the mechanical resonance ( @xmath115 ) to the induced optomechanical loading of the mechanical resonance ( @xmath250 ) by the optical cavity .
the matching condition for the simplified wavelength converter now must balance a mechanical resonance coupled on one side to an optical cavity with induced mechanical loading rate @xmath251 and on the other side to a second optical cavity with induced loading rate @xmath252 .
as such , assuming that @xmath253 we arrive at the _ photon - photon converter _ matching condition , = .
the optomechanical system as described would act as a quantum - limited optomechanical wavelength converter .
finally , we note that this particular implementation of the wavelength converter could also function in a wider array of optomechanical platforms since there is no longer a need for phononic waveguides .
two of the key requirements for a viable platform for quantum computation are the ability to do storage and communication of quantum information . for the case of superconducting phase qubits , promising theoretical proposals to provide such functionality
have involved electrical @xcite and mechanical resonators @xcite .
experimentally , an electromagnetic resonator quantum bus was demonstrated by sillanp et al .
@xcite in 2007 , while more recently oconnell et al .
@xcite demonstrated the strong coupling of a mechanical resonator to a superconducting qubit .
circuit qed ( cqed ) to date remains limited by the lack of a true long range state transfer mechanism , one which is readily available for the case of quantum - optical qubits , in the form of optical fibers and free - space links . using the ppt system , one
potentially could implement a version of the quantum state transfer protocol of cirac et al .
@xcite allowing for the high fidelity transfer of states between optical and superconducting qubits .
such a system would satisfy one of the original goals of hybrid quantum system @xcite by interfacing a quantum optical and solid - state qubit . by connecting the phononic waveguide of a ppt to a piezoelectric resonator strongly coupled to a superconducting qubit @xcite , and connecting the ppt at its optical end to an optical cavity qed system , as shown in figure [ fig : ciraczoller ] , the phonon - photon translator could be used as an intermediary in a state transfer protocol among two energy - disparate quantum systems . the optical system @xmath254 is composed of a fabry - perot cavity containing a three - level atom system in a @xmath255 configuration .
as shown in @xcite , in the correct limit , this system may be modeled with effective jaynes - cummings ( jc ) hamiltonian with a rabi frequency @xmath256 controlled by another beam . for the superconducting system @xmath257 , a mechanical resonance
is coupled to a phase qubit , with a bias current used to change the frequency of the resultant two - level system , which effectively changes the coupling rate between the qubit and mechanical resonator .
this leads again to a system with an externally controllable rabi frequency of @xmath258 .
the hamiltonian of each subsystem is then given by _
j = g_j(t ) e^-i _ j(t ) _ j ^_j + g_j(t ) e^+i _ j(t ) ^_j _ j(j = a , b ) , where @xmath259 are the annihilation operators of the photonic or phononic resonances external to the ppt , and @xmath260 are the level lowering operators for the respective qubits to which they are coupled .
each cavity mode , with annihilation operators @xmath259 , is coupled to its respective waveguide with a loss rate @xmath261 . to characterize the ppt ,
the intrinsic losses in these cavities are ignored .
additionally , the phononic waveguide is assumed to be loss - less , though losses may be taken into account through the minor readjustment of the scattering parameters studied in section [ app : waveguide_loss ] . using the input - output boundary conditions @xcite the frequency domain expression for the noise input into systems @xmath262 and @xmath263
are found to be , @xmath264 where @xmath265 and @xmath266 represent the noise being coupled into the system from the third input of each circulator in figure [ fig : ciraczoller ] . to convert this equation to the time - domain , the convolution between various operators and scattering matrix elements must be taken .
if the photon pulse is of sufficiently large temporal width , i.e. , with bandwidth less than the bandwidth of the ppt , the frequency dependence of each scattering matrix element can be removed , replacing it with its extremal value assuming that the system is operating at resonance @xmath267 .
this requires that the coupling rates @xmath268 should change slowly relative to the response of the ppt . under this condition ,
the input - output relations in the time - domain are then @xmath269 in modeling the noise of the system , it is assumed that the optical noise inputs are in the vacuum state , and that the phononic noise is thermal with thermal phonon occupation numbers @xmath270 and @xmath69 for @xmath271 and @xmath272 , respectively , where the ppt phonon spontaneous emission noise is combined with the intrinsic thermal bath coupling of the mechanical mode in @xmath69 as described in section [ ss : spont_emission ] . using standard operational methods of quantum stochastic differential equations @xcite , the master equation describing the evolution of systems @xmath262 and @xmath263
is found to be , @xmath273 + \frac{\gamma_a}{2}\mathcal{l}_{i , a}\rho + \frac{\gamma_b}{2}\mathcal{l}_{i , b}\rho \nonumber\\ & & + \frac{\gamma_b}{2}\left(|s_{22}|^2 \bar{n } + |n_{22}|^2 \bar{n}^\prime \right)\mathcal{l}_{t , b}\rho\nonumber\\ & & + \sqrt{\gamma_a\gamma_b}|s_{21}| \left ( [ c_b^\dagger , c_a \rho ] + [ \rho c_a^\dagger , c_b]\right)\label{eqn : ab_mastereqn}\end{aligned}\ ] ] with the liouvillians @xmath274 and @xmath275 are given by , @xmath276 the final term in the master equation ( [ eqn : ab_mastereqn ] ) is the cascading term @xcite which gives rise to the unidirectional coupling between the systems . for the ppt , parameters typical to an omc structure such as the one with scattering matrices plotted in fig .
[ fig : scattering_matrix ] are used , @xmath277 .
equation ( [ eqn : go_mod ] ) can be used to find the optimal matching optomechanical coupling rate , which for the assumed ppt parameters is @xmath278 . at this optomechanical coupling rate , the resonant noise and scattering matrix parameters of the ppt
are @xmath279 , with a spontaneous emission noise equivalent occupation number of @xmath280 .
assuming that the ppt is cooled to the same cryogenic temperature of the superconducting qubit system ( @xmath281 mk ) , the thermal bath component of the effective thermal occupancy of the ppt mechanical resonance can be neglected , and @xmath282 .
the exact functional form of the @xmath268 are found through numerical optimization .
this was done by taking the pulse - shape to be a step smoothed by a sinusoidal function with a rise ( fall ) time of @xmath283 ( @xmath284 ) .
optimization on the state transfer fidelity for an ideal ppt ( @xmath285 ) and for circulators running in the direction shown in fig .
[ fig : ciraczoller ] ( qubit transfer from optical to superconducting system ) leads to rise and fall times of @xmath286 and @xmath287 , which are within the modeled ppt s @xmath288 bandwidth . putting this all together , in fig .
[ fig : fidelity_plot ] we plot estimates of the fidelity of the quantum state transfer between system @xmath262 and @xmath263 via the connecting ppt .
the definition of fidelity used to calculate the state transfer efficiency is @xmath289 , where @xmath290 represent respectively the ground @xmath291 , excited @xmath292 , and @xmath293 states of the atomic and superconducting two - level systems . under these conditions , and
considering an ideal ppt ( @xmath285 ) , states are transfered with fidelities @xmath294 , @xmath295 and @xmath296 .
taking into account the actual scattering and noise matrix values given above for the ppt , and accounting for the spontaneous emission noise of the ppt we find that the fidelities are reduced to @xmath297 , @xmath298 and @xmath299 .
the inverse system , with circulators turning the opposite direction to transfer qubits from the superconducting to optical system was also studied , for which the same input pulses only time reversed , and yield fidelities @xmath300 , @xmath301 and @xmath302 .
we have introduced the concept and design of a traveling phonon - photon translator .
we have shown that with a realistic set of parameters and the use of existing silicon optomechanical crystal technology , efficient and reversible conversion between phonons and photons should be possible . by characterizing the noise processes experienced by such a device , both classically and quantum mechanically , we have shown the utility of traveling phonon - photon translation to important problems in both classical optical communication and quantum information processing .
the authors wish to thank darrick chang , thiago alegre , matt eichenfield , klemens hammerer , and peter zoller for useful discussions .
this work was supported by the darpa / mto orchid program through a grant from afosr , and the nsf through emt grant no . 0622246 and cian grant no .
asn acknowledges support through nserc of canada .
10 a. d. oconnell , m. hofheinz , m. ansmann , radoslaw c. bialczak , m. lenander , erik lucero , m. neeley , d. sank , h. wang , m. weides , j. wenner , john m. martinis , and a. n. cleland
. quantum ground state and single - phonon control of a mechanical resonator .
, 464(7289):697703 , april 2010 .
masaya notomi , akihiko shinya , satoshi mitsugi , goh kira , eiichi kuramochi , and takasumi tanabe . optical bistable switching action of si high - q photonic - crystal nanocavities . ,
13(7):26782687 , april 2005 .
j. v. sanchez - perez , d. caballero , r. mrtinez - sala , c. rubio , j. sanchez - dehesa , f. meseguer , j. llinares , and f. glvez .
sound attenuation by a two - dimensional array of rigid cylinders . , 80(24):5325 , june 1998 .
j. o. vasseur , a .- c .
hladky - hennion , b. djafari - rouhani , f. duval , b. dubus , y. pennec , and p. a. deymier .
waveguiding in two - dimensional piezoelectric phononic crystal plates . , 101(11):1149046 , june 2007 .
chang , j .- c .
shieh , and w .- p .
design and fabrication of 2d phononic crystals in surface acoustic wave micro devices . in _
micro electro mechanical systems , 2006 .
mems 2006 istanbul .
19th ieee international conference on _ , pages 686689 , 2006 .
saeed mohammadi , ali asghar eftekhar , abdelkrim khelif , william d. hunt , and ali adibi .
evidence of large high frequency complete phononic band gaps in silicon phononic crystal plates .
, 92(22):2219053 , june 2008 .
matt eichenfield , jasper chan , amir h. safavi - naeini , kerry j. vahala , and oskar painter . modeling dispersive coupling and losses of localized optical and mechanical modes in optomechanical crystals . , 17(22):2007820098 , october 2009 .
y. pennec , b. djafari rouhani , e. h. el boudouti , c. li , y. el hassouani , j. o. vasseur , n. papanikolaou , s. benchabane , v. laude , and a. martinez
. simultaneous existence of phononic and photonic band gaps in periodic crystal slabs .
, 18(13):1430114310 , june 2010 .
l. ju , s. gras , c. zhao , j. degallaix , and d.g .
multiple modes contributions to parametric instabilities in advanced laser interferometer gravitational wave detectors . , 354(5 - 6):360365 , june 2006 .
paul e. barclay , kartik srinivasan , and oskar painter .
design of photonic crystal waveguides for evanescent coupling to optical fiber tapers and integration with high - q cavities .
, 20(11):22742284 , november 2003 .
eiichi kuramochi , masaya notomi , satoshi mitsugi , akihiko shinya , takasumi tanabe , and toshifumi watanabe .
ultrahigh - q photonic crystal nanocavities realized by the local width modulation of a line defect .
, 88(4):0411123 , january 2006 .
steven g. johnson , m. ibanescu , m. a. skorobogatiy , o. weisberg , j. d. joannopoulos , and y. fink .
perturbation theory for maxwell s equations with shifting material boundaries .
, 65(6):066611 , june 2002 .
yasushi takahashi , yoshinori tanaka , hiroyuki hagino , tomoyuki sugiya , yoshiya sato , takashi asano , and susumu noda .
design and demonstration of high - q photonic heterostructure nanocavities suitable for integration . , 17(20):1809318102 ,
september 2009 . | in this article we describe a general optomechanical system for converting photons to phonons in an efficient , and reversible manner .
we analyze classically and quantum mechanically the conversion process and proceed to a more concrete description of a phonon - photon translator formed from coupled photonic and phononic crystal planar circuits .
applications of the phonon - photon translator to rf - microwave photonics and circuit qed , including proposals utilizing this system for optical wavelength conversion , long - lived quantum memory and state transfer from optical to superconducting qubits are considered . |
directional solidification of dilute alloys gives rise to complex out - of - equilibrium growth patterns .
the control of these patterns is a central issue in materials science @xcite and raises fundamental problems in nonlinear physics .
the basic phenomenon in the field is the bifurcation from a planar to a digitate growth front , which occurs when the solidification rate @xmath0 exceeds a critical value @xmath1 , where @xmath2 is the applied thermal gradient , @xmath3 is the solute diffusion coefficient in the liquid and @xmath4 is the thermal gap of the alloy @xcite .
the morphology of the fingers above the critical point evolves from rounded cells at @xmath5 to dendrites ( parabolic tip and sidebranches ) at @xmath6 @xcite .
the dominant factors in the process are the diffusion of the chemical species in the liquid , and the resistance of the solid - liquid interface to deformation , which is determined by @xmath2 and the physical properties of the interface itself , namely , its surface tension @xmath7 and kinetic coefficient @xmath8 , where @xmath9 is the kinetic undercooling . while the value of @xmath10 is approximately independent of @xmath7 and @xmath11
, the characteristics of the cellular or dendritic patterns at @xmath12 crucially depend on these properties , especially , on their anisotropy @xcite .
a fundamental distinction must be made between nonfaceted and faceted alloys , the latter being the alloys , in which @xmath13 and/or @xmath14 have cusp singularities for some orientations of * n * , where * n * is the normal to the interface referred to the crystalline axes .
the distinction between two - dimensional ( or thin ) and three - dimensional ( or bulk ) solidification is also important .
this article reports the results of an experimental investigation of pattern formation during thin directional solidification ( t - ds ) in a substance forming faceted ( monoclinic ) crystals , namely , impure biphenyl .
the geometry of the experiments is specified in figure [ setupf ] .
we focus on the growth patterns , called weakly faceted ( wf ) patterns , that bring into play only high - mobility facets . under usual experimental conditions ( _ i.e. _ far from any roughening transition ) ,
facet mobility is controlled by the motion of one - molecule - thick growth sk.jpg emitted from certain sites ( intersections with crystal dislocations , contacts with other crystals ) @xcite .
a given facet can have a high , or a low mobility depending on whether , or not , it contains such step sources .
the high - to - low - mobility transitions of a facet during growth , if any , are due to step sources entering or leaving the facet , and are quite sharp , and thus easily identified on a macroscopic scale @xcite . : thermal - gradient and growth axis .
@xmath15 : normal to sample plane and direction of observation .
@xmath16 : overall direction of the growth front.,width=264 ] interest in the theory of thin weakly faceted ( wf ) growth was aroused by experiments by maurer _
_ showing that the facet length of free - growth dendrites of nh@xmath17br followed the same @xmath18 scaling law , where @xmath0 is the dendrite tip growth rate , as the tip radius of nonfaceted dendrites @xcite .
ben amar and pomeau explained this finding by establishing analytically that the whole morphology of steady free - growth dendrites obeys a @xmath18 scaling law @xcite using a purely capillary ( @xmath19 ) 2d model of weak faceting at low undercooling @xmath20 of the liquid .
these authors then showed that introducing standard facet kinetics ( @xmath21 , where @xmath22 ) in the model should not alter these conclusions .
adda bedia and hakim @xcite gave approximate analytical solutions for free - growth dendrites with capillary facets .
recently , debierre and coworkers revisited this problem using phase - field numerical simulations @xcite , and extended the validity of the @xmath18 scaling law to arbitrary @xmath20 and capillary - anisotropy coefficients .
concerning directional solidification , the main theoretical contributions so far are two analytical studies , one dealing with the cellular transition in the particular case when a high - mobility facet is perpendicular to the growth direction by bowley _
_ @xcite and caroli _ et al . _
@xcite , and the other dealing with steady wf patterns at @xmath12 in crystals with two facets at @xmath23 from the growth axis by adda bedia and ben amar @xcite . among the numerous problems left unsolved today , most authors singled out the question of the respective roles of capillary and kinetic anisotropies in wf growth . to tackle
this problem is a current challenge for phase - field numerical simulations @xcite .
with regard to experimental investigations capable of casting light on the dynamics of thin wf patterns , we are aware of studies dealing with mesophase systems @xcite , but none of dealing with `` solid '' crystals .
a reason for this is the frequent occurrence of large - amplitude plastic deformations called growth - induced deformation ( gid ) thereafter breaking up faceted solid crystals into a multitude of small grains during t - ds .
the origin of gid , and the methods of keeping it from happening , if any , are still unclear .
fabietti and trivedi @xcite studied gid during t - ds in impure naphthalene , which has the same crystallographic structure as biphenyl , but their observations were inconclusive as regards the possible existence of deformation - free thin cellular patterns . in this article , we report that gid did not occur during thin free growth ( t - fg ) of biphenyl crystals indicating that gid basically is a thermal - stress effect generated by the externally applied thermal gradient .
most significantly , we found that gid was also lacking during t - ds when the ( deformation - free ) growth front was cellular . by triggering the cellular instability at an early stage of t - ds , it was thus possible to grow deformation - free steady wf patterns in impure biphenyl .
we present a detailed study of the spatiotemporal dynamics of gid in t - ds samples with single - crystal seeds .
this study reveals that the first stage of gid is a long - range process , which occurs only in large - width crystals , and not in the narrow cells of the cellular patterns .
finally , we report a first investigation of thin wf patterns in deformation - free biphenyl crystals for various orientations of the crystal with respect to the growth direction .
biphenyl ( @xmath24 ) is a transparent substance , which crystallizes into a biaxial birefringent monoclinic phase at @xmath25c .
the point group of the crystal has a twofold axis @xmath26 and a mirror plane normal to @xmath26 that contains the other two lattice translations @xmath27 and @xmath28 ( fig .
[ biphcryst ] ) .
the crystalline parameters are @xmath29 , @xmath30 and @xmath31 nm , and the angle between @xmath27 and @xmath28 is 95.1 at room temperature @xcite .
biphenyl crystals have a perfect ( 001 ) cleavage , and various glide systems involving dislocations with burgers vectors [ 100 ] and [ 010 ] , but not [ 001 ] @xcite .
we show below that , during melt growth , biphenyl exhibits only \{001 } ( `` basal '' ) , and \{110 } facets .
( it does not exhibit \{100 } facets , contrary to naphthalene ) . the [ 100 ] apex angle of the basal facet , _
i.e. _ the angle between [ 110 ] and [ 1@xmath320 ] , calculated from the above data is @xmath33 .
the terminology for the orientation of the crystals employed hereafter is as follows .
crystals with ( 001 ) parallel to the sample plane are called ( 001)-oriented crystals .
their in - plane orientation is either specified by the angle @xmath34 of @xmath27 and the growth axis ( called @xmath35 , see fig .
[ setupf ] ) , or designated as ( 001)[100]- or ( 001)[010]-orientation when @xmath36 or @xmath37 , respectively .
crystals deviating from the ( 001)-orientation are called `` tilted '' , and the angle of their basal plane and the sample plane is denoted @xmath38
. crystals with their basal plane nearly perpendicular to the sample plane are called `` edge - on '' .
the observations were performed in commercial biphenyl ( fluka , 99.9 % ) unless otherwise mentioned .
a few experiments were made with biphenyl doped with 1 wgt% of camphor .
the methods of preparation and observation of the samples are explained elsewhere @xcite , and need only to be briefly outlined here .
the crucibles consist of two parallel glass plates separated by polymer spacers .
their inner dimensions are of 60 mm along @xmath35 , 8 mm ( width ) along @xmath16 , and 12 ( thickness ) along @xmath15 .
a funnel - shaped grain selector is created near the cold end of the crucibles by using spacers of an appropriate shape ( fig .
[ setupf ] ) .
the crucibles are filled with the liquid compound under controlled argon atmosphere , quenched to room temperature , and sealed .
they are then inserted into a t - fg or a t - ds setup , and observed in side view by videomicroscopy .
polarized light was used in order to take benefit of the birefringence of the crystals .
t - fg was carried out with a commercial hot stage ( instec , hs1-i ) mostly for the purpose of preparing single - crystal seeds for t - ds .
the temperature distribution in the setup is not perfectly uniform , but has a shallow depression at the center of the window of observation .
this feature can be utilized to grow single - crystal seeds .
the temperature of a polycrystalline sample is increased until all the grains melt except the one ( the future seed ) that is located at the minimum of the temperature distribution .
the melting is pursued until the seed detaches itself from the glass plates , is carried away by flows existing in the liquid , and redeposited with , generally , its basal facet closely aligned with the sample plane . the temperature
is then decreased step by step in order to make this ( 001)-oriented single crystal grow without ( or with as little as possible ) morphological instability .
it should be noted that , during this process , the crystal is a thin plate limited by two blocked ( deprived of step sources ) basal facets , which are basically not in contact with the glass walls .
some crystals had a small residual misalignment and collided with a glass wall during growth .
crystals that filled the sample without hitting the glass walls ( @xmath39 ) were selected as seeds for t - ds .
these crystals kept a uniform optical contrast during growth indicating that no noticeable gid occurred during t - fg .
the fully solidified samples were then slowly cooled to room temperature , and transferred to the t - ds setup . a weakly contrasted rectilinear
striation parallel to the direction [ 010 ] appeared at @xmath4010k during cooling , and persisted in the non - melted part of the samples after they were inserted into the t - ds setup .
however , they were not transmitted to the grown crystal during t - ds indicating that they did not pertain to the bulk of the crystal , but to the layer of matter comprised between the crystal and the glass walls that solidified during cooling to room temperature .
the crystallographic orientation of the striae and the presence of microbubbles are suggestive of slip bands left by [ 010 ] dislocations gliding in the ( 100 ) plane .
t - ds experiments were performed using a home - built stage made of two independent thermally regulated copper blocks separated by a several millimeters - wide gap , in which the solidification front sits . during this study , @xmath41 , unless otherwise mentioned .
the translation velocity of the samples is stable to within @xmath42 in the explored range ( 0.1 - 30 ) .
in addition to a standard t - ds stage , we used a new `` rotating t - ds stage '' , to be presented in a future publication , with which it is possible to rotate the sample about @xmath15 , that is , to change the in - plane orientation of the crystal during solidification .
the microstructure of the non - melted part ( seed ) of the sample at the beginning of a t - ds run is an all - important experimental factor that can be largely controlled by an appropriate design of the first stages of t - ds .
the main alternative is to include , or not , a low-@xmath0 growth through the grain selector prior to the t - ds run proper .
a few as - quenched samples were solidified without grain selection , and exhibited a strongly faceted mode of growth . as a general rule ,
as - quenched samples were solidified with grain selection , which led to a complete elimination of all the tilted grains ( see section [ gipdasq ] ) . a cellular instability was observed during t - ds of ( 001)-oriented grains under conditions , which will be specified later on .
we measured @xmath10 at @xmath43 in these grains by the method explained in ref .
three types of samples must be distinguished : fresh samples of commercial biphenyl , samples of commercial biphenyl having undergone a t - fg or t - ds solidification / melting cycle at low @xmath0 ( as is the case of samples with single - crystal seeds ) designated as `` purified '' hereafter , and samples of camphor - doped biphenyl .
we found @xmath44 in fresh undoped samples , @xmath45 in purified samples , and @xmath46 in camphor - doped samples .
in t - fg as well as t - ds , biphenyl crystals exhibited only two types of facets , namely , low - mobility \{001 } facets and high - mobility \{110 } facets .
we give experimental evidence of the respective kinetics of these facets .
figure [ biphfreeseq]a shows a spatiotemporal ( st ) diagram _i.e . _ a time series of binarized profiles of the growth front displayed in the reference frame of the sample of a ( 001)-oriented biphenyl crystal during t - fg .
the crystal first grows from a circular to a rhombus shape limited by \{110 } facets , then undergoes an impurity - driven ( mullins - sekerka ) instability , and finally settles into a steady dendritic regime with dendrites pointing in the @xmath47 and @xmath48 directions .
the \{110 } facets are linked to each other by smooth rounded regions indicating that no forbidden orientation range exists in the non - faceted parts of the solid - liquid interface .
the faceted parts of the profile remained rectilinear within the measurement uncertainty during the process .
the measured angle between [ 110 ] and [ 1@xmath320 ] facets was of @xmath49 .
the small difference between this angle , and the one deduced from crystallographic data , if significant , is attributable to differences in temperature and composition .
the slight crystallographic tilt of the \{110 } facets with respect to @xmath15 was not resolved , but a difference in contrast between the faceted and rounded part of the interface was apparent , revealing a difference ( planar versus rounded ) in 3d shape between these two regions .
the time evolution of the tip velocity @xmath0 ( figure [ biphfreeseq].b ) shows that \{110 } facets developed through a perfectly smooth process indicating that no discontinuity in the kinetic coefficient is associated with \{110 } facets .
the same features were also observed for \{110 } facets during t - ds , as will be seen in section [ stabil ] .
k was applied to a quasi - circular seed at @xmath50 .
b ) growth rates of the [ 100 ] ( rightmost ) and [ 010 ] ( uppermost ) tips as a function of time .
the thermal time lag of the t - fg stage is about 1 min .
the growth of the leftmost tip was perturbed by a dust particle.,width=264 ] detailed information about the growth kinetics of \{001 } facets was given by t - ds experiments performed without grain selection in as - quenched samples .
figure [ nucleation ] shows a deep liquid pocket due to the mutual impingement of edge - on crystals adhering to the glass walls through their nonfaceted extremities .
a sporadic nucleation of macrosk.jpg on the basal facets bordering the liquid pocket took place as @xmath20 progressively increased until the whole pocket was suddenly filled by a polycrystal through an `` explosive '' nucleation process .
the distance of the nucleation site from the growth front was of about 1.5 mm corresponding to a temperature difference of about 9k . in conclusion ,
the known absence of dislocations capable of serving as step sources for ( 001 ) facets namely , dislocations with a [ 001 ] burgers vector in biphenyl crystals is a sufficient explanation of the low mobility of these facets .
reciprocally , the observed high mobility of \{110 } facets must be attributed to dislocations with [ 100 ] or [ 010 ] burgers vector intersecting these facets .
the only acceptable alternative would be the proximity of a roughening transition , but this seems very unlikely given the large extension of these facets , and the small growth rates considered in this study .
however , the mechanisms by which \{110 } facets are fed with dislocations during growth even when the growth morphology becomes very complex , are unknown . anticipating on the observations presented below , we note that one , and probably the most important of these mechanisms is the plastic deformation generated by thermal stresses illustrated by gid .
the formation of a stratified microstructure , or equivalently of ( 001 ) twist boundaries , during growth explained in section [ gipdsingle ] may also contribute , but it is not certain that these boundaries survive the morphological transitions of the system ( see fig . [ feuillets ] below ) .
we performed t - ds experiments in a series of samples with single - crystal seeds with various in - plane orientations at values of @xmath0 ranging from 0.5 to 30 . in these samples , @xmath51 , as previously mentioned .
we observed gid processes with similar characteristics in all the experiments .
this , and the fact that gid does not occur during t - fg , may suffice to establish that gid is not due to a cellular instability , but most probably to thermal stresses .
however , to substantiate this conclusion , and because of the remarkable features of gid in our experiments , we study the geometry of this process in detail .
gid processes that took place in two different crystals oriented , one asymmetrically , and the other symmetrically , with respect to the t - ds setup are shown in figures [ gid24 ] and [ gida ] , respectively .
both processes go through the following three successive stages : stage 1 , which starts from the beginning of the solidification , and consists of the progressive amplification of a smooth contrast modulation extending along @xmath16 ( fig . [ levelplots ] ) ; stage 2 , which begins with the sudden creation of grain boundaries at positions corresponding to the extrema of the contrast modulation ; stage 3 , during which crystal - glass collisions followed by the appearance of new grains lead to a fully polycrystalline microstructure . .
the seed can be recognized from the [ 010 ] striation inherited from the post - t - fg cooling down of the sample.,width=340 ] -oriented single - crystal seed ( @xmath53 ) . @xmath54 . *
a * ) t=0 s ( start of t - ds ) .
* b * ) t=22 s. * c * ) t=33 s. * d * ) t=46 s. the two arrows indicate the appearance of grain boundaries and the first collisions with the glass plates , respectively.,width=302 ] during the t - ds run shown in fig .
thin line : @xmath55 .
broken line : @xmath56 .
thick line and dots : @xmath57 .
the curves were obtained by smoothing the data points ( dots ) after subtraction of a linear function fitted onto the background.,width=302 ] stage 1 clearly is a long - range process of plastic deformation of the growing crystal .
a full understanding of such a process under the conditions of our experiments ( strong confinement of the system and elasto - plastic anisotropy of the solid ) is notoriously beyond reach at present , but some interesting semi - quantitative remarks can be made .
various observations indicated that the distortion field existing in the crystal during stage 1 mostly consisted of rotations of the crystal lattice about the [ 100 ] axis of the crystal .
such a distortion field favors the formation of tilt boundaries ( arrays of [ 010 ] edge dislocations ) in ( 010 ) planes , in agreement with the fact that the grain boundaries appearing at the onset of stage 2 were approximately parallel to [ 100 ] in all the samples studied ( figs .
[ funnel ] and [ gidstria ] ) . figs .
[ gida ] and [ levelplots ] , on the one hand , and fig .
[ gidb ] , on the other hand , correspond to two successive runs performed in the same ( 001)[100]-oriented sample at @xmath54and @xmath58 , respectively .
the crystal length solidified during the first run was entirely re - melted before the second run so that there was no possible influence of the first gid on the second one .
we note that @xmath0 was above the cellular instability threshold during the first run , which is unimportant for our present purpose , but has interesting consequences , which will be commented later on . during these experiments ,
stage 1 clearly exhibited two characteristic lengths , namely , a wavelength @xmath59 along @xmath16 , and a solidification length @xmath60 along @xmath35 . during both
runs , @xmath59 , defined as the spacing of the extrema of grey - level plots ( fig .
[ levelplots ] ) , ranged from @xmath61 to @xmath62 .
this scatter was mostly due to the existence of a lateral gradient of unknown origin ( but probably linked to some experimental imperfection ) .
the value of @xmath60 , defined as the solidification length between the start of the experiment and the first appearance of grain boundaries , was of @xmath63 in both experiments .
larger ranges of @xmath59 and @xmath60 were found in asymmetrically oriented crystals ( figs .
[ funnel ] and [ gidstria ] ) than in the ( 001)[100]-oriented ones , but the orders of magnitude remained the same .
these observations indicate that @xmath59 and @xmath60 were essentially independent of @xmath0 , as could be expected from the fact that the plastic deformation started from the beginning of the solidification .
this suggests that these lengths are mostly determined by the geometry of the experiment .
contrary to stage 1 , stages 2 and 3 presented features , which depended on the in - plane orientation of the crystal , and on additional , ill - known , geometrical factors .
this point is illustrated in figs .
[ gida ] and [ gidb ] , which show gid processes that occurred during two successive runs at different values of @xmath0 in the same ( 001)[100]-oriented sample .
the common features of the two runs are that the gid - induced microstructure kept the mirror symmetry about @xmath64 of the initial crystal as well as the periodicity inherited from stage 1 ( in contrast with what occurred in asymmetrically oriented samples ) , and that thickness fringes appeared near the grain boundaries at , or a short time after , the onset of stage 2 .
the most apparent differences between the two runs are the additional symmetry ( twofold axis at @xmath65 ) existing in fig .
[ gida ] compared to fig .
[ gidb ] , and the fact that the thickness fringes that appear on either side of a grain boundary are divergent in fig .
[ gida ] , but convergent in fig .
thickness fringes can only arise from ultra - thin crystal wedges .
figure [ croquis ] displays a schematic 3d reconstruction of the microstructures based on this remark , and the following additional conjectures : ( i ) the onset of stage 2 occurred when the ongoing deformation made the crystal come into contact with the glass plates . at each point of contact ,
a pair of misoriented crystal wedges attached to the glass were created under the effect of external forces ( linked with capillarity , flows in the liquid , changes in the thermal field ) ; ( ii ) the initial positioning of the growing crystal with respect to the glass plates was symmetric in fig .
[ gida ] , but strongly asymmetric in fig .
regardless of the details , fig .
[ croquis ] illustrates the difference in nature between the early stages of gid , which do not involve contacts or collisions with the container walls , and later stages , which are essentially driven by such events .
the specificity of gid , in the sense given to this term here , lies in stage 1 , while collision - induced deformations similar to those occurring during stage 3 are also generated by a misaligned seed . .
the slanting edge of a grain selector appears in the lower - right hand corner of the photograph.,width=188 ] .
@xmath0=0.3 . after t - ds arrest
, the heating of the ovens was switched off for a few hours , and then put on again .
the grain boundaries are perpendicular to the [ 010 ] striation and thus parallel to [ 100 ] .
, width=264 ] the nature of the growth process following crystal - glass collisions is most clearly illustrated in figs .
[ gidb ] and [ funnel ] . without discontinuity , and thus , probably through a plastic deformation of the crystal colliding the glass , solidification continues with the growth of ( 001)-oriented thin - film crystals in contact with , or very close to the glass walls .
the extreme thinness of these crystals is revealed by a slight recoil of their solid - liquid interfaces with respect to thicker parts of the crystal attributable to their strong curvature in the direction @xmath15 ( gibbs - thomson effect ) . .
@xmath68 . , width=302 ] ( * a * ) and [ gidb ] ( * b * ) . _ f _ : thin film crystal .
continuous lines : interfaces and grain boundaries .
broken lines : zones of rapid , but continuous change in orientation .
arrows : local orientations of the axis [ 100 ] .
the sketches show a single period ( along * x * ) of the gid microstructures.,width=283 ] a spontaneous grain - growth process leading to a complete elimination of tilted grains in favor of ( 001)-oriented grains took place in as - quenched samples during the grain - selection stage of t - ds ( fig .
[ rearr ] ) .
when their size permitted it , the ( 001)-oriented grains emerging from this process underwent a gid process similar to those observed in samples with single - crystal seeds .
the gid process repeated itself with a constant amplification rate during the continuation of the growth through the funnel - shaped selector ( fig .
[ funnel ] ) .
adjacent ( 001)-oriented grains competed and often overlapped , leading to stratified microstructures ( fig .
[ feuillets ] ) . .
the slanting edge of a grain selector is visible in the lower - left hand corner of the photographs.,width=283 ] .
a ) @xmath50 ; b ) @xmath69 s ; c ) @xmath70 s and d ) @xmath71 s. the difference in orientation of the stratums is about 8 .
horizontal dimension : @xmath72.,width=226 ] the rapid development of gid prevented us from observing deformation - free wf patterns in samples with single - crystal seeds . to bring about cellular transitions in deformation - free crystals , we applied upward @xmath0-jumps to ( 001)-oriented grains at an early stage of gid in fresh undoped samples .
quite significantly , we never observed gid to appear after such a cellular transition occurred . in camphor - doped samples , cellular transitions without gid
were observed at all the explored values of @xmath0 .
cellular - instability transients occurring concurrently with gid processes are worth briefly considering as a new instance of coupling between plastic deformation in the solid and impurity - driven dynamics at the growth front @xcite .
it is known that impurities rejected by the growing edge of a plate - like crystal partly segregate at the rear of this edge @xcite . in the confined geometry of t - ds
, this effect manifests itself by an increase of the equilibrium temperature , and thus an advance of the growth front , inversely related to the crystal thickness .
such a relation between the profile @xmath73 of the growth front and the local value of the crystal thickness was observed during gid except in the thinnest regions of the front where the gibbs - thomson effect predominated .
it should be noted that this impurity - driven effect , which remains small at low @xmath0 , amplifies as @xmath0 approaches @xmath10 ( fig .
[ gida ] ) conferring an imperfect - bifurcation character to the cellular instability . in conclusion ,
the core of the gid process is the progressive amplification of a long - range modulated plastic deformation of the growing crystals ( stage 1 ) .
the subsequent stages of the process are essentially geometrical consequences of this first stage . the lack of sensitivity of stage 1 to the control parameters of the solidification rules out the possibility that this process be strongly connected with an impurity - driven dynamics .
the fact , that in our system the growing crystals that underwent gid were not in contact with the container walls makes it very likely that the first stage of gid is basically due to the thermal gradient alone more precisely , the stresses engendered by an inhomogeneous temperature field in a crystal with strong elastic and plastic anisotropies .
this stress field depends only on the geometry of the growing crystal at fixed geometry of the t - ds setup .
this is consistent with the observed insensitivity of @xmath59 and @xmath60 to @xmath0 and @xmath34 , and suggests that the lack of gid in cellular fronts is simply due to the fact that stresses and plastic deformation are not transmitted from a cell to another . like their non - faceted homologs
, the wf patterns of biphenyl exhibited a broad ( @xmath74 , typically ) range of stable spacings @xmath75 at fixed @xmath0 .
this range was bordered by a cell elimination instability at small @xmath75 and a cell splitting instability at large @xmath75 .
we observed cell elimination processes during cellular - instability transients ( fig .
[ diffinstab ] ) , and could estimate the cell elimination threshold spacing to be about 40 in undoped biphenyl in the explored @xmath0 range ( @xmath76 ) . on the other hand , we noted various modes of instability at @xmath77 , in particular , oscillations and a propagating tip splitting ( fig .
[ splitting ] ) .
-oriented grain .
time interval : 2 s ( top ) , 0.08 s ( bottom ) .
@xmath0=6.5 .,width=226 ] in their recent numerical study of the free growth of thin wf systems in a channel , gurin _ et al .
_ identified an oscillatory symmetry - broken mode of growth , and argued that this mode belongs to the same branch of states as the steady asymmetric fingers that exist in low - anisotropy nonfaceted systems @xcite . in t - ds , such fingers , if any , should appear in the form of pairs of asymmetric fingers called _ doublons _ @xcite .
we performed a few experiments at very high @xmath78 in camphor - doped biphenyl samples , and indeed observed a dendrite - to-_doublon _ transition ( fig . [ biphdoublons ] ) lending support to gurin _ et al . _ s argument .
it must be noted , however , that we have not been able to ascertain the existence of facets at the tips of theses _ doublons _ , so that the possibility of a roughening transition occurring in our system at high @xmath0 can not be entirely excluded .
-oriented grain .
@xmath0=17 . note the transient oscillations in the wake of the solitary wave .
horizontal dimension : @xmath79 .
the diagram has been contracted by a factor of 2 vertically .
, width=226,height=222 ] = 150.,width=264 ] cellular patterns were symmetric ( _ i.e. _ did not drift in the direction @xmath16 ) in ( 001)[100]-oriented grains of undoped biphenyl ( fig . [ 100cells ] ) , as could be expected since a mirror plane of the crystal structure is parallel to @xmath64 in these grains .
the @xmath75-distribution ranged from about 40 to 100 at the end of the cellular transition , and was slowly relaxing toward a uniform distribution .
it was not possible to wait for this relaxation to be complete because the lifetime of the grains of interest was limited by the competition of the adjacent grains .
we performed measurements in quiet regions of the evolving patterns assuming the quasi - steady - state condition to be valid in these regions .
we measured the cell facet length and tip radius during three different t - ds runs in ( 001)[100]-oriented crystals of undoped biphenyl by the method explained in the legend to figure [ facet_def ] .
the results are displayed in figure [ l_facet ] .
the measurement error on @xmath80 was of @xmath81 , and thus of @xmath82 , at worst , which can account for the scatter of the data .
the major trend emerging from the data is an essentially linear increase of @xmath80 with increasing @xmath75 .
the concomitant increase in @xmath83 , which is precursor to tip splitting , is much weaker .
the dependence of @xmath80 and @xmath83 on @xmath0 is undetectable .
the fraction of the front occupied by facets ( @xmath84 ) is important , and increases as @xmath75 increases .
the sidebranching threshold is thus shifted to high values of @xmath75 compared to what it would be without facets .
this effect is most clearly illustrated by the long distance separating the first sidebranches from the tips of tilted dendrites in figure [ facetperp ] below .
( 100)-oriented crystals .
undoped biphenyl .
@xmath0 = 6.5 . ,
width=264 ] a by thresholding and skeletonizing the image of a cell .
the broken line is the linear best - fit function along a presumed facet .
this function was subtracted from the relevant part of the profile yielding the set of data points displayed in the inset .
the box encloses the data points whose deviation from a smooth curve drawn through these points is smaller than their scatter .
the length of the box ( corrected for the projection factor ) was taken as a measure of the facet length .
, width=264 ] ( open symbols ) and tip radius @xmath83 ( filled symbols ) versus cell spacing @xmath75 during t - ds of ( 001)[100]-oriented crystals .
undoped biphenyl .
squares : @xmath864.7 .
triangles : @xmath866.2 .
disks : @xmath8617.,width=260 ] the grain - growth process studied in section [ gipdasq ] yielded ( 001)-oriented grains with in - plane orientations belonging to a limited interval around @xmath36 .
we used a rotating t - ds stage to grow ( 001)-oriented crystals with arbitrary @xmath34 values in camphor - doped biphenyl samples .
the uncertainty on @xmath34 was of @xmath87 .
at all @xmath34 values cell tips exhibited well - defined facets , which were however too small to permit a quantitative study .
we observed a lateral drift of the cellular patterns for all the in - plane orientation of the crystal except the ( 001)[100]- and ( 001)[010]-orientations ( fig .
[ celldoped ] and [ highmag ] ) . in the last - named orientation ,
the crystal structure is not invariant to reflection with respect to the @xmath64 plane , but the absence of drift indicates that the whole system has such an invariance .
( it should be noted that monoclinic angle of biphenyl is small ) .
in other words , the behavior of the system was practically that of a 2d system with two orthogonal symmetry axis ( [ 100 ] and [ 010 ] ) .
it is worth noting that , contrary to [ 100 ] cells , [ 010 ] cells did not exhibit tip splitting but transformed to dendrites at large spacing values ( fig .
[ tilted010dend ] ) .
-oriented crystal .
@xmath88 ; below : ( 001)[010]-oriented crystal .
@xmath89.,width=264 ] . left : [ 100 ] cell .
right : [ 010 ] cell.,width=264 ] we studied the lateral drift of the cellular patterns in grains with @xmath90 and @xmath91 . like in nonfaceted systems @xcite , the tilt angle of the direction of growth of the cell tips was an increasing function of @xmath0 and approached the tilt angle of the nearest axis of symmetry of the system ( @xmath34 or @xmath92 ) at large values of @xmath78 ( fig .
[ tilted010dend ] ) .
dendrites during t - ds of a ( 001)-oriented grain with @xmath34= 82 .
camphor - doped biphenyl . @xmath0=10 .
the tilt angle of the dendrites is @xmath93.,width=245 ] figure [ drift_cells ] shows the st diagram of a cellular - instability transient in a crystal with @xmath94 . though the shape of the cells rapidly departed from a mere sine , no drift of the structure was observed until facets appeared .
given that kinetic anisotropy controls the drift velocity during the first stages of the cellular transient @xcite , this indicates that this anisotropy is relatively weak for *
n * belonging to the basal plane of biphenyl . finally , we note that grains with a \{110 } facet nearly perpendicular to the growth axis exhibited a singular dynamics displaying a coexistence between crenellated interfaces @xcite , [ 100]-dendrites and [ 010]-dendrites during cellular - instability transients ( fig .
[ facetperp ] ) . .
undoped biphenyl .
@xmath0=6.2 . ,
width=302 ] .
camphor - doped biphenyl . @xmath0=10 .
the angle of the facet to @xmath35 is about 88.5.,width=264 ]
a thin plate - like crystal confined between two walls and submitted to directional solidification is prone to plastic deformation under the effect of thermal stresses generated by the applied thermal gradient , and interactions with the walls .
this study has shown that a deformation - free directional solidification of plate - like crystals is feasible , at least in the case of impure systems , and provides a good experimental model of the dynamics of 2d weakly faceted directional solidification .
we have presented first elements of an experimental investigation of this dynamics , including a set of preliminary quantitative data about the facet lengths of steady wf cellular patterns as a function of @xmath75 and @xmath0 .
previous experimental and theoretical studies have established that the scaling laws of nonfaceted dendritic free growth still hold in weakly faceted systems .
similarly , although in a less precise way , this study has shown that the dynamics of weakly faceted directional - solidification systems is similar to that of anisotropic , but nonfaceted systems , except , perhaps , when the growth front is nearly parallel to a facet .
theoretical studies pointed out that the length of the facets near dendrite tips , which is directly connected to the capillary and kinetic coefficients of the solid - liquid interface , is the most specific feature of wf growth morphologies .
further experimental and numerical studies of the facet lengths of steady wf cellular patterns as a function of various control , and material parameters could cast light upon basic questions such as the respective roles of capillary and kinetic anisotropies in weakly faceted growth .
99 m. asta , c. beckermann , a. karma , w. kurz , r. napolitano , m. plapp , g. purdy , m. rappaz , r. trivedi , acta mater . * 57 * , 941 ( 2009 ) .
see also w.j .
boettinger , s.r .
coriell , a.l .
greer , a. karma , w. kurz , m. rappaz , r. trivedi , acta mater . * 48 * , 43 ( 2000 ) .
rutter and b. chalmers , can .
* 31 * , 15 ( 1953 ) .
mullins and r.f .
sekerka : j. appl .
phys * 35 * , 444 ( 1964 ) .
p. kurowski , c. guthmann and s. de cheveign , phys .
a * 42 * , 7368 ( 1990 ) .
m. ben amar and y. pomeau , europhys . lett .
* 2 * , 307 ( 1986 ) .
a. barbieri , d.c .
hong and j.s .
langer : phys .
a * 35 * , 1802 ( 1987 ) .
e. a. brener , h. mller - krumbhaar and d.e .
temkin , europhys .
* 17 * , 535 ( 1992 ) .
t. ihle and h. mller - krumbhaar , phys .
e * 49 * , 2972 ( 1994 ) .
s. akamatsu , g. faivre and t. ihle , phys .
e * 51 * , 4751 ( 1995 ) .
s. akamatsu and t. ihle , phys .
e * 56 * , 4479 - 4485 ( 1997 ) .
burton , n. cabrera and f.c .
frank : phil .
london * 243a * , 299 ( 1951 ) .
chernov , contemporary phys .
, * 30 * , 251 ( 1989 ) .
j. maurer , p. bouissou , b. perrin and p. tabeling , europhysics letters * 8 * , 67 ( 1989 ) .
m. ben amar and y. pomeau , europhys . lett .
* 6 * , 609 ( 1988 ) .
m. adda bedia and v. hakim , j. phys .
i * 4 * , 383 ( 1994 ) .
m. adda bedia and m. ben amar , phys .
e * 51 * , 1268 ( 1995 ) .
debierre , a. karma , f. celestini and r. gurin , phys .
e * 68 * , 041604 ( 2003 ) .
r. gurin , j .-
debierre , k. kassner , phys .
e * 71 * , 011603 ( 2005 ) .
r. bowley , b. caroli , c. caroli , f. graner p. nozires and b. roulet , j. phys .
( paris ) , * 50 * , 1377 ( 1989 ) .
b. caroli , c. caroli and b. roulet , j. phys .
( paris ) , * 50 * , 3075 ( 1989 ) .
m. adda bedia and m. ben amar , phys .
a * 43 * , 5702 ( 1991 ) .
t. uehara , r.f .
sekerka , journal of crystal growth * 254 * , 251 ( 2003 ) .
t. brzsnyi , s. akamatsu and g. faivre , lecture notes in computational science and engineering * 32 * , 166 ( 2003 ) .
f. melo and p. oswald , phys .
lett . , * 64 * , 1381 ( 1990 ) .
t. brzsnyi , s. akamatsu and g. faivre , phys .
e * 65 * , 011702 ( 2001 ) .
t. brzsnyi and s. akamatsu , phys .
e * 66 * , 051709 ( 2002 ) .
fabietti , r. trivedi , j. crystal growth * 173 * , 503 ( 1997 ) .
j. trotter , acta cryst . , * 14 * , 1135 ( 1961 ) . a. hargreaves and s.h .
rizvi , acta cryst . * 15 * , 365 ( 1962 ) .
marwaha and b.s .
shah , cryst .
. technol .
* 27 * , 1097 ( 1992 ) .
j. mergy , g. faivre , c. guthmann and r. mellet , j. cryst .
growth * 134 * , 353 ( 1993 ) .
s. akamatsu and g. faivre , j. phys .
i * 6 * , 503 ( 1996 ) .
s. bottin - rousseau , s. akamatsu , and g. faivre , phys .
b * 66 * , 054102 ( 2002 ) . t. fujioka and r.f .
sekerka , j. cryst .
growth * 24/25 * , 84 ( 1974 ) .
b. utter , r. ragnarsson and e. bodenschatz , phys .
* 86 * , 4604 ( 2001 ) .
j. deschamps , thse de doctorat , universit aix - marseille i , 2007 . s.r .
coriell and r.f .
sekerka , j. cryst .
growth * 34 * , 157 ( 1976 ) . | we present an experimental study of thin - sample directional solidification ( t - ds ) in impure biphenyl .
the plate - like growth shape of the monoclinic biphenyl crystals includes two low - mobility ( 001 ) facets and four high - mobility \{110 } facets . upon t - ds , biphenyl plates oriented with ( 001 )
facets parallel to the sample plane can exhibit either a strong growth - induced plastic deformation ( gid ) , or deformation - free weakly faceted ( wf ) growth patterns .
we determine the respective conditions of appearance of these phenomena .
gid is shown to be a long - range thermal - stress effect , which disappears when the growth front has a cellular structure .
an early triggering of the cellular instability allowed us to avoid gid and study the dynamics of wf patterns as a function of the orientation of the crystal . |
the confirmation of the temporal variation of the fundamental constants would be the first indication of the universe expansion influence on the micro physics @xcite .
shlyakhter was the first who showed that the variation of the fundamental constants could lead to measurable consequences on the sm isotops concentrations in the ancient reactor waste @xcite .
later damur and dyson @xcite for zones 2 and 5 and also fujii @xcite for zone 10 of reactor oklo made more realistic analysis of the possible shift of fundamental constants during the last @xmath6 years based on the isotope concentrations in the rock samples of oklo core . in this investigation
the idealized maxwell spectrum of neutrons in the core was used .
the efforts to take into account more realistic spectrum of neutrons in the core were made in works @xcite .
new severe constraints on the variation of the fine structure constant have been obtained from reactor oklo analysis in work @xcite : @xmath7 we investigate here how these constraints confine the parameter of bsbm model @xcite of varying @xmath0 .
this theory combines bekenstein extension of electrodynamics @xcite with varying alpha to include gravitational effects of new scalar field @xmath8 .
it respects covariance , gauge invariance , causality and has only two free parameters : the fraction of electromagnetic energy @xmath4 in the total energy of matter including dark matter as well as the dimensional parameter @xmath3 which is having sense of characteristic length . as a result of our analysis
we get the constraints on the combination of the parameters of bsbm model .
bsbm theory @xcite is the extension of the bekenstein @xcite theory to include dynamics of the gravitational field .
total action of this theory has a form : @xmath9 where @xmath10 and @xmath11 .
a parameter @xmath12 here is definite as @xmath13 where dimensional parameter @xmath3 is having sense of characteristic length .
fine structure constant expressed via @xmath8 with the equation : @xmath14 .
varying @xmath8 we get the following equation : @xmath15 for pure radiation @xmath16 , so @xmath8 remains constant during radiation domination epoch . only in matter domination epoch changes in @xmath0
take place .
the only contribution to variation of @xmath8 come mainly from pure electrostatic or magnetostatic energy .
it is convenient to work in the following parameter : @xmath17 and according to @xcite @xmath18 and @xmath19 .
varying the metric tensor and using friedmann metric we get the following friedmann equation : @xmath20,\ ] ] and the equation for @xmath8 takes form : @xmath21 where @xmath22
. we have also energy conservation equations : @xmath23 which have solutions : @xmath24 , and @xmath25 let use critical density : @xmath26 and use also the fractions of all densities relative to critical : @xmath27 .
index @xmath28 will denote the present values of these fractions .
we use the ordinary values for these fractions at present : @xmath29 , @xmath30 , and @xmath31 is determined from the condition that the universe is flat
. then the friedmann equation takes form : @xmath32,\ ] ] and equation for @xmath8 : @xmath33 here constant @xmath34 is equal to @xmath35 . for negative @xmath4
this constant is positive and has the following dependence on the ratio of characteristic and plank lengthes : @xmath36 the result of the numerical integration of these equations is presented of fig.1 for the variation of different components of energy density with red shift @xmath37 , and on fig.2 for the variation of fine structure constant @xmath0 .
here we use the notation : @xmath38 .
we took the value of the characteristic length @xmath3 equal to @xmath39 during this analysis , and assigned the following value for the the parameter @xmath4 : @xmath40 where @xmath41 - is the fraction of energy density in the universe due to ordinary baryonic matter .
the initial values of the scalar field @xmath8 for the second order differential equation ( 6 ) : the value of the scalar field @xmath8 and its derivative during the radiation epoch was taken in such a manner that the present value of the fine structure constant coincide with experiment , and it appeared that the initial value of the @xmath42 during the radiation domination epoch could be assigned a rather arbitrary value because the result of integration influenced rather weakly by this choice .
- dash - dot line.,width=453 ] experimental results for keck telescope @xcite , closed circles - experimental results from vlt telescope ( data were taken from work @xcite ) , red circle at @xmath43 - oklo result.,width=453 ] as it is followed from figure [ dens ] , the scalar field @xmath8 influence rather weakly on the variation of the different components of the energy density with red shift .
the total variation of alpha during the whole history of the universe is about @xmath44 ( as is followed from figure [ alpha ] ) which is not contradict big bang and radiation recombination constraints @xcite . on
the other side the oklo analysis predict about zero result for @xmath45 with the experimental error which could be seen in figure [ alpha ] ) if we increase the scale of figure [ alpha ] one hundred times .
we investigate the constraints on the parameters of bsbm model followed from oklo analysis in the next section .
in analysis of oklo data @xcite we obtained the following constraints on the variation of the fine structure constant @xmath46 during the past @xmath47 years .
the age of the reactor @xmath48 years corresponds to red shift parameter @xmath43 .
we use here also previous constraints obtained in @xcite : @xmath49 and in @xcite : @xmath50 all these constraints are shown on figure [ oklo ] . to provide the solution of the equations ( 5 ) and ( 6 ) which does nt contradict the result of work @xcite ( see figure [ oklo ] )
, we have to set rather severe constraints on the combinations of the parameters of bsbm model .
they have to satisfy the following inequality : @xmath51 for realistic value @xmath52 to fulfill this inequality we have to demand that : @xmath53
a theoretical framework under very general assumptions was worked out by bekenstein to admit the variation of the fine structure constant .
a characteristic length @xmath3 enters into it .
an experimental constraint rules out @xmath0 variability of any kind if it is in clear conflict with predictions of the framework for @xmath3 no shorter than the fundamental length @xmath39 ( @xcite ) . as a result of oklo analysis
we get @xmath54 the oklo geophysical constraints strongly rule out all @xmath0 variability . in this analysis
we have considered only the variation of electromagnetic fine structure constant @xmath0 .
if other fundamental constants also varies the picture would be more complicated as well as the analysis of the oklo phenomenon and the analysis of the cosmological variation of @xmath0 . to do such analysis in our opinion
would be too early because till now we havent had any convincing manifestations of the cosmological variations of the other fundamental constants @xcite .
the author would like to express his gratitude to s. karshenboim and m.s .
yudkevich for useful discussions and critical remarks .
this work was partly supported by the rscf grant ( project 14 - 22 - 00281 ) . | new severe constraints on the variation of the fine structure constant have been obtained from reactor oklo analysis in our previous work .
we investigate here how these constraints confine the parameter of bsbm model of varying @xmath0 . integrating the coupled system of equations from the big bang up to the present time and taking into account the oklo limits
we have obtained the following margin on the combination of the parameters of bsbm model : @xmath1 where @xmath2 cm is a plank length and @xmath3 is the characteristic length of the bsbm model .
the natural value of the parameter @xmath4 - the fraction of electromagnetic energy in matter - is about @xmath5 . as a result
it is followed from our analysis that the characteristic length @xmath3 of bsbm theory should be considerably smaller than the plank length to fulfill the oklo constraints on @xmath0 variation . |
variability of active galactic nuclei ( agn ) provides the clearest evidence for dynamic processes occurring in the central engines and in the jets of these objects .
its study is therefore a powerful way to investigate the innermost regions of agn and the emission mechanisms responsible for the huge observed luminosities .
the emission from blazars spans the range from radio to @xmath3-ray energies , and exhibits more rapid and higher amplitude variability than other agn ( bregman 1990 ; wagner & witzel 1995 ) .
therefore , simultaneous multiwavelength monitoring of blazars is particularly suited to estimating the sizes of the emitting regions ( as a function of wavelength ) and to understanding , through correlated variability at different frequencies , the radiation processes .
the most widely accepted picture for blazar emission at radio through uv wavelengths is the synchrotron process within an inhomogeneous jet .
the model is typically characterized by a spatial dependence of the magnetic field , electron density and maximum electron energy , and usually incorporates a relativistic velocity of the plasma within the jet , which causes beaming of the radiation . how the power is transported along the jet and transferred to the high energy electrons responsible for the observed emission is still unknown .
particle acceleration may take place at a single ( dominant ) shock front or in a quasi - continuous way ( small shocks ) along the jet . in the former case ,
the spectral energy distribution from the millimeter to the soft x - rays derives from the energy distribution of the relativistic electrons accelerated at the shock front , with lower energy particles extending farther from the shock due to their longer lifetimes . in the case of _ in situ _ acceleration ( marscher 1980 ; knigl 1981 ; ghisellini , maraschi , & treves 1985 ; hutter & mufson 1986 ) the maximum emitted synchrotron frequency usually decreases along the jet , with uv and soft x - rays being produced closest to the central active source . in pks 2155304 , the brightest known bl lac object at uv wavelengths , synchrotron emission produces the optical and uv continuum , as demonstrated by simultaneous spectropolarimetric observations in the two bands ( allen et al .
the synchrotron emission extends to the medium x - ray range ( kii et al .
1996 ) and has a maximum power per decade ( @xmath4 ) between the uv and soft x - ray range ( wandel & urry 1991 ) . the spectral steepening from optical to uv to x - rays can be attributed to radiative energy losses in the single shock model , or to the decreasing volume of the region emitting at higher frequencies in the inhomogeneous jet model . in either case
the highest amplitude synchrotron variability is expected to be observed at or above the peak power output , which is determined by the steady - state balance of electron acceleration and radiation , since small changes in the electron acceleration substantially alter the higher energy emission .
previous monitoring of pks 2155304 with iue probed its variability in the far - uv domain ( 1200 - 3000 ) on a range of time scales from years down to a few hours , though the sampling was usually sparse , uneven , or limited in time ( maraschi et al . 1986 ; urry et al . 1988 ; treves et al . 1989 ; edelson et al . 1991 ; urry et al .
1993 , henceforth u93 ) .
the iue campaign in november 1991 ( u93 ) , which was coordinated with rosat observations , had adequate time coverage ( 30 days ) and sampling to probe interday variability on an extended time interval , and even intraday variability during the continuous observing period ( @xmath05 days out of 30 ) .
the presence of recurrent flares on a @xmath00.7-day time scale prompted further iue intensive monitoring in may 1994 , coordinated with euve ( marshall et al .
1996 ) , asca ( kii et al . 1996 ) , and rosat ( urry et al . 1996 ) , as well as radio , near - ir , and optical coverage from ground - based telescopes ( pesce et al . 1996 ) .
the aim of the iue campaign was to obtain the longest and best sampled uv light curve ever , in order to test the shortest possible variation time scales , within the capabilities of the iue instruments , and to explore the correlation with emission at other wavelengths ( urry et al .
1996 ) .
in this paper we concentrate on the iue monitoring . in
2 we present the iue observations and data analysis , in 3 we describe the uv light curves and spectral variability , in 4 we discuss these results and in 5
we summarize our conclusions .
iue was scheduled for continuous observations ( three 8-hr shifts per day ) from 1994 may 15 to 25 inclusive , with 8 brief ( @xmath12 hr ) and 5 longer ( between 4 and 17 hr ) gaps due to earth occultation and to a few time - critical programs .
the target acquisition was done through a double blind offset : first we pointed to the nearby bright star sao 213406 ( v = 6.5 , at 44@xmath5 distance from the source ) , then to the fainter sao 213450 ( v = 9.2 , at 4.5@xmath5 ) , and finally to the target itself .
the swp and lwp cameras were exposed in low dispersion mode alternately for 55 and 25 minutes respectively , to achieve comparable signal - to - noise ratio in both cameras , for a typical uv spectral slope of pks 2155304 ( @xmath6 ) . in the absence of operational problems , we obtained one pair of spectra each 96 minutes , due to satellite maneuvering and camera preparation overheads .
this time interval was chosen to phase with the asca satellite orbital period to allow cleaner cross - correlation analysis between the uv and x - ray light curves ; depending on the overheads , some of the spectra had slightly longer or shorter integration times than the nominal 25 and 55 minutes .
one long - wavelength spectrum ( lwp 28222 ) and three short - wavelength spectra ( swp 50815 , 50840 , 50854 ) were very underexposed ( the exposure times were less than half the normal values ) and were discarded from the subsequent analysis .
the photometrically flat - fielded and geometrically corrected images were inspected to assure proper target centering .
one long - wavelength spectrum was unusable because of off - axis placement of the aperture during exposure ( lwp 28187 ) .
as of late 1992 , scattered solar light severely affects the iue field of view ( depending on the satellite position with respect to the sun ) and can significantly compromise the spectrophotometry longwards of @xmath02700 ( caplinger 1995 , and references therein ) .
the scattered light also precludes useful information about the source brightness at optical wavelengths from the fes , so no fes counts from the source were recorded .
spectra were extracted from each of the 236 good iue images using the tomsips routine ( ayres 1993 ; ayres et al . 1995 ) , a modified version of the signal - weighted extraction technique ( swet ; kinney , bohlin , & neill 1991a ) .
the extracted net fluxes were converted to absolute fluxes using calibration curves based on swp and lwp low dispersion spectra of the white dwarf wd g191-b2b .
no correction was applied for the sensitivity degradation of the cameras .
the complete log of the iue observations is reported in table 1 .
two typical spectra from the campaign are shown in fig . 1 along with the intrinsic error distribution of the spectral flux .
the best - fit power - law model for the continuum is shown as a solid line .
the results of the tomsips extraction were compared with alternative processings using the standard iuesips , gex ( urry & reichert 1988 ) , and the final archive newsips ( nichols & linsky 1996 ) . in the first two cases a general consistency was found within @xmath010 - 15% , though gex gave anomalous results in a few cases .
the newsips lwp spectra from the end of the us2 shift , which is the most heavily affected by background radiation , turned out to be unreliable .
for these cases , the signal - to - noise ratio of the spectral data is so low ( because of the high background level ) that the extraction technique uses only two spline nodes to fit the cross - dispersion profile ( imhoff 1996 ) . in presence of solar light contamination , which ramps up dramatically above 2800 ,
if only two spline nodes are used to fit the cross dispersion profile , the fit can not accurately follow the rapid decline in flux below 2800 , resulting in overestimated flux in a large portion of the spectrum . for the swp ,
the newsips and tomsips light curves at 1400 are in agreement , apart from the different applied calibrations .
the expected interstellar extinction of the uv flux due to galactic neutral hydrogen is @xmath7 = 0.08 mag , corresponding to a column density n@xmath8 @xmath9 ( lockman & savage 1995 ) , assuming a gas - to - dust ratio n@xmath10 = 5.2@xmath11 @xmath9 mag@xmath12 ( shull & van steenberg 1985 ) and a total - to - selective extinction ratio @xmath13 ( rieke & lebofsky 1985 ) . the 236 iue spectra are well fitted by simple power - law models plus this assumed reddening ( 2.2 ) .
however , with @xmath7 as a free parameter , the @xmath14 associated with a power - law model is minimized ( both over the full set of swp and over the set of merged swp+lwp spectral flux distributions ) for @xmath7 = 0.4 mag , with a high degree of significance ( @xmath15 99.99% , according to the f - test ) . this value is inconsistent with that deduced from the galactic column density , and in fact with the results from u93 , who found no such excess reddening .
this might indicate variable absorption at the source or in intervening material ( bruhweiler et al .
nevertheless , since pks 2155304 was observed by euve during the campaign ( which would have been unlikely if the extinction were so high ) and since the @xmath7 = 0.08 mag fits are still acceptable , we conservatively adopted @xmath7 = 0.1 mag for consistency with u93 .
( this was derived , under the same above assumptions , from the hydrogen column density determined by the @xmath16 21 cm survey of stark et al .
( 1992 ) , n@xmath17 @xmath9 . )
note that for @xmath7 = 0.1 mag , the swp+lwp spectral indices are consistent with a simple power - law , while @xmath7 = 0.4 mag would require that the swp slope is steeper by @xmath18 than the lwp ( fig .
2 ) . for the dereddening corrections we used the extinction curve of seaton ( 1979 ) , as in u93 .
the updated dereddening curve of cardelli , clayton , & mathis ( 1989 ) would imply an average discrepancy of the fitted parameters not exceeding their uncertainties . for the spectral analysis we followed a procedure similar to u93 .
through an iterative , chi - squared minimization fitting routine , the dereddened swp spectral flux distributions ( 1230 - 1950 ) were fitted in wavelength space to a simple power - law model of the form @xmath19 , ignoring the regions 1277 - 1281 , 1286 - 1290 , 1660 - 1666 , and 1780 - 1800 , which are affected by camera artifacts ( crenshaw , bruegman , & norman 1990 ) , as well as regions of individual spectra contaminated by cosmic ray hits .
the fit was normalized to 1560 , which is the flux - weighted mean wavelength of the chosen interval for a spectral index @xmath20 .
similarly , power - law spectral fits were made to the lwp spectra in the 2100 - 2700 region , which is not heavily affected by the solar scattered light . because the signal in the 2700 - 2800 region did not exceed the fit curve extrapolation , and was well represented by the same power - law
, we concluded that the effect of the iue baffle anomaly is negligible shortward of 2800 ( as suggested also by direct image inspection ) , so we extended the fitted region to 2800 .
the fiducial wavelength of the fit normalization , computed as above , was chosen to be 2580 .
the 1-@xmath21 uncertainties associated with the fitted fluxes are generally less than 1% for swp spectra , and a few percent for lwp due to the large intrinsic errors affecting the spectral signal between 2100 and 2400 .
to these uncertainties a @xmath01% photometric error was added in quadrature , following u93 and edelson et al .
the best - fit parameters , energy spectral indices @xmath22 ( where @xmath23 ) and fitted fluxes at 1400 and 2800 ( swp and lwp respectively ) , are given in table 1 , as are the reduced @xmath14 values for each fit ( @xmath24 ) .
we also fitted a power - law model to pairs of swp and lwp spectra taken close together in time .
the fit was done over the wavelength range 1230 - 2800 , excluding the same regions that were discarded in the swp spectra and the 1900 - 2150 interval , which is affected by large errors . during the first day of monitoring
, the extremely fast variability does not allow any meaningful match between swp and lwp exposures .
therefore , since the determination of the spectral index would yield unreliable values , due to lack of simultaneity , we excluded the first six pairs of spectra from computation of the combined energy index @xmath25 .
the combined swp+lwp fit results are presented in table 2 .
the present analysis leads to a steeper average spectral slope than reported for the 1991 data ( @xmath26 is larger by @xmath028% , @xmath27 by @xmath014% , and @xmath28 by @xmath044% ) , which were reduced with the swet method .
we investigated the cause of this difference by re - analyzing the 1991 data after extracting the spectra with the tomsips routine and found that the change is mostly due to the different adopted calibration curves , and to the fact that the 1991 data were corrected for the swp camera sensitivity degradation , with smaller effects due to the more limited fitting range at the long wavelengths and ( only marginally ) to the different extraction algorithms ( e.g. , profile normalizations ) .
fitting the 1991 tomsips extracted spectra to a power - law yields average spectral indices that are consistent with those obtained for the 1994 data .
( see table 3 for a synoptic comparison between the 1991 and 1994 sets . )
the average @xmath24 values for the swp , lwp , and merged spectral fits are 1.02 , 0.67 , and 1.00 , respectively .
this indicates that , given the derived flux errors , the power - law model is acceptable , therefore no rescaling was applied to the intrinsic flux errors ( as had been done in u93 ) . to determine errors on spectral index we looked at the differences between spectral indices for pairs of adjacent spectra taken closer in time than 0.1 days , excluding the first six points in each energy index curve ( where the variability was much faster than the exposure time ) , leaving 94 , 95 , and 91 pairs of swp , lwp , and merged spectra , respectively . assuming that the spectral index does not vary significantly between two observations spaced only @xmath02 hours apart in time ( which is not the case in the first part of the monitoring ) , the difference divided by the sum in quadrature of their individual errors should be normally distributed , with a unity variance . since we found variances larger than unity ( 2.13 , 1.17 , and 2.86 respectively for the swp , lwp , and merged spectra ) , we applied , as in u93 , a correction to the spectral index errors equivalent to such variance .
the spectral index distributions for the swp and lwp overlap ( though the former is much narrower ) , indicating that the assumed extinction produces a consistent result for both swp and lwp spectral shapes ( see fig .
for @xmath7 = 0.4 mag , the fitted fluxes at 1400 and 2800 would be larger by @xmath0110% and @xmath065% , respectively , and the mean swp and lwp spectral indices would be smaller , the differences being @xmath00.3 and @xmath01.2 , respectively , compared to the values given in table 1 for @xmath7 = 0.1 mag .
the swp and lwp spectral index distributions for @xmath7 = 0.4 mag ( fig .
2b ) differ by @xmath18 , suggesting that such a high extinction value is unlikely .
both long- and short - wavelength iue light curves clearly show strong fast variability ( fig .
the most striking result of the campaign is the high amplitude , extremely rapid flux variability detected during the first day of monitoring ( fig .
3b ) . at 2800 ,
flux variations as large as a factor 2.2 were found in 1.5 hours ( i.e. , between one lwp integration and the other ) , implying that significant variability can occur on time scales shorter than the typical iue integration time .
the flux at 1400 also varied remarkably in the first part of the monitoring , though with smaller observed amplitude ( @xmath025% in 1.5 hours ) .
this behavior is entirely new and unexpected , since the many previous iue observations showed only minor differences in swp and lwp variability , usually in the sense of a larger amplitude in swp . in fig . 3b the two light curves
are compared normalizing the 2800 and the 1400 fluxes to their respective mean values ( computed excluding the initial flares , i.e. the first six flux points of each curve ) .
the figure shows that the swp light curve can be reconciled with the hypothesis of a complete correlation with the lwp ( i.e. , no spectral variability ) given that the swp integrations are twice as long as for the lwp and allowing for the fact that the shortest time scale variability is probably not resolved in either band . because the observed variability at 2800 is of unprecedented amplitude for such short time scales , we investigated possible extrinsic causes .
first , we asked whether motion of the spacecraft could have caused a dip in flux as the source drifted out and back into the large aperture .
we examined carefully the line - by - line file for the image with the sharpest drop in spectral flux , lwp 28142 .
the spectrum is visible with a good signal - to - noise ratio in the central part of the line - by - line image .
this , together with proper centering of the adjacent swp spectra with respect to the geocoronal ly@xmath29 line ( which fills the aperture ) , exclude a drift of the target along the long dimension of the aperture .
similarly , a drift along the dispersion direction is excluded because of the location of the reseau marks at the same wavelengths as in the other images , and the accuracy of the spacecraft slews between the target and the offset star .
second , we ruled out a mis - registration of the optimal extraction slit with respect to the spectrum , since several independent extraction methods ( including the original boxcar ) give essentially the same spectrum .
third , this `` dip '' event might resemble some sudden flux drops detected in lwp data during iue intensive monitorings , which have been ascribed to voltage failures in the lwp camera ( lwp anomaly , t. teays , private communication , 1996 ) .
however , such spurious effects are very rare , and no firm conclusion has been achieved on the correlation between the occurrence of the phenomenon and electric current depletion during the exposure .
more importantly , these events are confined to the lwp , while the event in pks 2155304 has a ( less dramatic ) counterpart in the swp light curve , in the euve light curve ( marshall et al .
1996 ) , and in polarized optical light ( pesce et al . 1996 ) .
considering all these points , we conclude the rapid initial variations reflect actual events in the bl lac object . in both light curves , after the very active period recorded at the beginning of the campaign
, the flux stays relatively constant ( after may 16 ) , with a gentle increase and decline of @xmath010% over 3.5 days , and then there is a prominent 2-day flare , with a total increase of @xmath035% , starting on may 19 .
further variation in the last four days of the monitoring is @xmath120% in both bands . throughout the monitoring , small ( 5 - 10% ) and rapid ( @xmath0 hours )
flux variations are superimposed on the more dramatic flares .
a variability test ( edelson 1992 ) yields similar variability indices for the swp and lwp light curves , which are marginally consistent within the 10% error , calculated as in edelson et al .
( 1995 ; @xmath30 , where @xmath31 is the standard deviation of the flux , and @xmath32 its average value ) : @xmath33 = 0.11 , @xmath34 = 0.13 . performing the test on the light curves after removing their first , dramatically variable portions , results in a 0.10 variability index both for swp and lwp .
this is consistent with the results of the intensive november 1991 campaign on pks 2155304 , when comparable amplitude variability was found in both iue wavelength ranges .
systematically larger amplitude variability is found at shorter uv wavelengths than at longer ones for blazars observed over longer time scales ( kinney et al .
1991b ; edelson 1992 ) , but the difference is not significant given the errors ( treves & girardi 1991 ) , and such differences are not observed on short time scales , where rapid variations are resolved ( see u93 ) .
we computed auto- and cross - correlation functions for the light curves by means of the discrete correlation function method ( dcf ; edelson & krolik 1988 ) . since the character of the variability in the first day is extreme and unresolved , we systematically removed the first 6 flux points in both light curves prior to application of the dcf routine . the auto - correlation function of the present swp and lwp light curves ( fig .
4 ) does not show any evidence of periodicity ( as would be implied by characteristic `` humps '' on the auto - correlation function curve at non - zero lags ) nor is any visible in the light curves , in contrast to the previous findings .
the auto - correlation function of a data train shows features at the timescale corresponding to recurrence in its variations , such as periodicity or quasi - periodicity would imply .
this allowed u93 to find a quasi - periodic variation in the november 1991 iue light curves of pks 2155304 with a @xmath00.7 days time scale ( a strict periodicity was shown not to be statistically significant by edelson et al .
the swp and lwp light curves are well correlated with each other , with no apparent lag larger than @xmath00.1 days ( fig .
5 ) , which is the approximate temporal resolution . during the first day of monitoring the uv spectrum of pks 2155304 varied both in the lwp and , less prominently , in the swp range ( fig .
after may 16 , the slopes of the swp and lwp spectra have an overall fractional variability @xmath35 of 0.08 and 0.28 , respectively ( with a 10% uncertainty on these values ) , according to the variability test of edelson ( 1992 ) .
testing the spectral index behavior against a constant trend yields a @xmath24 of 1.8 for swp and 2.2 for lwp , which implies a probability of constancy of less than 0.1% .
the central resolved flare is accompanied by spectral variations of more modest amplitude : @xmath36 is found for both index curves , whereas @xmath24 = 53 and @xmath24 = 21 is associated with the flux variations in the swp and lwp , respectively .
the auto - correlation function of the lwp spectral index shows equally spaced peaks with @xmath01 day separation .
this periodicity ( also recognizable in fig .
6b ) is spurious , deriving from the periodic background contamination during the us2 shift .
the swp spectral index auto - correlation function does not exhibit any significant features .
no clear trend is visible between the fluxes at 1400 and the swp spectral slopes , but their cross - correlation has a minimum at a lag of @xmath37 day ( fig .
7 ) , which implies that spectral flattening ( steepening ) leads flux increases ( decreases ) . limiting the cross - correlation function computation to the segments of the light and index curves corresponding to the central flare yields no evidence that the effect might be dominated by the behavior during the outburst .
the slope @xmath25 of the swp+lwp spectra is significantly variable ( @xmath38 = 0.08 , with @xmath24 = 2.8 ; see fig .
6c , where @xmath25 is plotted only for observations taken after may 16 ) . during the central flare , a @xmath39
is associated with the @xmath25 variation .
cross - correlating @xmath25 with the flux at 2000 yields the same results as found for the swp : spectral hardening @xmath01 day in advance of flux rise .
iue monitoring of pks 2155304 in may 1994 has given us the best sampled uv light curve ever obtained for an agn , with 2.2 times the temporal extension of the intensive part of the november 1991 campaign .
on average , the state presently detected was @xmath020% lower than in 1991 in both the swp and the lwp .
the observed flux variability behavior was also remarkably different .
the light curves from the two intensive campaigns are shown for comparison in fig .
in particular , we do not see the quasi - periodicity seen in 1991 ( see flux auto - correlation functions , fig .
4 ) , which has therefore to be regarded as random or transitory .
no significant spectral change is seen between the two epochs ( see 2.2 and table 3 ) . compared to the iue archival data ( edelson et al .
1992 ; pian & treves 1993 ) both uv flux and spectral index were about average in may 1994 .
the 1994 data exhibit dramatic and unprecedented variability , which is still underresolved , during the first day of the monitoring and later a well sampled flare of a factor of @xmath035% in @xmath01.5 days , visible in both the 1400 and 2800 light curves without any significant difference .
the flux rise to maximum is longer than the fading ( @xmath01 day ) toward the previous `` quiescent '' state ( as indicated by the asymmetric shape of the flare , fig .
3a ) , in agreement with the finding that uv flux decrease in blazars is usually sharper than brightening ( edelson 1992 ) .
the longer and `` structured '' rise requires that the mechanism producing the flare is not `` instantaneous '' but rather intrinsically long or diluted , possibly by light travel time effects in the emission region , or by multiple smaller events .
the fast fluctuations seen at the beginning of the light curve are truly exceptional . during the first day the lwp light curve exhibits a variation of a factor 2.2 in 8 hours and a second flare of similar amplitude in 1.5 hours ( fig .
rapid variability is simultaneously seen in the swp and in the lwp . fast ( though unresolved ) variations have also been detected during the optical observations simultaneous to the present iue campaign ( pesce et al .
1996 ) . defining the flux doubling time scale as @xmath40 one obtains @xmath41 = 6.87
hr and @xmath42 = 1.36 hr as minimum values at 1400 and 2800 respectively during the monitoring .
this is the most rapid observed flux doubling for pks 2155304 either in the uv or optical range , where typical values are of the order of days ( see u93 ; carini & miller 1992 ; miller 1996 ) .
doubling time scales as short as @xmath01 hour have been seen in pks 2155304 optical light curves ( pesce et al .
1996 ; paltani et al .
1996 ) , but for typical amplitudes much smaller than a factor of 2 . in x - rays ,
1-hour doubling time scales for this source are relatively more common ( morini et al . 1986 ) .
similar events were observed in several x - ray bright blazars in the x - ray band ( see e.g. , h 0323 + 022 , feigelson et al .
1986 ; pks 2155304 , treves et al .
1989 , sembay et al . 1993
; pks 0716 + 714 , cappi et al .
1994 ) , and also in the @xmath3-rays ( mkn 421 , macomb et al . 1995 ; pks 1622 - 297 , mattox et al . 1997 ) .
conspicuous spectral variations corresponding to the big central flare are not seen , a behavior reminiscent of other bl lacs in optical and uv ( oj 287 , pian et al .
1996 ; sillanp et al . 1996
; pks 0716 + 714 , wagner & witzel 1994 , see however ghisellini et al .
1996 , who find opposite results for this object in the optical ) .
this result is to be compared with the outcome of longer term monitorings of blazars ( days to years ) , showing that uv spectral variations do occur but are generally modest compared to flux changes and often weakly or not clearly correlated with them ( edelson 1992 ; shrader et al .
1994 ; koratkar et al .
1996 ) . where correlation is found , it is generally in the sense of a harder spectrum for brighter flux ( urry et al .
1988 ; bonnell et al .
1994 ) , in qualitative agreement with models based on radiative cooling . a significant feature of the cross - correlation of flux and spectral index for the swp ( fig .
7 ) is a minimum at @xmath0 1 day , which represents anti - correlation of the two quantities , consistent with the longer term spectral variability mentioned above .
the negative time - lag means that the spectrum is hardening ( softening ) @xmath01 day before the flux increases ( decreases ) , a result which , as yet unexplained , was also found in the november 1991 campaign ( see discussion in u93 ) .
the variability in the initial part of the light curve is so rapid that we can not test color variations using the joint lwp and swp ranges during the event .
however the spectral index in the lwp range takes one of its lowest values ( @xmath43 = 0.5 ) and its maximum value ( @xmath43 = 1.7 ) in correspondence to the relative maximum flux observed on may 15.95 and to the deep minimum observed on may 16.08 , respectively .
thus the little information we have points to some spectral variability rather than to an achromatic event . the completely resolved central flare , together with the @xmath050% and @xmath080% correlated flares detected at the extreme uv and x - ray wavelengths , respectively ( marshall et al .
1996 ; kii et al .
1996 ; urry et al .
1996 ) , is consistent with a variability amplitude monotonically increasing with energy , as expected for a synchrotron flare in an inhomogeneous jet ( celotti , maraschi , & treves 1991 ; georganopoulos & marscher 1996 ) .
the excellent correlation of the swp and lwp light curves implies that the uv emission in the 1200 - 3000 range is produced within a unique emitting region , without difference in the electron cooling times at these wavelengths larger than an hour .
since these must be equal to or shorter than the uv fading time scales , we can estimate a lower limit for the magnetic field in the uv emitting portion of the jet ( blandford 1990 ) , locally approximated as a homogeneous region : @xmath44 where @xmath45 is the frequency in units of @xmath46 hz , @xmath47 is the observed variability timescale in hours , and @xmath48 represents the doppler factor of the relativistic bulk motion .
based on the shortest observed variability time scale ( 1.5 hr at 2800 ) and assuming @xmath49 , we derive @xmath50 g. the detection of gev @xmath3-rays from pks 2155304 ( vestrand , stacy , & sreekumar 1996 ) indicates that inverse compton radiation due to electrons scattering off the synchrotron photons or other soft seed photons is significant .
the seed photons are expected to emit at optical and uv wavelengths ( @xmath51 , where the maximum electron energy @xmath3 is typically @xmath52 ) . given the derived lower limit on the magnetic field and the measured synchrotron luminosity in the uv ( @xmath53 ) , we can estimate the expected @xmath3-ray luminosity ( @xmath54 ) in the homogeneous case of an emitting blob of radius @xmath55 .
we consider the following relation between the observed synchrotron and inverse compton luminosities : @xmath56 hence @xmath57 where @xmath58 is the magnetic energy density , and @xmath59 is the synchrotron radiation energy density . for @xmath60 erg s@xmath12 , corresponding to the uv emission alone , and @xmath50 g , @xmath61 hr , @xmath49 , one obtains @xmath62 erg s@xmath12 , consistent with the observed @xmath3-ray luminosity of @xmath63 erg s@xmath12 .
the only quantities not directly observed are @xmath64 and @xmath48 .
the approximation @xmath49 is supported by other observations ; given the @xmath3-ray limit here , @xmath64 could be an order of magnitude larger than 1 g. the extremely rapid variation observed at the beginning of the uv monitoring implies that the limit @xmath65 erg s@xmath66 ( fabian 1979 ) is slightly exceeded ( by a factor of @xmath01.2 ) if an accretion efficiency @xmath67 is assumed , for a redshift @xmath68 = 0.116 ( falomo et al .
1993 ) , a hubble constant @xmath69 km s@xmath12 mpc@xmath12 and a deceleration parameter @xmath70 = 0.5 , and hence the rapid variations support the idea of relativistic beaming . based only on the flux change detected in the 2100 - 2800 band , the beaming factor need not exceed unity , but it might be well larger considering the simultaneous ( and in some cases larger ) flux variation in other parts of the electromagnetic spectrum , which is unfortunately undersampled ( see urry et al .
the observed time scale corresponds to a very small emission region , only @xmath71 cm if beaming corrections are not applied , or @xmath72 cm for @xmath73 .
notice that the constraint on the presence of beaming derived only from the variability in the uv band is a factor of 4 more stringent than found in u93 , but weaker than that determined by morini et al .
( 1986 ) , who detected a rapid increase in x - ray ( 1 - 6 kev ) flux for which @xmath74 exceeded the above limit by a factor 10 .
the may 1994 iue monitoring on pks 2155304 , which was part of a simultaneous multiwavelength campaign from radio to x - rays , yielded the best sampled uv light curve for this or any blazar and revealed significant flux variability at different time scales , from hours to days .
spectral changes are generally modest and not clearly correlated with flux variations . a resolved central flare of @xmath035% amplitude was observed in both iue cameras , and was likely correlated with flares of different amplitude and duration at higher energies ( urry et al .
the 1.5-hour flux variation of a factor 2.2 seen at 2800 during the first day of monitoring is unprecedented for blazars as a class .
this event , which suggests the occurrence of variability on time scales even shorter than the iue time resolution , represents a definite violation of the limits on luminosity variability , therefore implying the presence of relativistic beaming .
both variability events are interpreted within a scenario in which synchrotron radiation is the primary emission mechanism , and a lower limit of 1 g on the intensity of the magnetic field is determined , which turns out to be consistent with a calculation of this physical quantity based on the multiwavelength data and with the @xmath3-ray flux observed by egret .
comparison with the results from the 1991 iue intensive monitoring of pks 2155304 shows that the variability characteristics detected at the two epochs are fundamentally different .
in particular , unlike the findings of the 1991 campaign , no periodicity is seen in the present data .
the @xmath01 day anti - correlation between flux and spectral variations is however maintained .
the fastest variability in the uv has been largely undersampled with iue for pks 2155304 , and in general for other blazars .
uv telescopes allowing a better time resolution and continuous coverage are clearly needed .
apart from hst , which has rarely been dedicated to long monitorings , none are presently available .
while the fast sporadic variability may be most apparent in the uv , it should be expected to manifest itself , possibly in a less extreme form , also at optical wavelengths .
some indication is already present in the optical data of the 1994 campaign , especially in the polarization measurements ( pesce et al .
these arguments point to the importance of intensive , systematic monitoring in the optical band , possibly with polarization information .
such programs , which have led to the discovery of intraday variability in a number of blazars ( wagner & witzel 1995 ; miller 1996 ; sillanp et al . 1996 ; smith 1996 ) , can be carried out with medium sized telescopes and standard instrumentation , but need long , uninterrupted observing runs , possibly coordinated among different sites .
ep , cmu , jep , and pg acknowledge support from nasa grants nag5 - 1034 and nag5 - 2499 ; ep acknowledges support from a nato - cnr advanced fellowship ; rik acknowledges support from nasa ltsa grant nagw-2120 .
we are grateful to t. ayres for clarification of the tomsips routine and to p. smith for critical comments .
the iue staff at vilspa and gsfc is acknowledged , and particularly we would like to thank r. arquilla , r. bradley , j. caplinger , d. de martino , m. england , c. gonzalez , a. groebner , c. imhoff , n. loiseau , c. loomis , d. luthermoser , b. mccollum , j. nichols , r. pitts , c. proffitt , l. rawley , p. rodriguez , m. schlegel , l. taylor , t. teays , and r. thompson for assistance with iue observations and data reduction . c. bailyn , r. bohlin , j. bregman , w. brinkmann , m. carini , l. chiappetti , m. donahue , e. feigelson , a. fruscione , a. knigl , l. kedzior ,
y. kondo , a. koratkar , j. krolik , a. lawrence , f. makino , p. martin , h. miller , p. obrien , g. reichert , a. sadun , m. sitko , p. smith , a. szymkowiak , g. tagliaferri , e. tanzi , s. wagner , r. warwick , a. wehrle are acknowledged for their support to the observing project .
ccccccccc + + iue image & observation & exposure & observatory & @xmath75 & @xmath76 & @xmath77 & @xmath78 & @xmath24 + & midpoint ( ut ) & time & ( goddard & ( mjy ) & ( mjy ) & & & + & ( day of may 94 ) & ( min ) & or vilspa ) & & & & & + swp 50773 & 15.71418 & 60
& g & 7.03 & 0.13 & 0.95 & 0.10 & 0.66 + swp 50774 & 15.78942 & 55 & g & 7.75 & 0.12 & 0.88 & 0.07 & 0.81 + swp 50775 & 15.85629 & 35 & g & 8.66 & 0.13 & 1.19 & 0.06 & 0.88 + swp 50776 & 15.98259 & 45 & v & 8.95 & 0.13 & 1.04 & 0.05 & 0.79 + swp 50777 & 16.05153 & 44 & v & 8.63 & 0.13 & 1.04 & 0.06 & 1.02 + swp 50778 & 16.11782 & 55 & v & 8.01 & 0.12 & 0.88 & 0.05 & 0.89 + swp 50779 & 16.18512 & 55 & v & 9.89 & 0.14 & 0.90 & 0.05 & 1.02 + swp 50780 & 16.25361 & 55 & v & 9.92 & 0.14 & 0.98 & 0.05 & 0.97 + swp 50781 & 16.32059 & 50 & g & 9.90 & 0.14 & 0.98 & 0.05 & 0.96 + swp 50782 & 16.38535 & 55 & g & 9.89 & 0.14 & 0.94 & 0.05 & 0.88 + swp 50783 & 16.45285 & 55 & g & 9.91 & 0.14 & 1.00 & 0.05 & 1.06 + swp 50784 & 16.51876 & 55 & g & 9.86 & 0.14 & 1.01 & 0.05 & 1.05 + swp 50785 & 16.58534 & 55 & g & 9.91 & 0.14 & 0.94 & 0.05 & 1.19 + swp 50786 & 16.65740 & 42 & g & 10.19 & 0.16 & 0.92 & 0.08 & 0.87 + swp 50787 & 16.71852 & 55 & g & 9.72 & 0.34 & 0.86 & 0.25 & 0.36 + swp 50788 & 16.78274 & 45 & g & 9.85 & 0.17 & 1.11 & 0.09 & 0.65 + swp 50789 & 16.84537 & 40 & g & 10.34 & 0.15 & 0.94 & 0.05 & 0.82 + swp 50790 & 16.98689 & 48 & v & 10.11 & 0.14 & 1.00 & 0.05 & 1.08 + swp 50791 & 17.05113 & 55 & v & 10.15 & 0.14 & 0.94 & 0.05 & 1.39 + swp 50792 & 17.11719 & 55 & v & 10.17 & 0.14 & 0.98 & 0.05 & 1.06 + ccccccccc + iue image & observation & exposure & observatory & @xmath75 & @xmath76 & @xmath77 & @xmath78 & @xmath24 + & midpoint ( ut ) & time & ( goddard & ( mjy ) & ( mjy ) & & & + & ( day of may 94 ) & ( min ) & or vilspa ) & & & & & + swp 50793 & 17.18391 & 55 & v & 10.05 & 0.14 & 1.05 & 0.05 & 1.02 + swp 50794 & 17.25061 & 55 & v & 10.21 & 0.14 & 0.97 & 0.05 & 1.09 + swp 50795 & 17.31883 & 55 & g & 10.37 & 0.15 & 0.95 & 0.05 & 1.08 + swp 50796 & 17.38483 & 55 & g & 10.24 & 0.14 & 1.03 & 0.05 & 1.07 + swp 50797 & 17.45105 & 53 & g & 10.33 & 0.14 & 1.00 & 0.05 & 1.26 + swp 50798 & 17.51777 & 55 & g & 10.25 & 0.14 & 1.00 & 0.05 & 1.27 + swp 50799 & 17.58441 & 55 & g & 10.42 & 0.15 & 0.98 & 0.05 & 1.09 + swp 50800 & 17.65082 & 55 & g & 9.96 & 0.15 & 1.01 & 0.07 & 0.81 + swp 50801 & 17.71312 & 40 & g & 10.02 & 0.23 & 0.89 & 0.15 & 0.68 + swp 50802 & 17.77571 & 30 & g & 10.03 & 0.20 & 1.19 & 0.12 & 0.70 + swp 50803 & 17.84478 & 35 & g & 9.89 & 0.14 & 1.07 & 0.06 & 1.09 + swp 50804 & 17.97949 & 40 & v & 9.59 & 0.14 & 1.08 & 0.06 & 0.91 + swp 50805 & 18.04629 & 55 & v & 9.93 & 0.14 & 1.00 & 0.05 & 1.31 + swp 50806 & 18.10763 & 40 & v & 9.61 & 0.14 & 1.06 & 0.06 & 1.04 + swp 50808 & 18.38371 & 55 & g & 9.56 & 0.14 & 1.09 & 0.05 & 1.00 + swp 50809 & 18.44950 & 57 & g & 9.50 & 0.13 & 1.16 & 0.05 & 1.08 + swp 50810 & 18.51637 & 55
& g & 9.44 & 0.13 & 1.14 & 0.05 & 1.05 + swp 50811 & 18.58464 & 53 & g & 9.46 & 0.14 & 1.02 & 0.05 & 1.10 + swp 50812 & 18.65004 & 55 & g & 9.40 & 0.15 & 0.92 & 0.07 & 0.84 + swp 50813 & 18.70978 & 35 & g & 9.08 & 0.21 & 1.03 & 0.15 & 0.70 + ccccccccc + iue image & observation & exposure & observatory & @xmath75 & @xmath76 & @xmath77 & @xmath78 & @xmath24 + & midpoint ( ut ) & time & ( goddard & ( mjy ) & ( mjy ) & & & + & ( day of may 94 ) & ( min ) & or vilspa ) & & & & & + swp 50814 & 18.77382 & 33 & g & 9.59 & 0.19 & 0.85 & 0.12 & 0.71 + swp 50815@xmath79&19.00072&17 & v & 7.75 & 0.14 & 1.36 & 0.10 & 0.98 + swp 50816 & 19.05049 & 55 & v & 9.28 & 0.13 & 1.09 & 0.05 & 0.93 + swp 50817 & 19.11502 & 50 & v & 9.23 & 0.13 & 1.06 & 0.05 & 1.03 + swp 50818 & 19.18376 & 55 & v & 9.39 & 0.13 & 1.03 & 0.05 & 1.11 + swp 50819 & 19.24659 & 47 & v & 9.51 & 0.14 & 1.00 & 0.06 & 1.14 + swp 50820 & 19.32086 & 42 & g & 9.22 & 0.14 & 1.03 & 0.06 & 0.93 + swp 50821 & 19.38240 & 58 & g & 9.49 & 0.13 & 0.96 & 0.05 & 1.02 + swp 50822 & 19.44929 & 57 & g & 9.38 & 0.13 & 1.03 & 0.05 & 0.85 + swp 50823 & 19.51545 & 56 & g & 9.68 & 0.14 & 0.97 & 0.05 & 1.07 + swp 50824 & 19.58221 & 58 & g & 9.76 & 0.14 & 0.95 & 0.05 & 1.30 + swp 50825 & 19.64909 & 54 & g & 9.59 & 0.15 & 0.81 & 0.07 & 0.66 + swp 50826 & 19.70796 & 33 & g & 9.20 & 0.23 & 0.94 & 0.17 & 0.56 + swp 50827 & 19.77311 & 40 & g & 9.37 & 0.18 & 1.12 & 0.11 & 0.84 + swp 50828 & 19.84450 & 40 & g & 9.51 & 0.14 & 0.96 & 0.05 & 1.06 + swp 50829 & 19.98050 & 43 & v & 10.04 & 0.14 & 1.01 & 0.05 & 1.15 + swp 50830 & 20.04891 & 48 & v & 10.45 & 0.15 & 0.91 & 0.05 & 0.85 + swp 50831 & 20.11530 & 53 & v & 10.72 & 0.15 & 0.98 & 0.05 & 0.97 + swp 50832 & 20.18150 & 51 & v & 10.95 & 0.15 & 0.99 & 0.05 & 0.94 + swp 50833 & 20.24868 & 53 & v & 11.12 & 0.16 & 0.91 & 0.05 & 1.06 + ccccccccc + iue image & observation & exposure & observatory & @xmath75 & @xmath76 & @xmath77 & @xmath78 & @xmath24 + & midpoint ( ut ) & time & ( goddard & ( mjy ) & ( mjy ) & & & + & ( day of may 94 ) & ( min ) & or vilspa ) & & & & & + swp 50834 & 20.31591 & 55 & g & 11.05 & 0.16 & 0.88 & 0.05 & 1.24 + swp 50835 & 20.38177 & 57 & g & 11.16 & 0.16 & 0.92 & 0.05 & 1.34 + swp 50836 & 20.44870 & 58 & g & 11.03 & 0.15 & 1.09 & 0.05 & 1.31 + swp 50837 & 20.51553 & 57 & g & 11.23 & 0.16 & 0.95 & 0.05 & 1.08 + swp 50838 & 20.58183 & 56 & g & 11.37 & 0.16 & 1.01 & 0.05 & 1.39 + swp 50839 & 20.64849 & 54 & g & 10.69 & 0.17 & 1.06 & 0.08 & 0.66 + swp 50840@xmath79&20.70360&23 & g & 10.99 & 0.29 & 1.22 & 0.18 & 0.61 + swp 50841 & 20.77465 & 35 & g & 11.62 & 0.23 & 0.93 & 0.12 & 0.63 + swp 50842 & 20.84296 & 40 & g & 12.47 & 0.17 & 1.03 & 0.05 & 1.16 + swp 50843 & 20.98156 & 50 & v & 12.44 & 0.17 & 1.05 & 0.04 & 1.29 + swp 50844 & 21.04835 & 55 & v & 12.68 & 0.17 & 1.05 & 0.04 & 1.00 + swp 50845 & 21.11513 & 55 & v & 12.66 & 0.17 & 1.00 & 0.04 & 1.36 + swp 50846 & 21.18191 & 55 & v & 12.87 & 0.18 & 0.96 & 0.04 & 0.88 + swp 50847 & 21.24768 & 52 & v & 12.90 & 0.18 & 0.94 & 0.05 & 1.19 + swp 50848 & 21.31563 & 55 & g & 12.82 & 0.18 & 0.98 & 0.04 & 1.14 + swp 50849 & 21.38151 & 53 & g & 12.48 & 0.17 & 1.04 & 0.05 & 1.15 + swp 50850 & 21.44734 & 55 & g & 12.39 & 0.17 & 0.99 & 0.05 & 1.40 + swp 50851 & 21.51433 & 56 & g & 12.60 & 0.17 & 0.92 & 0.04 & 1.30 + swp 50852 & 21.58095 & 54 & g & 12.49 & 0.18 & 0.90 & 0.05 & 0.91 + swp 50853 & 21.64798 & 52 & g & 10.98 & 0.18 & 1.15 & 0.08 & 0.82 + ccccccccc + iue image & observation & exposure & observatory & @xmath75 & @xmath76 & @xmath77 & @xmath78 & @xmath24 + & midpoint ( ut ) & time & ( goddard & ( mjy ) & ( mjy ) & & & + & ( day of may 94 ) & ( min ) & or vilspa ) & & & & & + swp 50854@xmath79&21.70336&23 & g & 10.81 & 0.29 & 1.03 & 0.18 & 0.71 + swp 50855 & 21.77585 & 35 & g & 10.51 & 0.21 & 1.13 & 0.12 & 0.65 + swp 50856 & 21.84078 & 30 & g & 9.99 & 0.15 & 1.16 & 0.06 & 1.38 + swp 50857 & 21.98225 & 50 & v & 9.72 & 0.14 & 1.08 & 0.05 & 1.18 + swp 50858 & 22.04740 & 53 & v & 9.61 & 0.14 & 1.06 & 0.05 & 1.09 + swp 50859 & 22.11338 & 45 & v & 9.01 & 0.13 & 1.14 & 0.06 & 0.95 + swp 50861 & 22.38183 & 55 & g & 9.05 & 0.13 & 1.19 & 0.05 & 0.81 + swp 50862 & 22.44815 & 55 & g & 8.95 & 0.13 & 1.09 & 0.05 & 1.04 + swp 50863 & 22.51313 & 55 & g & 9.13 & 0.13 & 1.06 & 0.05 & 1.40 + swp 50864 & 22.57973 & 55 & g & 9.33 & 0.13 & 0.98 & 0.05 & 0.90 + swp 50865 & 22.64640 & 55 & g & 8.83 & 0.15 & 0.98 & 0.09 & 0.68 + swp 50866 & 22.70536 & 35 & g & 8.87 & 0.20 & 1.08 & 0.15 & 0.58 + swp 50867 & 22.77708 & 48 & g & 9.00 & 0.16 & 0.97 & 0.09 & 0.64 + swp 50868 & 22.83884 & 30 & g & 8.61 & 0.13 & 1.14 & 0.07 & 1.14 + swp 50869 & 22.98010 & 52 & v & 9.71 & 0.14 & 1.00 & 0.05 & 1.32 + swp 50870 & 23.04433 & 45 & v & 9.78 & 0.14 & 1.08 & 0.05 & 1.17 + swp 50871 & 23.11241 & 55 & v & 9.85 & 0.14 & 1.03 & 0.05 & 1.66 + swp 50872 & 23.18152 & 53 & v & 9.83 & 0.14 & 1.06 & 0.05 & 1.05 + swp 50873 & 23.24588 & 50 & v & 9.73 & 0.14 & 1.06 & 0.05 & 1.10 + swp 50874 & 23.31339 & 55 & g & 8.90 & 0.13 & 1.19 & 0.05 & 1.34 + ccccccccc + iue image & observation & exposure & observatory & @xmath75 & @xmath76 & @xmath77 & @xmath78 & @xmath24 + & midpoint ( ut ) & time & ( goddard & ( mjy ) & ( mjy ) & & & + & ( day of may 94 ) & ( min ) & or vilspa ) & & & & & + swp 50875 & 23.37897 & 55 & g & 8.88 & 0.13 & 1.16 & 0.05 & 1.01 + swp 50876 & 23.44587 & 55 & g & 9.31 & 0.13 & 1.12 & 0.05 & 1.12 + swp 50877 & 23.51219 & 55 & g & 9.71 & 0.14 & 1.01 & 0.05 & 1.03 + swp 50878 & 23.57843 & 55 & g & 9.85 & 0.14 & 1.02 & 0.05 & 1.02 + swp 50879 & 23.64552 & 55 & g & 9.04 & 0.15 & 1.03 & 0.08 & 0.88 + swp 50880 & 23.70677 & 40 & g & 8.53 & 0.20 & 1.23 & 0.15 & 0.60 + swp 50881 & 23.77962 & 52 & g & 9.06 & 0.15 & 0.97 & 0.08 & 1.07 + swp 50882 & 23.83647 & 30 & g & 8.58 & 0.13 & 1.27 & 0.06 & 1.16 + swp 50883 & 23.97883 & 55 & v & 9.99 & 0.14 & 1.05 & 0.05 & 1.24 + swp 50884 & 24.04752 & 55 & v & 8.96 & 0.13 & 1.05 & 0.05 & 1.13 + swp 50885 & 24.11168 & 55 & v & 10.07 & 0.14 & 1.04 & 0.05 & 1.16 + swp 50886 & 24.17872 & 55 & v & 10.30 & 0.14 & 1.00 & 0.05 & 1.17 + swp 50887 & 24.24718 & 50 & v & 10.18 & 0.14 & 1.03 & 0.05 & 1.14 + swp 50889 & 24.71008 & 55 & g & 11.31 & 0.16 & 1.02 & 0.05 & 1.33 + swp 50890 & 24.77687 & 55 & g & 10.87 & 0.15 & 1.07 & 0.04 & 1.16 + swp 50891 & 24.83460 & 30 & g & 10.51 & 0.15 & 1.09 & 0.06 & 1.11 + swp 50894 & 25.71048 & 55 & g & 11.13 & 0.16 & 0.98 & 0.05 & 1.13 + swp 50895 & 25.77088 & 37 & g & 11.09 & 0.16 & 1.07 & 0.06 & 0.92 + ccccccccc + iue image & observation & exposure & observatory & @xmath75 & @xmath76 & @xmath77 & @xmath78 & @xmath24 + & midpoint ( ut ) & time & ( goddard & ( mjy ) & ( mjy ) & & & + & ( day of may 94 ) & ( min ) & or vilspa ) & & & & & + lwp 28137 & 15.67458 & 30 & g & 7.75 & 0.27 & 0.39 & 0.35 & 0.66 + lwp 28138 & 15.75657 & 30 & g & 12.04 & 0.37 & 0.81 & 0.29 & 0.57 + lwp 28139 & 15.83287 & 25 & g & 12.07 & 0.28 & 0.97 & 0.21 & 0.55 + lwp 28140 & 15.95287 & 25 & v & 17.68 & 0.34 & 0.53 & 0.15 & 0.66 + lwp 28141 & 16.02132 & 25 & v & 15.12 & 0.32 & 1.55 & 0.18 & 0.63 + lwp 28142 & 16.08396 & 25 & v & 7.87 & 0.24 & 1.72 & 0.30 & 0.65 + lwp 28143 & 16.15366 & 25 & v & 17.54 & 0.34 & 1.04 & 0.15 & 0.52 + lwp 28144 & 16.22040 & 25 & v & 17.30 & 0.33 & 0.84 & 0.15 & 0.60 + lwp 28145 & 16.28846 & 25 & g & 18.23 & 0.36 & 1.10 & 0.16 & 0.63 + lwp 28146 & 16.35406 & 25 & g & 17.72 & 0.34 & 0.82 & 0.15 & 0.71 + lwp 28147 & 16.42109 & 25 & g & 18.02 & 0.36 & 1.24 & 0.16 & 0.67 + lwp 28148 & 16.48764 & 25 & g & 17.43 & 0.34 & 1.01 & 0.16 & 0.59 + lwp 28149 & 16.55419 & 25 & g & 17.96 & 0.36 & 0.93 & 0.16 & 0.67 + lwp 28150 & 16.62071 & 25 & g & 18.54 & 0.37 & 0.70 & 0.16 & 0.74 + lwp 28151 & 16.68728 & 25 & g & 18.94 & 0.49 & 0.90 & 0.23 & 0.64 + lwp 28152 & 16.75374 & 25 & g & 18.37 & 0.60 & 0.85 & 0.32 & 0.77 + lwp 28153 & 16.81868 & 20 & g & 18.94 & 0.40 & 1.23 & 0.18 & 0.63 + lwp 28154 & 16.95624 & 25 & v & 17.91 & 0.35 & 1.21 & 0.15 & 0.77 + lwp 28155 & 17.01958 & 25 & v & 18.89 & 0.36 & 1.19 & 0.15 & 0.59 + lwp 28156 & 17.08590 & 25 & v & 18.33 & 0.35 & 1.18 & 0.15 & 0.87 + ccccccccc + iue image & observation & exposure & observatory & @xmath75 & @xmath76 & @xmath77 & @xmath78 & @xmath24 + & midpoint ( ut ) & time & ( goddard & ( mjy ) & ( mjy ) & & & + & ( day of may 94 ) & ( min ) & or vilspa ) & & & & & + lwp 28157 & 17.15254 & 25 & v & 18.09 & 0.34 & 1.09 & 0.15 & 0.75 + lwp 28158 & 17.21922 & 25 & v & 18.37 & 0.35 & 1.15 & 0.15 & 0.70 + lwp 28159 & 17.28583 & 25 & g & 18.61 & 0.35 & 0.98 & 0.15 & 0.63 + lwp 28160 & 17.35373 & 25 & g & 18.27 & 0.35 & 0.98 & 0.15 & 0.62 + lwp 28161 & 17.42004 & 25 & g & 18.67 & 0.36 & 1.21 & 0.15 & 0.73 + lwp 28162 & 17.48654 & 25 & g & 18.58 & 0.35 & 0.95 & 0.15 & 0.66 + lwp 28163 & 17.55325 & 25 & g & 18.67 & 0.36 & 0.91 & 0.15 & 0.61 + lwp 28164 & 17.61939 & 25 & g & 19.01 & 0.38 & 0.76 & 0.16 & 0.79 + lwp 28165 & 17.68635 & 25 & g & 18.74 & 0.54 & 0.91 & 0.27 & 0.76 + lwp 28166 & 17.75058 & 20 & g & 18.85 & 0.70 & 0.58 & 0.36 & 0.65 + lwp 28167 & 17.81951 & 25 & g & 18.99 & 0.38 & 0.84 & 0.16 & 0.66 + lwp 28168 & 17.95336 & 25 & v & 18.43 & 0.35 & 1.01 & 0.15 & 0.56 + lwp 28169 & 18.01455 & 25 & v & 18.50 & 0.35 & 1.24 & 0.15 & 0.70 + lwp 28170 & 18.08130 & 25 & v & 18.17 & 0.35 & 1.25 & 0.15 & 0.68 + lwp 28171 & 18.35250 & 25 & g & 17.31 & 0.34 & 0.91 & 0.16 & 0.62 + lwp 28172 & 18.41790 & 25 & g & 17.72 & 0.35 & 1.00 & 0.16 & 0.66 + lwp 28173 & 18.48543 & 25 & g & 17.28 & 0.35 & 1.09 & 0.16 & 0.67 + lwp 28174 & 18.55439 & 25 & g & 17.40 & 0.34 & 0.97 & 0.16 & 0.65 + lwp 28175 & 18.61839 & 25 & g & 17.89 & 0.37 & 1.00 & 0.17 & 0.75 + lwp 28176 & 18.68376 & 20 & g & 17.81 & 0.54 & 0.54 & 0.29 & 0.64 + ccccccccc + iue image & observation & exposure & observatory & @xmath75 & @xmath76 & @xmath77 & @xmath78 & @xmath24 + & midpoint ( ut ) & time & ( goddard & ( mjy ) & ( mjy ) & & & + & ( day of may 94 ) & ( min ) & or vilspa ) & & & & & + lwp 28177 & 18.74965 & 20 & g & 16.62 & 0.60 & 0.13 & 0.35 & 0.55 + lwp 28183 & 19.01873 & 25 & v & 17.09 & 0.34 & 1.24 & 0.16 & 0.63 + lwp 28184 & 19.08539 & 25 & v & 17.10 & 0.34 & 1.22 & 0.16 & 0.60 + lwp 28185 & 19.15266 & 25 & v & 16.75 & 0.33 & 1.10 & 0.16 & 0.66 + lwp 28186 & 19.21819 & 25 & v & 16.81 & 0.33 & 0.83 & 0.16 & 0.59 + lwp 28188 & 19.35045 & 25 & g & 16.78 & 0.33 & 0.79 & 0.16 & 0.59 + lwp 28189 & 19.41771 & 25 & g & 16.89 & 0.34 & 0.90 & 0.16 & 0.58 + lwp 28190 & 19.48420 & 25 & g & 16.67 & 0.33 & 1.03 & 0.16 & 0.60 + lwp 28191 & 19.54995 & 25 & g & 17.36 & 0.34 & 1.08 & 0.16 & 0.57 + lwp 28192 & 19.61765 & 25 & g & 16.86 & 0.34 & 0.41 & 0.16 & 0.63 + lwp 28193 & 19.68391 & 25 & g & 16.82 & 0.52 & 0.34 & 0.29 & 0.59 + lwp 28194 & 19.74867 & 23 & g & 17.99 & 0.75 & 1.17 & 0.42 & 0.65 + lwp 28195 & 19.81773 & 18 & g & 18.05 & 0.36 & 0.79 & 0.16 & 0.66 + lwp 28196 & 19.95100 & 25 & g & 17.67 & 0.34 & 0.88 & 0.15 & 0.60 + lwp 28197 & 20.01746 & 25 & v & 18.56 & 0.35 & 1.02 & 0.15 & 0.71 + lwp 28198 & 20.08449 & 25 & v & 18.88 & 0.36 & 1.00 & 0.15 & 0.64 + lwp 28199 & 20.15042 & 25 & v & 18.81 & 0.35 & 1.04 & 0.15 & 0.63 + lwp 28200 & 20.21753 & 25 & v & 19.57 & 0.37 & 0.91 & 0.15 & 0.61 + lwp 28201 & 20.28340 & 25 & v & 19.26 & 0.37 & 0.71 & 0.15 & 0.64 + lwp 28202 & 20.35010 & 25 & g & 19.69 & 0.38 & 0.82 & 0.15 & 0.66 + ccccccccc + iue image & observation & exposure & observatory & @xmath75 & @xmath76 & @xmath77 & @xmath78 & @xmath24 + & midpoint ( ut ) & time & ( goddard & ( mjy ) & ( mjy ) & & & + & ( day of may 94 ) & ( min ) & or vilspa ) & & & & & + lwp 28203 & 20.41673 & 25 & g & 20.28 & 0.38 & 1.07 & 0.14 & 0.68 + lwp 28204 & 20.48396 & 25 & g & 20.07 & 0.37 & 0.83 & 0.14 & 0.55 + lwp 28205 & 20.55041 & 25 & g & 20.07 & 0.37 & 0.89 & 0.14 & 0.66 + lwp 28206 & 20.61706 & 25 & g & 20.62 & 0.41 & 0.68 & 0.16 & 0.72 + lwp 28207 & 20.68032 & 25 & g & 22.14 & 0.77 & 0.94 & 0.34 & 0.63 + lwp 28208 & 20.74675 & 16
& g & 20.73 & 0.83 & 0.12 & 0.39 & 0.65 + lwp 28209 & 20.81692 & 15 & g & 22.24 & 0.43 & 0.43 & 0.15 & 0.64 + lwp 28210 & 20.94994 & 25 & g & 23.09 & 0.41 & 1.01 & 0.13 & 0.59 + lwp 28211 & 21.01677 & 25 & v & 22.71 & 0.40 & 0.83 & 0.13 & 0.65 + lwp 28212 & 21.08339 & 25 & v & 22.82 & 0.40 & 0.89 & 0.13 & 0.80 + lwp 28213 & 21.15039 & 25 & v & 23.02 & 0.41 & 1.03 & 0.13 & 0.68 + lwp 28214 & 21.21723 & 25 & v & 22.82 & 0.41 & 0.88 & 0.13 & 0.66 + lwp 28215 & 21.28325 & 25 & v & 22.70 & 0.41 & 0.81 & 0.13 & 0.66 + lwp 28216 & 21.35069 & 25 & g & 22.91 & 0.41 & 0.96 & 0.13 & 0.68 + lwp 28217 & 21.41593 & 25 & g & 22.44 & 0.40 & 0.77 & 0.13 & 0.53 + lwp 28218 & 21.48215 & 25 & g & 22.27 & 0.40 & 0.87 & 0.13 & 0.74 + lwp 28219 & 21.54946 & 25 & g & 22.25 & 0.40 & 0.71 & 0.13 & 0.56 + lwp 28220 & 21.61635 & 25 & g & 21.59 & 0.41 & 0.25 & 0.15 & 0.91 + lwp 28221 & 21.67990 & 25 & g & 22.72 & 0.75 & 0.91 & 0.32 & 0.64 + lwp 28222@xmath79&21.74510&16 & g & 21.35 & 0.96 & 1.16 & 0.45 & 0.61 + ccccccccc + iue image & observation & exposure & observatory & @xmath75 & @xmath76 & @xmath77 & @xmath78 & @xmath24 + & midpoint ( ut ) & time & ( goddard & ( mjy ) & ( mjy ) & & & + & ( day of may 94 ) & ( min ) & or vilspa ) & & & & & + lwp 28223 & 21.81732 & 13 & g & 19.85 & 0.40 & 0.45 & 0.16 & 0.63 + lwp 28224 & 21.95083 & 25 & g & 18.65 & 0.35 & 1.12 & 0.15 & 0.58 + lwp 28225 & 22.01694 & 25 & v & 18.46 & 0.35 & 1.14 & 0.15 & 0.64 + lwp 28226 & 22.08223 & 25 & v & 18.07 & 0.36 & 1.30 & 0.16 & 0.70 + lwp 28227 & 22.34867 & 25 & v & 17.06 & 0.34 & 1.09 & 0.16 & 0.61 + lwp 28228 & 22.41740 & 25 & g & 17.44 & 0.35 & 1.24 & 0.16 & 0.73 + lwp 28229 & 22.48184 & 25 & g & 17.45 & 0.35 & 1.52 & 0.17 & 0.65 + lwp 28230 & 22.54847 & 25 & g & 17.18 & 0.34 & 1.17 & 0.16 & 0.65 + lwp 28231 & 22.61508 & 25 & g & 17.92 & 0.38 & 0.73 & 0.18 & 0.80 + lwp 28232 & 22.67991 & 25 & g & 17.64 & 0.56 & 0.80 & 0.31 & 0.79 + lwp 28233 & 22.74640 & 20 & g & 17.19 & 0.65 & 0.96 & 0.38 & 0.69 + lwp 28234 & 22.81508 & 18 & g & 17.28 & 0.34 & 0.73 & 0.15 & 0.74 + lwp 28235 & 22.94809 & 25 & g & 17.63 & 0.34 & 1.04 & 0.15 & 0.63 + lwp 28236 & 23.01624 & 25 & v & 18.14 & 0.34 & 0.87 & 0.15 & 0.69 + lwp 28237 & 23.06579 & 25 & v & 18.45 & 0.35 & 1.30 & 0.15 & 0.73 + lwp 28238 & 23.14927 & 25 & v & 18.25 & 0.35 & 1.14 & 0.15 & 0.73 + lwp 28239 & 23.21620 & 25 & v & 18.34 & 0.35 & 1.09 & 0.15 & 0.55 + lwp 28240 & 23.28145 & 25 & v & 17.83 & 0.35 & 0.92 & 0.15 & 0.69 + lwp 28241 & 23.34773 & 25 & g & 18.28 & 0.35 & 1.20 & 0.15 & 0.72 + lwp 28242 & 23.41432 & 25 & g & 18.16 & 0.35 & 1.24 & 0.16 & 0.67 + ccccccccc + iue image & observation & exposure & observatory & @xmath75 & @xmath76 & @xmath77 & @xmath78 & @xmath24 + & midpoint ( ut ) & time & ( goddard & ( mjy ) & ( mjy ) & & & + & ( day of may 94 ) & ( min ) & or vilspa ) & & & & & + lwp 28243 & 23.48093 & 25 & g & 18.39 & 0.35 & 1.26 & 0.15 & 0.71 + lwp 28244 & 23.54752 & 25 & g & 18.39 & 0.36 & 1.28 & 0.16 & 0.63 + lwp 28245 & 23.61417 & 25 & g & 18.15 & 0.37 & 0.74 & 0.17 & 0.77 + lwp 28246 & 23.68069 & 25 & g & 17.59 & 0.54 & 0.52 & 0.29 & 0.69 + lwp 28247 & 23.74660 & 25 & g & 18.83 & 0.58 & 0.85 & 0.29 & 0.54 + lwp 28248 & 23.81392 & 21 & g & 18.95 & 0.36 & 1.12 & 0.15 & 0.54 + lwp 28249 & 23.94704 & 25 & g & 18.69 & 0.36 & 1.18 & 0.15 & 0.64 + lwp 28250 & 24.01432 & 25 & v & 18.64 & 0.35 & 1.17 & 0.15 & 0.69 + lwp 28251 & 24.08010 & 25 & v & 18.85 & 0.36 & 1.28 & 0.15 & 0.74 + lwp 28252 & 24.14703 & 25 & v & 18.66 & 0.35 & 1.06 & 0.15 & 0.66 + lwp 28253 & 24.21411 & 25 & v & 19.57 & 0.37 & 1.35 & 0.15 & 0.79 + lwp 28254 & 24.28127 & 25 & v & 19.09 & 0.36 & 1.14 & 0.15 & 0.84 + lwp 28256 & 24.67891 & 25 & v & 20.00 & 0.37 & 0.81 & 0.14 & 0.80 + lwp 28257 & 24.74557 & 25 & g & 20.43 & 0.38 & 0.80 & 0.14 & 0.77 + lwp 28258 & 24.81200 & 25 & g & 20.66 & 0.38 & 1.30 & 0.14 & 0.68 + lwp 28259 & 24.94266 & 25 & g & 19.59 & 0.42 & 1.16 & 0.18 & 0.57 + lwp 28262 & 25.67900 & 17 & g & 20.34 & 0.39 & 0.52 & 0.15 & 0.66 + lwp 28263 & 25.74547 & 25 & g & 20.86 & 0.40 & 0.65 & 0.15 & 0.62 + + + + + + cccccccc + & observation & @xmath80 & @xmath81 & @xmath25 & @xmath82 & @xmath24 + & midpoint ( ut ) & ( mjy ) & ( mjy ) & & & + swp & lwp & ( day of may 94 ) & & & & & + 50779 & 28143 & 16.16939 & 13.32 & 0.18 & 0.83 & 0.04 & 0.88 + 50780 & 28144 & 16.23701 & 13.43 & 0.18 & 0.84 & 0.04 & 0.95 + 50781 & 28145 & 16.30453 & 13.60 & 0.18 & 0.88 & 0.04 & 0.90 + 50782 & 28146 & 16.36971 & 13.51 & 0.18 & 0.87 & 0.04 & 0.85 + 50783 & 28147 & 16.43697 & 13.49 & 0.18 & 0.85 & 0.04 & 1.06 + 50784 & 28148 & 16.50320 & 13.38 & 0.18 & 0.84 & 0.04 & 1.03 + 50785 & 28149 & 16.56976 & 13.54 & 0.19 & 0.87 & 0.04 & 1.00 + 50786 & 28150 & 16.63905 & 14.01 & 0.20 & 0.89 & 0.05 & 0.85 + 50787 & 28151 & 16.70290 & 13.75 & 0.22 & 0.96 & 0.07 & 0.49 + 50788 & 28152 & 16.76824 & 13.92 & 0.22 & 0.95 & 0.07 & 0.74 + 50789 & 28153 & 16.83202 & 14.07 & 0.19 & 0.86 & 0.04 & 0.81 + 50790 & 28154 & 16.97156 & 13.62 & 0.18 & 0.82 & 0.04 & 1.16 + 50791 & 28155 & 17.03536 & 13.95 & 0.19 & 0.89 & 0.04 & 1.15 + 50792 & 28156 & 17.10154 & 13.79 & 0.18 & 0.84 & 0.04 & 1.12 + 50793 & 28157 & 17.16822 & 13.75 & 0.18 & 0.86 & 0.04 & 1.12 + 50794 & 28158 & 17.23492 & 13.84 & 0.19 & 0.84 & 0.04 & 1.03 + 50795 & 28159 & 17.30233 & 14.11 & 0.19 & 0.86 & 0.04 & 0.98 + 50796 & 28160 & 17.36928 & 14.01 & 0.19 & 0.86 & 0.04 & 1.07 + 50797 & 28161 & 17.43554 & 14.05 & 0.19 & 0.85 & 0.04 & 1.21 + 50798 & 28162 & 17.50215 & 14.10 & 0.19 & 0.88 & 0.04 & 1.15 + 50799 & 28163 & 17.56883 & 14.25 & 0.19 & 0.87 & 0.04 & 1.00 + 50800 & 28164 & 17.63511 & 14.03 & 0.20 & 0.96 & 0.04 & 0.81 + cccccccc + & observation & @xmath80 & @xmath81 & @xmath25 & @xmath82 & @xmath24 + & midpoint ( ut ) & ( mjy ) & ( mjy ) & & & + swp & lwp & ( day of may 94 ) & & & & & + 50801 & 28165 & 17.69974 & 13.85 & 0.23 & 0.90 & 0.08 & 0.70 + 50802 & 28166 & 17.76314 & 14.42 & 0.25 & 1.00 & 0.09 & 0.70 + 50803 & 28167 & 17.83215 & 14.11 & 0.19 & 0.98 & 0.04 & 0.98 + 50804 & 28168 & 17.96642 & 13.56 & 0.18 & 0.96 & 0.04 & 0.85 + 50805 & 28169 & 18.03042 & 13.70 & 0.18 & 0.89 & 0.04 & 1.20 + 50806 & 28170 & 18.09446 & 13.34 & 0.18 & 0.91 & 0.04 & 1.03 + 50808 & 28171 & 18.36811 & 13.25 & 0.18 & 0.90 & 0.04 & 1.02 + 50809 & 28172 & 18.43370 & 13.41 & 0.18 & 0.95 & 0.04 & 1.14 + 50810 & 28173 & 18.50090 & 13.13 & 0.18 & 0.90 & 0.04 & 1.16 + 50811 & 28174 & 18.56951 & 13.07 & 0.18 & 0.90 & 0.04 & 1.00 + 50812 & 28175 & 18.63421 & 13.05 & 0.18 & 0.92 & 0.04 & 0.83 + 50813 & 28176 & 18.69677 & 13.14 & 0.22 & 1.02 & 0.08 & 0.70 + 50814 & 28177 & 18.76174 & 13.09 & 0.22 & 0.88 & 0.09 & 0.68 + 50816 & 28183 & 19.03461 & 12.81 & 0.17 & 0.89 & 0.04 & 1.03 + 50817 & 28184 & 19.10021 & 12.74 & 0.17 & 0.89 & 0.04 & 1.01 + 50818 & 28185 & 19.16821 & 12.79 & 0.17 & 0.85 & 0.04 & 1.12 + 50819 & 28186 & 19.23239 & 12.96 & 0.18 & 0.85 & 0.04 & 1.03 + 50821 & 28188 & 19.36642 & 12.92 & 0.18 & 0.86 & 0.04 & 0.93 + 50822 & 28189 & 19.43350 & 12.90 & 0.18 & 0.88 & 0.04 & 0.86 + 50823 & 28190 & 19.49983 & 12.98 & 0.18 & 0.81 & 0.04 & 1.05 + 50824 & 28191 & 19.56608 & 13.20 & 0.18 & 0.84 & 0.04 & 1.13 + 50825 & 28192 & 19.63337 & 13.04 & 0.18 & 0.86 & 0.05 & 0.65 + cccccccc + & observation & @xmath80 & @xmath81 & @xmath25 & @xmath82 & @xmath24 + & midpoint ( ut ) & ( mjy ) & ( mjy ) & & & + swp & lwp & ( day of may 94 ) & & & & & + 50826 & 28193 & 19.69594 & 12.84 & 0.22 & 0.93 & 0.09 & 0.58 + 50827 & 28194 & 19.76089 & 13.21 & 0.23 & 0.96 & 0.09 & 0.73 + 50828 & 28195 & 19.83112 & 13.37 & 0.18 & 0.95 & 0.04 & 0.90 + 50829 & 28196 & 19.96575 & 13.60 & 0.18 & 0.84 & 0.04 & 1.01 + 50830 & 28197 & 20.03319 & 14.08 & 0.19 & 0.83 & 0.04 & 0.86 + 50831 & 28198 & 20.09990 & 14.49 & 0.19 & 0.83 & 0.04 & 0.99 + 50832 & 28199 & 20.16596 & 14.58 & 0.20 & 0.79 & 0.04 & 1.01 + 50833 & 28200 & 20.23311 & 15.03 & 0.20 & 0.83 & 0.04 & 0.93 + 50834 & 28201 & 20.29966 & 14.88 & 0.20 & 0.83 & 0.04 & 1.05 + 50835 & 28202 & 20.36594 & 15.13 & 0.20 & 0.84 & 0.04 & 1.14 + 50836 & 28203 & 20.43272 & 15.33 & 0.20 & 0.91 & 0.04 & 1.29 + 50837 & 28204 & 20.49974 & 15.37 & 0.20 & 0.87 & 0.04 & 0.94 + 50838 & 28205 & 20.56612 & 15.48 & 0.21 & 0.86 & 0.04 & 1.26 + 50839 & 28206 & 20.63278 & 15.26 & 0.21 & 0.99 & 0.05 & 0.72 + 50841 & 28208 & 20.76070 & 16.20 & 0.28 & 0.92 & 0.09 & 0.65 + 50842 & 28209 & 20.82994 & 17.44 & 0.23 & 0.93 & 0.04 & 1.02 + 50843 & 28210 & 20.96575 & 17.31 & 0.23 & 0.91 & 0.03 & 1.18 + 50844 & 28211 & 21.03256 & 17.49 & 0.23 & 0.89 & 0.03 & 1.06 + 50845 & 28212 & 21.09926 & 17.41 & 0.23 & 0.88 & 0.03 & 1.28 + 50846 & 28213 & 21.16615 & 17.44 & 0.23 & 0.84 & 0.03 & 0.90 + 50847 & 28214 & 21.23246 & 17.46 & 0.23 & 0.84 & 0.04 & 1.06 + 50848 & 28215 & 21.29944 & 17.52 & 0.23 & 0.86 & 0.03 & 1.06 + cccccccc + & observation & @xmath80 & @xmath81 & @xmath25 & @xmath82 & @xmath24 + & midpoint ( ut ) & ( mjy ) & ( mjy ) & & & + swp & lwp & ( day of may 94 ) & & & & & + 50849 & 28216 & 21.36610 & 17.28 & 0.23 & 0.90 & 0.03 & 1.09 + 50850 & 28217 & 21.43163 & 17.10 & 0.23 & 0.90 & 0.04 & 1.16 + 50851 & 28218 & 21.49824 & 17.04 & 0.22 & 0.84 & 0.03 & 1.16 + 50852 & 28219 & 21.56520 & 16.99 & 0.23 & 0.86 & 0.04 & 0.80 + 50853 & 28220 & 21.63216 & 16.16 & 0.23 & 1.07 & 0.05 & 0.93 + 50856 & 28223 & 21.82905 & 14.75 & 0.20 & 1.08 & 0.04 & 1.17 + 50857 & 28224 & 21.96654 & 13.73 & 0.19 & 0.95 & 0.04 & 1.06 + 50858 & 28225 & 22.03217 & 13.54 & 0.18 & 0.95 & 0.04 & 1.01 + 50859 & 28226 & 22.09781 & 12.92 & 0.18 & 1.00 & 0.04 & 0.96 + 50861 & 28227 & 22.36525 & 12.75 & 0.17 & 0.94 & 0.04 & 0.94 + 50862 & 28228 & 22.43277 & 12.69 & 0.17 & 0.96 & 0.04 & 1.03 + 50863 & 28229 & 22.49748 & 12.70 & 0.17 & 0.91 & 0.04 & 1.27 + 50864 & 28230 & 22.56410 & 12.81 & 0.17 & 0.88 & 0.04 & 0.86 + 50865 & 28231 & 22.63074 & 12.81 & 0.18 & 1.05 & 0.05 & 0.72 + 50866 & 28232 & 22.69263 & 12.81 & 0.22 & 1.02 & 0.09 & 0.67 + 50867 & 28233 & 22.76174 & 12.60 & 0.20 & 0.94 & 0.08 & 0.69 + 50868 & 28234 & 22.82696 & 12.55 & 0.17 & 1.05 & 0.04 & 1.01 + 50869 & 28235 & 22.96410 & 13.32 & 0.18 & 0.88 & 0.04 & 1.19 + 50870 & 28236 & 23.03028 & 13.68 & 0.18 & 0.93 & 0.04 & 1.10 + 50871 & 28237 & 23.08910 & 13.64 & 0.18 & 0.90 & 0.04 & 1.47 + 50872 & 28238 & 23.16539 & 13.63 & 0.18 & 0.90 & 0.04 & 1.07 + 50873 & 28239 & 23.23104 & 13.61 & 0.18 & 0.93 & 0.04 & 0.98 + cccccccc + & observation & @xmath80 & @xmath81 & @xmath25 & @xmath82 & @xmath24 + & midpoint ( ut ) & ( mjy ) & ( mjy ) & & & + swp & lwp & ( day of may 94 ) & & & & & + 50874 & 28240 & 23.29742 & 12.95 & 0.18 & 1.04 & 0.04 & 1.18 + 50875 & 28241 & 23.36335 & 12.95 & 0.17 & 1.05 & 0.04 & 0.98 + 50876 & 28242 & 23.43009 & 13.23 & 0.18 & 0.97 & 0.04 & 1.08 + 50877 & 28243 & 23.49656 & 13.46 & 0.18 & 0.91 & 0.04 & 0.99 + 50878 & 28244 & 23.56297 & 13.58 & 0.18 & 0.89 & 0.04 & 0.99 + 50879 & 28245 & 23.62984 & 13.08 & 0.19 & 1.03 & 0.05 & 0.86 + 50880 & 28246 & 23.69373 & 12.73 & 0.21 & 1.11 & 0.08 & 0.67 + 50881 & 28247 & 23.76311 & 13.20 & 0.20 & 1.06 & 0.07 & 0.90 + 50882 & 28248 & 23.82520 & 13.05 & 0.18 & 1.16 & 0.04 & 0.97 + 50883 & 28249 & 23.96293 & 13.88 & 0.18 & 0.91 & 0.04 & 1.14 + 50884 & 28250 & 24.03092 & 13.03 & 0.17 & 1.05 & 0.04 & 0.97 + 50885 & 28251 & 24.09589 & 13.91 & 0.18 & 0.90 & 0.04 & 1.14 + 50886 & 28252 & 24.16287 & 14.12 & 0.19 & 0.87 & 0.04 & 1.10 + 50887 & 28253 & 24.23064 & 14.17 & 0.19 & 0.92 & 0.04 & 1.12 + 50889 & 28256 & 24.69450 & 15.47 & 0.21 & 0.86 & 0.04 & 1.28 + 50890 & 28257 & 24.76122 & 15.38 & 0.20 & 0.96 & 0.03 & 1.11 + 50891 & 28258 & 24.82330 & 14.85 & 0.20 & 0.96 & 0.04 & 1.04 + 50894 & 28262 & 25.69474 & 15.53 & 0.21 & 0.93 & 0.04 & 0.98 + 50895 & 28263 & 25.75818 & 15.74 & 0.21 & 0.97 & 0.04 & 0.85 + cccc + range & + & 1991 , swet@xmath83 & 1991 , tomsips & 1994 , tomsips + 1230 - 1950 & [email protected]@xmath85 ( 98@xmath79 ) & [email protected] ( 98 ) & [email protected] ( 115 ) + 2100 - 2800 & & [email protected] ( 97 ) & [email protected] ( 117 ) + 2100 - 3100 & [email protected] ( 97 ) & [email protected] ( 97 ) & + 1230 - 2800 & & [email protected] ( 99 ) & [email protected] ( 107 ) + 1230 - 3100 & [email protected] ( 99 ) & [email protected] ( 99 ) & + + + + + + [ pian , e. , treves , a. , webb , j. , kazanas , d. , maraschi , l. , mccollum , b. , shrader , c. , & wamsteker , w. 1996 , proceedings of the oj-94 project meeting , tuorla observatory reports , no . 176 , ed . l. o. takalo , p. 26 ] [ wagner , s. j. , & witzel , a. 1994 , in the nature of compact objects in active galactic nuclei , proc . of the 33rd herstmonceux conference , eds .
a. robinson and r. terlevich , cambridge university press , p. 397 ] * fig . 1 * typical spectra from the may 1994 iue campaign not corrected for reddening ( upper panels ) . in both cases
the power - law fitting curve @xmath19 is shown as a solid line , with indices _ a ) _ @xmath86 = [email protected] ; _ b ) _
@xmath86 = [email protected] .
the lower panels represent the intrinsic error distributions of the spectral fluxes . * fig . 3 *
dereddened light curves at 1400 ( filled circles ) and 2800 ( open circles ) : _ a ) _ full observing period ( the circled points correspond to underexposed spectra , see text ) ; _ b ) _ expanded view of the initial portion . in the second plot , the light curves are normalized to their respective averages , calculated after excluding the flux points taken during the first day of monitoring . variability is detected on time scales comparable to the exposure times ( up to a factor @xmath02.2 flux change at 2800 in 1.5 hr ) , and more rapid variations are probably present but unresolved .
6 * spectral indices for the dereddened flux distributions in the _ a ) _ 1230 - 1950 band ; _ b ) _ 2100 - 2800 band ; _ c ) _ 1230 - 2800 band .
the circled points correspond to underexposed spectra .
the horizontal solid lines represent the average energy indices in each band . * fig .
7 * cross - correlation function between the flux at 1400 and the swp spectral index computed with the dcf of edelson & krolik ( 1988 ) .
the spectrum flattens @xmath01 day before the flux increases . *
8 * comparison of the light curves at 1400 ( filled ) and at 2800 ( open ) obtained during the present iue campaign ( circles ) and during the intensive monitoring period in november 1991 ( squares ) .
day 1 in the temporal scale corresponds to 10 november for the 1991 data and to 15 may for the 1994 data . for both epochs ,
the light curves have been normalized to the average swp and lwp fluxes in 1991 .
the flux level in 1991 was @xmath020% brighter than in 1994 .
the character of the variability is different at the two epochs : recurrent @xmath020% variations detected in 1991 are not seen in the 1994 data , which exhibit an extremely rapid flux doubling at the beginning of the light curve and a big central flare of @xmath035% amplitude . | pks 2155 - 304 , the brightest bl lac object in the ultraviolet sky , was monitored with the iue satellite at @xmath01 hour time - resolution for ten nearly uninterrupted days in may 1994 .
the campaign , which was coordinated with euve , rosat , and asca monitoring , along with optical and radio observations from the ground , yielded the largest set of spectra and the richest short time scale variability information ever gathered for a blazar at uv wavelengths .
the source flared dramatically during the first day , with an increase by a factor @xmath02.2 in an hour and a half . in subsequent days , the flux maintained a nearly constant level for @xmath05 days , then flared with @xmath035% amplitude for two days .
the same variability was seen in both short- and long - wavelength iue light curves , with zero formal lag ( @xmath1 2 hr ) , except during the rapid initial flare , when the variations were not resolved .
spectral index variations were small and not clearly correlated with flux .
the flux variability observed in the present monitoring is so rapid that for the first time , based on the uv emission alone , the traditional @xmath2 limit indicating relativistic beaming is exceeded . the most rapid variations , under the likely assumption of synchrotron radiation , lead to a lower limit of 1 g on the magnetic field strength in the uv emitting region .
these results are compared with earlier intensive monitoring of pks 2155304 with iue in november 1991 , when the uv flux variations had completely different characteristics . [
= 1.0in=1 ] = cmr7 |
complex decision making tasks over a distributed quantum network , a network including entangled nodes , can be analyzed with a quantum game theory approach .
quantum games extend the applicability of classical games to quantum networks , which may soon be a reality .
quantum game theory imports the ideas from quantum mechanics such as entanglement and superposition , into game theory .
the inclusion of entanglement leads to player outcomes that are correlated so that entanglement often behaves like mediated communication between players in a classical game .
this can lead to a game that has different nash equilibria with greater payoffs than the classical counterpart .
the analysis of quantum games with entanglement can resemble the correlated equilibria of classical games .
the entanglement is imposed by a referee , and acts like a contract that can not be broken between the players , and can persist non - locally after the initial entanglement has been performed and communication forbidden .
this is in contrast to classical correlated equilibria that rely on communication between the players , whose contracts can be broken , and can not exhibit the non - local behavior associated with quantum mechanics . the correlations produced by entanglement can achieve probability distributions over the payoffs that are not possible in the classical game , even when mixed strategies are used .
when interacting with a network , the agents will often have incomplete information about the other nodes .
quantum games with incomplete information can be treated within a bayesian approach . with this approach in mind
, we are interested in quantized games with classical priors , i.e. a statistical mixture of two quantum games . detailed analysis of bayesian quantum games can potentially lead to applications in quantum security protocols@xcite , the development of distributed quantum computing algorithms@xcite , or improving the efficiency of classical network algorithms @xcite .
experiments have begun to demonstrate the results of quantum game theory in nuclear magnetic resonance @xcite , quantum circuits in optical @xcite , and ion - trap platforms @xcite , which , in some cases , i.e. optical , can be easily imagined on a distributed quantum network . to quantize a classical game ,
we follow the approach given in the seminal einstein - wilkens - lewenstein scheme .
the scheme goes as follows ; both players qubits are initialized to the @xmath0 state , an entangling operation , @xmath1 , is applied , the players apply their strategy choice , @xmath2 , an un - entangling operation is applied , the payoffs are determined from the probability distribution of the final state @xmath3 .
this procedure can be encoded in the quantum circuit show in figure [ fig : qpd ] . [ cols="^ " , ] the amount of entanglement that occurs
can be varied by varying the parameter @xmath4 in the entangling operation : @xmath5 at maximal entanglement,@xmath6 , this operation produces a bell state , and at @xmath7 is the identity operator .
the game is defined by setting the possible strategies of the players .
for this we parametrize a single qubit rotation , @xmath8 , with three parameters,@xmath9 in : @xmath10 where @xmath11,\phi \in [ 0,2\pi],\alpha \in [ 0,2\pi]$ ] .
the outcome of the game is given by : @xmath12 and the average payoff @xmath13 is derived from the expectation values of a measurement performed at the end and the payoff vector @xmath14 @xmath15 there are four possible outcomes , @xmath16 .
correspondence to the classical game is made by associating each outcome as one of the classical strategy choices , such that @xmath0 corresponds to confess ( c ) , and @xmath17 corresponds to defect ( d ) , as is illustrated in the canonical prisoner s dilemma game with payoff matrix shown in table [ tab : pdmatrix ] . ' '' ''
@xmath18 & @xmath19 & @xmath20 + ' '' '' @xmath19 & @xmath21 & @xmath22 + ' '' '' @xmath20 & @xmath23 & @xmath24 + the bayesian game is constructed with the protocol laid out by harsanyi@xcite . in the bayesian game
the players have incomplete knowledge about their opponent s payoff matrices .
this is represented by having the players receive a statistical mixture of different payoff matrices .
below we analyze games that are represented by two different payoff matrices .
if , for example , player a s payoff is the same in both matrices while player b s vary , this represents player a having incomplete knowledge about player b s preferences .
if both have different payoffs , this could be interpreted as two players having incomplete knowledge about what game their playing .
this game can be represented by the quantum circuit shown in figure [ fig : quantumcircuit ] .
quantum circuit for bayesian game ] @xmath25 is a unitary operation on the control qubit . @xmath26 and @xmath27 are controlled entangling operations acting on @xmath28 and @xmath29 or@xmath30 and @xmath31 , depending on the state of the control qubit .
this representation is equivalent to playing a statistical mixture of two quantum circuits shown in figure [ fig : qpd ] with different two - player games . the average payoff for player a in the bayesian game is given by : @xmath32 the b player s average payoff is still calculated according to equation [ eq : payoff ] .
the primary solution concept used in game theory is the nash equilibrium .
a nash equilibrium is a set of strategies where neither player could benefit by unilaterally deviating .
the payoff to the player s at the nash equilibrium represents a stable payoff in a repeated game or large ensemble , because it is self - enforcing .
there are refinements to the concept of a nash equilibrium that are used to capture different types of games .
relevant to quantum games is the concept of a correlated equilibrium .
a correlated equilibrium is a game where the player s strategy choices are correlated in some way , such as reacting to advice or a contract , such that probability distributions are possible that are not in the image of the classical game with mixed strategies .
entanglement acts to correlate the player s outcomes in a similar way , except in quantum games the entanglement is imposed by a referee , and once the entanglement is produced , the player s can not break the contract .
the method of best responses is used to find the nash equilibria of the game .
there have been analytic solutions found for certain cases of quantum games(cite ) , but with the aim to examine a wide range of games , including a bayesian framework and asymmetric payoffs , and for these cases , analytic solutions remain elusive .
therefore we adopt a numerical procedure . first , the possible strategies must be chosen .
equation [ eq : strat ] represents a completely arbitrary strategy choice .
it is instructive to analyze the game with a more discretized set of strategies .
we chose a step size for each parameter @xmath33 .
then we compile a list that contains all possible combinations of integer multiples of these steps , within the bounds of the parameters : @xmath34 where @xmath35 represents the ith element of the list .
this set defines the possible strategies of a game .
several of these matrices are redundant , because for example , when @xmath36 , @xmath37 is undefined . to construct the best response list for player a , for each possible strategy choice of player
b , we compute the payoff for player a for each of their possible strategy choices
. then we select the elements which have the highest payoff , or best response , @xmath38 .
this produces a list of a s best responses to each of b s strategy choices : @xmath39 , where j ranges over all possible strategy choices .
then similarly b s best response list is composed,@xmath40 .
the nash equilibria are given by the intersection of the best response functions : @xmath41 .
for the bayesian game , this procedure is straightforwardly extended to three players .
many of the interesting features of the games we examine exist with the stepping parameters @xmath42 , which has a total of 8 unique strategy choices .
the majority of the data are presented with these stepping parameters .
next we examine the behavior of one game in more detail as we step finer in each of the strategy parameters .
this eventually becomes computationally impractical as the step sizes get too small .
for example , the stepping parameters @xmath43 yield 1824 unique strategy choices .
the code in mathematica with this many strategy choices takes 1 hour to compute all of the nash equilibria for the two - player game for all values of entanglement , making solutions to the bayesian game impractical to find with the current method .
here we present the solutions for several two - player games found in the literature @xcite and textbook games . for the sake of space , we will not discuss the real world interpretration of these games , rather we focus on their mathematical propeties .
the data are shown in figure [ fig : bayesiandata ] , where a pair of two - player games are presented with their payoff matrices inset in the form of table [ tab : pdmatrix ] . in the case of asymmetric games , in the two - player games , player a is plotted in blue , and player b is plotted in red .
the payoffs at the nash equilibria are plotted for both players as a function of the entanglement parameter @xmath4 .
the nash equilibria curves come in two types , constant with @xmath4 , or increasing with @xmath4 .
it is instructive to compare these to the solutions of the corresponding classical games . in all cases , the values of the payoff at no entanglement ,
i.e. @xmath44 , match those of the classical game . for games
only have one constant nash equilibria curve , such as in the games labeled ` type b , ' ` deadlock ' , and ` stag hunt , ' there is one nash equilibrium of the classical game and it is pareto optimal . as the entanglement increases
, the nash equilibrium vanishes above some critical value.for games that have nash equlibrium that grows with @xmath4 ( i.e. ` prisoner s dilemma ' ) there is only one nash equilibrium of the classical game but it is not pareto efficient .
the nash equilibra for these games also vanish for some critical value of entanglement .
the vanishing of the nash equilibrium at a critical entanglement has been compared to a phase transition - like behavior@xcite . for games that have two nash equilibrium in the classical game , one pareto optimal
, the pareto optimal solution remains for all values of entanglement , whereas the second nash equlibrium grows and converges with the optimal one at maximal entanglement . in these cases ,
the pareto optimal nash equilibrium does not vanish at some critical entanglement . for two player games with no nash equilibrium classically , such as the ` matching pennies ' game , there are no nash equilibria in the corresponding quantum game
. in short , games with one equilibrium seem to lose that equilibrium at a critical entanglement , and in games with two equilibria , those equilibria persist for all values of entanglement .
becuase our methods are numerical , these observations are not tantamount to formal proofs , and counter - examples may be found , but they are suggestive of a deeper structure .
the ` da s brother ' is an interesting outlier from this categorization .
the classical game has only one nash equilibrium , and it is pareto optimal .
however , as the entanglement increases , a second nash equilibrium appears and then converges to the pareto optimal solution as in the case of games with two equilibria mentioned above .
additionally , the pareto efficient solution does not vanish at some critical value of entanglement .
three - player bayesian games can be composed out of a pair of two - player games .
this can be interpreted as the player s having incomplete information about their opponents payoffs .
the solutions to the bayesian games composed of various two - player game combinations are plotted in 3d below the two - player results in figure [ fig : bayesiandata ] .
the payoff for only player a is plotted against the entanglement , @xmath4 , and the probability to play with each player , @xmath45 from equation [ eq : bpayoff ] .
the bayesian graphs are oriented so that the top two - player game is in the back of the 3d plot with the bottom game in the front , and no entanglement on the right with maximal entanglement on the left .
as expected , the @xmath46 and @xmath47 solutions to the bayesian game match the two - person game solutions for players a and b respectively . along the @xmath7 plane
, the results match those of the classical game with mixed strategies . in games with the same number of nash equilibria in the two component two - player games such as in ` deadlock ' vs. ` prisoner s dilemma ' , the solutions at @xmath46 continuously and linearly transform into the solutions at @xmath47 . when there are a different number of nash equilibria in the two component two - player games , the equilibria must vanish , or appear , at some value of @xmath45 .
this is similar to the vanishing , or appearance , of nash equilibria at a critical entanglement in the two - player games , only here , we also see them in the bayesian game as the degree of incomplete information , @xmath45 , changes . returning to a two - player game
, we now examine the game as the discretization of the strategy choices in equation [ eq : space ] becomes finer , approaching the limit of completely arbitrary su(2 ) rotations .
the nash equilibria of the ` da s brother ' are now calculated using the stepping parameters @xmath48 which yields 1824 unique strategy choices .
as seen in the left graph of figure [ fig : continuous ] , the space between the two nash equilibria becomes filled with many additional equilibria .
the nash equilibria found with @xmath49 form the upper and lower bounds of the new nash equilibria . in the right graph of figure
[ fig : continuous ] the strategy parameters of the nash equilibria for the players are compared by plotting @xmath50 vs @xmath51 against each other for each nash equilibrium .
this shows that the nash equilibria generally follow the trend @xmath52 . taking a slice of the payoff as a function of entanglement data at @xmath53 ,
a histogram of the payoffs achieved suggests that there is some structure within the distribution , as shown in the left hand graph of figure [ fig : continuous2 ] . in this data
the stepping parameters used were @xmath54 , yielding 7968 unique strategy choices .
the data suggest that more nash equilibria occur near the original pareto optimal solution that occured with @xmath55 .
there may be some indication that the nash equilibria are beginning to converge towards the two original nash equilibria however , as computation with finer steps is impractical with our current numerical method , further study is required .
there is also a relationship between the payoff that is rewarded and the strategy choice at each nash equilibrium . in the right side graph of figure
[ fig : continuous2 ] we plot the @xmath50 parameter of each nash equilibria against the payoff of player a. when @xmath56 , the payoff is the pareto optimal solution , which is expected from the results with @xmath55 .
then , as @xmath57 approaches @xmath58 , the other equilibrium of the original game , the payoff transforms to the payoff of the second equilibria .
further study is needed to understand these relationships .
the nash equilibria that arise in a quantum game , where entanglement produces correlations in the player s outputs , can be compared to the correlated equilibrium in classical game theory@xcite .
a correlated equilibrium in classical games can arise when mixed strategies are used and there is communication between the players in the form of advice from a referee or a contract . if players receive some piece of advice , or react in a predetermined way to a random event , they can employ strategies that are correlated with one another and realize self - enforcing equilibria that are different from those in the mixed game without communication .
when entanglement produces correlated outcomes for the players , the equilibria produced strongly resemble the correlated equilibria .
in contrast to the classical case , the role of advice is played by the initial entanglement . and
once that entanglement is imposed on the players by a referee , it forms an effective contract that can not be broken .
the correlations will persist even if the players are not allowed communication after the initial entanglement .
in addition , quantum correlations can exhibit probability distributions that are not allowed by classical correlations , and can persist non - locally .
the sudden dissapearance of a nash equilibrium as the entanglement is increased suggests that the correlations can benefit the player s up to a point , but when the correlations are too strong , the nash equilibrium no longer occurs
. it would be interesting to find examples of classical games where the enforcement of some contract produces a benefit for the players , but if it is enforced too strongly , it ceases to allow a nash equilibrium .
in addition , the abrupt changing of the structure of nash equilibrium as a function of the player s incomplete information could strongly effect any protocol on a network where the agents have some uncertainty about the payoffs or players in the game .
maitra , a. , _ et al .
_ : proposal for quantum rational secret sharing , phys .
a 92 , 022305 ( 2015 ) .
li , q. , he , y. , and jiang , j .-
p . : a novel clustering algorithm based on quantum games , j. phys . a : math . theor .
42 , 445303 ( 2009 ) .
zableta , o.g . ,
barrang , j. p. , and arizmendi c. m. : quantum game application to spectrum scarcity problems , physica a 466 ( 2017 ) .
du , j. , li , h. , xu , x. , shi , m. , wu , j. , zhou , x. , and han , r. : experimental realization of quantum games on a quantum computer , phys . rev .
lett . , 88 , 137902 ( 2002 ) .
prevedel , r. , andre , s. , walther , p. , and zeilinger , a. : experimental realization of a quantum game on a one - way quantum computer , new journal of physics 9 , 205 ( 2007 ) .
buluta , i. m. ; fujiwara , s. ; hasegawa , s. : quantum games in ion traps , physics letters a 358 , 100 ( 2006 ) .
harsanyi , j. c. : games with incomplete information played by bayesian players , mgt .
14 , 159 ( 1967 ) .
flitney , p. and d. abbott , advantage of a quantum player over a classical one in 2x2 quantum games , proc .
a , london , ( 2003 ) .
avishai , y. : some topics in quantum games , masters thesis , ben gurion university of the negev , beer sheva , israel ( 2012 ) .
du , j. , li , h. , xu , x. , zhou , x. , and han , r. : phase - transition - like behaviour of quantum games , j. phys . a : math .
gen 36 p. 6551 - 6562
solmeyer , n. , r. dixon , and r. balu , characterizing the nash equilibria of a three - player bayesian quantum game , _ forthcoming ( 2017 ) .
auman , r. : subjectivity and correlation in randomized strategies , journal of mathematical economics , 1 , p. 67 - 96 , ( 1974 ) .
_ * neal solmeyer * is a physicist the army research laboratory ( adelphi , md ) .
he received his ba degrees in physic and philosophy from carleton college ( northfield , mn ) in 2006 , and his phd degree in physics from penn state university ( state college , pa ) in 2013 .
his research interests include using rydberg excitations in ensembles of laser cooled and trapped rubidium atoms for the purposes of quantum communication , and applying quantum game theory to distributed quantum networks .
biographies and photographs of the other authors are not available . | quantum games with incomplete information can be studied within a bayesian framework .
we analyze games quantized within the ewl framework [ eisert , wilkens , and lewenstein , phys rev .
lett .
83 , 3077 ( 1999 ) ] .
we solve for the nash equilibria of a variety of two - player quantum games and compare the results to the solutions of the corresponding classical games .
we then analyze bayesian games where there is uncertainty about the player types in two - player conflicting interest games .
the solutions to the bayesian games are found to have a phase diagram - like structure where different equilibria exist in different parameter regions , depending both on the amount of uncertainty and the degree of entanglement .
we find that in games where a pareto - optimal solution is not a nash equilibrium , it is possible for the quantized game to have an advantage over the classical version .
in addition , we analyze the behavior of the solutions as the strategy choices approach an unrestricted operation .
we find that some games have a continuum of solutions , bounded by the solutions of a simpler restricted game
. a deeper understanding of bayesian quantum game theory could lead to novel quantum applications in a multi - agent setting . * * * * * neal solmeyer 1 |
ultracold samples of bi - alkali polar molecules have been created very recently in their ground electronic @xmath7 , vibrational @xmath8 , and rotational @xmath9 states @xcite .
this is a promising step before achieving bose - einstein condensates or degenerate fermi gases of polar molecules , provided that further evaporative cooling is efficient . for this purpose
, elastic collision rates must be much faster than inelastic quenching rates .
this issue is somewhat problematic for the bi - alkali molecules recently created , since they are subject to quenching via chemical reactions .
if a reaction should occur , the products are no longer trapped . for alkali dimers that possess electric dipole moments , elastic scattering appears to be quite favorable , since elastic scattering rates are expected to scale with the fourth power of the dipole moment @xcite .
inelastic collisions of polar species can originate from two distinct sources .
the long - range dipole - dipole interaction itself is anisotropic and can cause dipole orientations to be lost .
this kind of loss generally leads to high inelastic rates , and is regarded as the reason why electrostatic trapping of polar molecules is likely not feasible @xcite .
moreover , these collisions also scale as the fourth power of dipole moment in the ultracold limit @xcite , meaning that the ratio of elastic to inelastic rates does not in general improve at higher electric fields .
this sort of loss can be prevented by trapping the molecules in optical dipole traps .
more serious is the possibility that collisions are quenched by chemical reactions .
chemical reaction rates are known to be potentially quite high even at ultracold temperatures @xcite . indeed , for collision energies above the bethe wigner threshold regime
, it appears that many quenching rates , chemical or otherwise , of barrierless systems are well described by applying langevin s classical model @xcite . in this model
the molecules must surmount a centrifugal barrier to pass close enough to react , but are assumed to react with unit probability when they do so .
this model has adequately described several cold molecule quantum dynamics calculations @xcite . within the bethe wigner limit , scattering can be described by an elegant quantum defect theory ( qdt ) approach @xcite .
this approach makes explicit the dominant role of long - range forces in controlling how likely the molecules are to approach close to one another .
consequently , quenching rate constants can often be written in an analytic form that contains a small number of parameters that characterize short - range physics such as chemical reaction probability . for processes in which the quenching probability is close to unity
, the qdt theory provides remarkably accurate quenching rates @xcite . for dipoles
, however , the full qdt theory remains to be formulated . in this article
we combine two powerful ideas suppression of collisions due to long - range physics , and high - probability quenching inelastic collisions for those that are not suppressed to derive simple estimates for inelastic / reactive scattering rates for ultracold fermionic dipoles .
the theory arrives at remarkably simple expressions of collision rates , without the need for the full machinery of close - coupling calculations .
strikingly , the model shows that quenching collisions scale as the _ sixth _ power of the dipole moment for ultracold @xmath3wave collisions . on the one hand
, this implies a tremendous degree of control over chemical reactions by simply varying an electric field , complementing alternative proposals for electric field control of molecule - molecule @xcite or atom - molecule @xcite chemistry . on the other hand
, it also implies that evaporative cooling of polar molecules may become more difficult as the field is increased . in section
ii , we formulate the theoretical model for three dimensional collisions . in section iii , we apply this model to pure two dimensional collisions and conclude in section iv . in the following , quantities are expressed in s.i . units , unless explicitly stated otherwise .
atomic units ( a.u . ) are obtained by setting @xmath10 .
in quantum mechanics , the quenching cross section of a pair of colliding molecules ( or any particles ) of reduced mass @xmath11 for a given collision energy @xmath12 and a partial wave @xmath13 is given by @xmath14 where @xmath15 is the transition matrix element of the quenching process , @xmath16 represents the quenching probability , and the factor @xmath17 represents symmetrization requirements for identical particles @xcite .
if the two colliding molecules are in different internal quantum states ( distinguishable molecules ) , @xmath18 and if the two colliding molecules are in the same internal quantum state ( indistinguishable molecules ) , @xmath19 .
the total quenching cross section of a pair of molecules is @xmath20 . the quenching rate coefficient of a pair of molecules for a given temperature @xmath21 ( collisional event rate ) is given by @xmath22 where @xmath23 } \end{aligned}\ ] ] is the maxwell boltzmann distribution for the relative velocities for a given temperature and @xmath24 is the maxwell boltzmann constant .
the total quenching rate coefficient of a pair of molecules is @xmath25 . to avoid confusion ,
we will also write the corresponding rate equation for collisions between distinguishable and indistinguishable molecules . first , we consider collisions between two distinguishable molecules in quantum states @xmath26 and @xmath27 ( @xmath18 in eq . ) . during a time @xmath28 , where @xmath29 is the time of a quenching collisional event , the number of molecules @xmath30 lost in each collision is one and the number of molecules @xmath31 lost in each collision
is one . then @xmath32 and @xmath33 .
the volume per colliding pairs of molecules is @xmath34 , where @xmath35 stands for the volume of the gas . during the time @xmath29
, the quenching collisional event is associated with a volume @xmath36 . by definition of @xmath29
, this volume should be equal to the one occupied by just one colliding pair of molecules .
then we get @xmath37 . the rate equation for the number of molecule @xmath30 or @xmath31
is then given by @xmath38 if @xmath39 and @xmath40 are the densities of molecule @xmath26 and @xmath27 in the gas , then @xmath41 we consider now the case of collisions between two indistinguishable molecules ( @xmath19 in eq . ) . during the time @xmath28 the number of molecules @xmath42 lost in each collision is two .
then we get @xmath43 .
the volume per colliding pairs of molecules is @xmath44 where we have taken into account the indistinguishability of the molecules .
for the same reason explained above , the volume associated with the collisional event during the time @xmath29 should be equal to the volume occupied by just one colliding pair of molecules .
and then @xmath45 . the rate equation for the number of molecule @xmath42
is then given by @xmath46 if @xmath47 and @xmath48 , then @xmath49 [ t ] as a function of the intermolecular separation @xmath50 .
@xmath51 and @xmath52 denote the height and the position of the centrifugal barrier .
[ spag - fig ] , width=302 ] we consider the case of two identical ultracold fermionic polar molecules , as has been achieved very recently for krb dimers @xcite in their ro - vibronic ( @xmath53 ) ground state . under these circumstances , because of fermi exchange symmetry , the relative orbital angular momentum quantum number @xmath54 between the two molecules must take odd values @xmath55 .
these molecules are polar molecules and can be controlled by an electric field @xmath56 . in the usual basis set of partial waves @xmath57 , the long - range behavior of two colliding polar molecules in a presence of an electric field is governed by an interaction potential matrix whose elements are @xmath58 where @xmath50 denotes the distance between the two molecules .
the diagonal elements represent effective potentials for the colliding molecules and the non - diagonal elements represent couplings between them .
the coefficient @xmath59 is the van der waals coefficient , assumed to be isotropic in the present treatment .
the @xmath60 is the term corresponding to the electric dipole - dipole interaction expressed in the partial wave basis set @xmath61 between two polarized molecules in the electric field direction , with @xmath62 , where @xmath63 is the induced electric dipole moment , and @xmath64 represent the relative orientation between the molecules . in the basis set of partial waves , @xmath65 takes the form @xmath66 the large bracket symbols denote the usual 3@xmath67 coefficients .
the coefficient @xmath68 is introduced to simplify further notations . the combination between repulsive and attractive terms in the effective potentials ( diagonal terms ) of eq .
generate a potential barrier of height @xmath51 which is plotted schematically in fig .
[ spag - fig ] .
the height of this barrier plays a crucial role as it can prevent the molecule from accessing the short range region where reactive chemistry occurs .
the quantum threshold ( qt ) model consists of two conditions .
first , for @xmath69 , we use the bethe wigner threshold laws @xcite for ultracold scattering .
second , we use the classical capture model ( langevin model ) @xcite to estimate the probability of quenching for @xmath70 .
a classical capture model is indiferent to collision energies @xmath69 since the barrier prevents the molecules from coming close together . in real - life quantum scattering , collisions do occur at these energies due to quantum tunneling , and they are the ones relevant to ultracold collisions . moreover , collisions in this energy regime are dictated by the the bethe wigner quantum threshold laws .
for quenching collisions , the threshold laws @xcite state that @xmath71 . for @xmath72 , a classical capture model will guarantee to deliver the molecule pair to small values of @xmath50 , where chemical reactions are likely to occur with unit probability ( see fig .
[ spag - fig ] ) . following this classical argument
, we will assume that when @xmath73 , the quenching probability reaches unitarity @xmath74 .
using this assumption together with the quantum threshold laws , the qt quenching tunneling probability below the barrier can be written as @xmath75 consequently , the quenching cross section and rate coefficient are approximated by @xmath76 for @xmath69 .
the qt model has the simple and intuitive advantage of showing how the cross sections and rate coefficients scale with the height of the entrance centrifugal barrier . for @xmath72 ,
it is easy to find the corresponding expression of the cross section in eq . by setting @xmath74 .
the cross section @xmath77 will reach the unitarity limit at @xmath72 .
it is also easy to find the corresponding expression of the rate coefficient in eq . .
the qt model is general for any collision between two particles provided that there is a barrier in the entrance collision channel and that chemical reactions occur with near unit probability at short range .
the only information on short range chemistry is that chemical reactions occur at full and unit probability and the only information on long range physics is provided by the height of the entrance barrier @xmath51 .
the qt model describes the background scattering process , it does not take into account scattering resonances .
note that the model will not be appropriate in the present form for barrierless ( @xmath78wave ) collisions since @xmath79 .
for this particular type of collisions that do not possess a centrifugal barrier , the qdt theory can be usefully applied @xcite .
the present form of the qt model does not take into account the anisotropy of the intermolecular potential at intermediate range and/or the electronic and nuclear spin structure of the molecular complex but remains suitable as far as the entrance centrifugal barrier takes place at long range .
the qt model will have to be modified if longer range interactions takes place .
for example , collisions between @xmath9 and @xmath80 polar molecules can have long range interactions between hyperfine states due to dipolar and hyperfine couplings @xcite .
however , for collisions between rotationless @xmath9 polar molecules , the hyperfine couplings are weak and the qt model can be applied without further modifications .
[ h ] k + @xmath81k@xmath82 as a function of the collision energy for the partial wave @xmath83 : ( i ) calculated with a full quantum calculation ( solid lines ) , reproduced from ref .
@xcite ( ii ) using the qt model ( dashed lines ) ( iii ) fitting the qt model ( dotted line ) , using @xmath84 in eq . .
[ crossk3-fig ] , width=302 ] [ h ] k + @xmath81k@xmath82 as a function of the collision energy for the partial wave @xmath4 : comparison between the full quantum calculation ( solid line ) and the qt model ( dashed line ) .
the fitted qt model appears as a dotted line .
the height of the barrier @xmath51 and the corrected height @xmath85 ( @xmath86 ) appear as vertical lines .
[ probak3-fig ] , width=302 ] in the absence of an electric field in eq .
, the long range potential reduces to a diagonal term in the basis set of partial waves .
the position and height of the barrier are given by @xmath87 we can insert eq .
in eq . to get analytical forms of the quenching cross section or rate coefficient . for two indistinguishable fermionic polar molecules at ultracold temperatures
when @xmath4 , we get @xmath88 in eq .
, we used the fact that @xmath89 in three dimensions . note that to get the overall contribution for a given @xmath54 , we have to multiply eq . by the degeneracy factor
@xmath90 corresponding to all values of @xmath91 .
we can get similar expressions for any partial wave @xmath54 . to test the validity of the model , we compare in fig .
[ crossk3-fig ] the quenching cross sections of @xmath81k@xmath92 + @xmath81k@xmath93 collisions as a function of the collision energy for the partial waves @xmath83 : ( i ) calculated in ref .
@xcite with a full quantum time - independent close - coupling calculation based on hyperspherical democratic coordinates @xcite and the full potential energy surface of k@xmath94 ( solid lines ) ( ii ) using the simple qt model ( dashed lines ) with a value of @xmath95 a.u . given in ref .
@xcite ( 1 a.u .
= 1 @xmath96 where @xmath97 is the hartree energy and @xmath98 is the bohr radius ) . in this example
, the qt model provides an upper limit to the cross sections .
this is due to the fact that the quenching cross section does not reach a maximum value at the height of the barrier @xmath51 , but rather at somewhat higher energy , say @xmath85 , with @xmath99 ( see ref .
@xcite ) . for all partial waves ,
there is a worse agreement for collision energies in the vicinity of the height of the barrier where the passage from the ultralow regime to the unitarity limit is smoother than for the qt model .
this smoother passage is visible in fig .
[ probak3-fig ] for the full quantum @xmath4 quenching probability compared to the qt model , which has a sharp corner in the vicinity of @xmath51 . to account for more flexibility in the qt model
, one can replace @xmath51 in eq .
by @xmath85 ( @xmath99 ) , and use the coefficient @xmath100 as a fitting parameter to reproduce either full quantum calculations or experimental observed data .
alternatively , we can correct the qt quenching tunneling probability with an overall factor @xmath101 , @xmath102 with @xmath103 .
@xmath104 can be interpreted as the quenching probability reached at the height of the barrier @xmath51 in the qt model , rather than the rough full unit probability ( @xmath105 ) . as an example , we find that @xmath106 reproduces the quantum @xmath4 partial wave cross section for @xmath81k + @xmath81k@xmath82 ( dotted line in fig .
[ crossk3-fig ] and fig .
[ probak3-fig ] ) .
this yields a maximum quenching probability of @xmath107 instead of 1 .
in other words , the qt model is only a factor of @xmath108 higher than the full quantum calculation for @xmath81k + @xmath81k@xmath82 collisions at ultralow energies .
given the fact that full quantum calculations are computationally demanding @xcite and impossible at the present time for alkali molecule - molecule collisions , the accuracy of the qt model is satisfactory and can be a quick and powerful alternative way to estimate orders of magnitude for the scattering observables . besides
, agreement between the qt model with experimental data or full quantum calculations is expected to be satisfactory for collisions involving alkali species , because it is likely that short range quenching couplings will dominate and lead to high quenching probability in the region where the two particles are close together @xcite . very recently , eq . of the qt model has been applied for the evaluation of ultracold chemical quenching rate of collisions of two @xmath109k@xmath110rb molecules in the same internal quantum state , and provided good agreement with the experimental data @xcite .
[ h ] as a function of the induced dipole moment @xmath0 for the partial waves @xmath111 ( red curves ) and @xmath112 ( blue curves ) .
[ barriers - fig ] , width=302 ] [ h ] k@xmath110rb molecules as a function of the induced electric dipole moment for @xmath4 and for a temperature of @xmath113 nk ( black curves ) .
the rates have been calculated using the barrier heights of fig .
[ barriers - fig ] .
the red lines represent the @xmath111 partial wave contribution .
the blue lines represent the sum of @xmath114 and @xmath115 partial wave contributions .
the dashed lines represent the rates calculated with the diabatic barriers while the solid lines with the adiabatic barriers ( see text for detail ) .
the total , @xmath116 and @xmath117 curves have been indicated in the left hand side .
[ rate - num - fig ] , width=302 ] in the presence of an electric field in eq .
, the long - range interaction potential matrix is no more diagonal and couplings between different values of @xmath54 occur .
@xmath91 is still a good quantum number . a first approximation ( diabatic approximation )
consists of neglecting these couplings and using only the diagonal elements of the diabatic matrix directly
. then one can find numerically for which @xmath50 the centrifugal barriers are maximum and evaluate the height of the diabatic barriers @xmath118 .
this is repeated for all values of the induced dipole moment @xmath0 . a second approximation ( adiabatic approximation )
is to diagonalize this matrix ( including the non - diagonal coupling terms ) for each @xmath50 and again find the maximum of the centrifugal barriers to get the height of the adiabatic barriers @xmath119 .
as an example , we compute these barrier heights for @xmath109k@xmath110rb@xmath120@xmath109k@xmath110rb collisions , using a value of @xmath121 a.u . @xcite .
we plot in fig .
[ barriers - fig ] the heights of the diabatic ( dashed lines ) and adiabatic ( solid lines ) barriers for the quantum numbers @xmath116 ( red curves ) and @xmath117 ( blue curves ) .
the adiabatic barriers have been calculated using five partial waves @xmath122 in eq . .
the effect of the couplings can be clearly seen in this figure by comparing diabatic and adiabatic barriers . especially for the @xmath117 case for @xmath123 d ( 1 d = 1 debye = @xmath124 c.m ) , couplings with higher partial waves
make the adiabatic barrier decrease as the dipole increases while the diabatic barrier continues to increase .
using these heights of the barriers , we use eq .
to plot in fig . [ rate - num - fig ] the total quenching rate coefficients ( black curves ) as a function of @xmath0 for two indistinguishable fermionic polar @xmath109k@xmath110rb molecules in the same quantum state for @xmath4 and at a typical experimental temperature of @xmath113 nk @xcite .
for @xmath113 nk , the mean collision energy @xmath125 nk , and the maximum dipole moment for which @xmath126 nk ( that is for which eq
. does not apply anymore ) is around @xmath127 d ( see fig .
[ barriers - fig ] ) .
the dashed curves correspond to rates calculated with the diabatic approximation while the solid curves correspond to rates calculated with the adiabatic approximation .
the @xmath116 contribution is plotted in red and the contribution of @xmath128 and @xmath129 is plotted in blue .
the rates highly reflect the behavior of the centrifugal barriers in the entrance collision channel .
when the barrier increases with the dipole , it prevents the molecules from getting close together and the quenching rates decreases .
when the barrier decreases , the tunneling probability is increased allowing the molecules to get close together , and the quenching rates increases .
[ t ] but we use the analytical expressions for the rates ( see text for detail ) .
the total , @xmath116 and @xmath117 curves have been indicated in the left hand side .
the individual analytical curves have been indicated in the right hand side by roman numbers .
[ rate - anal - fig ] , width=302 ] in order to have an intuitive sense of how the chemical quenching rate scales with the induced dipole moment ( and the electric field ) , we evaluate analytical expressions of the barriers and the rates as it has been done in the previous section for a zero electric field . the analytical expression of the height of the diabatic barrier @xmath118 is complicated by the occurrence of two distinct long - range potentials in the diagonal matrix term of eq . .
we circumvent this difficulty by looking in the two limits where one dominates over the other . for small electric fields ,
we use the zero electric field limit discussed in the preceding section by setting @xmath130 . for larger electric fields
we ignore the @xmath59 coefficient in eq . if the electric dipole - dipole interaction is attractive ( positive @xmath65 ) .
we ignore the centrifugal term in eq . if the electric dipole - dipole interaction is repulsive ( negative @xmath65 ) .
these two cases are discussed below . in between , to accommodate the transition between the low - field and high - field limit , we will simply add the rate coefficients derived in the two limiting cases . for positive @xmath65 coefficients in eq .
, @xmath131 is attractive in eq . .
for @xmath4 partial waves for example , this occurs when @xmath116 , which favors an attractive orientation of dipoles .
we consider @xmath132 in eq . .
in this case , the position and height of the barrier are given by @xmath133 the position of the barrier in eq .
has to be in the region where eq .
is satisfied .
this happens for suitably large dipole moments , @xmath134 where @xmath135 the subscript _ a _ stands for the attractive interaction .
for two indistinguishable fermionic polar @xmath109k@xmath110rb molecules , and for @xmath4 and @xmath116 , @xmath136 and we get @xmath137 d. the threshold laws for quenching collisions in an electric field are the same as in the zero - field limit . consequently , the quenching cross sections and rate coefficients behaves as in eq . except that @xmath51 is given now by eq . and varies with @xmath0
. we can insert eq . in eq .
to get the corresponding analytical expressions . for a partial wave @xmath2 , the quenching rate scales as @xmath1 . for two indistinguishable fermionic polar molecules at ultracold temperatures when
@xmath4 and @xmath116 , we get @xmath138 thus the @xmath111 quenching rate increases as @xmath139 .
this is a more rapid dependence on dipole moment than for purely long - range dipolar relaxation in dipolar gases @xcite . for negative @xmath65 coefficients in eq .
, @xmath131 is repulsive in eq . .
for @xmath4 partial waves for example , this occurs when @xmath140 , which favors a repulsive orientation of dipoles .
we consider @xmath141 in eq . .
the long - range potential again experiences a barrier , but now it is generated by the balance between the repulsive dipole potential at large @xmath50 , and the attractive van der waals potential at somewhat smaller @xmath50 . in this case , the position and height of this barrier are given by @xmath142 for this approximation to hold , the position of the barrier in eq .
has to be in the region where eq .
is satisfied .
this requires that @xmath143 where @xmath144 the subscript _
_ stands for the repulsive interaction . for two indistinguishable fermionic polar @xmath109k@xmath110rb molecules , and for @xmath4 and @xmath145 or @xmath129 , @xmath146 and we get @xmath147 d. we can replace eq . in eq . to get the corresponding analytical expressions . for a partial wave @xmath2 , the quenching processes scale as @xmath148 . for two indistinguishable fermionic polar molecules at ultracold temperatures
when @xmath4 and @xmath117 , we get @xmath149 the @xmath112 quenching rate decreases as @xmath150 as the electric field grows . these analytical expressions use the diabatic barriers .
if we consider that at large @xmath0 , the total rate is mostly given by the @xmath116 contribution ( we neglect the @xmath117 contributions at large @xmath0 ) , one can have an analytical expression using the adiabatic barrier .
if we take into account the couplings between @xmath111 and @xmath151 , we can diagonalize analytically the @xmath152 matrix in eq . .
it can be shown that for each dipole moment @xmath0 , the coupling with @xmath153 lower the diabatic barrier of @xmath154 by a factor of 0.76 at the position of the barrier , to give rise to the adiabatic barrier . inserting this correction of the barrier in eq .
, this yields a correction of @xmath155 for @xmath156 .
the difference between diabatic and adiabatic calculations can be already seen in fig .
[ barriers - fig ] and fig .
[ rate - num - fig ] for the numerical barriers at large dipole moment . in fig .
[ rate - anal - fig ] the black curve corresponds to the total quenching rate coefficient as a function of @xmath0 for two indistinguishable fermionic polar @xmath109k@xmath110rb molecules in the same quantum state for @xmath4 and at a temperature of @xmath113 nk .
the analytical expressions i , ii , iii , iv , v ( green thin lines ) correspond respectively to eq . , 2 @xmath157 eq . , eq . , 2 @xmath157 eq . , 1.51 @xmath157 eq . .
the curves iii and iv are for the diabatic barriers , while curve v is to account for the adiabatic barrier . the red dashed line ( i+iii )
represents the @xmath111 partial wave contribution for the diabatic barriers while the blue dashed line ( ii+iv ) represents the sum of @xmath114 and @xmath115 partial wave contributions for the diabatic barriers .
the analytical sum i+ii+iii+iv is represented as a black dashed line . to account for the adiabatic barriers we assume that the correction for the total rate comes only from the @xmath111 partial wave , and we replace i+iii by i+v ( red solid line ) .
the analytical sum i+ii+v+iv is represented as a black solid line . neglecting the @xmath150 contribution at larger @xmath0
, the analytical @xmath3wave quenching rate ( taking into account the adiabatic barriers ) is given by the simple expression @xmath158 @xmath159 ( @xmath160 ) is the quenching probability reached at the height of the barrier in the qt model for the zero ( non - zero ) electric field regime .
the qt model assumes that @xmath161 but become fitting parameters ( @xmath162 ) when compared with full quantum calculations or experimental data . the limiting value @xmath163 d ( @xmath164 d ) , where the @xmath5 ( @xmath150 ) behavior begins , has also been indicated with an arrow .
it turns out that the total rates for @xmath4 calculated analytically ( for both the use of diabatic and adiabatic barriers ) are very similar to the numerical ones of fig .
[ rate - num - fig ] ( 10 % difference at most , around @xmath165 ) .
however , the sub - components @xmath111 and @xmath112 have different behaviors .
for example the numerical @xmath111 ( @xmath112 ) component starts to increase ( decrease ) at earlier dipole moment ( typically at 0.02 d ) than their analytical analogs ( typically after 0.06 d ) .
the use of the simple analytical expressions ( using the diabatic or adiabatic barriers ) can be useful to estimate the total rate coefficients , while the numerical ones are prefered to estimate the @xmath111 and @xmath112 individual rates .
[ t ] for the partial waves @xmath166 in two dimensions .
the green thin curves represent the analytical eq .
( constant ) and eq .
( @xmath167 ) .
the dashed black curve is the sum of them .
the solid black curve is the height of the barrier in eq . .
[ barriers2d - fig ] , width=302 ] [ h ] k@xmath110rb molecules as a function of the induced electric dipole moment for the @xmath168 and @xmath169 components at a temperature of @xmath113 nk .
the dashed lines represent the rate using analytical expressions while the solid line represents the rate using the numerical expression ( see text for detail ) .
the individual analytical curves have been indicated in the right hand side by roman numbers .
[ 2d - fig ] , width=302 ] in three dimensional collisions , the quenching loss is largely due to incident partial waves with angular momentum projection @xmath116 , emphasizing head - to - tail orientations of pairs of dipoles .
these are the kind of collisions that are largely suppressed in traps with a pancake geometry , with the dipole polarization axis orthogonal to the plane of the pancake @xcite .
if these collisions can be removed , then it is likely that increasing the electric field will suppress quenching collisions , making evaporative cooling possible .
if we assume an ideal pancake trap that confines the particles to move strictly on a plane , one can apply the present model to estimate the behavior of the quenching processes .
we assume that the molecules are polarized along the electric field axis , perpendicular to the two dimensional plane . in this case , the long range potential is given by @xmath170 where @xmath171 stands for the distance between 2 particles in a two dimensional plane , @xmath172 stands for the angular momentum projection on the electric field axis .
the last term comes from the repulsive dipole - dipole interaction when the dipoles are pointing along the electric field and approach each other side by side .
the height of this barrier has been plotted as a function of @xmath0 in fig .
[ barriers2d - fig ] ( black solid line ) . at ultralow
energy and large molecular separation , the bethe wigner laws for quenching processes depend only on the long - range repulsive centrifugal term @xmath173 @xcite .
the repulsive centrifugal terms are different in eq . and eq . . as the repulsive centrifugal term in eq
. leads to the threshold laws in eq . , the replacement @xmath174 ( that is @xmath175 ) in eq . leads to @xmath176 where @xmath51 denotes the height of the centrifugal barrier in two dimensions .
this result requires that the centrifugal potential is repulsive , i.e. , that @xmath177 . for @xmath178
the threshold law exhibits instead a logarithmic divergence @xcite . in two dimensions ,
quenching cross sections and rate coefficients have respectivelly units of length and length squared per unit of time , and are given by @xcite @xmath179 within this model , it follows that the quenching cross section and rate coefficient for @xmath180 are given by @xmath181 the energy dependence is in agreement with the one found in ref .
@xcite . in eq .
, @xmath182 for the zero - electric field regime and @xmath183 for the non - zero electric field regime .
the height of these barriers has been reported in fig .
[ barriers2d - fig ] ( green thin lines ) .
these results imply that for @xmath166 the quenching processes within this model will be independent of the dipole moment in the zero electric field regime , where @xmath184 and will scale as @xmath6 in the non - zero electric field regime , where @xmath185 we use the fact that @xmath186 in two dimensions .
the non - zero electric field regime is reached when @xmath187 where @xmath188 for two indistinguishable fermionic polar @xmath109k@xmath110rb molecules , and for @xmath166 , we get @xmath189 d. the behavior of the quenching rate ( black lines ) is shown in fig .
[ 2d - fig ] for two indistinguishable fermionic polar @xmath109k@xmath110rb molecules as a function of the induced electric dipole moment for @xmath168 and @xmath169 components at a temperature of @xmath113 nk .
the dashed line represents the analytical rate which is the sum of the analytical expression vi corresponding to @xmath190 eq . and analytical expression vii corresponding to @xmath190 eq . .
the solid line represents the rate using the general expression eq . and the numerical height of the barrier calculated in eq . . the limiting value @xmath191 d , where the @xmath6 behavior for the quenching rate begins , has also been indicated with an arrow .
the difference between the numerical calculation and the analytical expression reflects the difference in the calculation of the height of the barrier , already seen in fig .
[ barriers2d - fig ] .
the numerical calculation is more exact , while the other is analytical .
however , at large @xmath0 , the numerical rate tends to the analytical @xmath6 behavior .
the quenching rate decreases rapidly as the dipole moment increases and this may be promising for efficient evaporative cooling of polar molecules since the elastic rate is expected to grow with increasing dipole moment @xcite .
we have proposed a simple model which combines quantum threshold laws and a classical capture model to determine analytical expressions of the chemical quenching cross section and rate coefficient as a function of the collision energy or the temperature .
we also provide an estimate as a function of the induced electric dipole moment @xmath0 in the presence of an electric field .
we found that the quenching rates of two ultracold indistinguishable fermionic polar molecules grows as the sixth power of @xmath0 . for weaker electric field ,
quenching processes are independent of the induced electric dipole moment .
prospects for two dimensional collisions have been discussed using this model and we predict that the quenching rate will decrease as the inverse of the fourth power of @xmath0 .
this fact may be useful for efficient evaporative cooling of polar molecules .
this model provides a general and comprehensive picture of ultracold collisions in electric fields .
preliminary data suggest that this model gives good agreement with experimental chemical rates for three dimensional collisions in an electric field @xcite .
we acknowledge the financial support of nist , the nsf , and an afosr muri grant .
we thank k .- k .
ni , s. ospelkaus , d. wang , m. h. g. de miranda , b. neyenhuis , p. s. julienne , j. ye , and d. s. jin for helpful discussions . | we use the quantum threshold laws combined with a classical capture model to provide an analytical estimate of the chemical quenching cross sections and rate coefficients of two colliding particles at ultralow temperatures .
we apply this quantum threshold model ( qt model ) to indistinguishable fermionic polar molecules in an electric field . at ultracold temperatures and in weak electric fields , the cross sections and rate coefficients depend only weakly on the electric dipole moment @xmath0 induced by the electric field . in stronger electric fields , the quenching processes scale as @xmath1 where @xmath2 is the orbital angular momentum quantum number between the two colliding particles . for @xmath3wave collisions ( @xmath4 ) of indistinguishable fermionic polar molecules at ultracold temperatures ,
the quenching rate thus scales as @xmath5 .
we also apply this model to pure two dimensional collisions and find that chemical rates vanish as @xmath6 for ultracold indistinguishable fermions .
this model provides a quick and intuitive way to estimate chemical rate coefficients of reactions occuring with high probability
. = cmr7 |
as the downscaling of conventional cmos technology is bound to reach its fundamental limit new algorithms will be the answer to achieve increasingly higher performance and reduced power consumption .
reconfigurable digital circuits provide a way to extend the functionalities of conventional cmos by implementing in the same physical space multiple logic operations and therefore increasing the computational complexity .
reconfiguration of the logic functions at each individual device promises even more compact and flexible circuit design @xcite .
however , the implementation of such reconfigurable logic using single - electron transistors ( sets ) @xcite is appealing because sets have good scalability , one of the lowest energy - per - switching - event @xcite and the possibility to combine their electrical properties with magnetic elements @xcite .
there have been several proposals to implement programmable set logic by using the charge degree of freedom such as fixed gate voltages @xcite , non - volatile charge nodes @xcite and the spin degree of freedom @xcite . in this manuscript
, we show a proof of principle for reconfigurable boolean logic based on magnetically - gated circuit elements and we suggest multi - device reconfigurable logic architectures .
more particularly , we report the complementary logic operation of a aluminium mset .
the reconfigurable capability of our mset stems from the magnetization - dependent work function of gamnas back gate .
when the back - gate is kept at a constant potential the magnetic - field - induced chemical change causes a charge accumulation in the gate electrode which can be readily sensed by the set as a change in the coulomb oscillation phase @xcite .
although the concept is demonstrated here for mset devices , the operating principle is general and transferable to any field - effect transistor .
the mset has two modes of operation : it responds to gate voltage inputs ( electric mode ) as well as to the orientation of the magnetic moments ( magnetic mode ) . by reorienting the magnetization of the gamnas substrate we are able to switch from n - type to p - type mset . making use of the magnetic mode we demonstrate two sets of reprogrammable boolean logic gates implemented at the single device level .
finally , we suggest a strategy to scale the reconfigurable logic operation to the multiple device level . a schematic cross - section of the proposed aluminium mset is depicted in fig .
the key element of this structure is the magnetic gate electrode , an epitaxially grown ga@xmath0mn@xmath1as layer on gaas that acts as a back - gate with easy - axis directions [ 110 ] and [ 1 - 10 ] . a magnetic field @xmath2=0.7 t , larger that the saturation field [email protected] t , is applied to rotate the magnetization @xmath4 with respect to the [ 001 ] direction ( @xmath5=0@xmath6 ) .
the angle @xmath7 with respect to the [ 1 - 10 ] direction is kept equal to 90@xmath6 in all measurements .
the results presented here are independent of the magnitude of @xmath2 for @xmath8 .
a 100 nm alumina layer is used as gate dielectric on top of which the al - set is fabricated .
the source and drain leads are separated from the micron - sized island by aluminium oxide tunnel junctions , as it can be observed in the sem image in fig .
1(b ) . in this letter
the side gates were not used but could provide extra - functionality to the structure .
tunnel junctions .
side gates were not used in this experiment .
( c ) drain current ( @xmath9 ) oscillations as a function of the back gate voltage ( @xmath10 ) . ] in the electric mode of operation we use the back - gate voltage to control the induced - charge in the island of the mset .
1(c ) shows the drain current ( @xmath9 ) versus back - gate voltage ( @xmath10 ) at 300 mk .
the plot shows a periodic oscillatory behaviour of the current , namely the coulomb blockade oscillations .
several devices were studied with typical charging energies @xmath11=100 - 300 @xmath12ev and gate capacitances @xmath13=0.4 - 1.0 ff .
we obtain a i@xmath14/i@xmath15 contrast ratio of up to 1.33 and typical tunnel junction resistances of @xmath16=500 k@xmath17 .
the magnetic mode of operation is based on the dependence of the gamnas band structure on the orientation of the magnetization due to spin orbit coupling .
the magnetization reorientation yields a shift in the gamnas work function @xmath18 which in itself does not produce a response on the mset .
however , when the back gate is held at constant potential a change in @xmath12 causes an inward or outward flow of charge in the gate which in turn offsets the coulomb oscillations .
if the work function shift in the magnetic layer is adequate , the phase shift becomes half a period of the coulomb blockade oscillations and complementary msets are obtained . the equation that defines the necessary work function shift for maximum on - off current ratio
can be expressed as , @xmath19 as well , the minimum potential shift for complementary operation at temperature @xmath20 can be defined as a gate voltage shift equal to the fwhm of the coulomb peak @xcite , @xmath21 these two equations set the operational parameters in terms of temperature , magnetic field and physical dimensions of the structure . the complementary operation of the mset is experimentally demonstrated in fig
. 2 for a device with @xmath22=300@xmath12ev and @xmath23=1.0 ff . to switch between complimentary modes
we rotate continuously a saturating magnetic field from @xmath5=90@xmath6 , where the magnetization resides @xmath4 parallel to the plane along the easy - axis [ 110 ] ( fig.2(a ) ) to @xmath5=0@xmath6 where the magnetization is out - of - plane ( fig .
this can be seen in fig.2(c ) where we plot the source - drain current @xmath9 as a function of the magnetization angle @xmath5 and gate voltage @xmath10 for @xmath24=15@xmath12v and b=0.7 t. when the magnetization is in - plane , we select @xmath10=-61 @xmath12v as the logic 0 and @xmath10=+86 @xmath12v as the logic 1 for the n - type set ( see fig .
2(d ) ) . as the magnetic field
is rotated to @xmath5=0@xmath6 , the magnetization is re - oriented in the [ 001 ] direction and the gamnas work function increases producing a gate voltage shift of @xmath25=150 @xmath12v @xcite . the combination of work function potential shift in the ga@xmath0mn@xmath1as layer with the 300 @xmath12v gate voltage period of the device result in that the magnetically - induced gate voltage shift coincides with half a coulomb oscillation ( fulfilling eq.(1 ) ) .
the mset behaves effectively as a p - type set , fig .
therefore we can arbitrarily select the function of the set from an n - set for in - plane magnetization @xmath5=90@xmath6 to a p - type for out - of - plane magnetization @xmath5=0@xmath6 .
= 0@xmath6 the magnetization is in - plane and ( b ) @xmath5=90@xmath6 the magnetization is out - of - plane .
( c ) coulomb blockade oscillations as a function of the direction of the back - gate voltage @xmath10 and the applied magnetic field orientation @xmath5 for b=0.7 t. the - dashed blue and red lines indicate the operating points .
( d ) mset ids - vgs transfer function at @xmath5=0@xmath6 .
the logic 0 ( 1 ) has been selected at a low ( high ) current level , n - type set .
( e ) mset ids - vgs transfer function at @xmath5=90@xmath6 .
the logic outputs have been inverted , p - type set . ]
we now focus on the logic design that could be implemented at the single device level .
it has been shown that any boolean function of two variables can be implemented on a set @xcite . in this letter , by making use of the magnetic mode of operation , we demonstrate two sets of reprogrammable logic gates .
the inputs in this case are the drain voltage ( input a ) and the gate voltage ( input b ) and the output is the drain current . in fig .
3(a ) we plot the @xmath26 diagram of the mset showing the characteristic coulomb diamonds at @xmath5=0@xmath6 .
overimposed , we schematically show two logic gates framed in red and and nand .
the output is drawn as en empty dot if the results is 0 ( low current level ) and a full dot if the results is 1 ( high current level ) . upon changing the magnetization angle to @xmath5=90@xmath6
, the whole diagram is shifted horizontally by @xmath25=-150 @xmath12v and , therefore , at the same input voltages the gate output changes to the two blue - framed logic gates , from ( n)and to ( n)or as depicted by the arrow .
the reconfigurable logic gates are demonstrated in the histograms in fig .
3(b - e ) . in order to discriminate between logic outputs 0 and 1 we select the low(high ) current threshold at @xmath9=80(90 ) pa . in fig .
3(b ) we represent a histogram of the current output for an and gate implemented on the edge of the coulomb diamond . by rotating the magnetization to the in - plane direction ( @xmath5=90@xmath6 ) the coulomb diamonds shift and the logic gate switches function to or ( fig .
[ fig : fig3](c ) ) . moreover , in fig .
[ fig : fig3](d , e ) we demonstrate the logical complement set of reprogrammable boolean gates .
the set switches between nand ( @xmath5=0@xmath6 ) and nor ( @xmath5=90@xmath6 ) logic gates
. map of the drain current for @xmath5=0@xmath6 showing the characteristic coulomb diamonds .
red and blue frames sketch the implemented logic gates for @xmath5=0@xmath6 and 90@xmath6 respectively .
( b - c ) and - or set of reprogrammable logic gates .
and gate implemented at @xmath5=0@xmath6 ( b ) and or gate at @xmath5=90@xmath6 ( c ) with @xmath24 ( input a ) 0(1 ) defined as @xmath27 @xmath12v and @xmath10 ( input b ) 0(1 ) defined as @xmath28 @xmath12v .
( d - e ) nand - nor set of reprogrammable logic gates .
nand gate implemented at @xmath5=0@xmath6 ( d ) and nor gate at @xmath5=90@xmath6 ( e ) with @xmath29 ( input a ) 0(1 ) defined as 220(132 ) @xmath12v and @xmath10 ( input b ) 0(1 ) defined as 128(224 ) @xmath12v . ] finally , we briefly comment on the possibilities for multi - device reconfigurable logic design . as an example , we focus on pull - down networks formed by two elements with a common magnetic back - gate and independently addressable electrical gates .
we concentrate on the case of mset with identical @xmath9-@xmath10 transfer functions and logic inputs defined by fig .
2(c , d ) . the circuit in fig .
4(a ) consists of a resistor load and two msets in series . in the state @xmath5=0@xmath6
both mset are in the p - type state and the network realizes the operation or .
however this gate can be reconfigured by rotating the magnetization , @xmath5=90@xmath6 .
in this situation both mset are in the n - type state and the logic operation becomes nand .
similarly , for a pull - down network consisting of two msets in parallel , fig .
4(b ) the logic gate can be programmed to switch between and ( @xmath5=0@xmath6 ) and nor ( @xmath5=90@xmath6 ) operations .
= 0@xmath6 and nand at @xmath5=90@xmath6 .
( b ) parallel pull - down network performs the and operation at @xmath5=0@xmath6 and nor at @xmath5=90@xmath6 . ]
in conclusion , we have demonstrated the complementary logic operation of an aluminium mset by making use of magnetization - dependent work function of the gamnas magnetic back - gate . by using the source - drain voltage and the gate voltage as binary inputs we showed a set of boolean gates and
its logical complement implemented at the single device level , and - or and nand - nor gates .
finally , we have described a route to integrate multiple mset in to integrated reconfigurable logic circuits .
although here we present results on msets , magnetic gating could be exploited to switch function in more conventional circuit elements such as cmos field - effect transistors or novel electronic devices such as graphene ambipolar transistors @xcite or spin - hot carrier transistors @xcite .
for reliable room temperature switching , magnetic materials with work function shifts of the order of 60 mv will be required @xcite .
materials with large work function anisotropy such as copt alloys could provide that functionality @xcite .
moreover , non - volatile logic reconfiguration could be also achieved by making use of the stability of the magnetic moments along non collinear magnetic easy axesorientations without applying external magnetic fields or electrical currents .
the authors thank m. klein , d.a .
williams for fruitful discussion .
the research leading to these results has been supported by the european community s seventh framework under the grant agreement no.318397 .
http://www.tolop.eu , by the eu european research council ( erc ) advance grant no .
268066 , by the ministry of education of the czech republic grant no .
lm2011026 , and by the grant agency of the czech republic grant no .
14 - 37427 g .
takiguchi m , shimada h , mizugaki y. correlation between polarity of magnetoresistance ratio and tunnel resistance in ferromagnetic single - electron transistor with superconductive island .
j j app phys .
2014;53 : 043101 ishikuro h , hiramoto t. influence of quantum confinement effects on single electron and single hole transistors . in : electron devices meeting , 1998 .
technical digest .
, international .
1998 ; 119 - 122 .
uchida k , koga j , ohba r , toriumi a. programmable single - electron transistor logic for low - power intelligent si lsi . in : solid - state circuits conference , 2002 .
digest of technical papers .
2002 ieee international .
2002;1 : 206 - 460 .
wunderlich j , jungwirth t , irvine ac , kaestner b , shick ab , campion rp , et al .
coulomb blockade anisotropic magnetoresistance and voltage controlled magnetic switching in a ferromagnetic gamnas single electron transistor .
j magn magn mater .
2007;310 : 1883 - 1888 .
wunderlich j , jungwirth t , kaestner b , irvine ac , shick ab , stone n , et al .
coulomb blockade anisotropic magnetoresistance effect in a ( ga , mn)as single - electron transistor . phys rev lett .
2006;97 : 077201 .
klein m , lansbergen gp , mol ja , rogge s , levine rd , remacle f. reconfigurable logic devices on a single dopant atom - operation up to a full adder by using electrical spectroscopy .
chemphyschem 2009;10 : 162 - 173 .
betz ac , barraud s , wilmart q , placais b , jehl x , sanquer m , et al .
high frequency characterisation of thermionic charge transport in silicon - on - insulator nanowire transistors .
app phys lett 2014;104 : 043106 . | we propose a novel hybrid single - electron device for reprogrammable low - power logic operations , the magnetic single - electron transistor ( mset ) .
the device consists of an aluminium single - electron transistors with a gamnas magnetic back - gate . changing between different logic gate functions
is realized by reorienting the magnetic moments of the magnetic layer which induce a voltage shift on the coulomb blockade oscillations of the mset .
we show that we can arbitrarily reprogram the function of the device from an n - type set for in - plane magnetization of the gamnas layer to p - type set for out - of - plane magnetization orientation .
moreover , we demonstrate a set of reprogrammable boolean gates and its logical complement at the single device level .
finally , we propose two sets of reconfigurable binary gates using combinations of two msets in a pull - down network . |
contraction of thin liquid layers on solid surfaces due to dewetting or drying is a common phenomenon .
it is observable for instance , on plants leafs as the water breaks up into small droplets , in non - sticking pans as the oil layer shrinks or on an outdoor oil - polluted surface after rain .
another well - know example is the contraction of the liquid layer covering the eyeball , the characteristic time scale of a complete contraction being the time elapsed between two successive blinks @xcite .
dewetting plays an important role in the tire industry as well : when the contraction of the wetting layer on the tire s groove is too slow , aquaplaning is more likely to occur @xcite .
dewetting is also important in the lubricant manufacturing , however in this case exactly the opposite effect is desired : the more a lubrifient remains on the surface of sliding pieces , i. e. the larger its contraction time , the better .
along with the development of the polymer industry , contraction of polymer films started to gain interest @xcite .
dewetting turned out to be a useful investigative tool for determining various rheological and interfacial properties of thin polymer films due to the fact that molecular properties are reflected in the macroscopic shape of the solid - liquid - gas triple interface @xcite .
in other cases , liquids are used as carriers for certain substances ( nanoparticles , for example ) , thus dewetting eventually accompanied by drying on rough surfaces of such solutions , results in deposition of the dissolved substance on the substrate .
in fact , this deposition process can only be controlled through controlling the dynamics of the carrier liquid film , and , in particular , the evolution of the morphology of the triple line . in a recent study ,
dna molecules were deposited in a highly ordered array by dissolving them in a solvent and letting the solvent dewet a micropillar - structured surface @xcite .
the dynamics of wetting on flat solid and liquid surfaces is quite well understood @xcite , however , despite its applicability , only a few experiments were performed on inhomogeneous , either patterned or disordered surfaces @xcite , while the dynamics of a receeding contact line remains almost unexplored . in spite of the apparent simplicity of the phenomenon , there are no simple , easily manageable models for describing it .
although in the lubrication approximation the navier - stokes ( or , in the highly viscous regime the stokes ) equation reduces to two dimensions @xcite , the numerical modeling of layers with large planar extent is still computationally time consuming and cumbersome due to the discontinuities on the liquid - solid and liquid - gas interfaces .
these discontinuities are tackled within the framework of phase - field models @xcite , but it remains unclear however , how substrate inhomogeneities would be introduced in such models .
it is also also unsettled how the actual dynamics of the layer is influenced by the chosen particular form of the phase interface .
the continuous emergence of newer and newer schemes in the topic suggests that the demand for a convenient approach for modeling thin liquid layers dynamics is still unsatisfied @xcite .
based on the revolutionary paper of j. f. joanny and p. g. de gennes on the perturbed contact line shape @xcite , a series of depinning type models were constructed that aimed to describe interface dynamics in presence of disorder @xcite .
these models are not restricted to dewetting phenomena , as they apply to fracture front propagation or even magnetic domain wall motion . in the framework of these models , small deformations of the interface and a linear restoring force acting on the contact line resulting from a perturbative approach
are considered .
they are thus inherently linear , and the only source of nonlinearity is the disorder of the landscape they propagate in . although they have had a great success in the sampling of the depinning transition and determination of various critical exponents @xcite , they have the drawback that they neither allow for large deformations , nor for local backward movement of the line .
consequently , they are unable to account for the tearing up of the dewetting film , which , in fact , is a common phenomenon .
our purpose here is precisely to address the question of large deformations and the eventual tearing of the film with an efficient and easily manageable model for the contact line motion .
our method works best for viscous , flat and extended droplets with small wetting angle .
it is shown that in this regime , in contrast to the perturbative treatment @xcite , the line is soft and ductile , meaning that a localized perturbation of the line induces only short range forces .
considering a viscous regime , the line s equation of evolution becomes an overdamped one . in the following sections we will describe this method in detail
, we will show how to handle substrate inhomogeneities , and an application is presented .
let the upper surface of the contracting fluid layer be described by @xmath0 .
our approach is restricted to the description of large , flat layers in the highly viscous regime , the same assumption that is made when deriving the lubricant equations @xcite , i.e. @xmath1 .
one further assumption we make is that the relative change in the height of the droplet is small , therefore its height is almost constant in time , @xmath2 . under these considerations , the layer s free energy has two terms .
the first component is the joint contribution of the well - known liquid - solid and liquid - gas ( air ) surface tensions .
if the layer is flat , its upper and lower surface areas are approximately equal , @xmath3 . denoting by @xmath4 the appropriate surface tension coefficients , the surface energy writes as : @xmath5 the second contribution to the total free energy of the layer is the line energy which occurs due to the unbalanced forces acting on the layer boundaries on the molecules from the liquid - substrate - air triple interface .
this is a curve with finite thickness , thus this energy is comparable to the surface energy and it is proportional to the length of the triple interface , @xmath6 @xmath7 where @xmath8 is the line tension coefficient .
neither the interpretation of @xmath8 , nor its measurement is straightforward , in fact , there is still less consensus regarding its magnitude : values ranging from @xmath9 to @xmath10 were measured or computed in various experiments and simulations @xcite .
the major difficulty arises from the fact that dewetting is often accompanied by a precursor layer with a much smaller thickness than the rest of the layer . in our case , in term ( [ line1 ] ) a contribution resulting from the layer s side surface has to be also considered .
this yields an extra surface energy that is also proportional with @xmath6 , consequently , we believe that an effective @xmath8 has to be used instead
. therefore in calculations larger values than the presented range should be used . in the case of a real two dimensional flow ( for instance , flow in a hele - shaw cell @xcite ) , the line tension is well defined and it is clearly a result from the finite side surface of the layer between the plates . for complete wetting , i.e. zero wetting angle , @xmath11 , where @xmath12 is the distance between the plates of the hele - shaw cell @xcite .
alternatively , if a quantitative upscaling of the elastic type of energy introduced in @xcite was possible ( properly removing the third dimension from the model ) , it could provide the correct expression for the line tension for sufficiently flat droplets , bounded by one solid surface only .
such an expression however is not available , hence it remains an open question .
the total free energy of the system is the sum of these two contributions : @xmath13 .
our approach is based on the fact that both the surface and the line energies are functionals of the shape of the triple interface , which is a one - dimensional curve .
when inertial effects do not play an important role ( the highly viscous , low reynolds number regime ) , the total energy of the system is uniquely defined by the shape of the contact line , it is therefore enough to track solely its dynamics . in order to illustrate this
, we consider a simple example : the dynamics of a circular hole . due to the symmetry of the problem , an analytically study is possible . from energy terms
( [ surface1 ] ) and ( [ line1 ] ) the forces acting on the edge of the hole can be derived , which , due to symmetry considerations act in the radial direction @xmath14 where @xmath15 is the radius of the hole .
similarly , the force resulting from the line tension : @xmath16 assuming an overdamped motion of the edge of the hole ( the triple interface ) , the following equation of motion yields for its radius : @xmath17 in the above expression , @xmath18 is the mobility of the three - phase line and is inversely proportional to its length , i.e. the longer the line , the more sluggish it is : @xmath19 , where @xmath20 is the mobility of a line segment of length @xmath21 .
the equation of motion for the contact line is thus : @xmath22 it can be seen that the equilibrium radius of the hole is @xmath23 which is an intrinsic length scale of the system . for large radii ( @xmath24 )
the line energy can be neglected and the velocity of the contact line is constant : @xmath25 note that when @xmath15 is large , @xmath26 , which is in complete concordance with previous results , for instance @xcite .
so far the mobility of the triple interface has been introduced as a phenomenological parameter which , in turn , defines the time - scale of the problem . considering the case when no slippage of the interface occurs (
the flow of the interface is a poiseuille flow ) , in previous studies similar results to eq .
( [ veloc1 ] ) have been derived for the radial velocity of the triple interface for a drying patch nucleated into a liquid film @xcite : @xmath27 where @xmath28 is the equilibrium contact angle , @xmath6 is the rim width , @xmath29 is the extrapolation length ( the distance from the rim at which the velocity extrapolates to zero ) and @xmath30 is the viscosity . comparing eq .
( [ veloc1 ] ) to eq .
( [ veloc2 ] ) one can identify the mobility given now in terms of independently measurable quantities that are now properties of the contact line : @xmath31 in case of a curve - like interface with parametric equation @xmath32 ( where @xmath33 is some arbitrary parameter ) , the equation of motion writes as @xmath34\ \cdot \vec{f}[\vec{r}(\theta)],\ ] ] hence the mobility and the force in this case are both functionals of the shape of the interface .
in order to model the dynamics of contact lines of arbitrary shape , numerical methods are necessary . as a first step ,
the contact line is discretized into _
representative points_. after the contour is discretized , the points are connected through directed line segments ( vectors ) .
each of the points `` tracks '' its previous and upcoming neighbors and , by convention , the vectors are directed so that the liquid always lies on their left hand side . following the direction of the vectors connecting the points , a directed chain is established .
we denote by @xmath35 the index of the ensuing point corresponding to point @xmath36 and by @xmath37 the point preceding @xmath36 ( fig.[fig : model_sketch ] ) .
discretization of the contact line.,width=226 ] in terms of the representative points coordinates , the line and surface tension energies write as : @xmath38 once the energies are obtained , the forces acting on the representative points are computed as @xmath39 . in our two - dimensional approximation ,
the two components of this force are @xmath40 + \gamma(y_{s_i } - y_{w_i } ) \nonumber \\
f_{iy } & = & -\frac{\partial u}{\partial y_i } = \\ & = & - \alpha \left [ \frac{y_i - y_{w_i}}{d_{i , w_i } } + \frac{y_i - y_{s_i}}{d_{i , s_i } } \right ] - \gamma(x_{s_i } - x_{w_i } ) \nonumber,\end{aligned}\ ] ] where @xmath41 is the distance between points @xmath42 and @xmath6 .
it can be readily seen that each point interacts with its nearest neighbors only , therefore a molecular dynamics type simulation is suitable for investigating their dynamics .
we emphasize that the localized nature of the forces is a direct consequence of our primary hypothesis , i.e. the droplet is flat and its height profile does not change significantly during the movement of the contact line . either at lower scales , where the fine structure of the contact line becomes relevant or in the case of non - flat droplets the green function of the contact line ( its response to a localized perturbation ) is of long - range nature . as mentioned in the introduction , a perturbative treatment for small deformations of the contact line
is described in ref .
@xcite , while the propagation of such lines in random media resulting in depinning transition and a consequent advancing accompanied by avalanches are extensively studied in ref .
@xcite and @xcite . for the present case
, we stick to the lubricant approximation , thus proceed with eq .
( [ forces ] ) .
the overdamped equation of motion for the points is : @xmath43 the mobility @xmath44 associated to point @xmath36 is inversely proportional to the length element of the respective point on the triple interface : @xmath45 where we remind that @xmath46 is the distance between point @xmath36 and its upcoming neighbor , while @xmath47 is the distance between point @xmath36 and its previous neighbor . during their dynamics , the representative points will approach or move away from each other . in order to preserve numerical accuracy
, their density on the triple line should be kept constant . imposing a constant density
however , is incompatible with the movement of the individual points , therefore , an optimal fluctuation around an average value is necessary .
this issue is solved by inserting a new point between two neighboring points whenever they move farther than a predefined distance @xmath48 . in case
they come closer than another predefined distance @xmath49 , one of the points is removed . as a rule of thumb ,
we consider @xmath50 , which ensures that no insertion is necessary right after a removal . with this choice
, @xmath20 is then the mobility of one line segment .
note that continuous indexing of neighboring points is not possible due to the repeated insertions and removals .
whenever two segments intersect , the points are reconnected such that the line breaks up , hence allowing for tearing the layer .
the used reconnection mechanism is sketched on figure [ fig : tear_sketch ] .
similarly to previous descriptions , one may introduce inhomogeneities of the substrate in terms of pinning points .
whenever the contact line hits a pinning point , it is blocked as long as the force acting on it does not reach a given threshold .
. [ eqm3 ] then modifies to : @xmath51 \cdot \left(\vec{f}[\vec{r}(\theta ) ] + \vec{f}_{pin}[\vec{r}(\theta ) ] \right)\end{aligned}\ ] ] where @xmath52 is the pinning force resulting from inhomogeneities : @xmath53 here @xmath54 will characterize the pinning strength at site with position at @xmath55 . in case of point - like inhomogeneities ,
localized at spatial coordinates @xmath56 @xmath57 where @xmath58 are the thresholds of the pinning points . in the followings , spatially
uniformly distributed and uncorrelated inhomogeneities are considered . for simplicity reasons ,
the @xmath58 threshold values are considered also uniformly and uncorrelatedly distributed on the @xmath59 interval .
the concentration of the point - like inhomogeneities is @xmath60 while their average distance is given by @xmath61 .
note that the disorder is quenched , which means that in principle their positions would have to be generated and fixed right from the beginning of the simulation .
the line segments have to be tested at any instant of the simulation , whether they cross any of the pinning points , a procedure which is extremely time consuming . in order to avoid this ,
a simplified procedure is used to generate pinning points on the run , yet preserving their statistical properties .
( color online ) handling the substrate inhomogeneities . in this example , the line segment corresponding to point @xmath36 crosses @xmath62 pinning centers , each with its own threshold . the effective threshold experienced by point @xmath36 is the largest one out of those @xmath62 .
the pinning points are considered point - like , with no planar extension.,width=226 ] if the line segment belonging to point @xmath36 sweeps a small area @xmath63 within a time interval @xmath64 ( fig .
[ fig : pinning ] ) , the probability of finding exactly @xmath65 pinning points within that area has a poisson distribution : @xmath66 since the pinning is related to thresholds , whenever the line segment crosses @xmath65 pinning points , with thresholds @xmath67 , it will experience an effective threshold which is the maximum of all the thresholds of the points within @xmath63 : @xmath68 bearing in mind that @xmath58 is uniformly distributed on the @xmath59 interval , the probability distribution of the maximum is given by : @xmath69 at every time step , for each site , the number of pinning points is drawn according to the ( [ poisson ] ) distribution , while the thresholds is generated according to the ( [ maxd ] ) distribution .
as application to the previously discussed method , we will study the dynamics and topology of a moving dewetting line on a substrate with uniformly distributed quenched disorders .
disorders act as pinning centers , and we consider them point - like with the statistical properties described in the previous section .
the initial state of the interface is a straight line along the @xmath70 axis ( @xmath71 ) , and the liquid is considered to be under this line in the @xmath72 semiplane .
periodic boundary conditions are imposed along the @xmath70 axis , hence while the liquid contracts , the contact line moves towards the negative @xmath73 direction . after a transient period
, the line reaches a dynamic equilibrium state , in which its statistical properties are stationary .
@xmath74 is chosen as the unit length of the simulation .
all the lengths are then expressed in terms of dimensionless coordinates @xmath75 .
let us introduce @xmath76 , which would correspond to a flat line element subjected to a capillary force that would move it over a pinning dot with threshold @xmath77 .
its dimensionless form is @xmath78 .
the dimensionless time is @xmath79 .
the equation of motion ( [ eqm2 ] ) can then be rewritten in terms of these dimensionless quantities which leaves us with two parameters only : the length scale @xmath80 defined by the amplitude of the inhomogeneity thresholds and the length scale @xmath81 defined by their concentration .
consequently , the dynamics of the line is a result of the competition between these two length scales .
simulations were carried out for a system length along the @xmath70 direction @xmath82 , representative points distance @xmath83 and a time step @xmath84 . as mentioned previously ,
tearing up of the layer is possible , however , the resulting droplets ( droplets left behind as a result of tearing ) are disregarded as they have no more influence on the dynamics of the main line . wherever it was meaningful ,
an ensemble average over @xmath85 ensembles ( independent runs ) was considered .
we present now the results obtained for the dynamics of the model system described in the previous section ( section v. ) .
first , we study qualitatively the dynamics of the interface .
[ fig : coordinates ] shows the time evolution of the contact line for various parameters @xmath80 and @xmath86 . as the line s average velocity decreases , i. e. as it approaches the depinning transition , its length and roughness increases .
one will observe that the contact line reaches a statistically stable conformation , and its shapes are in good qualitative agreement with the experiments carried out by clotet et al.@xcite and paterson et al .
@xcite in a hele - shaw cell , although both experiments were carried out for wetting on disordered substrates , i.e. the opposite dynamics of the contact line . in the dynamic equilibrium ( stationary regime of the moving interface ) ,
the mean velocity of the interface along the @xmath73 direction presents a nontrivial , phase - transition like behavior as a function of @xmath87 .
there is a critical concentration , below which the line is depinned ( fig .
[ fig : velocity_vs_l0 ] ) and this is what we call _ depinning transition_. ( color online ) mean velocity of the interface along the @xmath73 axis , in the stationary regime , as a function of @xmath87 for different @xmath80 parameters .
the inset shows the mean velocity as a function of @xmath88 .
a reasonable collapse is obtained.,width=302 ] this critical concentration ( or , the associated length @xmath89 ) depends on the pinning strength . from fig .
[ fig : velocity_vs_l0 ] we also learn that @xmath89 increases with @xmath80 and converges to @xmath90 as @xmath91 .
this value is significantly lower than @xmath92 , which would be the critical length for a regular array of defects with infinite strength that would prevent tearing .
collective trapping of parts of the contact line thus is possible if the distance between the neighboring defects is less than @xmath93 .
the existence of such a threshold , lower than @xmath92 has been shown experimentally @xcite , however , since the experiment was carried out in gravity , its value is related to the capillary length . in our case ,
the obtained lower limit is merely a consequence of the competition between the line and surface tensions and the value @xmath94 is thus a consequence of the underlying disorder .
it is related to the percolation of the contact line between the localized defects .
as it is expected for a critical behavior , the mean velocity curves have a reasonable collapse if they are plotted as a function of @xmath88 . the inset in figure [ fig : velocity_vs_l0 ]
shows the results in such sense .
although the number of the simulated data points was rather limited for this purpose , we made an attempt to find the @xmath95 dependence .
we considered the mesh illustrated on fig .
[ fig : phase_diagram ] in the @xmath96 plane to detect the occurrence of the depinning transition .
the inset in fig .
[ fig : phase_diagram ] shows that @xmath97 is a reasonable fit for describing the boundary between the two phases in the mapped region .
interestingly , this fit suggests that for @xmath98 a total pinning is not possible .
( color online ) phase diagram of the contact line in the @xmath99 parameter space .
symbols indicate parameter values at which simulations were performed .
blue squares indicate the obtained pinning phase , green dots the depinning phase .
the inset derived from the separation points shows that the two phases are delimited by the curve @xmath97 , the dashed line indicating a slope @xmath100 .
please note the logarithmic scales for the inset graph.,width=302 ] for the high inhomogeneity and low threshold regime ( @xmath101 , @xmath102 ) , one would expect the possibility of a classical depinning transition , with small deformations of the contact line .
interestingly however , we could not observe such a transition , even for extremely low values of @xmath87 and @xmath80 . in their experiments , duprat et al .
@xcite investigated the depinning of a wetting contact line from an individual defect . they reported that depending on the pinning strength the contact line either jumped off the defect or completely wetted it , and advanced by tearing up and leaving an air hole behind . for individual or localized group of inhomogeneities
we observed the same behavior , however , it turned out to be impossible to recover a collective depinning transition without the tearing up of the film .
this is probably the result of the high ductility of the contact line .
the classical depinning transition occurs due to the competition between disorder and long range elastic restoring forces @xcite , while in our case , we lack the long range part , therefore , we encounter a new transition , which is mainly governed by large deformations and tearing up of the layer . in the experiments of paterson and fermigier @xcite , the authors distinguish between strong and weak pinning as a function of the spatial distribution of the inhomogeneities . in the strong pinning case , defects were spread randomly and uniformly over the whole surface , while in the weak pinning case , they were spread by positioning randomly only one defect in each unit cell of a larger square lattice , hence obtaining a more homogeneous pattern . for the same defect concentration , the second case results in smaller average distance @xmath87 between the defects .
the observation that in the strong pinning case ( small @xmath87 ) the contact line breaks up , and in the weak pinning case ( large @xmath87 ) it advances with a rather smooth shape , is compatible with our simulation results , even though we tuned @xmath87 by changing the defect concentration rather than changing their distribution or correlation .
another major difference compared to classical depinning models is that in our system local backward movements of the interface may appear , and , indeed , approaching the transition , positive velocities of the representative points occur , which plays an important role in the roughening mechanism .
figure [ fig : velocities ] shows how the distribution of the velocity components in the @xmath73 direction changes as we approach the transition point .
far from the transition point we experience an almost bimodal distribution ( one peak corresponding to the unpinned part , while the other one , at zero , to the pinned part ) , while close to it we obtain an almost zero - averaged symmetric distribution . clearly , it is due to the slight asymmetry that the contact line moves forward on average .
( color online ) distribution of the @xmath73 component of velocities along the contact line for @xmath103 .
note that when we approach the depinning transition ( @xmath104 ) a considerable local backward movement ( @xmath105 ) of the interface occurs . also , far from the transition ( @xmath106 ) , the pinned part of the line is quite well separated from the moving part.,width=302 ] in order to quantify the morphology of the contact line around the transition , we performed a classical rasterization analysis .
the length of the contact line @xmath107 was measured by taking into account only every @xmath108 representative point , and the scaling of @xmath107 with respect to @xmath109 was investigated .
this means that for @xmath110 , @xmath107 is computed by adding up the distance between each nearest neighboring point , for @xmath111 by summing the distance between each second neighbor points and so on , hence the length of the curve is approximated at different precisions .
figure [ fig : l_delta ] shows that as the system approaches the depinning transition , the scaling converges to a power law , @xmath112 .
this suggests a fractal - like structure and a scale - free morphology with a diverging total length as @xmath109 decreases .
this is again a direct consequence of the undergoing phase transition .
( color online ) development of the scale - free morphology as the system approaches the critical state .
the normalized length @xmath113 of the contact line as a function of @xmath109 ( see the text for the definitions ) .
results for @xmath114 and different values of @xmath87 .
the dashed line is a guide for the eye , and has a slope @xmath115 .
a natural upper cutoff arises due to the finite system size , and a lower cutoff from the discretization.,width=302 ] since @xmath109 can be used to parametrize the contact line @xmath116 , further information concerning its shape can be extracted by investigating the structure factor @xmath117 defined as the power spectrum of @xmath118 : @xmath119 where @xmath120 is the fourier transform of @xmath118 .
figure [ fig : s_delta ] shows the convergence of @xmath117 to a power law in the vicinity of the transition point : @xmath121 .
this suggests again the scale - free , fractal - like shape for the interface .
as expected , the main difference between the various curves @xmath117 arises from the low frequency , hence large wavelength values , showing that long range correlation develops close to the transition point .
( color online ) development of the scale - free morphology as the system approaches the critical state .
the structure factor @xmath117 as a function of @xmath109 ( see the text for definitions ) .
results for @xmath122 and different values of @xmath87 .
the dashed line has a slope @xmath123 and the range @xmath124 was used for the fourier transform.,width=302 ] ( color online ) average position of the contact line as a function of time , for @xmath114 and @xmath125 .
the arrow indicates increasing values of @xmath87 .
note how fluctuations increase as approaching the transition point and the dynamics becomes intermittent .
the inset shows the avalanche size distribution for @xmath114 and @xmath126 , while the solid line has a slope : @xmath123.,width=302 ] the average position of the contact line was also followed as a function of time .
results for a fixed @xmath114 value and a wide range of @xmath87 values are plotted on figure [ fig : y_avg ] .
when approaching the critical point , fluctuations increase and the sudden jumps in the average position become more and more dominating .
these jumps are the result of either the slip of the contact line over individual defects or the tearing up of the layer .
analogously to jumps in the magnetization ( barkhausen noise ) , these jumps are termed avalanches , since the average position of the line is governed by fast slips .
close to the transition , the sizes of the jumps exhibit a power - law distribution with an exponent @xmath127 ( inset of fig .
[ fig : y_avg ] ) .
our results along this line are however modest ( the scaling is on an interval less than two orders of magnitude ) , due to the lack of statistics for the large avalanche sizes .
it is important to note however that experimental data presented in ref @xcite clearly shows values around -2 , giving thus some confidence to the results of our model .
a novel and efficient , off - lattice molecular dynamics type simulation has been introduced in order to investigate the dynamics of thin and viscous liquid layers , dewetting on inhomogeneous surfaces . by using this simulation method the existence of an unusual depinning transition was revealed .
this transition is governed by large deformations of the interface and the breaking up of the layer .
the two - dimensional parameter space of the investigated system was thoroughly explored , and the obtained results were discussed in view of available experimental observations .
we learned that the contact line s dynamics is a result of an interplay between the capillary forces and the substrate disorder , however , with the appropriately introduced adimensional form , both relevant parameters are related to the inhomogeneities . in such an approach , the universal properties of the contact line can be viewed as a result of the competition between the inhomogeneities strength and their density .
the difference between the dynamics of a receeding and an advancing contact line ( dewetting vs. wetting ) , other than the contact angle hysteresis , remains an open question and could be investigated in the future by introducing pressure in our model .
the work of z. neda was supported by the romanian ideas pn - ii - id - pce-2012 - 4 - 0470 research grant .
the research of b. tyukodi was supported by the european union and the state of hungary , co - financed by the european social fund in the framework of tmop 4.2.4.a/2 - 11 - 1 - 2012 - 0001 ` national excellence program ' .
we are grateful for damien vandembroucq and etienne barthel for the careful reading and relevant comments on the manuscript . | a molecular - dynamics type simulation method , which is suitable for investigating the dewetting dynamics of thin and viscous liquid layers , is discussed .
the efficiency of the method is exemplified by studying a two - parameter depinning - like model defined on inhomogeneous solid surfaces .
the morphology and the statistical properties of the contact line is mapped in the relevant parameter space , and as a result critical behavior in the vicinity of the depinning transition is revealed .
the model allows for the tearing of the layer , which leads to a new propagation regime resulting in non - trivial collective behavior .
the large deformations observed for the interface is a result of the interplay between the substrate inhomogeneities and the capillary forces . |
suprathermal charged particles scattering back and forth across the surface of a shock wave gain energy .
the concept of stochastic energization due to randomly moving inhomogeneities was first proposed by fermi @xcite . in that original version ,
the acceleration process is easily shown to be efficient only at the second order in the parameter @xmath0 , the average speed of the irregularities in the structure of the magnetic field , in units of the speed of light . for non - relativistic motion , @xmath1 , the mechanism is not very attractive .
the generalization of this idea to the case of a shock wave was first proposed in @xcite and is nicely summarized in several recent reviews @xcite , where the efficiency of the process was found to be now at the first order in @xmath0 . since these pioneering papers the process of particle acceleration at shock waves has been investigated in many aspects and is now believed to be at work in a variety of astrophysical environments .
in fact we do observe shocks everywhere , from the solar system to the interplanetary medium , from the supernovae environments to the formation of the large scale structure of the universe .
all these are therefore sites of both heating of the medium crossing the shock surface and generation of suprathermal particles .
the two phenomena are most likely different aspects of the same process , also responsible for the formation of the collisionless shock itself
. one of the major developments in the theory of particle acceleration at astrophysical shock waves has consisted of removing the assumption of _ test particle _ , namely the assumption that the accelerated particles could not affect the dynamics of the shocked fluid .
two approaches have been proposed to treat this intrinsically non - linear problem : the _ two fluid models _ @xcite and the _ kinetic models _
@xcite , while numerous attempts to simulate numerically the process of particle acceleration have also been made @xcite .
the two fluid models treat the accelerated particles as a separate fluid , contributing a pressure and energy density which enter the usual conservation laws at the shock surface . by construction
, these models do not provide information about the spectrum of the accelerated particles , while correctly describing the detailed dynamics of the fluids involved .
the kinetic models on the other hand have a potential predictive power in terms of both dynamics and spectral shape of the accelerated particles .
all these considerations hold in principle for all shocks but in practice most of the work has been done for the case of newtonian shock waves ( however see @xcite for an extension to relativistic shocks ) .
astrophysical studies have shown that there are plenty of examples in nature of fluids moving at relativistic speeds , and generating shock waves .
the generalization of the process of particle acceleration to the relativistic case represents in our opinion the second major development of the theory ( baring , these proceedings ) . in this paper
, we will not present a review of all the current efforts in the investigation of shock acceleration .
we will rather concentrate our attention upon some recent work in the direction of accounting for the non - linear backreaction of the accelerated particles .
the original theory of particle acceleration was based on the assumption that the accelerated particles represent a _ passive _ fluid , with no dynamical backreaction on the background plasmas involved . within the context of this approximation
, several independent approaches @xcite give the spectrum of the accelerated particles in the form of a power law in momentum @xmath2 , where the slope @xmath3 is related in a unique way to the mach number @xmath4 of the upstream fluid as seen in the shock frame , through the expression @xmath5 ( here we asumed that the adiabatic index of the background gas is @xmath6 ) . this result
is easily shown by using the diffusion - convection equation in one dimension for a stationary situation ( namely @xmath7 ) : @xmath8 - u \frac{\partial f ( x , p)}{\partial x } + \ ] ] @xmath9 where @xmath10 is the diffusion coefficient , @xmath11 is the distribution function of accelerated particles in phase space and @xmath12 is the injection function , which we will assume to be a dirac delta function at the shock surface @xmath13 in the downstream fluid ( @xmath14 ) .
the function @xmath15 is normalized in such a way that the total number of accelerated particles is given by @xmath16 . as a first step ,
we integrate eq .
[ eq : trans ] around @xmath13 , from @xmath17 to @xmath14 , which we denote as points `` 1 '' and `` 2 '' respectively , so that we get @xmath18_2 - \left [ d \frac{\partial f}{\partial x}\right]_1 + \frac{1}{3 } p \frac{d f_0}{d p } ( u_2 - u_1 ) + q_0(p)= 0,\ ] ] where @xmath19 ( @xmath20 ) is the fluid speed immediately upstream ( downstream ) of the shock and @xmath21 is the particle distribution function at the shock location . by requiring that the distribution function downstream is independent of the spatial coordinate ( homogeneity )
, we obtain @xmath22_2=0 $ ] , so that the boundary condition at the shock can be rewritten as @xmath18_1 = \frac{1}{3 } p \frac{d f_0}{d p } ( u_2 - u_1 ) + q_0(p ) .
\label{eq : boundaryshock}\ ] ] we can now perform the integration of eq .
( [ eq : trans ] ) from @xmath23 to @xmath17 ( point `` 1 '' ) , in order to take into account the boundary condition at upstream infinity .
( [ eq : boundaryshock ] ) we obtain @xmath24 the solution of this equation for @xmath21 has the form of a power law @xmath25 with slope @xmath26 , where we introduced the compression factor @xmath27 at the shock . for a strong shock @xmath28 and we find the well known asymptotic spectrum @xmath29 , or @xmath30 in terms of energy ( here again we assumed that the adiabatic index of the background gas is @xmath31
. why should we expect this simple result to be affected by the assumption of test particles ?
there are three physical arguments that may serve as plausibility arguments to investigate the effects of possible backreactions : 1 ) the spectrum @xmath32 is logarithmically divergent in its energy content , so that even choosing a maximum momentum , it is possible that the energy density in the form of accelerated particles becomes comparable with the kinetic pressure , making the assumption of test particles untenable ; 2 ) if the non thermal pressure becomes appreciable , the effective adiabatic index can get closer to @xmath33 rather than @xmath34 , making the shock more compressive and the spectrum of accelerated particles even more divergent ; 3 ) more divergent spectra imply larger fluxes of escaping particles at the maximum momentum , which make the shock radiative - like , again implying a larger compression and flatter spectra .
all the three issues raised here point toward the direction of making the backreaction more severe rather than alleviating its effect , therefore a run - away reaction seems likely , which drives the shock toward a strongly non - linear cosmic ray modified configuration ( here the term _ cosmic rays _ is used in a general way to indicate the accelerated particles ) .
we can describe the expected effects on the basis of the following simple argument : if , as is usually the case , the diffusion coefficient increases with the momentum of the particles , we can expect that particles with larger momenta will diffuse farther from the shock surface in the upstream section of the gas . at large distances from the shock , only the high energy particles will be present , while lower energy particles will populate the regions closer to the shock surface .
there is some critical distance which corresponds to the typical diffusion length of the particles with the maximum momentum achievable , @xmath35 . at this distance ,
the pressure of the cosmic rays is basically zero and the fluid is unperturbed . on the other hand , moving inward , toward the shock , an increasing number of accelerated particles is present , and their pressure contributes to the local pressure budget by slowing down the fluid ( in the shock frame ) .
this effect causes the fluid speed upstream to be space - dependent , and decreasing while approaching the shock surface .
the region of slow decrease of the fluid velocity is usually called the _ precursor_. the shock , which may now be substantially weakened by the effect of the accelerated particles , is usually called _
subshock_. it is useful to introduce the two quantities @xmath36 and @xmath37 , which are respectively the compression factor at the gas subshock and the total compression factor between upstream infinity and downstream . here
@xmath38 , @xmath19 and @xmath20 are the fluid speeds at upstream infinity , upstream of the subshock and downstream respectively .
the two compression factors would be equal in the test particle approximation .
for a modified shock , @xmath39 can attain values much larger than @xmath40 and more in general , much larger than @xmath41 , which is the maximum value achievable for an ordinary strong non - relativistic shock .
the shape of the particle spectrum is still determined by some jump in the velocity field , but this quantity is now local : at low energies , the compression felt by the particles is @xmath42 , while at @xmath43 the effective compression is @xmath44 .
it follows that , since @xmath45 , the spectrum at low energies is steeper than that at higher energies : the overall spectrum at cosmic ray modified shocks is therefore expected to have a concave shape . in the following
we will describe the effects of the particle backreaction following the kinetic semi - analytical approach proposed in @xcite , and we will use the most general formalism , which includes the possible presence of seed pre - accelerated particles in the environment in which the shock propagates .
we repeat here the steps illustrated above for the linear case . integrating again eq .
[ eq : trans ] around @xmath13 , from @xmath17 to @xmath14 , we get eq .
[ eq : boundaryshock ] , after invoking the homogeneity of the particle distribution downstream .
performing now the integration of eq .
[ eq : trans ] from @xmath23 to @xmath17 we obtain @xmath46 @xmath47 here @xmath48 represents the distribution of seed pre - accelerated particles possibly present at upstream infinity .
we can now introduce the quantity @xmath49 defined as @xmath50 whose physical meaning is instrumental to understand the nonlinear backreaction of the accelerated particles .
the function @xmath49 is the average fluid velocity experienced by particles with momentum @xmath51 while diffusing upstream away from the shock surface . in other words , the effect of the average is that , instead of a constant speed @xmath19 upstream , a particle with momentum @xmath51 experiences a spatially variable speed , due to the pressure of the accelerated particles that is slowing down the fluid . since the diffusion coefficient is in general @xmath51-dependent , particles with different energies _ feel _ a different compression coefficient , higher at higher energies if , as expected , the diffusion coefficient is an increasing function of momentum . with the introduction of @xmath49 , eq .
( [ eq : stepnl ] ) becomes @xmath52 + u_0 f_\infty + q_0(p ) = 0 , \label{eq : step1}\ ] ] the solution of this equation can be written in the following implicit form : @xmath53 @xmath54}{u_{\bar p } - u_2 } \times\ ] ] @xmath55 \right\}. \label{eq : solut}\ ] ] in the case of monochromatic injection with momentum @xmath56 at the shock surface , we can write @xmath57 where @xmath58 is the gas density immediately upstream ( @xmath17 ) and @xmath59 parametrizes the fraction of the particles crossing the shock which are going to take part in the acceleration process . in terms of @xmath40 and @xmath39 , introduced above , the density immediately upstream is @xmath60 .
we can introduce the dimensionless quantity @xmath61 so that @xmath62 @xmath63 @xmath64 @xmath65 the structure of the fluid upstream of the shock and the corresponding spectrum of accelerated particles is determined if the velocity field @xmath61 is known .
the nonlinearity of the problem reflects in the fact that @xmath66 is in turn a function of @xmath21 as it is clear from the definition of @xmath49 . in order to solve the problem we need to write the equations for the thermodynamics of the system including the gas , the reaccelerated cosmic rays , the cosmic rays accelerated from the thermal pool and the shock itself . the velocity
, density and thermodynamic properties of the fluid can be determined by the mass and momentum conservation equations , with the inclusion of the pressure of the accelerated particles and of the preexisting cosmic rays .
we write these equations between a point far upstream ( @xmath23 ) , where the fluid velocity is @xmath38 and the density is @xmath67 , and the point where the fluid upstream velocity is @xmath49 ( density @xmath68 ) .
the index @xmath51 denotes quantities measured at the point where the fluid velocity is @xmath49 , namely at the point @xmath69 that can be reached only by particles with momentum @xmath70 .
the mass conservation implies : @xmath71 conservation of momentum reads : @xmath72 where @xmath73 and @xmath74 are the gas pressures at the points @xmath23 and @xmath75 respectively , and @xmath76 is the pressure in accelerated particles at the point @xmath69 ( we used the symbol @xmath77 to mean _ cosmic rays _ , in the sense of _ accelerated particles _ ) . the mass flow in the form of accelerated particles has reasonably been neglected .
our basic assumption , similar to that used in @xcite , is that the diffusion is @xmath51-dependent and more specifically that the diffusion coefficient @xmath78 is an increasing function of @xmath51 .
therefore the typical distance that a particle with momentum @xmath51 moves away from the shock is approximately @xmath79 , larger for high energy particles than for lower energy particles increases with @xmath51 faster than @xmath49 does , therefore @xmath80 is a monotonically increasing function of @xmath51 . ] . as a consequence
, at each given point @xmath69 only particles with momentum larger than @xmath51 are able to affect appreciably the fluid . strictly speaking
the validity of the assumption depends on how strongly the diffusion coefficient depends on the momentum @xmath51 .
the cosmic ray pressure at @xmath69 is the sum of two terms : one is the pressure contributed by the adiabatic compression of the cosmic rays at upstream infinity , and the second is the pressure of the particles accelerated or reaccelerated at the shock ( @xmath81 ) and able to reach the position @xmath69 . since only particles with momentum larger than @xmath51 can reach the point @xmath75 , we can write @xmath82 @xmath83 where @xmath84 is the velocity of particles with momentum @xmath51 , @xmath35 is the maximum momentum achievable in the specific situation under investigation , and @xmath85 is the adiabatic index for the accelerated particles . in eq .
[ eq : cr ] the first term represents the adiabatic compression of the pressure of the seed particles advected from upstream infinity , while the second term represents the pressure in the freshly accelerated particles at the position @xmath69 . in the following we use @xmath86 ( see @xcite for a detailed discussion of the reasons for this choice ) . the pressure of cosmic rays at upstream infinity is simply @xmath87 where @xmath88 is some minimum momentum in the spectrum of seed particles . from eq .
( [ eq : pressure ] ) we can see that there is a maximum distance , corresponding to the propagation of particles with momentum @xmath35 such that at larger distances the fluid is unaffected by the accelerated particles and @xmath89 .
we will show later that for strongly modified shocks the integral in eq .
( [ eq : cr ] ) is dominated by the region @xmath43 .
this improves even more the validity of our approximation @xmath90 .
this also suggests that different choices for the diffusion coefficient @xmath78 may affect the value of @xmath35 , but at fixed @xmath35 the spectra of the accelerated particles should not change in a significant way . assuming an adiabatic compression of the gas in the upstream region , we can write @xmath91 where we used mass conservation , eq .
( [ eq : mass ] ) .
the gas pressure far upstream is @xmath92 , where @xmath93 is the ratio of specific heats for the gas ( @xmath6 for an ideal gas ) and @xmath94 is the mach number of the fluid far upstream . we introduce now a parameter @xmath95 that quantifies the relative weight of the cosmic ray pressure at upstream infinity compared with the pressure of the gas at the same location , @xmath96 . using this parameter and the definition of the function @xmath66 , the equation for momentum conservation becomes @xmath97 + \ ] ] @xmath98 using the definition of @xmath99 and multiplying by @xmath51 , this equation becomes @xmath100 = \ ] ] @xmath101 where @xmath21 depends on @xmath66 as written in eq .
( [ eq : laeffe ] ) .
( [ eq : eqtosolve ] ) is therefore an integral - differential nonlinear equation for @xmath66 .
the solution of this equation also provides the spectrum of the accelerated particles .
the last missing piece is the connection between @xmath40 and @xmath39 , the two compression factors appearing in eq .
( [ eq : solut ] ) .
the compression factor at the gas shock around @xmath13 can be written in terms of the mach number @xmath102 of the gas immediately upstream through the well known expression @xmath103 on the other hand , if the upstream gas evolution is adiabatic , then the mach number at @xmath17 can be written in terms of the mach number of the fluid at upstream infinity @xmath94 as @xmath104 so that from the expression for @xmath40 we obtain @xmath105^{\frac{1}{\gamma_g+1}}. \label{eq : rsub_rtot}\ ] ] now that an expression between @xmath40 and @xmath39 has been found , eq .
( [ eq : eqtosolve ] ) basically is an equation for @xmath40 , with the boundary condition that @xmath106 . finding the value of @xmath40 ( and the corresponding value for @xmath39 ) such that @xmath106 also provides the whole function @xmath66 and , through eq .
( [ eq : solut ] ) , the distribution function @xmath107 for the particles resulting from acceleration and reacceleration in the nonlinear regime . when the backreaction of the accelerated particles is small , the _ test particle _ solution is recovered . in fig .
1 we show an example of the spectrum calculated for parameters which are typical of a supernova remnant ( solid line ) , as compared with the spectrum estimated according with the simple model of @xcite ( broken line ) and the result of a numerical simulation ( dashed line ) , also reported in@xcite . in this calculation
no seed particles have been assumed to be present in the shock environment .
the good agreement between the semi - analyical approach discussed here and the montecarlo simulations proves that the semi - analytical approach discussed here is quite effective in describing the behaviour of cosmic ray modified shock waves as particle accelerators .
however the situation is in general more complex than this : previous approaches to the problem of cosmic ray modified shocks had already shown the appearance of multiple solutions .
this was first discussed in @xcite in the context of two - fluid models and in @xcite by using a kinetic approach .
multiple solutions are found with the method proposed here as well . in @xcite
it was pointed out how the multiple solutions appear also in the case of reacceleration of seed particles .
an example of the phenomenon is illustrated for the case of no seed particles in fig .
2 , where we plot @xmath108 , bound to be unity for the physical solutions , as a function of the total compression factor @xmath39 . here
@xmath56 , @xmath35 and the shock mach number are all fixed .
the solutions are identified by the points of intersections of the curves ( obtained for different values of @xmath59 , as indicated ) with the horizontal line at @xmath106 .
one can see that for low values of @xmath59 ( approximately unmodified shock ) there is only one intersection at @xmath109 .
however , when @xmath59 is increased the intersections may become three .
all the three solutions are fully acceptable from the point of view of conservation laws . for large values of @xmath59
the shock is always strongly modified ( @xmath110 ) .
for these cases , the asymptotic shape of the spectrum at large momenta is well described by the power law @xmath111 ( or @xmath112 ) . the comparison between the method described above and that of @xcite has been discussed in @xcite . in fig .
3 , extracted from @xcite , we illustrate the spectra and @xmath66 for a case in which three solutions appear ( in both approaches ) .
the case corresponds to mach number @xmath113 , gas temperature at upstream infinity @xmath114 , injection momentum @xmath115 and maximum momentum @xmath116 . in the calculations of @xcite a specific form for the diffusion coefficient as a function of momentum
is required . for reference we adopted a bohm diffusion coefficient @xmath117 . in fig .
3 , each panel corresponds to one solution .
we plot in each panel the spectrum @xmath118 multiplied by @xmath119 ( the linear theory would predict @xmath120 ) .
the solid lines show the spectra as calculated with the approach of @xcite , while the dashed lines are the corresponding spectra as obtained using the calculations of @xcite with bohm diffusion .
the agreement between the two methods is excellent , despite the fact that the approach presented here does not require the detailed knowledge of the diffusion coefficient .
the question arises of whether the appearance of multiple solutions is an artifact of our ignorance of the parameter @xmath59 , which defines the efficiency of the shock in injecting particles from the thermal pool . although this is probably not the all story , as confirmed by the fact that multiple solutions are present even in the case of reacceleration of pre - accelerated particles ( in that case @xmath59 is no longer a free parameter ) @xcite , it is likely that injection plays a crucial role . in order to show this
, we adopt a simple physical recipe for the injection of particles at the shock .
_ real _ shock fronts are not one - dimensional sheets but rather complex surfaces with a typical thickness that for collisionless shocks is expected to be of the order of the larmor radius of the thermal protons downstream of the shock .
one should keep in mind that the temperature of the downstream gas is also affected by the non - linear modification induced by the accelerated particles , therefore the shock is expected to be thinner when the subshock is weaker .
our recipe is the following : we assume that the particles which are injected at the shock are those with momentum @xmath121 , where we choose @xmath122 and @xmath123 is the momentum of the thermal particles in the downstream plasma ( we assume that the gas distributions are maxwellian ) , determined as an output of the non - linear calculations from the rankine- hugoniot relations at the subshock .
this approximation is sometimes called _ thermal leakage _ @xcite . in physical terms , this makes @xmath59 an output of the calculations rather than a free parameter to be decided _ a priori_. in fig .
4 we plot @xmath108 calculated as described above , in the case in which @xmath59 is evaluated self - consistently from the prescription of thermal leakage .
the different curves are obtained for @xmath124 ( from top to bottom ) for a fixed mach number @xmath125 .
one can see that only single intersections with the horizontal line @xmath106 are present , namely the multiple solutions disappear if the shock is allowed to determine its own level of efficiency in particle acceleration .
this calculation was repeated for different values of the parameters , but the conclusion was confirmed for all cases of physical interest @xcite .
one can also see that large values of @xmath35 typically correspond to more modified shocks , and that the compression factor can reach large values , far from the test particle prediction .
we discussed some aspects of particle acceleration in astrophysical collisionless shock waves , and showed that even when the fraction of particles that participate in the acceleration process is relatively small ( one in @xmath126 of the particles crossing the shock surface ) a large fraction of the incoming energy can be channelled into few non - thermal particles .
this result , found previously by using several different approaches , is of the greatest importance for the physics of cosmic rays .
not only the accelerated particles can keep a substantial fraction of the energy available at the shock , but the spectrum of the accelerated particles may substantially differ from a power law , showing a concavity which appears to be the clearest evidence for the appearance of cosmic ray modified shocks . despite the passive role that electrons are likely to play in the shock dynamics , the spectrum of accelerated electrons is expected to be determined by the ( cosmic ray modified ) velocity profile determined by the accelerated hadrons in the shock vicinity .
a concavity in the spectrum of the radiation generated by relativistic electrons appears to be one of the possible evidences for shock acceleration in the non - linear regime . in the case of supernova remnants , there are hints that this concavity might have been observed @xcite .
one of the aspects of particle acceleration that are more poorly understood is the injection of particles from the thermal pool of particles crossing the shock .
this ignorance reflects in the difficulty of determining the fraction of particles that takes part in the acceleration process , and we argued that this might be the reason ( or one of the reasons ) why calculations of the non - linear shock structure may show the appearance of multiple solutions . on the other hand , assuming a simple recipe for the injection process ( thermal leakage ) is shown to result in the existence of only one solution . in other words , if the heating and acceleration processes are interpreted as two aspects of the same physical phenomenon , there seem to be no ambiguities in the way the shock is expected to behave . in this case , there is no doubt that strongly modified shocks are predicted .
the efficient particle acceleration at strong shocks is also expected to result in the reduced heating of the downstream plasma , as compared with the heating achieved in the absence of accelerated particles .
this effect should be visible in those cases in which it is possible to measure the temperatures of the upstream and downstream fluids separately , for instance through the x - ray emission of the thermal gases .
when the shock is strongly modified by the accelerated particles , a large fraction of gas heating is due to adiabatic compression in the shock precursor , rather than to shock heating at the gasous subshock . in @xcite
it was pointed out that if the shock propagates in a medium which is populated by seed pre - accelerated particles , the non - linear modification of the shock can be dominated by such seeds rather than by the acceleration of fresh particles from the thermal pool .
this might be the case for shocks associated with supernova remnants , which move in the interstellar medium where the cosmic rays are known to be in rough pressure balance with the gas .
the spectra of re - accelerated particles for modified shocks were calculated in @xcite and showed the usual concavity that is typical of cosmic ray modified shocks .
there is an additional aspect of particle acceleration at shock waves that has not been discussed so far , namely the generation of a turbulent magnetic field in the upstream section , due to the streaming instability induced by the accelerated particles .
the fact that the pressure in the form of accelerated particles may reach an appreciable fraction of the kinetic pressure at upstream infinity , @xmath127 , suggests that the magnetic field can also be amplified to a turbulent value which may widely exceed the background magnetic field , and approach the equipartition level . in @xcite the process of amplification has been studied numerically , and this naive expectation has been confirmed .
one should however notice that the non - linear effects in particle acceleration , discussed in this paper , and in particular the spectral modification , are not included self - consistently in the calculations of the field amplification in the shock vicinity .
all these issues are relevant for the investigations of the origin of ultra - high energy cosmic rays in many ways : 1 ) strongly modified shocks can be very efficient accelerators , so that the energy requirements for the sources we know might be substantially relaxed ; 2 ) the spectra of particles accelerated at strongly modified shocks are flatter than those expected in the linear theory .
flat spectra generate a gzk feature which is milder than that due to steep spectra , therefore it may be a less severe problem to explain possible excesses of events at the highest energies ; 3 ) magnetic field amplification in the shock vicinity has been invoked in the case of snr s as a possible way to accelerate particles up to the ankle in these sources @xcite . for other classes of sources
this may imply that it is easier to reach ultra - high energies in cases that are currently believed to have too low magnetic fields .
this last point deserved deeper investigation . | the acceleration of charged particles at astrophysical collisionless shock waves is one of the best studied processes for the energization of particles to ultrarelativistic energies , required by multifrequency observations in a variety of astrophysical situations . in this paper
we discuss some work aimed at describing one of the main progresses made in the theory of shock acceleration , namely the introduction of the non - linear backreaction of the accelerated particles onto the shocked fluid .
the implications for the investigation of the origin of ultra high energy cosmic rays will be discussed . |
topological insulators ( tis ) are novel materials in which even though the bulk system is insulating , the surface can support spin - polarized gapless states with dirac - cone - like linear energy dispersion.@xcite the topological surface states are unique in being robust against scattering from non - magnetic impurities , and display spin - momentum locking , which results in helical spin textures .
@xcite tis not only offer exciting possibilities for applications in spintronics , energy and information technologies , but also provide platforms for exploring in a solid state setting questions which have traditionally been considered to lie in the realm of high energy physics , such as the weyl semimetal phases and the higgs mechanism.@xcite two dimensional ( 2d ) topological insulators , also referred to as the quantum spin hall ( qsh ) insulators , were predicted theoretically , before being realized experimentally in hgte / cdte quantum wells.@xcite the three - dimensional ( 3d ) tis were identified later in bismuth - based thermoelectrics , bi@xmath1sb@xmath0 , bi@xmath2se@xmath5 , bi@xmath2te@xmath5 , and sb@xmath2te@xmath5,@xcite although transport properties of these binary tis are dominated by intrinsic vacancies and disorder in the bulk material . by
now a variety of 3d tis have been proposed theoretically and verified experimentally in a number of cases . @xcite in sharp contrast , to date , the only experimental realizations of the qsh state are hgte / cdte and inas / gasb / alsb quantum well systems.@xcite no standalone thin film or a thin film supported on a suitable substrate has been realized as a qsh state , although various theoretical proposals have been made suggesting that 2d tis could be achieved through the reduced dimensionality in thin films of 3d tis.@xcite the need for finding new qsh insulator materials is for these reasons obvious .
a topological phase transition ( tpt ) from a trivial to a non - trivial topological phase in 2d is an interesting unexplored issue , although in 3d a tpt has been demonstrated in tlbi(se , s)@xmath2 solid solutions.@xcite despite the theoretical prediction for the existence of nontrivial 2d tis in this family of materials@xcite , no experimental realization has been reported , which may be due to stronger bonding in the tl - compounds compared to the weaker van der waals type bonding between quintuple layers in the bi@xmath2se@xmath5 family .
interestingly , rhombohedral sb@xmath2se@xmath5 has been predicted to be a trivial insulator , implying that a tpt could be realized in ( bi@xmath1sb@xmath6)@xmath2se@xmath5 solid solutions .
however , the real sb@xmath2se@xmath5 material exhibits an orthorhombic structure , and a structural phase transition intervenes before the tpt point is reached , as the sb concentration increases .
these considerations suggest that a strategy for realizing a thin - film material exhibiting a 2d tpt is to begin with an existing 3d topological material in which the layers are weakly bonded and trivial and non - trivial topological phases can be achieved without encountering a structural instability . here
we recall that gebi@xmath2te@xmath3 ( gbt124 ) was theoretically predicted @xcite and experimentally verified@xcite as a 3d ti , but gesb@xmath2te@xmath3 ( gst124 ) is predicted @xcite to be a trivial insulator with a large bulk band - gap .
gbt124 and gst124 thus support different topological characters , but possess a similar lattice structure .
notably , insulating samples of bi@xmath2te@xmath5 and bi@xmath2se@xmath5 have proven difficult to realize experimentally , but this problem may prove more tractable in the ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 system .
in particular , the dirac point in gbt124 , for example , is well isolated from the bulk bands,@xcite while it lies very close to the valence band maximum or it is buried in the bulk bands in bi@xmath2te@xmath5/bi@xmath2se@xmath5 . along this
line , experimental studies of gbt124 show a larger surface state spin - polarization ( @xmath770% ) @xcite compared to bi@xmath2te@xmath5 and bi@xmath2se@xmath5 ( @xmath750%-60%).@xcite also , being ternary compounds , there is greater flexibility in substitutions , for example , the ge site in gbt124 can be replaced by sn or pb to tune the lattice constant and electronic structure .
gbt124 and gst124 thus are good candidate parent compounds for investigating a tpt .
also , if we consider related 2d slabs thinner than the surface state decay length@xcite , these slabs may yield a material supporting a 2d ti as well as a 2d tpt.@xcite moreover , since gbt124 is known to be @xmath8-type and topologically nontrivial , adding sb could reduce electron carriers , leading to a more insulating compound . accordingly
, this paper examines the evolution of topological characteristics of ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 by systematically varying the concentration @xmath4 of bi atoms for various layer thicknesses .
we find n - layer ( nl ) films of ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 to be 2d tis as follows : 28l and 35l films at @xmath4 = 0.6 ; 21l , 28l , and 35l films at @xmath4 = 0.8 ; and 14l , 21l and 28l films at @xmath4 = 1.0 .
the material undergoes a 3d phase transition to topological regime near @xmath4 = 0.6 .
we have also constructed a topological phase diagram for 2d ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 thin films with varying thickness and bi - concentrations .
the present article is organized as follows .
section ii gives computational details . in sec .
iii , we delineate the bulk crystal and electronic structures of gbt124 and gst124 .
the tpt in ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 with varying @xmath4 and the existence of 2d tis in thin films of ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 with various concentrations are also discussed .
finally , sec .
iv summarizes our conclusions .
our electronic structure calculations were carried out within the density functional theory ( dft)@xcite framework with the projector - augmented - wave ( paw ) basis@xcite . the generalized - gradient approximation ( gga)@xcite was used to include exchange - correlation effects , and the spin - orbit coupling ( soc ) effects were included as implemented in vasp ( vienna ab initio simulation package).@xcite the bulk calculations used a plane wave cut - off energy of 350 ev and a @xmath9-centered 8@xmath108@xmath108 k - mesh with conjugate gradient algorithm@xcite .
since sb and bi atoms possess a similar outermost electronic configuration , evolution of electronic structure could be tracked by varying either bi or sb concentration , and we have chosen to do so by varying the concentration of bi atoms for the sake of definiteness .
specifically , a bulk hexagonal supercell with 35-atomic layers ( five septuple layers ) was prepared with values of @xmath4 varying between 0.0 and 1.0 .
the corresponding bulk parameters , including the structure were optimized , until all components of hellman - feynman forces on each ion were less than 0.001 ev/. the relaxed structures of ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 examined for various bi concentrations were also found to be hexagonal .
the surface electronic - structure calculations are based on a slab geometry with a plane wave cut - off energy of 350 ev , a @xmath9-centered 8@xmath108@xmath101 k - mesh , and relaxed bulk parameters with a vacuum greater than 12 .
the gbt124 and gst124 belong to the rhombohedral crystal structure@xcite , composed of seven - layer ( 7l ) or septuple blocks , with layers in the sequence te - bi(sb)-te - ge - te - bi(sb)-te . as an example
, the arrangement of layers in a unit cell in gbt124 is shown in fig .
1(a ) . in a septuple block ,
the ge and bi atoms are sandwiched between the te - atoms , and the ge atom can be chosen as the inversion center .
the bonding within the septuple blocks is strong , being mainly of ionic - covalent type , whereas across the septuple blocks bonding is of van der waals type .
@xcite with this in mind , we take the surface termination to occur between two septuple blocks and regard one septuple block as a 2d thin film with the smallest thickness considered .
te@xmath3 ( gbt124 ) with inversion - symmetry .
a seven atomic layer block is shown .
( b ) the corresponding bulk brillouin zone ( bz ) with four time reversal invariant points @xmath9 , @xmath11 , @xmath12 and @xmath13 , and the 2d brillouin zone of the ( 111 ) surface with two time reversal invariant points @xmath14 and @xmath15 .
( c - d ) the bulk electronic structure of gebi@xmath2te@xmath3(gbt124 ) and gesb@xmath2te@xmath3(gst124 ) , respectively with bi / sb ( blue dots ) and te ( green dots ) atomic weights for different bands .
signs of @xmath16 at trim are also shown.,scaledwidth=45.0% ] the bulk band structures of gbt124 and gst124 are shown in figs .
1(c ) and 1(d ) , respectively .
gbt124 is an indirect band - gap semiconductor in which the conduction band minimum ( cbm ) and valence band maxima ( vbm ) lie along the @xmath17z direction .
the bulk valence and conduction bands at the @xmath12 point are composed of bi @xmath18 states and te @xmath18 states , respectively , with an inverted band order .
in contrast , gst124 is a direct band - gap material with cbm and vbm at @xmath12 point and a normal ordering of bands . since the structure possesses inversion symmetry in both cases , it is possible to calculate the @xmath19 invariants @xmath20;(@xmath21 ) ( where @xmath20 is a strong and @xmath22 is a weak topological invariant)@xcite from the bulk band structure .
the @xmath19 invariants are determined from the parity @xmath23 of the occupied bulk bands at the time reversal invariant momentum ( trim ) points @xmath24 = ( @xmath25)/2 , where b@xmath26 , b@xmath27 , and b@xmath28 are the reciprocal lattice vectors and n@xmath29 = 0 or 1.@xcite the @xmath19 invariants can then be calculated using @xmath30 and @xmath31 where @xmath32 here , n is the number of occupied bulk bands and @xmath33 is the parity of the 2m@xmath34 occupied energy band at the point @xmath35 .
there are eight trim points in the rhombohedral brillouin zone , but only four of these points [ @xmath9 , z , f and l ; see fig .
1(b ) ] , are inequivalent . the product of the parity eigenvalues ( @xmath36 ) of the occupied bands at the trim points
are shown in figs .
1(c)-1(d ) . in the case of gbt124 ,
interestingly , the band inversion occurs at the z - point , which leads to @xmath19 invariants being equal to 1;(111 ) , and is different from bi@xmath2se@xmath5 ti family@xcite , where the band inversion occurs at the @xmath9-point , yielding @xmath19 invariants to be 1;(000 ) . on other hand
, gst124 lacks band inversion at any of the trim points , indicating that the system is a normal insulator with all @xmath19 invariants zero . for investigating the tpt in ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 compounds
, we used a supercell geometry with a relaxed hexagonal supercell having 35 atomic layers for @xmath4 = 0.0 , 0.2 , 0.4 , 0.5 , 0.6 , 0.8 and 1.0.@xcite as examples , the bulk hexagonal supercells for @xmath4=1.0 and @xmath4 = 0.8 are shown in fig .
2 . for @xmath4 = 1.0 ,
the ge atom in the supercell remains an inversion center , whereas for @xmath4 = 0.8 this is not the case and the inversion symmetry is broken .
= 1.0 ( with inversion symmetry ) and ( b ) @xmath4 = 0.8 ( without inversion symmetry ) .
red boxes identify the septuple layer arrangement.,scaledwidth=45.0% ] the bulk band structures were computed using the fully relaxed structures .
figure 3(a ) shows the variation of bulk band gap of ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 as a function of @xmath4 .
the gap is seen to start decreasing from 0.13 ev at @xmath4 = 0.0 , with the valence band and conduction band mainly composed of te p - states and bi / sb p - states , respectively [ fig .
3(b ) ] , to a minimum value around [email protected] .
as we further increase @xmath4 , the gap opens up again and attains a value of @xmath380.07 ev at @xmath4=1.0 with valence and conduction bands swapping their orbital characters at the @xmath12-point .
this closing and reopening of the bulk gap with an inverted band order indicates that there is a tpt between @xmath4 = 0.5 to @xmath4 = 0.6.@xcite figures .
3(c ) -3(h ) show electronic structures of 35-atomic - layer slabs for various bi concentrations .
figure 3(c ) shows that the @xmath4 = 0.0 compound gst124 is a normal insulator with a large band gap without a gapless surface state inside the bulk energy gap region , which is consistent with the trivial insulator found in bulk calculations . on the other hand ,
a clear dirac cone surface state at the @xmath39 point is seen in fig .
3(h ) for the nontrivial phase gbt124 at @xmath4=1 .
the system becomes metallic at @xmath4 = 0.2 and 0.4 due to a surface conduction band crossing the fermi level . between the @xmath39 point and the @xmath40 point
, this surface band crosses the fermi level twice , an even number , consistent with the trivial phase found in bulk calculations . the metallic character decreases with increasing @xmath4 and at @xmath4 = 0.6 and 0.8 , it becomes insulating with a very small gap ( @xmath38 10 and 18 mev at @xmath4 = 0.6 and @xmath4 = 0.8 , respectively ) .
this gap is due to quantum confinement effects , i.e. , the interaction between the two dirac cones residing on the top and bottom surfaces . without such interaction , each side of the surface has gapless dirac - cone surface states centered at the @xmath39 point , which arises from the nontrivial bulk band topology .
we now turn to discuss the evolution of electronic structure with slab thickness in terms of multiples of septuple layer blocks .
reference has previously shown that when the thickness of a slab is smaller than the surface state decay length , states on the two surfaces of the slab become coupled via quantum tunneling , leading to a small thickness - dependent gap in the electronic structure.@xcite this coupling between the two surfaces for thin slabs is responsible for opening a gap at the dirac point and is the key for realizing the insulating phase . since thin films of ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 with @xmath4 = 0.0 ( gst124 ) as well as @xmath4 = 1.0 ( gbt124)[fig .
2(a ) ] are symmetric under inversion , we used parity analysis @xcite to determine their topological character .
on the other hand , thin films for @xmath4 = 0.2 , 0.4 , 0.5 , 0.6 and 0.8 are asymmetric under inversion [ fig .
2(b ) ] , and therefore parity - analysis can not be used . instead
, we varied the soc strength and monitor the band gap to assess the topological character .
we have further verified this non - trivial character via edge state computations .
sb@xmath1)@xmath2te@xmath3 at @xmath4 = 1.0 for ( a ) 7l , ( b ) 14l , ( c ) 21l , and ( d ) 28l thick films .
signs of @xmath16 at the trim points of the 2d brillouin zone are also shown .
panel ( e ) shows the variation of band gap for various film thicknesses in ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 at @xmath4 = 0.8 as a function of the size of spin - orbit coupling .
the inset zooms in on the shaded region .
the energy gap decreases to zero and reopens as soc increases from zero to 100% for 21l , 28l , and 35l slabs.,scaledwidth=45.0% ] figures 4(a)-(d ) show electronic structures of gbt124 films ( @xmath4 = 1.0 ) for various thicknesses .
it is evident that the slab with 7l displays an indirect band gap , whereas thicker slabs with 14l , 21l , and 28l have a direct band gap .
since all these slabs are insulating , we examined the possibility of these slabs being 2d tis .
note that slabs with @xmath4 = 1.0 are symmetric under inversion , so that the bloch wave functions have well defined parity at the trim points , which can be used to compute the 2d topological invariants , which are also shown in figs .
4(a)-(d ) .
the 2d @xmath19 invariant assumes the nontrivial value equal to 1 for 14l , 21l , and 28l films , but has the trivial value of zero for the 7l film .
as already pointed out , thin films with @xmath4 = 0.2 , 0.4 , 0.5 , 0.6 , and 0.8 are asymmetric under inversion . in order to check their topological character , we varied the soc strength from zero to 100% .
since the topological phase for all insulators without soc is trivial , we can monitor the gap size to determine if a tpt takes place as the strength of the soc is increased .
results for @xmath4 = 0.8 are shown in fig .
4(e ) . for 7l and 14l films the band gap decreases with increasing values of the soc strength , being 0.336 ev for 7l and 0.019 ev for 14l at 100% soc without closing at any value of soc strength . therefore , band structures with and without soc are adiabatically connected , implying that they are both topologically trivial . on the other hand , for 21l , 28l and 35l films , the gap closes at 86% , 84% , and 89% soc strength , respectively , and re - opens to a value of 0.071 ev for 21l , 0.062 ev for 28l , and 0.032 ev for the 35l film at 100% soc strength .
we thus conclude that the slabs of ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 at @xmath4 = 0.8 , with 21l , 28l and 35l thickness are topologically non - trivial .
sb@xmath1)@xmath2te@xmath3 at @xmath4 = 0.8 .
the continuum of bands is shown in violet color .
an odd number of crossings of the edge states at the fermi level establishes their nontrivial character .
( d ) band gaps of slabs with various thicknesses and bi concentrations @xmath4 .
shaded region shows slabs which are non trivial and predicted to provide new candidates for realizing 2d tis .
, scaledwidth=45.0% ] the existence of gapless edge states is the hallmark of 2d topological insulators .
thus to verify the topological character of 2d thin films of ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 , we also computed the edge state energy dispersion for inversion symmetric(@xmath4 = 1.0 ) and inversion asymmetric(@xmath4 = 0.8 ) slabs .
as an example , the unit cell for edge state computations of gbt124 ( @xmath4 = 1.0 ) is shown in fig .
5(a ) , where the two edges considered are marked with pink and green vertical lines . the edge state energy dispersion in gbt124 ( @xmath4 = 1.0 ) for 14l [ fig .
5(b ) ] shows that there are three fermi level crossings , labeled by @xmath41 .
this odd number of crossings between two time - reversal - invariant points confirms the nontrivial nature of these films .
we also computed the edge states of ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 with @xmath4 = 0.8 for 21l and the results are shown in fig .
since these slabs are asymmetric under inversion , the states associated with opposite edges are not degenerate .
the labels @xmath41 and @xmath42 in fig .
5(c ) indicate the fermi level crossings from states related to the left and right edges , respectively .
in this case also the edge states cross the fermi level three times , implying that these thin films are topologically non - trivial . in order to explore topological phases of 2d ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 films
, we carried out calculations for 7l , 14l , 21l , 28l , and 35l slabs for @xmath4 = 0.0 , 0.2 , 0.4 , 0.5 , 0.6 , 0.8 , and 1.0 .
the topological nature of the films was determined by the methods already described above .
the computed variation of the band gap at the @xmath43-point for slabs of different thicknesses and compositions is summarized in fig .
the shaded area in the figure identifies slabs with non - trivial character .
all slabs at x = 0.0 ( gst124 ) are topologically trivial . excepting the 7l film , thin films for @xmath4 = 0.2 , 0.4 , and 0.5 are metallic , where the metallic character decreases with increasing bi concentration .
an insulating phase is found for slabs with larger values of @xmath4 .
in particular , we predict that the 28l and 35l films for @xmath4 = 0.6 , 21l , 28l , and 35l films for @xmath4 = 0.8 , and 14l , 21l and 28l films with @xmath4 = 1.0 are topologically non - trivial .
thus , these slabs are possible candidates for realizing 2d tis , and their solid solutions could realize a 2d tpt .
we have investigated electronic structures of thin films of ge - based system ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 over the full range of bi concentrations @xmath4 within the framework of the density functional theory . by using parity analysis and surface state computations ,
we show that the @xmath4=1.0 bulk compound , gbt124 , is a topological insulator with @xmath19 = 1;(111 ) harboring a single metallic dirac cone surface state at the @xmath43 point lying within the bulk energy - gap . on the other extreme , at @xmath4 = 0 , the material ( gst124 ) transforms into a standard insulator without metallic surface states .
computations as a function of @xmath4 indicate that a topological phase transition ( tpt ) takes place for @xmath4 values between 0.5 and 0.6 .
our analysis further suggests that a 2d - tpt could be realized in thin films of ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 by varying bi concentration .
we predict that 28l and 35l films at @xmath4 = 0.6 , 21l , 28l , and 35l films at @xmath4 = 0.8 , and 14l , 21l and 28l films at @xmath4 = 1.0 , would be 2d tis .
our study opens up the possibility of identifying the qsh state in thin films of a large materials family , along with that of realizing a 2d topological phase transition .
this work was supported by the department of science and technology , new delhi ( india ) through project sr / s2/cmp-0098/2010 .
the work at northeastern university was supported by the us department of energy , office of science , basic energy sciences contract no .
de - fg02 - 07er46352 , and benefited from northeastern university s advanced scientific computation center ( ascc ) , theory support at the advanced light source , berkeley , and the allocation of time at the nersc supercomputing center through doe grant no .
de - ac02 - 05ch11231 .
acknowledges the singapore national research foundation ( nrf ) for support under nrf award no .
nrf - nrff2013 - 03 .
m. neupane , s .- y .
xu , l. a. wray , a. petersen , r. shankar , n. alidoust , c. liu , a. fedorov , h. ji , j. m. allred , y. s. hor , t .-
chang , h .-
jeng , h. lin , a. bansil , r. j. cava , and m. z. hasan , phys . rev .
b * 85 * , 235406 ( 2012 ) . s. v. eremeev , g. landolt , t. v. menshchikova , b. slomski , y. m. koroteev , z. s. aliev , m. b. babanly , j. henk , a. ernst , l. patthey , a. eich , a. a. khajetoorians , j. hagemeister , o. pietzsch , j. wiebe , r. wiesendanger , p. m. echenique , s. s. tsirkin , i. r. amiraslanov , j. h. dil , and e. v. chulkov , nat commun * 3 * , 635 ( 2012 ) .
k. okamoto , k. kuroda , h. miyahara , k. miyamoto , t. okuda , z. s. aliev , m. b. babanly , i. r. amiraslanov , k. shimada , h. namatame , m. taniguchi , d. a. samorokov , t. v. menshchikova , e. v. chulkov , and a. kimura , phys .
b * 86 * , 195304 ( 2012 ) .
note that we have modeled non - stoichiometric compositions by using appropriate supercells of ordered phases .
it will be interesting to consider disorder effects using first - principles approaches within the dft framework ( ref . ) , although virtual crystal approximation ( ref . )
might provide a reasonable scheme for describing effects of bi / sb substitution .
since the dft correctly predicts that ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 is a trivial insulator at x=0 and a non - trivial insulator at x=1 , we can have confidence in the prediction that a transition will take place for some value of x over the range 0 - 1 , uncertainty in estimating the precise transition point due to inherent limitations of the dft in underestimating gap sizes notwithstanding .
figure 3(b ) indicates that the error in estimating the transition point will depend on details of how the corrections to the dft affect the valence and conduction levels .
for example , if the positions of the conduction and valence levels should undergo similar corrections ( compared to the dft ) as a function of x , we may see relatively little change in the transition point in fig . | we discuss possible topological phase transitions in ge - based thin films of ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 as a function of layer thickness and bi concentration @xmath4 using the first principles density functional theory framework .
the bulk material is a topological insulator at @xmath4 = 1.0 with a single dirac cone surface state at the surface brillouin zone center , whereas it is a trivial insulator at @xmath4 = 0 . through a systematic examination of the band topologies we predict that thin films of ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 with @xmath4 = 0.6 , 0.8 and 1.0 are candidates for two - dimensional ( 2d ) topological insulators , which would undergo a 2d topological phase transition as a function of @xmath4 .
a topological phase diagram for ge(bi@xmath0sb@xmath1)@xmath2te@xmath3 thin films is presented to help guide their experimental exploration . |
numerical methods for approximating variational problems or partial differential equations ( pdes ) with solutions defined on surfaces or manifolds are of growing interests over the last decades .
finite element methods , as one of the main streams in numerical simulations , are well established for those problems .
a starting point can be traced back to @xcite , which is the first to investigate a finite element method for solving elliptic pdes on surfaces . since then , there have been a lot of extensions both in analysis and in algorithms , see for instance @xcite and the references therein . in the literature , most of the works consider the _ a priori _ error analysis of various surface finite element methods , and only a few works , up to our best knowledge , take into account the _ a posteriori _ error analysis and superconvergence of finite element methods in a surface setting , see @xcite .
recently , there is an approach proposed in @xcite which merges the two types of analysis to develop a higher order finite element method on an approximated surface , where a gradient recovery scheme plays a key role .
gradient recovery techniques , which are important in _ post processing _ solutions or data for improving the accuracy of numerical simulations , have been widely studied and applied in many aspects of numerical analysis . in particular for planar problems ,
the study of gradient recovery methods has reached already a mature stage , and there is a massive of works in the literature , to name but only a few @xcite .
we point out some significant methods among them , like the classical zienkiewicz zhu ( @xmath0 ) patch recovery method @xcite , and a later method called polynomial preserving recovery ( ppr ) @xcite .
the two approaches work with different philosophies in methodology .
the former method first locates positions of certain points in the given mesh , and then recovers the gradients themselves at those points to achieve a higher order approximation accuracy , while the latter one first recovers the function values by polynomial interpolations in a local patch at each nodal points , and then takes gradients at the nodal points from the previously recovered functions .
both the methods can produce comparable superconvergence results , but do not require the same assumptions on the discretized meshes .
gradient recovery methods for data defined on curved spaces have only recently been investigated . in @xcite ,
several gradient recovery methods have been adapted to a general surface setting for linear finite element solutions which are defined on polyhedrons by triangulation .
there a surface is concerned to be a zero level set of a smooth function defined in a higher dimensional space , which is from the point of view of an ambient space of the surface .
it has been shown that most of the properties of the gradient recovery schemes for planar problems are maintained in their counterparts for surface problems .
in particular , in their implementation and analysis , the methods ask for exact knowledge of the surface , e.g. the nodal points are located on the exact surface , and the tangent spaces or in another word the normal vector field are given . however , this information is usually not available in reality , where we have only the approximations of surfaces , for instance , polyhedrons , splines or polynomial surfaces . on the other hand , the generalized @xmath0 scheme for gradient recovery with surface elements gives the most competitive results in @xcite , including several other methods ,
their superconvergence are proved with the assumption that the local patch is @xmath1symmetric on the discretized surfaces , which is restrictive in applications . in the planar case
, the @xmath1symmetric condition is also asked for the superconvergence by these methods which have been generalized to a surface setting in @xcite , but it is not necessary for the ppr method .
this difference gives us the motivation to generalize the ppr method for problems with data defined on manifolds .
a follow - up question would be what are the polynomials in the domains of curved manifolds .
using the idea from the literature , e.g. @xcite , one could consider polynomials locally on the tangent spaces of the manifolds .
obviously , a direct generalization of ppr to a manifold setting based on tangent spaces will again fall into the awkward situation : the exact manifold and its tangent spaces are unknown . to overcome these difficulties , we go back to the original definition of a manifold .
we take the manifold as patches locally parametrized by euclidean planar domains , but not necessarily by their tangent spaces .
this has no interruption for us to define patch - wise polynomials in such planar parameter domains . in this manner
, we are able to recover the unknown surfaces from the given sampling points in these local domains , as well as the finite element solutions iso - parametrically .
our proposed method is thus called parametric polynomial preserving recovery ( pppr ) which _ does not _
rely on the @xmath1 symmetric condition for the superconvergence , just like its genetic father ppr . to this end
, it will be revealed that pppr is particularly useful to _ address the issue of unavailable tangent spaces _ , and thus it enables us to solve the open issues in @xcite .
another benefit of the pppr method for data on a surface is that it is relatively _ curvature stable _ in comparing with the methods proposed in @xcite .
this is verified by our numerical examples , but a quantitative analysis will be open in the paper . moreover , the original ppr method @xcite does not preserve the function values at the nodal points in its pre - recovery step . in this paper , we provide an alternative method which can achieve this goal . with this option , the pppr can not only preserve _ parametric polynomial _ , but also preserve the _ surface sampling points _ and the _ function values _ at the given points simultaneously .
that means the given data is invariant during the recovery by using the pppr method .
the rest of the paper is organized as follows : section [ sec : background ] gives a preliminary account on relevant differential geometry concepts and an analytic pde problem .
section [ sec : spaces ] introduces discretized function spaces and collects some geometric notations used in the paper .
section [ sec : pppr ] presents the new algorithms especially the pppr for gradient recovery on manifolds .
there we make remarks on the comparison of algorithms and the idea of preserving function values , and provide an argument for its curvature stable property .
section [ sec : analysis ] gives a brief analysis of the superconvergence properties of the proposed method .
section [ sec : estimator ] analyze the recovery - based _ a posteriori _ estimator using the new gradient recovery operators . finally , we present some numerical results and the comparisons with existing methods in section [ sec : numerics ] .
we have a proof of a basic lemma in appendix [ appendix ] .
we will only show some basic concepts which are relevant to our paper . for a more general overview on the topic of riemannian geometry or differential geometry
, one could refer to for instance @xcite . in the context of the paper
, we shall consider @xmath2 as an oriented , connected , @xmath3 smooth and compact riemannian manifold without boundary , where @xmath4 denotes the riemann metric tensor .
the idea we are going to work should be no restriction for general @xmath5 dimensional manifolds , but we will focus on the case of two dimensional ones , which are also called surfaces , in the later applications and numerical examples .
our concerns are some quantities @xmath6 which are scalar functions defined on manifolds .
first , let us mention the differentiation of a function @xmath7 in a manifold setting , which is called covariant derivatives in general .
it is defined as the directional derivatives of the function @xmath7 along an arbitrarily selected path @xmath8 on the manifold @xmath9 here @xmath10 is a tangential vector field .
the gradient then is an operator such that @xmath11 it is not harm to think of the gradient as a tangent vector field on the manifold @xmath12 . in a local coordinate
, the gradient has the form @xmath13 where @xmath14 is the entries of the inverse of the metric tensor @xmath4 , and @xmath15 denotes the tangential basis .
let @xmath16 be a local geometric mapping , then we can rewrite into a matrix form with this local parametrization , that is @xmath17 in , @xmath18 is the pull back of function @xmath7 to the local planar parameter domain @xmath19 , @xmath20 denotes the gradient on the planar domain @xmath19 , @xmath21 is the jacobian matrix of @xmath22 , and @xmath23 on this patch .
[ rem : surface_gradient ] @xmath22 is not specified here , and we will make it clear when it becomes necessary later .
we actually have a relation that @xmath24 where @xmath25 denotes the moore - penrose inverse of @xmath21 .
see ( * ? ? ?
* appendix ) for a detailed explanation .
note that the parametrization map @xmath22 is not unique , typical ones can be constructed through function graphs which will be used in our later algorithms .
we have the following lemma of which the proof is given in appendix [ appendix ] .
[ lem : invariant ] the gradient is invariant under different chosen of regular isomorphic parametrization functions @xmath22 . let @xmath26 be the volume form on @xmath12 , and @xmath27 be the tangential bases . for every tangent vector field @xmath28 , @xmath29
, we have a @xmath30 form defined by the interior product of @xmath31 and the volume form @xmath32 through the following way @xmath33 where @xmath34 are @xmath30 indexes with @xmath35 taking out from @xmath36 .
the divergence of the vector field @xmath31 then satisfies @xmath37 where @xmath38 denotes the exterior derivative . since both the left hand side and the right hand side of are @xmath5 forms
, @xmath39 is a scalar field . using the local coordinates ,
we can write the volume form explicitly @xmath40 applying equation , the divergence of the vector field @xmath41 can be computed by @xmath42 it is revealed that the divergence operator is actually the dual of the gradient operator . with the above preparation
, we can now given the definition of the laplace - beltrami operator , which is denoted by @xmath43 in our paper , as the divergence of the gradient , that is @xmath44 we mention that if the manifold @xmath12 is a hyper - surface , that is @xmath45 which has co - dimension @xmath46 .
the gradient and divergence of the function @xmath7 can be equally calculated through projecting the gradient and divergence of an extended function in ambient space @xmath47 to the tangent spaces of @xmath12 respectively .
this type of definitions has been applied in many references which consider problems in an ambient space setting , i.e. @xmath48 where @xmath49 and @xmath50 are the extended scalar and vector fields defined in the ambient space of the hypersurface , which satisfies @xmath51 and @xmath52 for all @xmath53 .
note that @xmath54 is the gradient operator defined in the ambient euclidean space @xmath47 , @xmath55 is the tangential projection operator @xmath56 and @xmath57 is a unit normal vector field of @xmath12 . it can be showed that the gradient and divergence by projections are independent of the way of the extension of the scalar or vector fields , and they are equivalent to the former definitions . with the generalized notions of the differentiation on manifolds , the function spaces based on manifold domains can be studied analogously to euclidean domains .
sobolev spaces on manifolds @xcite are one of the mostly investigated spaces , which provide a breeding ground to study pdes .
we are interested in numerically approximating pdes of which the solutions are defined on @xmath12 . even though our methods are _ problem independent _ , in this paper , our analysis will be mainly based on the laplace - beltrami operator , and its generated pdes . for the purpose of both analysis and applications
, we consider an exemplary problem @xcite the _ laplace - beltrami equation _ , that is for a given @xmath58 satisfying @xmath59 to solve the equation @xmath60 where @xmath61 denotes the manifold volume measure .
the discretization of a smooth manifold @xmath12 has been widely studied in many settings , especially in terms of surfaces @xcite .
a discretized surface , in most cases , is a piecewise polynomial surface .
one of the most simple case is the polygonal approximation to a given smooth surface , especially with triangulations .
finite element methods for triangulated meshes on surfaces have firstly been studied in @xcite by using linear elements . in @xcite , a generalization of @xcite to high order finite element method
is proposed based on triangulated surfaces . in order to have an optimal convergence rates , it is showed that the geometric approximation error and the function approximation error has to be compatible with each other . in fact
, the balance of geometric approximation errors and function approximation errors is also the key point in the development of our recovery algorithm . in this paper
, we will denote @xmath62 the triangulated surface , where @xmath63 is the set of triangles , and @xmath64 is the maximum diameter .
we restrict ourselves to the first order finite element methods , thus the nodes consist of simply the vertices of @xmath65 , which we denote by @xmath66 . in the following , we define transform operators between the function spaces on @xmath12 and function spaces on @xmath67 , where @xmath67 denotes some perturbation of @xmath12 . @xmath68 and its inverse @xmath69 where @xmath70 is a continuous and bijective projection map from @xmath67 to @xmath12 .
we have the following lemma with triangulated approximation @xmath71 .
[ lem : transform ] for @xmath72 @xmath73 , the transform operators @xmath74 are uniformly bounded between the spaces @xmath75 and @xmath76 as long as the space @xmath76 is compatible with the smoothness of @xmath65 . for every @xmath77 , denote @xmath78 .
each triangular faces @xmath79 of @xmath65 corresponding to a curved triangle faces on @xmath12 , and we denote it as @xmath80 .
if @xmath81 , every function @xmath82 and its derivatives are uniformly bounded on @xmath12 , then we can always find constants @xmath83 and @xmath84 such that @xmath85 for @xmath86 , using the results in @xcite , we have the equivalence of @xmath87 and @xmath88 .
that is there exists positive and uniformly bounded constants @xmath89 and @xmath90 , such that @xmath91 holds on each pair of the triangular faces .
since @xmath92 we have then the estimate @xmath93 which gives the conclusion .
[ rem : transform ] the statement of lemma [ lem : transform ] hold also for higher order continuous piece - wise polynomial approximation of @xmath12 .
we give here an assumption on the triangulations of surfaces , which is a common condition to have the so - called supercloseness .
[ ass : irregular ] @xmath65 is a quasi - uniform and shape regular triangulation of @xmath12 , and it satisfies the @xmath94 irregular condition ( cf .
* definition 2.4 ) , or ( * ? ? ?
* definition 3.2 ) ) .
+ for convenience , table [ tab : geometry ] collects some notations in the paper .
.notations [ cols="<,<",options="header " , ] in the example , we consider a benchmark problem for adaptive finite element method for laplace - beltrami equation on the sphere @xcite .
we choose the right hand side function @xmath58 such that he exact solution in spherical coordinate is given by @xmath95 in case of @xmath96 , it easy to see that the solution @xmath7 has two singularity points at north and south poles and the solution @xmath7 is barely in @xmath97 .
in fact , @xmath98 . [ fig : sphere_init ] [ fig : sphere_adaptive ] [ fig : sphere_err ] [ fig : sphere_idx ] to obtain optimal convergence rate , adaptive finite element method ( afem ) is used .
different from existing methods in the literature , recovery - based _ a posteriori _ error estimator is adopted .
we start with the initial mesh given as in fig [ fig : sphere_init ] . the mesh is adaptively refined using the dorfler @xcite marking strategy with parameter equal to @xmath99 .
fig [ fig : sphere_adaptive ] plots the mesh after the 18 adaptive refinement steps .
clearly , the mesh successfully resolves the singularities .
the numerical errors are displayed in fig [ fig : sphere_err ] . as expected , optimal convergence rate for @xmath100 error can be observed .
in addition , we observe that the recovery is superconvergent to the exact gradient at a rate of @xmath101 . to test the performance of our new
recovery - based _ a posterior _ error estimator for laplace - beltrami problem , the effectivity index @xmath102 is used to measure the quality of an error estimator @xcite , which is defined by the ratio between the estimated error and the true error @xmath103 the effectivity index is plotted in fig [ fig : sphere_idx ] .
we see that @xmath102 converges asymptotically to @xmath46 which indicates the posteriori error estimator or is asymptotically exact . in this example , we consider the following laplace - beltrami type equation on dziuk surface as in @xcite : @xmath104 where @xmath105 .
@xmath58 is chosen to fit the exact solution @xmath106 note that the solution has an exponential peak . to track this phenomena ,
we adopt afem with an initial mesh graphed in fig [ fig : dziuk_init ] .
fig [ fig : dziuk_adaptive ] shows the adaptive refined mesh .
we would like to point out that the mesh is refined not only around the exponential peak but also at the high curvature areas .
fig [ fig : dziuk_err ] displays the numerical errors .
it demonstrates the optimal convergence rate in @xmath100 norm and a superconvergence rate for the recovered gradient .
the effective index is showed in fig [ fig : dziuk_idx ] , which converges to 1 quickly after the first few iterations .
again , it indicates the error estimator ( or ) is asymptotically exact .
[ fig : dziuk_init ] [ fig : dziuk_adaptive ] [ fig : dziuk_err ] [ fig : dziuk_idx ]
in this paper , we have proposed a curvature stable gradient recovery method ( pppr ) for data defined on manifolds . in comparing with existing methods for data on surfaces in the literature , cf .
@xcite , the proposed method has several improvements : the first highlight is that it does not require the exact surfaces , which makes it a realistic and robust method for practical problems ; second , it does not need the element patch to be @xmath101 symmetric to achieve superconvergence .
third , it is the most curvature stable methods in comparing with the existing methods . aside from that
, we have evolved the traditional ppr method ( for planar problems ) to function value preserving at the mean time . by testing with some benchmark examples ,
it is quite evident that the proposed method numerically performs better than the methods in the state of the art .
we have also shown the capability of the recovery operator for constructing _ a posterior _ error estimator . even though we only develop the methods for linear finite element methods on triangulated meshes , the idea should be applicable to higher order fem on more accurate approximations of surfaces , e.g. piece - wise polynomial surfaces , b - splines or nurbus .
we leave this as a potential work for future .
gradient recovery has other applications , like enhancing eigenvalues , pre - processing data in image science , simplifying higher order discretization of pdes , or even designing new numerical methods for higher order pdes , and so on .
it would also be interesting to further investigate the full usage of the pppr method for problems with solutions defined on a manifold domain .
the authors thank dr .
pravin madhavan , dr .
bjorn stinner and dr .
andreas dedner for their kind help and discussions on a numerical example .
gd acknowledges support from the austrian science fund ( fwf ) : geometry and simulation , project s11704 .
hg acknowledges support from the center for scientific computing from the cnsi , mrl : an nsf mrsec ( dmr-1121053 ) and nsf cns-0960316 for providing computing resources .
in general , there are infinitely many isomorphic parametrizations for a given patch @xmath107 .
let us pick arbitrarily two of them , which are denoted by @xmath108 respectively , where @xmath19 and @xmath109 are planar parameter domains , then there exist @xmath110 to be a bijective , differentiable mapping , such that @xmath111 .
that means for an arbitrary but fixed position @xmath112 , we have @xmath113 and @xmath114 , such that @xmath115 then we have @xmath116 and consequently , for every function @xmath117 @xmath118 we have @xmath119 using chain rule on both sides of the former equation of , then we get @xmath120 which gives the latter equation in since @xmath121 is non - degenerate . using the same process but consider @xmath122 , we can show the reverse implication .
thus , we have shown that any two arbitrary parameterizations @xmath22 and @xmath123 lead to the same gradient values at same positions . , _ best constants in the sobolev imbedding theorem : the yamabe problem _ , in seminar on differential geometry , vol . 102 of ann . of math .
stud . , princeton univ . press , princeton , n.j .
, 1982 , pp . 173184 . , _ finite elements for the beltrami operator on arbitrary surfaces _ , in partial differential equations and calculus of variations , vol .
1357 of lecture notes in math . ,
springer , berlin , 1988 , pp .
142155 .
, _ nonlinear analysis on manifolds : sobolev spaces and inequalities _ , vol . 5 of courant lecture notes in mathematics , new york university , courant institute of mathematical sciences , new york ; american mathematical society , providence , ri , 1999 .
height 2pt depth -1.6pt width 23pt , _ the superconvergent patch recovery and a posteriori error estimates .
error estimates and adaptivity _ , internat .
methods engrg . , 33 ( 1992 ) ,
. 13651382 . | this paper investigates gradient recovery schemes for data defined on discretized manifolds . the proposed method , parametric polynomial preserving recovery ( pppr ) , does not ask for the tangent spaces of the exact manifolds which have been assumed for some significant gradient recovery methods in the literature .
another advantage of the proposed method is that it removes the symmetric requirement from the existing methods for the superconvergence .
these properties make it a prime method when meshes are arbitrarily structured or generated from high curvature surfaces . as an application ,
we show that the recovery operator is capable of constructing an asymptotically exact posteriori error estimator .
several numerical examples on 2dimensional surfaces are presented to support the theoretical results and make comparisons with methods in the state of the art , which show evidence that the pppr method outperforms the existing methods .
.3 cm * ams subject classifications . * primary 65n50 , 65n30 ; secondary 65n15 , 53c99 .3 cm * key words .
* gradient recovery , manifolds , superconvergence , parametric polynomial preserving , function value preserving , curvature stable . |
at energies below the electroweak scale the weak interactions are described by local four - fermi operators multiplied by effective coupling constants , the wilson coefficients . the formal framework to achieve this is the operator product expansion ( ope ) which allows one to separate the calculation of a physical amplitude into two distinct parts : the short distance ( perturbative ) calculation of the wilson coefficients and the long distance ( generally non - perturbative ) calculation of the hadronic matrix elements of the operators @xmath11 .
we calculate on the lattice @xmath0 and @xmath1 .
this allows us to calculate the low energy constants in chiral perturbation @xcite which , after incorporating the non - perturbative renormalization factors are then translated into @xmath12 matrix elements .
the cp - pacs collaboration has also presented a very similar calculation at this meeting @xcite .
we have used the wilson gauge action , quenched , at @xmath7 on a @xmath13 lattice which corresponds to an inverse lattice spacing @xmath14 . the domain wall fermion height @xmath8 and fifth dimension @xmath15 give a residual symmetry breaking @xmath16 @xcite ; 400 configurations separated by 10000 heat - bath sweeps were used in this analysis .
@xmath0 matrix elements were calculated in the @xmath17 limit for 5 light quark masses @xmath18 .
since the @xmath1 matrix elements vanish in the @xmath19 limit these matrix elements were calculated with non - degenerate quark propagators for 10 mass combinations subject to the constraint @xmath20 .
we have also calculated the so called eye diagrams with an active charm quark for @xmath21 ( the physical charm quark is around 0.5 ) . however ,
the analysis for charm - in is still in progress ; in this presentation we concentrate on the case with 3-active flavors wherein charm is integrated out assuming it is very heavy .
the calculation took about 4 months on 800 gflops ( peak ) .
quark propagators were calculated using the conjugate gradient method with a stopping residual of @xmath22 with periodic and anti - periodic boundary conditions which amounts to doubling the lattice size in time direction .
the two wall source propagators at @xmath23 and @xmath24 were fixed to coulomb gauge . for eye diagrams we employed random wall sources
spread over time slices @xmath25 with 2 hits per configuration .
dividing the three - point correlation functions by the wall - wall pseudoscalar - pseudoscalar correlation function yields the desired matrix elements up to a factor of @xmath26 which is determined from a covariant fit to the wall - point two - point function in the range @xmath27 for each mass .
since our results unambiguously show that re@xmath4 and re@xmath5 come essentially from the current - current operators ( recall these have the largest wilson coefficients ) we will concentrate on these operators from now on .
quenched chiral perturbation theory predicts @xmath28 % \end{equation } % \end{small}\ ] ] we find a quenched chiral logarithm coefficient @xmath29 which has a negligible contribution in our matrix element calculation . unlike the quenched chiral logarithms , the conventional logarithms coming from quenched chiral perturbation theory induce large corrections to the @xmath3 @xmath0 matrix element as can be seen in figure [ fig : o2_ktopi_3_2 ] .
we fit these amplitudes to @xcite @xmath30 + b_2^{(27,1 ) } m_m^4 % \end{equation}\ ] ] where @xmath31 , @xmath32 , @xmath33 .
the conventional chiral logarithm @xmath34 is almost linear over the mass range we have used so the fitting routine can not distinguish this term from the linear term if we leave the coefficient of the logarithm as a free parameter . since the large coefficient -6 of the logarithm makes the contribution of this term comparable to the contribution of the linear term omitting this term would change @xmath35 by almost a factor of two .
the quenched chiral log contribution is very small .
= -0.3 in -0.2 in @xmath2 @xmath0 matrix elements mix with @xmath36 with a power divergent coefficient @xmath37 .
we define a subtracted matrix element @xmath38 by @xmath39 where @xmath40 is obtained from a linear fit to @xmath41 . for an explanation of this subtraction of the power divergence
we refer the reader to @xcite .
the quenched chiral perturbation theory corrections to @xmath42 our data is consistent with a linear fit @xmath43 with the slope @xmath44 determining the low energy constants @xmath45 and the intercept @xmath46 arising from residual chiral symmetry breaking .
-0.1 in = -0.3 in -0.2 in
we use chiral perturbation theory to compute the lattice @xmath49 matrix elements . using non - perturbative z factors
we obtain the continuum matrix elements which are then multiplied by wilson coefficients to yield the physical amplitudes .
we present an extrapolation to the kaon mass scale to lowest order in chiral perturbation theory and a second extrapolation which includes one loop logarithmic effects .
we multiply the pseudoscalar masses by @xmath50 so that for @xmath51 the chiral perturbation theory extrapolation is increasingly accurate but we need the extrapolation at @xmath52 , the physical point . in figure
[ fig : re_a0_re_a2 ] we present re@xmath47 and re@xmath48 as a function of the parameter @xmath50 .
the chiral logarithm correction for re@xmath47 is large ( about @xmath53 ) .
in addition one expects a large correction ( not included here ) coming from the tree level @xmath54 terms necessary to cancel the dependence on the chiral perturbation theory scale @xmath55 .
-0.1 in = -0.3 in -0.2 in if the z factors and the wilson coefficients were calculated to all orders in perturbation theory the physical amplitudes that we calculate would not depend on the scale @xmath56 where the transition between the lattice and the continuum operators is made .
to a good approximation this is what we find , even though at @xmath57 one expects non - perturbative effects in the z factors and at @xmath58 the discretization errors may be large . in figure
[ fig : delta_i_1_2_mu_dep ] we present the ratio re@xmath47/re@xmath48 , the so called @xmath59 rule which shows a large enhancement in the @xmath60 channel in accord with experiment ( note , the chiral logarithm corrections largely cancel in the ratio so a large enhancement is seen for both extrapolation choices ) .
the residual scale dependence in the physical amplitudes is slight ( see figure [ fig : delta_i_1_2_mu_dep ] ) .
= -0.3 in -0.2 in
in conclusion , re@xmath47 , re@xmath48 and especially the ratio re@xmath47/re@xmath48 were found reasonably close to the experimental values .
we see this as an important success of the lattice method .
however there were a number of major approximations in our calculation , the hardest to quantify is the use of quenched qcd .
also the chiral logarithms in quenched @xmath0 , @xmath61 are not known and we have included only the logarithmic portion of the next - to - leading - order , 1-loop corrections in @xmath12 extrapolations . 6 c. bernard , _ et .
* d32 * ( 1985 ) 2343 .
j. noaki , _ et .
hep - lat/0108013 , j. noaki , these proceedings . t. blum , _ et .
al . _ , hep - lat/0007038 m. golterman and e. pallante , jhep * 08 * , 023 ( 2000 ) , hep - lat/0006029 .
t. blum , _ et .
al . _ ( rbc ) , hep - lat/0110075 , r. mawhinney , these proceedings .
j. bijnens , phys . lett .
b * 152 * ( 1985 ) 226 . | we have used domain wall fermions to calculate @xmath0 and @xmath1 matrix elements which can be used to study the @xmath2 rule for k decays in the standard model .
nonlinearities in the @xmath3 matrix elements due to chiral logarithms are explored and the subtractions needed for the @xmath2 matrix elements are discussed .
using renormalization factors calculated using non - perturbative renormalization then yields values for real @xmath4 and @xmath5 .
we present the details of our quenched @xmath6 , @xmath7 , @xmath8 simulation , where a previous calculation showed that the finite @xmath9 chiral symmetry breaking effects are small ( @xmath10 ) . |
an important breakthrough occurred in the early 2000 s after realizing a rich integrable structure of the laplacian growth problem @xcite .
remarkable connections of the laplacian growth with integrable hierarchies and random matrices provide ample opportunities to address long - standing problems with novel methods .
a particularly important example is a consolidation of the two - dimensional dyson gas theory @xcite , quantum hall effect @xcite , diffusion - limited aggregation , and laplacian growth within a single framework of the random matrix ensembles with complex eigenvalues . when the size of the matrices , @xmath0 , becomes large
some new features emerge , and the language of statistical equilibrium thermodynamics provides an adequate description of the matrix ensemble .
different aspects of the @xmath1 expansion of the free - energy and density correlation functions were discussed in @xcite . at large @xmath0
the eigenvalues densely fill a domain in the complex plane ( the support of eigenvalues ) , and their density steeply drops down at the edge . under a change of parameters of the random matrix ensemble and the size of matrices the support of eigenvalues grows in accordance with the laplacian growth law @xcite .
this growth , however , is an _
( quasi - static ) process , as the passages between states in thermodynamic equilibrium are only considered this is a limitation of the equilibrium thermodynamics .
_ this paper aims _ to apply common methods of non - equilibrium thermodynamics to the laplacian growth problem by taking into account small fluctuations of the parameters of the matrix ensembles , which occur close enough to equilibrium .
this branch of thermodynamics , also known as the `` linear thermodynamics '' @xcite , allows us to focus on transitions between thermodynamic states , represented by equilibrium distributions of eigenvalues within planar domains .
the probability of transitions between different thermodynamic states ( or , equivalently , supports of eigenvalues ) is an essential step towards the path - integral formulation of the laplacian growth .
_ structure of the paper is straightforward . _ after introducing the grand - canonical partition function of the two - dimensional dyson gas in the large @xmath0 limit , we derive the laplacian growth equation by maximizing the transitional probability between equilibrium states .
afterwards , following einstein s theory of thermodynamic fluctuations @xcite we determine the probability for occurring fluctuations of the parameter of the matrix ensemble .
a particular class of fluctuations generates the _
stochastic laplacian growth _ of the support of eigenvalues .
remarkably , the growth probability is determined by the free - particle propagator on the infinite - dimensional complex manifold with the khler metric .
finally , we draw our conclusion and indicate some open problems .
the grand - canonical partition function of the two - dimensional coulomb plasma ( often called the dyson gas @xcite ) in the thermal equilibrium at `` temperature '' @xmath2 in the external potential , @xmath3 , is @xmath4 where @xmath5 is the lagrange multiplier , which fixes the total number of eigenvalues , drop the inessential term @xmath6 , and rescale @xmath7 afterwards . ] , and the energy of the gas equals : @xmath8 a ) the electronic droplet in the external potential ; the point charges ( impurities ) are marked by @xmath9 .
b ) the support of eigenvalues @xmath10 ( solid line ) , and the background charge occupying the domain @xmath11 ( dashed line ) . ] a particularly important special case arises if @xmath3 is a quasiharmonic potential : @xmath12 , where @xmath13 is a holomorphic function in the domain which includes the support of eigenvalues @xmath10 .
the two valuable interpretations of @xmath3 are : 1 ) @xmath14 is the electric potential created by _ impurities _ ( point electric charges ) located in the exterior of @xmath10 ( fig
. [ droplet]a ) , and 2 ) @xmath3 is the potential generated by the _ background charge _ , which fills the domain @xmath15 with a unit density ( fig .
[ droplet]b ) , i.e. @xmath16 in the interior and exterior of @xmath11 the potential @xmath3 admits the series expansions : @xmath17 the coefficients of the expansions , @xmath18 and @xmath19 , are the harmonic moments of @xmath11 : @xmath20 _ the mean values _ of symmetric functions of particle coordinates are defined in a usual way : @xmath21 some important examples of the symmetric functions are the logarithmic function , @xmath22 , and polynomials @xmath23 : @xmath24 in the large @xmath0 limit these functions become the harmonic moments of the support of eigenvalues @xmath10 , defined in a similar way to .
by @xmath25 and @xmath26 we will denote the compliment set of the harmonic moments of @xmath10 .
at large @xmath0 and low enough temperature the dyson gas behaves as an incompressible charged liquid .
it is convenient to introduce the short distance cutoff , @xmath27 , and define the quasiclassical limit as @xmath28 and @xmath29 while @xmath30 is kept fixed .
then , the eigenvalues densely fill the domain @xmath10 in a complex plane , so that the mean density in the exterior of @xmath10 is exponentially small as @xmath28 .
since the density of eigenvalues , is a smooth function in the large @xmath0 limit , the partition function of the dyson gas can be symbolically rewritten as : @xmath31)}[d\rho],\ ] ] where is the area of @xmath10 , and we rescale @xmath32 for convenience .
the energy of the dyson gas can be written in terms of the density function . in the leading order
we have , which results from the atomic structure of the dyson gas , can be neglected at large scales . ] : @xmath33 minimizing the functional , @xmath34 $ ] , in , we obtain the following equation for charge distribution : @xmath35 the classical ( equilibrium ) solution to this equation , which is , @xmath36 , if @xmath37 , and @xmath38 otherwise , determines the support of eigenvalues @xmath10 in what follows .
all harmonic moments of the domains @xmath10 and @xmath11 are equal , except for @xmath39 ( or , equivalently , @xmath40 ) because of the term @xmath5 in . if @xmath41 the domains @xmath10 and @xmath11 coincide ( fig .
[ droplet]b ) , while in the limit @xmath42 the support of eigenvalues is the mother body of @xmath11 @xcite .
the extremum condition also fixes the lagrange multiplier @xmath5 . comparing the potential of the dyson gas at the origin with @xmath43 we obtain from : @xmath44 after substituting this extremal value of @xmath5 into
, the grand - canonical partition function takes the form : @xmath45)}[d\rho].\ ] ] the free - energy admits the @xmath1 expansion @xcite , and the leading contribution to @xmath46 is determined by the classical configuration @xmath47 .
the @xmath1 corrections to the free energy emerge as the atomic structure of the dyson gas is taken into account @xcite .
the language of statistical thermodynamics provides an adequate description of the dyson gas in the large @xmath0 limit .
one can introduce the entropy @xmath48 does not contribute to the entropy , as the charges are frozen at their equilibrium positions .
non - trivial contributions to the entropy result from the discrete `` atomic '' structure of the dyson gas . ] , the internal energy @xmath49 , and external coordinates , which determine the configuration of the system @xcite .
these coordinates are the parameters of the external potential , i.e. @xmath50 , @xmath51 , @xmath52 , etc . by differentiating
we obtain the _ fundamental thermodynamic relation _ , which relates changes of the entropy , the energy , and the generalized `` forces '' @xmath53 : @xmath54 where we took into account the mean values of the `` forces '' , @xmath55 and @xmath56 , in the thermodynamic equilibrium .
the _ adiabatic _ ( quasi - static ) variations of the parameters of the potential , @xmath3 , cause the evolution of the support of eigenvalues , that is a motion of @xmath10 .
the dynamical law , which governs the evolution , is incorporated in the extremum condition . by differentiating w.r.t .
@xmath57 and replacing @xmath47 by the characteristic function of @xmath10 , we obtain the functional equation for the support of the eigenvalues @xmath10 : @xmath58 taking time derivative of we arrive to the celebrated laplacian growth equation @xcite , @xmath59 written in terms of the schwarz function , @xmath60 , for the boundary @xmath61 away from the contour @xcite . ] , and of the complex potential , @xmath62 , which determines the change of the external potential with time if @xmath63 . ] .
following the einstein s theory of thermodynamic fluctuations @xcite the probability of the fluctuation , @xmath64 , is determined by the entropy difference @xmath65 between two thermodynamic states .
the change of the entropy of the system , @xmath66 , can be calculated using the minimal reversible work , @xmath67 , which has to be applied to bring the system out of equilibrium . for @xmath67
one can use the expression @xcite : @xmath68 where @xmath69 , @xmath70 , @xmath19 , @xmath26 are the equilibrium values of temperature , harmonic moments , and the area of the liquid droplet . using the fundamental thermodynamic relation we find in the quadratic approximation the probability for occurring fluctuations of the thermodynamic variables : @xmath71 the large @xmath0 limit of the dyson gas corresponds to a very low temperature , when the charges are frozen at their equilibrium positions and their fluctuations are negligible .
thus , we will consider the classical fluctuations at the constant temperature , @xmath72 , only .
remarkably , the highly - unstable and non - equilibrium laplacian growth can be considered as small ( gaussan ) fluctuations near equilibrium in the beta - ensemble of complex eigenvalues .
different ways of adding new particles to the system result into different growth processes .
a particular important example is an aggregation of diffusing particles issued from the source at infinity .
since the particles can freely arrive to the system to compensate the background charge , we have @xmath73 and @xmath74 at the thermodynamics equilibrium . _ the deterministic growth _
, we introduce first , serves as a reference point in the space of all growth processes .
it is caused by a specific variation of the background potential , when all @xmath18 s ( except @xmath70 ) are kept fixed .
the new particles arrive to the system from the distant source to compensate the change of @xmath3 .
being attached to the boundary of @xmath10 with a rate proportional to the local electric field , they cause the growth of the support of eigenvalues . since changing of @xmath3 is a much faster process than adjusting of the domain @xmath10 , the chemical potential for newly incoming particles is non - zero and is given by . expanding this formula in @xmath75
, we obtain : @xmath76 where @xmath77 , and @xmath78 is the conformal radius of @xmath11 .
usually , the conformal radius is defined by the conformal map , @xmath79 , from the compliment to the unit disk to the exterior of @xmath11 .
the map is normalized so that @xmath80 , and @xmath81 is real and positive , i.e. @xmath82 representing the logarithm of the conformal radius as the solution to the dirichlet boundary problem , where @xmath83 is the green s function of the dirichlet boundary problem in the exterior of @xmath11 @xcite , and @xmath84 at the boundary . ]
, we recast the chemical potential in the form : @xmath85 where @xmath86 is the inverse to the conformal map , and @xmath87 is a normal displacement of the boundary , @xmath88 , due to the change @xmath89 .
the newly attached particles cause the growth of @xmath10 with the normal velocity , @xmath90 , which prompts the darcy s law , where @xmath91 is the rate of the growth .
thus , the evolution of the interface follows the _ deterministic laplacian growth dynamics _ : @xmath92 where @xmath93 parametrizes the boundary . by plugging @xmath94 , @xmath95 , and @xmath96 to
, we conclude that the deterministic laplacian growth maximizes the probability . _ the stochastic growth processes _ deviate from the deterministic laplacian growth dynamics by random fluctuations of the interface during the evolution .
these fluctuations are generated by newly arrived particles , which tend to compensate the change of the parameters of the external potential , @xmath97 , caused by variations of the impurities ( fig .
[ droplet]a ) .
the latter are related by the richardson theorem @xcite : @xmath98 where @xmath99 is the position of @xmath100-th impurity , and @xmath101 is its rate of change of the charge .
the corresponding interface dynamics is then described by the _
stochastic laplacian growth equation _ : @xmath102 where @xmath103 are the inverse pre - images of @xmath104 inside the unit disk .
this equation is well known in the classical laplacian growth problem in the presence of point - like hydrodynamical sources @xmath101 at the points @xmath99 @xcite .
the general formula determines the small fluctuations of various thermodynamic quantities .
the harmonic moments , @xmath18 and @xmath19 , as the coefficients of the multipole expansion of the potential , are interrelated via a certain potential function @xcite : @xmath105 where @xmath106 is the logarithm of the tau - function of the boundary curve @xcite , @xmath107 which also provides a solution to the dispersionless limit of the 2d integrable toda hierarchy @xcite .
the symmetry relations between the harmonic moments suggest to chose @xmath18 s as independent variables .
variations of @xmath18 s also cause the deviation of the chemical potential for newly incoming particles from its `` deterministic '' value .
since @xmath108 we transform the line integral in in the contour one .
the laplacian growth equation relates the change of schwarz function with the variation of the electric potential : @xmath109 is a regular in @xmath11 part of the schwarz function @xmath110 . ] .
thus , evaluating the contour integrals we determine the variation of the chemical potential : @xmath111 now , using the symmetry relation one can express @xmath112 s in terms of @xmath113 s to transform the probability for occurring fluctuation in the quadratic form in @xmath113 and @xmath114 only .
as for the term @xmath115 , .
] we use and take into account description of the conformal map , @xmath86 , in terms of the tau - function @xcite , which leads to the formula : @xmath116 thus allowing to express @xmath117 through @xmath113 s . as discussed earlier , the variations of @xmath18 s are caused by fluctuations of the impurities in the exterior of @xmath10 ( fig . [
droplet]a ) .
thus , the quadratic form in the exponent of ( upon rewriting in terms of @xmath18 s ) includes the cross terms of the form @xmath118 .
this terms , however , do not contribute to the mean fluctuation values of any quantities , if the statistical independence of impurities , @xmath119 for @xmath120 , is assumed . under these circumstances
the probability for occurring fluctuations can be written in the form : @xmath121 _ the mean square fluctuations of the harmonic moments _
, @xmath122 then are determined by the coefficients of the quadratic form in the exponent of , @xmath123 and @xmath124 are the elements of the inverse matrix
. the mean square fluctuation of @xmath112 s are obtained from by virtue of the symmetry relation .
it also follows from that the area of the droplet does not fluctuate with time , as we consider fluctuations of the shape of the interface against the background of the growth process with a constant rate . if growth continues until time @xmath125 ( @xmath126 is integer ) the associated probability of the process
is given by the product of the _ conditional probabilities _
: @xmath127 since @xmath128 depends solely ( through the tau - taunction ) on the present domain at time @xmath129 , the product in is a markovian chain . in the limit @xmath130 by summing the exponents of the probabilities we obtain : @xmath131 remarkably , the exponent in is the action for a non - relativistic massless particle on the infinite dimensional complex manifold with the khler metric , so @xmath106 is the khler potential @xcite .
although the growth of the support of eigenvalues was treated as response to the variation of the external potential , one can forget the details of this derivation and consider the growth probability on its own . then , assuming that fluctuations of the interface during evolution are inevitable features of the unstable laplacian growth , formula allows to compare probabilities of different growth scenarios , which lead to the same final domain starting , e.g. , from the ideal circle .
under these circumstances the path - integral formulation did what it was supposed to do by indicating the most probable growth scenario .
application of methods of the statistical mechanics to stochastic growth phenomena might throw new light on the long - standing problems . in this paper
we have touched only the laplacian growth , because of its impressively wide applicability ranging from solidification and oil recovery to biological growth . also , it possesses beautiful and powerful properties , unusual for most of nonlinear pdes , such as infinitely many conservation laws and closed form exact solutions .
a particularly important new result is an explicit calculation of the growth probability , which turns out to be the propagator of a free - particle on the infinite dimensional complex manifold with the khler metric . in the light of this simple result
several natural questions come to the surface .
does this manifold possesses new interesting mathematics , owing to the fact that the khler potential is the logarithm of the tau - function of the two - dimensional integrable toda hierarchy ?
what are the geodesics in this space ? what is their role from the laplacian growth point of view ? and so on .
remarkably , the fluctuation theory of the laplacian growth also has a rich mathematical structure if to use a dual description of lg ( known as the inverse potential problem ) in terms of the conformal maps .
here we only mention some results , while the details can be found elsewhere @xcite .
the probability of the fluctuations generalizes the result of @xcite , where the semiclassical one - particle wave function at the boundary of the electronic droplet for @xmath132 was calculated , to the case of the non - vanishing ( in the limit @xmath133 ) stochastic growth of the droplet .
being written in terms of the singularities of the conformal map , the growth probability resembles the semiclassical correlation function of the exponential operators in the liouville conformal field theory on a pseudoshere .
thus , the semiclassical corrections to growth probability might provide a deep insight in the link between laplacian growth and conformal field theories first noted in @xcite .
the most important missing pieces of the construction is a distribution function for stochastic impurities .
there are strong reasons to believe that fluctuations in the laplacian growth do not decay exponentially with time in contrast to the usual fluctuation theory .
once this issue is resolved , the role of the fluctuation - dissipation theorem in the laplacian growth will be much more clear .
i. kostov , i. krichever , m. mineev - weinstein , p. wiegmann and a. zabrodin , _
@xmath134-function for analytic curves _ , random matrices and their applications , msri publications , * 40 * , cambridge university press , 285 ( 2001 ) ; http://arxiv.org/abs/hep-th/0005259[arxiv:hep-th/0005259];t r. teodorescu , e. bettelheim , o. agam , a. zabrodin and p. wiegmann , _ normal random matrix ensemble as a growth problem _ , nucl . phys .
b * 704 * , 407 ( 2005 ) ; https://arxiv.org/abs/hep-th/0401165v2[arxiv:hep-th/0401165 ] ; l. takhtajan , _ free bosons and tau - functions for compact riemann surfaces and closed smooth jordan curves i. current correlation functions _ , lett.math.phys .
* 56 * , 181 ( 2001 ) ; https://arxiv.org/abs/math/0102164v1[arxiv:math/0102164 [ math.qa ] ] ; o. agam , e. bettelheim , p. wiegmann and a. zabrodin , _ viscous fingering and a shape of an electronic droplet in the quantum hall regime _ , phys.rev.lett . * 88 * , 236801 ( 2002 ) ; https://arxiv.org/abs/cond-mat/0111333v3 [ arxiv : cond - mat/0111333 ] ; | we develop statistical mechanics for stochastic growth processes as applied to laplacian growth by using its remarkable connection with a random matrix theory .
the laplacian growth equation is obtained from the variation principle and describes adiabatic ( quasi - static ) thermodynamic processes in the two - dimensional dyson gas . by using einstein s theory of thermodynamic fluctuations we consider transitional probabilities between thermodynamic states , which are in a one - to - one correspondence with planar domains .
transitions between these domains are described by the stochastic laplacian growth equation , while the transitional probabilities coincide with the free - particle propagator on the infinite dimensional complex manifold with the khler metric . |
the astrophysically ubiquitous keplerian accretion disks should be unstable and turbulent in order to explain observed data , but are remarkably rayleigh stable .
they are found in active galactic nuclei ( agns ) , around a compact object in binary systems , around newly formed stars etc .
( see , e.g. , * ? ? ?
the main puzzle of accreting material in disks is its inadequacy of molecular viscosity to transport them towards the central object .
thus the idea of turbulence and , hence , turbulent viscosity has been proposed .
similar issue is there in certain shear flows , e.g. plane couette flow , which are shown to be linearly stable for any reynolds number ( @xmath0 ) but in laboratory could be turbulent for @xmath0 as low as @xmath1 . therefore , linear perturbation can not induce the turbulent viscosity to transport matter inwards and angular momentum outwards , in the keplerian disks .
note that the issue of linear instability of the couette - taylor flow ( when accretion disks are the subset of it ) is a century old problem .
although in the presence of vertical shear and/or stratification , keplerian flow may reveal rayleigh - taylor type instability ( e.g. @xcite ) , convective overstability ( @xcite ) and the zombie vortex instability ( @xcite ) , we intend here to solve the classic century old problem of the origin of linear instability with the exponential growth of perturbation in purely hydrodynamical rayleigh - stable flows with only radial shear
. the convective overstability does not correspond to an indefinitely growing mode and it has some saturation ( @xcite ) .
in addition , the zombie vortex instability is not sufficient to transport angular momentum significantly in a small domain of study .
in fact , all of them could exhibit only smaller shakura - sunyaev viscosity parameter ( @xcite ) @xmath2 than that generally required to explain observation .
the robustness of our work is that , it can explain the turbulent behavior of any kind of rayleigh - stable shear flows , starting from laboratory to astrophysical flows .
while many realistic non - magnetized and keplerian flows could be stratified in both the vertical and radial directions of the disks , it is perhaps impossible to prove that all the non - magnetized accretion disks have significant amount of vertical shear and/or stratification to sustain the above mentioned instabilities .
note that indeed many accretion disks are geometrically thin .
moreover , the laboratory taylor - couette flows have no vertical shear and/or stratification . in 1991 , with the application of magnetorotational instability ( mri ; * ? ? ?
* ; * ? ? ?
* ) to keplerian disks , @xcite showed that initial weak magnetic field can lead to the perturbations growing exponentially . within a few rotation times
, such exponential growth could reveal the onset of turbulence .
however , for charge neutral flows mri should not work .
note also that for flows having strong magnetic fields , where the magnetic field is tightly coupled with the flow , mri is not expected to work ( e.g. @xcite ) .
it is a long standing controversy ( see , e.g. , * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
* ; * ? ? ?
? * ; * ? ? ?
* ) , whether the matter in rayleigh stable astrophysical disks is stable or unstable .
the answer has profound significance for our understanding of how stars and planets form .
it is argued , however , that some types of rayleigh stable flows certainly can be destabilized @xcite .
based on ` shearing sheet ' approximation , without @xcite and with @xcite explicit viscosity , some authors attempted to tackle the issue of turbulence in hot accretion disks .
however , other authors argued for limitations in this work @xcite .
based on the simulations including explicit viscosity , the authors could achieve @xmath3 and concluded that keplerian like flows could exhibit very weak turbulence in the absence of magnetic field . nevertheless , the recent experimental results by @xcite clearly argued for the significant level of transport from hydrodynamics alone .
moreover , the results from direct numerical simulations @xcite and exploration of transient amplification , in otherwise linearly stable flows , with and without noise ( e.g. * ? ? ?
* ; * ? ? ?
* ) also argued for ( plausible ) hydrodynamic instability and turbulence at low @xmath0 .
interestingly , accretion disks have huge @xmath0 ( @xmath4 ) @xcite , prompting to the belief that they are hydrodynamically unstable .
we show here that linearly perturbed apparently rayleigh stable flows driven stochastically can be made unstable even in the absence of any magnetic field .
we also argue , why stochastic noise is inevitable in such flows .
they exist in the flows under consideration inherently .
we develop our theory following the seminal concept based on fluctuating hydrodynamics of randomly stirred fluid , pioneered by @xcite and @xcite , which , however , was never applied in the context of accretion flows or other shear flows .
this work provides a new path of linear hydrodynamic instability of shear flows , which will have vast applications from accretion disks to laboratory flows , for the first time .
the plan of the paper is the following .
in the next section , we introduce equations describing the system under consideration . then
3 describes the evolution of various perturbations in stochastically driven hydrodynamic flows .
subsequently , we discuss the relevance of white noise in the context of shear flows in 4 . finally we summarize with conclusions in 5 . in appendix
, we demonstrate in detail the generation of white noise from random walk , particularly in the present context .
the linearized navier - stokes equation in the presence of background plane shear @xmath5 and angular velocity @xmath6 , when @xmath7 being the distance from the center of the system , in a small section approximated as incompressible flow with @xmath8 , has already been established @xcite . here , any length is expressed in units of the size @xmath9 of the system in the @xmath10direction , the time in units of @xmath11 , the velocity in @xmath12 ( @xmath13 ) , and other variables are expressed accordingly ( see , e.g. , @xcite , for detailed description of the choice of coordinate in a small section ) .
hence , in dimensionless units , the linearized navier - stokes equation and continuity equation ( for an incompressible flow ) can be recasted into the well - known orr - sommerfeld and squire equations , but in the presence of stochastic noise and coriolis force @xcite , given by @xmath14 @xmath15 where @xmath16 are the components of noise arising in the linearized system due to stochastic forcing such that @xmath17 @xcite , where @xmath18 is a constant for white noise and @xmath19 ; @xmath20 is the @xmath21-component of velocity perturbation vector and @xmath22 the @xmath21-component of vorticity perturbation vector .
now , we can resort to a fourier series expansion of @xmath20 , @xmath22 and @xmath23 as @xmath24 where @xmath25 can be any one of @xmath20 , @xmath22 and @xmath23 ; @xmath26 and @xmath27 are the wavevector and frequency respectively in the fourier space such that @xmath28 and @xmath29=@xmath30 .
writing down equations ( [ hydroorrv ] ) and ( [ hydroorrzeta ] ) in fourier space by using equation ( [ fourier ] ) , and taking ensemble average , we obtain the equations involving the evolution of mean values of perturbations in the presence of noise as @xmath31 @xmath32 where fourier transformations of @xmath16 are basically @xmath33 multiplied with a random number and on ensemble average it appears to be a constant @xmath34 which is the mean value of the white noise ( we get @xmath35 and @xmath36 when the drift coefficient of the brownian motion or wiener process corresponding to the white noise is zero and nonzero respectively , see appendix for details ) , and @xmath37 , @xmath38 are the fourier transforms of @xmath39 and @xmath40 which are the mean or the ensemble averaged values of @xmath20 and @xmath22 respectively .
now let us take the trial solutions , @xmath41 , where @xmath42 are the constant , in general complex , amplitudes of perturbation and @xmath43 , is a vertical wavevector ( one should not confuse this @xmath44 with the shakura - sunyaev viscosity parameter ) .
vertical wave vector is chosen since it will be unaffected by shear @xcite .
this gives @xmath45 ( using equation ( [ fourier ] ) ) .
substituting these trial solutions in equations ( [ hydrofourierumean ] ) and ( [ hydrofourierzetamean ] ) , integrating with respect to @xmath26 and @xmath27 we obtain @xmath46 @xmath47 now eliminating @xmath34 and assuming @xmath48 we obtain the dispersion relation @xmath49 if we find any pair of @xmath44 and @xmath50 satisfying equation ( [ vertmeli ] ) for which the imaginary part of @xmath50 positive , then we can say that the mean value of perturbation is unstable .
equation ( [ vertmeli ] ) is the hydrodynamic counter part of the dispersion relation obtained due to mri @xcite , leading to the avenue of pure hydrodynamic instability . for @xmath35 , from equations ( [ hydroudispersion ] ) and ( [ hydrozetadispersion ] ) , either @xmath51 and @xmath52 both turn out to be zero or there is no instability for non - trivial @xmath51 and @xmath52 .
overall , @xmath35 gives rise to stable solutions like the zero magnetic field for mri .
figure [ vertm ] shows the ranges of @xmath44 giving rise to linear instability .
it is easy to understand that similar results could be obtained with the choice of unequal ensemble averages of white noise in equations ( [ hydroudispersion ] ) and ( [ hydrozetadispersion ] ) and a more general phase difference between @xmath52 and @xmath51 . and the imaginary part of @xmath50 , for vertical perturbation in case i with @xmath53 .
we consider @xmath54 , however we obtain almost the same results for other admissible values of @xmath55 . ] 1.5 cm now for a given @xmath51 and @xmath34 , after eliminating @xmath52 from equations ( [ hydroudispersion ] ) and ( [ hydrozetadispersion ] ) , we obtain a dispersion relation between @xmath44 and @xmath50 as @xmath56 which is second order in @xmath50 and hence has two roots @xmath57 and @xmath58 .
if we find any pair of @xmath44 and @xmath50 for which the imaginary part of @xmath50 positive , then we can say that the mean value of perturbation is unstable . for @xmath35 in equation ( [ dispersion ] ) , there is no instability , like the zero magnetic field for mri .
@cc@ and the imaginary part of one of the solutions of @xmath50 , for vertical perturbation in case ii .
we consider @xmath59 , however we obtain almost the same results for other admissible values of @xmath55 . other solution of @xmath50 is stable.,title="fig : " ] and the imaginary part of one of the solutions of @xmath50 , for vertical perturbation in case ii .
we consider @xmath59 , however we obtain almost the same results for other admissible values of @xmath55 .
other solution of @xmath50 is stable.,title="fig : " ] + + + + and the imaginary part of one of the solutions of @xmath50 , for vertical perturbation in case ii .
we consider @xmath59 , however we obtain almost the same results for other admissible values of @xmath55 . other solution of @xmath50 is stable.,title="fig : " ] and the imaginary part of one of the solutions of @xmath50 , for vertical perturbation in case ii .
we consider @xmath59 , however we obtain almost the same results for other admissible values of @xmath55 .
other solution of @xmath50 is stable.,title="fig : " ] 1.5 cm in fig . [ unstable ] , for @xmath60 and different values of @xmath0 above a certain value , we show that for keplerian flows , there are modes for which the mean values of perturbation are unstable . if the amplitude of perturbations decreases , the value of @xmath61 increases for any fixed nonzero @xmath34 , leading to a larger range of @xmath44 for instability .
however for @xmath35 , i.e. for the white noise with zero mean ( which also corresponds to the hydrodynamic accretion flows without any noise ) , we obtain no such unstable modes .
while modes are stable for smaller @xmath0 , with the increase of @xmath0 they become unstable and range of @xmath44 giving rise to instability increases with increasing @xmath0 and for @xmath62 unstable modes arise all the way upto @xmath63 . [
cols="^,^ " , ] in fig .
[ unstablemsp ] , we show how spherical perturbation modes in keplerian flows vary with @xmath64 .
this is very similar to as shown in fig .
[ unstablem ] for vertical perturbations , except that the modes are stable for a very small but non - zero @xmath64 , while for vertical perturbation the modes remain unstable for @xmath65 . for plane couette flows , making @xmath66 in equation ( [ dispersionsp ] )
, we obtain the corresponding dispersion relation as @xmath67 while the first root always corresponds to the stable mode for a real @xmath44 , the second one will lead to the unstable solution for a negative @xmath34 satisfying @xmath68 . and imaginary part of one of the solutions of @xmath50 , for spherical perturbation in case ii , where @xmath59 .
it shows the unstable modes for negative @xmath34 ( drift velocity).,width=321 ] figure [ negmsp ] shows that for spherical perturbations , the keplerian flows remain unstable upto @xmath69 , as shown in fig .
[ negm ] for vertical perturbation cases .
now we shall discuss that how relevant and how likely the white noise is to be present in shear flows .
the rayleigh stable flows under consideration have a background shear profile , with some molecular viscosity however small that may be , and hence some drag ( e.g. , in protoplanetary disks , it could be due to the drag between gas and solid particles ) . for plane couette flow , such shear is driven in the fluids by moving the boundary walls by externally applied force . if the external force ( cause ) is switched off , the shearing motion ( effect ) dies out .
similarly , in accretion disks , the central gravitational force plays the role of driving force ( cause ) producing differential velocity ( shear ) in the flow .
hence , by fluctuation - dissipation theorem of statistical mechanics ( see , e.g. , * ? ? ?
* ; * ? ? ?
* ) , there must be some thermal fluctuations in such flows , with some temperature however low be , and that cause the fluid particles to have brownian motion . therefore the time variation ( derivative ) of this brownian motion , which is defined as white noise , plays the role of extra stochastic forcing term in the orr - sommerfeld equations ( equations ( [ hydroorrv ] ) , ( [ hydroorrzeta ] ) ) which are present generically , in particular when perturbation is considered .
now , due to the presence of background shear in some preferential direction , it is very likely for the fluid particles to have brownian motion with nonzero drift , however small it may be .
the detailed technical description of generation of white noise ( with zero and nonzero mean ) from brownian motion has been included in appendix .
therefore , if @xmath70 is the random displacement variable of a brownian motion with drift coefficient @xmath34 , its probability density function @xmath71 can be written as @xmath72 , \label{brownian0}\ ] ] where @xmath73 is the standard deviation of the distribution and @xmath74 the time .
taking the stochastic time derivative of @xmath70 , we obtain the white noise process which we denote by @xmath75 ( @xmath76 ) .
since the stochastic variable @xmath70 is not differentiable in the usual sense , we consider a finite difference approximation of @xmath75 using a time interval of width @xmath77 as @xmath78 therefore , the presence of infinitesimal molecular viscosity ( and shear ) , which is there always , would be enough just to give rise to a nonzero ( infinitesimal ) temperature , leading to thermal noise which can do the rest of the job of governing instability .
note that a very tiny mean noise strength , due to tiny asymmetry in the system , is enough to lead to linear instability , as demonstrated in previous sections . here , the externally applied force ( for plane couette ) or the force arising due to the presence of strongly gravitating object ( accretion disk ) introduces the asymmetry in the system , just like , e.g. , the brownian ratchets which has several applications in soft condensed matter and biology ( see , e.g. , @xcite ) .
the measure of asymmetry and drag determines the value of @xmath34 , which furthermore controls the growth rate of perturbation .
the corresponding power spectrum appears to be almost flat / constant ( for ideal white noise it is purely flat ) .
although in our chosen shearing box , the azimuthal direction is assumed to be periodic , every such small box always encounters drag and hence thermal fluctuation , which assures the presence of nonzero mean noise . as a result , every such sharing box reveals exponential growth of perturbation .
we have shown that linearly perturbed hydrodynamic apparently rayleigh stable rotating shear flows , including accretion disks , and plane couette flow , driven stochastically , can indeed be unstable , since the averaged values of the perturbations grow exponentially . due to background shear and hence
drag , thermal fluctuations arise in these flows which induce brownian motion of the fluid particles and hence stochastic forcing by white noise .
therefore the accretion flows , in particular due to perturbation , are inevitably driven by white noise which can not be neglected .
it is indeed shown in experiments that the stochastic details decide whether turbulence will spread or eventually decay @xcite , which furthermore argues for the determining factor played by stochastic forcing , which we demonstrate here for the first time .
since the forcing term in this system is a random variable , the solutions of the perturbations @xmath79 are also random variables and hence have some distributions whose averaged values are investigated . hence , we have shown that even in the absence of magnetic field , accretion disks can be made unstable and plausibly turbulent if they are driven by stochastic noise which is very likely to be present in the disks due to thermal fluctuations .
in fact , we argue that neglecting the stochastic noise in accretion flows and any other shear flows is vastly an inappropriate assumption .
this is because , some shear is always there ( because those are always driven externally by definition ) , which leads to some temperature ( however be the magnitude ) and a small temperature is enough to reveal stochastic noise , which is the basic building block of our work .
hence , the presence of ( asymmetric ) drag and stochastic noise in shearing flows is inherent .
hence , this work inevitably presents the origin of pure hydrodynamic instability of rotating shear flows and plane couette flow .
therefore , this sheds enormous light on to the understanding of formation of planets and stars .
evidently this mechanism works for magnetized shear flows as well , because thermal fluctuations are available there also .
for example , a background field of the order of unity with @xmath80 can easily lead to unstable modes of perturbation for @xmath81 in the limit of very large @xmath0 and @xmath82 which is the case in accretion disks . in future
, we will report this result in detail .
indeed , earlier we studied stochastically driven magnetized flows and showed them to be plausibly unstable and turbulent by calculating the correlation functions of perturbations ( @xcite ) .
hence the pure hydrodynamic instability explored here is generic .
this is , to the best of our knowledge , the first solution to the century old problem of hydrodynamic instability of apparently rayleigh stable flows . in due courses , one has to investigate how exactly the required value of stochastic forcing strength could be arised in real systems and if the growth rates of unstable modes could adequately explain data . in certain cases ,
only high @xmath0 reveals instability which might be difficult to achieve in laboratory experiments and numerical simulations as of now .
we have assumed here that the white noise has a nonzero mean value .
the term _ white noise _ is ambiguous . to shed light on this matter , here
we point out the two definitions of white noise .
@xcite defines it as + + . . .
a stationary random process having a constant spectral density function .. + + @xcite defines it as + + we shall say that a process @xmath83 is white noise if its values @xmath84 and @xmath85 are uncorrelated for every @xmath86 and @xmath87 : @xmath88 .. + + the following subsections explore the implications of each definition with respect to the mean of the resulting process .
let @xmath70 is an _ ergodic stochastic process _ with the property that it has a constant power spectral density , i.e. @xmath89 where @xmath90 is the power spectral density of the random variable @xmath70 and @xmath44 is a constant .
then the corresponding autocorrelation function for the process is @xmath91=\phi_{xx}(\tau)=\alpha \delta(\tau ) , % ~{\rm by~taking~inverse~fourier } \\ % \nonumber \label{autocorr}\end{aligned}\ ] ] by taking inverse fourier transform of @xmath90 , where @xmath92 $ ] denotes the expectation value .
now let us assume that @xmath70 is a zero mean white noise process and @xmath93 is a nonzero mean process
. then @xmath94=e[(x(t)+m)(x(t+\tau)+m)]=\alpha\delta(\tau)+m^2 .
\\ % \label{brownian}\end{aligned}\ ] ] therefore @xmath95 which is not constant , thus @xmath96 violates the requirement of the white noise process by this definition . representing the papoulis definition of white noise in our notations ,
we can write , a stochastic process @xmath70 is called a white noise process if any two distinct random variables of this stochastic process are independent and uncorrelated , i.e. , the autocovariance function @xmath97 when @xmath98 . in mathematical notation ,
@xmath99 \\
\nonumber = e[x(t)]e[x(t+\tau)]-m_tm_{t+\tau } \\ = m_tm_{t+\tau}-m_tm_{t+\tau}=0 , \label{cov}\end{aligned}\ ] ] where @xmath100 and @xmath101 are the corresponding mean values of the random variables @xmath70 and @xmath102 respectively .
we can write the second equality in equation ( [ cov ] ) since @xmath70 , for different values of @xmath74 , are independent random variables by definition .
_ thus it is not necessary that a white noise process always has to have a zero mean , from papoulis definition_. that is , a stochastic process having nonzero mean can be a white noise process according to this definition .
_ in the present work , we have used papoulis definition of white noise which can indeed have a non - zero mean_. now let us explain why we have chosen papoulis definition over brown s definition .
let us consider a signal @xmath103 with constant power spectral density @xmath104 .
that is , @xmath105 , \label{corrf}\end{aligned}\ ] ] and the fourier transform of @xmath106 is @xmath107 a constant . now
the _ parseval s theorem _ tells us that @xmath108 where @xmath109 is the fourier transform of @xmath103 . since @xmath104 and
consequently @xmath110 has a constant positive value according to brown s definition of white noise , the equation ( [ energyf ] ) tells us that the total power of the signal is infinity ( see , e.g. , @xcite ) . in mathematical terminology
, the energy norm of the signal @xmath103 is infinity and hence the function @xmath103 is not @xmath111-integrable .
therefore , driving a system by a stochastic noise with constant power spectral density is same as injecting infinite amount of energy into the system , which is unphysical @xcite . in this section ,
we outline the derivation of white noise starting from the random walk , via brownian motion .
1.5 cm figure [ rwdiagram ] shows an array of positions @xmath112 where @xmath113 etc .
and @xmath114 is the spacing between points . at each interval of time , @xmath115 ,
a hop is made with probability @xmath116 to the right and @xmath117 to the left .
the distribution of @xmath7 , of hops to the right , in @xmath118 steps is given by the bernoulli distribution @xmath119 the first moment ( mean ) and the second moment ( variance ) of the bernoulli distribution in equation ( [ bar ] ) is given by @xmath120 a particle that started at @xmath121 and took @xmath7 steps to the right and @xmath122 steps to the left arrives at the position @xmath123 with mean value @xmath124 notice that , if @xmath125 , or equal probability to jump to the right and the left , the average position after @xmath118 steps will remain @xmath121 .
the second moment about the mean is given by @xmath126 therefore , from the central limit theorem , the limiting distribution after many steps is gaussian , with the first and second moments just obtained ( in equations ( [ meanposition ] ) and ( [ varposition ] ) ) , given by @xmath127 ^ 2}{8npq}\right\rbrace .
\label{distposition}\end{aligned}\ ] ] if we introduce the position and time variables by the relations @xmath128 the moments of @xmath21 are given by @xmath129 the factor @xmath130 in the definition of diffusion coefficient @xmath131 is appropriate for one dimension , and would be replaced by @xmath132 if we consider the random walk in a space of dimension @xmath133 . thus the distribution moves with a driftvelocity @xmath134 and spreads with a diffusion coefficient defined by @xmath135 thus the probability distribution of the displacement @xmath21 of a particle under this random walk is @xmath136 , \label{distdisp}\end{aligned}\ ] ] where @xmath137 . a stochastic process in which the random variables @xmath70s are stationary and independent and have distribution as in equation ( [ distdisp ] ) , is called a brownian motion or wiener process .
it is very clear from the equation ( [ driftvel ] ) that , when @xmath125 , then the drift velocity is @xmath121 , which means if some random walk is fully symmetric without any bias , then only we obtain the zero drift velocity of the corresponding brownian motion ( which is known as standard brownian motion in literature ) .
however , if some process has any asymmetry ( for example hydrodynamic flows with shear in a particular direction , bulk hydrodynamic flows , flows under gravity etc . ) , the random walk of particles in that process will have some bias ( i.e. @xmath138 ) , which eventually introduces a brownian motion with _ nonzero drift velocity_. if we take stochastic time derivative of a brownian motion or wiener process , we obtain a _ white noise _ process . if @xmath70 is the random displacement variable of a brownian motion with drift velocity @xmath34 , its probability density function @xmath71 can be written as ( using equation ( [ distdisp ] ) ) @xmath139 , \label{browniansupp}\end{aligned}\ ] ] where @xmath73 is the standard deviation of the distribution and @xmath74 the time . taking the stochastic time derivative of @xmath70
, we obtain the white noise process which we denote by @xmath75 ( @xmath76 ) .
since the stochastic variable @xmath70 is not differentiable in the usual sense , we consider a finite difference approximation of @xmath75 using a time interval of width @xmath77 as @xmath140 since the stochastic random variables @xmath70 corresponding to a brownian motion process are stationary and independent , from equations ( [ browniansupp ] ) and ( [ whitenoisesupp ] ) we obtain that the white noise process has mean / averaged value @xmath34 and variance @xmath141 . as @xmath142 , the variance @xmath143 , and this white noise tends to the ideal white noise having a constant power spectral density @xcite .
however , since brownian motion is not differentiable anywhere , the ideal white noise does not exist , as also explained above from the energy norm point of view .
now we will show that the white noise defined in equation ( [ whitenoisesupp ] ) satisfies the papoulis definition of white noise , i.e. , the process is an uncorrelated stochastic process . to establish this ,
let us first note that if @xmath70 and @xmath144 are two random variables of a brownian motion with @xmath145 , then @xmath146 \\
\nonumber = e[\lbrace(x(t)-mt)-(x(s)-ms)+(x(s)-ms)\rbrace ( x(s)-ms ) ] \\ \nonumber = e[\lbrace(x(t)-x(s))-(mt - ms)\rbrace(x(s)-ms)]+e[(x(s)-ms)^2 ] \\ = 0+\sigma^2 s=\sigma^2 { \rm min}\lbrace t , s\rbrace . \label{browncov}\end{aligned}\ ] ] the third equality is possible since @xmath147 and @xmath144 are independent random variables for a brownian motion . having the result of equation ( [ browncov ] ) in hand
, we now calculate the autocovariance of white noise .
it is very easy to verify that the autocovariance function @xmath148 of two random variables @xmath149 and @xmath150 is a linear function in both of its arguments .
therefore , @xmath151 \\ \nonumber = \frac{1}{\delta t^2 } \left[c(x(t+\delta t),x(s+\delta t))-c(x(t+\delta t),x(s ) ) \right .
\\ \left .
-c(x(t),x(s+\delta t))+c(x(t),x(s))\right ] .
\label{whitecov}\end{aligned}\ ] ] when @xmath152 , i.e. @xmath153 , then using equation ( [ browncov ] ) , from equation ( [ whitecov ] ) we obtain @xmath154 now let us consider the cases when @xmath155 , i.e. when @xmath156 or @xmath157 . for @xmath156 , equations ( [ browncov ] ) and ( [ whitecov ] ) imply @xmath158 and also for @xmath157 , @xmath159 therefore , @xmath160 i.e. @xmath161 and @xmath162 are uncorrelated .
let us define @xmath163 @cc@ for two small values of @xmath77.,title="fig : " ] + + for two small values of @xmath77.,title="fig : " ] + + figure [ twodelta ] shows the variation of @xmath164 for two different small values of @xmath77 .
the function @xmath164 defined in equation ( [ whitecovdelta ] ) is an approximation of the well known delta function @xmath165 , because @xmath166 as it is seen from fig .
also the function @xmath164 satisfies the integral property of the delta function as shown below , @xmath167 hence , when @xmath142 , from equation ( [ whitecovfinal ] ) we obtain @xmath168 therefore , the noise with nonzero mean , obtained from the stochastic time derivative of brownian motion with nonzero drift , is a white noise process according to the papoulis definition and also has the correlators as defined below equation ( [ hydroorrzeta ] ) .
the authors acknowledge partial support through research grant no . istc / pph / bmp/0362 .
the authors thank amit bhattacharjee of iisc and debashish chowdhury of iit kanpur for discussions related to possible nonzero mean of white noise and the brownian ratchets .
thanks are also due to the anonymous referee and ramesh narayan of harvard for suggestions which have helped to improve the presentation of the paper .
afshordi , n. , mukhopadhyay , b. , & narayan , r. 2005 , , 629 , 373 avila , m. 2012 , phys .
lett . , 108 , 124501 avila , m. _ et al .
_ 2011 , science , 333 , 192 balbus , s. a. 2011 , nature , 470 , 475 balbus , s. a. , & hawley , j. f. 1991 , , 376 , 214 balbus , s. a. , hawley , j. f. , & stone , j. m. 1996 , , 467 , 76 barker , a. j. , & latter , h. n. 2015 , mnras , 450 , 21 barkley d. _ et al .
_ 2015 , nature , 526 , 550 bottin , s. , & chat , h. 1998 , eur .
j. b , 6 , 143 brown , r. g. 1983 , _ introduction to random signal analysis and kalman filtering .
_ , john wiley and sons cantwell , c. d. , barkley , d. , & blackburn , h. m. 2010 , phys .
flud . , 22 , 034101 chandrasekhar , s. 1960 , proc . nat . acad .
, 46 , 53 dauchot , o. , & daviaud , f. 1995 , phys .
fluids , 7 , 335 de dominicis , c. , & martin , p. c. 1979 , phys
a , 19 , 419 dubrulle , b. , dauchot , o. , daviaud , f. , longaretti , p. -y . ,
richard , d. , & zahn , j. -p .
2005 , phys .
fluids , 17 , 095103 dubrulle , b. , marie , l. , normand , c. , hersant , f. , richard , d. , & zahn , j. -p .
2005 , a&a , 429 , 1 forster , d. , nelson , d. r. , & stephen , m. j. 1977 , phys . rev .
a , 16 , 732 fromang , s. , & papaloizou , j. 2007 , a&a , 476 , 1113 gardiner , c. w. 1985 , _ handbook of stochastic methods for physics , chemistry and natural sciences _ , 2nd edition , springer gu , p. -g . ,
vishniac , e. t. , & cannizzo , j. k. 2000 , apj , 534 , 380 hawley , j. f. , balbus , s. a. , & winters , w. f. 1999 , , 518 , 394 kliemann , w. , & namachchivaya , s. 1995 , _ nonlinear dynamics and stochastic mechanics _ , crc press kim , w. -t . , & ostriker , e. c. 2000 , apj , 540 , 372 klahr , h. h. , & bodenheimer , p. 2003
, , 582 , 869 klahr , h. , & hubbard , a. 2014 , , 788 , 21 kuo , h. h. 1996 , _ white noise distribution theory _ , crc press latter , h. n. 2016 , mnras , 455 , 2608 lesur , g. , & longaretti , p. -y .
2005 , a&a , 444 , 25 lin , m .- k . , & youdin , a. n. 2015 , , 811 , 17 luki , b. , jeney , s. , tischer , c. , kulik , a. j. , forr , l. , & florin , e .-
2005 , phys .
lett . , 95 , 160601 lyra , w. 2014 , , 789 , 77 mahajan , s. m. , & krishan , v. 2008 , , 682 , 602 marcus , p. s. , pei , s. , jiang , c .- h . , & barranco , j. a. 2015 , , 808 , 87 marcus , p. s. , pei , s. , jiang , c .- h . , & hassanzadeh , p. 2013 , phys .
, 111 , 084501 miyazaki , k. , & bedeaux , d. 1995 , physica a , 217 , 53 mukhopadhyay , b. 2013 , phys .
b , 721 , 151 mukhopadhyay , b. , afshordi , n. , & narayan , r. 2005 , , 629 , 383 mukhopadhyay , b. , & chattopadhyay , a. k. 2013 , j. phys . a , 46 , 035501 mukhopadhyay , b. , mathew , r. , & raha , s. 2011 , njph , 13 , 023029 nath , s. k. , & chattopadhyay a. .k .
2014 , phys .
e , 90 , 063014 nath , s. k. , & mukhopadhyay , b. 2015 , phys .
e , 92 , 023005 nath , s. k. , mukhopadhyay , b. , & chattopadhyay a. .k .
2013 , phys .
e , 88 , 013010 nelson , r. p. , gressel , o. , & umurhan , o. m. 2013 , mnras , 435 , 2610 paoletti , m. s. , van gils , d. p. m. , dubrulle , b. , sun , c. , lohse , d. , & lathrop , d. p. 2012
, a&a , 547 , a64 papoulis , a. 1991 , _ probability , random variables , and stochastic processes , 3rd ed .
_ , wcb / mcgraw - hill poor , h. v. 2013 , _ an introduction to signal detection and estimation _ , springer science & business media pringle , j. e. 1981 , , 19 , 137 pumir , a. 1996 , phys .
fluids , 8 , 3112 richard , s. , nelson , r. p. , & umurhan , o. m. 2016 , mnras , 456 , 3571 richard , d. , & zahn , j. -p . 1999 ,
a&a , 347 , 734 rudiger , g. , & zhang , y. 2001 , a&a , 378 , 302 shakura , n. i. , & sunyaev , r. a. 1973 , a&a , 24 , 337 stoll , m. h. r. , & kley , w. 2014 , a&a , 572 , a77 stoll , m. h. r. , & kley , w. 2016 , arxiv:1607.02322 trefethen , l. n. , trefethen , a. e. , reddy , s. c. , & driscoll , t. a. 1993 , science , 261 , 578 van oudenaarden , a. , & boxer , s. g. 1999 , science , 85 , 1046 umurhan , o. m. , nelson , r. p. , & gressel , o. 2016 , a&a , 586 , a33 velikhov , e. 1959 , j. exp . theor .
, 36 , 1398 yecko , p. a. 2004 , , 425 , 385 zhong , w. x. 2006 , _ duality system in applied mechanics and optimal control _ , springer science & business media | we provide the possible resolution for the century old problem of hydrodynamic shear flows , which are apparently stable in linear analysis but shown to be turbulent in astrophysically observed data and experiments .
this mismatch is noticed in a variety of systems , from laboratory to astrophysical flows .
there are so many uncountable attempts made so far to resolve this mismatch , beginning with the early work of kelvin , rayleigh , and reynolds towards the end of the nineteenth century . here
we show that the presence of stochastic noise , whose inevitable presence should not be neglected in the stability analysis of shear flows , leads to pure hydrodynamic linear instability therein .
this explains the origin of turbulence , which has been observed / interpreted in astrophysical accretion disks , laboratory experiments and direct numerical simulations .
this is , to the best of our knowledge , the first solution to the long standing problem of hydrodynamic instability of rayleigh stable flows . |
microlensing is one of the promising techniques that can be used to search for extra - solar planets , especially located at large distances @xcite .
planet detection via microlensing is possible because the lensing event caused by a lens system containing a planet can produce noticeable anomalies in the resulting light curve when the source passes close to the lens caustics , which represents the source positions on which the lensing magnification of a point source event becomes infinity @xcite . for a lens system with a planet , there exist two or three disconnected sets of caustics . among them , one is located close to the primary lens ( central caustic ) and the other(s ) is ( are ) located away from the primary lens ( planetary caustic[s ] ) . accordingly ,
there exist two types of planet - induced anomalies : one affected by the planetary caustic ( type i anomaly ) and the other affected by the central caustic ( type ii anomaly ) @xcite .
compared to the frequency of type i anomalies , type ii anomaly occurs with a relatively low frequency due to the smaller size of the central caustic compared to the corresponding planetary caustic .
however , the efficiency of detecting type ii anomalies can be high because intensive monitoring is possible due to the predictable time of anomalies , i.e. near the peak of magnification , and the known type of candidate events for intensive follow - up monitoring , i.e. very high magnification events @xcite .
keeping the high efficiency of type ii anomaly detections in mind , @xcite pointed out that if an event is caused by a lens system having multiple planets located in the lensing zone , within which the chance for the occurrence of planet - induced anomalies is maximized , all planets will affect the central region of magnification pattern , and thus the existence of multiple planets can be inferred by detecting additionally deformed anomalies from intensive monitoring of high magnification events .
this method , however , has important limitations in identifying the existence of multiple planets and determining their parameters ( the mass ratio and the instantaneous separation between each planet and the host star ) .
this is because the anomalies induced by multiple planets are in many cases qualitatively degenerate from that induced by a single planet and even if the existence of the multiple planets is known , accurate determination of the individual planet parameters will be difficult due to the complexity of multiple planet lensing models . in this paper , we propose a new channel to search for extra - solar systems composed of multiple planets via microlensing .
the method is based on the finding of @xcite that the lensing light curve anomalies induced by multiple planets are well approximated by the superposition of those of the single planet systems where the individual planet - primary pairs act as independent lens systems .
then , if the source trajectories passes both of the outer deviation regions around the planetary caustics of the individual planets , one can identify the existence of the multiple planets .
we illustrate that the probability of successively detecting lensing light curve anomalies induced by two jovian - mass planets located in the lensing zone through this channel will be substantial .
we discuss the advantages of the proposed method over the previous method of monitoring high magnification events .
if a source located at @xmath0 on the projected plane of the sky is lensed by a lens system composed of @xmath1-point masses , where the individual components masses and locations are @xmath2 and @xmath3 , the positions of the resulting images , @xmath4 , are obtained by solving the lens equation of the form @xmath5 where @xmath6 is the total mass of the lens system and @xmath7 is the angular einstein ring radius .
the angular einstein ring radius is related to the physical parameters of the lens system by @xmath8 where @xmath9 and @xmath10 are the distances to the lens and the source from the observer , respectively .
since the lensing process conserves the surface brightness of the source , the magnification of each image is simply given by the surface area ratio between the image and the unamplified source , and mathematically its value corresponds to the inverse of the jacobian of the lens equation evaluated at the image position @xmath11 ; @xmath12 the total magnification is then given by the sum of the individual images magnifications , i.e. @xmath13 , where @xmath14 is the total number of images .
for a single point - mass lens ( @xmath15 ) , there are two solutions for the lens equation ( @xmath16 ) and the resulting total magnification is expressed in a simple analytical form of @xmath17 where @xmath18 is the dimensionless lens - source separation vector normalized by @xmath7 . for a rectilinear lens - source transverse motion ,
the separation vector is related to the single lensing parameters by @xmath19 where @xmath20 represents the time required for the source to transit @xmath7 ( einstein time scale ) , @xmath21 is the closest lens - source separation in units of @xmath7 ( impact parameter ) , @xmath22 is the time at that moment , and the unit vectors @xmath23 and @xmath24 are parallel with and normal to the direction of the relative lens - source transverse motion , respectively .
the lens system with planets is described by the formalism of multiple lensing with very low mass companions . for a binary lens system ( @xmath25 ) , the lens equation becomes a fifth order polynomial in @xmath26 and the positions of the individual images are obtained by numerically solving the equation @xcite .
this yields three or five solutions depending on the source position with respect to the lenses .
the main new feature of a multiple lens system is the formation of caustics . for a binary lens , the set of caustics
form several disconnected closed curves .
the number and locations of the caustic curves are dependent on the separation between the planet and the primary lens .
if the separation is larger than @xmath7 , there exist two sets of caustics and one is the central caustic located near the primary lens and the other is the planetary caustic located away from the primary on the planet side with respect to the primary .
the lens system having a planet with a separation smaller than @xmath7 also has a single central caustic , but has two planetary caustics located on the opposite side of the planet .
the planetary caustic(s ) is ( are ) located within the einstein ring when the projected separation between the planet and the primary ( normalized by @xmath7 ) is in the lensing zone of @xmath27 . since the sizes of both the central and planetary caustics are maximized at around this separation , planet detection probability is also maximized for systems having planets located in this lensing zone @xcite . for a lens system with triple lenses ( @xmath28 ) ,
the lens equation becomes a tenth order polynomial and it is still numerically solvable . for these lens systems , there are a maximum of ten images and a minimum of four images , and the number of images changes by a multiple of two as the source crosses a caustic . unlike the caustics of a binary lens system forming separate sets of closed curves ,
those of a triple lens system can exhibit self - intersection and nesting . 13.0 cm for lens systems with even larger number of lenses ( @xmath29 ) , numerically solving the lens equation becomes very difficult .
however , one can still obtain the magnification patterns of these lens systems by using the inverse ray - shooting technique @xcite . in this method ,
a large number of light rays are uniformly shot backwards from the observer plane through the lens plane and then collected ( binned ) in the source plane .
then , the magnification pattern is obtained by the ratio of the surface brightness ( i.e. , the number of rays per unit area ) on the source plane to that on the observer plane .
then the light curve resulting from a particular source trajectory corresponds to the one - dimensional cut through the constructed magnification pattern .
this technique has an advantage of allowing one to construct magnification patterns regardless of the number of lenses , but it has a disadvantage of requiring large computation time to obtain smooth magnification patterns .
recently , @xcite showed that the light curve anomalies induced by multiple planets are well approximated by the superposition of those of the single planet lens systems where the individual planet - primary pairs act as independent lens systems .
as pointed out by gaudi et al .
( 1998 ) , then , the anomaly pattern in the central region caused by one of the planets can be significantly affected by the existence of other planet(s ) because the central deviation regions caused by the individual planets occur in this same region . however , the anomaly pattern in the outer deviation region caused by each planet is hardly affected by other planet(s ) because the outer deviation regions of the individual planets occur , in general , at different locations .
the fact that the anomaly patterns in the outer deviation regions are scarcely affected by other planet(s ) provides a new channel to search for multiple planets and to determine their parameters . in this channel ,
multiple planets are detected when the source trajectory passes both of the outer deviation regions around the planetary caustics of the individual planets .
the greatest advantage of this method over the method of monitoring high magnification events is that one can unambiguously identify the existence of multiple planets from the unique pattern of consecutive anomalies and accurately determine the planet parameters due to the applicability of much simpler single planet lensing models to the individual anomalies .
in this section , we illustrate that the efficiency of detecting multiple planets through the new channel will be substantial . to demonstrate this
, we estimate the probability of successively detecting lensing light curve anomalies induced by multiple planets . the probability is determined for a lens system having two jovian - mass planets ( i.e. jupiter and saturn ) orbiting a host star with a mass of @xmath30 .
then , the mass ratios of the planets ( to that of the host star ) are @xmath31 and @xmath32 , for the jupiter - mass and the saturn - mass planets , respectively .
if the lens system is located located at @xmath33 kpc ( and with a source at @xmath34 kpc ) and the component planets have intrinsic orbital separations similar to those of jupiter and saturn of our solar system , the orbital separations ( normalized by @xmath7 ) of the planets correspond to @xmath35 and @xmath36 , respectively . to estimate the probability , we first construct maps of magnification excess .
the magnification excess is defined by @xmath37 where @xmath38 and @xmath39 represent the magnifications expected with and without the planets , respectively .
figure 1 shows the constructed map when the projected separations of the individual planets are @xmath40 and @xmath41 and the angle between the position vectors to the planets fro the host star ( orientation angle ) is @xmath42 .
we note that the projected separation , @xmath43 , is related to the intrinsic orbital separation , @xmath44 , by @xmath45 where @xmath46 is the phase angle and @xmath47 is the inclination of the orbital plane .
once the map is constructed , the probability is estimated by computing the ratio of events whose source trajectories pass both of the outer deviation regions induced by the individual planets ( for example , the event resulting from the source trajectory designated by ` 1 ' in fig .
1 ) during the measurements to the total number of trial events . for given lens positions ,
the source trajectory orientations ( with respect to the jupiter - primary axis ) and the impact parameter ( with respect to the primary ) of the trial events are randomly selected in the ranges of @xmath48 and @xmath49 , respectively .
measurements for each event are assumed to be carried out during @xmath50 with a frequency of 10 times / night , which corresponds to that of the current microlensing followup observations @xcite .
we consider each planet is detected if excesses greater than a threshold value of @xmath51 are consecutively detected more than 5 times during the measurements .
since we are interested in anomalies occurred only in the outer deviation regions , we do not count the detections of anomalies occurred in the central deviations region ( the region enclosed by the solid circle around the primary with a radius @xmath52 ) . in addition , since identifying the existence of both planets and determining their parameters will be difficult if both of the planets outer deviation regions are located at a similar place due to the resulting complexity of the interfered pattern of anomalies , we also do not count detections if the time interval between the anomalies induced by the individual anomalies is shorter than @xmath53 .
6.8 cm in figure 2 , we present the distribution of the _ absolute _ probabilities of detecting both planets as functions of their projected separations , @xmath54 .
the presented probabilities are the values averaged over the random orientation angles in the range of @xmath55 . in figure 3
, we also present the distribution of _ conditional _ probabilities to successively detect the second planet under the condition that the first planet is detected , @xmath56 .
one finds that if the two planets are located in the lensing zone , detecting both of them is possible with absolute probabilities of @xmath57 .
one also finds that once a planet is detected , the probabilities to successively detect the second one are @xmath58 .
we propose a new channel of detecting extra - solar systems composed of having multiple planets by using microlensing . in this method ,
multiple planets are detected when the source trajectory passes both of the outer regions of deviations induced by the individual planets . from the estimation of the probabilities to detect both planets of an example galactic lens system composed of two jovian - mass planets around a star with @xmath30
, we find that if they are located within the lensing zone , both planets can be detected with a non - negligible absolute probabilities ( @xmath59 ) and a substantial conditional probabilities ( @xmath60 ) of successively detecting the second planet under the condition that the first planet is detected .
the proposed method has an important advantage of allowing one to accurately determine the planet parameters because the light curve anomalies induced by the individual planets can be well described by simple single planet lensing models . | microlensing is one of the promising techniques that can be used to search for extra - solar systems .
planet detection via microlensing is possible because the event caused by a lens system having a planet can produce noticeable anomalies in the lensing light curve if the source star passes close to the deviation region induced by the planet .
gaudi , naber & sackett pointed out that if an event is caused by a lens system containing more than two planets , all planets will affect the central region of the magnification pattern , and thus the existence of the multiple planets can be inferred by detecting additionally deformed anomalies from intensive monitoring of high magnification events .
unfortunately , this method has important limitations in identifying the existence of multiple planets and determining their parameters ( the mass ratio and the instantaneous projected separation ) due to the degeneracy of the resulting light curve anomalies from those induced by a single planet and the complexity of multiple planet lensing models . in this paper
, we propose a new channel to search for multiple planets via microlensing .
the method is based on the finding of han et al . that the lensing light curve anomalies induced by multiple planets are well approximated by the superposition of those of the single planet systems where the individual planet - primary pairs act as independent lens systems .
then , if the source trajectory passes both of the outer deviation regions induced by the individual planets , one can unambiguously identify the existence of the multiple planets .
we illustrate that the probability of successively detecting light curve anomalies induced by two jovian - mass planets located in the lensing zone through this channel will be substantial .
since the individual anomalies can be well modeled by much simpler single planet lensing models , the proposed method has an important advantage of allowing one to accurately determine the parameters of the individual planets . |
in the past years and decades , several models of nova shells have been presented in the literature .
often they were adapted to describe the state and evolution of specific objects , and often remarkable agreement between model and observation was achieved .
nevertheless it should be kept in mind that a nova shell is a rapidly evolving object , and its properties change significantly with time .
furthermore , a plethora of different types of novae are observed , which is accompanied by an amazing variety of nova shells of various morphologies and physical properties in different stages of temporal development .
although studies of nova shells have been carried out since the first bright nova of the 20th century , gk persei in 1901 , most of these studies were carried out in a qualitative way .
this approach permitted the calculation of nebular expansion parallaxes and the morphological study of shells . since the shells were usually faint , and the observations were carried out with photographic plates , hardly any quantitative results are available . only in the first phases of the outburst ,
when the shells document themselves in the form of emission lines , were the line fluxes estimated and derived for a few cases , notably by payne - gaposchkin and collaborators .
replacement of the photographic plate by digital receivers has facilitated the task of studying the evolution of nova remnants , both spectroscopically and by means of direct imaging through narrow - band filters .
in fact , quite a number of studies have even been carried out for extragalactic novae , where h@xmath0-images can more easily detect the objects above the stellar background ( see , e.g. ciardullo et al .
1987 ) . in this paper
, we report on the results of a recent imaging survey of nova remnants , carried out at the kitt peak and eso la silla observatories .
we also use a hitherto unpublished survey of nova shells carried out in 1984 at calar alto , and the images from the _ hubble space telescope _ archive .
furthermore , we have collected and homogenized the existing quantitative record of nova shell observations . because the survey attempted to cover as many objects in as many evolutionary stages as possible , hardly any detailed information on a given object , or any detailed modelling of shells will be given ( i.e. the distribution of line flux between various specific parts of a nova shell ) .
we rather attempt to describe the `` average '' or global evolutionary track of a nova shell , in order to derive expected values for faint shells of ancient novae
. a theoretical interpretation of the observed behavior will be the subject of a forthcoming paper ( duerbeck & downes 2002 ) .
section 2 describes our observations and reductions .
section 3 briefly describes the classification of novae according to speed class , which is the base for merging our shell luminosity data into groups .
section 4 gives the derivation of global trends in luminosity evolution for the lines h@xmath0 , h@xmath1 and [ o iii ] @xmath25007 in novae of different speed classes ( including , besides classical novae , recurrent ones ) .
section 5 summarizes our results .
old data of nova shell line fluxes , derived both from spectral observations or direct images , were collected from the literature .
besides many data scattered in the literature , the early photographic studies of payne - gaposchkin and collaborators deserve special mentioning , as well as the recent tololo nova survey , carried out by williams and collaborators , and kindly put at our disposal by him .
the new observations were obtained at the european southern observatory , la silla , chile , and at the the kitt peak national observatory . on 1998 march 21 23 ,
the dutch 0.9 m telescope at eso , equipped with a tek tk512cb chip ( @xmath3 pixels ) with a scale of @xmath4 pixel@xmath5 was used . on 1998 may 28 june 1 , observations were obtained with the kpno 2.1 m telescope using the tek `` t1ka '' chip ( @xmath6 pixels with a scale of @xmath7 pixel@xmath5 ) , and on 1998 june 30 with the kpno 0.9 m telescope using the tek `` t2ka '' chip ( @xmath8 pixels with a scale of @xmath9 pixel@xmath5 ) .
a final run was carried out at the danish 1.54 m telescope at eso on 2000 july 16 .
the dfosc was used , which has a loral / lesser chip ( @xmath10 pixels with a scale of @xmath11 pixel@xmath5 ) .
the data were obtained with narrow - band filters centered at h@xmath0 ( 80 and 62 fwhm at the eso dutch and danish , 36 at kpno ) and [ o iii ] @xmath25007 ( 55 and 57 at the eso dutch and danish , 31 at kpno ) , as well as off - band and _ ubvr _ filters ; see downes & duerbeck ( 2000 ) for details .
note that the offband [ o iii ] @xmath25007 filter for the objects observed at eso is a strmgren @xmath12 filter .
the data were reduced in the standard manner .
flux calibration of the novae were obtained via `` standard '' planetary nebulae .
ngc 6833 was used for the kpno observations .
the h@xmath0 and @xmath13 $ ] flux of ngc 6833 was determined by tony keyes ( private communication ) based on _ hubble space telescope _
faint object spectrograph observations , and we adopted values of @xmath14 erg @xmath15 s@xmath5 arcsec@xmath16 and @xmath17 erg @xmath15 s@xmath5 arcsec@xmath16 for h@xmath0 and @xmath13 $ ] , respectively . for the `` dutch '' observations , sp 1 was used .
the h@xmath0 and @xmath13 $ ] flux of sp 1 was measured by perinotto et al .
( 1994 ) and the authors of the strasbourg / eso catalog of galactic planetary nebulae ( acker et al . 1992 ) spectroscopically , and by webster ( 1969 ) , copetti ( 1990 ) and shaw & kaler ( 1989 ) through interference filters .
unfortunately , the results show some scatter , so we assume fluxes of @xmath18 erg @xmath15 s@xmath5 arcsec@xmath16 and @xmath19 erg @xmath15 s@xmath5 arcsec@xmath16 for h@xmath0 and @xmath13 $ ] , respectively .
for the `` danish '' observations , three objects from the list of dopita & hua ( 1997 ) were used : pn 327.5 + 13.3 , pn [email protected] , and pn [email protected] . on the second night of the eso `` dutch '' observations ,
the flux of sp 1 was fainter by @xmath21 and @xmath22 in h@xmath0 and @xmath13 $ ] , an effect that was traced in the magnitudes of stars measured in these filters as well .
thus , we assume that the overall transmission , and not the central wavelength , underwent a change .
broadband transmission was very similar in the three nights .
we thus took the nightly aperture magnitude of the pn , corrected to airmass 1 , as the `` zero - point '' of the system , which corresponds to the fluxes given .
the targets were partly starlike , and normal daophot photometry was carried out for most of the nebulae . in the @xmath13 $ ] data , v842 cen and v1974 cyg
were slightly resolved , and rr pic and cp pup were clearly resolved ( downes & duerbeck 2000 ) . in these cases ,
the onband and offband frames were aligned , the offband frame scaled appropriately to cancel the majority of stars , and subtracted from the onband frame . in most cases , a starlike center remained ( because of emission in the region of @xmath13 $ ] ) , and this central emission was again fitted with a psf - profile of the onband frame , yielding an emission - line magnitude of the central star .
the detached nebular shell emission was measured in an annulus , where the central `` remnant '' that remained from the subtraction and fitting attempts , was duly neglected .
more objects were slightly resolved in h@xmath0 , and a similar process was performed .
for the starlike images , the nova magnitudes measured from the images are a combination of the line flux from the shell plus a continuum flux from the stellar remnant .
this continuum flux needs to be removed if we are to study the h@xmath0 and [ o iii ] @xmath25007 line fluxes themselves . to correct for the continuum flux
, a fit was made to all objects ( excluding the nova ) in the field of the form @xmath23 these fits allowed us to estimate the flux in the continuum of the novae , and thus to derive net line fluxes . for the eso observations ,
the use of the broad - band strmgren @xmath12 filter as the [ o iii ] off - band resulted in a more involved continuum removal .
we note that this filter has practically the same central wavelength as a broadband johnson @xmath24 filter , without , however , including strong emission lines like the [ o iii ] lines . for all field stars ,
@xmath24 magnitudes in the standard system are available .
the magnitude difference for the @xmath24- and the ( uncalibrated ) @xmath12-observations was determined , by establishing the relation @xmath25 i.e. by determining the zero - point difference @xmath26 .
furthermore , for all stars ( excluding the nova ) the relation @xmath27 was established and @xmath28 determined .
thus the magnitude difference between the @xmath24 magnitude and the [ o iii]-magnitude of an object without emission lines is @xmath29 field stars of all colours scatter from these linear relations by @xmath30 to @xmath31 , including m - stars with tio bands .
thus it is likely that unusually blue stars ( like nova star continua in their late stages ) also will not deviate significantly . under this assumption , the presumably emission - free @xmath12 magnitude of the nova was used to calculate its hypothetical `` emission - free '' @xmath24-magnitude , and from this magnitude , its `` emission - free '' [ o iii]-magnitude was derived .
the flux corresponding to this magnitude was subtracted from the flux calculated from the observed [ o iii]-magnitude , yielding an `` emission - only '' [ o iii ] flux . in most cases ,
the correction due to stellar contamination was below 10% of the total flux .
we also evaluated direct images of nova shells , taken in august 1984 through h@xmath0 and [ o iii ] @xmath25007 filters , with a ccd camera in the cassegrain focus of the calar alto 2.2 m telescope ( operated by the centro astronomico hispano - aleman , almeria , and the max - planck institut fr astronomie , heidelberg ; observers were h.w .
duerbeck and w.c .
seitter ) . only resolved shells were considered , the central stars ( and others in the field ) were reduced with the daophot allstar routine within iraf , and aperture photometry was carried out using midas .
the planetary nebula m57 ( ngc 6720 ) served as a flux standard .
furthermore , we used flux - calibrated spectroscopic observations of several novae , taken ( by us ) at several telescopes in the past 25 years .
determination of the [ o iii ] @xmath25007 and h@xmath0 fluxes was straightforward , since the shells were usually unresolved .
we were also given access to the extensive data set obtained in the tololo nova survey ( cf .
williams et al .
1994 ) , from which direct line flux measurements were made .
in addition to the ground - based data , we used archival [ o iii ] @xmath25007 and h@xmath0 images obtained with the _ hubble space telescope_. finally , we collected shell fluxes from the literature , which were derived from direct images or spectroscopy of nova shells .
to be able to compare line strengths from different objects , we converted the observed fluxes to luminosities ; the distances and reddenings necessary for the conversion were mostly taken from downes & duerbeck ( 2000 ) ; a few additional determinations or re - determinations are discussed in the appendix of this paper .
the list of objects , with adopted distances , reddenings , and references to these data , as well as references for the flux measurements , is given in table 1 .
the complete set of data are given in table 2 , table 3 and table 4 for [ o iii ] @xmath25007 , h@xmath0 , and h@xmath1 , respectively .
in order to merge sparse observations of many objects into a single diagram , a reasonable grouping of objects seems desirable .
since the photospheric growth and shrinking is controlled by the mass loss in the course of the outburst , and the photospheric shrinking is reflected in the light curve , we have found it appropriate to sort the data according to nova speed class .
the speed classes adopted were those defined in payne - gaposchkin ( 1957 ) : very fast , fast , moderately fast , slow and very slow , depending on the average decline rate over the first two magnitudes ( @xmath32 ) .
her classification supersedes previous schemes by mclaughlin ( 1939 ) ( fast , average , slow , rt ser - type ) and bertaud ( 1948 ) ( fast , slow , very slow ) , which were usually based on @xmath33 times .
however , payne - gaposchkin s classification can lead to serious errors in some slow novae which show a pronounced maximum of somewhat more than two magnitudes , superimposed on a plateau of almost constant brightness ( e.g. rr pic ) . in a few cases ,
we have deliberately modified the assignment to group together similar objects .
similarly , a few reassignments were made to better fit the shell flux data of a single object to the general trend prevailing in a neighboring speed class .
such reassignments are noted below .
in addition to the speed classes , we have also tried to discriminate between the spectral classes fe- and he / n novae according to the criteria given by williams ( 1992 ) ; a source of this information is della valle & livio ( 1998 ) . finally , confirmed and probable onemg novae were identified , either from recent work in the literature , or from earlier descriptions ( for an early identification of such objects , see mclaughlin ( 1944 ) ) . the objects for which shell flux data were obtained are listed in table 1 , together with decline times @xmath34 , speed class , spectral class , and peculiarity ( i.e. onemg group , recurrent nova , or nova with noticeable dust formation ) .
the nova distances and reddenings of table 1 and the observed nova fluxes of tables 2 , 3 and 4 were used to calculate line luminosities , which are also given in tables 2 , 3 and 4 .
depending on the distance , reddening and type of nova , the luminosity of the stellar remnant , as well as the light gathering and resolving power of the telescope used for the study , there is a lower limit for the line luminosity that can still be recorded with certainty . in general , we have assumed a luminosity of @xmath35 erg s@xmath5 as the lower limit that was achieved with the telescopes used in this study .
this should be regarded as an averaged a posteriori value . to derive a correct estimate of the lower limit
, one would not only have to take into account the distance , but also source confusion in the case of an extended shell , and the contribution of accretion disk emission in the case of a pointlike object . in the following
, the nova is said to have `` switched off '' emission in a specific line if its luminosity was below the value @xmath35 erg s@xmath5 .
the behavior of very fast novae is illustrated in figs .
1 3 . in these diagrams ,
the logarithm of the luminosity ( measured in ergs @xmath36 ) is plotted versus logarithm of time after maximum ( time measured in years ) .
ordinary galactic novae are shown as filled circles , and relations are determined exclusively for this galactic dataset , unless stated otherwise .
extragalactic novae are shown as open circles , and are merely displayed for comparison .
unusual novae ( mentioned in the text and the figure caption ) are shown with special symbols .
the h@xmath0 evolution of very fast novae appears to be quite homogeneous , unlike the evolution of the other spectral lines considered below .
several objects , however , have to be considered separately .
one is m31-c31 , which is similar in evolution than the other novae , but systematically fainter at earlier stages ( see below ) .
the long series of measurements of v838 her ( indicated as asterisks in figs .
13 ) shows that it is much fainter and faster in its evolution than other very fast novae , except v4160 sgr , which is marked with encircled asterisks .
many of the famous bright novae of the 20th century ( v603 aql , gk per and cp lac ) are not present in this diagram , since no spectrophotometry was obtained longward of the photographic ( blue ) region .
the general evolution can be approximated by straight lines . at times
@xmath37 ( @xmath38 days ) , the novae have @xmath39 or an almost constant luminosity , @xmath40 at times @xmath41 ( 10 days to 1 year ) @xmath42 and at late times ( @xmath43 ) : @xmath44 the fit was restricted to @xmath45 , since at very late times emission from the accretion disk may contaminate the measurements .
a few points were omitted , which belong to single , poorly - observed objects .
the data points of v838 her lie all on a straight line which can be approximated by @xmath46 the class of ` superbright novae ' ( v1500 cyg , n lmc 1991 ) by della valle ( 1991 ) is not obvious in the present diagrams . if there is anything peculiar , it is the _ faintness _ of m31-c31 .
since della valle used the m31 novae as a template ( and m31-c31 appears to fit well into this group ) , it appears that _ all _ very fast galactic novae fall into the region of the superbright novae , a fact also clearly seen in the s - calibration of downes & duerbeck ( 2000 ) , as applied to galactic novae .
why there is such a marked discrepancy in the observed behaviour of very fast novae in the galaxy and in m31 remains to be explained . at late times
( typically @xmath47 , i.e. 50 years after outburst ) , fast novae still show h@xmath0 emission at a level of a few @xmath48 .
two objects , however , are noticeably brighter : cp pup and gk per .
it is obvious that the emission does not originate from the accretion disk , since the shell is spatially resolved .
the h@xmath1 flux shows a large scatter ( this is possibly due to the fact that the old novae v603 aql , gk per , cp lac etc .
are now included ) .
it appears as if there are two well - defined bands , an upper and a lower one , with only a few points in between .
the extragalactic ( lmc , m31 ) novae seem to cluster near the lower band . as in the previous diagram ,
v838 her is notably fainter than the rest . as in the case of h@xmath0 , between @xmath49 and @xmath50 ( i.e. 1 day and 15 days after maximum ) , flux levels
are almost constant at @xmath51 and @xmath52 ( @xmath53 ergs @xmath36 ) ; afterwards , a decline sets in . at times @xmath54 @xmath55 at times @xmath41
, a dichotomy seems indicated .
the upper line is @xmath56 and the lower one @xmath57 at late times @xmath58 , the slope is only poorly documented , and at times @xmath59 , emission from the disk may dominate .
the slope at late times is : @xmath60 the time interval @xmath61 can also be fitted with one straight line : @xmath62 finally , the nova v838 her follows its own relation : @xmath63 the x - ray turnoff time of such objects is typically of the order one to several hundred days according to vablandingham et al .
( 2001 ) , so that the break in h@xmath1 ( and [ o iii ] ) around 1 year may have to do with the switching off of nuclear burning on the central object . at late stages
( @xmath64 of the order of 50 years ) , objects cluster at some @xmath65 .
no late h@xmath1 observations exist for gk per and cp pup .
generally , [ o iii ] @xmath25007 appears around age @xmath66 ( about a month after optical maximum ) , and remains within a factor 10 constant until @xmath67 ( i.e. one year after outburst ) . as in the case of h@xmath1 ,
two well - expressed bands seem to be present , with the extragalactic novae populating the lower one .
after two years , the [ o iii ] @xmath25007 emission declines dramatically and for most objects is not recorded any more after 10 years .
there are two noticeable exceptions : cp pup and gk per , which show emission at almost @xmath68 50 85 years after outburst . however , the luminosity from cp pup is due to n ii 5005 , as was found spectroscopically by williams ( 1982 ) , while the emission in gk per is indeed due to [ o iii ] @xmath25007 ( bode et al .
1988 ) .
the dichotomy that appeared in the h@xmath1 luminosities is again clearly seen . a fit through the points with @xmath69 yields @xmath70 for the upper line , and @xmath71 for the lower one .
subsequently , both bands merge and a drop @xmath72 is observed .
the last two data points , which belong to cp pup and gk per , were neglected in this fit . for the remaining novae
, a luminosity of @xmath65 is reached at @xmath73 , i.e. 11 years after outburst .
the maximum phase appears to reveal the presence of two groups , which show almost the same temporal behavior , but are clearly separated by a factor @xmath74 in emission line luminosity , with an amazingly low scatter . among the objects in the bright group are v476 cyg , gk per , v603 aql and v977 sco ; the faint group includes cp lac , v1500 cyg , v4157 sgr , and v351 pup .
interestingly , both groups contain the same mixture of fe ii / he - n novae , co / onemg novae , so that there is no obvious parameter that influences the [ o iii ] luminosity .
a clear outlier is v838 her , whose [ o iii ] @xmath25007 luminosity is about 100 times fainter than all other objects , except v4160 sgr .
the [ o iii ] @xmath25007 lines emerge earlier than in any other object .
a fit through the points of v838 her yields : @xmath75 the strong late emission of h@xmath0 and [ o iii ] @xmath25007 in gk per should be pointed out .
gk per is a peculiar nova shell which seems to interact with interstellar material or a fossil planetary nebula ( seaquist et al .
thus the ejected material shows shock interaction with the stationary circumstellar material , and remains in a hot state .
the noticeable emission in the [ o iii ] @xmath25007 band of cp pup is due to n ii .
cp pup has , however , also an unusually strong h@xmath0 luminosity .
smits ( 1991 ) has argued for a small distance ( between 525 and 850 pc ) for cp pup , which would lower the luminosity by factors 10 and 4 , respectively .
the intricate geometry ( see downes & duerbeck ( 2000 ) for a detailed discussion ) makes these values less likely , although they would bring the h@xmath0 luminosity closer to the average value found in other very fast novae of this age .
the peculiar case of v838 her has already been mentioned .
this was the nova with the shortest @xmath32 and @xmath33 times of the light curve decay .
v838 her appears to have less mass ejection , and thus a thin photosphere , which made the central object and its light changes appear after an unusually short time ( leibowitz et al . 1992 ) .
it is important to note that the poorly observed nova v4160 sgr represents a second object of this rare class of rapidly fading neon novae with he - spectral characteristics .
the three spectra available for v4160 sgr ( williams 2000 ) resemble those of v838 her ( vanlandingham et al .
1996 ) at similar phases . the trends in the evolution of line luminosities in the class of fast novae ( figs .
4 6 ) differ somewhat from those of very fast novae .
the scatter between individual objects in this group is less pronounced .
there is only one extragalactic nova ( n lmc 1988 - 1 ) in the sample , which will be considered at the end .
h@xmath0 emission is first recorded at @xmath76 ( 9 days after maximum ) , and slowly declines till @xmath77 ( 115 days after maximum ) by a factor 5 .
after that , a steeper decline sets in . at times
@xmath78 ( 115 days ) : @xmath79 with an average value @xmath80 , and at later times @xmath81 h@xmath1 emission is first recorded at @xmath82 ( 7 days after maximum ) , and slowly declines till @xmath77 ( 115 days after maximum ) by a factor 3 .
after that , a steeper decline sets in .
the extragalactic nova lies noticeably above the other objects .
if it would be placed among the very fast novae ( and practically all other lmc novae are very fast novae ) , it would fit wery well into the corresponding diagrams .
such a case occurs from time to time : an object assigned to a group because of its light curve better fits into another group according to its line luminosity evolution .
we will discuss the few cases in detail after the general discussion . at times
@xmath78 ( 115 days ) : @xmath83 with an average value @xmath84 , and at later times @xmath85 in fast novae , [ o iii ] @xmath25007 shows up at @xmath86 ( 46 days after maximum ) .
it displays a continuous decline in strength ( a factor of about 50 ) until a breakpoint that can only be approximately determined because of the large scatter .
the breakpoint can be put at @xmath87 as well as at @xmath88 , i.e. 1 to 3.2 years after outburst .
we will assume @xmath89 as the position of the breakpoint .
after that , a steeper decline sets in . at early stages @xmath90 , the average @xmath91 , a fit through these points with yields @xmath92 subsequently , a drop @xmath93 is observed .
this means that , on the average , a luminosity of @xmath94 is reached at @xmath95 , i.e. 32 years after outburst .
the x - ray turnoff time of onemg novae of the fast speed class like qu vul and v1974 cyg is of the order @xmath96 years according to vanlandingham et al .
this might coincide with the time when the [ o iii ] @xmath25007 emission starts to fade noticeably .
some moderately fast novae have been caught in very early stages , with fairly strong or weak h@xmath0 emission lines , but the sample is too small to draw any conclusions . between @xmath97 and @xmath98 ( 12 days and 115 days ) , a fairly well - expressed maximum is reached . afterwards , a decline with a slope of @xmath99 sets in . at times @xmath78 , @xmath100 with an average value @xmath101 , and at later times @xmath102 figures 7 9 show two objects with special symbols .
open circles denote data for the slow nova pw vul , which fits quite well into the group of moderately fast novae .
asterisks denote fluxes of the peculiar nova gq mus . as in the case of h@xmath0 , in h@xmath1 the time between @xmath97 and @xmath98 ( 12 days and 115 days )
is marked by a large scatter , but no obvious general decline is noticeable .
afterwards , the decline in luminosity is obvious , however also marked by large scatter .
the slope after 4 months is @xmath103 , slightly shallower than for fast and very fast novae . at times between @xmath104 and @xmath105 , @xmath106 with an average value @xmath107 , and at later times @xmath108 again , the luminosities of the slow nova pw vul emulate those of moderately fast novae , while the peculiar x - ray active nova gq mus shows a much more rapid decline than other novae . for moderately fast novae , the [ o iii ]
@xmath25007 data are somewhat sparse .
[ o iii ] @xmath25007 shows up after @xmath109 ( 145 days ) , and persists with fairly constant strength until @xmath110 ( 2.5 years ) .
afterwards , a quite dramatic decline sets in . between @xmath111 , the average @xmath112 , while a fit through the points with @xmath113 , excluding the novae gq mus and pw vul and the late value of dq her , yields @xmath114 on the average
, a luminosity of @xmath94 is reached at @xmath115 , i.e. 24 years after outburst .
the similarity of the evolution of the slow nova pw vul with the majority of moderately fast novae has already been pointed out .
it is also interesting to note that the decline in [ o iii ] @xmath25007 is very similar in fast and moderately fast novae ( although [ o iii ] @xmath25007 appears much earlier in fast novae ) .
an important point is the unusually bright [ o iii ] @xmath25007 luminosity determined for dq her 50 years after outburst .
the [ o iii ] @xmath25007 filter exposure has certainly picked up emission of the line n ii 5005 , which was observed spectroscopically by williams et al .
( 1978 ) at the endpoints of the major axis of the shell .
indeed , the direct image of 1984 shows a deficit of emission along the minor axis .
flux data are only available for a handful of novae , making general statements quite uncertain .
the main string of data ( see figs . 10 12 ) , is from m31-c32 , a well - monitored extragalactic very slow nova .
a few data points from v723 cas fall remarkably close to this object .
thus it seems that the luminosity discrepancy between novae in m31 and the galaxy occurs only among very fast novae .
note that some novae , like rr pic , hr del and v723 cas have an extended pre - maximum halt .
because of the logarithmic plot , emission line fluxes are plotted only after maximum light .
the number of studied objects is somewhat scarce , and the behavior of slow novae is quite varying , so these fits are not as representative as those for the other speed classes .
the luminosity remains essentially constant between @xmath116 and @xmath117 ( @xmath118 days after maximum ) , and declines afterwards . at times
@xmath119 , @xmath120 with an average value @xmath121 , and at later times @xmath122 h@xmath0 emission from the shell is still observed in very old remnants .
here we have a long data series of rr pic , from the presumable maximum onward , plus data from hr del and v868 cen . between @xmath123 and @xmath117 ( 3 days 460 days after maximum ) , the behavior of the h@xmath1 flux declines very slowly by a factor 4 .
after 460 days , the slope becomes steeper . at times
@xmath119 , @xmath124 with an average value @xmath125 , and at later times @xmath126 h@xmath1 emission from the shell is still observed in very old remnants . in slow novae , [ o iii ]
@xmath25007 emission appears at time @xmath127 , about 230 days after maximum .
rr pic and hr del show almost constant [ o iii ] @xmath25007 flux between the first appearance at 0.6 years up to @xmath88 , 3 years after maximum .
the average @xmath128 , a fit through the points with @xmath129 yields @xmath130 in the following century , a drop @xmath131 is observed .
note that the decline from maximum is documented only by data of hr del .
a few very old remnants , rr pic , do aql and x ser still show [ o iii ] @xmath25007 emission , which is , in two cases , documented by spectroscopy ( and thus confusion with n ii , as in cp pup and dq her , can be ruled out ) .
such a behavior is at variance with all other types of classical novae , and indicates that a source of high energy photons , presumably from the nuclear burning in the outer layers of the white dwarf , is still active @xmath132 years after outburst .
most notable are the [ o iii ] @xmath25007-bright shells around old slow novae .
at least three distinct groups of recurrent novae can be discriminated : very fast novae with dwarf companions , very fast novae with giant companions , and the slow recurrent nova t pyx .
the most extensive data set exists for the second group , and it is obvious that the other groups behave differently : the fast recurrent novae with dwarf secondaries show an extremely rapid decline of emission line strength in the balmer lines , and [ o iii ] @xmath25007 is basically absent ( see figs .
13 15 ) .
the slow recurrent nova t pyx shows , more than thirty years after its most recent outburst , persistently strong h@xmath0 and [ o iii ] @xmath25007 emission , likely because of shock interaction of ejecta from different outbursts . in the interval @xmath133 , recurrent novae with
giant companions have @xmath134 recurrent novae with dwarf companions show a more rapid decrease in flux : @xmath135 at times @xmath133 , recurrent novae with giant companions show @xmath136 recurrent novae with dwarf companions show a more rapid decrease in flux : @xmath137 no [ o iii ] @xmath25007 flux measurements in recurrent novae with dwarf companions are known ; these features are either very weak or do not appear . early weak
[ o iii ] @xmath25007 emission , caused by photoionization of the giant wind by the energetic radiation from the early outburst , has been reported for v745 sco . noticeable emission in [ o iii ]
@xmath25007 starts to appear at @xmath138 , reaches a maximum at @xmath77 , and declines afterwards .
a fit is uncertain for lack of secure data , but the average flux at times before one year is @xmath139 .
t pyx is still bright in [ o iii ] @xmath25007 30 years after the last outburst : interaction with previous shells keeps the ejecta hot , which appears to be untypical for other , faster recurrent novae .
thus we refrain from merging data of very different objects , as we have also considered rn with dwarf and giant companions separately .
we have collected about 1200 available line fluxes of 96 classical and recurrent novae of various speed classes and have studied the evolution of the luminosity in the h@xmath0 , h@xmath1 , and [ o iii ] @xmath25007 line as a function of time after outburst . in general , novae of a given speed class follow similar patterns , so that functional relations for the average evolution of line luminosity could be derived .
general trends for novae of various speed classes are shown in figs .
15 , 16 and 17 . a few novae turn out to be unusual : gk per and t pyx , which interact with circumstellar material , v838 her and v4160 sgr , which have unusually small mass ejection , and gq mus , an x - ray emitting classical nova .
a general discussion of the material presented here in the framework of nova properties and shell evolution will be presented elsewhere ( duerbeck & downes 2002 ) .
the data of tables 1 4 are also included as ascii files in table1.dat , table2.dat , table3.dat and table4.dat .
* do aql * - this is a very interesting slow nova ; unfortunately the light curve has a large seasonal gap , and the true maximum may have been missed .
the reddening is low even for large distances , @xmath140 .
application of the mmrd yields a distance @xmath141 kpc ; if do aql has a luminosity that is similar to the faintest nova in the mmrd sample , a minimum distance @xmath142 kpc follows .
a distance @xmath143 is plausible .
* v365 car * - the light curve and its parameters ( liller & henize 1975 ) indicate a slow nova for which an average absolute magnitude @xmath144 is assumed .
since the light curve is photographic , a @xmath145 correction of 0.25 was applied . the galactic extinction program ( hakkila et al .
1997 ) yields an @xmath146 for a distance of 4.7 kpc , which yields the expected apparent magnitude . * v868 cen * - the nova until now had a poor light curve coverage . there were also discrepant reports on dust formation ( smith et al . 1995 ; harrison et al .
the photometric observations published in the iau circulars were supplemented by an extended series of visual estimates ( jones 2001 ) .
they lead us : ( 1 ) to reclassify v868 cen as a medium fast nova , ( 2 ) to time the dust formation at the epoch of the ir observations of harrison et al .
( 1998 ) ( but certainly later than the ir spectroscopy of smith et al .
( 1995 ) , which however , did not develop into a dramatic visual minimum , but a short depression , similar to those observed in nq vul ) , and ( 3 ) to derive a consistent distance and reddening .
* v888 cen * - light curve data were taken from the iau circulars , and @xmath147 , @xmath148 days was estimated . using the mmrd , a consistent distance and reddening ( @xmath149 pc , @xmath150 ) was found . * by cir
* - light curve data were taken from the iau circulars , and @xmath147 , @xmath151 days was estimated . using the mmrd , a consistent distance and reddening ( @xmath152 pc , @xmath153 ) was found . *
v2295 oph * - light curve data were taken from the iau circulars , and @xmath154 , was estimated .
the light curve seems to indicate a plateau , followed by dust formation , and a light curve type c was assumed , which is typical for novae with @xmath155 .
a consistent distance and reddening ( @xmath156 kpc , @xmath157 ) was found .
* v4361 sgr * - light curve data were taken from the iau circulars , and @xmath158 , was estimated .
the nova curve is fragmentary , but the nova appears to evolve slowly , thus @xmath159 was assumed .
a consistent distance and reddening ( @xmath160 pc , @xmath161 ) was found .
* v4633 sgr * - light curve data were taken from liller & jones ( 1999 ) , and @xmath162 , @xmath163 days was estimated . using the mmrd , an @xmath164 was estimated , and a consistent distance and reddening ( @xmath165 pc , @xmath166 ) was found . * v4642 sgr * - light curve data were taken from the iau circulars .
the very fragmentary light curve indicates a nova which is not fast ; maximum @xmath167 .
the absolute magnitude was assumed as @xmath168 . a consistent distance and reddening ( @xmath169 pc , @xmath170 ) was found .
* v1141 sco * - light curve data were taken from the iau circulars .
the light curve indicates a very fast nova with @xmath171 , the mmrd yields an absolute magnitude of @xmath172 .
a consistent distance and reddening ( @xmath173 pc , @xmath174 ) was found .
* v1142 sco * - light curve data were taken from liller & jones ( 1999 ) .
the light curve indicates a very fast nova with @xmath175 , maximum light was assumed to be @xmath176 .
the mmrd yields an absolute magnitude of @xmath177 .
a consistent distance and reddening ( @xmath178 pc , @xmath179 ) was found .
we thank bob williams ( stsci ) for providing the data for the tololo nova survey objects , and albert jones for providing his series of visual observations of v868 cen .
we also thank the referee , a. jorissen , for his comments .
this research has made use of nasa s astrophysics data system bibliographic services , as well as of the simbad database , operated at cds , strasbourg , france .
this project was supported by the flemish ministry for foreign policy , european affairs , science and technology .
acker , a. , marcout , j. , ochsenbein , f. , stenholm , b. , and tylenda , r. 1992 , strasbourg - eso catalogue of galactic planetary nebulae ( eso , germany ) anderson , c.m . and
gallagher , j.s .
1977 , pasp , 89 , 264 andre , j. , drechsel , h. , and starrfield , s. 1994 , a&a , 291 , 869 anupama , g.c . ,
duerbeck , h.w . ,
prabhu , t.p . , and jain , s.k .
1992 , a&a , 263 , 87 anupama , g.c . and dewangan , g.c .
2000 , aj , 119 , 1359 anupama , g.c . and sethi , s. 1994 , mnras , 269 , 105 bahng , j.d.r . , 1972 , mnras , 158 , 151 bertaud , c. 1948 , annales dastrophysique , 11 , 1 bode , m.f . , duerbeck , h.w . , seitter , w.c . , albinson , j.s . , and evans , a. 1988 , in a decade of uv astronomy with the iue satellite , esa - sp 281 , vol . 1 , 183 bohigas , j. , echevarra , j. , diego , f. , and sarmiento , j.a .
1989 , mnras , 238 , 1395 ciardullo , r. , ford , h.c . ,
neill , j.d . ,
jacoby , g.h . , and shafter , a.w .
1987 , apj , 318 , 520 ciardullo , r. , shafter , a.w . ,
ford , h.c . ,
neill , j.d . ,
shara , m.m . and
tomaney , a.b .
1990 , apj , 356 , 472 ciatti , f. and rosino , l. 1974 , a&as , 16 , 305 copetti , m.v.s .
1990 , pasp , 102 , 77 de freitas pacheco , j.a .
, da costa , r.d.d . , and codina , s.j .
1989 , apj , 347 , 483 della valle , m. 1991 , a&a , 252 , l9 della valle , m. and duerbeck , h.w .
1993 , a&a , 275 , 239 della valle , m. , gilmozzi , r. , bianchini , a. , and esenoglu , h. 1997 , a&a , 325 , 1151 della valle , m. and livio , m. 1998 , apj , 506 , 818 diaz , m.p . , williams , r.e . , phillips , m.m . , and hamuy , m. 1995 , mnras , 277 , 959 dopita , m.a . , and hua , c.t . 1997 , apjs , 108 , 515 downes , r.a . and duerbeck , h.w . 2000 ,
aj , 120 , 2007 institute , erlangen - nuremberg university .
duerbeck , h.w . , and downes , r.a .
2002 , in preparation ferland , g.j . ,
lambert , d.l . , and woodman , j.h .
1986 , apjs , 60 , 375 ferland , g.j . , williams , r.e . ,
lambert , d.l . ,
slovak , m. , gondhalekar , p.m. , truran , j. w. , and shields , g. a. 1984 , apj , 281 , 194 gehrz , r.d . ,
jones , t.j . , woodward , c.e . ,
greenhouse , m.a . ,
wagner , r.m . ,
harrison , t.e . ,
hayward , t.l . , and benson , j. 1992 , apj , 400 , 671 gilmozzi , r. , selvelli , p. , and cassatella , a. 1998 , ultraviolet astrophysics beyond the iue final archive , esa sp-413 , eds .
w. wamsteker and r. gonzlez riestra , p. 415
gonzlez - riestra , r. 1992 , a&a , 265 , 71 hachisu , i. and kato , m. 2000a , apj , 540 , 447 hachisu , i. and kato , m. 2000b , apj , 536 , l93 hachisu , i. , kato , m. , kato , t. , matsumoto , k. 2000 , apj , 234 , l189 hakkila , j. , myers , j.m . ,
stidham , b.j . , and hartmann , d.h .
1997 , aj , 114 , 2043 hang , h. , zhu , z. , and liu , q. 1999 , acta astron .
sinica 40 , 247 harrison , t.e . , johnson , j.j . ,
mason , p.a . , and stringfellow , g.s .
1998 , in asp conf .
137 , wild stars in the old west , ed .
s. howell , e. kuulkers , and chick woodward ( san francisco : asp ) , 489 hauschildt , p.h . ,
starrfield , s. , shore , s.n .
, gonalez - riestra , r. , sonneborn , g. , and allard , f. 1994 , aj , 108 , 1008 honeycutt , r.k . ,
robertson , j.w . , and turner , g.w .
1995 , apj , 446 , 838 jones , a. 2001 , private communication kaluny , j. and chlebowski , t. 1988 , apj , 332 , 287 kamath , u.s . ,
anupama , g.c . ,
ashok , n.m . , and chandrasekhar , t. 1997 , aj , 114 , 2671 kawabata , k.s . ,
hirata , r. , ikeda , y. , akitaya , h. , seki , m. , matsumura , m. , and okazaki , a. 2000 , apj , 540 , 429 larsson - leander , g. 1954 , stockholm obs .
18 , no . 4 leibowitz , e.m . , mendelson , h. , mashal , e. , prialnik , d. , and seitter , w.c .
1992 , apj , 385 , l49 liller , w. and henize , k.g . 1975 , apj , 200 , 694 liller , w. and jones , a. 1999 , ibvs , 4664 liu , w. , and hu , j.y .
2000 , apjs , 128 , 387 lynch , d.k . ,
rossano , g.s . , rudy , r.j . , and puetter , r.c .
1995 , aj , 110 , 227 lynch , d.k . , rudy , r.j . ,
rossano , g.s . ,
erwin , p. , and puetter , r.c .
1989 , aj , 98 , 1682 matheson , t. , filippenko , a.v . , and ho , l.c .
1993 , apj , 418 , l29 mclaughlin , d.b . 1939 , pop .
astr . , 47 , 410 , 481 , 538 mclaughlin , d.b .
1944 , aj , 51 , 20 mclaughlin , d.b .
1953 , apj , 117 , 279 morisset , c. and pquignot , d. 1996 , a&a , 312 , 135 munari , u. , goranskij , v.p . ,
popova , a.a .
1996 , a&a , 315 , 166 munari , u. , yudin , b.f . ,
kolotilov , e.a . ,
shenavrin , v.i . ,
sostero , g. , and lepardo , a. 1994 , a&a , 284 , l9 mustel , e.r . and
boyarchuk , a.a .
1970 , ap&ss , 6 , 183 payne - gaposchkin , c. 1957 , the galactic novae ( interscience publishers , inc .
: new york ) payne - gaposchkin , c. and menzel , d.h .
1938 , harvard college circ .
428 payne - gaposchkin , c. and gaposchkin , s. 1942 , harvard college circ .
445 pquignot , d. , petitjean , p. , boisson , c. , and krautter , j. 1993 , a&a , 271 , 219 perinotto , m. , purgathofer , a. , pasquali , a. , and patriarchi , p. 1994
, a&as , 107 , 481 popper , d.m .
1940 , apj , 92 , 262 rafanelli , p. rosino , l. , and radovich , m. 1995 , a&a , 294 , 488 retter , a. , leibowitz , e.m . , and naylor , t. 1999 , mnras , 308 , 140 ringwald , f.a .
naylor , t. and mukai , k. 1996 , mnras , 281 , 192 rosino , l. , iijima , t. , benetti , s. , dambrosio , v. , di paolantonio , a. , and kolotilov , e.e .
1992 , a&a , 257 , 603 saizar , p. , pachoulakis , i. , shore , s.n . ,
starrfield , s. , williams , r.e . ,
rothschild , e. , and sonneborn , g. , 1996 , mnras , 279 , 280 saizar , p. , starrfield , s. , ferland , g.j . ,
wagner , r.m . ,
truran , j.w . ,
kenyon , s. j. , sparks , w. m. , williams , r. e. , and stryker , l. l. 1991 , apj , 367 , 310 saizar , p. , starrfield , s. , ferland , g.j . , wagner , r.m . ,
truran , j.w . ,
kenyon , s.j .
, sparks , w.m . , williams , r.e . , and stryker , l.l .
1992 , apj , 398 , 651 schwarz , g.j . ,
hauschildt , p.h . ,
starrfield , s. , baron , e. , allard , f. , shore , s.n . , and sonneborn , g. 1997 , mnras , 284 , 669 scott , a.d . , evans , a. , and rawlings , j.j.m .
1994 , mnras , 269 , l21 scott , a.d . ,
duerbeck , h.w . , evans , a. , chen , a .-
, de martino , d. , hjellming , r. , krautter , j. , laney , d. , parker , q.a . , rawlings , j.m.c . , and van winckel , h. 1995 , a&a , 296 , 439 seaquist , e.r . ,
bode , m.f . ,
frail , d.a . ,
roberts , j.a . , evans , a. , and albinson , j.s .
1989 , apj , 344 , 805 sekiguchi , k. , whitelock , p.a . , feast , m.w . , barrett , p.e . ,
caldwell , j.a.r . ,
carter , b.s . , catchpole , r.m . , laing , j.d .
, laney , c.d . ,
marang , f. , and van wyk , f. 1990a , mnras , 246 , 78 sekiguchi , k. , stobie , r.s . , buckley , d.a.h . , and caldwell , j.a.r .
1990b , mnras , 245 , 28p shafter , a.w .
1997 , apj , 487 , 226 shaw , r.a . and kaler , j.b . 1989 , apjs , 69 , 495 shore , s.n .
sonneborn , george ; starrfield , sumner g. ; hamuy , m. ; williams , r. e. ; cassatella , a. ; drechsel , h.et al .
1991 , apj , 370 , 193 smits , d.p .
1991 , mnras , 248 , 217 smith , c.h . ,
aitken , d.k . , roche , p.f . , and wright , c.m .
1995 , mnras , 277 , 259 snijders , m.a.j . ,
batt , t.j . ,
seaton , m.j . , blades , j.c . , and morton , d.c .
1984 , mnras , 211 , 7p snijders , m.a.j . ,
batt , t.j . ,
roche , p.f . ,
seaton , m.j .
, morton , d.c . ,
spoelstra , t.a.t . , and blades , j.c .
1987 , mnras , 228 , 329 somers , m.w . and naylor , t. 1999 , a&a , 352 , 563 szkody , p. and howell , s.b .
1992 , apjs , 78 , 537 thorstensen , j.r . and
taylor , c.j .
2000 , mnras , 312 , 629 tolbert , c.r . ,
pecker , j.c . , and pottasch , s.r .
1967 , bull .
netherlands , 19 , 17 tomaney , a.b . and shafter , r.w .
1993 , apj , 411 , 640 tylenda , r. 1977 , acta astr .
, 27 , 389 vanlandingham , k.m , starrfield , s. , wagner , r.m . , shore , s.n . , and sonneborn , g. 1996 , mnras , 282 , 563 vanlandingham , k.m , schwarz , g.j . , shore , s.n . , and starrfield , s. 2001 , aj , 121 , 1126 webster , b.l .
1969 , mnras , 143 , 79 weiler , e.j . and bahng , j.d.r .
1976 , mnras , 174 , 563 whipple , f.h . and payne - gaposchkin , c. 1937 , harvard college circ . 414
whipple , f.h . and payne - gaposchkin , c. 1939 , harvard college circ .
433 whitney , b.a . and
clayton , g.c .
1989 , aj , 98 , 297 williams , r.e . 1982 , apj , 261 , 170 williams , r.e .
1992 , aj , 104 , 725 williams , r.e .
1994 , apj , 426 , 279 williams , r.e .
2000 , private communication williams , r.e . ,
woolf , n.j . , hege , e.k . ,
moore , r.l . , and kopriva , d.a .
1978 , apj , 224 , 171 williams , r.e . ,
phillips , m.m . , and hamuy , m. 1994 , apjs , 90 , 297 zhen - xi , z. and heng - rong , h. 2000 chin .
a&a , 24 , 71 zwitter , t. and munari , u. 1995 , a&as , 114 , 575 zwitter , t. and munari , u. 1996 , a&as , 117 , 449 | the evolution of the luminosity of nova shells in the century following the nova outburst is studied for the lines h@xmath0 , h@xmath1 , and [ o iii ] @xmath25007 .
about 1200 flux measurements from 96 objects have been collected from the literature , from unpublished observations , from the hst archive , or from new narrow - band filter imaging . for most objects ,
the distance and reddening is known ( or newly determined ) , and luminosities were calculated from the observed fluxes .
the luminosity data were combined in five groups , according to nova light curve type ( very fast , fast , moderately fast , slow , recurrent ) ; some objects were re - assigned to other groups for a better fit of the luminosity data to the general trend . for very fast , fast and moderately fast novae ,
the slope of the [ o iii ] @xmath25007 decline is very similar , leading to a basic ` switchoff ' of [ o iii ] @xmath25007 emission after 11 , 23 and 24 years , respectively .
for the same speed classes , the slope of the balmer luminosity is quite similar .
in contrast to all types of fast novae , the decline in balmer luminosity is more rapid in slow novae .
however , the slope in [ o iii ] @xmath25007 is more gentle ; slow novae still show [ o iii ] @xmath25007 emission after 100 years .
thus shells of slow novae are still hot after one century ; the same applies for the shells of the very fast nova gk per and the recurrent nova t pyx , which interact with circumstellar material . in recurrent novae , [ o iii ]
@xmath25007 is usually inconspicuous or absent . in objects with giant companions ,
the balmer luminosity decreases very slowly after an outburst , which may be an effect of line blending of material from the ejecta and the giant wind . on the other hand , objects with dwarf companions show a very rapid decline in balmer luminosity .
* keywords : * novae : shells , novae : decline , cataclysmic variables |
the goal of this paper is to establish ( graded ) affine cellularity in the sense of koenig and xi @xcite for the khovanov - lauda - rouquier algebras @xmath0 of finite lie type .
in fact , we construct a chain of affine cell ideals in @xmath0 which are generated by idempotents .
this stronger property is analogous to quasi - heredity for finite dimensional algebras , and by a general result of koenig and xi ( * ? ? ?
* theorem 4.4 ) , it also implies finiteness of the global dimension of @xmath0 .
thus we obtain a new proof of ( a slightly stronger version of ) a recent result of kato @xcite and mcnamara @xcite ( see also @xcite ) . as another application
, one gets a theory of standard and proper standard modules , cf .
it would be interesting to apply this paper to prove the conjectural ( graded ) cellularity of cyclotomic klr algebras of finite types .
our approach is independent of the homological results in @xcite , @xcite and @xcite ( which relies on @xcite ) .
the connection between the theory developed in @xcite and this paper is explained in @xcite .
this paper generalizes @xcite , where analogous results were obtained for finite type @xmath1 .
we now give a definition of ( graded ) affine cellular algebra from ( * ? ? ?
* definition 2.1 ) . throughout the paper , unless otherwise stated , we assume that all algebras are ( @xmath2)-graded , all ideals , subspaces , etc .
are homogeneous , and all homomorphisms are homogeneous degree zero homomorphisms with respect to the given gradings . for this introduction
, we fix a noetherian domain @xmath3 ( later on it will be sufficient to work with @xmath4 ) .
let @xmath1 be a ( graded ) unital @xmath3-algebra with a @xmath3-anti - involution @xmath5 .
a ( two - sided ) ideal @xmath6 in @xmath1 is called an _ affine cell ideal _ if the following conditions are satisfied : 1 .
@xmath7 ; 2 .
there exists an affine @xmath3-algebra @xmath8 with a @xmath3-involution @xmath9 and a free @xmath3-module @xmath10 of finite rank such that @xmath11 has an @xmath1-@xmath8-bimodule structure , with the right @xmath8-module structure induced by the regular right @xmath8-module structure on @xmath8 ; 3 .
let @xmath12 be the @xmath8-@xmath1-bimodule with left @xmath8-module structure induced by the regular left @xmath8-module structure on @xmath8 and right @xmath1-module structure defined by @xmath13 where @xmath14 ; then there is an @xmath1-@xmath1-bimodule isomorphism @xmath15 , such that the following diagram commutes : + @xmath16 \ar^{\tau}[d ] & & \delta \otimes_b\delta ' \ar^{v \otimes b \otimes b ' \otimes w \mapsto w \otimes \si(b ' ) \otimes \si(b ) \otimes v } [ d ] \\
j \ar^-{\mu}[rr]&&\delta \otimes_b\delta'.}\ ] ] the algebra @xmath1 is called _ affine cellular _ if there is a @xmath3-module decomposition @xmath17 with @xmath18 for @xmath19 , such that , setting @xmath20 , we obtain an ideal filtration @xmath21 so that each @xmath22 is an affine cell ideal of @xmath23 . to describe our main results
we introduce some notation referring the reader to the main body of the paper for details .
fix a cartan datum of finite type , and denote by @xmath24 the set of positive roots , and by @xmath25 the positive part of the root lattice .
for @xmath26 we have the klr algebra @xmath0 with standard idempotents @xmath27 .
we denote by @xmath28 the set of root partitions of @xmath29 .
this is partially ordered with respect to a certain bilexicographic order ` @xmath30 ' .
to any @xmath31 one associates a proper standard module @xmath32 and a word @xmath33 .
we fix a distinguished vector @xmath34 , and choose a set @xmath35 so that @xmath36 is a basis of @xmath32 .
we define polynomial subalgebras @xmath37these are isomorphic to tensor products of algebras of symmetric polynomials .
we also explicitly define elements @xmath38 and set @xmath39
. then we set @xmath40 @xmath41 , and @xmath42 .
our main results are now as follows * main theorem .
* _ the algebra @xmath0 is graded affine cellular with cell chain given by the ideals @xmath43 . moreover , for a fixed @xmath44 , we set @xmath45 and @xmath46 for any @xmath47
. we have : _ 1 . @xmath48 ; 2 .
@xmath49 is an idempotent in @xmath50 ; 3 .
the map @xmath51 is an isomorphism of graded algebras ; 4 .
@xmath52 is a free right @xmath53-module with basis @xmath54 ; 5 .
@xmath55 is a free left @xmath53-module with basis @xmath56 ; 6 .
multiplication provides an isomorphism @xmath57 7 .
@xmath58 .
main theorem(vii ) shows that each affine cell ideal @xmath59 in @xmath60 is generated by an idempotent .
this , together with the fact that each algebra @xmath61 is a polynomial algebra , is enough to invoke ( * ? ? ?
* theorem 4.4 ) to get * corollary . *
_ if the ground ring @xmath3 has finite global dimension , then the algebra @xmath0 has finite global dimension . _
this seems to be a slight generalization of @xcite,@xcite,@xcite in two ways : @xcite assumes that @xmath3 is a field of characteristic zero ( and the lie type is simply - laced ) , while @xcite,@xcite assume that @xmath3 is a field ; moreover , @xcite,@xcite,@xcite deal with categories of graded modules only , while our corollary holds for the algebra @xmath0 even as an ungraded algebra .
the paper is organized as follows .
section 2 contains preliminaries needed for the rest of the paper .
the first subsection contains mostly general conventions that will be used .
subsection [ slie ] goes over the lie theoretic notation that we employ .
we move on in subsection [ ssklr ] to the definition and basic results of khovanov - lauda - rouquier ( klr ) algebras .
the next two subsections are devoted to recalling results about the representation theory of klr algebras .
then , in subsection [ ssqg ] , we introduce our notation regarding quantum groups , and recall some well - known basis theorems .
the next subsection is devoted to the connection between klr algebras and quantum groups , namely the categorification theorems .
finally , subsection [ ssdf ] contains an easy direct proof of a graded dimension formula for the klr algebras , cf .
* corollary 3.15 ) .
section 3 is devoted to constructing a basis for the klr algebras that is amenable to checking affine cellularity .
we begin in subsection [ sssswid ] by choosing some special weight idempotents and proving some properties they enjoy .
subsection [ snota ] introduces the notation that allows us to define our affine cellular structure .
this subsection also contains the crucial hypothesis [ hprop ] .
next , in subsection [ sspower ] , we come up with an affine cellular basis in the special case corresponding to a root partition of that is a power of a single root .
finally , we use this in the last subsection to come up with our affine cellular basis in full generality . in section 4
we show how the affine cellular basis is used to prove that the klr algebras are affine cellular . finally , in section 5
we verify hypothesis [ hprop ] for all positive roots in all finite types .
we begin in subsection [ sshomog ] by recalling some results concerning homogeneous representations . in subsection [ ssslo ]
we recall the definition of special lyndon orders and lyndon words , which will serve as the special weights of subsection [ sssswid ] .
the next subsection is devoted to verifying hypothesis [ hprop ] in the special case when the cuspidal representation corresponding to the positive root is homogeneous .
we then employ this in subsection [ ssade ] to show that the hypothesis holds in simply - laced types .
finally , we have subsection [ ssbcfg ] , wherein we verify the hypothesis by hand in the non - symmetric types .
in this section we set up the theory of klr algebras and their connection to quantum groups following mainly @xcite and also @xcite .
only subsection [ ssdf ] contains some new material . throughout the paper we work over the ground ring @xmath62 which is assumed to be either @xmath2 or an arbitrary field @xmath63 .
most of the time we work over @xmath63 and then deduce the corresponding result for @xmath2 using the following standard lemma [ pfieldtoz ] let @xmath64 be a finitely generated @xmath2-module , and @xmath65 a subset of @xmath64 . then @xmath66 is a spanning set ( resp . basis ) of @xmath64 if and only if @xmath67 is a spanning set ( resp .
basis ) of @xmath68 for every field @xmath63 .
let @xmath69 be an indeterminate , @xmath70 the field of rational functions , and @xmath71\subseteq \q(q)$ ] .
let @xmath72 be the @xmath73-algebra involution with @xmath74 , referred to as the _ bar - involution_. for a graded vector space @xmath75 , with finite dimensional graded components its _ graded dimension _ is @xmath76 $ ] . for any graded @xmath63-algebra @xmath77
we denote by @xmath78 the abelian category of all graded left @xmath77-modules , with morphisms being _ degree - preserving _ module homomorphisms , which we denote by @xmath79 .
let @xmath80 denote the abelian subcategory of all _ finite dimensional _ graded @xmath77-modules and @xmath81 denote the additive subcategory of all _ finitely generated projective _ graded @xmath77-modules .
denote the corresponding grothendieck groups by @xmath82 $ ] and @xmath83 $ ] , respectively .
these grothendieck groups are @xmath84-modules via @xmath85:=[m\langle m\rangle ] , $ ] where @xmath86 denotes the module obtained by shifting the grading up by @xmath87 : @xmath88 for @xmath89 , let @xmath90 denote the space of homomorphisms of degree @xmath91 .
set @xmath92 a _ cartan datum _ is a pair @xmath93 consisting of a set @xmath94 and a @xmath2-valued symmetric bilinear form @xmath95 on the free abelian group @xmath96 $ ] such that @xmath97 for all @xmath98 and @xmath99 for all @xmath100 in @xmath94 . set @xmath101 for @xmath102 and define the _ cartan matrix _ @xmath103 . throughout the paper ,
unless otherwise stated , we assume that @xmath1 has _ finite type _ , see @xcite .
we have simple roots @xmath104 , and we identify @xmath105 with @xmath106 .
let @xmath107 . for @xmath108 , we write @xmath109 for the sum of its coefficients when expanded in terms of the @xmath105 s .
denote by @xmath110 the set of _ positive _ roots , cf .
@xcite , and by @xmath111 the corresponding weyl group .
a total order on @xmath112 is called _ convex _ if @xmath113 and @xmath114 imply @xmath115 .
given @xmath116 $ ] , denote @xmath117_\beta:=(q_\beta^{n}-q_\beta^{-n})/(q_\beta - q_\beta^{-1}),\ [ n]^!_\beta:=[n]_\beta[n-1]_\beta\dots[1]_\beta.\end{aligned}\ ] ] in particular , for @xmath98 , we have @xmath118_i,[n]_i^!$ ] .
let @xmath1 be a @xmath25-graded @xmath70-algebra , @xmath119 for @xmath120 , and @xmath121 .
we use the standard notation for quantum divided powers : @xmath122_\al^!.$ ] denote by @xmath123 the set of all tuples @xmath124 of elements of @xmath94 , which we refer to as _ words_. we consider @xmath125 as a monoid under the concatenation product . if @xmath126 , we can write it in the form @xmath127 for @xmath128 such that @xmath129 for all @xmath130 . we then denote @xmath131!:=[m_1]^!_{j_1}\dots[m_r]^!_{j_r}.\ ] ] for @xmath124 set @xmath132 the symmetric group @xmath133 with simple transpositions @xmath134 acts on @xmath135 on the left by place permutations .
the @xmath133-orbits on @xmath135 are the sets @xmath136 parametrized by the elements @xmath108 of height @xmath137 .
let @xmath1 be a cartan matrix .
choose signs @xmath138 for all @xmath139 with @xmath140 so that @xmath141 , and define the polynomials @xmath142\mid i , j\in i\}$ ] : @xmath143 in addition , fix @xmath120 of height @xmath137 .
let @xmath144 be an associative graded unital @xmath62-algebra , given by the generators @xmath145 and the following relations for all @xmath146 and all admissible @xmath147 : @xmath148 @xmath149 @xmath150 @xmath151 @xmath152 @xmath153 @xmath154 @xmath155 the _ grading _ on @xmath0 is defined by setting : @xmath156 in this paper _ grading _ always means _ @xmath2-grading _ , ideals are assumed to be homogeneous , and modules are assumed graded , unless otherwise stated .
it is pointed out in @xcite and @xcite that up to isomorphism the graded @xmath62-algebra @xmath0 depends only on the cartan datum and @xmath29 .
we refer to the algebra @xmath0 as an _ ( affine ) khovanov - lauda - rouquier algebra_. it is convenient to consider the direct sum of algebras @xmath157 note that @xmath158 is non - unital , but it is locally unital since each @xmath0 is unital . the algebra @xmath159 possesses a graded anti - automorphism @xmath160 which is the identity on generators . for each element @xmath161
fix a reduced expression @xmath162 and set @xmath163 in general , @xmath164 depends on the choice of the reduced expression of @xmath165 .
[ tbasis ] ( * ? ? ?
* theorem 2.5 ) , ( * ? ? ?
* theorem 3.7 ) the following set is an @xmath62-basis of @xmath0 : @xmath166 in view of the theorem , we have a polynomial subalgebra @xmath167\subseteq r_\al.\ ] ] let @xmath168 be elements of @xmath25 with @xmath169 .
then we have a natural embedding @xmath170 of algebras , whose image is the _ parabolic subalgebra _ @xmath171 .
this is not a unital subalgebra , the image of the identity element of @xmath172 being @xmath173 an important special case is where @xmath174 is a multiple of a simple root , in which case we have that @xmath175 is the @xmath176 nilhecke algebra @xmath177 generated by @xmath178 subject to the relations @xmath179 the grading is so that @xmath180 and @xmath181 .
note that here the elements @xmath164 do not depend on a choice of reduced decompositions .
let @xmath182 be the longest element , and define the following elements of @xmath177 : @xmath183 it is known that @xmath184 and in particular @xmath185 is an idempotent , see for example @xcite .
the following is a special case of our main theorem for the case where @xmath174 , which will be used in its proof .
it is known that the center @xmath186 consists of the symmetric polynomials @xmath187^{{\mathfrak{s}}_d}$ ] .
[ tnilheckecellbasis ] ( * ? ? ?
* theorem 4.16 ) let @xmath188 be a @xmath62-basis of @xmath187^{{\mathfrak{s}}_d}$ ] and let @xmath189 be a basis of @xmath187 $ ] as an @xmath187^{{\mathfrak{s}}_d}$]-module . then @xmath190 is a @xmath62-basis of @xmath177 . by @xcite , every irreducible graded @xmath0-module is finite dimensional , and there are finitely many irreducible @xmath0-modules up to isomorphism and grading shift . for @xmath191 and @xmath192 ,
the _ @xmath193-word space _ of @xmath64 is @xmath194 we have a decomposition of ( graded ) vector spaces @xmath195 we say that @xmath193 is a _ word of @xmath64 _ if @xmath196 .
we identify in a natural way : @xmath197=\bigoplus_{\al\in q_+}[\mod{r_\al}],\quad [ \proj{r}]=\bigoplus_{\al\in q_+}[\proj{r_\al}].\ ] ] recall the anti - automorphism @xmath5 from ( [ star ] ) .
this allows us to introduce the left @xmath0-module structure on the graded dual of a finite dimensional @xmath0-module @xmath64the resulting left @xmath0-module is denoted @xmath198 . on the other hand ,
given any left @xmath0-module @xmath64 , denote by @xmath199 the right @xmath0-module with the action given by @xmath200 for @xmath201 . following ( * ? ? ?
* ( 14 ) ) , define the _ khovanov - lauda pairing _ to be the @xmath84-linear pairing @xmath202\times[\proj{r_\al } ] \rightarrow \laurent\cdot \prod_{i\in i}\prod_{a=1}^{m_i}\frac{1}{(1-q_i^{2a } ) } \ ] ] such that @xmath203,[q ] ) = \qdim(p^\tau\otimes_{r_\al } q)$ ] .
let @xmath204 .
recalling the isomorphism @xmath205 , consider the functors @xmath206 for @xmath207 and @xmath208 , we denote @xmath209 the functors of induction define products on the grothendieck groups @xmath210 $ ] and @xmath211 $ ] and the functors of restriction define coproducts on @xmath210 $ ] and @xmath211 $ ] .
these products and coproducts make @xmath210 $ ] and @xmath211 $ ] into twisted unital and counital bialgebras ( * ? ? ?
* proposition 3.2 ) .
let @xmath98 and @xmath212 . as explained in (
* @xmath2132.2 ) , the algebra @xmath214 has a representation on the polynomials @xmath215 $ ] such that each @xmath216 acts as multiplication by @xmath216 and each @xmath217 acts as the divided difference operator @xmath218 let @xmath219 denote this representation of @xmath220 viewed as a graded @xmath220-module with grading defined by @xmath221 by ( * ? ? ?
* @xmath2132.2 ) , the left regular @xmath220-module decomposes as @xmath222^!_i \cdot p(i^{(n)})$ ] .
in particular , @xmath219 is projective .
set @xmath223 where @xmath224 .
these functors induce @xmath84-linear maps on the corresponding grothendieck groups : @xmath225 \rightarrow [ \proj{r_{\alpha+n\al_i}}],\quad ( \theta_{i}^*)^{(n)}:[\mod{r_{\alpha+n\al_i } } ] \rightarrow [ \mod{r_{\alpha}}].\ ] ] standard module theory for @xmath0 has been developed in @xcite .
here we follow the most general approach of mcnamara @xcite . fix a reduced decomposition @xmath226 of the longest element @xmath227 .
this gives a convex total order on the positive roots @xmath228 with @xmath229 . to every positive root @xmath230 of the corresponding root system @xmath231 , one associates a _
_ @xmath232 .
this irreducible module is uniquely determined by the following property : if @xmath233 are non - zero elements such that @xmath234 and @xmath235 , then @xmath236 is a sum of positive roots less than @xmath237 and @xmath238 is a sum of positive roots greater than @xmath237 . a standard argument involving the mackey theorem from @xcite and convexity as in the proof of ( * ?
* lemma 2.11 ) , yields : [ lbkm ] let @xmath230 and @xmath239 .
all composition factors of @xmath240 are of the form @xmath241 .
let @xmath120 .
a tuple @xmath242 is called a _
root partition of @xmath29 _ if @xmath243 .
we also use the notation @xmath244 .
for example , if @xmath245 for @xmath230 , we have a root partition @xmath246 . denote by @xmath28 the set of all root partitions of @xmath29 .
this set has two total orders : @xmath247 and @xmath248 defined as follows : @xmath249 ( resp .
@xmath250 ) if there exists @xmath251 such that @xmath252 and @xmath253 for all @xmath254 ( resp .
@xmath255 ) .
finally , we have a _ bilexicographic partial order _ : @xmath256 the following lemma is implicit in @xcite ; see also ( * ? ? ?
* lemma 2.5 ) . [ lminrp ]
given any @xmath257 , we have @xmath258 . for a root partition @xmath259 as above , set @xmath260 , and define the corresponding _ proper standard module _
@xmath261 for @xmath244 , we denote @xmath262 [ tstand]@xcite for any convex order there exists a cuspidal system @xmath263 . moreover : 1 . for every @xmath31 , the proper standard module @xmath264 has irreducible head
; denote this irreducible module @xmath265 .
2 . @xmath266 is a complete and irredundant system of irreducible @xmath0-modules up to isomorphism .
3 . @xmath267 .
4 . @xmath268_q=1 $ ] , and @xmath269_q\neq 0 $ ] implies @xmath270 .
@xmath271 is irreducible for every @xmath272 and every @xmath273 .
@xmath274 implies @xmath275 , and @xmath276 .
note that the algebra @xmath277 is defined over @xmath2 , i.e. @xmath278 .
we will use the corresponding indices when we need to distinguish between modules defined over different rings .
the following result shows that cuspidal modules are also defined over @xmath2 : [ lcuspint]let @xmath230 , and @xmath279 be a non - zero homogeneous vector .
then @xmath280 is an @xmath281-invariant lattice such that @xmath282 as @xmath283-modules for any field @xmath63 .
note using degrees that @xmath284 is finitely generated over @xmath2 , hence it is a lattice in @xmath285 .
furthermore @xmath286 , whence by definition of the cuspidal modules , all composition factors of @xmath287 are of the form @xmath288 .
but there is always a multiplicity one composition factor in a reduction modulo @xmath289 of any irreducible module over a klr algebra , thank to ( * ? ? ?
* lemma 4.7 ) . following ( * ? ? ?
* section 1.2 ) , we define the algebra @xmath290 to be the free @xmath70-algebra with generators @xmath291 for @xmath292 ( our @xmath69 is lusztig s @xmath293 , in keeping with the conventions of @xcite )
. this algebra is @xmath25-graded by assigning the degree @xmath105 to @xmath291 for each @xmath292 , so that @xmath294 . if @xmath295 , we write @xmath296 . for @xmath297 , write @xmath298 .
then @xmath299 is a basis for @xmath300 . in particular , each
@xmath300 is finite dimensional .
consider the graded dual @xmath301 .
we consider words @xmath302 as elements of @xmath303 , so that @xmath304 . that is to say , @xmath305 is the basis of @xmath306 dual to the basis @xmath307 .
let @xmath308 be the @xmath309-subalgebra of @xmath290 generated by @xmath310 .
this algebra is @xmath25-graded by @xmath311 , where @xmath312 . given @xmath313 with @xmath314 for @xmath315 , denote @xmath316 .
then @xmath317 is an @xmath309-basis of @xmath318 .
we also define @xmath319 , and assign it the induced @xmath25-grading . for every @xmath26 ,
the @xmath309-module @xmath320 is free with basis @xmath321 ! \bi \mid
\bi \in \words_\al\}\ ] ] dual to ( [ embasisa ] ) .
there is a _ twisted _ multiplication on @xmath322 given by @xmath323 for homogeneous @xmath324 .
let @xmath325 be the algebra homomorphism determined by @xmath326 for all @xmath292 .
* proposition 1.2.3 ) there is a unique symmetric bilinear form @xmath327 on @xmath290 such that @xmath328 and @xmath329 where the bilinear form on @xmath322 is given by @xmath330 .
define @xmath331 to be the quotient of @xmath290 by the radical of @xmath332 .
denote the image of @xmath291 in @xmath331 by @xmath333 .
the @xmath25-grading on @xmath290 descends to a @xmath25-grading on @xmath331 with @xmath334 .
let @xmath335 be the @xmath309-subalgebra of @xmath331 generated by @xmath336 for @xmath337 .
this algebra is @xmath25-graded by @xmath338 . given @xmath313 with @xmath314 for @xmath315 , denote @xmath339 and @xmath340 we recall the definition of the _ pbw basis _ of @xmath335 from (
* part vi ) .
recall that a reduced decomposition @xmath226 yields a total order on the positive roots @xmath341 , with @xmath229 .
now , embed @xmath335 into the upper half of the full quantum group via @xmath342 and take the braid group generators @xmath343 from @xcite . for @xmath344 , we define @xmath345 for a sequence @xmath346 , we set @xmath347 and also define @xmath348 the next theorem now gives a pbw basis of @xmath349 .
[ tpbw ] the set @xmath350 is an @xmath309-basis of @xmath349 .
furthermore : @xmath351 this follows from corollary 41.1.4(b ) , propositions 41.1.7 , 38.2.3 , and lemma 1.4.4 of lusztig @xcite .
consider the graded dual @xmath352 .
the map @xmath353 gives @xmath354 the structure of an associative algebra .
let @xmath355 be the map dual to the quotient map @xmath356 .
set @xmath357 with the induced @xmath25-grading .
given @xmath292 , we denote by @xmath358 the dual map to the map @xmath359 .
then the divided power @xmath360 is dual to the map @xmath361 .
clearly @xmath362 stabilizes @xmath363 . for @xmath364 , define @xmath365 to be dual to @xmath366 .
we define @xmath367 for @xmath368 , and any sequence @xmath346 .
the next well - known result gives the _ dual pbw basis _ of @xmath363 .
[ tdualpbw ] the set @xmath369 is the @xmath309-basis of @xmath370 dual to the pbw basis of theorem [ tpbw ] .
it easily follows from the properties of the lusztig bilinear form , the definition of the product on @xmath354 and theorem [ tpbw ] that the linear functions @xmath371 and @xmath372 on @xmath331 are equal .
it remains to apply theorem [ tpbw ] one more time .
let @xmath373 , and @xmath374
. then @xmath375 , and , switching back to @xmath376 s , the pbw basis of @xmath377 is @xmath378 . using the defining properties of lusztig s bilinear form , one can easily check that @xmath379 , @xmath380 , and @xmath381 .
finally the dual basis is @xmath382 .
now we state the fundamental categorification theorem proved in @xcite , see also @xcite .
we denote by @xmath383 $ ] the class of the left regular representation of the trivial algebra @xmath384 .
[ klthm ] there is a unique @xmath84-linear isomorphism @xmath385 $ ] such that @xmath386 $ ] and @xmath387 for all @xmath388 , @xmath292 , and @xmath389 . under this isomorphism : * @xmath390 $ ] ; * the multiplication @xmath391 corresponds to the product on @xmath211 $ ] induced by the exact functor @xmath392 ; * for @xmath393 we have @xmath394 $ ] ; * for @xmath395 we have @xmath396 .
let @xmath64 be a finite dimensional graded @xmath0-module .
define the _
@xmath69-character _ of @xmath64 as follows : @xmath397 the @xmath69-character map @xmath398 factors through to give an @xmath399-linear map from the grothendieck group @xmath400\to { \prescript{\prime}{\a}{{\mathbf{f}}}}^*.\ ] ] we now state a dual result to theorem [ klthm ] , see ( * ? ? ?
* theorem 4.4 ) .
[ dklthm ] there is a unique @xmath84-linear isomorphism @xmath401\stackrel{\sim}{\rightarrow}{\prescript{}{\a}{{\mathbf{f}}}}^ * $ ] with the following properties : * @xmath402)= 1 $ ] ; * @xmath403 for all @xmath404,\ i \in i,\ n \geq 1 $ ] ; * the following triangle is commutative : @xmath405 } \arrow[2]{e , t}{\ga^ * } \arrow{ne , t}{\ch } \node{}\node{{\prescript{}{\a}{{\mathbf{f}}}}^ * } \arrow{nw , t}{\kappa } \end{pb - diagram}\ ] ] * @xmath406)={\prescript{}{\a}{{\mathbf{f}}}}^*_\al$ ] for all @xmath407 ; * under the isomorphism @xmath408 , the multiplication @xmath409 corresponds to the product on @xmath210 $ ] induced by @xmath392 ; we conclude with mcnamara s result on the categorification of the dual pbw - basis ( see also @xcite for simply laced lie types ) : [ lstdpbw ] for every @xmath410 we have @xmath411 ) = e_\pi^*$ ] . by (
* theorem 3.1(1 ) ) , we have @xmath412 ) = e_\be^*$ ] for all @xmath230 .
the general case then follows from theorem [ dklthm](5 ) and the definition ( [ estand ] ) of @xmath413 . in this section
we obtain a dimension formula for @xmath0 , which can be viewed as a combinatorial shadow of the affine quasi - hereditary structure on it .
the idea of the proof comes from ( * ? ? ?
* theorem 4.20 ) .
an independent but much less elementary proof can be found in ( * ? ? ?
* corollary 3.15 ) . recall the element @xmath414 from ( [ ethetai ] ) and the scalar @xmath415!\in\a$ ] from ( [ eifact ] ) .
we note that lemma [ lmonpbwgen ] and theorem [ tdimgen ] _ do not require the assumption that the cartan matrix @xmath1 is of finite type _ , adopted elsewhere in the paper .
[ lmonpbwgen ] let @xmath416 , and let @xmath417)\in{\prescript{}{\a}{{\mathbf{f}}}}^*_\al$ ] for @xmath418 .
assume that @xmath419 is an @xmath309-basis of @xmath420 .
let @xmath421 be the dual basis of @xmath349 .
then for every @xmath393 , we have @xmath422 ! } v_n.\ ] ] recall the map @xmath423 from ( [ eiota ] ) dual to the natural projection @xmath356 . by theorem [ dklthm ] we have for any @xmath424 : @xmath425))=\ch([v^n])=\sum_{\bi \in \words_\al}(\qdim v^n_\bi ) \bi
= \sum_{\bi \in \words_\al}\frac{\qdim v^n_\bi}{[\bi ] ! } [ \bi ] ! \bi.\end{aligned}\ ] ] recalling ( [ embasisa ] ) and ( [ edualmbasisa ] ) , @xmath426 and @xmath427 ! \bi \mid \bi \in \words_\al\}$ ] is a pair of dual bases in @xmath428 and @xmath429 .
so , using our expression for @xmath430 , we can now get by dualizing : @xmath431 ! } [ \bj ] !
\bj ( { \prescript{\prime}{}{\theta}}_{(\bi)})v_n \\ & = \sum_{n=1}^m \sum_{\bj \in \words_\al}\frac{\qdim v^n_\bj}{[\bj ] ! } \de_{\bi,\bj}v_n = \sum_{n=1}^m\frac{\qdim v^n_\bi}{[\bi ] ! } v_n,\end{aligned}\ ] ] as required .
[ tdimgen ] with the assumptions of lemma [ lmonpbwgen ] , for every @xmath432 , we have @xmath433 in particular , @xmath434 theorem [ klthm](3 ) shows that @xmath435 = \ga(\theta_\bi ) = \ga([\bi ] !
\theta_{(\bi)})$ ] . using the definitions and theorem [ klthm](4 )
, we have @xmath436 , [ r_\al e(\bj)])= ( [ \bi ] !
\theta_{(\bi ) } , [ \bj ] ! \theta_{(\bj)}).\end{aligned}\ ] ] now , by lemma [ lmonpbwgen ] we see that @xmath437 !
\theta_{(\bi ) } , [ \bj ] !
\theta_{(\bj ) } ) & = \big ( \sum_{n=1}^m(\qdim v^n_\bi ) v_n , \sum_{k=1}^n(\qdim v^k_\bj ) v_k \big ) , \ ] ] which implies the theorem .
recall the scalar @xmath438 from ( [ elpi ] ) , the module @xmath413 from ( [ estand ] ) , and pbw - basis elements @xmath439 from [ ssqg ] . [ tdim ] for every @xmath432 , we have @xmath440 in particular , @xmath441 by lemma [ lstdpbw ] , we have @xmath442 for all @xmath31 .
moreover , by theorem [ tdualpbw ] , @xmath443 and @xmath444 is a pair of dual bases in @xmath370 and @xmath349 .
finally , @xmath445 by theorem [ tpbw ] .
it remains to apply theorem [ tdimgen ] .
throughout this section we fix @xmath120 and a total order @xmath30 on the set @xmath28 of root partitions of @xmath29 , which refines the bilexicographic partial order ( [ ebilex ] ) .
recall from section [ sssmt ] that for each @xmath446 , we have a cuspidal module @xmath232 .
every irreducible @xmath0-module @xmath447 has a word space @xmath448 such that the lowest degree component of @xmath448 is one - dimensional , see for example ( * ? ? ?
* lemma 2.30 ) or ( * ? ? ?
* lemma 4.5 ) for two natural choices . from now on , for each @xmath230 we make an arbitrary choice of such word @xmath449 for the cuspidal module @xmath232 . for @xmath450 ,
define @xmath451 the sums being over @xmath452 .
we also consider the ( non - unital ) embedding of algebras : @xmath453 whose image is the parabolic subalgebra @xmath454 [ lallidem ] if a two - sided ideal @xmath6 of @xmath0 contains all idempotents @xmath455 with @xmath31 , then @xmath456 . if @xmath457 , let @xmath94 be a maximal left ideal containing @xmath6 .
then @xmath458 for some @xmath459
. then @xmath460 , which contradicts the assumption that @xmath461 .
this argument proves the lemma over any field , and then it also follows for @xmath2 . [ lbadwordsnew ] let @xmath31 and @xmath462 a homogeneous idempotent . if @xmath463 for all @xmath270 , then @xmath464 .
let @xmath94 be any maximal ( graded ) left ideal containing @xmath465 .
then @xmath466 for some @xmath452 such that @xmath270 .
indeed , if we had @xmath467 then by definition @xmath468 , and so @xmath469 , which is a contradiction . we have shown that @xmath470 is contained in every maximal left ideal containing @xmath465 . by a standard argument , explained in ( * ? ? ?
* lemma 5.8 ) , we conclude that @xmath464 .
[ lbadwords ] suppose that @xmath471 for some @xmath472 and @xmath364 .
let @xmath473 . if @xmath474 , then @xmath475 .
this follows from lemma [ lminrp ] together with proposition [ lbadwordsnew ] .
[ liotaimage ] let @xmath476 .
then @xmath477 . by lemma [ lallidem ]
, we have @xmath478 for all @xmath479 .
therefore the image of @xmath480 equals @xmath481 where the sum is over all @xmath482 .
fix @xmath483 for all @xmath479 .
if @xmath484 for every @xmath91 , then @xmath485 , and the corresponding term of ( [ eiotaimage ] ) is in @xmath486 by definition .
let us now assume that @xmath487 for some @xmath3 .
in view of lemma [ lminrp ] , we have @xmath488 . for any @xmath452 we have @xmath489 , and by theorem [ tstand](vi ) , if @xmath490 then @xmath491 .
furthermore , for @xmath492 we have , by theorem [ tstand](vi ) applied again @xmath493 which is zero since @xmath494 by theorem [ tstand](vi ) again .
we have shown that for all @xmath495 we have @xmath496 , and consequently @xmath497 .
applying lemma [ lbadwordsnew ] , we have that @xmath498 .
the following result will often allow us to reduce to the case of a smaller height .
[ lparaideal1new ] let @xmath499 , @xmath500 , and @xmath501 .
assume that @xmath502 is such that all idempotents from the set @xmath503 annihilate the irreducible modules @xmath504 for all @xmath270 .
then @xmath505 we may assume that @xmath506 since otherwise @xmath507 , and the result is clear .
by lemma [ lallidem ] , we have @xmath508 for all @xmath418 , and by definition , we have @xmath509 therefore the left hand side of ( [ eiotaofr ] ) equals @xmath510 the result now follows by applying lemma [ lbadwordsnew ] .
recall from lemma [ liotaimage ] that @xmath511 .
[ lparaideal2 ] let @xmath512 and @xmath251 .
then @xmath513 in particular , the composite map @xmath514 factors through the quotient @xmath515 .
apply proposition [ lparaideal1new ] with @xmath516 , @xmath517 , for @xmath518 , @xmath519 , and @xmath520 .
we have to prove that any @xmath521 annihilates all @xmath504 for @xmath270 .
we prove more , namely that @xmath470 annihilates @xmath522 for all @xmath523 . by theorem [ tstand](vi ) : @xmath524 which is zero since @xmath525 by theorem [ tstand](vi ) again .
[ lparaideal1 ] for @xmath272 and @xmath526 we have @xmath527 we apply proposition [ lparaideal1new ] with @xmath528 , @xmath529 , @xmath530 , @xmath531 , @xmath532 , @xmath533 and @xmath534 .
pick an idempotent @xmath535 .
since @xmath459 is the minimal element of @xmath536 , it suffices to prove that @xmath537 .
note that @xmath538 , so using lemma [ lbkm ] , we just need to show that @xmath539 .
but @xmath540 is zero , since @xmath541 by theorem [ tstand](vi ) .
repeated application of corollary [ lparaideal1 ] gives the following result .
[ conerootideal ] for @xmath272 and @xmath542 we have @xmath543 let @xmath237 be a fixed positive root of height @xmath137 .
recall that we have made a choice of @xmath449 so that in the word space @xmath544 of the cuspidal module , the lowest degree part is @xmath545-dimensional .
we fix its spanning vector @xmath546 defined over @xmath2 , see lemma [ lcuspint ] .
similarly , the highest degree part is spanned over @xmath2 by some @xmath547 .
we consider the element of the symmetric group @xmath548 @xmath549 which permutes the @xmath550th and the @xmath551st ` @xmath137-blocks ' .
now define @xmath552 moreover , for @xmath553 with a fixed reduced decomposition @xmath554 , define the elements @xmath555 in section [ sverif ] , we will explicitly define homogeneous elements @xmath556 and @xmath557 so that the following hypothesis is satisfied : [ hprop ] we have : 1 .
2 . @xmath559 and @xmath560 are @xmath5-invariant .
3 . @xmath561 and @xmath562 , 4 .
@xmath560 has degree @xmath563 and commutes with @xmath564 and @xmath565 , 5 .
the algebra @xmath566 is generated by @xmath567 . 6 .
@xmath568 . from now on until we verify it in section [ sverif ]
, we will work under the assumption that hypothesis [ hprop ] holds .
it turns out that this hypothesis is sufficient to construct affine cellular bases .
[ cprop ] @xmath569 this follows as in the proof of lemma [ lallidem ] using @xmath570 . using lemma [ lcuspint ]
, we can choose a set @xmath571 of elements defined over @xmath2 such that @xmath572 is an @xmath62-basis of @xmath573 .
fix @xmath574 and define the set @xmath575 and the element @xmath576 further , define the elements of @xmath577 @xmath578 where @xmath579 is the longest element .
it will be proved in corollary [ c030413_3 ] that @xmath580 generalizing part ( i ) of hypothesis [ hprop ] .
it is easy to see , as in ( * ? ? ?
* lemma 2.4 ) , that there is always a choice of a reduced decompositon of @xmath581 such that @xmath582 we have the algebras of polynomials and the symmetric polynomials : @xmath583\quad\text{and}\quad { \lambda}_{(\be^p)}={p}_{(\be^p)}^{{\mathfrak{s}}_p}\ ] ] while it is clear that the @xmath584 commute , we do not yet know that they are algebraically independent , but this will turn out to be the case . for
now , one can interpret @xmath585 as the algebra generated by the elementary symmetric functions in @xmath586 .
note using hypothesis [ hprop](iv ) that @xmath587 given @xmath26 of height @xmath137 and a root partition @xmath588 we define the parabolic subgroup @xmath589 and we denote by @xmath590 ( resp .
@xmath591 ) the set of minimal left coset representatives of @xmath592 ( resp .
@xmath593 ) in @xmath594 . set @xmath595 using the natural embedding of @xmath596 , we define the elements @xmath597 which belong to the word space corresponding to the words @xmath598 from definitions we have [ ldebasis]let @xmath31 .
then @xmath599 is a basis for @xmath32 .
define @xmath600 note by ( [ eupperboundp ] ) and ( [ elpi ] ) that @xmath601 choose also a homogeneous basis @xmath602 for @xmath61 .
the following lemma is a consequence of hypothesis [ hprop](ii),(vi ) and ( [ etauinv1 ] ) .
[ lddeltau ] we have @xmath603 and @xmath604 . throughout this subsection @xmath364 and
@xmath605 are fixed .
define @xmath606 , and @xmath607 .
define @xmath608 , and given @xmath609 write @xmath610 for its image in @xmath611 .
the following proposition is the main result of this subsection .
[ lpowerbasis ] we have that 1 . @xmath612 is an @xmath62-basis for @xmath613 .
@xmath614 is an @xmath62-basis for @xmath615 .
@xmath616 is an @xmath62-basis for @xmath611 .
4 . the elements @xmath617 are algebraically independent .
the proof of the proposition will occupy this subsection .
it goes by induction on @xmath618 . if @xmath237 a simple root , then @xmath619 is exactly the nil - hecke algebra , and we are done by theorem [ tnilheckecellbasis ] . for the rest of the section , we assume the proposition holds with @xmath620 whenever @xmath621 and @xmath622 and prove that it also holds for @xmath623 .
we shall also assume that @xmath624 is a field , and then use lemma [ pfieldtoz ] to lift to @xmath2-forms .
[ l050413_1 ] assume that @xmath625 .
then proposition [ lpowerbasis ] holds .
since @xmath232 is the unique simple module in @xmath626 and @xmath627 is one - dimensional by hypothesis [ hprop ] , it follows that @xmath628 is the projective cover of @xmath232 in @xmath626 under the map @xmath629 .
all composition factors of @xmath628 are isomorphic to @xmath232 .
therefore , lifting the basis @xmath630 of @xmath232 to @xmath628 we see that @xmath628 is spanned by @xmath631 by hypothesis [ hprop](v ) , @xmath632 is generated by @xmath633 .
thus @xmath634 analogously , @xmath635 is the projective cover of @xmath636 as right @xmath637-modules under the map @xmath638 .
as above , lifting the basis @xmath639 of @xmath636 to @xmath635 we see that @xmath640 therefore by lemma [ cprop ] and hypothesis [ hprop](iv ) , @xmath641 let @xmath642 . by definition and
* proposition 2.16 ) we have @xmath643 because @xmath644 .
the opposite inclusion follows from lemma [ liotaimage ] . for @xmath479 ,
define @xmath645 by part ( iii ) of the induction hypothesis , for @xmath479 , the image of @xmath646 in @xmath647 is a basis .
let @xmath648 by corollary [ lparaideal2 ] and definitions from section [ snota ] , @xmath649 and therefore @xmath650 by definition of @xmath651 we have @xmath652 using ( [ edimupper ] ) and the equality @xmath653 for all @xmath654 , we get @xmath655 by corollary [ tdim ] .
the inequalities are therefore equalities , and this implies that the spanning set @xmath656 of @xmath657 is a basis and @xmath658 for all @xmath459 . these yield ( iii ) and ( iv ) of proposition [ lpowerbasis ] in our special case @xmath625 . to show ( i ) and ( ii )
, we have already noted that the claimed bases span @xmath628 and @xmath635 , respectively .
we now apply part ( iii ) to see that they are linearly independent .
[ cfree ] we have 1 . @xmath659 is a polynomial algebra in the variable @xmath660 .
@xmath628 is a free right @xmath659-module with basis @xmath661 .
3 . @xmath635 is a free left @xmath659-module with basis @xmath662 . by the lemma
, we have proposition [ lpowerbasis ] for @xmath625 .
now , ( i ) follows from parts ( i ) and ( iv ) of the proposition .
the remaining statements follow from parts ( i ) and ( ii ) of the proposition .
[ cgrot ] in the grothendieck group , we have @xmath663 = [ l(\be)]/(1-q_\be^2).$ ] [ linduct ] up to a degree shift , @xmath664 . by corollary [ conerootideal ]
we have a map @xmath665 by frobenius reciprocity , we obtain a map @xmath666 we now show that @xmath667 .
it is enough to prove that @xmath668 for all @xmath669 .
since all composition factors of @xmath628 are isomorphic to @xmath232 , it follows that all composition factors of @xmath670 are isomorphic to @xmath671 . by theorem [ tstand ] , @xmath672 , which proves the claim .
since @xmath673 , we obtain a map @xmath674 the homomorphisms @xmath675 map the evident cyclic generators to each other , and so are inverse isomorphisms .
[ lmackeypsi ] there exists an endomorphism of @xmath676 which sends @xmath677 to @xmath678 .
apply the mackey theorem to @xmath679 .
we get a short exact sequence of @xmath680-modules @xmath681 where @xmath682 is a preimage of the standard generator of @xmath683 .
we now show that this is actually a sequence of @xmath684-modules .
it is sufficient to show that for any @xmath685 , we have that @xmath686 we show the first equality , the second being similar .
all composition factors of @xmath628 are isomorphic to @xmath232 , so all composition factors of @xmath676 are isomorphic to @xmath687 , and thus all composition factors of @xmath679 are isomorphic to @xmath688 .
theorem [ tstand ] now tells us that @xmath689 for all @xmath685 . by the projectivity of @xmath628 as @xmath637-module , the short exact sequence splits , giving the required endomorphism by frobenius reciprocity .
[ c030413_1 ] @xmath690 .
let @xmath691 be the endomorphism of @xmath676 constructed in lemma [ lmackeypsi ] , regarded as an endomorphism of @xmath692 by lemma [ linduct ] .
then @xmath693 as required .
[ c030413_2 ] we have @xmath694
. follows from ( [ edesi ] ) , corollary [ c030413_1 ] and hypothesis [ hprop ] .
[ lendbasis ] the set @xmath695 gives a linear basis of @xmath696 .
the elements above are linearly independent by lemmas [ linduct ] and [ lmackeypsi ] , and corollary [ cfree ] .
we use frobenius reciprocity , corollary [ cgrot ] , and ( * ? ? ?
* lemma 2.11 ) to see that @xmath697\\ & = [ \res_{\be,\dots,\be } l(\be)^{\circ p } : l(\be)^{\boxtimes p}]/(1-q_\be^2)^p\\ & = q_\be^{-\frac{1}{2}p(p-1 ) } [ p]_\be^ ! / ( 1-q_\be^2)^p.\end{aligned}\ ] ] by the formula for the poincar polynomial of @xmath698 , we have shown that @xmath699 showing that the proposed basis also spans .
the next two lemmas are proved using ideas that already appeared in the proofs of ( * ? ? ?
* lemmas ) .
[ lnhrel1 ] we have that @xmath700 we use lemma [ linduct ] to identify @xmath701 with @xmath670 .
it is enough to prove the first relation in the case @xmath702 .
the mackey theorem analysis in the proof of lemma [ lmackeypsi ] shows that , as a graded vector space @xmath703 the vector @xmath704 is of minimal degree , and thus @xmath705 is of minimal degree in @xmath706 .
the degree of @xmath707 is smaller by @xmath708 , so the vector is zero .
the second relation is clear from the definitions . to prove the third relation , it is sufficient to consider @xmath709
. let @xmath710 , and set @xmath711 .
using the defining relations of @xmath712 , we deduce that @xmath713 is an element of degree @xmath714 in @xmath715 , where @xmath716 denotes the bruhat order . by a mackey theorem analysis as in the proof of lemma [ lmackeypsi ]
, we see that @xmath717 the lowest degree of an element in @xmath718 is therefore @xmath719 , and the third relation is proved .
[ lnhrel2 ] there exists a unique choice of @xmath720 such that @xmath721 the first relation is clear from the definitions .
it is enough to prove the remaining relations for @xmath702 . using the defining relations of @xmath722 and a mackey theorem analysis as in the proof of lemma [ lmackeypsi ]
, we deduce that @xmath723 and the only vector of the correct degree is @xmath724 .
therefore ( working over @xmath2 ) we must have that @xmath725 for some @xmath726 .
similarly , we obtain @xmath727 for some @xmath728 .
we compute @xmath729 and since @xmath730 and @xmath730 are linearly independent by lemma [ lendbasis ] , we must have @xmath731 .
we now fix a prime @xmath289 and extend scalars to @xmath732 .
suppose that @xmath733 , so that @xmath734 define @xmath718 to be the submodule of @xmath676 generated by @xmath730 and @xmath735 .
the above equations show that the endomorphism defined by right multiplication by @xmath736 leaves @xmath718 invariant . on the other hand
, @xmath737 is irreducible .
since the endomorphism algebra of an irreducible module is one dimensional , we have a contradiction .
therefore @xmath738 when reduced modulo any prime , i.e. @xmath720 . [ c030413_3 ] the homomorphism from the nilhecke algebra @xmath739 determined by @xmath740 is an isomorphism . under this isomorphism
the idempotent @xmath741 is mapped onto @xmath742 . using lemmas [ lnhrel1 ] and [ lnhrel2 ]
, we see that the map exists . by lemma [ lendbasis ] ,
the map is an isomorphism .
the second statement now follows using corollary [ c030413_2 ] .
[ csymm ] given @xmath743 , @xmath744 commutes with @xmath745 , @xmath742 , and @xmath746 .
it follows directly from hypothesis [ hprop](iv ) and the definitions that @xmath747 commutes with every element of @xmath748 , and in particular with every element of the subalgebra @xmath749 .
denote by @xmath581 the longest element of @xmath698 . then by corollaries [ c030413_1 ] and [ c030413_3 ] @xmath750 any @xmath743 commutes with @xmath751 by hypothesis [ hprop](iv ) .
it is well known that the center of the nilhecke algebra @xmath739 is given by the symmetric functions @xmath752 . in particular , every element of @xmath752 commutes with @xmath753 .
let @xmath754 be such that @xmath755
. then @xmath756 .
this implies the claim .
we can now finish the proof of proposition [ lpowerbasis ] .
corollary [ c030413_3 ] provides an isomorphism @xmath757 under which the idempotent @xmath758 corresponds to right multiplication by @xmath742 .
but @xmath758 is a primitive idempotent , so the image @xmath759 of this endomorphism is an indecomposable projective @xmath50-module .
we may identify @xmath760 where the action of @xmath749 on @xmath761 is given by right multiplication which makes sense in view of corollary [ csymm ] . therefore @xmath762 is a projective cover in @xmath763 .
furthermore , since @xmath763 has only one irreducible module , every composition factor of @xmath613 is isomorphic to @xmath504 with an appropriate degree shift . we can lift the basis @xmath764 for @xmath504 to the set @xmath765 . using the basis @xmath766 for @xmath749
, we get a basis @xmath767 for @xmath613 .
similarly , @xmath768 is a projective cover in @xmath769 .
it is immediate that @xmath770 is a basis of @xmath771 .
lifting as above , we have that @xmath772 is a basis for @xmath615 . finally , applying the multiplication map and corollary [ csymm ] we have that @xmath773 spans @xmath611 .
therefore by induction , @xmath774 and comparing graded dimensions with corollary [ tdim ] as in the proof of lemma [ l050413_1 ] , this set is therefore a basis . in this section
we use the results of the previous subsections to obtain affine cellular bases of the klr algebras of finite type .
fix @xmath26 and @xmath588 .
define @xmath45 , and write @xmath775 for the image of an element @xmath776 .
we begin with some easy consequences of the previous section .
[ csymmgen ] we have 1 . given @xmath777 , @xmath744 commutes with @xmath778 , @xmath49 , and @xmath779 .
2 . up to a grading shift , @xmath780 .
the map @xmath781 sending @xmath782 to right multiplication by @xmath783 is an isomorphism of algebras .
the map @xmath784 sending @xmath782 to left multiplication by @xmath783 is an isomorphism of algebras .
claim ( i ) follows directly from corollary [ csymm ] and the definitions .
the proof of claim ( ii ) is similar to that of lemma [ linduct ] . to be precise ,
by corollary [ lparaideal2 ] we have a map @xmath785 which by frobenius reciprocity determines a homomorphism @xmath786 we now claim that @xmath787 .
it is enough to prove that @xmath788 for all @xmath467 . by exactness of induction
, it follows that @xmath789 has an exhaustive filtration by @xmath790 . by theorem [ tstand ] , @xmath791 , which proves the claim . since @xmath792 we obtain a map @xmath793 the homomorphisms @xmath675 map the evident cyclic generators to each other , and so are inverse isomorphisms .
we use claim ( ii ) to identify @xmath794 with @xmath795 .
as noted in the proof of claim ( ii ) , @xmath794 has an exhaustive filtration by @xmath796 by theorem [ tstand](vi ) , @xmath797 picks out the summand corresponding to @xmath798 .
therefore @xmath799 .
applying frobenius reciprocity and ( [ eendo ] ) , we obtain @xmath800 this proves claim ( iii ) , and claim ( iv ) is shown similarly .
[ lgenbasis ] we have that 1 . @xmath801 is an @xmath62-basis for @xmath794 , 2 .
@xmath802 is an @xmath62-basis for @xmath803 , and 3 .
@xmath804 is an @xmath62-basis for @xmath805 . for @xmath479 ,
define @xmath806 by proposition [ lpowerbasis ] , @xmath646 is a basis of @xmath807 for each @xmath479 .
let @xmath808 be the map induced by @xmath480 , as in corollary [ lparaideal2 ] . using ( * ? ? ?
* proposition 2.16 ) , and computing as in the proof of lemma [ l050413_1 ] , we have @xmath809 we have shown that the set in ( i ) spans @xmath794 .
a similar argument shows that the set in ( ii ) spans @xmath803 .
now , applying the multiplication map @xmath810 and using corollary [ csymmgen](i ) yields the spanning set of ( iii ) . letting @xmath459 vary over @xmath28 , we have @xmath811 using ( [ edimupper ] ) and the equality @xmath653 for all @xmath410 , we get @xmath812 by corollary [ tdim ] .
the inequalities are therefore equalities , and this implies that the spanning set @xmath813 of @xmath0 is a basis and @xmath658 for all @xmath459 . to show ( i ) and ( ii ) , we have already noted that the claimed bases span @xmath628 and @xmath635 , respectively .
we now apply part ( iii ) to see that they are linearly independent .
[ ccellbasis]the set @xmath814 is an @xmath62-basis for @xmath0 .
apply proposition [ lgenbasis](iii ) and the fact that the filtration by the ideals @xmath486 exhausts @xmath0 , which follows from lemma [ lallidem ] .
recall the notion of an affine cellular algebra from the introduction . in this section ,
we fix @xmath26 and prove that @xmath0 is affine cellular over @xmath2 ( which then implies that it is affine cellular over any @xmath3 ) . for any @xmath410 , we define @xmath815 by corollary [ ccellbasis ] , we have @xmath816 . moreover , @xmath817 . indeed , @xmath818 commutes with elements of @xmath61 in view of hypothesis [ hprop](iv ) .
so by lemma [ lddeltau ] , we have @xmath819 by proposition [ lgenbasis ] , we have @xmath820 , and we have a nested family of ideals @xmath821 . to check that @xmath0 is affine cellular , we need to verify that @xmath822 is an affine cell ideal in @xmath823 . as usual we denote @xmath824 for @xmath825
. the affine algebra @xmath8 in the definition of a cell ideal will be the algebra @xmath61 , with the automorphism @xmath9 being the identity map .
the @xmath2-module @xmath10 will be the formal free @xmath2-module @xmath826 on the basis @xmath651 . by corollary [ csymmgen](i ) and proposition [ lgenbasis ] ,
the following maps are isomomorphisms of @xmath61-modules .
@xmath827 this allows us to endow @xmath828 with a structure of an @xmath829-bimodule and @xmath830 with a structure of an @xmath831-bimodule . in view of corollary [ csymmgen](iii),(iv )
we see that @xmath794 ( resp .
@xmath803 ) is a right ( resp
. left ) @xmath61-module , and so we may define an @xmath0-bimodule homomorphism @xmath832 by proposition [ lgenbasis ] , @xmath833 is an isomorphism .
. this will be the map @xmath835 in the definition of a cell ideal .
the above data make @xmath0 into an affine cellular algebra .
to verify that @xmath805 is a cell ideal in @xmath50 , we first check that our @xmath831-bimodule structure on @xmath836 comes from our @xmath829-bimodule structure on @xmath828 via the rule ( [ erightcell ] ) .
let @xmath837 be the swap map .
this is equivalent to the fact that the composition map @xmath838 is an isomorphism of right @xmath0-modules .
we already know that this is an isomorphism of @xmath2-modules , and so it suffices to check that @xmath839 for all @xmath777 , @xmath840 , and @xmath841 .
note that @xmath842 .
so we have to check @xmath843 by proposition [ lgenbasis](ii ) we can find @xmath844 such that @xmath845 also , by corollary [ csymmgen](i ) , we have @xmath846 using this and the @xmath5-invariance of @xmath847 and @xmath818 , we get ( [ e120613 ] ) as follows : @xmath848 which equals the left hand side of ( [ e120613 ] ) by definition of @xmath691 . to complete the proof , it remains to verify the commutativity of ( [ ecellcd ] ) .
this is equivalent to @xmath849 for all @xmath850 and @xmath851 .
the left hand side equals @xmath852 which equals @xmath853 , as required .
in this section we verify hypothesis [ hprop ] for all finite types . in ade types
( with one exception ) this can be do using the theory of homogeneous representations developed in @xcite .
this theory is reviewed in the next subsection .
we use the cuspidal modules of @xcite . in this section
we assume that the cartan matrix @xmath1 is symmetric . in this subsection
we fix @xmath120 with @xmath857 .
a graded @xmath0-module is called _ homogeneous _ if it is concentrated in one degree .
let @xmath191 .
we call @xmath858 an _ admissible transposition _ for @xmath193 if @xmath859 . the _ word graph
_ @xmath860 is the graph with the set of vertices @xmath861 , and with @xmath146 connected by an edge if and only if @xmath862 for some admissible transposition @xmath863 for @xmath193 . a connected component @xmath864 of @xmath860 is called _ homogeneous _
if for some @xmath865 the following condition holds : @xmath866 [ thomog ] ( * ? ? ?
* theorems 3.6 , 3.10 , ( 3.3 ) ) let @xmath864 be a homogeneous connected component of @xmath860 .
let @xmath867 be the vector space concentrated in degree @xmath868 with basis @xmath869 labeled by the elements of @xmath864 .
the formulas @xmath870 define an action of @xmath0 on @xmath867 , under which @xmath867 is a homogeneous irreducible @xmath0-module .
furthermore , @xmath871 if @xmath872 , and every homogeneous irreducible @xmath0-module , up to a degree shift , is isomorphic to one of the modules @xmath867 . we need to push the theory of homogeneous modules a little further . in proposition
[ phomgenrel ] below we give a presentation for a homogeneous module as a cyclic modules generated by a word vector .
let @xmath864 be a homogeneous component of @xmath860 and @xmath873 .
an element @xmath874 is called _
@xmath193-admissible _ if it can be written as @xmath875 , where @xmath876 is an admissible transposition for @xmath877 for all @xmath878 .
we denote the set of all @xmath193-admissible elements by @xmath879 .
note that if @xmath881 are admissible elements , then @xmath882
if and only if @xmath883 . indeed , it suffices to prove that @xmath884 implies @xmath798 , which follows from the property ( [ enc ] ) .
the lemma follows .
note that the elements in ( [ ej(c ) ] ) annihilate the vector @xmath888 , which generates @xmath867 , whence we have a ( homogeneous ) surjection @xmath889 to prove that this surjection is an isomorphism it suffices to prove that the dimension of @xmath890 is at most @xmath891 , which follows easily from lemma [ ladm ] .
recall the theory of standard modules reviewed in [ sssmt ] .
we now specialize to the case of a _ lyndon _ convex order on @xmath112 as studied in @xcite . for this
we first need to fix a total order ` @xmath30 ' on @xmath94 .
this gives rise to a lexicographic order ` @xmath30 ' on the set @xmath125 .
in particular , each finite dimensional @xmath0-module has its ( lexicographically ) highest word , and the highest word of an irreducible module determines the irreducible module uniquely up to an isomorphism .
this leads to the natural notion of _ dominant words _ ( called good words in @xcite ) , namely the elements of @xmath861 which occur as highest words of finite dimensional @xmath0-modules .
the dominant words of cuspidal modules are characterized among all dominant words by the property that they are _ lyndon words _ , so we refer to them as _
dominant lyndon words_. there is an explicit bijection @xmath892 uniquely determined by the property @xmath893 .
note that this notation @xmath449 will be consistent with the same notation used in [ sssswid ]
. setting @xmath894 if and only if @xmath895 for @xmath896 defines a total order on @xmath112 called a _ lyndon order_. it is known that each lyndon order is convex , and the theory of standard modules for lyndon orders , developed in @xcite , fits into the general theory described in [ sssmt ] .
however , working with lyndon orders allows us to be a little more explicit .
in particular , given a a root partition @xmath259 , set @xmath897 from now on , we fix the notation for the dynkin diagrams as follows : @xmath899{$a_\ell\quad(\ell \geq 1)$ } ; \draw ( 0,0)node[above]{$1 $ } circle ( 0.10 ) ; \draw ( 0.15,0)--(0.85,0 ) ; \draw ( 1,0)node[above]{$2 $ } circle ( 0.10 ) ; \draw [ dotted ] ( 1.15,0)--(2.85,0 ) ; \draw ( 3,0)node[above]{$\ell-1 $ } circle ( 0.10 ) ; \draw ( 3.15,0)--(3.85,0 ) ; \draw ( 4,0)node[above]{$\ell$ } circle ( 0.10 ) ; \end{dynkin}\quad \begin{dynkin } \draw ( .5,.8)node[anchor = west , font=\normalsize]{$b_\ell\quad(\ell \geq 2)$ } ; \draw ( 0,0)node[above]{$1 $ } circle ( 0.10 ) ; \draw ( 0.15,0)--(0.85,0 ) ; \draw ( 1,0)node[above]{$2 $ } circle ( 0.10 ) ; \draw [ dotted ] ( 1.15,0)--(2.85,0 ) ; \draw ( 3,0)node[above]{$\ell-1 $ } circle ( 0.10 ) ; \draw ( 3.15,.05)--(3.85,.05 ) ; \draw ( 3.15,-.05)--(3.85,-.05 ) ; \draw ( 3.5,0)node{$>$ } ; \draw ( 4,0)node[above]{$\ell$ } circle ( 0.10 ) ; \end{dynkin}\quad \begin{dynkin } \draw ( .5,.8)node[anchor = west , font=\normalsize]{$c_\ell\quad(\ell \geq 3)$ } ; \draw ( 0,0)node[above]{$1 $ } circle ( 0.10 ) ; \draw ( 0.15,0)--(0.85,0 ) ; \draw ( 1,0)node[above]{$2 $ } circle ( 0.10 ) ; \draw [ dotted ] ( 1.15,0)--(2.85,0 ) ; \draw ( 3,0)node[above]{$\ell-1 $ } circle ( 0.10 ) ; \draw ( 3.15,.05)--(3.85,.05 ) ; \draw ( 3.15,-.05)--(3.85,-.05 ) ; \draw ( 3.5,0)node{$<$ } ; \draw ( 4,0)node[above]{$\ell$ } circle ( 0.10 ) ; \end{dynkin}\ ] ] @xmath899{$d_\ell\quad(\ell \geq 4)$ } ; \draw ( 0,0)node[above]{$1 $ } circle ( 0.10 ) ; \draw ( 0.15,0)--(0.85,0 ) ; \draw ( 1,0)node[above]{$2 $ } circle ( 0.10 ) ; \draw [ dotted ] ( 1.15,0)--(2.85,0 ) ; \draw ( 3,0)node[above]{$\ell-2 $ } circle ( 0.10 ) ; \draw
( 3,-.15)--(3,-.85 ) ; \draw ( 3,-1)node[right]{$\ell$ } circle ( 0.10 ) ; \draw ( 3.15,0)--(3.85,0 ) ; \draw ( 4,0)node[above]{$\ell-1 $ } circle ( 0.10 ) ; \end{dynkin } \begin{dynkin } \draw ( .75,.8)node[anchor = west , font=\normalsize]{$e_\ell\quad(\ell = 6,7,8)$ } ; \draw ( 0,0)node[above]{$1 $ } circle ( 0.10 ) ; \draw ( 0.15,0)--(0.85,0 ) ; \draw ( 1,0)node[above]{$2 $ } circle ( 0.10 ) ; \draw [ dotted ] ( 1.15,0)--(2.85,0 ) ; \draw ( 3,0)node[above]{$\ell-3 $ } circle ( 0.10 ) ; \draw ( 3,-.15)--(3,-.85 ) ; \draw ( 3,-1)node[right]{$\ell$ } circle ( 0.10 ) ; \draw ( 3.15,0)--(3.85,0 ) ; \draw ( 4,0)node[above]{$\ell-2 $ } circle ( 0.10 ) ; \draw ( 4.15,0)--(4.85,0 ) ; \draw ( 5,0)node[above]{$\ell-1 $ } circle ( 0.10 ) ; \end{dynkin } \begin{dynkin } \draw ( 1,.8)node[anchor = west , font=\normalsize]{$f_4 $ } ; \draw ( 0,0)node[above]{$1 $ } circle ( 0.10 ) ; \draw ( 0.15,0)--(0.85,0 ) ; \draw ( 1,0)node[above]{$2 $ } circle ( 0.10 ) ; \draw ( 1.15,.05)--(1.85,.05 ) ; \draw ( 1.15,-.05)--(1.85,-.05 ) ; \draw ( 1.5,0)node{$>$ } ; \draw ( 2,0)node[above]{$3 $ } circle ( 0.10 ) ; \draw ( 2.15,0)--(2.85,0 ) ; \draw ( 3,0)node[above]{$4 $ } circle ( 0.10 ) ; \end{dynkin } \ ] ] ( 0,.8)node[anchor = west , font=]@xmath900 ; ( 0,0)node[above]@xmath545 circle ( 0.10 ) ; ( 0.15,.05)(0.85,.05 ) ; ( 0.15,0)(0.85,0 ) ; ( 0.15,-.05)(0.85,-.05 ) ; ( 0.5,0)node@xmath716 ; ( 1,0)node[above]@xmath901 circle ( 0.10 ) ; also , we choose the signs @xmath138 as in [ ssklr ] and the total order @xmath30 on @xmath94 so that @xmath902 and @xmath903 if the corresponding labels @xmath106 and @xmath904 satisfy @xmath903 as integers .
we stick with the choices made in [ ssslo ] . throughout the subsection
, we assume that the cartan matrix is of @xmath905 type and @xmath364 is such that @xmath449 is homogeneous .
let @xmath906 .
the module @xmath232 is concentrated in degree 0 , and each of its word spaces is one dimensional .
set @xmath907 .
then we can take @xmath908 .
let @xmath909 , and define @xmath910 .
all parts of hypothesis [ hprop ] are trivially satisfied , except ( v ) . in the rest of this subsection
we verify hypothesis [ hprop](v ) .
we have @xmath913 for a reduced decomposition @xmath914 .
let @xmath3 be the largest index such that @xmath915 is not an admissible transposition of @xmath916 . by theorem [ thomog ]
, @xmath917 is not a word of @xmath232 .
so by corollary [ lbadwords ] , @xmath918 whence @xmath919 .
we prove by induction on @xmath922 that @xmath923 for all @xmath924 .
the base case @xmath925 is trivial .
let @xmath926 , and write @xmath927 . if @xmath928 for all @xmath929 , then @xmath930 is a word of @xmath232 .
on the other hand , lemma [ lhighestwt ] says that @xmath449 is the largest word of @xmath232 and so @xmath931 .
but then @xmath449 is not a lyndon word , which is a contradiction .
thus there exists some @xmath932 with @xmath933 . since the cartan matrix is assumed to be of ade type , either @xmath934 or @xmath935 . in the second case , by homogeneity ( [ enc ] ) we can find @xmath936 with @xmath937
this shows that the definition @xmath938 makes sense . once again by homogeneity we must have that @xmath928 for any @xmath550 with @xmath939 .
therefore , using defining relations in @xmath0 , we get @xmath940 on the other hand , the cycle @xmath941 is not an element of @xmath942 . by lemma [ lbadinblock ]
we must have @xmath943 .
this shows that @xmath944 , and therefore by induction that @xmath923 for every @xmath550 with @xmath924 . by theorem [ tbasis ] ,
an element of @xmath948 is a linear combination of terms of the form @xmath949 such that @xmath950 .
if @xmath951 , then @xmath952 by lemma [ lbadinblock ] .
otherwise , lemma [ ladm ] shows that @xmath798 .
therefore , @xmath659 is spanned by terms of the form @xmath953 . in view of lemma [ lyeq ]
, we see that @xmath659 is generated by @xmath954 . throughout the subsection
, we assume again that the cartan matrix is of @xmath905 type .
by @xcite , with a correction made in ( * ? ? ?
* lemma a7 ) , if @xmath364 is any positive root , except the highest root in type @xmath955 , then @xmath449 is homogeneous .
we have proved in the previous subsection that hypothesis [ hprop ] holds in this case .
now , we deal with the highest root @xmath956 in type @xmath955 . by (
* example a.5 ) , the corresponding lyndon word is @xmath957 define the positive roots @xmath958 then the root partition @xmath959 is a minimal element of @xmath960 .
moreover , @xmath961 and @xmath962 .
indeed , one sees by inspection that these words are highest words in the corresponding homogeneous representations and are lyndon .
finally , we have @xmath963 .
denote by @xmath964 and @xmath965 non - zero vectors in the @xmath966- and @xmath967-word spaces in the homogeneous modules @xmath968 and @xmath969 , respectively .
note that @xmath970 is naturaly a submodule of @xmath971 , so we can consider @xmath972 as a cyclic vector of @xmath971 , and similarly @xmath973 as a cyclic vector of @xmath974 . by definition , @xmath971 is the proper standard module @xmath975 , and let @xmath976 be the image of @xmath972 under the natural projection @xmath977 .
denote by @xmath978 the element of @xmath979 which sends @xmath980 to @xmath981 .
the following has been established in @xcite , see especially ( * ? ? ?
* theorem a.9 , proof ) , but we sketch its very easy proof for the reader s convenience .
[ lses ] the multiplicity of the highest word @xmath982 in @xmath983 is one .
moreover , there is a non - zero vector @xmath984 in the @xmath376-word space of @xmath983 and homogeneous @xmath985-module maps @xmath986 such that the sequence @xmath987 is exact .
finally , @xmath988 by ( * ? ? ?
* theorem 7.2(ii ) ) , the multiplicity of the word @xmath989 in @xmath971 is @xmath545 .
moreover , an explicit check shows that the multiplicity of @xmath982 in @xmath971 is @xmath69 .
we conclude using theorem [ tstand ] and the minimality of @xmath990 in @xmath991 that the standard module @xmath971 is uniserial with head @xmath992 and socle @xmath993 .
the result follows from these observations since @xmath992 is @xmath994-self - dual and @xmath995 in view of ( * ? ? ?
* theorem 2.2 ) . by theorem [ tbasis ] ,
an element of @xmath1011 is a linear combination of terms of the form @xmath1012 such that @xmath1013 . if @xmath1014 , then @xmath1015 by lemmas [ pparae8 ] and [ lbadinblock ] .
so we may assume that @xmath1016 with @xmath1017 , @xmath1018 .
it is easy to check that the only such permutation that fixes @xmath982 is the identity .
we therefore see that @xmath1009 is generated by @xmath1019 .
note that @xmath1020 and @xmath1021 . using the cases @xmath1022 and @xmath1023 proved above and lemma [ pparae8 ]
, we have that @xmath1024 if @xmath1025 or @xmath1026 .
it remains to show that @xmath1024 for some @xmath1027 and @xmath1028 .
let @xmath1029 be the cycle @xmath1030 .
by considering words and using corollary [ lbadwords ] , one can verify that @xmath1031 on the other hand , by the formula for the character of @xmath983 from lemma [ lses ] , we have that @xmath1032 is not a word of @xmath983 .
therefore , by corollary [ lbadwords ] , we have that @xmath1033 , so @xmath1034 , and we are done .
[ lhelper ] suppose that @xmath1036 have been chosen so that hypothesis [ hprop](iii ) is satisfied .
if the minimal degree component of @xmath1037 is spanned by @xmath565 , then hypothesis [ hprop](i ) and ( vi ) are satisfied .
since @xmath1038 has the same degree as @xmath565 , the assumption above implies that @xmath1038 is proportional to @xmath565 .
acting on @xmath547 and using hypothesis [ hprop](iii ) gives @xmath1039 , which upon multiplication by @xmath564 on the right gives the property @xmath1040 , which is even stronger than ( i ) .
it will be clear in almost all cases that the condition of lemma [ lhelper ] will be satisfied , and moreover hypothesis [ hprop](ii ) and ( iv ) are easy to verify by inspection .
this leaves hypothesis [ hprop](v ) to be shown in each case .
let @xmath1047 .
then @xmath1048 , and the irreducible module @xmath1049 is one - dimensional with character @xmath1050 .
define @xmath1051 and @xmath1052 . using corollary [ lbadwords ]
one sees that @xmath1053 for all @xmath550 , which by theorem [ tbasis ] shows that @xmath1054 \bar e_\be$ ] .
this also shows that for @xmath1055 we have the elements of @xmath1056 : @xmath1057 it follows that @xmath1058 \bar e_\be$ ] , and thus @xmath1059 is generated by @xmath1060 .
consider @xmath1061 . in this case ,
@xmath1062 , and @xmath1063 .
define @xmath1064 , @xmath1065 , and @xmath1066 . using corollary [ lbadwords
] , one sees that @xmath1067 for @xmath1068 .
it is also clear that @xmath1069 , and therefore by theorem [ tbasis ] , @xmath1070 e_{\be}$ ] .
we also have the following elements of @xmath1056 : @xmath1071 taken together , these show that @xmath1072 \bar e(\bi_{\be})$ ] . multiplying on both sides by @xmath1073 and using the klr / nil - hecke relations
, we have @xmath1074 \bar e_{\be}.\ ] ] furthermore , @xmath1075 , and so in fact @xmath1076 \bar e_{\be } = f[\bar y_1 ] \bar e_{\be}.\ ] ] consider @xmath1082 . then @xmath1083 and @xmath1084
define @xmath1085 .
define @xmath1086 .
using corollary [ lbadwords ] one sees that @xmath1053 for all @xmath550 , which by theorem [ tbasis ] shows that @xmath1054 \bar e_\be$ ] .
this also shows that for @xmath1055 we have the elements of @xmath1056 : @xmath1087 consequently , @xmath1059 is generated by @xmath1060 .
consider @xmath1088 . then @xmath1089 and @xmath1090 .
define @xmath1085 , and @xmath1086 . using corollary [ lbadwords ]
one sees that @xmath1053 for all @xmath550 , which by theorem [ tbasis ] shows that @xmath1054 \bar e_\be$ ] .
this also shows that for @xmath1055 we have the elements of @xmath1056 : @xmath1091 it follows that @xmath1092 \bar e_\be$ ] .
furthermore , by the relation ( [ r7 ] ) , @xmath1093 and therefore @xmath1094 e_\be = f[\bar y_\be ] \bar e_\be$ ] .
consider @xmath1095 .
then @xmath1096 and @xmath1097 let @xmath1098 be the permutation that sends @xmath1099 to @xmath1100 , and define @xmath1101 .
define also @xmath1102 and @xmath1103 .
set @xmath1104 .
since @xmath1105 is generated by idempotents @xmath1106 with @xmath1107 , and @xmath1108 is the highest weight of @xmath1109 , we see that @xmath1110 let @xmath1111 be the induced map . note that every weight of @xmath232 ends with @xmath1112 , so that @xmath1113 unless @xmath1114 , by corollary [ lbadwords ] .
therefore , applying ( [ eendo ] ) in the type @xmath1 case of @xmath1115 ( which has already been verified ) , we obtain @xmath1116 \bar e_\be.\ ] ] furthermore , @xmath1117 and @xmath1118 thus @xmath659 is generated by @xmath1119 .
the following table shows the choice of data for the remaining roots , except for the highest root @xmath1161 , which we discuss separately . in each of these cases ,
the hypotheses may be verified by employing the same methods used above .
for example , in each case either hypothesis [ hprop](i)-(iv),(vi ) may be verified directly or with the help of lemma [ lhelper ] when it applies .
consider now @xmath1161 , where @xmath1162 .
let @xmath1163 be the permutation that sends @xmath1164 to @xmath1165 , and set @xmath1166 .
let @xmath1167 , and @xmath1168 .
define @xmath1169 .
there is a map @xmath1170 .
this map is not surjective , but one can show that @xmath1171 and thus @xmath1172 \bar e_{(\ga^2 ) } \otimes r_{\al_2 } ) \\ & = \o[\bar y_5 + \bar y_{10 } , \bar y_5 \bar y_{10 } , \bar y_{11 } ] \bar e_\be.\end{aligned}\ ] ] we also compute ( cf .
@xcite ) : @xmath1173 which is zero because it contains the weight @xmath1174 , and this is not a weight of @xmath232 .
since @xmath1175 , we see that @xmath1176 \bar e_\be$ ] , as required .
let @xmath835 be the composition @xmath1194 .
if @xmath1195 , then @xmath1196 , and so @xmath835 is surjective .
furthermore , @xmath1197 .
thus @xmath1198^{{\mathfrak{s}}_{3,1 } } \bar e_\be$ ] .
since @xmath1199 , we have that @xmath659 is generated by @xmath1200^{{\mathfrak{s}}_3 } \bar e_\be$ ] .
observe using ( * ? ? ?
* theorem 4.12(i ) ) that @xmath1201 and @xmath1202 therefore @xmath659 is generated by @xmath1203 .
this is clearly true unless @xmath165 is one of the twelve permutations that stabilizes the weight @xmath449 .
of these , six produce a negative degree . since @xmath565 spans the smallest degree component of @xmath1037 we the claim holds for these six permutations .
two of the remaining six permutations end with the cycle @xmath1211 .
since @xmath1212 , this implies that the claim holds for them too .
finally , reduced decompositions for the remaining non - identity permutations may be chosen so that @xmath1213 by lemmas [ lbadwordsnew ] and [ lhighestwt ] .
now we combine the claim with theorem [ tbasis ] to see that @xmath659 is generated by @xmath1214 \bar e_\be$ ] .
next , by weights and quadratic relations , @xmath1215 .
thus @xmath1216 \bar e_\be$ ] .
this can then be seen to be equal to @xmath1217^{{\mathfrak{s}}_3 } \bar e_\be$ ] , arguing as in the case of the root @xmath1218 above .
again as in the case of the root @xmath1218 , one then shows using specific elements of @xmath1056that @xmath1219 and @xmath1220 , so that @xmath1221 e_\be$ ] .
m. khovanov and a. lauda , a diagrammatic approach to categorification of quantum groups i , _ represent .
* 13 * ( 2009 ) , 309347 .
m. khovanov and a. lauda , a diagrammatic approach to categorification of quantum groups ii , _ trans .
* 363 * ( 2011 ) , 26852700 .
a. kleshchev , cuspidal systems for affine khovanov - lauda - rouquier algebras , _ math .
z. _ , to appear ; arxiv:1210.6556 .
a. kleshchev and a. ram , representations of khovanov - lauda - rouquier algebras and combinatorics of lyndon words , _ math .
* 349 * ( 2011 ) , no .
4 , 943975 . a. kleshchev and a. ram , homogeneous representations of khovanovlauda algebras , _ j. eur . math . soc . _
* 12 * ( 2010 ) , 12931306 .
b. leclerc , dual canonical bases , quantum shuffles and @xmath69-characters , _ math .
z. _ * 246 * ( 2004 ) , 691732 .
g. lusztig , _ introduction to quantum groups _ ,
progress in mathematics 110 , birkhuser , boston , 1993 . | we prove that the khovanov - lauda - rouquier algebras @xmath0 of finite type are ( graded ) affine cellular in the sense of koenig and xi .
in fact , we establish a stronger property , namely that the affine cell ideals in @xmath0 are generated by idempotents .
this in particular implies the ( known ) result that the global dimension of @xmath0 is finite . |
simulation studies of nambu - jona lasinio models have proven to be much more quantitative than those of other field theories @xcite .
in particular , the logarithmic triviality of these models has been demonstrated , although determining logarithmic singularities decorating mean field scaling laws is a daunting numerical challenge .
the reason for this success lies in the fact that when one formulates these four fermi models in a fashion suitable for simulations , one introduces an auxiliary scalar field @xmath3 in order to write the fermion terms of the action as a quadratic form . in this formulation @xmath3 then acts as a chiral order parameter which receives a vacuum expectation value , proportional to the chiral condensate @xmath4 , in the chirally broken phase .
most importantly , the auxiliary scalar field @xmath3 becomes the dynamical mass term in the quark propagator .
the dirac operator is now not singular for quarks with vanishing bare mass and its inversion @xcite , @xcite is successful and very fast .
the algorithm for nambu - jona lasinio models is `` smart '' it incorporates a potential feature of the solution of the field theory , chiral symmetry breaking and a dynamical fermion mass , into the field configuration generator . the good features of the simulation algorithm for the nambu - jona lasinio model can be generalized to lattice qcd @xcite and qed @xcite by incorporating a weak four fermi term in their actions .
these generalized models now depend on two couplings , the familiar gauge coupling and a new four fermi coupling . by choosing the four fermi coupling small
we can be confident that all the dynamics resides in the gauge and fermi fields and the four fermi term just provides the framework for an improved algorithm which allows us to simulate the chiral limit of massless quarks directly .
we shall find a line of spontaneously broken chiral symmetry transition points in the two dimensional coupling constant parameter space of the u(1)-gauged nambu - jona lasinio model . by simulating the model at several regions along the transition line
, we will see that the theory is logarithmically trivial and that the four fermi term is irrelevant in the continuum limit .
our conclusions will be supported by fits with very high confidence levels .
because of the irrelevance of the pure four fermi interaction , this model will make `` textbook '' qed accessible and this paper will address the classic problem of whether qed suffers from complete charge screening .
our measurements will show that the theory is logarithmically trivial and the systematics of the logarithms of triviality follow those of the nambu - jona lasinio model rather than the scalar @xmath5 model as usually assumed . simulating
the @xmath6 case directly has substantial advantages , both theoretical and practical .
when @xmath7 is set to zero , the theory has the exact chiral symmetry of the interaction terms in the action and this forbids chiral symmetry breaking counterterms from appearing in its effective action .
this simplicity can lead to a large scaling window in the direction of the gauge or four fermi coupling in the theory s parameter space .
our simulation results will support this point .
however , when @xmath7 is not zero , as in most past studies of lattice qed and qcd , the effective action has no protection from dangerous symmetry breaking counterterms .
in fact we will find that the scaling window of the lattice theory in the @xmath7-direction is very small and this fact is responsible for the failure of past approaches to lattice qed to address the question of triviality in a straightforward , convincing fashion . in fact , @xcite claimed non - triviality for the theory while @xcite found triviality and backed up their claim further in @xcite by calculating the sign of the beta function , which is directly relevant to the question of triviality .
in addition , we shall check that the algorithm used in this work generates gauge field configurations for couplings near the chiral transition line which are free of lattice artifacts , such as monopoles @xcite and dirac strings , etc . in this paper
we will present data and analyses .
preliminary results have already appeared in letter form @xcite , but this article will contain new data , analyses and discussions .
other applications of the use of a four fermi term to speed lattice gauge theory simulations are also under development and are being applied to qcd @xcite .
it is important to note that in these applications the strength of the four fermi term is weak , so it is not responsible for chiral symmetry breaking .
it just acts as scaffolding which leads to an algorithm that converges efficiently in the limit of massless quarks .
the dynamics resides in the gauge and fermion field interactions .
this paper is organized as follows . in the next section we present the formulation of the lattice action and discuss its symmetries and general features . in the third section we test the algorithm and
tune its parameters . in the next three sections we present data and analyses over a range of gauge couplings for three choices of the irrelevant four fermi coupling on @xmath1 lattices .
the irrelevance of the four fermi coupling is demonstated explicitly and equation of state fits are presented which show that the theory is logarithmically trivial with the same systematics as the nambu - jona lasinio model .
the confidence levels of these fits range from approximately @xmath8 to @xmath9 percent .
analyses of the order parameter s susceptibility reinforce our conclusions . in the seventh section
we consider simulations at nonzero bare fermion masses in order to make contact with past work on pure lattice qed .
we find that subdominant scaling terms are needed to fit the data .
in other words , the usual assumption that the scaling window is wide enough to address the issue of triviality by simulating the model at nonzero fermion masses and fitting to a logarithmically improved mean field form is shown to be incorrect . in section eight
we present data on lattices ranging in size from @xmath0 to @xmath2 to check that our data for the chiral condensate is not influenced significantly by finite size effects for the range of couplings used in the fits . in section nine
we consider measurements of lattice monopole observables to check that they are not critical at the chiral transition points as long as the bare four fermi coupling is nonzero .
in section ten we discuss the possible role of lattice artifacts in simulations of pure lattice qed and address some concerns in the literature .
finally , in section eleven we suggest additional work in this field .
we considered the @xmath10gauged nambu jona lasinio model with four species of fermions .
the lagrangian for the continuum nambu - jona lasinio model is , @xmath11 the symmetries and other properties of @xmath12 have been discussed in @xcite and we refer the reader to that and related references for details
. we will be brief here and just review a few conceptually important points . the pure nambu - jona lasinio model ( eq
. 1 . with @xmath13
set to zero ) has been solved at large @xmath14 by gap equation methods @xcite , and an accurate simulation study of it has been presented @xcite . the lattice action for eq .
1 reads : @xmath15 where @xmath16 where @xmath3 is an auxiliary scalar field defined on the sites of the dual lattice @xmath17 @xcite , and the symbol @xmath18 denotes the set of the 16 lattice sites surrounding the direct site @xmath19 .
the factors @xmath20 are the gauge connections and @xmath21 are the staggered phases , the lattice analogs of the dirac matrices .
@xmath22 is a staggered fermion field and @xmath7 is the bare fermion mass , which will be set to 0 .
note that the lattice expression for @xmath23 is non - compact in the lattice field @xmath24 , while the gauge field couples to the fermion field through compact phase factors which guarantee local gauge invariance .
this point will be discussed further in sec.10 below .
interesting limiting cases of the above action are the pure @xmath25 nambu - jona lasinio model ( @xmath26 ) , which has a phase transition at @xmath27 @xcite and the pure lattice qed ( @xmath28 ) limit , whose chiral phase transition is near @xmath29 for four flavors @xcite , @xcite .
the pure qed ( @xmath28 ) model also has a monopole percolation transition which is probably coincident with its chiral transition at @xmath30 @xcite .
past simulations of this lattice model have led to contradictory results @xcite , @xcite .
since the gauged nambu - jona lasinio model can be simulated at @xmath6 for all gauge couplings , the results reported here will be much more precise and decisive than those of the pure lattice qed ( @xmath28 ) limit .
we scanned the 2 dimensional parameter space ( @xmath31,@xmath32 ) using the hybrid molecular dynamics algorithm tuned for four continuum fermion species @xcite and measured the chiral condensate and monopole susceptibility as a function of @xmath31 and @xmath32 .
we found that as we increased @xmath32 and moved off the @xmath28 axis , the peak of the monopole susceptibility shifted from @xmath30 at @xmath28 to @xmath33 at large @xmath34 .
by contrast the chiral transition point shifted to a larger @xmath31 than the monopole percolation transition for a given value of @xmath34 and became distinct from the monopole percolation point as soon as @xmath34 became nonzero , as shown in the phase diagram , fig.1 .
before turning to physically interesting measurements , we should address some technical issues concerning the algorithm . unlike the hybrid monte carlo algorithm , the hybrid molecular dynamics algorithm is not exact .
the molecular dynamics equations of motion can be found in the literature @xcite . in order to evolve the noisy equations of motion and generate an ensemble of field configurations , one must choose a monte carlo time step @xmath35 @xcite .
the discretization errors have been exhaustively studied and it has been shown that systematic errors in observables behave as @xmath36 @xcite . therefore , we must choose @xmath35 small enough that these systematic errors are no larger than the statistical errors we will encounter . in table 1 and fig .
2 we show the order parameter @xmath3 evaluated on a @xmath0 lattice at gauge coupling @xmath37 and four fermi coupling @xmath38 .
( we write @xmath3 here as a shorthand for @xmath39 , the expectation value of the field .
this is a standard notational shortcut which , hopefully , should nt lead to confusion . ) the table shows that as long as @xmath40 the systematic error in @xmath3 is negligible .
the figure shows that the theoretically expected quadratic dependence of the systematic error on @xmath36 has been confirmed numerically .
the error bars quoted in the table have been obtained using the usual binning techniques , so they reflect the correlations in the measurements .
the last column of the table gives the number of trajectories in each data set .
a trajectory here means an interval of one monte carlo time unit of the algorithm ( for @xmath41 a trajectory consists of one hundred sweeps ) . after each trajectory
a single measurement of @xmath3 was made .
most of our production runs were done using @xmath42 . particularly close to the critical point where these systematiic errors are most dangerous
, we checked our results with runs having @xmath43 .
no problems were encountered .
as stated in the introduction , we made accurate measurements on the chiral critical line for many choices of couplings ( @xmath31 , @xmath32 ) and lattice sizes ranging from @xmath0 to @xmath2 . in this section
we review our data collected varying @xmath44 at fixed @xmath38 on a @xmath1 lattice .
a discussion and presentation of this data has appeared in @xcite , so we will be brief . the data is presented in table 2 .
the columns list the average values of @xmath3 , @xmath45 which is the longitudinal susceptibility of the order parameter @xcite , @xmath46 which is the monopole percolation order parameter and @xmath47 which is the susceptibility of the monopole order parameter @xcite .
the monopole observables will be discussed later .
the data for the order parameter was fit to a form which could accomodate either @xmath5 or nambu - jona lasinio triviality : @xmath48 , where the parameter @xmath49 , the critical point @xmath50 , the amplitude @xmath51 and the scale @xmath52 are determined by the fitting routine . recall that @xmath5 triviality gives @xmath53 and nambu - jona lasinio triviality gives @xmath54 . for the scaling window of gauge couplings
@xmath31 between @xmath55 and @xmath56 , we found the parameters @xmath57 , @xmath58 , @xmath59 and @xmath60 with a confidence level of 34 percent .
the reader should consult the figures and discussison in @xcite for more detail and perspective .
as will be discussed below in sec.9 , these simulations also measured topological observables for the system s vacuum and we confirmed that monopoles and related objects were critical near the chiral transition @xmath57 , @xmath38 .
( we shall see there that the monopole percolation transition is very narrow and occurs at @xmath61 for @xmath38 . )
are other fitting forms possible for this data ?
this is certainly true , of course .
the point we are making , however , is that log - improved mean field theory fits the data with very high confidence levels and there are compelling theoretical reasons for it .
the data and the fits support the conventional wisdom that qed is a trivial field theory and that the logarithms of triviality follow the systematics of the nambu - jona lasinio model rather than the scalar @xmath5 model .
this last point is different from that usually assumed .
in retrospect , it is very plausible that the nambu - jona lasinio model represents the triviality of qed better than @xmath5 , but the differences between the two models have not been emphasized or appreciated in the past .
let s end our discussion of @xmath3 with some examples of other fits .
simple power laws are the first ones to try . for example , a fit of the form @xmath62 is expected to work rather well with @xmath63 slightly larger than @xmath64 since the nambu - jona lasinio fits have worked so well . if we choose the range of @xmath31 to extend from @xmath65 to @xmath56 , we find @xmath66 , @xmath67 , @xmath68 , but the confidence level is very poor , ( @xmath69 ) .
if we accept only a smaller range of couplings closer to the critical point , @xmath31 extending from @xmath55 to @xmath56 , the quality of the fit improves ( @xmath70 , confidence level of @xmath71 percent ) while the critical index @xmath63 rises to @xmath72 .
this is the trend we find in the data : powerlaw estimates of the critical index @xmath63 increase as the range of couplings is restricted closer and closer to the critical coupling .
this systematic drift in the fitting results suggests that a simple power is not an adequate representation of the full data set , but is simply mocking up the logarithm of the nambu - jona lasinio fit , which has a higher confidence level and is stable as different ranges of @xmath31 are considered .
in @xcite we also analyzed the susceptibility associated with @xmath3 .
in mean field theory , the singular piece of the longitudinal susceptibility @xmath73 diverges at the critical point @xmath50 as @xmath74 , @xmath75 , as @xmath76 approaches zero from above in the broken phase , and as @xmath77 in the symmetric phase @xcite .
the critical index @xmath78 is exactly unity in mean field theory . in logarithmically trivial models
@xmath78 remains unity , but the amplitudes @xmath79 and @xmath80 develop weak logarithmic dependences @xcite . in the two component @xmath5 model , @xmath81 , while in the @xmath25 nambu - jona lasinio model , @xmath82 @xcite , where the scale @xmath52 comes from the order parameter fit .
constrained linear fits to the data @xcite produced the amplitude ratio @xmath83 . since @xmath3 varies from @xmath84 to @xmath85 over the @xmath31 range @xmath55 - @xmath56 of the scaling window in the broken phase , the logarithm in the theoretical prediction of the nambu - jona lasinio model states that @xmath86 should range from @xmath87 to @xmath88 .
again , the agreement between the simulation data and theory is very good .
we find no support for the approximate analytic schemes discussed in @xcite which predicted that gauged @xmath89 nambu - jona lasinio models with a four fermi term with continuous chiral symmetry are nontrivial , have powerlaw critical singularities with indices that vary continuously with the couplings @xmath31 and @xmath32 .
additional simulations in sec.5 and 6 below will give strong evidence for the irrelevance of the four fermi term contrary to the results of @xcite .
the reader should recall that truncated @xmath89 nambu - jona lasinio models which account for only restricted sets of feynman diagrams produce nontrivial `` theories '' with critical indices that vary continuously as @xmath31 and @xmath32 are varied .
for example , this occurs if only `` rainbow graphs '' of gauged nambu - jona lasinio models are summed @xcite .
some of these exercises may be relevant to technicolor model building . on the basis of the work here
, however , we suspect that when fermion vacuum polarization is accounted for , one would find complete charge screening and every gauged nambu - jona lasinio model based on continuum noncompact @xmath89 gauge dynamics would be trivial for all couplings .
we suspect that nontriviality and lines of nontrivial field theories are aspects of truncation procedures only .
we suspect , on the basis of the present work and past triviality investigations in scalar qed @xcite , that only models with dynamics beyond continuum noncompact @xmath89 gauge fields and fermions can be nontrivial and have a renormalization group fixed point at nonzero gauge coupling .
an example might be afforded by @xmath89 theories with fundamental monopoles @xcite .
in this section we consider new data collected varying @xmath91 at fixed @xmath90 on a @xmath1 lattice .
the purpose of this series of simulations is to 1 . to verify that the four fermi coupling is irrelevant , and 2 . to accumulate more evidence that the theory is logarithmically trivial in the sense of the nambu - jona lasinio model .
the data is presented in table 3 in the same format as table 2 . in fig .
3 we show the data for the chiral condensate @xmath3 , at fixed @xmath90 and variable @xmath31 .
we use the same fitting procedures as used in sec.4 : @xmath48 , where the parameter @xmath49 , the critical point @xmath50 , the amplitude @xmath51 and the scale @xmath52 are determined by the fitting routine . for the scaling window of gauge couplings @xmath31 between @xmath92 and @xmath93 , we found the parameters @xmath94 , @xmath95 , @xmath96 and @xmath97 with a confidence level of 87 percent .
this excellent fit is the one shown in the figure .
note that eight data points for @xmath31 between @xmath98 and @xmath99 were used in the fit while the figure has two additional points at stronger coupling .
those points lie slightly below the fit , are slightly outside the scaling window and show the extent of the scaling window .
we plot the eight data points between @xmath31 of @xmath98 and @xmath99 and the fit as shown in fig.4 .
, @xmath100 vs. @xmath101 to illustrate the importance and numerical significance of the logarithm .
the dashed line is the previous fit redrawn in this format , where we have zoomed in on the scaling window for emphasis .
clearly this fit is stable to further cuts on the data set since all the data points lie on the fit .
we conclude that nambu - jona lasinio triviality accommodates the lattice data at @xmath90 with very good confidence levels .
this success also shows the irrelevance of the four fermi term in the lattice action : the scaling law for the order parameter is the same as that at the larger @xmath32 value although the lattice parameters , such as the location of the critical point , have changed .
next , in fig.5 we show the inverse of the longitudinal susceptibility of the auxiliary field @xmath3 at fixed @xmath90 and variable @xmath31 .
we follow the same procedures as used in sec.4 to analyze and plot the data here .
the plot picks out a critical point @xmath102 and is consistent with the mean field value of the critical index @xmath103 .
the constrained linear fits determine the amplitude ratio , @xmath104 . since @xmath3 varies from @xmath105 to @xmath106 over the @xmath31 range @xmath92 - @xmath93 of the scaling window in the broken phase , the logarithms in the theoretical prediction of the nambu - jona lasinio model for the amplitude ratio predict that @xmath86 should range from @xmath87 to @xmath88 .
again , the agreement between the simulation data and theory is good , but is not comparable in quality or decisiveness to our other fits .
in this section we consider new data collected varying @xmath108 at fixed @xmath107 on a @xmath1 lattice . in this case
the four fermi coupling is four times stronger than the data discussed in the previous section , but far too weak to cause chiral symmetry breaking in the absence of the gauge coupling .
the analysis and plots here are identical to the previous discussions of @xmath38 and @xmath90 , so we will be brief .
the data is presented in table 4 in the same format as table 1 .
in fig.6 we show the data for the chiral condensate @xmath3 , at fixed @xmath107 and variable @xmath31 .
we use the same fitting procedures as used in sec.3 : @xmath48 , where the parameter @xmath49 , the critical point @xmath50 , the amplitude @xmath51 and the scale @xmath52 are determined by the fitting routine . for the scaling window of gauge couplings
@xmath31 between @xmath109 and @xmath110 , we found the parameters @xmath111 , @xmath112 , @xmath113 and @xmath114 with a confidence level of 99.9 percent .
this impressive fit is the one shown in the figure .
we plot the data and the fit as shown in fig.7 , @xmath100 vs. @xmath101 to illustrate the importance and numerical significance of the logartihm .
the dashed line is the previous fit redrawn in this format .
the success of this fit reiterates the irrelevance of the four fermi term : the scaling law for the order parameter is the same as that at the @xmath32 values of @xmath115 and @xmath116 although the lattice parameters , such as the location of the critical point , have changed .
next , in fig.8 we show the inverse of the longitudinal susceptibility of the auxiliary field @xmath3 at fixed @xmath107 and variable @xmath31 .
the plot picks out a critical point @xmath117 and the constrained linear fits to the data shown in the figure produced the amplitude ratio @xmath118 . which compares well to the theoretical prediction @xmath119 .
again , the agreement between the simulation data and theory is good , but is not comparable in quality or decisiveness to our order parameter fits .
past simulations of lattice qed had to be done at nonzero fermion mass @xcite .
the standard algorithms fail to converge in the limit @xmath120 because the lattice dirac operator becomes singular in the chiral limit @xcite , @xcite .
this algorithmic problem has led to indecisive results for lattice qed because of large statistical and systematic errors .
it is interesting to use the algorithm of this paper to discover , assess and clarify the problems in past work in this field .
we chose to do simulations at nonzero fermion mass at the critical coupling @xmath50 determined by our fits presented in the previous section . in this way we can look for the width of the scaling window in the @xmath7-direction in a particularly simple fashion .
recall that at criticality the order parameter @xmath3 should scale with the fermion mass @xmath7 , an explicit symmetry breaking parameter , as @xmath121 where the critical index @xmath122 should be @xmath123 in a logarithmically trivial theory and @xmath124 , the power of the logarithm should be @xmath125 for a @xmath5 theory and should be @xmath126 for a nambu - jona lasinio model @xcite . by accumulating data over a range of small @xmath7 values , we can look for the region where eq .
6 might apply and determine the width of the scaling window .
it is important to keep the number of variables and parameters to a minimum in this sort of investigation .
this is the reason we work at criticality .
the critical coupling has been determined to be @xmath127 in sec.4 .
the data for the order parameter and its susceptibility are shown in table 5 for @xmath7 ranging from @xmath128 to @xmath129 .
note that the statistics for this data set is particularly high as smaller and smaller @xmath7 values are considered and the critical point is approached .
the error bars in @xmath3 recorded there account for critical slowing down which forced us to accumulate such high statistics .
the statistics are at least an order of magnitude greater than those of past studies and produce @xmath3 values with errors ranging from @xmath64 percent to @xmath130 percent .
we learned in past studies of the pure nambu - jona lasinio that small @xmath7 values , typically below @xmath131 , are needed to find a scaling window @xcite .
however , in this case the dynamics is controlled by the gauge coupling which alone is driving chiral symmetry breaking .
the four fermi coupling is tiny and is not affecting the dynamics in a numerically significant fashion .
therefore , the width of the scaling window must be determined anew from the data in table 5 . in fig .
9 we plot @xmath132 vs. @xmath7 in order to assess visually the relevance of the leading logarithm result eq
the data clearly pick out the value @xmath133 for the dominant powerlaw singularity of the scaling law for very small @xmath7 values , all less than @xmath131 .
however , we also see that the deviations from the mean field result are numerically significant over the entire mass range shown .
in fact , they are far too large to be accommodated by a weak logarithmic scaling violation as expected in eq . 6 .
in fact , a fit of that form to the data ranging from @xmath134 to @xmath135 produces a huge value for the power of the logarithm , @xmath136 , and a very small confidence level of @xmath137 percent ( @xmath138 ) .
therefore , the data rule out the applicability of logarithmic improved mean field scaling to describe the data at nonzero @xmath7 except for the very smallest values of @xmath7 , @xmath139 .
unfortunately , most data used to study the potential triviality of qed using the conventional action employed @xmath7 values considerably larger than @xmath140 in order to run efficiently and generate sufficient statistics .
typical ranges of @xmath7 have been between @xmath131 and @xmath141 @xcite and are very sensitive to data taken with @xmath142 , @xmath143 and @xmath144 .
this criticism applies to all past studies of noncompact qed , for example , @xcite .
it also means that the methods of analysis introduced in @xcite do not apply to this data set because those methods require data in a scaling window , controlled by a single asymptotic form .
higher precision data taken at the smallest values of @xmath7 , @xmath139 , are required apparently and , in fact , larger lattices than @xmath1 might be necessary also because of the possibility of significant finite size effects .
two possible explanations for the data come to mind : 1 . perhaps the real critical point is significantly different from @xmath145 as determined by our fits at @xmath146 , or 2 .
perhaps subdominant singularities in the scaling law are numerically significant over this range of @xmath7 .
it is easy to rule out option 1 . ignoring logarithms ,
the mean field equation of state reads @xmath147 .
this implies that if @xmath31 were different from @xmath50 , then @xmath132 would behave as @xmath148 , and the correction term would be large for small @xmath3 , which is just the opposite of the behavior observed in fig .
9 . now consider option 2 .
if a subdominant singularity contributes to the equation of state , then at criticality the relation @xmath149 should be replaced by , @xmath150 where @xmath122 should be @xmath123 and @xmath151 should be considerably larger @xcite .
this hypothesis fits the data beautifully : the curved dashed line in fig .
9 shows the fit which has a confidence level of @xmath152 percent ( @xmath153 ) .
in fact this fitting form can be well approximated in a fashion that is useful for practical purposes , @xmath154 eq .
8 approximates eq . 7 because the correction to the constancy of @xmath132 is less than @xmath155 percent over the range of @xmath7 values in the figure .
the fit gives @xmath156 , @xmath157 , and @xmath158 .
we learn several lessons from this exercise .
previous simulations of pure qed at nonzero @xmath7 could not possibly have detected the logarithms of triviality decorating mean field singularities . for the present range of @xmath7 values and lattice sizes , data at nonzero @xmath7
have contributions from subdominant critical singularities which are larger numerically than logarithmic corrections to mean field theory .
since we are using a new algorithm which works in the limit of massless quarks , we should be careful to monitor finite size effects .
some of our data are taken very near to critical points in order to find critical indices that control continuum limits of the lattice models . at these points
the model s correlation length diverges and there are potentially dangerous finite size effects which could mimic finite temperature effects , for example . we need to check that the lattice is large enough to contain correlations larger than the lattice spacing but smaller than the system s spatial extent in order to work within a scaling window where we can extract continuum features of the field theory . in table 6
we show data for @xmath3 taken for gauge couplings @xmath31 ranging from @xmath159 to @xmath160 at fixed four fermi coupling @xmath161 for @xmath0 , @xmath1 and @xmath2 lattices .
the comparison of the three data sets shows coincidence everywhere except at @xmath162 between the smallest lattice @xmath0 and the other two .
@xmath162 was our closest approach to the critical point in the symmetry broken phase and it appears that our @xmath1 lattice was sufficient , given our @xmath64 percent statistical errors .
reliance on a @xmath0 lattice would have failed us . in the next table
we show @xmath0 data for a simulation where the four fermi coupling is fixed at @xmath38 and @xmath31 ranges from @xmath159 to @xmath163 .
the data consists of @xmath3 as well as monopole observables that will be discussed in a later section . comparing the @xmath3 data here to that in table 3 ,
we confirm the absence of finite size effects within our statistical errors . in summary ,
the @xmath1 data we have used to extract scaling laws from @xmath3 measurements appears free of significant finite size effects .
the significance of finite size effects depends strongly on the observable being simulated .
we also checked that the longitudinal susceptibility data that was used to extract the logarithmic violations of scaling in the amplitudes was not distorted by finite size effects .
since these susceptibilities are determined with much larger statistical error bars , this test was less demanding .
certainly the finite size effects in @xmath45 are much larger than those in @xmath3 itself . however , since @xmath3 was determined within a fraction of a percent while the statistical uncertainty in @xmath45 was typically several percent , a @xmath1 lattice was adequate for the range of couplings used in this study .
noncompact lattice qed was first studied with the goal of simulating the dynamics of @xmath89 gauge fields without the monopoles that accompany compact lattice qed @xcite .
it was found , however , that even the noncompact formulation has monopole - like dislocations in its lattice formulation because of the space - time cutoff itself @xcite .
these dislocations can undergo a percolation transition where long range correlations develop @xcite . because of this transition , it is not obvious that simulation results in pure noncompact lattice qed reflect the physics of textook qed in which field configurations are smooth and have no topological excitations .
the formulation of noncompact lattice qed with a four fermi term is free of the issues raised in @xcite .
the point is , as discussed in the section 2 above , the monopole percolation transition does not coincide with the chiral transition as long as the four fermi coupling is nonzero . therefore the gauge field vacuum is free of critical dislocations at the gauge couplings of interest , so we know that we are studying a model free of topological excitations , as we wish .
let s find the monopole percolation transition in the model with a fixed four fermi coupling @xmath38 .
the data for the monopole concentration @xmath46 and the associated monopole percolation susceptibility @xmath47 , both defined exactly as in @xcite , are given in table 2 .
in fig.10 we plot the monopole concentration against the gauge coupling and find a percolation transition at @xmath164 .
we determined in sec.4 that the chiral transition occurs at considerably weaker coupling , @xmath165 , where the monopole concentration is insignificant , as we read off fig.10 .
it is also informative to confirm this conclusion by considering the monopole percolation susceptility , @xmath47 . in fig.11
we plot this susceptibility against the gauge coupling and see that it appears to diverge in the vicinity of @xmath164 ( we also confirmed this impression with powerlaw fits ) .
in addition , in fig.12 we plot the longitudinal susceptibility of the chiral transition and confirm that it diverges near @xmath165 , as already determined in sec.4 .
the two susceptibility peaks are cleanly separated : @xmath164 vs. @xmath165 .
we end this section with a minor remark about the finite size effects observed in the monopole observables . comparing table 2 and table 7
, we see that as the monopole percolation transition s critical coupling is approached , there are numerically significant differences between the @xmath0 and the @xmath1 data sets for both the concentration @xmath46 as well as its associated susceptibility @xmath47 .
as expected , the percolation susceptibility @xmath47 is strongly suppressed by the lattice size near the transition . in fact , as we have discussed elsewhere @xcite , finite size scaling of the peak of the susceptibility is an effective and accurate means to measure the percolation critical indices .
it would take simulations on a series of lattice sizes to carry out such a program for this model .
the only point we wish to make here , however , is that the percolation and chiral transitions are well separated inside the phase diagram fig .
it is interesting ( and fortunate for the success of this project ) that the finite size effects in the chiral order parameter @xmath3 are significantly smaller than those in the monopole concentration .
although the major topic in this research is the behavior of the gauged nambu - jona lasinio model for @xmath167 , we will briefly discuss the present confusing state of theory and simulations at the edge of the phase diagram @xmath166 where past simulations have been carried out .
as we have already emphasized , the real problem with studies at @xmath166 is that they must be done at nonzero fermion mass away from the chiral limit and this has caused several problems : 1 .
the simulations become excessively slow for small @xmath7 values because the lattice dirac operator is singular in that limit . therefore , at low values of @xmath7 where the best statistics are required , the statistics of the data sets are typically the poorest .
the scaling window in the @xmath7-direction is extremely narrow , so fitting forms which only account for the leading critical behavior are inadequate and misleading .
attempting to go beyond leading order critical singularities in fits leads to a vast proliferation of parameters which undermines firm conclusions .
another potential problem concerning the @xmath166 edge of the phase diagram concerns lattice monopoles .
recall that one motivation for inventing and studying noncompact lattice qed @xcite was to make a model free of monopoles in order to understand the relation between chiral symmetry breaking and single gluon exchange .
however , hands and wensley @xcite pointed out that even the the noncompact model has monopole - like lattice dislocations because of gauge invariance of the pure gauge field piece of the action and because of the lattice cutoff itself .
these authors also pointed out that these lattice monopoles experience a percolation transition as the gauge coupling becomes strong and in the case of quenched simulations , the monopole percolation transition is very close to the chiral transition experienced by light fermions @xcite .
this led these authors to speculate that noncompact lattice qed might not be a sound framework for studying `` textbook '' qed at strong coupling @xcite . what does this possibility mean for this paper ?
since we work at @xmath167 where the critical line of monopole percolation is distinct from the chiral transition line , these lattice artifacts are not relevant to our conclusions .
we believe that we have a firm theoretical and numerical grasp of gauged nambu - jona lasinio models everywhere within the phase diagram fig.1 but not along the edge @xmath166
. how could this be ?
following hands and wensley , the gauge field piece of the action eq.2 is invariant under local gauge transformations defined by the group of real numbers @xmath168 , while the fermionic piece of the action , which describes the gauge invariant hopping of the fermion around the lattice , has a gauge symmetry based on phases , @xmath89 .
the cutoff theory described by the pure gauge piece of the action has monopole excitations attached by dirac strings @xcite .
these are singular field configurations whose actions diverge when the lattice spacing is taken to zero
. they would be of no concern if it were nt for the fact that as the coupling increases they experience a percolation transition where monopole clusters develop macroscopic dimensions .
since the fermions are sensitive to monopole clusters through their @xmath89 phase , hands and wensley speculated that they could affect the chiral transition in the quenched and unquenched model .
this speculation could be wrong for several reasons : 1 .
the underlying gauge action is just a quadratic form , so it is a perfectly solvable free field theory .
a free field theory ca nt have a phase transition , as emphasized in @xcite and 2 .
percolation transitions need not affect the bulk properties of the underlying field theory .
many examples of this sort can be cited .
these complaints can be answered in part : 1 .
the phase transition of percolation is not in local observables constructed out of the gauge fields , but rather is in nonlocal matrix elements .
it is not unusual in statistical mechanics to make models where non - local matrix elements experience phase transitions when the underlying local field theory has no transition itself .
condensed matter physics provides many examples of enormous practical importance including , for example , the localization - delocalization transition of single electrons in background fields of varying degrees of disorder .
the chiral transition is sensitive to loops of the @xmath89 phase and is of this type .
2 . since fermions flip their chirality in the presence of monopoles , it is plausible that a percolating network of monopole - like excitations can induce chiral symmetry breaking in the bulk system . there is a possibility that the @xmath169 pure qed model has qualitatively different physics than that found anywhere within the phase diagram in fig.1 . only at the edge of the diagram
would the percolating monopole - like excitations be critical where chiral symmetry is broken .
only there are new degrees of freedom , percolating monopoles , relevant so only there could there be a new universality class
. it might be that on the edge of the phase diagram , the chiral condensate is driven by monopole percolation and the chiral transition inherits a correlation length critical index @xmath170 from the percolating network and becomes the basis for a nontrivial field theory @xcite .
it has been noticed that as the number of fermions is varied , both the chiral and monopole percolation transitions move in unison @xcite .
in addition , in unquenched models , such as the four flavor model on the edge of the phase diagram fig.1 , the fermions induce @xmath89 plaquette terms into the theory s action which can support conventional lattice monopoles @xcite .
we have nothing to add to the pros and cons of these qualitative arguments .
we hope that the physics issues brought up here could be answered by striking out in new directions and finding approaches or arguments which are more precise and quantitative .
the monopole percolation picture may contain only half truths , but some of those ideas might be testable in the context of models with real monopoles , generalizations of compact @xmath89 lattice qed @xcite , perhaps .
we presented numerical evidence for the triviality of textbook qed using a new algorithm which converges for massless quarks .
past simulations using the action with massive quarks but no four fermi term produced controversial results . recall that @xcite claimed non - triviality for the theory while @xcite found triviality and backed up their claim further in @xcite by calculating the sign of the beta function , which is directly relevant to the question of triviality .
one could calculate the theory s renormalized couplings and their rg trajectories in the chiral limit , extending the work of @xcite to a two parameter space . both the gauge and the four fermi couplings should vanish as the reciprocal of the logarithm of the ultra - violet cutoff . as discussed in @xcite
this calculation has some technical challenges specific to lattices of finite extent which necessitate the extrapolation of raw lattice data to achieve physical results
. it would be worthwhile to investigate improved strategies here to avoid crude , indecisive results .
the high quality of the equation of state fits in sec.4 , 5 and 6 should lead to improved determinations of the renormalized couplings because the lattice critical couplings are determined with excellent precision .
one could also simulate the model with the @xmath25 chiral group replaced by a continuous group so the model would have goldstone bosons even on a coarse lattice @xcite .
it would then be possible to test the approach and results of @xcite more quantitatively .
it would also be interesting to generalize the results of sec.7 , that a subdominant critical singularity is needed to describe the data at nonzero @xmath7 , away from the critical coupling .
in other words , fit the finite @xmath7 data points of previous investigations such as @xcite to equations of state with both a dominant and subdominant singularity and check that improved confidence levels are achieved with simple hypotheses .
unfortunately , there will be a proliferation of fitting parameters in such a program , so its numerical significance might be questioned .
nonetheless , it would definitely be worth consideration .
such a program would also influence the determination of renormalized couplings because these calculations use critical couplings inferred from equation of state fits @xcite .
finally , it would be interesting to simulate compact qed with a small four fermi term and study the interplay of monopoles , charges and chiral symmetry breaking .
since the @xmath171 limit of the compact model is known to have a first order transition @xcite , generalizations of the action will be needed to find a continuous transition where a continuum limit of the lattice theory might exist .
since the parameter space of the generalized model would be at least three dimensional , this interesting problem would be quite challenging .
this work was partially supported by nsf under grant nsf - phy96 - 05199 .
s. k. is supported by the korea research foundation .
l. wishes to thank the _
ect@xmath172 _ , trento , for hospitality during the final stages of this project .
the simulations were done at npaci and nersc .
s. kim , a. koci@xmath173 and j.b .
kogut , nucl . phys . *
b429 * , 407 ( 1994 ) . phys . *
b220 * , 102 ( 1983 ) .
s. duane , a.d .
kennedy , b.j .
pendleton and d. roweth , phys .
lett . * b195 * , 216 ( 1987 ) .
s. duane and j.b .
kogut , phys .
* 55 * , 2774 ( 1985 ) .
s. gottlieb , w. liu , d. toussaint , r.l .
renken and r.l .
sugar , phys . rev . * d35*,2531 ( 1987 ) .
kogut , j .- f .
lagae , and d.k .
sinclair , phys . rev . *
d58 * , 34004 ( 1998 ) .
kogut , and d.k .
sinclair , hep - lat/0005007 .
s. kim , j.b .
kogut and m .-
lombardo , phys .
* b55 * , 2774 ( 2001 ) .
s. hands and r. wensley , phys .
. lett . * 63 * , 2169 ( 1989 ) .
b. rosenstein , b. warr and s. park , phys .
* 205 * , 497 ( 1991 ) .
y. cohen , s. elitzur and e. rabinovici , nucl . phys . *
b220 * , 102 ( 1983 )
. v. azcoiti , g. di carlo , a. galante , a.f .
grillo , v. laliena , and c.e .
piedrafita , phys . lett . *
b353 * , 279 ( 1995 ) ; * b379 * , 179 ( 1996 ) .
kogut and k.c .
wang , phys.rev . * d53 * , 1513 ( 1996 ) .
e. dagotto and a. koci@xmath173 and j.b .
b317 * , 271 ( 1989 ) .
s. duane and j.b .
. phys . * b275 * [ fs17 ] , 398 ( 1986 ) .
a. koci@xmath173 , j.b .
kogut and m .-
b398 * , 376(1993 ) ; a. koci@xmath173 and j.b .
. phys . * b422 * [ fs ] , 593 ( 1994 ) . c. itzykson and j .- m .
drouffe , statistical field theory ( cambridge university press , 1989 . ) v. azcoiti , g. di carlo , a. galante , a.f .
grillo , v. laliena , and c.e .
piedrafita , phys.lett . *
b355 * , 270 ( 1995 ) .
c.n leung , s.t .
love and w.a .
bardeen , nucl .
phys.*b327 * , 649(1986 ) . m. baig , h. fort , s. kim , j.b .
kogut and d.k .
phys . rev .
* d48 * , r2385 ( 1993 ) ; m. baig , h. fort , j.b .
kogut and s. kim .
phys . rev . * d51 * , 5216 ( 1995 ) .
a. koci@xmath173 , j.b .
kogut and s. hands .
b357 * , 467 ( 1991 ) ; s. hands and j.b .
kogut , nucl . phys . *
b462 * , 291 ( 1996 ) .
a. koci and j. kogut , nucl .
b * 422 * , 593 ( 1994 ) .
r. myerson , t. banks and j.b .
* b129 * , 493 ( 1977 ) .
s. hands , a. koci@xmath173 and j.b . kogut .
289b * , 400 ( 1992 ) .
j. bartholomew , j.b .
kogut , s. h. shenker , j. sloan , j. shigemitsu , d. k. sinclair and h. w. wyld .
b230 * , [ fs10 ] , 222 ( 1984 ) . m. gckeler , r. horsley , p.e.l . rakow and g. schierholz ,
d * 53 * , 1508 ( 1996 ) .
m. gockeler , r. horsley , v. linke , p. rakow , g. schierholz and h. stuben , phys .
* 80 * , 4119 ( 1998 ) .
e. dagotto and j.b .
kogut , phys .
59 * , 617 ( 1987 ) .
cdd @xmath35 & @xmath3 & @xmath174 + @xmath175 & 0.1279(6 ) & 1200 + @xmath176 & 0.1280(5 ) & 2000 + @xmath177 & 0.1284(4 ) & 2400 + @xmath178 & 0.1293(4 ) & 2800 + @xmath179 & 0.1301(3 ) & 3500 + @xmath180 & 0.1314(2 ) & 4800 + cddddd @xmath181 & @xmath3 & @xmath45 & @xmath46 & @xmath47 & @xmath174 + @xmath182 & 0.11980(7 ) & 0.1756(5 ) & 0.97676(1 ) & 0.1245(5 ) & 930 + @xmath183 & 0.11248(8 ) & 0.1843(20 ) & 0.9566(1 ) & 0.261(1 ) & 950 + @xmath184 & 0.10438(8 ) & 0.1993(50 ) & 0.9221(1 ) & 0.552(2 ) & 1030 + @xmath185 & 0.09531(9 ) & 0.2160(50 ) & 0.8644(2 ) & 1.220(5 ) & 1010 + @xmath186 & 0.08520(8 ) & 0.2621(30 ) & 0.7669(3 ) & 3.04(2 ) & 1500 + @xmath187 & 0.0738(1 ) & 0.3230(40 ) & 0.5976(7 ) & 10.3(1 ) & 1310 + @xmath93 & 0.0674(1 ) & 0.3671(30 ) & 0.463(1 ) & 25.3(6 ) & 1012 + @xmath188 & 0.0609(1 ) & 0.4301(30 ) & 0.250(2 ) & 124.0(9 ) & 1130 + @xmath189 & 0.0537(2 ) & 0.4849(40 ) & 0.0812(8 ) & 122.4(9 ) & 2701 + @xmath190 & 0.0456(2 ) & 0.6591(40 ) & 0.0338(5 ) & 69.7(5 ) & 1120 + @xmath56 & 0.0367(2 ) & 0.9356(40 ) & 0.0192(2 ) & 43.5(2 ) & 1670 + @xmath191 & & & 0.0130(2 ) & 30.2(1 ) & 810 + @xmath192 & -0.00008(9 ) & 3.360(90 ) & 0.00798(8 ) & 19.57(5 ) & 810 + @xmath193 & 0.0002(5 ) & 1.903(80 ) & 0.00681(4 ) & 16.90(2 ) & 2518 + @xmath194 & -0.00002(3 ) & 1.316(70 ) & 0.00601(5 ) & 14.87(3 ) & 960 + @xmath195 & 0.0003(4 ) & 0.919(50 ) & 0.00544(2 ) & 13.34(1 ) & 3350 + @xmath196 & -0.0007(3 ) & 0.786(40 ) & 0.00483(4 ) & 12.07(2 ) & 820 + @xmath197 & -0.0006(3 ) & 0.625(50 ) & 0.00407(4 ) & 10.28(1 ) & 1010 + @xmath198 & -0.0003(2 ) & 0.525(10 ) & 0.00303(6 ) & 9.051(9 ) & 1070 + @xmath199 & -0.0002(2 ) & 0.484(10 ) & 0.00166(6 ) & 8.185(7 ) & 1230 + @xmath200 & 0.0002(2 ) & 0.432(10 ) & 0.00074(5 ) & 7.521(6 ) & 1150 + cddddd @xmath181 & @xmath3 & @xmath45 & @xmath46 & @xmath47 & @xmath174 + @xmath183 & 0.05201(4 ) & 0.066(2 ) & 0.9479(1 ) & 0.327(1 ) & 730 + @xmath201 & 0.04932(4 ) & 0.068(1 ) & 0.9293(1 ) & 0.487(2 ) & 770 + @xmath184 & 0.04642(5 ) & 0.075(2 ) & 0.9047(2 ) & 0.729(3 ) & 660 + @xmath202 & 0.04332(5 ) & 0.074(2 ) & 0.8729(3 ) & 1.106(5 ) & 600 + @xmath185 & 0.03996(6 ) & 0.082(2 ) & 0.8314(4 ) & 1.72(1 ) & 500 + @xmath203 & 0.03644(6 ) & 0.089(3 ) & 0.7765(5 ) & 2.81(2 ) & 640 + @xmath186 & 0.03267(5 ) & 0.096(3 ) & 0.7054(6 ) & 4.82(3 ) & 960 + @xmath204 & 0.02848(7 ) & 0.110(4 ) & 0.6091(9 ) & 9.35(9 ) & 770 + @xmath187 & 0.02401(9 ) & 0.136(4 ) & 0.4751(16 ) & 24.2(9 ) & 640 + @xmath93 & 0.01872(9 ) & 0.190(5 ) & 0.2677(29 ) & 115(4 ) & 780 + @xmath188 & 0.01250(9 ) & 0.38(1 ) & 0.08801(19 ) & 127(2 ) & 960 + @xmath189 & -0.00013(64 ) & 1.62(9 ) & 0.03625(95 ) & 74.1(9 ) & 310 + @xmath190 & -0.00027(38 ) & 0.75(9 ) & 0.02333(52 ) & 51.3(7 ) & 320 + @xmath56 & 0.00009(15 ) & 0.33(4 ) & 0.01576(26 ) & 35.6(2 ) & 490 + @xmath191 & 0.00009(9 ) & 0.23(4 ) & 0.01172(15 ) & 27.9(1 ) & 660 + @xmath205 & 0.00009(9 ) & 0.19(4 ) & 0.00937(12 ) & 22.8(8 ) & 590 + @xmath192 & 0.00010(8 ) & 0.16(2 ) & 0.00779(9 ) & 19.22(6 ) & 640 + @xmath193 & 0.00010(6 ) & 0.145(9 ) & 0.00665(6 ) & 16.63(4 ) & 910 + @xmath194 & 0.00009(6 ) & 0.133(7 ) & 0.00586(5 ) & 14.67(3 ) & 790 + cddd @xmath181 & @xmath3 & @xmath45 & @xmath174 + @xmath182 & 0.2525(2 ) & 0.39(5 ) & 1500 + @xmath183 & 0.2434(2 ) & 0.43(2 ) & 1500 + @xmath184 & 0.2339(2 ) & 0.43(5 ) & 1500 + @xmath185 & 0.2237(2 ) & 0.44(5 ) & 1500 + @xmath186 & 0.2129(2 ) & 0.48(3 ) & 1500 + @xmath187 & 0.2012(2 ) & 0.51(4 ) & 1500 + @xmath188 & 0.1885(2 ) & 0.61(3 ) & 1500 + @xmath190 & 0.1751(3 ) & 0.68(3 ) & 1500 + @xmath191 & 0.1606(3 ) & 0.84(2 ) & 1500 + @xmath192 & 0.1450(3 ) & 1.04(5 ) & 1500 + @xmath194 & 0.1281(4 ) & 1.26(4 ) & 1500 + @xmath196 & 0.1095(4 ) & 1.68(4 ) & 1500 + @xmath197 & 0.0881(5 ) & 2.23(5 ) & 1500 + @xmath206 & 0.000(6 ) & 5.66(10 ) & 1500 + @xmath207 & 0.000(7 ) & 3.68(10 ) & 1500 + @xmath208 & 0.000(7 ) & 3.17(10 ) & 1500 + @xmath209 & 0.000(6 ) & 2.61(10 ) & 1500 + @xmath210 & 0.000(4 ) & 2.33(10 ) & 1500 + @xmath211 & 0.000(5 ) & 2.31(7 ) & 1500 + @xmath212 & 0.000(5 ) & 1.80(6 ) & 1500 + cddd @xmath7 & @xmath3 & @xmath45 & @xmath174 + @xmath213 & 0.0286(1 ) & 0.927(9 ) & 3400 + @xmath214 & 0.0315(1 ) & 0.779(2 ) & 3194 + @xmath215 & 0.0340(1 ) & 0.681(4 ) & 3040 + @xmath216 & 0.0361(1 ) & 0.639(5 ) & 3397 + @xmath217 & 0.0380(1 ) & 0.582(8 ) & 3207 + @xmath218 & 0.03944(9 ) & 0.538(5 ) & 2174 + @xmath219 & 0.04247(9 ) & 0.474(5 ) & 1837 + @xmath220 & 0.04890(8 ) & 0.415(5 ) & 2521 + @xmath221 & 0.05357(8 ) & 0.377(4 ) & 1740 + @xmath222 & 0.06110(9 ) & 0.325(8 ) & 1260 + @xmath223 & 0.06688(8 ) & 0.290(7 ) & 1380 + @xmath224 & 0.07160(7 ) & 0.269(7 ) & 1400 + @xmath225 & 0.07571(8 ) & 0.255(6 ) & 960 + @xmath226 & 0.07916(7 ) & 0.240(6 ) & 1020 + @xmath227 & 0.08214(8 ) & 0.231(7 ) & 1000 + @xmath228 & 0.08480(8 ) & 0.217(6 ) & 910 + @xmath229 & 0.08724(7 ) & 0.211(6 ) & 970 + @xmath182 & 0.09636(7 ) & 0.193(3 ) & 970 + @xmath187 & 0.10237(8 ) & 0.177(3 ) & 730 + cddd @xmath181 & @xmath230 & @xmath231 & @xmath232 + @xmath182 & & 0.2525(2 ) & + @xmath183 & & 0.2434(2 ) & + @xmath184 & 0.2341(4 ) & 0.2339(2 ) & + @xmath185 & 0.2239(4 ) & 0.2237(2 ) & + @xmath186 & 0.2130(4 ) & 0.2129(2 ) & + @xmath187 & 0.2013(5 ) & 0.2012(2 ) & + @xmath188 & 0.1885(5 ) & 0.1885(2 ) & + @xmath190 & 0.1747(6 ) & 0.1751(3 ) & 0.1748(4 ) + @xmath191 & 0.1606(6 ) & 0.1606(3 ) & 0.1617(5 ) + @xmath192 & 0.1451(7 ) & 0.1450(3 ) & 0.1454(3 ) + @xmath194 & 0.1281(6 ) & 0.1281(4 ) & 0.1283(4 ) + @xmath196 & 0.1089(8 ) & 0.1095(4 ) & 0.1093(4 ) + @xmath197 & 0.0866(8 ) & 0.0881(5 ) & 0.0885(4 ) + cdddd @xmath181 & @xmath3 & @xmath46 & @xmath47 & @xmath174 + @xmath182 & 0.1203(2 ) & 0.9770(1 ) & 0.123(1 ) & 1000 + @xmath183 & 0.1129(2 ) & 0.9568(2 ) & 0.261(2 ) & 1000 + @xmath184 & 0.1047(2 ) & 0.9223(4 ) & 0.551(5 ) & 1000 + @xmath185 & 0.0956(2 ) & 0.8641(6 ) & 1.21(1 ) & 1000 + @xmath186 & 0.0854(3 ) & 0.7660(10 ) & 3.10(5 ) & 1000 + @xmath187 & 0.0741(3 ) & 0.5954(22 ) & 10.6(3 ) & 1000 + @xmath188 & 0.0610(4 ) & 0.2669(49 ) & 71.5(9 ) & 1000 + @xmath190 & 0.0449(7 ) & 0.0651(16 ) & 51.0(7 ) & 1000 + @xmath191 & & 0.0291(5 ) & 26.6(2 ) & 1000 + @xmath192 & & 0.0203(6 ) & 18.4(2 ) & 1000 + @xmath194 & & 0.0152(4 ) & 14.1(1 ) & 1000 + | by adding a small , irrelevant four fermi interaction to the action of noncompact lattice quantum electrodynamics ( qed ) , the theory can be simulated with massless quarks in a vacuum free of lattice monopoles .
simulations directly in the chiral limit of massless quarks are done with high statistics on @xmath0 , @xmath1 and @xmath2 lattices at a wide range of couplings with good control over finite size effects , systematic and statistical errors .
the lattice theory possesses a second order chiral phase transition which we show is logarithmically trivial , with the same systematics as the nambu - jona lasinio model .
the irrelevance of the four fermi coupling is established numerically .
our fits have excellent numerical confidence levels .
the widths of the scaling windows are examined in both the coupling constant and bare fermion mass directions in parameter space . for vanishing fermion mass
we find a broad scaling window in coupling which is essential to the quality of our fits and conclusions . by adding a small bare fermion mass to the action we find that the width of the scaling window in the fermion mass direction is very narrow .
only when a subdominant scaling term is added to the leading term of the equation of state are adequate fits to the data possible .
the failure of past studies of lattice qed to produce equation of state fits with adequate confidence levels to seriously address the question of triviality is explained .
the vacuum state of the lattice model is probed for topological excitations , such as lattice monopoles and dirac strings , and these objects are shown to be non - critical along the chiral transition line as long as the four fermi coupling is nonzero .
our results support landau s contention that perturbative qed suffers from complete screening and would have a vanishing fine structure constant in the absence of a cutoff .
-1 truecm -1 truecm |
in hadron - hadron scattering , interactions are classified by the characteristics of the final states . in elastic scattering ,
both hadrons emerge unscathed and no other particles are produced . in diffractive dissociation , the energy transfer between the two interacting hadrons remains small , but one ( single dissociation ) or both ( double dissociation ) hadrons dissociate into multi - particle final states , preserving the quantum numbers of the associated initial hadron .
the remaining configurations correspond to inelastic interactions .
the most difficult conceptual aspect of diffractive scattering is to provide a unique and concise definition .
this will not be attempted here and diffraction will be understood as an interaction between projectile and target that generates a large rapidity gap between the respective final states , which is not exponentially suppressed .
diffractive interactions are mediated by the exchange of a colorless object , with quantum numbers of the vacuum .
this definition fits very well the framework of soft interactions , where diffractive scattering is mediated by the exchange of the universal pomeron trajectory ( @xmath0 ) , introduced by gribov @xcite .
ingelman and schlein @xcite proposed to use diffractive scattering in the presence of a large scale to establish the partonic content of the pomeron . in qcd ,
the candidate for vacuum exchange with properties similar to the soft pomeron is two gluon exchange @xcite . as a result of interactions between the two gluons , a ladder structure develops . in perturbative qcd ( pqcd ) , the properties of this ladder depend on the energy and scales involved in the interaction , implying its non - universal character .
each of the approaches mentioned above leads to definite predictions , which can be tested in high energy diffractive interactions in the presence of a hard scale .
this has been pursued in @xmath1 scattering at hera and in @xmath2 scattering at the tevatron .
the purpose of this talk is to summarize the recently achieved progress .
the variables used to analyze diffractive scattering will be introduced for deep inelastic @xmath1 scattering ( dis ) .
since dis is perceived as a two - step process , in which the incoming lepton emits a photon which then interacts with the proton target , the relevant variables can be readily generalized to @xmath2 interactions . interactions . _ _ ] a diagram for diffractive scattering in dis , where the diffracted state is separated from the scattered proton by a large rapidity gap ( lrg ) , is presented in figure [ fig : dis - diag ] and all the relevant four vectors are defined therein .
the usual dis variables are the negative of the mass squared of the virtual photon , @xmath3 , the square of the center of mass energy of the @xmath4 system , @xmath5 , the bjorken scaling variable , @xmath6 , which in the quark parton model constitutes the fraction of the proton momentum carried by the interacting quark , and the inelasticity , @xmath7 .
in addition to the usual dis variables , the variables used to described the diffractive final state are , @xmath8 @xmath9 is the fractional proton momentum which participates in the interaction with @xmath10 .
it is sometimes denoted by @xmath11 .
@xmath12 is the equivalent of bjorken @xmath13 but relative to the exchanged state .
@xmath14 is the invariant mass of the hadronic final state recoiling against the leading proton , @xmath15 .
the approximate relations hold for small values of the four - momentum transfer squared @xmath16 and large @xmath17 , typical of high energy diffraction .
to describe diffractive dis , it is customary to choose the variables @xmath9 and @xmath16 in addition to the usual @xmath13 and @xmath18 in the cross section formula . the diffractive contribution to @xmath19
is denoted by @xmath20 and the corresponding differential contribution , integrated over @xmath16 , is @xmath21 the three - fold differential cross section for @xmath1 scattering can be written as @xmath22 \sigma_r^{d(3)}(x , q^2,{x_{{{\scriptscriptstyle { i\!\!p } } } } } ) \ , , \label{eq : f2d4}\ ] ] where @xmath23 @xmath24 stands for the diffractive longitudinal structure function , which may not be small .
the structure function @xmath19 is related to the absorption cross section of a virtual photon by the proton , @xmath25 . for diffractive scattering , in the limit of high @xmath17 ( low @xmath13 ) ,
@xmath26 this relation allows predictions for diffractive scattering in dis based on regge phenomenology applied to @xmath4 scattering .
in fact many of the questions that are addressed in analyzing diffractive scattering are inspired by regge phenomenology as established in soft hadron - hadron interactions .
the scattering of two hadrons , @xmath27 and @xmath28 , at squared center of mass energy @xmath29 , is described by the exchange of the universal @xmath0 trajectory parameterized as @xmath30 .
the @xmath0 trajectory determines the @xmath31 dependence of the total cross section , @xmath32 .
the ratio of elastic and diffractive to total cross sections , is expected to rise like @xmath33 .
a steep and universal @xmath9 dependence of the diffractive cross section is expected , @xmath34 .
values of @xmath35 @xcite and @xmath36 @xcite were derived based on total hadron - proton interaction cross sections and elastic proton - proton data .
recently the @xmath0 intercept has been reevaluated @xcite leading to a value of @xmath37 .
the positive value of @xmath38 implies that the slope of the @xmath16 distribution is increasing with @xmath39 .
this fact , borne out by the hadron - hadron and photoproduction data ( for a review and new data see @xcite ) , is known as shrinkage of the @xmath16 distribution .
it is due to the fact that @xmath40 and has been explained by gribov @xcite as diffusion of particles in the exchange towards low transverse momenta , @xmath41 , with @xmath42 ( see also @xcite ) .
qcd factorization for the diffractive structure function of the proton , @xmath43 , is expected to hold @xcite , while it can not be proven for hadron - hadron interactions @xcite .
@xmath43 is decomposed into diffractive parton distributions , @xmath44 , in a way similar to the inclusive @xmath19 , @xmath45 where @xmath46 is the universal structure function for dis on parton @xmath47 , @xmath48 is the factorization scale at which @xmath44 are probed and @xmath49 is the fraction of momentum of the proton carried by the diffractive parton @xmath47 .
diffractive partons are to be understood as those which lead to a diffractive final state .
the dglap evolution equation applies in the same way as for the inclusive case . for a fixed value of @xmath9 ,
the evolution in @xmath13 and @xmath18 is equivalent to the evolution in @xmath12 and @xmath18 .
if , following ingelman and schlein @xcite , one further assumes the validity of regge factorization , @xmath43 may be decomposed into a universal @xmath0 flux and the structure function of the @xmath0 , @xmath50 where the normalization of either of the two components is arbitrary .
it implies that the @xmath9 and @xmath16 dependence of the diffractive cross section is universal , independent of @xmath18 and @xmath12 , and given by @xmath51 one of the expectations which is subject to experimental tests .
the mechanism for producing lrg is assumed to be present at some scale and the evolution formalism allows to probe the underlying partonic structure .
the latter depends on the coupling of quarks and gluons to the pomeron .
at hera the diffractive candidate events are selected either by requiring a large rapidity gap @xcite , or by requiring a leading proton @xcite .
the various analyzes differ in the way the non - diffractive contributions are treated and in the way the proton dissociative events are subtracted .
the comparison between the various measurements is shown in figure [ fig : f2dcomp ] . the @xmath9 dependence of @xmath52 , in the region of @xmath53 ,
is expected to be dominated by @xmath0 exchange .
the corresponding values of @xmath54 are shown as a function of @xmath18 in figure [ fig : f2dcomp ] .
a dependence of @xmath55 on @xmath18 can not be excluded , however the errors are large enough , so that a constant value of @xmath56 fits the data .
however averaged over the whole @xmath18 range the value of @xmath55 is definitely larger than the intercept of the soft pomeron .
in addition , the diffractive @xmath57 value is only half of that for inclusive @xmath19 measurements , also shown in the figure .
this means that the ratio of diffractive to total @xmath4 cross sections is constant with @xmath17 .
those are indications that the connection of diffractive dis to the simple soft @xmath0 picture is not straight forward .
the h1 measurements of @xmath52 @xcite , which cover by far the largest phase space , have been used to perform a qcd evolution fit to extract diffractive parton distribution functions ( dpdf ) .
the results of the fit are shown in figure [ fig : dpdf ] .
a very good description of the data is obtained , provided the parton distributions are dominated by gluons , which carry about 80% of the momentum of partons leading to diffractive events .
this latest extraction of dpdfs by h1 is called h1 2002 dpdfs .
qcd factorization can be tested in high transverse momenta , @xmath58 , jet production in @xmath4 , @xmath59 and @xmath2 diffractive production .
if factorization holds , the cross section for production of jets @xcite and charm @xcite in @xmath4 should be well reproduced by nlo calculations with dpdfs , extracted from structure function measurements .
this is indeed the case as demonstrated in figure [ fig : jetscc ] .
qcd factorization breaking is observed in hard diffractive scattering in @xmath2 .
measurements @xcite of two jet production accompanied by the presence of a leading anti - proton have been used to extract the effective diffractive structure function for two jet production @xmath60 , which can then be compared with the expectations from dpdf extracted at hera .
as shown in figure [ fig : fnaljj ] , even for the h1 2002 dpdfs , where the abundance of gluons is lesser compared to earlier dpdfs @xcite , the expectations are by about a factor 10 above the measurements .
interactions as a function of @xmath12 , compared to expectations of different sets of dpdfs : left , for single pomeron exchange , right , double pomeron exchange.__,width=207 ] it should be stressed however , that the range of @xmath11 ( that is @xmath9 ) covered by the @xmath2 measurements is beyond the range probed by the h1 data , from which the dpdfs originate . in extrapolating the h1 parameterization into this high @xmath11 region ,
the reggeon contribution is estimated to be of the order of 30 to 40% .
a lesser qcd factorization breaking is observed for diffractive dijet production in @xmath2 events in which both baryons remain unscathed - the so called double pomeron exchange ( dpe ) process @xcite .
compared to expectations , the rate of dijet production is only by about factor two less abundant .
this is shown in figure [ fig : fnaljj ] .
effects of factorization breaking are also expected for quasi - real @xmath59 interactions , due to the presence of the resolved photon component .
the measurements of diffractive dijets by the h1 experiment have been compared to nlo calculations @xcite based on the h1 2002 dpdfs .
, for the diffractive @xmath61 reaction , compared to nlo calculations without ( @xmath62 ) or with ( @xmath63 ) suppression of the resolved photon contribution .
@xmath64 denotes the diffracted system accompanying the two jets , separated by a large rapidity gap from the proton or its dissociative state @xmath65 .
_ , width=207 ] as shown in figure [ fig : xgam ] , a good agreement with data is obtained if the resolved photon contribution is suppressed relative to the direct contribution . the factor 0.34 that multiplies the resolved component in the calculation was motivated by the recent work of kaidalov et al .
kaidalov et al .
@xcite investigated what fraction of the gluon distribution in the proton leads to diffractive final states .
the ratio of diffractive to inclusive dijet production cross sections as a function of @xmath13 of the gluon , for different hard scattering scales and for the h1 2002 dpdfs is presented in figure [ fig : xgam2 ] . of the gluon for different scales of the hard scattering , for the h1 2002 dpdfs . also shown
is the unitarity limit , called pumplin bound . _ _ ] this ratio should be smaller than 0.5 @xcite , while for scales @xmath66 this limit is exceeded for @xmath67 .
this indicates that unitarity effects may already be present in diffractive scattering and may explain why the rise of diffractive scattering cross section with @xmath17 is slower than expected ( see section [ sec : incl ] ) .
the dynamics behind diffractive dis can be easier understood if the process is viewed in the rest frame of the proton .
the virtual photon develops a partonic fluctuations , whose lifetime is @xmath68 @xcite . at the small @xmath13 typical of hera , where @xmath69 , it is the partonic state rather than the photon that scatters off the proton .
if the scattering is elastic , the final state will have the features of diffraction .
the fluctuations of the @xmath10 are described by the wave functions of the transversely and longitudinally polarized @xmath10 which are known from perturbative qcd .
small and large partonic configurations of the photon fluctuation are present . for large configurations
non - perturbative effects dominate in the interaction and the treatment of this contribution is subject to modeling . for a small configuration of partons ( large relative @xmath41 ) the total interaction cross section of the created color dipole on a proton target
is given by @xcite @xmath70 where @xmath71 is the transverse size of the color dipole and @xmath72 is the scale at which the gluon distribution @xmath73 of the proton is probed .
the corresponding elastic cross section is obtained from the optical theorem . in this picture , the gluon dominance in diffraction results from the dynamics of perturbative qcd ( see equation ( [ eq : qqgp ] ) ) .
models of diffraction that follow this approach are quite successful in describing both the inclusive @xmath19 and the diffractive @xmath20 measurements , where the former are used to parameterize the dipole - proton cross section . an example taken from @xcite
is shown in figure [ fig : dipole ] .
measurements.__,width=321 ] it is interesting to note that the two models @xcite which predict a relatively mild increase of @xmath52 with decreasing @xmath9 are the ones which explicitly include effects of unitarization through saturation of cross sections for large size dipoles .
the saturation scale is increasing with increasing @xmath17 .
the dynamical origin of this form of saturation can be derived from a new form of qcd matter , called color gluon condensate @xcite . in this approach
, the data suggest that the saturation scale @xmath74 for @xmath75 .
the presence of small size @xmath76 configurations in the photon can be tested in exclusive vector meson ( vm ) production as well as for deeply inelastic compton scattering . at high energy ( low @xmath13 ) and in the presence of a large scale ( large @xmath18 or heavy flavor )
, these reactions are expected to be driven by two - gluon exchange
. a closer look at the theory of exclusive processes in qcd shows that the two partons taking part in the exchange do not carry the same fraction of the proton momentum .
that makes these processes sensitive to correlations between partons , which are encoded in the so - called generalized parton distributions , gpds @xcite .
these new constructs relate in various limits to the parton distributions , form factors and orbital angular momentum distributions .
the motivation behind studies of exclusive processes is to establish the region of validity of pqcd expectations and ultimately to pursue a full mapping of the proton structure , which can not be achieved in inclusive measurements .
the cross section for the exclusive processes is expected to rise with @xmath17 , with the rate of growth increasing with the value of the hard scale .
a compilation of logarithmic derivatives @xmath77 , for @xmath78 @xcite , @xmath79 @xcite and @xmath80 @xcite exclusive production , as a function of the scale defined as @xmath81 , where @xmath82 is the mass of the vm , is presented in figure [ fig : delta ] . with decreasing transverse size of the dipole
, the @xmath16 distribution is expected to become universal , independent of the scale and of the vm .
the exponential slope of the @xmath16 distribution , @xmath28 , reflects then the size of the proton .
a compilation @xcite of measured @xmath28 values is presented in figure [ fig : delta ] . around @xmath83 of about @xmath84 indeed the @xmath28 values become universal .
another important manifestation of the perturbative nature of exclusive processes , related to the universality of the @xmath16 distribution , is that the slope , @xmath38 , of the corresponding regge trajectory should become small .
the parameters of the effective regge trajectory can be determined in the study of the @xmath17 dependence of the differential cross section for exclusive processes at fixed @xmath16 .
the results obtained for exclusive vector meson production @xcite are compiled in figure [ fig : apom ] .
( dots ) and @xmath38 ( open squares ) values , extracted in exclusive vm production , as a function of @xmath83 .
_ , width=207 ] the deeply virtual compton scattering ( dvcs ) process , @xmath85 , has been advocated as one of the exclusive processes for which theoretical calculations are free of uncertainties due to hadronic wave function uncertainties @xcite .
in addition , the interference of the dvcs and qed bethe - heitler amplitudes for prompt @xmath86 production is proportional to the real part of the qcd amplitude , which in turn is sensitive to gpds .
the extraction of the dvcs cross section in @xmath1 scattering has been performed by the h1 @xcite and zeus @xcite experiments
. a clear rise of the dvcs cross section with @xmath17 has been observed @xcite as shown in figure [ fig : dvcs ] . as a function of @xmath17 ( left ) for fixed @xmath18 and as a function of @xmath18 for fixed @xmath17 ( right).__,width=283 ] a comparison of the dvcs cross section dependence on @xmath17 and @xmath18 with two approaches , one based on the dipole model @xcite and the other on nlo evolution of gpds with postulated initial conditions @xcite ,
is shown in figure [ fig : dvcs ] . a good agreement with data
is obtained .
diffractive production of vm or prompt @xmath86 at large values of @xmath16 accompanied by proton dissociation , form another class of interactions which are of interest for understanding the high energy regime of pqcd .
the large value of @xmath16 accompanied by a large rapidity gap , suggests the applicability of the leading logarithmic bfkl dynamics @xcite . the @xmath16 distribution for the process @xmath87
@xcite , where the lrg separates the photon from the dissociated proton state @xmath65 is shown in figure [ fig : largetp ] . at the proton vertex in the process @xmath88 .
the lines represent theoretical expectations based on bfkl , as described in the figure .
right : @xmath17 dependence of @xmath89 in bins of @xmath16 measured at the proton vertex .
the lines represent theoretical expectations based on the bfkl or dglap dynamics , as described in the figure .
_ ] the @xmath16 distribution is well represented by calculations based on leading logarithmic bfkl approximation @xcite .
the bfkl approach also describes well the @xmath17 dependence of the cross section for exclusive @xmath80 production at large @xmath16 , while expectations based on the leading logarithmic dglap dynamics fail to describe the observed dependence for @xmath90 .
this is shown in figure [ fig : largetp ] .
this is yet another indication that the dglap dynamics , which is successfully used to describe the measurements of @xmath19 at hera @xcite , may not be sufficient to describe all the features of dis at high energy .
the exclusive diffractive production of the higgs boson has been proposed @xcite as a potential background - free method to search for the light higgs at lhc . a process similar to exclusive higgs production is the exclusive @xmath91 production , for which at the tevatron the cross section is predicted to be about @xmath92 @xcite .
the diagram corresponding to the proposed diffractive process is shown in figure [ fig : higgs ] .
production in @xmath2 scattering .
right : invariant mass distribution of the candidate exclusive @xmath93 system .
_ , width=132 ] production in @xmath2 scattering .
right : invariant mass distribution of the candidate exclusive @xmath93 system .
_ , width=245 ] the cdf experiment @xcite has searched for the exclusive process @xmath94 and the invariant mass of the dimuon - photon system for candidate events is shown in figure [ fig : higgs ] . under the assumption that all found events originate from exclusive @xmath91 production , the measured cross section was estimated to be @xmath95 , which is far from the expected number .
i am very thankful to my colleagues from the hera and fnal experiments for their help in collecting the material for this talk .
i would also like to acknowledge the hospitality of the max planck institute in munich , where this writeup was prepared , and the humboldt foundation for making my stay in munich possible .
the research was partly supported by the israel science foundation .
99 v. n. gribov , jetp lett . *
41 * , 667 ( 1961 ) .
g. ingelman and p. e. schlein , phys .
* b152 * , 256 ( 1985 ) . f. e. low , phys .
* d12 * , 163 ( 1975 ) . s. nussinov , phys . rev .
lett . * 34 * , 1286 ( 1975 ) .
a. donnachie and p. v. landshoff , phys .
* b296 * , 227 ( 1992 ) [ hep - ph/9209205 ] .
a. donnachie and p. v. landshoff , nucl .
* b231 * , 189 ( 1984 ) . k. kang et al . , hep - ph/9812429 . j. r. cudell et al .
d * 65 * , 074024 ( 2002 ) .
j. c. collins , phys .
rev . * d57 * , 3051 ( 1998 ) [ hep - ph/9709499 ] ; erratum phys .
rev . * d61 * , 2000 ( 1998 ) .
a. berera and d. e. soper , phys .
rev . * d53 * , 6162 ( 1996 ) [ hep - ph/9509239 ]
. l. trentadue and g. veneziano , phys .
lett . *
b323 * , 201 ( 1994 ) .
h1 collaboration , paper 089 and 090 , submitted to the international europhysics conference on high energy physics , eps03 , aachen 2003 .
t. affolder _ et al . _
[ cdf collaboration ] , phys .
* 85 * , 4215 ( 2000 ) ; k. goulianos , hep - ph/0407035 .
h1 collaboration , paper 87 , submitted to the international europhysics conference on high energy physics , eps03 , aachen 2003 .
l. n. lipatov , sov .
j. nucl .
* 23 * , 338 ( 1976 ) ; e. a. kuraev , l. n. lipatov and v. s. fadin , sov .
jetp * 45 * , 199 ( 1977 ) ; i. i. balitsky and l. n. lipatov , sov . j. nucl
. phys .
* 28 * , 822 ( 1978 ) .
h1 collaboration , paper 6 - 0183 submitted to the 32nd international conference on high energy physics , ichep04 , beijing 2004 . | recent experimental results on inclusive diffractive scattering and on exclusive vector meson production are reviewed .
the dynamical picture of hard diffraction emerging in perturbative qcd is highlighted .
= 14.5pt |
the proposal of materials with simultaneous negative electric permittivity and magnetic permeability by veselago in 1967 @xcite has opened the door toward the design of novel and remarkable optical devices based on the use of metamaterials or photonic crystals , such as the perfect flat lens @xcite or the invisibility cloak @xcite .
recently , we have shown how these negative electromagnetic properties can be revisited through the admittance formalism @xcite , which is widely used in the thin - film community @xcite and defined the computational rules for the effective indices and phase delays associated with wave propagation through negative - index layers @xcite .
we have demonstrated that we can simulate the optical properties of negative index material ( nim ) layer by replacing it with a positive index material ( pim ) with the same effective index ( @xmath0 ) , provided that we use for this pim layer a * virtual * thickness * opposite * to that of the nim layer ( @xmath1 ) , which is reminiscent of optical space folding in complementary media @xcite .
this computational rule is easily implementable in standard thin - film software and has allowed us to analyze the spectral properties of some standard multilayer stacks , such as the antireflection coating , the quarter - wavelength bragg mirror and the fabry - perot bandpass filter , in which one or more layers of these stacks involve negative index materials @xcite . among the presented results , the most spectacular concerns the large increase in the spectral bandwidth of a quarter - wavelength bragg mirror induced by the use of a negatively refracting material ( either the high - index layers or the low - index layers ) and the ability to tailor the phase properties of such multilayer structures by adjusting the number and the features of nim layers within the stack .
the objective of this work is to define an optimization method for such improvements and to identify the design of a _ white _ fabry - perot , i.e. a multilayer cavity that spontaneously exhibits * resonant behavior over a very large spectral range*.
let us consider a bragg mirror that contains @xmath2 alternated quarter - wavelength layers as described by the following formula incident medium / @xmath3 / substrate the refractive index of the semi - infinite glass substrate is denoted by @xmath4 , whereas that of the semi - infinite incident medium is denoted by @xmath5 .
the incident medium and the substrate are non - absorbing positive - index materials .
each layer of the stack can be composed of a positive ( high - index @xmath6 , low - index @xmath7 ) or a negative index material ( @xmath8 , @xmath9 ) , each of which should be non - absorbing .
moreover , in this first approach we neglect the dispersion law of refractive indices for all materials under study .
though this assumption may appear too simplistic in the case of negative index materials , recently published results @xcite showed that it was possible to efficiently control this dispersion in a wide spectral range . to determine the reflection properties of such a stack , we use the following basic formula @xmath10 where @xmath11 is the effective index of the incident medium , @xmath12 is the complex admittance of the stack , @xmath13 is the amplitude reflection coefficient , @xmath14 is the corresponding reflectance and @xmath15 is the phase change at the reflection .
the effective index of a medium is given by the general relation @xcite @xmath16 where @xmath17 is defined by @xmath18 for a plane wave passing through a multilayer stack that contains @xmath2 layers , @xmath19 is an invariant quantity that is defined by the angle of incidence ( aoi ) @xmath20 in the incident medium @xmath21 where @xmath22 is the layer number , while @xmath17 depends on the layer and is defined , in the propagating mode , by ( @xmath23 ) @xmath24 consequently , we can rewrite relation ( [ eq : effectiveindexalpha ] ) in the form @xmath25 where @xmath26 is the vacuum impedance .
relation ( [ eq : effectiveindextheta ] ) is independent of the type of material ( pim or nim ) within the layer because @xmath27 and @xmath28 are simultaneously negative in the case of negative index materials .
the computation of the @xmath12 factor is based on the application of a recursive formula that links the admittances at two consecutive boundaries @xcite @xmath29 where @xmath30 is the phase delay introduced by the crossing of the layer @xmath22 .
the initialization of this recursive formula occurs in the substrate where only the outgoing plane wave is present @xmath31 the phase delay @xmath30 is given by @xmath32 where @xmath33 is the effectivel thickness of the layer @xmath22 and @xmath34 a binary coefficient equal to + 1 ( -1 ) for a layer that consists of a positive index ( negative index ) material .
this last relation justifies the statement in section [ sec : introduction ] and indicating that we can replace each nim with an equivalent pim that is characterized by a virtual negative thickness @xcite .
all of the layers in the bragg mirror stack are quarter - wavelength ; hence , at zero aoi , we can write @xmath35 where @xmath36 is the central wavelength of the mirror .
if we use a linear approximation of this last relation near @xmath36 , we have @xmath37 consequently , the recursive formula ( [ eq : recursiveformula ] ) becomes , at the same level of approximation @xmath38\ ] ] by initializing this new recursive formula using relation ( [ eq : initialization ] ) , we find @xmath39 with @xmath40^q\enskip\text{for } p=2q\\ \frac{\tilde{n}_h^2}{\tilde{n}_s}&\left[\frac{\tilde{n}_h^{2}}{\tilde{n}_l^{2}}\right]^q\enskip\text{for } p=2q+1 \end{aligned } \right .
\label{eq : y0lambda0}\ ] ] and @xmath41^l-\frac{\tilde{n}_h}{\tilde{n}_s}\sum\limits_{l=0}^{p-1}\gamma_{p - l}\left[\frac{\tilde{n}_h}{\tilde{n}_l}\right]^l \label{eq : alambda0}\ ] ] by combining ( [ eq : reflectioncoefficient ] ) and ( [ eq : linearadmittance ] ) , we finally obtain analytical expressions for the reflectance @xmath14 and the spectral derivative of the phase change at the reflection @xmath42 , both at the design wavelength @xmath36 @xmath43 ^ 2\ ] ] @xmath44
a multilayer fabry - perot ( fp ) cavity is composed of a thin spacer ( with refractive index @xmath45 and thickness @xmath46 ) surrounded by two quarter - wavelength bragg mirrors deposited at the surface of a semi - infinite substrate .
the overall round trip phase @xmath47 of this planar cavity is defined by @xmath48 where @xmath49 is the optical path difference corresponding to the round trip of the light in the spacer layer ( @xmath50 ) and @xmath51 ( @xmath52 ) the phase change at the reflection on the upper mirror ( lower mirror ) .
a fabry - perot resonance is defined by a central wavelength @xmath36 for which the overall round - trip phase @xmath53 is a multiple of @xmath54 and by a spectral bandwidth @xmath55 which is given , at leading order approximation , by @xmath56_{\lambda_0}|}\cdot\frac{1-\sqrt{r_{\text{ls}}^+r_{\text{us}}^-}}{(r_{\text{ls}}^+r_{\text{us}}^-)^{\frac{1}{4}}}\ ] ] in other words , the spectral bandwidth of such a fp resonance can become extremely large if the linear dependence of the overall round - trip phase in the cavity @xmath57_{\lambda_0}=-\frac{4\pi n_{\text{sp}}e_{\text{sp}}}{\lambda_0 ^ 2}+\left.\frac{\partial\rho_{\text{us}}^-}{\partial\lambda}\right|_{\lambda_0}+\left.\frac{\partial\rho_{\text{ls}}^+}{\partial\lambda}\right|_{\lambda_0 } \label{eq : philineardependence}\ ] ] is equal or very close to zero .
to fulfill this condition and thus obtain a resonant behavior over a very large spectral range , it is absolutely required that at least * one of the cavity mirrors includes nim layers * , which is the only way to obtain a * positive * linear dependence of the phase change at reflection @xcite . to determine the optimal design of such a * _ white _ multilayer fabry - perot cavity * , we systematically investigate the variation of @xmath58_{\lambda_0}$ ] of all the symmetric multilayer stacks described by the general formula air / h@xmath59 2l @xmath60h / glass when the refractive index @xmath6 of the high - index material varies in the range between 2.00 and 3.00 , while the other refractive indices remain constant ( @xmath5 = 1.00 , @xmath4 = 1.52 and @xmath7 = 1.48 ) . in the previous stack formula
, @xmath61 is an integer equal to 1 , 2 or 3 , while h and l represent either pim ( h , l ) or nim ( @xmath62 , @xmath63 ) quarter - wavelength layers . the calculus is performed for zero aoi . for each stack formula that exhibits a cancellation of the spectral dependence of the overall round - trip phase at the design wavelength @xmath36 , we thus calculate its spectral transmittance @xmath64 , and the variation of the ratio between the square modulus @xmath65 of the electric field within the stack and the square modulus of the incident field @xmath66 , where @xmath67 is the coordinate along a vertical axis perpendicular to the substrate plane , whose origin is taken to be at the top of the stack .
these two quantities are computed using a recursive relation between the electric fields at the boundaries , completed by the related initialization condition @xmath68\thinspace\vec{\mathcal{e}}_{j}\\ & \vec{\mathcal{e}}_{0}=(1+r)\vec{\mathcal{e}}_{0}^+ \end{aligned } \right.\ ] ] and extended in the thickness of each layer by @xmath69\thinspace\vec{\mathcal{e}}_{j}\\ \text{with}\enskip\delta_j(z)=\alpha_j(e_j - z)\end{gathered}\ ] ] to quantify and compare the strength and the spectral width of the resonance behavior of these various fabry - perot configurations , we introduce the merit factor @xmath70 @xmath71 where @xmath72 and @xmath73 are the limits of the selected spectral range ( here , @xmath72 = 600 nm and @xmath73 = 800 nm , cf . fig .
[ fig : whitefp ] ) , and @xmath74 is the spectral dependence of the amplitude of the electric field in the middle of the spacer .
the table [ tab : fabryperotconfigurations ] summarizes the results provided by this systematic screening and gives , for each optimized configuration , the stack formula , the refractive index of the h - layer , the value of the merit factor @xmath70 and the ratio @xmath75 between this merit factor and that of the corresponding all - pim cavity ( for all these cavities , the design wavelength @xmath36 is equal to 700 nm ) .
.[tab : fabryperotconfigurations]main features of three optimized symmetric _ white _ fabry - perot configurations [ cols="<,^,^,^ " , ] 2l @xmath76 , @xmath6 = 2.61 ; red line : @xmath76l@xmath62 2l @xmath62l@xmath76 , @xmath6 = 2.19 ; blue line : @xmath77l@xmath62 2l @xmath62l@xmath77 , @xmath6 = 2.37 . ]
figure [ fig : whitefp ] shows the spectral transmittance of these three optimized fabry - perot configurations ; the most attractive wide - band behavior seems provided by the first two listed ( m3 2l m3 and m5 2l m5 ) .
the variations in the normalized square modulus of the electric field within these stacks are presented in fig .
[ fig : pimm32lm3 ] for a standard all - pim m3 2l m3 configuration ( @xmath6 = 2.61 , @xmath7 = 1.480 ) and in fig .
[ fig : nimpimm32lm3 ] for the optimized symmetric nim / pim fabry - perot configuration ( @xmath76 2l @xmath76 ) .
the use of nim layers provides , as expected , a spectacular increase of the spectral range in which resonant behavior is achieved .
= 2.61 , @xmath7 = 1.48 ) . ]
2l @xmath76 fabry - perot configuration ( the refractive indices of the nim layers are opposite to those of the standard configuration ) . ]
we have shown how negative index materials can be used to design a novel type of planar multilayer cavity , which we propose to call a _ white _ fabry - perot cavity , and which is resonant over a very large range of wavelengths in comparison to standard devices .
more detailed analysis demonstrates that the choice of the refractive index @xmath6 of each optimized configuration is quite tolerant , which is advantageous for possible practical implementations .
moreover , by slightly modifying the features of the optimized m5 2l m5 fabry - perot cavity described in the previous section ( centering wavelength 600 nm , modified formula @xmath62l@xmath62l@xmath62 2l @xmath62l@xmath62l@xmath62 , @xmath6 = -2.24 for all high - index nim layers ) and by stacking 3 identical cavities linked by a pim low - index quarter - wavelength layer , we are able to obtain a filter with a nice rectangular profile , a bandwidth of approximately 90 nm , and a high level of rejection throughout the entire remaining spectral range ( approximately -30 db ) . the addition of negative index materials to the data - base of standard thin - film software allows us to define optimized designs for many filtering applications . v. g. veselago , `` the electrodynamics of substances with simultaneously negative values of @xmath78 and @xmath79 , '' sov .
* 10 * , 509514 ( 1968 , russian text 1967 ) .
j. b. pendry , negative refraction makes a perfect lens , phys .
lett . * 85 * , 3966 - 3969 ( 2000 ) .
d. schurig , j. j. mock , b. j. justice , s. a. cummer , j. b. pendry , a. f. starr , d. r. smith , `` metamaterial electromagnetic cloak at microwave frequencies '' . science * 314 * , 977980 ( 2006 ) .
m. lequime , b. gralak , s. guenneau , m. zerrad , and c. amra , `` negative indices and the admittance formalism in multilayer optics , '' in optical interference coatings conference ( optical society of america , 2013 ) , paper tb9 . h. a. macleod , _ thin - film optical filters _ , 4th ed .
( crc press , 2010 ) .
m. lequime , b. gralak , s. guenneau , m. zerrad , and c. amra , `` optical properties of multilayer optics including negative index materials , '' submitted for publication ( 2013 ) j. b. pendry and s. a. ramakrishna , `` focusing light using negative refraction , j. phys .
[ condensed matter ] , * 15 * , 6345 - 6364 ( 2003 ) .
z. h. jiang , s. yun , l. lin , j. a. bossard , d. h. werner , and t. s. mayer , `` tailoring dispersion for broadband low - loss optical metamaterials using deep - subwavelength inclusions , '' scientific reports * 3 * , 01571 ( 2013 ) .
c. amra , ' ' first - order vector theory of bulk scattering in optical multilayers , `` j. opt .
am . a * 10 * , 365 - 374 ( 1993 ) m. lequime and c. amra , ' ' broadband emitters within multilayer micro - cavities : optimization of the light extraction efficiency , " in _ optical interference coatings conference _
( optical society of america , 2010 ) , paper td3 . | the use of negative index materials is highly efficient for tayloring the spectral dispersion properties of a quarter - wavelength bragg mirror and for obtaining a resonant behavior of a multilayer fabry - perot cavity over a very large spectral range .
an optimization method is proposed and validated on some first promising devices . |
lowered excitation energies of the first 1@xmath9 states , fast e1 transitions between the k@xmath10 and ground state bands and high @xmath11 values observed in the even - even @xmath12nd isotopes constitute an evidence that these nuclei belong to the octupole deformation region in lanthanides .
also theory assigns these isotopes to the octupole region @xcite .
this same one should expect for the odd - n neodymium isotopes from this mass region . in these isotopes
one should observe parity doublet bands connected by strong e1 transitions with high @xmath11 moments .
however in ref .
@xcite in which the properties of @xmath13nd isotope have been studied we have obtained very low @xmath11 values for the lowest pair of opposite parity states which should constitute parity doublet in this isotope . in the present work we have undertaken investigations of octupole correlations in @xmath0nd .
excited states in @xmath0nd have been previously studied in the @xmath3-decay of @xmath0pr @xcite , in the neutron capture reaction @xcite and in the transfer reactions @xcite .
recently , high - spin states in @xmath0nd have been studied with the use of the heavy ion induced fusion - fission reaction @xcite .
multipolarities of several @xmath2-transitions in @xmath0nd have been established in refs .
@xcite by internal - conversion electron measurements .
lifetimes of the lowest levels in @xmath0nd have been measured in refs .
experimental data concerning the @xmath0nd nucleus are gathered in a recent compilation @xcite . despite of a rather big body of experimental data on @xmath0nd no firm configuration assignment for excited levels
was proposed and there is no experimental information on octupole correlations in @xmath0nd . in order to obtain a better understanding of the low energy structure and to get information on octupole strength in of @xmath0nd we have measured lifetimes of the excited states in this nucleus using the advanced time - delayed @xmath14 method @xcite .
the experimental methods are briefly described in section [ sec : method ] , while a new level scheme for @xmath0nd and the lifetime results are presented in section [ sec : results ] of this paper . in section [ sec : discussion ]
experimental results are discussed and results of theoretical calculations are presented . in particular ,
results of potential energy calculations on the ( @xmath5,@xmath6 ) plane and theoretical values of @xmath11 moments are shown .
measurements were carried out at the osiris on - line fission - product mass separator at studsvik in sweden @xcite .
excited states in @xmath0nd were populated in the @xmath3-decay of @xmath0pr , which was obtained via a chain of @xmath3-decays starting from the @xmath0cs and @xmath0ba isotopes , @xmath15 @xmath16 the a=147 nuclei were produced in the fission reaction of @xmath17u induced by the thermal neutrons from the r2 - 0 reactor at studsvik . the @xmath17u target consisted of about 1 g of uranium dispersed in graphite .
the a=147 activity , mass separated from other fission products , was deposited onto an aluminized mylar foil in a moving - tape collection system at the center of the experimental setup .
each measuring cycle was divided into eight sequential time - bins , each lasting 40 s. to clean up spectra from the activities coming from the @xmath0pr predecessors our radioactive samples were cumulated during first 135 s of each cycle . then the beam was deflected and the data were collected during the rest of the cycle .
two experiments have been performed . in the first one the multispectra ( mss ) and @xmath3-gated @xmath1 coincidence data
have been collected . in this experiment one lep ge detector with energy resolution fwhm of 0.6 kev at 81 kev , one 30% hpge detector with fwhm of 2.1 kev and one 80% hpge detector with fwhm of 2.9 kev at 1333 kev have been used .
a 3 mm thick ne111a plastic scintillator was used as a @xmath3-detector .
about 1.2@xmath1810@xmath19 double coincident events have been collected . in the second experiment lifetime measurements
have been performed using the advanced time - delayed @xmath20(t ) method @xcite . in this method
the fast timing information was derived from coincidences between fast - response @xmath3- and baf@xmath21 @xmath2-detectors , while an additional coincidence with a ge @xmath2-detector was used to select the desired @xmath2-cascade . in this experiment
the setup consisted of one baf@xmath21 detector , one hpge detector with efficiency of 30% and fwhm of 2.1 kev at 1333 kev , and one @xmath3-detector .
the latter was a 3 mm thick @xmath22e ne111a plastic detector to ensure almost constant , independent of the @xmath3-particle energy , time response of the fast timing system .
about 2.2@xmath1810@xmath23 @xmath3-ge - baf@xmath21(t ) coincident events have been collected .
several sets of the @xmath3-gated coincidence @xmath2-ray spectra from the ge and baf@xmath21 detectors and triple coincidence @xmath4(t ) time - delayed spectra have been sorted in the off - line analysis . gating conditions set on the @xmath3-spectrum
were chosen to keep the time response constant in the whole range of selected @xmath3 particle energies .
examples of the decay curves of @xmath2 rays from the @xmath3 decay of @xmath0pr obtained in the mss measurements.,width=340 ] [ cols="^,^,^,^,^,^,^,^,^,^ " , ] recently new high spin data have been reported for the @xmath0ce nucleus in ref .
the @xmath11 moments calculated from the b(e1)/b(e2 ) values given in this reference are 0.18(1 ) and 0.21(2 ) @xmath24 for the 31/2@xmath9 , 2703.1 kev and 35/2@xmath9 , 3264.0 kev levels , respectively .
these values are comparable to the average @xmath11 values of 0.19(3 ) and 0.21(2 ) @xmath24 obtained from the b(e1)/b(e2 ) values in the neighboring @xmath25ce @xcite and @xmath26ce @xcite isotopes , respectively .
it seems that considered opposite parity states in @xmath27nd and in @xmath13ce may not constitute parity doublets and the main octupole strength is located at higher excitation energies .
it is possible that the presence of odd neutron blocks octupole correlations at low excitation energies in the odd n isotopes . more experimental information on the parity doublets and dipole moments in @xmath0nd and in other odd n isotopes in lanthanides is needed to understand the @xmath11 moments behavior in nuclei from this region .
energy surfaces over ( @xmath28,@xmath29 ) plane were calculated for the lowest single - particle configurations in @xmath0nd by using the macroscopic - microscopic method .
macroscopic energy was calculated using the yukawa plus exponential model @xcite with parameters specified in @xcite .
a deformed woods - saxon potential potential @xcite with the universal set of parameters @xcite was used to calculate the microscopic energy .
we used a @xmath3 parametrization , @xmath3 = ( @xmath30 @xmath31 ) , in which the shape of a nucleus is described by the formula : @xmath32 where @xmath33 is the nuclear radius and @xmath34 is the volume - fixing factor . considered single particle configurations were chosen by blocking the odd neutron on a desired single particle state with a given @xmath35 .
energies were calculated on a grid of 1361367 points defined by deformation values : @xmath36 where the step in each deformation is given in parenthesis .
for each pair ( @xmath37 ) energy was then minimized with respect to @xmath38 @xmath39 @xmath40 .
energy surfaces obtained for four single - particle configurations in @xmath0nd are presented in fig .
[ fig : maps ] .
they show equilibrium deformations : @xmath28 of 0.13 - 0.18 and @xmath41 .
two configurations , @xmath42 and 3/2 ( energy levels no .
42 and 43 ) with shallow octupole minima may be called octupole - soft .
slightly more pronounced , @xmath43 mev deep , octupole minima were obtained for configurations with @xmath42 and 5/2 ( levels no 44 and 45 ; the fermi level in @xmath0nd lies in - between levels no 43 and 44 ) .
energy calculations for two even - a neighbors of @xmath0nd show octupole minima for both @xmath25nd and @xmath26nd ground states .
thus , our calculations suggest that @xmath0nd lies inside the lanthanide octupole collective region .
reduced probabilities of electromagnetic ( em ) transitions between the rotational band built on the one - phonon state and the g.s .
band can be calculated assuming the fixed structure of both the phonon and the collective rotor @xcite . for an operator @xmath44 of the multipolarity @xmath45
one has in odd - a nucleus @xmath46 with @xmath47 the intrinsic spherical component . for dipole transitions between pear - shaped parity - doublet bands
@xmath48^{1/2 } { \hat d}_z$ ] , where the dipole moment @xmath49 . in the strong coupling limit with two octupole minima at @xmath50 , the transition matrix element @xmath51
is calculated as the intrinsic dipole moment at this deformation of equilibrium . in this case ,
@xmath52 , the expectation value of @xmath29 in the first excited state of negative parity , nearly degenerate with the g.s . for shallow minima , closer to oscillation scenario , the degeneracy between parity - doublet bands is shifted . the transition matrix element in the intrinsic frame between the g.s . and the lowest excited state of negative parity @xmath53 may be approximated by a diagonal matrix element of the transition operator in the mean - field state with the deformations @xmath54 fixed as the most probable by the transition density . for a harmonic lowest - lying phonon one has the relation @xmath55 which follows from the one - dimensional harmonic oscillator .
we calculated expectation values of the electric dipole moment in the intrinsic states as a sum of the macroscopic and shell - correction parts , see e.g. @xcite .
the macroscopic part , derived within the droplet model in @xcite , was calculated as in @xcite , i.e. without assuming small deformations @xmath56 .
it turns out ( table [ tab : d0 ] ) that intrinsic dipole moments at equilibrium deformations for two configurations : @xmath42 and 3/2 ( levels no 43 and 44 ) interpolate the values for @xmath57nd .
both macroscopic and shell correction contributions have the same sign , so no reduction of @xmath58 due to their cancellation is possible
. thus , the dipole moments inferred from experiment do not fit the picture of parity doublet bands .
the advanced time - delayed @xmath59(t ) method has been used to measure half - lives of 8 excited states in @xmath0nd .
reduced transition probabilities were obtained for 30 transitions .
twenty - four new @xmath2-lines and 3 new levels have been introduced into the decay scheme of @xmath0nd based on the results of the @xmath3-gated @xmath1 coincidence measurement .
the potential energy surfaces on the ( @xmath5,@xmath6 ) plane and theoretical @xmath11 values suggest the presence of octupole deformation in @xmath0nd at low excitation energies for two configurations with k=1/2 and k=5/2 .
this suggestion is supported by high experimental @xmath11 values in two even - even neighbors of @xmath0nd . for the k=1/2 configuration
we were able to determine only lower limit of the dipole moment , @xmath70.02 e@xmath8 .
however non - observation of the k=5/2 parity doublet band may suggest that in @xmath0nd the @xmath11 values are as low as observed in other odd n isotopes at low excitation energies .
probably strong octupole correlations should be searched at higher excitation energies in the odd n isotopes from the lanthanides octupole region .
one of us ( e.r . ) would like to thank the osiris group for their generous hospitality and for financial support during her stay in studsvik .
this work was partially supported by narodowe centrum nauki under grant no .
2011/01/b / st2/05131 .
m.k . and j.s .
were co - financed by lea copigal funds .
venkova , m .-
porquet , a. astier , i. deloncle , p. petkov , a. prvost , f. azaiez , a. bogachev , a. buta , d. curien , o. dorvaux , g. duch@xmath60ne , j. durell , b.j.p .
gall , m. houry , f. khalfallah , r. lucas , m. meyer , i. piqueras , n. redon , a. roach , m. rousseau , o. stzowski , and ch .
theisen , eur .
j. a * 26 * , 315 ( 2005 ) .
li , s.j .
zhu , j.h .
hamilton , e.h .
wang , a.v .
ramayya , y.j .
chen , j.k .
hwang , j. ranger , s.h .
liu , z.g .
xiao , y. huang , z. zhang , y.x .
luo , j.o .
rasmussen , i.y .
lee , g.m .
ter - akopian , yu.ts .
oganessian , and w.c .
ma , phys .
c * 90 * , 047303(2014 ) . | properties of excited states in @xmath0nd have been studied with multispectra and @xmath1 coincidence measurements .
twenty - four new @xmath2-lines and three new levels have been introduced into the level scheme of @xmath0nd .
lifetimes of eight excited levels in @xmath0nd , populated in the @xmath3 decay of @xmath0pr , have been measured using the advanced time - delayed @xmath4(t ) method . reduced transition probabilities have been determined for 30 @xmath2-transitions in @xmath0nd .
potential energy surfaces on the ( @xmath5,@xmath6 ) plane calculated for @xmath0nd using the strutinsky method predict two single quasiparticle configurations with nonzero octupole deformation , with k=1/2 and k=5/2 .
we do not observe parity doublet bands with k=5/2 .
for pair of opposite parity bands that could form the k=1/2 parity doublet we were able only to determine lower limit of the dipole moment , @xmath70.02 e@xmath8 . |
materials placed in a close vicinity to each other modify the modes of the electromagnetic field .
this results in a change of the vacuum energy , which eventually manifests itself as a net force known as the casimir force @xcite .
the casimir force has been the subject of a number of experimental investigations at object separations ranging from tens of nanometers to a few micrometers .
starting with the experiments by lamoreaux @xcite and mohideen @xcite , the casimir effect has experienced an enormous increase in experimental activities in recent years @xcite .
theoretical approaches to the casimir force are usually built on an effective medium description of the interacting materials .
within such an approximation , the local details of the materials microstructure are neglected and the objects are described by macroscopic , spatially - independent dielectric constants . while the effective medium description is in general quite satisfactory for describing dense materials that indeed look homogenous at the typical scales of the casimir force ,
this is not necessarily the case for strongly heterogeneous ( `` disordered '' ) media that are made of many constituting elements ( `` scatterers '' ) well separated from one another .
examples of such heterogeneous systems include nanoporous materials @xcite , clouds of cold atoms @xcite and , in a slightly different context , corrugated surfaces @xcite . from a theoretical viewpoint ,
interaction phenomena involving strongly heterogeneous materials have been little studied .
seminal works on that subject considered the thermal casimir interaction between slabs made of piled layers separated from random distances ( one - dimensional disorder ) @xcite .
the question of disorder was also addressed recently @xcite in the context of the casimir - polder ( cp ) interaction @xcite between a sphere and a plate @xcite . in a recent work finally , the cp interaction between a dielectric sphere ( or an atom ) and a three - dimensional disordered dielectric material was also investigated @xcite .
this is the scenario we consider in the present paper .
when a probe sphere or an atom interacts with a spatially heterogeneous material such as a semi - infinite disordered medium , the cp potential naturally fluctuates in space . in other words ,
the casimir interaction depends on the specific statistical realization of the disorder . a shared conclusion of refs .
@xcite is that when the two objects are far enough from each other , precisely when the distance between them is large compared to the typical separation between two heterogeneities , the value of the casimir potential from a realization to another is well captured by its configuration average , which coincides with the prediction of the effective medium prescription . in strong contrast , at smaller distances fluctuations of the potential become larger than its mean , which is consequently no longer representative . in practice , this conclusion is crucial for measurements of quantum reflection @xcite , and more generally for any measurement of the casimir force involving heterogeneous materials . in our previous work @xcite
, we developed an exact mathematical treatment of the fluctuations of the cp interaction between a dielectric sphere and a dilute disordered dielectric medium , and applied it to the calculation of the mean value of the cp potential and of its standard deviation . in this paper , we consider the same geometry ( recalled in sec . [ framework ] ) , for which we perform extensive numerical simulations of the cp potential .
the results of these simulations confirm the predictions of @xcite ( sec .
[ numerics_sec ] ) , and additionally allow us to compute the full probability distribution of the cp potential which , for a given distribution of the scatterers , does not depend on the microscopic properties of the latter . in a second time ( sec . [ theory_sec ] ) , we present a simple statistical model based on a pairwise summation of the individual contributions of the scatterers , and confront it with the simulations . concluding remarks are collected in sec .
[ conclusion_sec ] .
we address the cp interaction between a probe dielectric sphere ( placed in vacuum ) of static polarizability @xmath0 ( here and in the following , polarizabilities are expressed in si units divided by @xmath1 ) and a semi - infinite , three - dimensional disordered medium consisting of a collection of many scatterers , as illustrated in fig .
[ scheme ] .
we denote by @xmath2 the distance between the sphere and the surface of the disordered medium . for definiteness , in this paper
we restrict our discussion to the retarded regime of the casimir interaction where @xmath2 much exceeds the resonance wavelength @xmath3 of the probe sphere ( the treatment of the opposite limit @xmath4 is analogous ) .
scatterers are also modeled by dielectric spheres of size @xmath5 and of static polarizability @xmath6 . throughout the paper
, we assume that they are uniformly distributed in space with density @xmath7 , and we consider the case of a dilute disordered medium , such that the average distance between the scattering spheres is larger than their size , @xmath8 .
this is the typical regime where the statistical fluctuations of the cp potential are the largest @xcite . in the opposite limit @xmath9 of scatterers very close to each other , the approach developed below
does not apply but we expect the statistics of the casimir - polder potential to be approximately captured by a model where the atom interacts with a rough surface @xcite .
( color online ) we consider the casimir - polder interaction between a dielectric sphere ( placed in vacuum ) and a semi - infinite disordered medium .
the disordered medium consists of a collection of dielectric spheres ( size @xmath5 , density @xmath7 ) whose positions are uniformly distributed in space . ] in @xcite , the question of fluctuations in the limit @xmath10 was tackled with the help of a statistical description of the disordered material , in which the cp potential @xmath11 becomes a random variable .
its mean , @xmath12 , and its variance , @xmath13 , were calculated from an exact treatment of radiation - matter interaction , based on the scattering approach to casimir forces @xcite combined with a diagrammatic description of radiation scattering off the disordered medium @xcite . in the limit @xmath14 , the following expression for the mean
was found @xcite : @xmath15 where @xmath16 is the casimir potential between the probe sphere and a perfect mirror .
as announced , the result ( [ ubar_theo ] ) coincides with the prediction of an effective medium description where the probe sphere interacts with an homogeneous surface of relative permittivity @xmath17 .
the amplitude of fluctuations , quantified by the ratio @xmath18 of the standard deviation of the cp potential and its mean , was found to be ( for @xmath14 ) : @xmath19 with @xmath20 .
equation ( [ gamma_theo ] ) indicates that @xmath12 , i.e. the prediction of the effective medium theory , is well representative of @xmath11 only when @xmath21 . at smaller scales , @xmath18 becomes larger than unity and @xmath12 no longer provides a trustful estimation of the interaction .
we now propose to investigate the statistical properties of the cp potential from exact numerical simulations in the geometry of fig .
[ scheme ] . for this purpose , we proceed as follows .
we generate an ensemble of @xmath22 dielectric spheres of radius @xmath5 and frequency - dependent permittivity @xmath23 , uniformly distributed in a cube of side @xmath24 .
this system constitutes a disordered medium of average density @xmath25 .
an additional probe sphere is placed above this cube , at a distance @xmath2 to the center of one face , as in fig .
[ scheme ] . denoting by @xmath26 the total , internal casimir energy between the @xmath27 spheres @xcite ,
the cp interaction @xmath11 is by definition @xmath28 the strategy thus consists in calculating the interaction energy as the difference between the internal energies of @xmath27 and @xmath22 spheres . within the scattering formalism
@xcite , the total casimir energy between @xmath22 spheres is given by @xcite @xmath29 @xmath30 is a block - square matrix of dimension @xmath22 with the following structure : @xmath31 the diagonal blocks of @xmath30 are the inverse of the spheres reflection operators @xmath32 .
the @xmath33 off - diagonal block of @xmath30 contains the translation operator @xmath34 , which relates an outgoing spherical wave centered on @xmath35 to an incoming spherical wave centered on @xmath36 @xcite .
finally , @xmath37 is the block - diagonal matrix @xmath38 . for the simulations , we express the scattering and translation operators @xmath39 and @xmath34 in a basis of spherical vector waves @xmath40 , with @xmath41 , @xmath42 and @xmath43 . in this basis ,
the matrix elements of @xmath39 are given by the standard mie scattering amplitudes @xcite .
we compute these amplitudes without any approximation , taking into account the full multipole expansion .
finally , we evaluate the matrix elements of @xmath34 using the formalism of ref .
@xcite . making use of the approach described in sec .
[ methodology ] , we compute the cp potential @xmath11 between a dielectric sphere of radius @xmath44 nm and a disordered medium consisting of @xmath45 other , identical spheres with the same radius @xmath5 , uniformly distributed in a cube of side @xmath46 m . for these parameters , the disordered medium is dilute , @xmath47 , and we are effectively describing the geometry of a semi - infinite bulk system as long as @xmath48 . for definiteness , we give to all the spheres the permittivity of silicon : @xmath49\omega_0 ^ 2/(\omega^2-\omega_0 ^ 2+i\gamma\omega)$ ] , where @xmath50 with @xmath51 nm , @xmath52 and @xmath53 @xcite .
figure [ averagez_cp ] displays the absolute value of the disorder - averaged cp potential , @xmath12 , computed with these parameters , for several values of @xmath2 ( red dots ) .
each dot is obtained by generating thousands ( typically between @xmath54 and @xmath55 depending on @xmath2 ) of disorder realizations , computing @xmath11 for each of them and finally averaging the results .
( color online ) absolute value of the mean cp potential between a dielectric sphere and a semi - infinite disordered medium , as a function @xmath2 .
dots are the results of exact numerical simulations and the dashed line is the theoretical prediction ( [ ubar_theo ] ) . ] in fig .
[ averagez_cp ] , we also show the theoretical prediction ( [ ubar_theo ] ) ( dashed line ) , which is in very good agreement with the numerics .
the small disagreement visible at large @xmath2 stems from deviations to the geometry of the semi - infinite medium : when @xmath2 becomes of the order of @xmath56 , the probe sphere starts to be sensitive to the boundaries of the system , and a cross - over toward the sphere - cube geometry is expected .
we also show in fig .
[ varz_cp ] the standard deviation of the cp potential relative to its mean , @xmath18 , as a function of @xmath57 .
red dots are the numerical results , and the dashed curve is the theoretical prediction ( [ gamma_theo ] ) .
( color online ) relative fluctuations of the cp potential as a function of @xmath57 .
dots are the results of exact numerical simulations and the dashed line is the theoretical prediction ( [ gamma_theo ] ) . ]
the agreement between theory and numerics is very good , up to small finite - size effects at large @xmath2 .
as was pointed out in @xcite , the fluctuations of @xmath11 become significant at distances @xmath58 , when @xmath18 becomes larger than unity , see eq .
( [ gamma_theo ] ) and fig .
[ varz_cp ] .
this suggests that at small distances , the mean @xmath12 is no longer representative of @xmath11 . in order to confirm this picture
, we compute the full probability distribution function @xmath59 of the cp potential normalized to its mean , @xmath60 , by constructing histograms of the numerical data . in fig .
[ scattered_u ] , we show as an example a scatter plot of the data obtained for 5500 disorder realizations , for @xmath61 . the associated histogram @xmath59 is displayed in fig .
[ ps_fig ] , together with the histograms corresponding to three other values of the parameter @xmath57 .
( color online ) scatter plot of the cp potential normalized to its mean value , @xmath60 , computed for 5500 disorder realizations , for @xmath61 . ]
( color online ) probability distribution function @xmath62 , for four increasing values of @xmath57 .
histograms ( vertical lines ) are the results of exact numerical simulations , and solid curves are the theoretical prediction , eq . ( [ ps_general ] ) .
the lower - right panel also displays as a dashed curve the gaussian distribution expected in the limit of very large distances , eq .
( [ gaussian_ps ] ) . ] a quick look at the distributions in fig .
[ ps_fig ] forthwith confirms the property already outlined by the analysis of @xmath18 : as the sphere gets closer to the disordered medium , the distribution function becomes more and more peaked around a value @xmath63 , corresponding to a cp potential much smaller than its mean .
in other words , @xmath11 is no longer a self - averaging quantity .
only when @xmath64 does the maximum of the distribution approaches @xmath65 .
such a phenomenon was previously observed in the context of the interaction between plates with one - dimensional disorder @xcite .
we see here that it is a quite general property , not restricted to one - dimensional systems .
to conclude our numerical study , we address the question of the sensitivity of @xmath59 with respect to changes in the microscopic properties of the scatterers . for this purpose ,
we perform additional numerical simulations involving scatterers with a different radius @xmath66 nm and made of a different material with frequency - independent permittivity @xmath67 .
we also set the radius of the probe sphere to @xmath66 nm , but keep the same value of the permittivity of silicon as in the previous section .
( color online ) numerical probability distribution @xmath59 for two different sets of parameters of the scattering spheres , for @xmath68 ( histograms , vertical lines ) .
no visible difference is seen between the two histograms .
this is confirmed by the theoretical prediction ( solid curve ) , eq .
( [ ps_general ] ) , which depends on the single parameter @xmath57 . ] the distribution @xmath59 obtained for these new parameters is shown in the right panel of fig .
[ universality_fig ] , for @xmath68 . for comparison
, we also display the distribution @xmath59 computed with the parameters of the previous section .
no visible difference is seen between the two histograms , which indicates that in the dilute limit @xmath59 is in fact a function of the parameter @xmath57 only . in particular
, the parameter @xmath69 is irrelevant .
this could have been anticipated since in the limit @xmath10 of independent scatterers , @xmath69 enters both @xmath70 and @xmath71 within the same prefactor , which thus cancels out when considering the ratio @xmath60 ( see sec .
[ theory_sec ] for a general proof ) .
this property is in particular fulfilled by the second moment of the distribution , @xmath72 , see eq .
( [ gamma_theo ] ) .
we now develop a simplified statistical description of the cp interaction between a dielectric sphere and a disordered bulk medium , based on a pairwise summation ( pws ) approximation @xcite .
this approximation describes the total cp interaction @xmath11 as a sum of the pair interaction @xmath73 between the probe sphere and each of the @xmath22 scatterers .
it has to be distinguished from the perturbative expansion , as it can in principle be used for non perturbative pair interactions . in the problem
studied in this paper however , the validity of the two approximations is a consequence of the same assumption of a dilute disorder ( @xmath10 ) . as in the numerical simulations ( sec .
[ numerics_sec ] ) , we consider a situation where the distance @xmath2 much exceeds the sphere resonance wavelength @xmath3 ( in the opposite limit @xmath4 , the reasoning follows exactly the same lines ) .
consequently , the interaction potential between the probe sphere and a scatterer located at distance @xmath74 takes the simple form @xmath75 , where @xmath76 is a constant characteristic of the microscopic properties of the two interacting objects .
let us consider a small spherical cap of volume @xmath77 containing @xmath78 scatterers , as illustrated in the left panel of fig .
[ parametrization ] . the elementary cp interaction between them and
the probe sphere is @xmath79 . as in the simulations
, we assume the positions of the scatterers to follow a uniform distribution , so the random variable @xmath78 is poisson distributed .
the @xmath80 cumulant of @xmath78 , @xmath81 , thus fulfills : @xmath82 where @xmath83 is the average density of scatterers . within the pws approximation ,
the cumulant of the elementary cp potential @xmath84 involving @xmath78 scatterers reads : @xmath85 where we have used the property @xmath86 in the first equality , and eq .
( [ kmdn ] ) in the second .
the cumulant of the total cp potential between the probe sphere and the @xmath22 scatterers is finally obtained by using the parametrization @xmath87 , see fig .
[ parametrization ] , and integrating over @xmath74 from @xmath88 to @xmath89 .
( color online ) left : parametrization of the geometry for the statistical approach of sec .
right : typical ( rare ) disorder realization contributing to the lifshitz tail : the atom is above a large region free of scatterers , of size @xmath90 . ]
this yields @xmath91 from which various statistical properties can be deduced , as we now discuss . as a preliminary application of our statistical approach , we propose to re -
derive the predictions ( [ ubar_theo ] ) and ( [ gamma_theo ] ) , previously obtained from an exact treatment of the radiation - matter interaction . for simplicity we assume the scatterers to be spheres of static polarizability @xmath0 , identical to the probe sphere .
the coefficient @xmath76 then describes the large - distance interaction between two identical spheres .
it can be readily evaluated , for instance from the casimir - polder law @xcite for the energy between two atoms @xcite : @xmath92 the mean @xmath12 is by definition the first - order cumulant : @xmath93 , where we have used eq .
( [ cumulants ] ) for @xmath94 . combining this expression with eq .
( [ gammasept ] ) , we recover eq .
( [ ubar_theo ] ) .
the variance @xmath13 is , on the other hand , given by the second - order cumulant , @xmath95 . taking the ratio with @xmath96
, we find @xmath97 which is nothing but eq .
( [ gamma_theo ] ) , with an analytic expression of the coefficient @xmath98 .
let us now derive the probability distribution function @xmath62 that has been studied numerically in sec .
[ ps_numerics ] .
@xmath59 is given by the inverse laplace transform @xmath99 where @xmath100 is greater than the real part of all singularities of @xmath101 .
@xmath102 is the cumulant generating function of @xmath103 , and can be expressed as a power series of the cumulants ( [ cumulants ] ) : @xmath104 making use of eq .
( [ cumulants ] ) and of the relation @xmath105 obtained above , we find : @xmath106,\ ] ] where we have introduced @xmath107 and where @xmath108 is the lower incomplete gamma function . since @xmath101 has no singularities in the complex plane , we can set @xmath109 in eq .
( [ lmoinsun ] ) .
furthermore , we have the property @xmath110 , such that after the substitution @xmath111 , eq . ( [ lmoinsun ] ) simplifies to @xmath112dx.\ ] ] inserting eq .
( [ phit ] ) into this relation , we finally obtain @xmath113 \right\}d\tau.\ ] ] distributions @xmath59 obtained from numerical evaluation of eq .
( [ ps_general ] ) are displayed in fig .
[ ps_fig ] as solid curves on top of the numerical results of sec .
[ numerics_sec ] .
the agreement is excellent for all the values of @xmath57 .
furthermore , we notice that eq .
( [ ps_general ] ) confirms the conclusion drawn from the simulations in sec .
[ universality ] : @xmath59 depends on the parameter @xmath57 only , being completely independent of the microscopic details of the probe sphere and of the scatterers . although the distribution @xmath59 , eq .
( [ ps_general ] ) , has no evident analytic expression , several simple asymptotic limits can be readily examined . _
large-@xmath103 limit _ we describe the limit of large @xmath103 by expanding the term inside the square brackets in eq .
( [ ps_general ] ) up to second order in @xmath114 .
since essentially the values of @xmath115 such that @xmath116 contribute to the integral , this expansion is a good approximation provided @xmath117 .
it results in a gaussian integral which is straightforwardly performed to give : @xmath118.\ ] ] making use of eq .
( [ gamma_theo_pws ] ) , we rewrite eq .
( [ gaussian1 ] ) as @xmath119.\ ] ] at large @xmath103 , @xmath59 is thus simply a ( normalized ) gaussian distribution of mean @xmath71 and variance @xmath120 .
the expansion used to derive eq .
( [ gaussian_ps ] ) being valid as long as @xmath121 , the gaussian shape is a very good approximation of the whole distribution when @xmath122 , which is a direct consequence of the weakness of fluctuations at large distances .
we expect this gaussian distribution to be universal at large distances , as a consequence of the central - limit theorem ( many scatterers contribute to @xmath11 when @xmath122 ) , regardless the nature of the disordered medium .
this conclusion is supported by similar predictions previously made in the context of the thermal casimir effect in one - dimensional disordered media @xcite , as well as in recent studies of the cp interaction involving quasi two - dimensional disordered metals @xcite . for comparison , we show eq .
( [ gaussian_ps ] ) in the lower - right panel of fig .
[ ps_fig ] as a dashed curve , for @xmath123 .
note that in the chosen geometry this limit is difficult to reach in the numerical simulations , because it requires the generation of a significant number of scattering spheres in order to satisfy the condition @xmath124 , while maintaining @xmath125 to avoid finite - size effects .
when @xmath126 , eq .
( [ gaussian_ps ] ) still holds but only in the very far tail of the distribution , @xmath127 .
the physical reason for which we recover a gaussian tail even for small values of @xmath57 is the following .
large values of @xmath103 correspond to particular disorder realizations @xmath128 for which the casimir potential is very large , i.e. for which the density of scatterers @xmath129 underneath the probe atom is very high .
thus , for these specific disorder realizations the condition @xmath130 is effectively fulfilled .
_ moderate values of @xmath103 _ ( color online ) distribution @xmath59 for @xmath131 ( solid red curve ) , together with the approximate form ( [ ps_moderate ] ) at moderate @xmath103 ( dashed curve ) . ] to describe the limit @xmath132 , we expand the term inside the square brackets in eq .
( [ ps_general ] ) for large @xmath115 : @xmath133\right\}.\end{aligned}\ ] ] we then perform a wick rotation @xmath134 , and express both exponential terms @xmath135 and @xmath136 as power series .
this gives @xmath137^n \left[\pi n z^3\gamma(5/7)\left(-x\right)^{2/7}\right]^m.\end{aligned}\ ] ] if we additionally assume @xmath138 , the terms @xmath139 and @xmath140 give the leading contribution to @xmath59 ( the term @xmath141 is purely real and does not contribute ) . keeping only these two terms and computing the remaining integral , we find @xmath142.\ ] ] equation ( [ ps_moderate ] ) holds for @xmath143 .
it is shown in fig .
[ ps_moderates ] as a dashed curve for @xmath131 , together with the exact distribution calculated from eq .
( [ ps_general ] ) ( solid red curve ) . _
small-@xmath103 limit _ we finally consider the low - potential tail @xmath144 of @xmath59 . in order to find an asymptotic expansion in that limit , we come back to eq .
( [ ps_smallnzc ] ) and apply the change of variables @xmath145 .
this yields @xmath146\right\}.\end{aligned}\ ] ] in the limit of very small @xmath103 , we are thus led to evaluate @xmath147,\ ] ] where @xmath148 and @xmath149 .
equation ( [ i_def ] ) naturally calls for the method of steepest descent .
there are five saddle points , solutions of @xmath150
. only one of them , denoted by @xmath151 , turns out to give a nonzero contribution to @xmath59 : @xmath152^{1/4}e^{-i\pi/14}.\ ] ] using cauchy s theorem , we then deform the path of integration to a path coinciding with the path of steepest descent in the vicinity of @xmath151 .
this is achieved by expanding @xmath153 up to second order around @xmath154 and performing the change of variables @xmath155 , leaving us with a gaussian integral whose evaluation leads to : @xmath156,\end{aligned}\ ] ] with prefactors @xmath157 and @xmath158 . the asymptotic form ( [ lifshitz_tail ] )
is completely analogous to the so - called lifshitz tail that describes the band edge of the density of states of disordered conductors in solid - state physics @xcite .
physically , it can be understood from the following qualitative argument .
low values of @xmath103 are achieved for rare disorder realizations where the probe sphere stands above a large region free of scatterers , as illustrated in the right scheme of fig .
[ parametrization ] . since the distribution of scatterers is poissonian ( sec .
[ theory_sec ] ) , the distance between the events of this poisson process follows an exponential distribution .
consequently , the probability to find a large region of size @xmath159 free of scatterers is @xmath160 , where @xmath161 is a numerical constant . in such configuration ,
the casimir potential felt by the atom can be estimated as ( see fig .
[ parametrization ] ) : @xmath162 to leading order in @xmath163 . on the other hand ,
we have seen in sec .
[ mean_var_pws ] that the average casimir potential is @xmath164 .
therefore , for the rare disorder realization displayed in fig . [ parametrization ] , we have @xmath165 , such that @xmath166\nonumber\\ & = & \exp\left(-\dfrac{cnz^3}{s^{3/4}}\right),\end{aligned}\ ] ] which is nothing but the asymptotic form ( [ lifshitz_tail ] ) .
in this paper , we have developed a statistical description of casimir - polder potentials from both a numerical and an analytical perspective .
this approach is well suited for describing the casimir interaction between a simple dielectric object and a strongly heterogenous medium made of a large number of independent constituting elements .
it can be readily extended to other geometries and to heterogeneous media characterized by a more complex statistics involving , for instance , a non - uniform or polydisperse distribution of scatterers . as a first extension of our work , it would be interesting to investigate deviations to the dilute limit @xmath8 where the scatterers can no longer be systematically considered independent .
we expect these deviations to primarily affect the far tails of the distribution @xmath59 .
a second open question concerns the change in the statistics of the casimir - polder potential when the host ( homogeneous ) medium has a dielectric constant differing from unity .
this problem is more difficult to treat since it now involves a surface , which implies multiple reflections inside and outside the medium .
the presence of a host medium may therefore strongly affect the distribution @xmath59 [ in the obvious limit where the dielectric constant of the host medium goes to infinity , one recovers a perfectly reflecting interface and @xmath59 should tend to the dirac function @xmath167 . in practice , the distribution @xmath59 could be experimentally accessed either by moving the sphere over a static disordered medium to record different disorder distributions , or by taking advantage of a brownian motion of the scatterers if the measurement process is fast enough .
indeed , in that case different disorder realizations can be obtained by detecting the casimir force and then letting the scatterers move before carrying out the next measurement .
if the measurement process is slow , the effect of the motion of the scatterers is to average the casimir potential , giving him a value well approximated by eq .
( [ ubar_theo ] ) since doppler shifts have a negligible effect at thermal velocities @xcite . | we explore the statistical properties of the casimir - polder potential between a dielectric sphere and a three - dimensional heterogeneous medium , by means of extensive numerical simulations based on the scattering theory of casimir forces .
the simulations allow us to confirm recent predictions for the mean and standard deviation of the casimir potential , and give us access to its full distribution function in the limit of a dilute distribution of heterogeneities .
these predictions are compared with a simple statistical model based on a pairwise summation of the individual contributions of the constituting elements of the medium . |
the lensed b0218 + 357 @xcite is one of the rapidly growing but still small number of gravitational lens systems with an accurately known time - delay @xcite .
it is therefore a candidate to apply refsdal s method to determine the hubble constant @xmath4 @xcite . besides the time - delay and an at least partial knowledge of the other cosmological parameters ,
only good mass models of the lens are needed to accomplish this task .
no other complicated and potentially incompletely known astrophysics enters the calculation or contributes to the errors .
since the cosmological parameters are believed to be known with sufficient accuracy now , e.g. from the wmap project @xcite , the only significant source of errors lies in the mass models themselves
. the difficulties in constraining the mass distribution of lenses should not be underestimated , but the lens method is still much simpler to apply than classical distance - ladder methods , where errors difficult to estimate can enter at each of the numerous steps . the classical methods still fight with the problem of results which are incompatible within the formal error bars . the hst key project @xcite e.g. obtains a value of @xmath5 ( @xmath6 ) which is often used as a reference .
it should not be forgotten , however , that other groups obtained results which are significantly different , even though the general distance ladder used is very similar .
@xcite determines a value of @xmath7 ( @xmath6 ) , @xcite obtain @xmath8 ( 90% ) .
the differences in the analyses are various , ranging from the correction of selection effects and the cepheid period - luminosity relation to instrumental effects .
the fact that a number of world - expert groups spent such an amount of effort but still do not agree on the final result proves how difficult the problems of distance ladder methods really are .
see the references above for a discussion of possible reasons for the discrepancies .
in addition we refer to @xcite for an extensive discussion of this subject and an additional result of @xmath9 . after finishing a first version of this paper ,
the analysis of the results of the first year of wmap observations have been published .
although @xmath4 is not the primary goal of cmb observations , determinations of this parameter are possible with additional assumptions . assuming an exactly spatially flat universe ( @xmath10 ) and a cosmological constant for the ` dark energy ' ( equation of state parameter @xmath11 ) , @xcite obtain a value of @xmath12 .
for different @xmath13 , the result for @xmath4 can be significantly smaller .
the assumptions of @xmath14 and @xmath10 can only be dropped if additional information from other astronomical fields are included . with the local information from the 2df galaxy redshift survey @xcite , a constraint on @xmath13
is obtained .
wmap alone can constrain @xmath15 only weakly , leading to wide allowed ranges for the density of dark energy or matter . for a determination of all free parameters ,
more astronomical data has to be included as discussed in detail by @xcite .
the constraints for all cosmological parameters obtained from the combination of many data sets are very impressive , but it should be kept in mind that the situation now becomes similar to the distance ladder methods where model assumptions in many astrophysical fields are necessary , making realistic error estimates of the final result extremely difficult .
therefore the more parameters that go into cmb models have accurate independent determination , the better .
this allows the cmb analyses to concentrate on other parameters which can not be determined in any other way .
taking these difficulties into account , we believe that the lens method is superior to most other methods not only in terms of the possible accuracy of @xmath4 but also , and not less important , in terms of estimating the possible errors reliably .
we emphasize that the effort needed is small when compared with large projects like the hst key project or wmap plus 2df survey etc .
the only serious problem when using the lens effect to determine the hubble constant lies in the mass models for the lenses which directly affect the resulting @xmath4 . to be able to constrain a mass model accurately , two conditions should be met .
first , the lens should be simple by which we mean the galaxy should be normal without nearby companions and without a surrounding cluster . only in these cases
can the mass distribution be described with confidence by a small number of parameters .
second , as many observational constraints as possible should be provided by the system .
advantageous is a high number of highly structured images .
extended lensed sources and einstein rings can provide constraints for a wide range of positions and are therefore preferred to point - like multiple images . in the case of more than two images ,
multiple time - delays can also be used as constraints for the mass model .
in addition to these pure lensing constraints , additional information from stellar dynamics or surface photometry of the lens can be included in the modelling .
radio loud lensed sources have the potential to provide a whole set of independent observational constraints additional to those provided by optical observations .
flux ratios of the images are less affected by microlensing and extinction at radio wavelengths than optical ones .
radio observations also have the advantage that they can reach much higher levels of resolution than in the optical .
much smaller substructures in the images can therefore be detected and used for the lens models .
the lens b0218 + 357 , discovered in the jvas survey @xcite , meets most of the criteria for a ` golden lens ' which allows an accurate determination of @xmath4 .
the lens is an isolated single galaxy , and no field galaxies nearby contribute to the lensing potential significantly @xcite .
the lens shows only two images of the core of the source , but substructure of the images can be used as further constraint .
the main bonus is the presence of an einstein ring which shows a lot of substructure in detailed radio maps @xcite .
this paper deals with utilizing the information from the ring to constrain the mass models and to determine @xmath4 .
b0218 + 357 has only one disadvantage : it is the system with the smallest image separation known , which makes useful direct measurements of the lensing galaxy s position relative to the lensed images extremely difficult .
since this parameter is of fundamental importance for the determination of @xmath4 , no reliable results have been possible so far .
@xcite present a galaxy position which may be close to the correct value , but their error bars are highly underestimated as was already stated by @xcite .
in this paper we will first discuss ` classical ' lens models which only use parametrized data of the compact images to constrain the models .
we will see that with this approach the constraints are not sufficient to determine the galaxy position with any accuracy . to be able to determine the hubble constant
, other information has to be included which is naturally provided by the highly structured radio ring .
the best method available to utilize this information is lensclean@xcite . because the method in its original form has serious shortcomings which prevent its use in a system like b0218 + 357 , where the dynamic range between the compact images and the ring is very high , we had to improve the algorithm considerably .
the lensclean method itself is discussed in paper i @xcite while the results are presented here ( paper ii ) .
we will use the improved version of lensclean to determine the galaxy position for isothermal models with an accuracy that is sufficient to achieve a result for the hubble constant which is competitive with other methods but avoids their possible systematic errors .
possible deviations from isothermal mass distributions , which are the main source of systematic error when combining the results from many lenses , will be discussed by presenting a preliminary analysis of vlbi observations of the substructure of the images .
these will allow us to constrain the radial mass profile for power - law models .
we will learn that deviations from isothermal mass distributions do _ not _ play a major role in b0218 + 357 , because the constraints are already quite strong and in addition the effect on @xmath4 is in this special case , where the galaxy position is determined indirectly with lensclean , much weaker than in most other systems .
we nevertheless discuss future work which will improve the results even more . as a secondary result we will use our best lens models and newly developed methods to produce maps of the source plane as it would look like without the action of the lens .
we will show the source as well as its appearance in the lensed image plane on scales from milli - arcseconds to arcseconds .
finally we use lensclean to shed some light on the question of scattering induced effective ` extinction ' in the a component of b0218 + 357 by comparing results for data sets of different frequencies .
this paper ii as well as paper i are condensed versions of major parts of @xcite . for more details the reader is referred to that work .
in the absence of an accurate position of the lens relative to the position of the lensed images there is no need to use very complicated and general lens models .
the effect of the lens position shows its full impact already with simple standard models .
we therefore restrict most of our work to isothermal elliptical mass models . to introduce the notation
, we write the lens equation as @xmath16 the true source position @xmath17 is the image position @xmath18 shifted by the apparent deflection angle @xmath19 . in static single - plane lenses ,
the latter can be written as gradient of a potential @xmath20 : @xmath21 we use this potential to parametrize the mass models .
general power - law models have @xmath22 the radial power - law index is @xmath3 in the isothermal case .
the arbitrary azimuthal function @xmath23 describes the asymmetry .
mass distributions _ can be described by the formalism presented in @xcite or @xcite .
we prefer the approximation of elliptical _ potentials _
@xcite , which have practical advantages over the true elliptical mass distributions , because the inversion of the lens equation is possible analytically for elliptical potentials . for sufficiently small ellipticities ,
both approaches are equivalent .
a singular elliptical power - law potential ( ` siep ' in the isothermal case ) is given by @xmath24 for small ellipticities
@xmath25 and moderate @xmath26 , this model corresponds to an elliptical mass distribution with axial ratio @xmath27 , or @xmath28 for isothermal models .
we do not include external perturbations , because the external shear and convergence introduced by field galaxies and large scale structure are expected to be very small .
@xcite estimate them to be of the order @xmath29 to @xmath30 , which is negligible in our context .
the time - delay @xmath31 between images @xmath32 and @xmath33 is generally related to the potential and the apparent deflection angle in the following way : @xmath34 here @xmath35 is the redshift of the lens and @xmath36 , @xmath37 and @xmath38 are the normalized angular size distances to the lens , source and from lens to source , respectively .
isothermal models have an interesting analytical property , first described by @xcite . for fixed positions of the images relative to the mass centre ,
the time - delay for given @xmath4 is uniquely determined and does not depend on any other parameters of the model , as long as the images are fitted exactly . in our case
this means that , given the time - delay , @xmath4 for isothermal models can be calculated directly from the lens position @xmath39 using the equation @xmath40 the dependence of @xmath41 on @xmath39 is linear .
these simple relations show the importance of measuring the lens position directly .
once this position is known with sufficient accuracy , no explicit model fits are necessary , but the hubble constant can be derived directly from the positions of the images and the lens .
the resulting hubble constant for isothermal models using the time - delay of @xmath42 @xcite is shown in fig .
[ fig : h0 ] as a function of @xmath39 .
these values are calculated for a low - density flat universe ( @xmath43 , @xmath44 ) .
they would be smaller by 5.5 per cent for an einstein - de sitter universe ( eds ) .
we use a lens redshift of @xmath45 ( @xcite ; and others ) and a source redshift of @xmath46 @xcite .
the older value of @xmath47 @xcite would lead to a 4.2 per cent smaller hubble constant .
included in the plot are estimates for the lens position from different sources .
the optical positions from @xcite are clearly not accurate enough .
the centre of the ring is relatively well defined , but it can deviate from the position of the mass centre by 003 or even more , depending on the ellipticity of the lens and the size and shape of the source .
we will see later that the error bars of @xcite are underestimated significantly so that their result for the lens position can not be used .
the alternative time - delay of @xmath48 ( @xmath49 ) from @xcite is compatible with the result from @xcite which we used for our work because of the smaller error bars .
the result for @xmath4 obtained from the @xcite time - delay would be 4 per cent higher .
our first modelling attempts for b0218 + 357 followed the classical route of using only parameters of the two compact components as constraints .
this is the same approach as presented in @xcite
. we will learn that it is not possible to determine uniquely the lens position with this information .
the results shown in this section were obtained using isothermal models because the ( small , see below ) deviations from isothermality are not of importance as long as the galaxy position is not known with high accuracy . once this parameter is determined with lensclean , constraints on these deviations as well as the impact on @xmath4 will be discussed in sec .
[ sec : non - iso ] .
all coordinates in this paper are measured with respect to the a image .
the @xmath50 coordinate is measured eastwards ( increasing right ascension ) , @xmath51 is measured northwards .
table [ tab : pos ] shows a compilation of known relative positions of the a and b images as well as of the subcomponents 1 and 2 which were revealed by vlba observations @xcite .
there are significant differences in the given positions , but they are too small to affect the results seriously . for the b@xmath52a separation , we used the 15ghz vlba positions for the core components a1 and b1 with their formal accuracy .
the relative positions of the _ _
sub__components offer an additional possibility to constrain the lens models . for the modelling we assumed an accuracy of 0.1mas as in @xcite and made independent calculations for the data from @xcite and @xcite .
the difference between the two sets of positions is of the same order as the assumed error bar .
the new vlbi data presented in @xcite show , in addition to the already known two close subcomponents , the inner jet in both images with subcomponents at separations of up to ca .
10mas which could potentially provide additional constraints .
these data are not used for our classical modelling because the accuracy of parametrized model fits to the additional components are influenced significantly by the underlying smooth emission of the jet .
the true errors of such fits are expected to be much higher than formal estimates so that we decided to exploit this new data set only later with lensclean .
rl@ cd3.2d3.2c & & & ( sub)comp + 5ghz & evn & accuracy ] & 308.5 & 130.3 & b@xmath52a + 8.4ghz & vlbi & & 309.00 & 127.30 & b1@xmath52a1 + & & & 309.32 & 126.37 & b2@xmath52a2 + 15ghz & vlba & accuracy ] & 309.2 & 127.4 & b1@xmath52a1 + & & & 309.6 & 126.6 & b2@xmath52a2 + 15ghz & vla & & 310.56 & 127.11 & b@xmath52a + optical & hst & & 307 & 126 & b@xmath52a + 8.4ghz & vlbi & & 1.18 & 0.87 & a2@xmath52a1 + & & & 1.50 & -0.06 & b2@xmath52b1 + 15ghz & vlba & & 1.072 & 0.868 & a2@xmath52a1 + & & & 1.470 & 0.000 & b2@xmath52b1 the magnification ratio of the two compact images is more difficult to obtain .
several effects can influence the results .
one problem is scattering in the lensing galaxy which seems to decrease the flux density of a. this effect is stronger for lower frequencies and ( with regard to the fluxes ) thought to be almost absent at 15ghz .
secondly , flux densities of compact images embedded in a smooth surface brightness background can not be determined unambiguously .
we always have to expect an uncertainty of about the background surface brightness integrated over the beam .
this error should be smaller than 10 per cent for the lower resolution data sets at lower frequencies and much less than that in other cases .
finally we observe the source at different epochs in the two images because of the time - delay . from measured light curves and the known time - delay @xcite , the typical error of this effect
is estimated to be less than 5 per cent .
table [ tab : flux ] shows flux density ratios for different frequencies , epochs and resolutions .
there seems to be a systematic increase in the ratio with increasing frequency ( discounting optical ) , which hints at scattering .
also the spread in values for 8.4ghz and above is relatively small .
for both these reasons , and because there are values with the effects of delay removed , we chose to use a value of 3.75 which is compatible with most of the measurements at higher frequencies
. please note that the results depend only very weakly on small changes of this ratio .
for the lensclean work presented below , no explicit measurement of the flux density ratio is required .
resolution effects and contamination by the ring are then taken into account automatically .
d1@ll@ ccd1.3@l & & epoch & + 1.7&ghz & vlbi & & 1992 jun 19 & 2.62 + 5&ghz & merlin & & 1991 aug 26 & 2.976 + 5&ghz & merlin & & 1992 jan 13 & 3.23 + 5&ghz & merlin & & 1992 mar 27 & 3.35 + 5&ghz & evn & & 1990 nov 19 & 3.185 + 8.4&ghz & vla & & 1991 aug 1 & 3.247 + 8.4&ghz & vla & & 1996/1997 & 3.57 & @xmath53 + 8.4&ghz & vla & & 1996/1997 & 3.2 & @xmath54 + 8.4&ghz & vlbi & & 1995 may 9 & 3.18 & @xmath55 + & & & & & 3.72 & @xmath56 + 15&ghz & vla & & 1991 aug 1 & 3.690 + 15&ghz & vla & & 1992 nov 18 & 3.79 + 15&ghz & vla & & 1996/1997 & 3.73 & @xmath53 + 15&ghz & vla & & 1996/1997 & 4.3 & @xmath57 + 15&ghz & vlba & & 1994 oct 3 & 3.623 & @xmath58 + 22&ghz & vla & & 1991 aug 1 & 3.636 + & hst & & & 0.14 & the subcomponents are only marginally resolved and even the formal uncertainties of fitted gaussians are too high to serve as additional constraints .
furthermore , the apparent sizes and shapes might be influenced by scatter broadening in the interstellar medium of the lensing galaxy , which is known to be gas rich @xcite .
we nevertheless tried to include the shapes of the subcomponents into the modelling , without any gain in accuracy of @xmath4 .
the actual observational data as given by @xcite and @xcite are shown in table [ tab : subcomp shapes ] .
the fit of the best source shapes for each lens model is performed analytically with a cartesian parametrization of the ellipses .
mathematical details can be found in appendix [ sec : app shapes ] .
.shapes of the subcomponents [ cols="^,^,^,^,>,>",options="header " , ] the hubble constant @xmath4 can now be determined from the distances and from the parameter @xmath41 given by the lens model . for the final result we only use the naturally weighted data since the uncertainties are smaller here . for an eds world model
, the result is @xmath59 . for a low - density flat standard model with @xmath43 and @xmath44
, it increases by about 6 per cent : . ]
@xmath60 the error bars are formal @xmath49 limits which include the uncertainty of the lens model ( 6 per cent , contributed almost exclusively by the uncertainty of the lens position ) and the time - delay ( 4 per cent ) . we stress that they do not include possible systematic errors which might result if the true mass distribution can not be described by our isothermal models ( siep ) . as we show below
, no large effects from this are expected .
using the time - delay from @xcite , our result becomes @xmath61 with the largest error contribution ( 15 per cent ) coming from the time - delay .
standard quintessence models ( e.g. * ? ? ?
* ; * ? ? ?
* ) have parameters between the eds and the low - density @xmath62 world models and would lead to hubble constants between the two extremes . for @xmath63 , the result would be indistinguishable from the eds case ( equivalent to @xmath64 ) while for @xmath65 the hubble constant would be 2.1 per cent higher than for eds .
the cosmological constant corresponds to @xmath11 .
it is a well known fact the the radial mass distribution is the most important source of uncertainties in the @xmath4@xmath66 relation for lens systems where the lens position is known accurately ( see e.g. references in the introduction of * ? ? ?
* ) . for power - law models a simple scaling law of @xmath67 is valid in many cases @xcite while in quadruple lenses
an even stronger dependence of @xmath68 was found under certain circumstances by @xcite for generalized power - law models following eq . .
as a generalization of the siep models , we use elliptical power - law potentials which ( if not too far from isothermal and not too elliptical ) are a good approximation to elliptical power - law mass distributions .
the two families of models differ only in the special form of the azimuthal function @xmath23 .
we will see that deviations from isothermality are very small so that our lenscleanresults for isothermal models can still be used as a first approximation .
the deviations can thus be treated as a small perturbation .
the subcomponents of both images are oriented more or less radially with respect to the expected lens position .
this has the disadvantage that their positions can not be used to constrain the galaxy position for isothermal models as shown before . for non - isothermal models ,
on the other hand , this insensitivity can be used to determine constraints for the radial mass profile from these data relatively independent of the true lens position .
if this position is taken to be close to the lensclean result and compatible with the vlbi constraints , @xmath26 can be estimated with some accuracy using the method presented in appendix [ sec : app vec ] . if we use the 8.4ghz vlbi data from @xcite , the result is @xmath69 ( @xmath6 limits ) .
the error bar does not include uncertainties of the galaxy position but the values are relatively independent of @xmath39 anyway . from the 15ghz data from @xcite , a slightly larger value of @xmath70 can be estimated .
@xcite were able to detect the jet with many subcomponents over a length of 10mas in both images with global vlbi observations at 8.4ghz .
a systematic relative stretching of the b jet image by about 10 per cent can be accounted for by a non - isothermal lens model with @xmath71 , in very good agreement with the other results .
we conclude that the lensing galaxy seems to be close to isothermal with a deviation in @xmath26 of only about 4 per cent , but have to keep in mind that these results are only preliminary estimates which are not based on a self - consistent lens model but used the lens position from lensclean with isothermal models to estimate deviations from isothermality .
the errors of this method are expected to be only small ( see also below ) but a combined fit of non - isothermal lens models to the vla and vlbi data should be aimed for in the future . as explained in paper i , lensclean relies on a very robust method for the inversion of the lens equation . using the newly developed lentil method implemented into lensclean , we were now able to perform a few tests with non - isothermal lens models .
we repeated the calculations described before with the 15ghz vla data but used different fixed values of @xmath26 and compared the results afterwards .
we noticed that the best lens position does indeed shift slightly eastwards when increasing @xmath26 , by about 1mas for 1 per cent change of @xmath26 .
see the arrow in fig .
[ fig : vla i ] for an illustration . to our relief , this effect acts on @xmath4 in the opposite direction as the scaling with @xmath72 , which we expect for fixed galaxy positions , so that the two cancel almost completely .
a rather conservative estimate is that the @xmath72 scaling is reduced by at least a factor of two as a result of the shifting lens centre .
this means that for a value of @xmath73 a change of the hubble constant by at most 2 per cent is expected compared to the isothermal models .
this is much smaller than the statistical uncertainties of the result for isothermal models and thus not of concern yet .
when _ increasing _
@xmath26 slightly , starting from the isothermal @xmath3 , the resulting shift of the best lensclean lens position has the tendency to _ decrease _ the optimal @xmath26 in classical fits .
combined fits of both data sets should therefore be possible and lead to very stable results . in the near future we will use lensclean and lentil to obtain more accurate constraints on the radial mass profile from the existing very high - quality vlbi observations at 8.4ghz @xcite . in combination with the existing vla data which are more sensitive to the galaxy position , a self - consistent lens model can be found and accurate error estimates will be possible .
in paper i we discussed a new method to reconstruct source plane maps and superresolved lens plane maps from emission models determined with lensclean .
we present the resulting lens plane maps of b0218 + 357 in fig .
[ fig : lens map ] .
one version ( a ) is restored with the nominal clean beam while the other ( b ) is a superresolved version with anisotropic and varying beams calculated as explained in paper i. the superresolved map is not easy to interpret because of the varying and anisotropic beams .
the resolution is generally higher close to the critical curve i.e. in parts of the ring .
the elliptical shape of the b image is also a result of the local elliptical beams .
the reconstructed source plane map is shown in fig .
[ fig : src map ] .
the left map ( a ) is reconstructed with the source plane beams calculated from eqs . and in paper i. the radial streaks emerging from the region near the caustic are artifacts of the varying and highly anisotropic beams .
the elliptical shape of the core is also a result of the highly eccentric local beam and should not be misinterpreted as a resolved elliptical core . the right map ( b ) was restored with varying but circular beams ( the smallest circles which still cover the local elliptical beams completely ) to make interpretations easier .
unfortunately the greatest part of the lensing magnification and improvement of resolution is lost in this map .
we see that the jet , which by coincidence emerges from the inner core in a direction very similar to the major axis of the elliptical core image in fig .
[ fig : src map](a ) , bends southwards and crosses the caustic of the lens at the bend .
it is this part of the source which forms the ring in the lens plane and which is used as constraint by lensclean . on larger scales ( shown especially well by longer wavelength observations , e.g. the 8ghz vla data in fig .
[ fig : comp ] ) , the jet proceeds in a southern direction only to bend eastwards again at a distance of about @xmath74 .
the jet then seemingly ends in a relatively bright blob @xmath75 south - east of the core .
[ [ sec : comp ] ] the composite colour map + + + + + + + + + + + + + + + + + + + + + + + + in fig . [ fig : comp ] shows b0218 + 357 on different scales both in the lens plane and in the source plane ( reconstructed with lensclean ) .
the main maps showing the outer parts of the jet on arcsec scales are made from a 8.4ghz data set obtained at the vla simultaneously with a global vlbi observation ( * ? ? ?
we show a regular lens plane map and a source plane reconstruction with circular beams .
the insets on the left are made from the 15ghz vla data set that was used for our main lensclean work .
the lens and source plane maps show the same data as fig .
[ fig : lens map](a ) and [ fig : src map](b ) .
we also show the critical curve in the vla lens plane maps and the tangential caustic ( diamond shape ) and cut ( elliptical ) in the vla source plane maps .
the insets on the right ( upper : b , lower : a ) are regular lens plane cleanmaps of a global vlbi data set @xcite .
the source plane map is a projection into the source plane of the a image alone .
the b image is less resolved and would not provide much additional information .
the map was restored from the clean components with the projected clean beam .
5 in @xcite shows projections of images a _ and _ b restored with equal beams to show the similarities .
true lenscleansource plane maps of the vlbi data have not been produced yet .
the magnifications relative to the main maps are 3 for the left insets and 130 for the jet insets on the right .
the scales are the same for lens plane and source plane .
the contour lines start at three times the rms noise and double for each new level .
the surface brightness levels in the lens plane and source plane maps are directly comparable .
the magnified sections are marked by black rectangles in the vla maps .
it is a generally accepted fact that the apparent flux ratio a@xmath76b changes systematically with the frequency at which it is measured . at the high end
the value seems to reach its final value of 3.8 for frequencies @xmath77 . on the other
end the ratio goes down to 2.6 at @xmath78 ( compare table [ tab : flux ] ) . at even lower frequencies it is difficult to separate the compact emission from the ring which becomes more dominant at low frequencies .
there is unambiguous evidence that the ism of the lensing galaxy is very rich and that optical and radio emission are affected to a significant degree @xcite .
@xcite estimated the scattering measure from the broadening of image a relative to b ( taking into account the magnifying effect of the lens ) .
they obtain a value of @xmath79 which is very high compared to typical lines of sight along our galaxy but comparable to lines of sight through the galactic centre .
one possible explanation for the low flux ratios at low frequencies is therefore the action of the very rich ism of the lensing galaxy in front of image a. this can cause an effective extinction of a either by a high amount of scattering or by other physical effects .
the refractive index of an interstellar plasma goes proportional to @xmath80 which would explain the stronger effects for larger @xmath81 .
another possible explanation is a very strong frequency dependence of the source structure which would effectively result in a position shift along the jet for lower frequencies . because the spectrum of the jet is much steeper than the spectrum of the core , such an effect is not implausible .
when the source is shifted in this direction , the situation becomes more symmetric and the magnification ratio decreases . for a sufficiently large shift ( @xmath82 ) the differential magnification gradient could
, in combination with this frequency dependent source position , well explain the observed flux ratio changes .
there are , however , strong arguments against this scenario .
if the shift is as large as required , there should be significant differences in the appearance of the innermost components in vlbi maps at 8.4ghz ( see e.g. * ? ? ?
* ; * ? ? ?
* ) and 15ghz ( see e.g. * ? ? ?
* ) , but these maps look remarkably similar .
both show two strong inner components with similar shapes and compatible relative positions .
their separation of @xmath83 is far too small to explain the demanded shift by different spectral indices .
if the shift is caused by another component which is not seen at 15ghz but becomes stronger at low frequencies , such a component should be detectable on a scale of @xmath84 to @xmath85 in the existing very deep 8.4ghz vlbi observations .
@xcite do indeed detect more jet components but these are far too weak to account for a shift of the required magnitude .
a more exotic alternative scenario would be refraction caused by large scale systematic trends in the column density of the lens galaxy s ism .
to investigate the question of extinction and/or source shifts with lensclean , we compare our best lens models fitted to the 15ghz data which are known to be unaffected by both effects with a merlin data set at 5ghz .
this multi - frequency data set was used to produce the maps in @xcite and is described in detail there .
we used only the merlin part because the vla resolution at 5ghz is too low to be of any help here .
unfortunately , this data set has some disadvantages for lensclean .
while the resolution is better than in the 15ghz vla data , and the flux in the ring is higher due to the lower frequency , it was not possible to obtain reliable direct constraints for the lens model with lensclean .
one problem is the possibility of extinction itself .
another very serious problem is the frequency - dependence of the emission , which is different for the ring and the compact images .
@xcite approximately corrected for this effect by first mapping the three frequencies independently .
the clean components responsible for the compact images were then subtracted from the data so that only the ring remains .
the three data sets were rescaled in magnitude to obtain a consistent total flux density at all frequencies .
finally , the compact components from the central frequency were added back to the complete data set .
this process may introduce some distortions to the lowest and highest frequencies , because it combines data for the ring from these frequencies with data for the compact components from the central frequency .
the combined data were then successively mapped and self - calibrated to obtain the final data set and map .
we used the resulting @xmath86 data set as basis for our computations .
the spectral index correction seems to work very well for making maps , but introduces errors in lensclean which are not fully understood .
this shows especially in incompatible results from the individual three frequencies and the combination of all three . on the other hand
, tests showed that the data can nevertheless be used to fit a subset of the lens model parameters if the other ones are fixed . at most three parameters
can be determined accurately because they are already well defined by the two bright images alone and the ring is only a relatively small correction .
we therefore used a variety of lens models compatible with the 15ghz vla data and fitted the same models to the 5ghz data set allowing for a relative shift of the two data sets and a possible extinction in the a image at 5ghz .
these fits are much more robust than complete lens model fits from this data set alone . in particular
, the results do not depend on which of the frequencies is used .
the combination of all three also leads to the same result . our strategy to determine shift and extinction works like this .
first we took one of the best lens models from lenscleaning the 15ghz data .
then we fixed all parameters but @xmath39 and made a fit with the 5ghz data including a ( for the moment fixed ) extinction in a. the difference between the two values of @xmath39 is then a measurement of the shift in the lens plane because in reality the lens should have the same position at both frequencies but the source may shift .
we then varied the extinction to find the best value by minimizing the residuals .
vla data.,width=302 ] the results for one of the best vla 15ghz lens models are shown in fig .
[ fig : vlarefit ] .
the position of the minimum changes somewhat when other lens models still compatible with the 15ghz data are used .
the most extreme values for the extinction are 15 and 35 per cent ; most probable is a value between 20 and 30 per cent .
the data sets where initially registered to obtain the same position for the a image . the relative shift in the fit ( measured as a shift of the lens centre merlin @xmath52 vla , equivalent to a shift of the source by the same amount with opposite direction ) in the region of
the minimum is negligible .
the fact that the shifts in @xmath50 and @xmath51 have their zeros at almost the same extinction which itself is furthermore compatible with the residual minimum gives strong evidence that the real shift is indeed small and that the brightness of a at 5ghz is reduced by @xmath87 per cent by some propagation effect , possibly scattering . for zero extinction
, the required shift would be about @xmath88 eastwards , but the residual curve seems to be incompatible with this scenario .
it seems as if extinction in a , caused by whatever physical process , accounts for a least most of the frequency dependent flux ratio .
this has an important consequence for further modelling work .
although the ring becomes brighter at lower frequencies , only high frequencies can be used in a simple way to fit lens models .
observations at medium frequencies may on the other hand be used to study the propagation effects in the ism of the lensing galaxy in more detail now that the lens model is known with sufficient accuracy .
b0218 + 357 is one of the most promising lens systems for determining the hubble constant with the method of @xcite .
this method has , compared with other approaches , the advantage that it is a very simple one - step determination and relies on the understanding of only very little astrophysics . in this way the systematic uncertainties can not only be minimized but ( equally important ) also be estimated much better than in distance ladder methods which have their problems with several astrophysical processes at each step .
the only serious uncertainty left is the mass distribution of the lens . in this respect b0218
+ 357 is a close to optimal ` golden lens ' .
the lensing galaxy seems to be a regular isolated galaxy without contributions from a group or cluster nearby .
this not only avoids the inclusions of further parameters to describe the external mass distribution but also allows the assumption that the mass distribution of the galaxy itself is that of an unperturbed smooth ellipsoid .
b0218 + 357 provides a wealth of potential constraints for the mass models in the form of highly accurate image positions , a measurable flux density ratio with probably negligible influence by microlensing , substructure on scales resolved by vlbi and a well resolved and structured einstein ring on the scale of the image separation . without using the ring , ` classical ' lens modelling allows us to tightly constrain the ellipticity and mass of the galaxy as long as its position is known .
the vlbi structure of several subcomponents along the jet in a radial direction relative to the lens allows us to determine the radial mass distribution with an accuracy not possible in most other lenses ( see e.g. * ? ? ? * and references therein ) .
this most important general degeneracy of the lens method can therefore be broken in the case of b0218 + 357 .
the only real disadvantage is the small size ( @xmath89 ) of the system .
b0218 + 357 is indeed the lens with the smallest image separation known to date .
although the images and the lens are detected with optical observations , it was until now not possible to use these observations to accurately measure the galaxy position directly . _
hst_/acs observations obtained recently are currently analysed to allow a first useful direct measurement .
the model lens positions from @xcite are probably not too far from the truth and are indeed consistent with our results but their error bars are seriously underestimated .
we have shown that , without using the ring , no useful estimate of the galaxy position and thus @xmath4 is possible .
previous efforts have not utilized the most striking feature of b0218 + 357 , the beautiful einstein ring shown by radio observations ( e.g. * ? ? ?
doing this is much more difficult than using parameters of compact components because the source itself has to be modelled as well , either implicitly or explicitly .
the method best suited for this task is lensclean which surprisingly has been used only for very few cases before .
this is partly a result of the high numerical demands and partly of serious shortcomings of the original method as it was implemented by @xcite and @xcite . in paper
i we discuss the development of our new improved variant of lensclean .
one of the most important improvements is the implementation of a new unbiased selection of components .
the problem of determining the lens position is almost degenerate because the bright components , which contribute most of the signal , provide no information for this parameter .
it is only the relatively weak and fuzzy ring which can be used as constraint . in other words
the residuals have a very strong dependency on two directions ( defined by the compact components ) of the five dimensional parameter space but change only weakly in the other three directions as a result of the constraints provided by the ring .
finding minima of functions like this is a numerically difficult problem and relies on the very accurate knowledge of the residual function .
any numerical noise produces local minima which make the finding of the global minimum difficult .
special care was therefore needed to produce the best possible results with the given limited computing resources .
the modified lenscleanalgorithm was applied to a 15ghz vla data set resulting in good constraints for all parameters of an _ isothermal _ lens model , including the lens position of @xmath90mas , @xmath91mas relative to the a image .
this model was then used to determine the hubble constant from the time - delay of @xcite to be @xmath92 .
the error bars are @xmath49 confidence limits which for the hubble constant include the error of the lens position ( 6 per cent ) and the time delay ( 4 per cent ) . the accuracy for all other lens model parameters is much higher as long as the position is fixed .
the value for @xmath4 is in agreement with the results from the _ hst _ key project @xcite and the wmap project @xcite but incompatible with the lower distance - ladder results of @xcite , @xcite or @xcite .
recently , a series of papers was published by @xcite , see also @xcite , in which it is claimed that a number of gravitational lens time - delays is compatible with the hubble constant from the _ hst _ key project only if the mass concentration is as compact as the light distribution ( @xmath93 in the picture of power - law models ) .
our result for b0218 + 357 does not confirm this view and does not give rise to a ` new dark matter problem ' since it is significantly higher than other results from lenses .
our isothermal models lead to a value absolutely compatible with the one preferred in @xcite .
one could now in a similar way compare with the lower @xmath4 results of @xcite , @xcite or @xcite .
however , as long as the local determinations do not agree with each other , we consider such an exercise to be of only limited value . rather than using locally measured values for @xmath4 to constrain lens mass distribution ,
we prefer a more direct approach , either using the lens effect itself ( see below ) , or by including additional information .
the latter approach is followed by thy ` lenses structure and dynamics ' survey lsd @xcite .
the general idea is simple : the total mass within the einstein radius is well constrained by lensing , in contrast to the radial mass _
profile_. for the given mass , the stellar velocity dispersion of the lensing galaxy depends strongly on this profile .
the more the mass is concentrated in the centre , the deeper the potential well and the higher the velocity dispersion has to be . with some additional assumptions ,
it is indeed possible to obtain valuable constraints which can then again be included in the lens models to determine @xmath4 . for the important system pg 1115 + 080 , @xcite obtain a power - law index of @xmath94 which increases the otherwise very low estimate of @xmath4 for this system to a value of @xmath95 ( @xmath6 ) .
this steep mass profile is quite unusual and moves the galaxy significantly off the fundamental plane .
nevertheless it is not as steep as the _ light _ profile and the result for @xmath4 correspondingly lies between the isothermal and constant @xmath96 model results of @xcite . for b1608 + 656 ,
@xcite are able to constrain the mass profile of the main lensing galaxy quite well and find that it is consistent with isothermal .
the resulting hubble constant is @xmath97 ( @xmath98 , inconsistent with the very low mean value of @xmath99 obtained by @xcite for isothermal models for a number of lenses , but consistent with our result for b0218 + 357 .
however , b1608 + 656 has the major disadvantage of a second lensing galaxy very close to the primary .
the two galaxies might even be interacting which could cause complicated deviations from usual galaxy mass distributions .
these examples show that the general picture of lensing constraints for @xmath4 and mass profiles is not as consistent as it appeared not long ago ( e.g. * ? ? ?
* ) . in our opinion
the only way to resolve the discrepancies is to try and use the best available constraints for all applicable lens systems individually .
our work on b0218 + 357 shows this for one example , although there are still many open questions .
our alternative way of constraining mass profiles relies on the lens effect alone and has the advantage that additional astrophysics ( galactic dynamics in lsd work ) and corresponding additional model assumptions are not needed , keeping the method simple and clean .
it is easy to understand that lenses showing only two or four images of one compact source can not provide sufficient information to constrain the mass profile tightly , especially since quads usually have all images close to the einstein ring so that they probe the potential at only one radius .
far better suited are systems with extended sources or at least with substructure in the compact images .
b0218 + 357 is a good example which offers both .
using our galaxy position as an estimate , non - isothermal power - law models could be constrained with the vlbi substructure . these are not the final results since no self - consistent fitting of all available data has been performed so far , but the weak dependency of the radial mass exponent @xmath26 on the galaxy positions shows that the estimate is nevertheless quite good .
we learned that the mass distribution in the lens is close to isothermal ( which would be @xmath3 ) but is slightly shallower ( @xmath100 ) .
preliminary calculations with lensclean showed that the shift in the best lens position caused by this small deviation partly compensates for the expected scaling with @xmath72 for a fixed position .
we therefore do not expect very significant changes of the result for @xmath4 from the slight non - isothermality in this system .
a very conservative error estimate disregarding this compensation effect would be a possible 4 per cent error for @xmath4 .
it has to be kept in mind , however , that all results presented so far depend on the assumption of a power - law for the mass profile ( and potential and deflection angle ) . for more general mass models the constraints on @xmath26 are still interesting , but they only measure the _ local _ slope of the profile rather than the global mass profile .
it is expected that power - laws can be used as a very good approximation for more general models for limited ranges of radii , and our model fits confirm this view in the case of b0218 + 357 .
the lensclean results for the lens position presented here are therefore also valid for other mass profiles , but the value for the hubble constant may well have to be modified slightly . the mass models discussed here do not include external shear because it seems to be very small .
the expected true external shear of 12 per cent @xcite would change the result of @xmath4 by about the same amount at most . more difficult to estimate
is the possible influence of differences in the ellipticity of the inner and outer parts of the lensing galaxy which can effectively also act as external shear .
the possible impact of the remaining higher - order aspects of the mass profile degeneracies and the effects of multi - component mass models ( bulge+disc+halo ) are currently under investigation , especially the aspect of what apparent isothermality in the central part of the galaxy , which should be dominated by the luminous matter in the bulge , means for the true mass distribution . in the future
we will avoid the model fitting of vlbi components which are then used as constraints and instead use lensclean itself also on the existing vlbi data that show a wealth of structure in the images ( see fig .
[ fig : comp ] ) . to be able to use non - isothermal models with lensclean
, we developed the new method lentil for the inversion of the lens equation .
with these two methods combined , a simultaneous fit of medium resolution vla data ( sensitive to the lens position ) and the existing high resolution vlbi data of @xcite , which are more sensitive to the radial mass profile , will not only improve the results for all parameters but will , by using self - consistent lens models , also provide realistic error estimates including all uncertainties . to be able to use the 8.4ghz data
set it will be necessary to include the extinction and scatter broadening in the a image in the models .
this is possible because the 15ghz data are only little affected by extinction so that the combination of the two allows estimates of this effect .
since the vlbi constraints for the lens models are mainly given by the positions of subcomponents but not by their size , which is relatively independent of the lens model , the different sizes of components in a and b can be used to obtain better constraints on the scattering measure than the simple estimates from @xcite .
the application of lensclean to the vlbi data will also show whether any significant amount of substructure in the mass distribution of the lens is needed to explain all features of the jet .
the first results from @xcite seem to be compatible with no substructure , but a quantitative analysis can be done in the future . if significant substructure effects are present , it will also be necessary to correct the medium resolution data for them before using lensclean to determine the galaxy position .
clumps of matter close to one of the images could change the magnification significantly which would then mimic a different lens model .
the analysis of the vlbi data will show whether this may be the explanation for the ( compared with some other lenses ) relatively high result for @xmath4 .
lensclean can also be used to study substructure from vlbi observations of other lenses .
it will then be possible to do an analysis similar to e.g. @xcite but without any assumption about the true source structure . to improve the medium resolution side , we recently performed observations of b0218 + 357 at 15ghz using the fibre link between the vla and the vlba telescope pie town .
this combination provides a resolution 1.52 times better than the vla alone .
the long track observation of the total accessible hour angle range of 14 hours also improves the sensitivity which in combination with the improved resolution can reduce both statistical and ( even more important ) systematical errors .
simulations performed before the observations showed that an accuracy of about 12@xmath101 for the lens position should be achievable .
this would in comparison to the existing data be an improvement of a factor of @xmath102 . the remaining uncertainty in @xmath4 from the lens position alone
will then shrink to below 2 per cent .
possible uncertainties caused by calibration errors will also be reduced significantly .
this new data set is currently being calibrated and will be analysed with lenscleansoon , alone and in combination with the vlbi data . as a secondary result of lensclean we presented maps of the brightness distribution of the source as well as improved lens plane maps , both produced with a method newly developed together with lensclean in paper i. this allowed the first view of b0218 + 357 s source as it would be seen without the distortion of the lens , but with improved resolution .
the ring itself is caused by a bending jet which crosses the tangential caustic of the lens . in the future
we will improve these results and also produce source plane polarization maps .
the ring shows an interesting radial polarization pattern @xcite and the data will , when projected to the source plane , give us a very detailed and magnified view of the polarization structure in the jet of the source .
finally we used lensclean to investigate the puzzling changes of the flux density ratio a@xmath76b with frequency . to test the two theories of either scattering induced extinction caused by the ism in the lensing galaxy or of an effective shift of the source with frequency , we compared our best lensclean models , which were fitted to 15ghz data where the fluxes are expected to be unaffected by propagation effects , with merlin data at 5ghz taking into account a possible relative shift and extinction in the 5ghz data .
although the 5ghz data set has some calibration problems , the evidence for a significant extinction ( @xmath103 per cent ) of the a component at 5ghz is very strong .
the relative shift of the source between these two frequencies seems to be negligible .
this work will be extended in the future by comparing medium resolution data sets at different frequencies in order to study the propagation effects in the lensing galaxy in more detail .
observations with merlin at different frequencies will allow us to measure the position dependent faraday rotation and depolarisation , providing invaluable information about the ism of the lensing galaxy . a direct measurement of the galaxy position in b0218 + 357 by
_ hst_/acs observations will be available soon .
the formal uncertainty of this measurement may be larger than the lensclean model constraints but it will be an absolutely independent and therefore complementary direct measurement . the comparison will either confirm the lens models or give information about deviations of the true mass distribution from the relatively simple models .
we will then reach the point were the uncertainty in @xmath4 is no longer dominated by the lens models but by the time - delay uncertainties themselves .
a reanalysis of existing monitoring data @xcite and new monitoring campaigns will therefore allow further improvements .
with all the constraints available now or soon and the simplicity of the possible mass models , b0218 + 357 has the potential to lead to the most robust measurement of @xmath4 of all time - delay lenses .
the authors would like to thank the anonymous referee for a very helpful report and the royal astronomical society for covering the cost of the colour figure .
o.w . was funded by the ` deutsche forschungsgemeinschaft ' , reference no . re 439/261 and 439/264 ; european commission , marie curie training site programme , under contract no .
hpmt - ct-2000 - 00069 and tmr programme , research network contract erbfmrxct96 - 0034 ` ceres ' ; and by the bmbf / dlr verbundforschung under grant 50 or 0208 .
a. d. , browne i. w. a. , helbig p. , koopmans l. v. e. , wilkinson p. n. , perley r. a. , 1999 , , 304 , 349 a. d. , browne i. w. a. , muxlow t. w. b. , wilkinson p. n. , 2001 , , 322 , 821 biggs a. d. , wucknitz o. , porcas r. w. , browne i. w. a. , jackson n. , mao s. , patnaik a. r. , wilkinson p. n. , 2003 , , 338 , 599 i. w. a. , patnaik a. r. , walsh d. , wilkinson p. n. , 1993 , , 263 , l32 c. l. , rupen m. p. , yanny b. , 1993 , , 412 , l59 a. s. , hewitt j. n. , moore c. b. , haarsma d. b. , 2000 , , 545 , 578 cohen j. g. , lawrence c. r. , blandford r. d. , 2002 , , in print , also available as astro - ph/0209457 e. a. , browne i. w. a. , wilkinson p. n. , 1996 , in @xcite , p. 37
j. d. , kochanek c. s. , hewitt j. n. , 1996 , , 464 , 556 w. l. , et al . , 2001 , , 553 , 47 a. , kovner i. , 1993 , , 417 , 450 kemball a. j. , patnaik a. r. , porcas r. w. , 2001 , , 562 , 649 kochanek c. s. , 2002a , , submitted , also available as astro - ph/0204043 kochanek c. s. , 2002b , , 578 , 25 kochanek c. s. , 2003 , , 583 , 49 c. s. , blandford r. d. , lawrence c. r. , narayan r. , 1989 , , 238 , 43 kochanek c. s. , falco e. e. , impey c. , lehr j. , mcleod b. , rix h .- w . , , 2002 , castles website , http://cfa-www.harvard.edu/castles/ c. s. , hewitt j. n. , eds , 1996 , astrophysical applications of gravitational lensing iau symp .
173 c. s. , narayan r. , 1992 , , 401 , 461 kochanek c. s. , schechter p. l. , 2003 , in freedman w. l. , ed . , measuring and modeling the universe , carnegie observatories astrophysics series , vol .
2 . in print , also available as astro - ph/0306040 the hubble constant from gravitational lens time delays l. v. e. , fassnacht c. d. , 1999 , , 527 , 513 l. v. e. , treu t. , 2002 , , 568 , l5 l. v. e. , treu t. , 2003a , astro - ph/0308056 l. v. e. , treu t. , 2003b , , 583 , 606 l. v. e. , treu t. , fassnacht c. d. , blandford r. d. , surpi g. , 2003 , astro - ph/0306216 r. , schneider p. , bartelmann m. , 1994 , , 284 , 285 i. , 1987 , , 312 , 22 c. r. , 1996 , in @xcite , p. 299
j. , et al . , 2000 , , 536 , 584 e. v. , 1988a , , 206 , 175 e. v. , 1988b , , 206 , 190 metcalf r. b. , 2002 , , 580 , 696 j. r. , et al . , 2000 , , 529 , 786 b. r. , saha a. , sandage a. , tammann g. a. , 2000 , , 540 , 634 a. r. , browne i. w. a. , king l. j. , muxlow t. w. b. , walsh d. , wilkinson p. n. , 1993 , , 261 , 435 a. r. , porcas r. w. , 1999 frequency dependent radio structure of the gravitational lens system b0218 + 357 .
156 , p. 247
a. r. , porcas r. w. , browne i. w. a. , 1995 , , 274 , l5 w. j. , et al .
, 2001 , , 327 , 1297 s. , 1964 , , 128 , 307 a. , 1999 , , 527 , 479 d. n. , et al . , 2003 , , 148 , 175 g. a. , reindl b. , thim f. , saha a. , sandage a. , 2002 , in metcalfe n. , shanks t. , eds , a new era in cosmology cepheids , supernovae , @xmath4 , and the age of the universe , also available as astro - ph/0112489 .
283 t. , koopmans l. v. e. , 2002a , , 575 , 87 t. , koopmans l. v. e. , 2002b , , 337 , l6 l. , et al . , 2002 , , 335 , 432 s. , kochanek c. s. , narayan r. , 1996 , , 465 , 64 j. , paczyski b. , 1994 , , 108 , 1156 t. , combes f. , 1995 , , 299 , 382 h. j. , mao s. , keeton c. r. , 2000 , , 544 , 98 h. j. , mao s. , schechter p. l. , 1995 , , 443 , 18 wucknitz o. , 2002a , , 332 , 951 wucknitz o. , 2002b , phd thesis , universitt hamburg , germany , available from the author or from http://www.astro.physik.uni-potsdam.de/~olaf/ wucknitz o. , 2003 , , also available as astro - ph/0312262 ( paper i ) o. , refsdal s. , 2001 , in brainerd t. g. , kochanek c. s. , eds , gravitational lensing : recent progress and future goals , asp conf . series , vol .
237 degeneracies of the radial mass profile in lens models .
applying the lens equation for two nearby positions gives to first order : @xmath104 here @xmath105 denotes the magnification matrix @xmath106 clearly any information about the small scale structure of the images a and b ( flux density ratio , subcomponent positions and shapes ) can only provide constraints for the _ relative _ magnification matrix @xmath107 . it would be possible to use the subcomponents of a and b as images of two independent sources to constrain the models in the usual way .
since the separation of the subcomponents is much smaller than the typical scale of the system , however , these data can only constrain the first derivative of the deflection angle , i.e. the magnification matrix . a different approach to constrain this matrix directly is therefore more appropriate . given a model value for the relative position vector of @xmath108 in the source plane ,
the corresponding vectors in the lens plane are given by eq .
, leading to a contribution to @xmath109 of @xmath110 with the measured lens plane vectors of @xmath111 , @xmath112 and corresponding covariance matrices @xmath113 and @xmath114 .
this equation defines an ordinary linear least squares problem leading to a weighted mean of the backprojected @xmath115 for the solution : @xmath116 with the common polar ellipticity parameters , fits can only be done numerically and are potentially very unstable .
the following cartesian formalism allows a direct analytical fit for the elliptical shape of the source to minimize the deviations from the observed images .
an ellipse with major and minor axis @xmath117 and @xmath118 with direction of the major axis @xmath119 ( mathematical sense ) can be described by the matrix @xmath120 .
for all points @xmath121 located on the ellipse , the following equation holds : @xmath122 { { \mathbfss{e } } } = \begin{pmatrix } r_+ + r_- \cos2\phi & r_- \sin2\phi \\
r_-\sin2\phi & r_+ - r_-\cos2\phi \end{pmatrix } { } \end{gathered}\ ] ] @xmath123 the astronomical position angle ( north through east ) is related by @xmath124 .
transformation between source and lens plane is linear for the ellipticity parameters @xmath125 : @xmath126 we can therefore use the same simple general linear least squares formalism as in appendix [ sec : app vec ] with transformation matrices for the three - dimensional ellipticity vector @xmath127 . in this way
the best source plane ellipticity matrix / vector and the residual contributions can be calculated analytically .
the cartesian approach also avoids singularities and degeneracies for certain shapes , e.g. for circular sources . | b0218 + 357 is one of the most promising systems to determine the hubble constant from time - delays in gravitational lenses .
consisting of two bright images , which are well resolved in vlbi observations , plus one of the most richly structured einstein rings , it potentially provides better constraints for the mass model than most other systems . the main problem left until now was the very poorly determined position of the lensing galaxy . after presenting detailed results from classical lens modelling
, we apply our improved version of the lensclean algorithm which for the first time utilizes the beautiful einstein ring for lens modelling purposes . the primary result using isothermal lens models is a now very well defined lens position of @xmath0mas relative to the a image , which allows the first reliable measurement of the hubble constant from the time - delay of this system .
the result of @xmath1 is very high compared with other lenses .
it is , however , compatible with local estimates from the _ hst _ key project and with wmap results , but less prone to systematic errors .
we furthermore discuss possible changes of these results for different radial mass profiles and find that the final values can not be very different from the isothermal expectations .
the power - law exponent of the potential is constrained by vlbi data of the compact images and the inner jet to be @xmath2 , which confirms that the mass distribution is approximately isothermal ( corresponding to @xmath3 ) , but slightly shallower . the effect on @xmath4
is reduced from the expected 4 per cent decrease by an estimate shift of the best galaxy position of ca .
4mas to at most 2 per cent .
maps of the unlensed source plane produced from the best lensclean brightness model show a typical jet structure and allow us to identify the parts which are distorted by the lens to produce the radio ring .
we also present a composite map which for the first time shows the rich structure of b0218 + 357 on scales ranging from milli - arcseconds to arcseconds , both in the image plane and in the reconstructed source plane .
finally we use a comparison of observations at different frequencies to investigate the question of possible weakening of one of the images by propagation effects and/or source shifts with frequency .
the data clearly favour the model of significant ` extinction ' without noticeable source position shifts .
the technical details of our variant of the lensclean method are presented in the accompanying paper i. quasars : individual : jvas b0218 + 357 gravitational lensing distance scale techniques : interferometric |
in the 1980 s , belavin and drinfeld classified solutions @xmath0 of the classical yang - baxter equation ( cybe ) for simple lie algebras @xmath3 satisfying @xmath4 @xcite .
they proved that all such solutions fall into finitely many continuous families and introduced combinatorial objects to label these families , belavin - drinfeld triples ( see section [ bd ] ) . in 1993 , gerstenhaber , giaquinto , and schack attempted to quantize such solutions for lie algebras @xmath5 as a result , they formulated a conjecture stating that certain explicitly given elements @xmath6 satisfy the quantum yang - baxter equation ( qybe ) and the hecke relation @xcite .
specifically , the conjecture assigns a family of such elements to any belavin - drinfeld triple of type @xmath7 .
this conjecture is stated in section [ ggsss ] .
recently , etingof , schiffmann , and the author found an explicit quantization of all @xmath0-matrices from the belavin - drinfeld list .
they did so by twisting the coproduct in the drinfeld - jimbo quantum group @xmath8 . for @xmath9
, one can evaluate the universal @xmath10-matrix of the twisted @xmath8 in the vector representation of @xmath8 .
this gives an element @xmath11 of @xmath12 which satisfies the qybe and the hecke relation .
this element is presented in section [ ptp ] . in this paper
i show that the elements @xmath11 and @xmath13 from @xcite and @xcite coincide .
this proves the ggs conjecture .
this is done by explicitly expanding the formula for @xmath11 using combinatorial techniques involving belavin - drinfeld triples .
the proof occupies all of section 2 .
note that the ggs conjecture was proved in some special cases ( the cremmer - gervais and orthogonal disjoint cases ) by hodges in @xcite and @xcite .
the ggs conjecture was proved in some additional cases ( orthogonal generalized disjoint ) by the author in @xcite .
also , the disjoint case was completed in @xcite by pavel etingof and the author .
the author actually found the matrix @xmath11 with the help of a computer before the general twist given in @xcite was found .
the matrix was constructed to coincide with the ggs matrix in many cases , and motivated in part the general construction given in @xcite .
see @xcite for details .
also , many steps of this proof were motivated by and checked with computer programs .
let @xmath14 be the standard basis for @xmath15 .
let @xmath16 be the set of simple roots of @xmath17
. we will use the notation @xmath18 .
let @xmath19 denote the inner product on @xmath15 having @xmath20 as an orthonormal basis .
@xcite a _ belavin - drinfeld triple of type @xmath7 _ is a triple@xmath21 where @xmath22 and @xmath23 is a bijection , satisfying two relations : \(a ) @xmath24 preserves the inner product : @xmath25 , @xmath26 .
\(b ) @xmath24 is nilpotent : @xmath27 such that @xmath28 .
let @xmath29 be the lie algebra of complex @xmath30 matrices .
let @xmath31 be the subspace of diagonal matrices .
elements of @xmath15 define linear functions on @xmath32 by @xmath33 .
let @xmath34 be the casimir element inverse to the standard form on @xmath3 .
it is easy to see that @xmath35 , for any @xmath36 .
let @xmath37 be the projection of @xmath38 to @xmath39 . for any belavin - drinfeld triple
, consider the following equations for @xmath40 : @xmath41 s = \frac{1}{2 } \bigl[(\alpha + t \alpha ) \otimes 1\bigr ] p^0.\end{gathered}\ ] ] belavin and drinfeld showed that solutions @xmath42 of the cybe satisfying @xmath43 , up to isomorphism , are given by a discrete datum ( the belavin - drinfeld triple ) and a continuous datum ( a solution @xmath44 of ) .
we now describe this classification . for @xmath45 ,
set @xmath46 .
define @xmath47 .
for any @xmath48 , set @xmath49 ( the set of positive roots of the subalgebra of @xmath17 having @xmath50 as the set of simple roots ) .
in particular we will often use the notation @xmath51 .
we extend @xmath24 additively to a map @xmath52 , i.e. @xmath53 .
whenever @xmath54 for @xmath55 , we say @xmath56 .
clearly @xmath57 is a partial ordering on @xmath58 .
we will also use @xmath59 to denote @xmath56 or @xmath60 .
suppose @xmath61 for @xmath45 and @xmath62 .
then there are two possibilities on how @xmath63 sends @xmath64 to @xmath65 , since @xmath24 is an automorphism of the dynkin diagram .
namely , either @xmath66 and @xmath67 , or @xmath68 and @xmath69 . in the former case , call @xmath63 _ orientation - preserving on @xmath64 _ , and in the latter , _ orientation - reversing on @xmath64_. let @xmath70 now we define @xmath71 ( @xmath72 is the standard solution of the cybe satisfying @xmath73 . ) the element @xmath0 is the solution of the cybe corresponding to the data @xmath74 .
it follows from @xcite that any solution @xmath75 is equivalent to such a solution @xmath0 under an automorphism of @xmath76 .
the ggs conjecture suggests a quantization of the matrix @xmath0 given in , given by a matrix @xmath77 conjectured to satisfy the quantum yang - baxter equation ( qybe ) , @xmath78 , and the hecke relation , @xmath79 .
this may be formulated and justified as follows ( which is more or less the original motivation ) . if we write @xmath80 , where @xmath81 , then we can consider the constraints imposed by the qybe and the hecke relation modulo @xmath82 .
one may easily check that the qybe becomes the cybe for @xmath0 , while the hecke relation becomes the condition @xmath83 . since it is not difficult to see that @xmath84 is symmetric , the unique symmetric choice for @xmath85 is @xmath86 where @xmath87 [ ggs1 ] there exist unique polynomials @xmath88 of the form + @xmath89 such that @xmath90 .
the proof is easy .
define @xmath91 , with the @xmath88 uniquely determined by proposition [ ggs1 ] .
the matrix @xmath13 is called the ggs @xmath10-matrix .
we will use the notation @xmath92 for elements @xmath93 . define the following matrices : @xmath94 where @xmath95 is the standard drinfeld - jimbo solution to the qybe , which is a quantization of @xmath96 .
the matrix @xmath13 equals @xmath97 .
this is a straightforward computation .
we see that @xmath98 , although @xmath99 in general . * `` the ggs conjecture '' @xcite * [ ggs ] the matrix @xmath13 satisfies the qybe and the hecke relation .
[ or0 ] it is sufficient to check the qybe for one value of @xmath100 since the space of solutions to the homogeneous equation corresponding to is exactly the space @xmath101 where @xmath102 is the space of all @xmath103 such that @xmath104 for any @xmath105 . indeed , it is easy to see that @xmath106 implies @xmath107 = 0 $ ] , and it follows that @xmath108 satisfies the qybe iff @xmath13 does , for any @xmath109 .
our formulation is from @xcite , correcting misprints .
the original formulation in @xcite is somewhat different .
we will write @xmath110 to denote the matrix @xmath103 with @xmath111 substituted for @xmath112 .
define @xmath113 where @xmath114 is the transpose of @xmath103 , for @xmath115 .
then , the original form of @xmath13 can be written as follows : @xmath116 we have @xmath117 .
thus , @xmath13 satisfies the hecke relation iff @xmath10 satisfies the hecke relation . in this case
, we have @xmath118 , so @xmath119 , and thus @xmath10 satisfies the qybe iff @xmath13 does .
thus , the two formulations are equivalent . in this section
we give a combinatorial formula for @xmath120 .
first let us introduce some definitions , which will be used in this formula as well as in the proof of the main theorem .
a _ positive @xmath24-pair _ is a pair @xmath121 for @xmath122 .
we define the _ order _ to be @xmath123 .
the set of positive @xmath24-pairs is denoted @xmath124 . in section 2
, we will also define negative @xmath24-pairs , but so far we do nt need them .
if @xmath125 satisfies the property that @xmath126 for some @xmath122 , we say its _ right - passing order _
is @xmath127 , and denote it by @xmath128 . if there is no such @xmath127 , say @xmath129 .
similarly define _ left - passing order _
@xmath130 .
a positive @xmath24-pair @xmath131 is _ right - passed _
if@xmath132 and @xmath133 .
we say that @xmath134 is _ half right - passed _ if @xmath135 .
similarly define the left versions , and we denote the sets of all such @xmath24-pairs by @xmath136 , and @xmath137 ( right - passed , half right - passed , left - passed , half left - passed , respectively ) .
the @xmath138 subscripts indicate positive @xmath24-pairs . for a positive @xmath24-pair @xmath131
, we define @xmath139 to be @xmath140 if @xmath141 , @xmath142 if @xmath143 , and otherwise @xmath144 .
similarly define @xmath145 .
it will be useful to define _ symmetric _ and _ anti - symmetric _ versions : @xmath146 , @xmath147 .
this allows us to state a simple combinatorial formula for @xmath120 : [ fep ] we may rewrite @xmath120 as follows : @xmath148 ( e_\beta \o e_{-\alpha } + e_{-\alpha } \o e_{\beta}).\end{gathered}\ ] ] this is proved in section [ es ] and also follows from the proof of the main theorem , theorem [ mt ] ( see remark [ efr ] for details ) .
[ frg ] @xmath149 is given as follows : @xmath150 + r_{st}\end{gathered}\ ] ] clear .
[ gcge ] for a given @xmath151 , there are exactly @xmath152 triples ( @xmath153 is the euler @xmath153-function ) in which @xmath154 @xcite .
these are called _ generalized cremmer - gervais _ triples .
these are indexed by @xmath155 , where @xmath156 , and given by @xmath157 , @xmath158 , and @xmath159 , where @xmath160 gives the residue modulo @xmath151 in @xmath161 .
for these triples , there is a unique @xmath100 with first component having trace 0 , which is given by @xmath162 , and @xmath163 for @xmath164 ( this is easy to verify directly and is also given in @xcite ) . with this @xmath100
, @xmath13 has a very nice combinatorial formula , which was conjectured by giaquinto and checked in some cases .
we now state and prove this formula .
[ gp ] @xmath13 is given as follows for generalized cremmer - gervais triples : @xmath165.\end{gathered}\ ] ] see appendix a. in @xcite , an explicit quantization is given for any classical @xmath0-matrix described in section [ bd ] .
this is given by a twist of the standard coproduct on the quantum universal enveloping algebra @xmath166 .
in particular , in the @xmath151-dimensional representation , this gives an element @xmath167 so that @xmath168 satisfies the qybe and the hecke relation .
in fact , @xmath169 is triangular , i.e. @xmath170 and @xmath171 .
suppose we are given @xmath24 so that @xmath172 .
define the following matrices : @xmath173 [ esst ] @xcite the element @xmath11 satisfies the qybe and the hecke relation .
now , we state the main theorem of this paper : [ mt ] for any given @xmath24 and @xmath100 , @xmath174 .
this theorem clearly implies the ggs conjecture .
in this section we will introduce combinatorial objects and arguments which are sufficient to prove the combinatorial formula for @xmath120 . to do this ,
we wish to directly expand .
this involves expanding a quadratic expression in terms of the form @xmath175 and @xmath176 for @xmath177 .
most of the monomials in the expansion are zero .
thus , the first step is to restrict our attention to those that are not . in this vein
, we define _ compatible @xmath24-quadruples _ ( definition [ ctd ] ) . to further simplify the formula for @xmath178
, we will have to show that most of these monomials cancel pairwise .
this is accomplished with bijections @xmath179 and @xmath180 ( definitions [ phd]-[pssd ] ) between the corresponding @xmath24-quadruples .
for all of section 2 we will not need to refer to the dimension of the representation ( formerly @xmath151 ) , so we will reuse @xmath151 for other purposes . for any subset @xmath48 ,
define @xmath181 .
let @xmath182 . these will be called _ negative @xmath24-pairs_.
we define @xmath183 , and @xmath184 for @xmath185 . for negative @xmath24-pairs
, we allow the order to be zero , but not for positive ones ! for @xmath186 ,
define @xmath187 if @xmath188 and @xmath189 if @xmath190 .
we define @xmath191 and @xmath192 just as in the positive case , but with components in @xmath24-pairs permuted .
let @xmath193 and @xmath194 .
note that all passed elements must have positive order , so there is nothing new in the case of negative passed @xmath24-pairs . for @xmath195 , set @xmath196 and @xmath197 ( _ lower , higher , _ respectively .
) for @xmath198 , similarly set @xmath199 and @xmath200 .
an element @xmath186 is uniquely given by @xmath201 , and the sign of @xmath103 .
this is obvious . for convenience
, we will often give elements @xmath186 in terms of @xmath201 , and the sign of @xmath103 .
for convenience , we will say that @xmath202 and @xmath203 for @xmath24-pairs @xmath103 . for any pair @xmath204 , define @xmath205 to be @xmath206 .
we will , however , only consider cases where @xmath207 and @xmath208 .
[ ql ] suppose @xmath209 .
then , if @xmath210 , then there is a unique @xmath211 such that @xmath212 .
namely , this is given by componentwise addition of all @xmath213 .
this follows from nilpotency .
it is easy to see that we need only show that @xmath214 for any @xmath215 and @xmath216 .
equivalently , we have to show that the componentwise addition of all @xmath213 is not @xmath217 .
first , we generalize @xmath218 and @xmath219 .
clearly , every @xmath220 can be written as @xmath221 for some @xmath222 .
we say that @xmath223 for @xmath224 if @xmath221 and @xmath225 for nonnegative integers @xmath226 . in this case
, we define @xmath227 .
note that , when @xmath186 , @xmath228 .
now , we set @xmath229 and @xmath230 .
let @xmath231 and let @xmath232 .
we see that componentwise addition of two elements of @xmath233 , if in @xmath234 , yields another element of @xmath233 , where the orders are summed . in particular , this means that , when @xmath235 , for @xmath236 , then @xmath237 . since @xmath238
, it follows that any componentwise sum of elements of @xmath239 , if in @xmath240 , yields a nonzero element of @xmath241 .
this is all we need .
define @xmath242 as follows . when @xmath243 , @xmath244 , for @xmath245 , @xmath246
. otherwise , @xmath247 we now apply these general definitions to the task of setting aside the important subsets of @xmath248 .
these subsets parameterize nonzero terms that arise upon expansion of @xmath178 .
[ ctd ] define @xmath249 and @xmath250 ( _ compatible @xmath24-quadruples of same / opposite sign _ ) by @xmath251 define @xmath252 ( _ increasing , constant , _ and _ decreasing _ in * order * ) by @xmath253 for any @xmath254 , @xmath255 .
if @xmath256 , then it is clear from @xmath254 that @xmath257 , negating in each component .
but then @xmath258 , which is impossible .
define @xmath259 and @xmath260 ( _ increasing _ and _ decreasing _ in * size * ) by @xmath261 @xmath250 and @xmath262 are partitioned into increasing and decreasing quadruples by different criteria ( size and order , respectively ) !
it turns out that these criteria are the useful ones .
now we define _ good _ and _ bad _ subsets of the sets we have defined . @xmath263 and @xmath264 moreover , set @xmath265 and similarly for @xmath266 .
the definitions above are symmetric in the two components of @xmath24-quadruples because @xmath267 iff @xmath268 iff @xmath269 iff @xmath270 in the following definitions , particularly in the orientation - reversing cases , it may not be completely obvious that the maps have the images indicated ( i.e. are well - defined )
. see proposition [ qc ] for well - definition .
[ phd ] we define @xmath271 . we first consider the increasing case .
take @xmath272 and suppose @xmath273 , @xmath274 , @xmath275 , and @xmath276 .
suppose that @xmath277 for @xmath278 .
if @xmath279 ( i.e. @xmath280 is good ) , then set @xmath281 where @xmath282 and @xmath283 have the same sign as @xmath284 and are given by @xmath285 , @xmath286 , @xmath287 , and @xmath288 .
in other words , @xmath282 and @xmath283 are the unique same - sign @xmath24-pairs such that @xmath289 and @xmath290 . in the case that @xmath280 is bad , we simply set @xmath291 .
in the good decreasing case , we take @xmath292 , set @xmath112 and @xmath293 as above , and again let @xmath294 where @xmath295 are the unique @xmath24-pairs with the same sign as @xmath284 such that @xmath289 and @xmath296 . in the bad decreasing case
, we again set @xmath297 .
we define @xmath298 as follows .
take @xmath299 .
if @xmath300 is good , i.e. @xmath301 does not divide @xmath302 , then write @xmath303 + q$ ] for @xmath304 . in the case
@xmath300 is bad , i.e. @xmath301 divides @xmath302 , we again write @xmath303 + q$ ] , this time choosing @xmath305 when @xmath306 and @xmath307 when @xmath308 .
then , we define @xmath309 where @xmath284 has the same sign as @xmath282 and @xmath283 , while @xmath103 has the opposite sign , and @xmath310 , @xmath311 } + \ldots + t^{p(\o(v)-\o(u))})(u_l+v_l)+t^{(p+1)(\o(v)-\o(u ) ) } v_l$ ] , @xmath312 , and @xmath313 . in the case of decreasing quadruples , we begin with @xmath314 and set @xmath315 , with @xmath316 and @xmath112 all defined as above .
[ pssd ] we similarly define @xmath317 and @xmath318 .
suppose @xmath319 .
if @xmath283 is good , i.e. @xmath320 does not divide @xmath321 , we write @xmath322 for @xmath323 .
if @xmath283 is bad , i.e. @xmath320 divides @xmath321 , we set @xmath305 when @xmath324 and @xmath325 when @xmath326 , and again write @xmath327 .
now , we define @xmath328 where @xmath284 has the same sign as @xmath283 , @xmath103 has the opposite sign , and @xmath329 and @xmath330 . for @xmath331 , we define @xmath332 , with @xmath316 and @xmath112 all defined as above .
we had to define two separate maps @xmath180 and @xmath333 because sometimes @xmath334 and @xmath335 intersect , and @xmath333 and @xmath180 do not agree .
we say that a pair @xmath195 reverses orientation if @xmath336 , and in this case , we set @xmath337
. otherwise @xmath103 preserves orientation and @xmath338 . for negative pairs
@xmath339 , we say @xmath103 preserves orientation and @xmath340 if @xmath60 ; otherwise , we set @xmath341 and say that @xmath103 preserves / reverses orientation iff @xmath131 does .
define @xmath342 $ ] as follows : @xmath343 and @xmath344 .
[ nc ] if @xmath345 then @xmath282 preserves orientation . in the event @xmath283
reverses orientation , then @xmath346 . if @xmath272 , then @xmath284 preserves orientation . in the event @xmath103 reverses orientation , @xmath347 . the same results hold considering @xmath348 and @xmath349 , respectively .
in the first case , if @xmath282 reversed orientation , then @xmath350 , thus @xmath351 by nilpotency this is a contradiction . on the other hand ,
when @xmath283 reverses orientation , @xmath352 must reverse orientation , and thus by nilpotency can not be defined on all of @xmath353 .
thus @xmath354 , hence the desired result .
the second case follows easily from nilpotency .
[ qc ] the maps @xmath355 and @xmath356 are inverse to each other .
the maps @xmath357 and @xmath358 are inverse to each other .
finally , the maps @xmath359 and @xmath360 are inverse to each other .
all maps @xmath361 and @xmath333 preserve @xmath362 and @xmath363 .
the fact that the maps are well - defined , inverse to each other , and preserve @xmath362 is easy to see from construction when orientations are preserved ( it helps to draw a picture ) .
also , when orientations are preserved , @xmath363 is trivially preserved .
so it remains to consider orientation - reversing cases .
given @xmath364 , if orientation is reversed in @xmath103 or @xmath284 , it can only be reversed in @xmath103 , and @xmath347 by lemma [ nc ] . in the case
@xmath365 , it follows that @xmath366 , where @xmath282 preserves orientation , @xmath283 reverses orientation , @xmath367 , and @xmath368 .
so @xmath363 and @xmath362 are preserved , and @xmath309 . if , instead , @xmath369 , then @xmath370 , where latexmath:[$|v| = reverses orientation .
again , @xmath363 and @xmath362 are preserved and @xmath328 .
now consider an element @xmath372 in which either @xmath282 or @xmath283 reverses orientation . by lemma [ nc ] , only @xmath283 reverses orientation and @xmath373 .
so , we get @xmath374 where @xmath375 . in this case
, @xmath103 reverses orientation and @xmath284 does not , so @xmath362 and @xmath363 are preserved , and clearly @xmath281 .
finally , suppose @xmath376 reverses orientation . by definition
, this means that @xmath377 reverses orientation on @xmath283 . by nilpotency , @xmath378 , and it follows that @xmath379 .
we then have @xmath380 and @xmath381 .
it follows that @xmath103 reverses orientation , and so @xmath363 and @xmath362 are preserved , and @xmath370 .
the decreasing and left cases follow in exactly the same way as the increasing and right cases .
[ psq ] suppose @xmath272 .
if @xmath281 and not all of @xmath382 preserve orientation , then @xmath383 reverse orientation , @xmath384 preserve orientation , and @xmath385 .
if @xmath370 and not all of @xmath386 preserve orientation , then @xmath383 reverse orientation , @xmath284 preserves orientation , and @xmath385 .
the same facts hold under the assumptions @xmath387 with @xmath294 or @xmath283 .
this follows directly from the argument above .
@xmath388 fix a choice of sign @xmath389 for this proof .
clearly , whenever @xmath390 , @xmath391 , and @xmath392 , then @xmath393 . in this case , @xmath394 , as @xmath395 also has reversed orientation and the same order as @xmath103 and @xmath284 .
it remains only to see that , for any @xmath396 with @xmath397 , there are @xmath398 ways of writing @xmath399 for @xmath400 , and they all are of this form .
the formula follows immediately .
* direct proof of proposition [ fep ] . * set @xmath401 and + @xmath402 , so that @xmath403
. then , using @xmath153 and @xmath404 , @xmath405 equation follows immediately .
[ efr ] the combinatorial formula [ fe ] for @xmath178 also follows from the proof of theorem [ mt ] .
namely , in the proof we actually show that @xmath11 has the form of @xmath13 but we use the combinatorial formula for @xmath178 instead of the original one . on the other hand , since the combinatorial formula is symmetric , and because @xmath11 satisfies the qybe and the hecke relation , @xmath11 must be the unique element satisfying the hypotheses of proposition [ ggs1 ] by the discussion in section [ ggsss ] .
the proof of proposition [ fep ] above is , however , given for pedagogical reasons and because the results used will be needed later . by
_ passing properties _ of a @xmath24-pair we mean information about its left- and right - passing order .
in particular , usually we will be concerned with whether a pair is ( half ) right- or ( half ) left - passed . in this section
, we will list all possible passing properties of compatible increasing and decreasing quadruples , in connection with those properties of their images under @xmath153 or @xmath404 .
these results , lemmas [ pq ] and [ ppp ] , are essential in order to consider quadratic terms which arise in the hecke condition for @xmath13 , which are similar to those in the formula for @xmath178 but include powers of @xmath112 which depend on the passing properties . as a consequence of these results
, one can prove the hecke condition for @xmath13 directly ( see @xcite ) . in order to prove these results
, we will first need to develop some more powerful combinatorial tools and notation regarding belavin - drinfeld triples .
the combinatorics can best be pictured on the dynkin diagram for @xmath17 .
we picture this diagram as the line segment @xmath406 $ ] with integer vertices .
we then picture the positive root @xmath407 , for @xmath408 , as the line segment @xmath409 $ ] . in this context , @xmath410 and @xmath411
can be thought of as subsets of the graph consisting of the union of all the length-1 segments which make them up , and maps @xmath24 are nilpotent graph isomorphisms @xmath412 .
assume @xmath413 and @xmath414 .
we say @xmath415 if @xmath416 ( @xmath407 is to the left of @xmath417 ) . as subcases of this , we say that @xmath418 for @xmath419 , @xmath420 if @xmath421 , and @xmath422 for @xmath423
. similarly define @xmath424 by @xmath425 whenever @xmath426 , and the same for @xmath427 and @xmath428 .
if @xmath429 or @xmath430 , then we say that @xmath431 ( meaning there are orthogonal subsets @xmath432 so that @xmath433 and @xmath434 ) . here and in the sequel we will make frequent use of the following key combinatorial lemma : take @xmath435 .
let @xmath436 be the smallest positive integer such that @xmath437 .
let @xmath438 be the smallest positive integer less than @xmath436 such that @xmath439 , if such an integer exists .
otherwise set @xmath440 .
if @xmath438 is finite and there is a positive integer @xmath441 which is not a multiple of @xmath438 satisfying @xmath442 , then let @xmath443 be the smallest such @xmath444 .
otherwise set @xmath445 .
[ c1 ] suppose @xmath435 and @xmath446 .
( i ) for any positive integer @xmath447 such that @xmath442 , either @xmath444 is a multiple of @xmath438 , or @xmath448 .
( ii ) @xmath449 and @xmath450 take some @xmath444 such that @xmath442 .
we show in the following paragraph that either @xmath438 divides @xmath444 , or @xmath451 and @xmath452 .
this proves the lemma all that remains is to see that , in the latter case , @xmath448 . if , instead , @xmath453 , then applying the above result also to @xmath443 , we find that both @xmath454 and @xmath455 are perpendicular to @xmath456 . by space concerns on the diagram
, it follows that @xmath457 , but then @xmath458 , which would show that @xmath438 divides @xmath459 , in contradiction to @xmath460 .
so , take any @xmath444 such that @xmath442 , and assume that @xmath444 is not a multiple of @xmath438 .
define @xmath461 by @xmath462 if @xmath463 preserves orientation on @xmath464 , and @xmath465 otherwise .
define @xmath466 as follows . for any @xmath467 ( @xmath151 is the length of the dynkin diagram ) ,
let @xmath112 be given by @xmath468 .
then we define @xmath469 .
define @xmath470 .
clearly , @xmath471 is defined so that if @xmath472 is a connected segment of the diagram , then @xmath471 is constant on @xmath473 , for each fixed @xmath293 , @xmath474 . since @xmath475 for @xmath476 , it follows that @xmath471 is @xmath438-periodic in the first component . for the same reason
, @xmath471 is @xmath444-periodic in the first component .
if @xmath477 , @xmath471 must be @xmath478-periodic in the first component .
this follows since @xmath479 , @xmath480 implies @xmath481 whenever @xmath482 , @xmath483 . by minimality , @xmath484 , which is impossible .
hence , @xmath485 .
furthermore , @xmath486 , because otherwise @xmath487 , which would imply that @xmath488 by the above results applied to @xmath489 , which is clearly contradictory .
[ c3 ] suppose that @xmath490 for @xmath491 where @xmath492 and @xmath493 .
then for any @xmath494 , @xmath495 , where @xmath496 and @xmath497 , then @xmath498 , and @xmath56 implies that @xmath499 . by reversing all directions , given @xmath500 , @xmath501 , with @xmath502 , then for any @xmath503 , @xmath504 , where @xmath496 and @xmath505 , then @xmath498 , and @xmath56 implies @xmath506 . first , it is clear that @xmath0 divides @xmath507 iff @xmath508 for @xmath122 an integer . in this case ,
the theorem is satisfied ; so suppose not . by applying @xmath509 some number of times to @xmath64 or @xmath65
, it suffices to assume @xmath510 .
now , if @xmath511 is defined on @xmath64 , then @xmath512 together with the lemma gives the desired result .
so assume @xmath511 is not defined on @xmath64 , and hence it is not defined on @xmath65 either .
suppose @xmath513 for some positive integer @xmath514 .
write @xmath515 .
now , by applying @xmath516 some number of times to each @xmath517 , we can obtain @xmath518 for some @xmath519 , showing that @xmath511 is defined on @xmath65 , which is a contradiction .
but then @xmath516 is defined the direction - reversed case is the same . for convenience ,
let @xmath520 for any or no superscript @xmath521 .
when @xmath63 acts on some segment of the dynkin diagram ( i.e. some subset of @xmath522 ) by sending @xmath464 to @xmath523 , we say it acts by _ shifting to the right by @xmath127 _ when @xmath127 is positive , and by _ shifting to the left by @xmath127 _ when @xmath127 is negative . in particular , on each segment , @xmath63 acts by _ shifting _ iff orientation is preserved .
the next two lemmas summarize all of the possible passing properties of an opposite - sign quadruple and its image under @xmath153 .
[ pq ] take @xmath524 and @xmath281 .
then exactly one of the following must hold : @xmath525a@xmath526 @xmath527 + @xmath525b@xmath526 @xmath528 similarly , exactly one of the following must hold : @xmath525c@xmath526 @xmath529 + @xmath525d@xmath526 @xmath530 . these results , after interchanging superscripts of @xmath531 with @xmath0 , also hold when one considers @xmath292 and @xmath294 , instead of the original hypothesis .
let @xmath532 and @xmath533 for some @xmath127 ( which exists by construction ) .
suppose that @xmath534 .
we will analyze all possible cases by considering the value of @xmath535 .
write @xmath536 + q$ ] where @xmath537 .
first , i claim that @xmath538 .
suppose instead that @xmath539 . in this case , @xmath540 , which immediately implies from the proof of proposition [ qc ] that @xmath541 preserves orientation on @xmath353 .
so @xmath283 preserves orientation , and by definition @xmath542 preserves orientation on @xmath283 .
now , if we set @xmath543 , we find that @xmath544 because @xmath545 .
now , @xmath540 shows that @xmath546 .
finally , @xmath547 , because @xmath539 .
these facts , however , contradict lemma [ c1 ] .
so , it must be that @xmath548 .
we divide into the two cases , ( 1 ) @xmath549 and ( 2 ) @xmath550 .
first consider the case @xmath551 .
set @xmath552 - po^l(v)$ ] .
then we have three cases : ( i ) @xmath553 , ( ii ) @xmath554 , and ( iii ) @xmath555 . first consider ( i ) .
now , @xmath556 , so @xmath557 .
conversely , whenever @xmath557 , clearly @xmath558 with @xmath559 - po^r(y)$ ] ( we use that @xmath284 always preserves orientation ) . in this case ,
@xmath553 and @xmath549 , as desired .
this situation , characterized by @xmath560 , falls into ( a ) and we will call it ( a1 ) .
next , take ( ii ) . in this case , @xmath561 , so @xmath562 and @xmath563 .
conversely , it is clear that @xmath563 iff @xmath562 from the construction of @xmath153 , and in this case , @xmath558 with @xmath564 .
thus , @xmath565 and @xmath549 , as desired .
this is a different case of ( a ) , so let us call it ( a2 ) . finally , consider ( iii ) . in this case , @xmath566 .
set @xmath567 $ ] .
then it follows that @xmath568 , so that @xmath569 with @xmath570 .
conversely , if @xmath571 , then @xmath572 with @xmath573 + po^l(x)$ ] , as desired ( this can be checked separately when @xmath103 reverses orientation here
@xmath574 so there is no difficulty . )
hence , @xmath575 and @xmath549 .
this is the final case of ( a ) , so let us call it ( a3 )
. next , consider the case @xmath576 .
set @xmath577 $ ] .
because @xmath576 , it follows that @xmath578 , hence @xmath579 with @xmath580 . since @xmath581 iff @xmath582 , we have @xmath583 .
conversely , whenever @xmath579 , then @xmath534 with @xmath584 $ ] . hence , @xmath550 and @xmath583 .
this accounts for case ( b ) .
we have proved the first part of the lemma , because we have considered all possible nonzero values of @xmath585 and @xmath586 , and grouped them into the cases ( a1 ) , ( a2 ) , ( a3 ) , and ( b ) .
we have shown that each of these is associated with different values of @xmath535 , which justifies the zero values of @xmath585 and @xmath586 in each case .
next , we apply the same analysis used in the first part to show that exactly one of ( c),(d ) holds .
let @xmath587 and @xmath588 .
we suppose that @xmath589 and divide into the cases @xmath590 and @xmath591 .
first consider @xmath590 .
then it is clear that @xmath592 , with equality iff @xmath593 .
now , @xmath594 is defined on @xmath595 , and @xmath596 , so @xmath597 preserves orientation on @xmath595 .
this shows that @xmath598 , so that @xmath599 with @xmath600 .
that is , @xmath601 .
conversely , whenever @xmath599 , we know from the fact that @xmath282 preserves orientation that @xmath602 preserves orientation on @xmath603 , and hence that @xmath604 with @xmath605 .
it is then clear that @xmath606 .
this accounts for case ( d ) .
now , suppose that @xmath589 with @xmath591 .
we divide into the cases @xmath540 and @xmath574 .
first suppose @xmath540 .
in particular , this implies that @xmath377 is defined on @xmath595 , and by nilpotency , it must preserve orientation .
also , @xmath541 must preserve orientation on @xmath595 .
so , we see that @xmath607 , thus @xmath608 with @xmath609 .
conversely , if @xmath540 and @xmath610 , we see from @xmath611 and lemma [ c1 ] that @xmath612 .
it follows that @xmath613 with @xmath614 , and @xmath615 , as desired .
this accounts for one situation of ( c ) ; call this ( c1 ) . next , suppose @xmath574 and @xmath591 .
we further divide into the three cases ( i ) @xmath616 , ( ii ) @xmath617 , and ( iii ) @xmath618 . in case ( i )
, we use that @xmath284 preserves orientation to see that @xmath619 , so that @xmath620 with @xmath621 .
conversely , whenever @xmath620 , it follows that @xmath574 using lemma [ c1 ] : otherwise we would have @xmath622 with @xmath623 while + @xmath624 .
hence , @xmath625 with @xmath626 .
call this situation ( c3 ) . in case ( ii )
, we have that @xmath627 .
conversely , whenever @xmath628 , we must have @xmath574 by lemma [ c1 ] , and then @xmath629 and @xmath630 , as desired .
if @xmath631 , then it also follows from lemma [ c1 ] considering @xmath595 that @xmath574 , and then @xmath632 .
call this situation ( c2 ) .
finally , we consider case ( iii ) .
now , @xmath618 shows that @xmath633 . by nilpotency
, @xmath594 must preserve orientation on @xmath634 , and it follows that @xmath635 with @xmath636 .
conversely , if @xmath635 with @xmath574 , then it follows that @xmath637 , so@xmath638 .
hence , @xmath625 with @xmath639 , and @xmath640 .
this has the same passing properties as ( c1 ) , so call this situation ( c1 ) .
we have finished the second half of the lemma , since we have accounted for all possible values of @xmath641 , and @xmath642 in cases ( d ) , ( c1 ) , ( c1 ) , ( c2 ) , and ( c3 ) .
each of these are associated with distinct values of @xmath643 with rexspect to the @xmath644 , and @xmath321 , once @xmath127 is fixed .
to obtain the result for decreasing quadruples , simply reverse all directions and permute the components of all @xmath24-quadruples ( not pairs ! ) in this proof .
[ ppp ] take @xmath645 and @xmath370 .
then @xmath646 , and exactly one of the following hold : @xmath525a@xmath526 @xmath527 .
+ @xmath525b@xmath526 @xmath583 , @xmath647 . under the hypotheses @xmath648 and @xmath649
, these results still hold upon interchanging superscripts of @xmath531 and @xmath0 .
clearly @xmath593 iff @xmath632 .
suppose @xmath650 . hence @xmath651 .
suppose for a contradiction that @xmath620 .
then , @xmath652 .
however , @xmath653 and lemma [ c1 ] imply that @xmath654 , a contradiction .
so , the identity @xmath655 easily follows ( since @xmath656 ) .
the rest of the proof is almost exactly the same as the proof of the first part of lemma [ pq ] , getting rid of @xmath282 .
again , the results follow with simple modifications in the decreasing case . in this section , we consider the longer monomial terms that arise in the expansion of the formula for @xmath11 . as in the previous cases ,
most terms cancel ; we therefore explicitly give the groupings which cancel ( corollary [ top ] ) or almost cancel , and show that what remains is simply @xmath13 , proving theorem [ mt ] . in order to do this ,
we need to define the notion of a @xmath24-chain , which generalizes @xmath24-quadruples to the objects needed to handle the longer monomials , and prove some more combinatorial results regarding these .
@xmath657 @xmath239-chain _ is a chain @xmath658 , so that @xmath210 .
let @xmath659 denote the set of @xmath657 tp - chains .
let @xmath660 denote the set of all @xmath239-chains , i.e. @xmath661 . a chain @xmath662 is said to be _ outer _ if@xmath663 and @xmath664 .
let @xmath665 denote the set of such @xmath657-chains , and let @xmath666 denote the set of all outer chains for any @xmath667 .
lemmas [ ail]-[onp ] below have obvious analogues obtained by changing the sign of all @xmath24-pairs , reversing the order of @xmath24-chains , and replacing @xmath659 with @xmath668 in all forms . these results are not stated but will be referred to in the same manner as the actual results stated .
[ ail ] @xmath525i@xmath526 suppose @xmath669 .
suppose @xmath670 and @xmath671 where @xmath672
. then @xmath673 iff @xmath674 , @xmath675 iff @xmath676 , and @xmath677 iff @xmath678 .
@xmath525ii@xmath526 suppose @xmath679 .
suppose @xmath680 and @xmath681 where @xmath682 and @xmath683 .
then @xmath684 iff @xmath685 for @xmath686 any order relation @xmath687 or @xmath424 .
the proof is the same for both ( i ) and ( ii ) , does not mention @xmath688 , and is given in the following paragraphs .
first , we note that @xmath675 iff @xmath689 and @xmath690 reverse orientation , which is true iff @xmath691 and @xmath692 reverse orientation , which is true iff @xmath676 .
now , we show that @xmath693 implies @xmath694 . to reach a contradiction ,
suppose that @xmath693 and @xmath695 .
by nilpotency , @xmath689 and @xmath692 preserve orientation .
thus @xmath691 and @xmath690 also preserve orientation .
write @xmath696 .
first suppose @xmath697 .
then @xmath698 , so that @xmath699 is defined on @xmath700 and therefore on @xmath701 .
however , this implies that @xmath702 , while @xmath703 .
this contradicts lemma [ c1 ] .
so @xmath704 .
then @xmath705 and @xmath706 .
since @xmath693 , it must be that @xmath707 , which implies that @xmath708 .
so @xmath709 \overline { < } ( z_2)_l + ( y_3)_l$ ] .
now , write @xmath710 + m$ ] , for @xmath711 .
since @xmath712 , it follows that @xmath713 ( @xmath714 ) or @xmath715 ( @xmath716 ) . by lemma [ c1 ]
, it follows that @xmath717 divides both @xmath514 and @xmath718 . but then @xmath719 divides @xmath720 and @xmath721 , which shows that @xmath722 , contrary to assumption . next , suppose @xmath722 .
we show @xmath678 .
find @xmath293 and @xmath112 such that @xmath723 and @xmath724 . in this case , @xmath725 .
hence , @xmath726 . by assumption , @xmath727 .
since @xmath728 , this in particular implies that @xmath678 , as desired .
[ ail2 ] @xmath525i@xmath526 suppose @xmath729 with @xmath730 and + @xmath731 .
let @xmath732 and suppose @xmath733 .
. then @xmath694 .
@xmath525ii@xmath526 similarly , suppose @xmath735 with @xmath736 and @xmath737 .
then set @xmath738 and suppose @xmath733 . setting @xmath739
, it follows that @xmath694 .
again , parts ( i ) and ( ii ) have nearly the same proof , which follows .
suppose , on the contrary , that @xmath740 .
clearly @xmath741 , else @xmath692 would have reversed orientation , which is not possible by lemma [ nc ] .
so @xmath678 .
now , @xmath742 is defined on @xmath701 since , in case ( i ) , @xmath742 is defined on @xmath743 and @xmath744 , which follows from @xmath745 , and in case ( ii ) , @xmath742 is defined on @xmath743 . in particular , @xmath746 is defined on @xmath701 .
but now , @xmath747 but @xmath748 , contradicting lemma [ c1 ] . [ pmol ]
suppose @xmath749 with @xmath750 and @xmath751 . then @xmath752 iff @xmath753 and @xmath754 iff @xmath755 .
first we note that , if @xmath752 , then @xmath756 , for the following reason .
suppose @xmath752 and write @xmath757 or @xmath758 , depending on whether @xmath759 is bad or good . then @xmath760 , so lemma [ ail2 ] applies and shows that @xmath761 , as desired . by symmetry
, @xmath754 implies that @xmath755 .
so , it suffices to show that @xmath752 if @xmath762 and @xmath754 if @xmath763 . by the symmetry of the situation , we need only prove the first . assume , for sake of contradiction , that @xmath754 and @xmath764 .
first suppose that @xmath765 . then we note that @xmath766 is defined on @xmath767 because it is defined on @xmath768 and @xmath769 .
next we note that @xmath770 because @xmath771 .
also , @xmath772 , and @xmath773 .
this contradicts lemma [ c1 ] . on the other hand ,
it is impossible that @xmath774 shifted @xmath775 to the right by @xmath776 , while @xmath777
. this would contradict corollary [ c3 ] .
[ bps ] suppose @xmath735 and @xmath778
. then @xmath779 .
clearly @xmath780 would imply that @xmath781 and @xmath689 reverse orientation , which is not possible since latexmath:[$|x_1| > that @xmath742 , and hence @xmath783 , is defined on @xmath743 , and hence @xmath784 . since @xmath785
, we find that @xmath786 \overline { < } t^{\o(x_2 ) } [ ( x_2)_l + ( y_2)_l ] \gtrdot t^{\o(y_2 ) } [ ( x_2)_l + ( y_2)_l]$ ] , contradicting lemma [ c1 ] .
so @xmath779 .
[ onp ] suppose @xmath787 with @xmath788
. then @xmath789 and @xmath790 .
first , note that @xmath766 is defined on @xmath768 since it is defined on @xmath791 , and the former is a subset of the latter on the diagram . if @xmath792 , then set @xmath793 . if @xmath794 and @xmath795 , set @xmath796 . in either case ,
@xmath797 is defined on @xmath798 .
indeed , @xmath799 is defined on @xmath798 , and @xmath800 .
note that , by lemma [ nc ] , @xmath688 and @xmath690 must preserve orientation .
hence , @xmath801 and @xmath799 must preserve orientation on @xmath802 .
set @xmath803 .
we have @xmath804 , contradicting lemma [ c1 ] .
now , we are ready to define maps @xmath805 , and @xmath806 which parameterize terms which cancel in the expansion of @xmath807 ( given in corollary [ top ] ) . for
any map @xmath808 taking two arguments , let @xmath809 be the map @xmath808 applied to the @xmath215 and @xmath216-th components of some larger @xmath127-tuple .
similarly , for any map @xmath810 taking only one argument , define @xmath811 to be the map @xmath810 applied to the @xmath215-th component of a larger @xmath127-tuple .
define @xmath812 for @xmath813 as follows . for @xmath814 ,
let @xmath815 .
define @xmath816 for @xmath817 by @xmath818 . for @xmath819 , define @xmath820 by @xmath821 note that , unlike in the case of the map @xmath404 ( i.e. the case of @xmath822 chains ) , we can have chains on which both maps @xmath823 and @xmath824 are defined .
we see , however , that in many cases ( namely , the hypotheses of lemma [ pmol ] ) , only one map will yield an outer chain .
when @xmath825 and @xmath826 , @xmath103 is said to be _ left - passed _ and we write @xmath827 . if @xmath828 and @xmath829 or @xmath830 , then @xmath103 is said to be _ right - passed _ or _ half - passed _ , and @xmath831 or @xmath832 , respectively .
note that , given @xmath828 with @xmath833 , @xmath834 iff @xmath835 , so in these cases the half - passed chains not only have a half right - passed pair but also a half left - passed pair .
define @xmath836 by @xmath837 @xmath838 similarly , define @xmath839 by @xmath840 . as in the case of @xmath841 ,
we have the following : [ ppi ] let @xmath842 .
then @xmath843 iff @xmath844 for some sign @xmath389 or @xmath845 for some superscript @xmath846 .
this follows easily from proposition [ qc ] . a chain @xmath828 is said to be _ negatively special _ if @xmath847 and @xmath848 .
the set of such chains is denoted @xmath849 .
similarly , @xmath850 is _ positively special _ ( or in @xmath851 ) if @xmath852 and latexmath:[$|x_{m}| > general term will be simply _
special_. a chain @xmath828 is said to be _ negatively reversed _ if @xmath854 and @xmath855 and @xmath856 both reverse orientation .
the set of such chains is @xmath857 .
a chain @xmath828 is _ positively reversed _ ( or in @xmath858 ) if @xmath859 and @xmath860 and @xmath861 both reverse orientation .
as before , @xmath862 is the set of _ reversed _ chains .
[ rcl ] if @xmath863 , then @xmath864 or @xmath865 .
if @xmath866 , then @xmath867 for any @xmath868 .
in fact , if @xmath869 , then @xmath824 is not defined on @xmath103 , and @xmath870 satisfies @xmath871 .
additionally , @xmath872 or @xmath873 where @xmath874 . similarly ,
if @xmath875 , then @xmath823 is not defined on @xmath103 , and @xmath876 satisfies @xmath877 .
additionally , @xmath878 or @xmath879 where @xmath880 .
this easily follows from lemmas [ nc ] and [ psq ] .
[ rs ] no outer chain is both reversed and special .
we define maps @xmath881 as follows . if @xmath882 where @xmath883 , set @xmath884 . if @xmath885 , set @xmath886 , and if @xmath887 , set @xmath888 .
similarly , if @xmath889 where @xmath890 , set @xmath891 .
if @xmath892 , set @xmath893 , and if @xmath894 , then set @xmath895 . [ pg1 ] ( i ) take @xmath896 for some fixed sign @xmath389 .
then @xmath897 or @xmath898 , but not both . in the former case , @xmath899 as well .
\(ii ) if @xmath900 , for @xmath901 , then either ( a ) @xmath902 , ( b ) @xmath903 , or ( c ) neither . in case ( c ) , @xmath904 . in case ( a ) , @xmath905 , and in case ( b ) , @xmath906 .
\(iii ) if @xmath907 $ ] , then @xmath908 . additionally , @xmath909 if @xmath910 and @xmath911 if @xmath912 $ ] .
\(i ) this follows immediately from lemma [ ail ] in the cases that @xmath913 and @xmath914 .
( note that it is impossible to have @xmath904 by corollary [ rs ] . ) in the case that @xmath915 and @xmath916 or in the case that @xmath917 and @xmath918 , it is clear that @xmath898 .
\(ii ) it follows that only one of the @xmath919 can be in @xmath666 from lemma [ pmol ] .
if @xmath920 , then no @xmath921 is outer by lemma [ rcl ] . by lemma [ pmol ] , if @xmath922 , there is nothing to prove . if @xmath923 and @xmath924 , then lemma [ ail2 ] gives the desired result . for @xmath925
the result is easy .
\(iii ) the two statements follow from lemmas [ onp ] , [ bps ] , respectively .
we define the following subsets of @xmath666 : + ( a ) @xmath926 + ( a ) @xmath927 + ( b ) @xmath928 \mid \psi_{\pm}(x ) \in tpc^o \}$ ] + furthermore , let @xmath929 , @xmath930 , and @xmath931 . [ ds ] the subsets @xmath932 and @xmath933 are disjoint and their union is all of @xmath666 . that @xmath934 , and @xmath935 are disjoint from @xmath936 follows from corollary [ rs ] , lemma [ rcl ] , and the definition , respectively .
the other facts regarding disjointedness are obvious .
to check that the union is all of @xmath666 , we apply lemma [ pg1].ii , which shows that @xmath937 $ ] .
it is clear , though , that @xmath938 .
this proves the desired result .
[ ctc ] @xmath525i@xmath526 for each choice of sign , @xmath939 maps @xmath940 injectively to @xmath941 .
+ @xmath525ii@xmath526 @xmath942 maps @xmath943 \cap [ rpc \cup hpc]$ ] injectively to @xmath944 $ ] and @xmath945 maps @xmath946 \cap lpc$ ] injectively to @xmath947 .
+ @xmath525iii@xmath526 @xmath948 = \emptyset$ ] .
+ @xmath525iv@xmath526 @xmath949 $ ] and @xmath950 $ ] are both disjoint from @xmath951 .
+ @xmath525v@xmath526 @xmath952 is disjoint from @xmath666 for each choice of sign .
\(i ) this follows immediately from lemmas [ pg1].i and [ ppi ] .
\(ii ) this follows immediately from lemmas [ pg1].ii , [ ppi ] , and [ ppp ] .
\(iii ) this follows immediately from lemma [ onp ] .
\(iv ) this follows immediately from the definition of @xmath953 .
\(v ) this follows from lemma [ pg1].i , since we know @xmath951 is disjoint from @xmath936 .
[ top ] @xmath666 is partitioned by all sets of the form : + @xmath525a@xmath526 for @xmath954 , @xmath955 @xmath525a@xmath526 those of the form , replacing @xmath138 with @xmath956 and @xmath531 with @xmath0 .
( call these also @xmath957 , distinguished from the above by the `` sign '' of @xmath103 . )
+ @xmath525b@xmath526 for a fixed @xmath958 , @xmath959 @xmath525b@xmath526 those of form , replacing @xmath138 with @xmath956 , @xmath531 with @xmath0 , and @xmath960 with @xmath933 .
call these @xmath957 as well . here
it is meant that , if @xmath961 is not defined on @xmath103 , then @xmath962 for any @xmath284 . for every @xmath954 , we see from the lemma , part ( i ) , that @xmath963 . furthermore , @xmath964 and @xmath965 where these are defined ( lemma , part ii ) .
now take any @xmath966 .
first note that either @xmath967 or @xmath968 because @xmath969 .
note that @xmath970 because @xmath971 and @xmath972 are disjoint from @xmath666 ( lemma [ rcl ] ) .
now , if @xmath973 , then using lemma [ ds ] , @xmath974 ( of type a or b , depending on whether @xmath975 or @xmath968 , respectively ) . if @xmath976 , then clearly @xmath966 . in this case ,
@xmath977 by the lemma , parts ( ii ) , ( iv ) , and ( v ) , using lemma [ ppi ] . furthermore , in this case @xmath978 , again from the lemma , parts ( iv ) and ( v ) . then , @xmath979
. thus , we see that all elements @xmath966 fit into at least one of the sets ( a ) or ( b ) . to see that all elements in @xmath980 fit into at most one , note that the way in which @xmath284 appears is uniquely determined by which of the following holds : ( a ) @xmath973 , @xmath981 ; ( b ) @xmath973 , @xmath982 ; ( c ) @xmath976 , @xmath983 .
now , it remains to consider those elements of @xmath936 . for any @xmath984 , we know that @xmath985 by lemma [ rcl ] .
the former two are subsets of @xmath935 , so we know that @xmath986 for any @xmath984 , and clearly @xmath987 is unique .
the negative case is almost identical to this one . for @xmath988 , set @xmath989 if @xmath990 or @xmath991 and @xmath144 otherwise .
now , given @xmath992 , set @xmath993 .
[ jrjl ] the terms in the expansion of @xmath807 are @xmath995 for each @xmath996 , as well as the terms @xmath997 for @xmath164 and @xmath998 for each @xmath215 . [ cop ] @xmath525i@xmath526 if @xmath999 , then @xmath1000 .
@xmath525ii@xmath526 if @xmath1001 , then @xmath1002 .
\(i ) let @xmath954 .
the negative case is similar ( see comments at the end of the proof of this part ) .
let @xmath828 and let @xmath1003 be the part of the formula for @xmath1004 which will not change upon applying @xmath823 , @xmath945 , and @xmath1005 ( where applicable ) .
let @xmath1006 if @xmath827 and @xmath1007 otherwise .
set @xmath1008 .
first we show that @xmath1009 first , by lemma [ pq ] , @xmath1010 hence , @xmath1011 if @xmath1012 , then @xmath1007 and @xmath1013 . in this case , follows from the definition of @xmath1014 . if @xmath827 , then @xmath1015 , and implies @xmath1016 .
then lemma [ ppp ] shows that @xmath1017 so that the left - hand side of is @xmath1018 . by , this is the same as the right - hand side .
next , we show that @xmath1019 f.\ ] ] naturally , we may assume that @xmath860 reverses orientation . since , in this case , @xmath1020 also reverses orientation , it must be that @xmath1021 .
suppose now that @xmath1022 and @xmath1023 .
now , for every @xmath293 such that @xmath1024 , set @xmath1025 , @xmath1026 , and @xmath1027 . then @xmath1028 are exactly those reversed chains such that @xmath1029 .
now , we consider the possible passing properties of the @xmath24-pairs involved .
note first that , since @xmath1030 , @xmath1031 for all @xmath293 .
for the same reason , @xmath1032 for all @xmath293 .
next , note that @xmath1033 for all @xmath293otherwise , applying @xmath1034 to @xmath689 would contradict nilpotency .
for the same reason , @xmath1035 for all @xmath293 .
next , note that @xmath1036 for all @xmath293 .
this follows from the fact that @xmath1037 for all @xmath293 , making use of lemma [ pq ] .
now , follows readily .
let @xmath1038 when @xmath1039 and @xmath1040 otherwise .
finally , we show that @xmath1041 the proof is similar to the proof two paragraphs back . if @xmath1042 , then @xmath1040 . by lemma [ pq ] , @xmath1043 .
thus @xmath1044 , proving in this case ( note that @xmath1045 and recall the definition of @xmath1014 . ) if @xmath1039 , then @xmath1046 and by lemma [ pq ] ( or the equation above ) , @xmath1047 and @xmath1048 .
if we set @xmath1049 , then it follows from lemma [ ppp ] that neither @xmath1050 nor @xmath1051 is passed , so @xmath1052 .
hence , @xmath1053 , so the left - hand side of is @xmath1054 , which is equal to the right - hand side by corollary [ psq ] and the above analysis . now , putting , , and together
, we get @xmath1055 as desired .
the negative case is almost the same as the above , except that the `` error term '' @xmath1014 over - corrects , but this is counteracted by the fact that now @xmath942 is defined on @xmath832 as well as @xmath1056 .
( the details are omitted . )
\(ii ) this is almost the same as the proof above .
first we note that the result is clear when @xmath1057 , as @xmath1058 and @xmath1004 gives the desired formula ( bearing in mind the definition of @xmath1059 ) .
suppose @xmath1060 .
let @xmath1006 and let @xmath1061 if @xmath827 and @xmath1062 , and @xmath1040 otherwise .
let @xmath1063 f.\ ] ] lemma [ ppp ] shows that @xmath1064 and @xmath1065 first suppose @xmath1040 . in this case
, implies easily .
if @xmath1066 , then @xmath1067 , and @xmath1068 . then , @xmath1069 .
furthermore , @xmath1070 . from this follows .
next , we claim that @xmath1071 f.\ ] ] this follows by exactly the same arguments as used in the second paragraph of the proof of part ( i ) .
but , @xmath1064 , so , putting and together , the result easily follows .
the negative/@xmath933 case is almost the same as this , bearing in mind the final comments in the proof of the previous part .
_ proof of the main theorem , [ mt ] .
_ this follows immediately from lemmas [ jrjl ] and [ cop ] and corollary [ top ] , using for @xmath13 .
i would like to thank pavel etingof for introducing me to this problem and advising me .
i would also like to thank the harvard college research program for their support .
finally , i am indebted to gerstenhaber , giaquinto , and hodges for valuable discussions and for sharing some unpublished results .
in this section we explicitly compute @xmath13 for generalized cremmer - gervais triples , the only triples satisfying @xmath1072 . recall from example [ gcge ] the results given in @xcite : these triples are indexed by @xmath1073 where @xmath1074 is the dimension , @xmath514 is relatively prime to @xmath151 , @xmath1075 , @xmath1076 , and @xmath1077 , where @xmath160 give the residue mod @xmath151 in @xmath161 . moreover when @xmath100 is taken to have trace zero in the first component , it is uniquely given by @xmath1078 and @xmath1079 where @xmath1080 .
then , the only difficulty is in computing @xmath1081 , so here we use to prove .
clearly we have @xmath1082 where @xmath1083 , for @xmath1084 , since @xmath338 for all @xmath103 .
it suffices , then , to show @xmath1085 for all @xmath103 .
take @xmath1086 .
below we use the notation @xmath1087 = 1 $ ] if `` statement '' is true and 0 if it is false .
one sees that @xmath1088 - \frac{1}{2}[i = k ] - \\
\frac{1}{n}\text{res}\bigl(\frac{i+k-2j}{m}\bigr ) - \frac{1}{n}\text{res } \bigl(\frac{k - i}{m}\bigr ) \\ = 1 - \frac{1}{2}[2j = i+k ] - \frac{1}{2}[i = k ] - \frac{2}{n } \text{res}\bigl(\frac{k - j}{m}\bigr ) + m_{i , j , k},\end{gathered}\ ] ] where @xmath1089[\text{res}\bigl(\frac{k - j}{m}\bigr ) > \text{res}\bigl(\frac{i+k-2j}{m } ) ] \\ - [ i \neq k][\text{res}\bigl(\frac{k - j}{m}\bigr ) < \text{res}\bigl(\frac{k - i}{m}\bigr)].\end{gathered}\ ] ] thus , since @xmath1090 , it suffices to show @xmath1091 + \frac{1}{2}[i = k ] + p^r(x ) + p^l(x ) = 1 + m_{i , j , k}$ ] . note that @xmath1092 iff @xmath632 and @xmath1093 iff @xmath562 , and in these cases @xmath1094 , so it suffices to consider @xmath1095 . in this case
we need to show @xmath1096 . now , @xmath1097 + [ \text{res}\bigl(\frac{k - j}{m}\bigr ) > \text{res}\bigl(\frac{k - i}{m}\bigr)],\end{gathered}\ ] ] and it is not difficult to see that @xmath1098 = p^l(x)$ ] while @xmath1099 = p^r(x)$ ] .
this finishes the proof .
m. gerstenhaber and a. giaquinto , and s. d. schack , _ construction of quantum groups from belavin - drinfeld infinitesimals _ , israel math .
* 7 * ( 1993 ) , 4564 .
a. giaquinto and t. j. hodges , _ nonstandard solutions of the yang - baxter equation _
* 44 * ( 1998 ) , 6775 . | we prove the ggs conjecture @xcite ( 1993 ) , which gives a particularly simple explicit quantization of classical @xmath0-matrices for lie algebras @xmath1 , in terms of a matrix @xmath2 which satisfies the quantum yang - baxter equation ( qybe ) and the hecke condition , whose quasiclassical limit is @xmath0 .
the @xmath0-matrices were classified by belavin and drinfeld in the 1980 s in terms of combinatorial objects known as belavin - drinfeld triples .
we prove this conjecture by showing that the ggs matrix coincides with another quantization from @xcite , which is a more general construction .
we do this by explicitly expanding the product from @xcite using detailed combinatorial analysis in terms of belavin - drinfeld triples . |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.